UNIVERSITEIT GENT
FACULTEIT ECONOMIE EN BEDRIJFSKUNDE ACADEMIEJAAR 2009 – 2010
COUNTERFEITING AND CONSUMER BEHAVIOUR
Masterproef voorgedragen tot het bekomen van de graad van Master in de Toegepaste Economische Wetenschappen
Dennis De Cat onder leiding van Prof. Dr. I. Vermeir
UNIVERSITEIT GENT
FACULTEIT ECONOMIE EN BEDRIJFSKUNDE ACADEMIEJAAR 2009 – 2010
COUNTERFEITING AND CONSUMER BEHAVIOUR
Masterproef voorgedragen tot het bekomen van de graad van Master in de Toegepaste Economische Wetenschappen
Dennis De Cat onder leiding van Prof. Dr. I. Vermeir
PERMISSION Ondergetekende verklaart dat de inhoud van deze masterproef mag geraadpleegd en/of gereproduceerd worden, mits bronvermelding. Dennis De Cat
ACKNOWLEDGEMENTS First and foremost, I would like to express my gratitude to my promoter Prof. Dr. Iris Vermeir. She was always prepared scheduling a face-to-face meeting to discuss the progress of my research. Above all, I want to thank her for replying to dozens of emails immediately. Her help in analyzing and interpreting the results of my experiment was indispensable. I am grateful for the opportunity to have studied under her direction. Further, I would like to thank Prof. Boonghee Yoo, Prof. Aron O’Cass, Prof. Mateja Kos Koklic, Prof. Judy W. Spain, Prof. Elfriede Penz, Prof. Giacomo Gistri, Prof. Dr. Celso Augusto de Matos, Prof. Dr. Judy Zaichkowsky, Prof. Ian Phau and his assistant Min Teah for answering my questions about the papers they have written and for giving me additional literature tips. By doing so, they gave me the opportunity to gain a very proficient insight in the subject of counterfeiting and consumer behaviour. I also want to mention our head of department, Prof. Dr. Patrick Van Kenhove, as he was a great help in analyzing my research results. I also want to thank all my closest friends who helped me gathering participants for my online questionnaire. Especially Philippe Lescornez did a great effort in making sure I had enough conscious respondents. Kevin De Cock was a great help in analyzing the results of my research with SPSS. My special and appreciative thanks must be given to Tony Van Gulck for providing me the opportunity to have an in-depth interview with Miss. Hagenaers (Trade Mark and Design attorney) about the legal topic of intellectual property right violation. On top of this, as CEO of BOO! Fashion, he sponsored my research by providing me with purses to win in order to gather participants for my online questionnaire. Last but not least I owe my appreciation and thanks to my parents, Viviana De Letter and Marc De Cat as well as to my girlfriend, Ine Lescornez. As this thesis is the masterpiece after four years of hard work I would like to thank them for all emotional, psychological and financial support during my study of Applied Economics. Without them I would not be the person I am right now.
I
SUMMARY (NEDERLANDS) Verantwoording De grootte van de (illegale!) namaakindustrie neemt snel toe. Volgens gegevens van de Europese Commissie stijgt het aantal namaakartikelen dat men onderschept aan de grenzen van Europese lidstaten elk jaar. De OECD schat de waarde van de namaakgoederen die in 2005 internationaal verhandeld werden op 200 miljard dollar. Het aantal productcategorieën (bv. elektrische onderdelen, medicijnen, cd’s, etc.) waarvoor men namaakartikelen op de markt vindt, kent een continue groei. Desalniettemin
is
het
probleem
nog
steeds
het
grootst
in
de
mode-
en
luxegoederenindustrie. Namaak wordt gezien als dé misdaad van de 21ste eeuw omwille van de negatieve gevolgen die hieraan verbonden zijn: productrisico’s, ondersteuning van terroristische organisaties, verlies van werkgelegenheid etc. Het is dan ook niet verwonderlijk dat veel (academisch) onderzoek zich richt op het bepalen van determinanten van vraag- en aanbodzijde van de (luxe)namaakindustrie, evenals op het zoeken naar tactieken die de handel in namaakgoederen kan terugdringen. Vele auteurs wijzen op de noodzaak om dit probleem product-, industrie- en cultuurspecifiek te onderzoeken. Focus Dit onderzoek richt zich op de vraagzijde van namaakhandel. Indien er geen vraag meer is, zal immers ook het aanbod wegebben. De context van dit onderzoek wordt toegespitst op ‘niet bedrieglijke namaakartikelen’ nl. namaakartikelen waarvan men op het moment van aankoop weet dat ze effectief namaak zijn. De literatuur over het ontwikkelen van strategieën om de namaakindustrie te bestrijden, suggereert dikwijls dat consumenten meer bewust gemaakt moeten worden van de negatieve gevolgen die geassocieerd worden met namaakhandel. In dit onderzoek wordt de impact nagegaan van boodschappen met negatieve gevolgen van de namaakhandel op de attitude t.o.v. het kopen van namaak luxemodegoederen. De onderzoeker gaat na of er een verschil bestaat in attitude tussen consumenten die bewust gemaakt worden van persoonlijke risico’s (bv. prestatierisico, fysiek risico, etc.) verbonden aan namaakhandel en consumenten die geconfronteerd worden met II
maatschappelijke risico’s (bv. kinderarbeid in de productie van deze items) verbonden aan deze illegale praktijk. Verder zal ook de relatie tussen attitudes en aankoopintentie onderzocht worden. Recente academische bronnen beweren dat de prijs en kwaliteit van namaakartikelen zijn toegenomen. Daarom zullen we de impact nagaan van deze elementen
op
de
gepercipieerde
kwaliteit
van
luxueuze
modegerelateerde
namaakgoederen. De auteur onderzoekt tevens de impact van deze gepercipieerde kwaliteit op attitudes en aankoopintentie. Het feit dat namaak steeds meer en makkelijker beschikbaar wordt, wordt opgenomen als mogelijke determinant van de aankoopintentie. Dit gebeurt door integratie van het construct ‘perceived behavioural control’. Verder neemt dit onderzoek ook modebewustzijn op als factor die aankoopintentie mogelijk beïnvloedt. Subjectieve normen en vroeger aankoopgedrag worden eveneens onderzocht. Onderzoeksopzet Er wordt een pre-test opgezet die ervoor moet zorgen dat we in het echte onderzoek ‘persoonlijk risico’ en ‘maatschappelijk risico’ correct manipuleren. Met betrekking tot het hoofdonderzoek worden de data verkregen via online vragenlijsten. De populatie bestaat zowel uit studenten als niet-studenten. Er wordt gebruik gemaakt van een 2 (kwaliteit: laag/gemiddeld) x 2 (prijs: laag/gemiddeld) x 2 (type boodschap: persoonlijk risico/maatschappelijk risico) ‘between-subjects’ experimenteel design. Respondenten worden ‘at random’ aan één van de acht condities toegewezen. De onderzoeker gebruikt bestaande meetschalen om de constructen te meten. Resultaten Een
manipulatiecontrole
geeft
aan
dat
de
manipulaties
van
persoonlijk
en
maatschappelijk risico gelukt zijn. Bovendien werden alle boodschappen in de acht condities als zeer geloofwaardig beschouwd. De betrokkenheid bij de boodschap was echter hoger in het geval dat deze informatie bevatte over de maatschappelijke risico’s van namaakhandel. Er worden twee ‘multiple hierarchical’ regressies uitgevoerd: één op attitude en één op aankoopintentie. De groep waartoe men behoort (student/werkend), de gepercipieerde kwaliteit van modegerelateerde namaakgoederen (+), het type boodschap waarmee men geconfronteerd wordt (-) en subjectieve norm (-) blijken significante voorspellers III
van de attitude t.o.v. het kopen van namaakgoederen. De negatieve invloed van het type boodschap duidt aan dat attitudes negatiever zijn bij boodschappen over maatschappelijke
risico’s
dan
bij
boodschappen
over
persoonlijke
risico’s.
Modebewustzijn (+), vroeger aankoopgedrag (+), attitude (+), gepercipieerde kwaliteit (+) en subjectieve norm (-) blijken significante voorspellers van de aankoopintentie voor modegerelateerde luxegoederen. Praktische implicaties De resultaten dragen bij tot het beter begrijpen van de attitude van Belgische consumenten t.o.v. het kopen van modegerelateerde luxueuze namaakartikelen. Er worden ook belangrijke inzichten verworven in de determinanten van aankoopintentie. Vooral de bevindingen betreffende de belangrijke invloed van vroeger gedrag, subjectieve norm, gepercipieerde kwaliteit en type boodschap als determinanten van attitude en intentie kunnen van groot belang zijn in het ontwikkelen van campagnes tegen namaakhandel. Beperkingen Er werd geen controlegroep opgenomen in het onderzoek. Hierdoor kunnen we de attitude van respondenten die geconfronteerd werden met een boodschap niet vergelijken met mensen die niet blootgesteld waren aan dergelijke boodschap. Bijgevolg kan dus enkel de invloed van de verschillende boodschappen relatief t.o.v. elkaar beschouwd worden. Het onderzoek werd enkel uitgevoerd bij mensen met een Belgische nationaliteit. Verder onderzoek is nodig in andere landen en over andere productcategorieën om te testen of de bevindingen stand houden.
IV
TABLE OF CONTENTS ACKNOWLEDGEMENTS................................................................................................. I SUMMARY (NEDERLANDS) .......................................................................................... II TABLE OF CONTENTS ..................................................................................................V ABBREVIATIONS .........................................................................................................VII LIST OF TABLES .........................................................................................................VIII LIST OF FIGURES .........................................................................................................IX 1. Introduction.............................................................................................................. 1 2. Defining counterfeit trade in luxury fashion brands. ........................................... 4 3. Exploring the supply- and demand-side of counterfeiting .................................. 7 3.1. 3.2.
Supply side of counterfeiting ............................................................................................... 7 Demand side of counterfeiting ............................................................................................ 8
4. Determinants of purchase intention for counterfeit luxury fashion items: theories, constructs and hypothesis development.............................................. 9 4.1. 4.1.1. 4.1.2. 4.1.3. 4.1.4. 4.1.5. 4.1.6. 4.1.7. 4.1.8. 4.1.9. 4.1.10. 4.1.11. 4.2. 4.2.1. 4.3.
Intrinsic Determinants ...........................................................................................................10 Attitude toward buying CLFI, and intention to purchase CLFI........................................... 10 Subjective norms....................................................................................................................................... 11 Perceived behavioural control............................................................................................................ 12 Perceived risk.............................................................................................................................................. 13 Perceived harm and actor-proximity ............................................................................................... 14 Perceived Personal harm ..................................................................................................................... 15 Perceived Societal harm ....................................................................................................................... 15 Fashion consciousness.......................................................................................................................... 16 Past behaviour............................................................................................................................................ 17 Price/quality inference ............................................................................................................................ 18 Perceived quality ....................................................................................................................................... 18 Extrinsic Determinants..........................................................................................................19 Price ................................................................................................................................................................. 19 Conceptual model for intention to purchase CLFI ....................................................21
5. Summary of the research focus ........................................................................... 22 6. Pre-test ................................................................................................................... 23 6.1. 6.2. 6.3. 6.4.
Products and brands .............................................................................................................23 Perceived personal harm .....................................................................................................24 Perceived societal harm .......................................................................................................24 Price, quality, perceived harm and message credibility .........................................25
7. Main research......................................................................................................... 27 7.1. 7.1.1. 7.1.2. 7.1.3. 7.1.4. 7.2. 7.2.1.
Methodology..............................................................................................................................27 Data collection ............................................................................................................................................ 27 Sample............................................................................................................................................................ 27 Stimuli.............................................................................................................................................................. 28 Measures....................................................................................................................................................... 29 Results .........................................................................................................................................31 Manipulation check .................................................................................................................................. 31
V
7.2.2. 7.2.3. 7.2.4.
Descriptive statistics ................................................................................................................................ 33 Message credibility and message involvement......................................................................... 34 Development of regression models for attitude toward purchasing CLFI and for intention to purchase CLFI................................................................................................................... 34
7.2.4.1. 7.2.4.2. 7.2.4.3.
7.2.5. 7.2.6.
Regression model for attitude toward purchasing CLFI....................................................................35 Regression model for purchase intention of CLFI ................................................................................37 Mediation................................................................................................................................................................40
The influence of price and quality messages on perceived quality ................................ 41 Summary ....................................................................................................................................................... 42
8. Special topic: explorative questions about counterfeiting and its harming effects on society, businesses and individuals.................................................. 43 9. Discussion and implications ................................................................................ 45 10. Research limitations and recommendations ...................................................... 50 REFERENCES............................................................................................................... 52 ANNEX 1.1: Questionnaire used for pre-test............................................................. 60 ANNEX 1.2: Statistical analyses of pre-test............................................................... 68 1. 1.1 1.2 2. 3. 4. 5.
Products and brands .............................................................................................................68 Product relevance..................................................................................................................................... 68 Brands............................................................................................................................................................. 68 Personal harm...........................................................................................................................69 Societal harm ............................................................................................................................70 Price, quality, perceived harm and message credibility .........................................71 Credibility ...................................................................................................................................72
ANNEX 2: Scenarios (stimuli) used in the main research........................................ 73 ANNEX 3.1: Questionnaire used in the main research............................................. 77 ANNEX 3.2: Statistical analyses of main research ................................................... 86 1. 2. 3. 4. 5.
Manipulation check.................................................................................................................86 ANOVA and Post Hoc test to evaluate possible differences in Ad credibility88 T-test for identifying differences in Message involvement according to message type ............................................................................................................................91 T-test for evaluating differences in attitude depending on the message one has been exposed to. .............................................................................................................91 ANOVA and Post Hoc test for evaluating differences in attitude depending on the group one finds him/herself in. ..................................................................................92
ANNEX 4: Explorative questions ................................................................................ 94
VI
ABBREVIATIONS CLFI
Counterfeit Luxury Fashion Item
EC
European Commission
IPR
Intellectual Property Rights
OECD
Organisation for Economic Co-operation and Development
PBC
Perceived Behavioural Control
RFID
Radio Frequency Identification
SN
Subjective Norm
VII
LIST OF TABLES TABLE 1: PERCEIVED PERSONAL HARM PER RISK TYPE & GENDER DIFFERENCES ...............................................................24 TABLE 2: PERCEIVED SOCIETAL HARM PER SOCIETAL CONSEQUENCE & GENDER DIFFERENCES ..................................25 TABLE 3: MESSAGE CREDIBILITY PER PRICE AND QUALITY SITUATION & GENDER DIFFERENCES ...................................26 TABLE 4: DEMOGRAPHIC PROFILE OF RESPONDENTS .....................................................................................................................28 TABLE 5: CRONBACH'S ALPHA ANALYSIS FOR PERCEIVED SOCIETAL HARM AND PERCEIVED PERSONAL HARM ........32 TABLE 6: MANIPULATION CHECK FOR PERCEIVED HARM ................................................................................................................32 TABLE 7: DESCRIPTIVE STATISTICS FOR OUR DATA SET .................................................................................................................33 TABLE 8: REGRESSION MODEL OF PREDICTORS FOR ATTITUDE TOWARDS PURCHASING CLFI ......................................35 TABLE 9: ATTITUDE AND MESSAGE TYPE DIFFERENCES .................................................................................................................37 TABLE 10: REGRESSION MODEL FOR PREDICTORS OF PURCHASE INTENTION FOR CLFI..................................................38 TABLE 11: PERCEIVED QUALITY AND PRICE & QUALITY CONDITION DIFFERENCES ...............................................................41 TABLE 12: SUMMARY OF RESEARCH RESULTS ...................................................................................................................................42
VIII
LIST OF FIGURES FIGURE 1: CLASSIFICATION SCHEME OF COUNTERFEITING AND RELATED ITEMS (STAAKE ET AL., 2009)..................... 4 FIGURE 2: CONCEPTUAL MODEL OF CONSUMER COMPLICITY (CHAUDRY AND ZIMMERMAN, 2008; ADAPTED FROM CHAUDRY AND STUMPF, 2007)...........................................................................................................................................10 FIGURE 3: CONCEPTUAL MODEL FOR INTENTION TO PURCHASE COUNTERFEIT LUXURY FASHION ITEMS ....................21 FIGURE 4: CHANCE OF ARREST INCENTIVIZING RESPONDENTS NOT TO BUY CLFI ANYMORE ..........................................44
IX
1. Introduction Many reports state that the overall magnitude of counterfeiting is obviously on the rise. For instance, the Taxation and Customs Union from the European Commission (EC) reports that the number of counterfeit articles detained in EU member states only, has risen from 25 million in 1999 to 178 million in 2008 (European Commission, 2008). The most frequently cited number to ‘value’ counterfeiting is the one the OECD proposes in their analysis of international trade data. They suggest that, worldwide, up to USD 200 billion of internationally traded products could have been counterfeit in 2005. However, one must remain critical. As the OECD report suggests itself, “available information on counterfeiting and piracy falls far short of what is needed for robust analysis and for policymaking”. Chaudry and Zimmerman (2008) even state it is virtually impossible to determine the real size of the worldwide counterfeit product market as it concerns an illegal activity. Green and Smith (2002) blame the fact that there exists no exact standard or agreement about the factors that should be taken into account when calculating the monetary value of counterfeiting for this non-transparency. Nevertheless, they also suggest that product counterfeiting is significant and growing. This statement is confirmed by previous research on the reasons of counterfeit growth by Vagg and Harris (2000). Not only the magnitude of counterfeiting is increasing. The same goes for the scope of counterfeiting. Chaudry and Zimmerman (2008) state that the types of products being counterfeited are broadening. Not only CD’s, DVD’s, computer equipment, clothes and shoes are being counterfeited. Other product categories frequently being imitated are toys, pharmaceuticals, automobile component parts, electrical equipment, food and beverages, tobacco and personal care products. This is confirmed by the OECD (2008) which even finds a shift from luxury goods to common consumer goods. These results are also congruent with the findings of the Taxation and Customs Union from the EC. Gentry et al. (2006) even put it more extreme: “If one can attach some value to a consumer brand, one is likely to find counterfeit imitations of it, somewhere.” This view is supported by Lewis (2009) who ads the aspect of the illegal profit margin that has to be high enough before a product is attractive for being counterfeited. Despite the fact that other product categories are on the rise, the OECD (2008) and the EC (2008)
1
report that fashion items (i.e. clothing, jewellery, accessories and footwear) still account for the largest part of counterfeit trade, e.g. the textile sector and jewellery together make up 66,2% of all interventions by European Customs. Gessler (2009) states that counterfeiting is the major obstacle the luxury fashion industry is facing today. These astonishing numbers explain why the author opted for investigating non-deceptive counterfeiting (i.e. people are fully aware of the fact they are buying a counterfeit at the time of purchase) and consumer behaviour in the luxury fashion industry. The consequences of counterfeiting are enormous at various levels. Gessler (2009) divides ‘the true costs of counterfeiting’, i.e. the consequences of the phenomenon, in six categories: the cost to brand owners, government burdens, the effects counterfeiting has on consumers, child and forced labour issues in the production of these counterfeits, organised crime and terrorist funding activities of counterfeiters and the moral cost of counterfeiting. Later on in this paper the author will make a distinction on the basis of personal and societal harm counterfeit trade causes. One thing is clear: counterfeiting may no longer be seen as a victimless crime as it has a damaging effect on businesses, national economies, consumers and on society as a whole (UNICRI, 2009; Santos and Ribeiro, 2006). In the anti-counterfeiting literature many authors propose different strategies to counter product and brand counterfeiting. These anti-counterfeiting tactics range from the use of RFID tags (Tuyls and Batina, 2006) to the development of new legal frameworks (Bush et al., 2001). However, Berman (2008) states also companies can contribute to the reduction of the counterfeiting problem through the development of consumer education programs that publicize the personal and societal dangers counterfeiting causes. A part of this education process is that ‘marketers need to develop advertising campaigns that focus on the significant safety, performance and financial risks associated with the purchase of counterfeit merchandise’. To the author’s best knowledge, no such research has been conducted before that classifies and investigates the potential consequences of counterfeiting in such an extensive way. Furthermore, by integrating the construct of ‘societal harm’, this research responds to the critical remark Gessler (2009) made: ‘Unfortunately, the societal impact of counterfeiting remains largely under researched and often neglected’. In this research 2
we will assess the impact of consumers’ awareness of societal consequences on their attitudes toward purchasing CLFI. Altogether, the main interest of this research goes out to examining consumers’ attitude toward purchasing counterfeit luxury fashion items (CLFI) and their intention to purchase CLFI. More specifically, the author will be investigating if there exists a difference in attitude toward purchasing CLFI if one is being informed about the personal harm counterfeits cause rather than being informed about the societal consequences bound to counterfeit trade. In addition, several factors proven important in previous research (e.g. subjective norms, perceived behavioural control, etc.) will be reinvestigated in a Belgian context. The relationship between the price level, the quality level and perceived quality will be examined as well. Finally, the linkage between attitudes and intentions is assessed in a counterfeit-related context.
3
2. Defining counterfeit trade in luxury fashion brands The author uses the ‘classification scheme of counterfeiting and related terms’ created by Staake, Thiesse and Fleisch (2009). It provides deeper insight into the fuzziness and buzzwords associated with product counterfeiting. Figure 1 gives a graphical representation. Figure 1: Classification scheme of counterfeiting and related items (Staake et al., 2009)
First and foremost, it is important to stress the fact that we find ourselves in the area of illicit or illegal trade. Herein the distinction is made between the smuggling of arms (contrabands), illegal trafficking of drugs (controlled goods), the trade in stolen goods and the trade in goods that are infringing intellectual property rights (IPR). IPR are ‘rights granted to creators and owners of works that are the result of human intellectual creativity’ (JISC Legal, 2008). There are different types of intellectual property rights, the most important being copyrights, trademarks, industrial designs and patents. In short, copyrights mostly serve to protect films, music, and literary and artistic works. Trademarks are distinctive signs that identify certain goods as those produced by a specific person or enterprise (e.g. logos). Industrial designs serve to protect the aesthetic aspects of an article (e.g. the shape of a purse). Patents are limited in time 4
and geographically bound rights that enable the patent holder to exclude unauthorised parties from using the patented inventions (OECD, 2008; Interview with Miss Hagenaers, 16 February 2010, Antwerp). Whereas piracy is often linked with copyright infringement and counterfeiting with trademark infringement, the distinction between both concepts is often not clear. Staake et al. (2009) state that IPR infringements frequently overlap as companies often may protect their products by several IPR simultaneously. Therefore, counterfeiting and piracy are frequently used to describe the same phenomenon. Finally, Illicit parallel imports cannot be unequivocally classified as counterfeits since it concerns the illegal distribution of ‘third shift products’ or ‘overruns’. The latter are genuine products that are manufactured and sold without knowledge and permission of the legitimate IPR owner (Gessler, 2009). In this research we will be using the definition of counterfeit trade proposed by Staake et al. (2009): “Counterfeit trade is the trade in goods that, be it due to their design, trademark, logo, or company name, bear without authorization a reference to a brand, a manufacturer, or any organization that warrants for the quality or standard conformity of the goods in such a way that the counterfeit merchandise could, potentially, be confused with goods that rightfully use this reference”. Focussing on physical goods, in particular luxury fashion counterfeits, Staake et al. (2009) propose to make a distinction with regard to consumer’s perception. Consumers believe they may have bought a genuine article when in fact it is a counterfeit. Consequently they are not aware of the underlying IPR infringement. This is the case in deceptive counterfeiting (Vida, 2007; Staake et al., 2009). However, this research will focus on non-deceptive counterfeiting in which the consumer is fully aware that the product purchased is a counterfeit product at the time of purchase (Nia and Zaichkowsky, 2000; Grossman and Shapiro, 1988). Gentry et al. (2006) give a more practical definition: ‘in the case of non-deceptive counterfeiting, the buyer recognizes that the product is not authentic according to specific information cues such as price, purchase location or materials used.’ These goods are often also called ‘knockoffs’ (Berman, 2008). The author wishes to focus on the non-deceptive form of counterfeiting as this is acknowledged to be particularly prevalent in luxury brand markets (Nia and Zaichowsky, 2000). However, one must remain critical in classifying counterfeits only in deceptive and non-deceptive products. Different authors state there exists a continuum 5
of deceptiveness, rather than a dichotomy (Eisend and Schuchert-Güler, 2006; Lee and Yoo, 2009; Bosworth, 2006; Berman, 2008). This is particularly true in the present time frame, as the quality of knockoffs has improved dramatically recent years, even resulting in the existence of ‘supercopies’ (Prendergast et al., 2002; Wilcox et al., 2009; Hilton et al., 2004; Gessler, 2009).
6
3. Exploring the supply- and demand-side of counterfeiting 3.1. Supply side of counterfeiting Many authors (Chaudry and Zimmerman, 2008; Yoo and Lee, 2009; Gessler, 2009; OECD, 2008) suggest there are several supply-side factors that make it attractive for producers to take part in the illegal act of counterfeiting: • The potential of attaining very high profit margins: counterfeiters benefit from the R&D and marketing expenses invested by legitimate trademark owners. They are in fact free riding on the economic value associated with IPR ownership. • Low wages and almost no existing IPR enforcement in (developing) production countries (e.g. China, South-Africa). • Counterfeiters face far lower risks in terms of consequences than other illegal activities (e.g. fines and prosecution). Inadequate penalties form the basis of this phenomenon. • The global availability of low-cost high technological equipment creates the opportunity for counterfeiters to copy and produce nearly every product category imaginable. • The existence of free trade zones en free ports allows counterfeiters to engage in origin-laundering activities by means of which the true origin of these products can be obscured so that there exists no link towards the actual producers of these knock-offs. • The rise of the Internet is a major contributor toward the high availability rate of counterfeits. Counterfeiters can easily sell their products via e-mail solicitations (direct marketing) and it can also serve as a powerful marketing tool in reaching customers in a more disguised way. Counterfeiters now face unprecedented distribution opportunities. In his book “Knockoff: The Deadly Trade in Counterfeit Goods”, Tim Phillips (2005) puts it even more extreme: “Counterfeiting is thousands of years old, but conditions have never been better for it.”
7
The counterfeiting of luxury fashion brands is particularly attractive if one considers the inherent characteristics of the luxury industry. As brands are considered to be the most valuable assets these legitimate companies (e.g. LVMH Group) possess, counterfeiters hit them right in the heart by taking advantage of the consumer’s trust that has been established by legitimate owners who have been creating brand equity for many years and have thus been making considerable investments in gaining a prominent position in consumers’ minds (Green and Smith, 2002; Gessler, 2009). Gentry et al. even state the following: “While the purchase of a counterfeit represents the consumption of the brand (brand decision), it does not appear to represent a product decision”. As such, counterfeiters of luxury fashion brands are able to let consumers enjoy the snob appeal associated with the branded original without paying for it (Wee et al., 1995). On top of this, Hilton et al. (2004) state ‘production of the (counterfeit fashion) good and copying of designs are relatively easy (compared to counterfeiting other product categories)’.
3.2. Demand side of counterfeiting In the present study, we will focus on the demand side of counterfeiting. After all, it is basic economic reasoning that if no demand for counterfeit products exist, supply will erode automatically. Thus, as consumers play a leading and growing role in the existence of counterfeit trade (Yoo and Lee, 2009; Bian and Moutinho, 2008), it is important to gain a deeper insight in potential determinants of consumer’s willingness to purchase non-deceptive counterfeit products. This insight becomes even more important if we consider the fact that consumers indicate they have a clear picture of what they buy with the purchase of a counterfeit article. According to Penz and Stöttinger (2008), consumer’s mental maps of counterfeit goods and their original counterparts do not seem to overlap.
8
4. Determinants of purchase intention for counterfeit luxury fashion items: theories, constructs and hypothesis development Resulting from earlier research, there is no doubt that price plays a major role in the appeal of fake products (Ang et al., 2001; Albers-Miller, 1999; Bloch et al., 1993; Tom et al., 1998; Penz et al. 2009). In addition, Wee et al. (1995) were among the first researchers to investigate the impact of non-price determinants on intention to purchase counterfeit goods. These have been classified as psychographic (attitude toward counterfeiting, brand status and novelty seeking), demographic (age, educational attainment and income) and product-attribute (appearance, image, perceived fashion content, purpose and perceived quality) variables. In order to give the reader the opportunity to develop a more extensive understanding of factors influencing consumer complicity to buy counterfeits, the author would like to refer to the conceptual model developed by Chaudry and Stumpf (2007), published and discussed in “The Economics of Counterfeit Trade” by Chaudry and Zimmerman (2008). Figure 2 gives a graphical representation. A distinction is made between intrinsic and extrinsic determinants influencing consumer’s complicity to purchase counterfeit goods. The author will use this reasoning as a basis for variable classification and hypothesis development. Although cultural values and ethical perspectives are mentioned in the model, they will not be discussed in this paper. However, we will investigate various intrinsic determinants proven important in other research: subjective norm, perceived behavioural control, perceived risk, fashion consciousness, price quality inference and perceived quality.
9
Figure 2: Conceptual model of consumer complicity (Chaudry and Zimmerman, 2008; adapted from Chaudry and Stumpf, 2007)
4.1. Intrinsic Determinants 4.1.1. Attitude toward buying CLFI, and intention to purchase CLFI Many researchers suggest that attitudes toward behaviour are more accurate in predicting intentions to perform that behaviour than attitudes toward a product (Ajzen, 1991; Penz et al., 2005; Smith et al. 2008). Thus, in the context of counterfeiting, the attitude towards buying counterfeit luxury fashion brands can be used as a good predictor for purchase intention of CLFI (Chaudry and Zimmerman, 2008; Staake et al., 2009), which in turn can be a good predictor for the actual purchasing of CLFI. However, many researchers (Koklic and Vida, 2009; Santos and Ribeiro, 2006) state counterfeiting has to be examined in a country-specific way. Therefore we hypothesize: H1: There exists a positive relationship between the attitude towards buying CLFI and the intention to purchase CLFI.
10
4.1.2. Subjective norms Ajzen (1991) defines ‘subjective norm’ (SN) as “the perceived social pressure to perform or not to perform the behaviour in question”.
In turn, the SN construct is
determined by normative beliefs, which are defined by Armitage and Conner (2001) as: “underlying normative beliefs are concerned with the likelihood that specific individuals or groups (referents) with whom the individual is motivated to comply will approve or disapprove of the behaviour”. In short, SN serves to measure social influences. Social influences refer to the potential effect ‘significant others’ (e.g. family, friends, teachers, employers) have on consumer behaviour, in casu purchasing CLFI. There are two kinds of social influences, i.e. informational and normative social influences (Bearden et al., 1989; Ang et al. 2001). Informational social influences refer to the fact one’s decisions might be based on the expert opinion of others. In this case, the person is ‘informational susceptible’. Normative social influences refer to the fact one’s decisions might be based on expectations of what would impress others. Deutsch and Gerard (1955) describe normative social influence as the influence to conform to the expectations of another group or person. In this case, a person is ‘normative susceptible’. Ang et al. (2001) found that a person’s informational susceptibility was not a significant predictor of attitude towards counterfeiting. Therefore, the author will solely focus on normative social influences in this paper. It is particularly relevant to add this variable to our research about CLFI as Large (2009) stresses the importance of consumer fashion goods as a means of projecting social status within a social group. Ang et al. (2001) also found a negative relationship between normative susceptibility and the attitude one holds toward buying counterfeits. However, one might only expect this negative relationship to be true if there exists a norm among one’s reference group not to buy CLFI. Earlier research indicates there exists not only an impact of subjective norm on attitude towards purchasing counterfeits. Penz and Stöttinger (2005) report the SN construct has a significant effect on purchase intention. Therefore the author expects the disapproval of important others will have a direct negative impact on purchase intention for CLFI if a person is normative susceptible. Reasons for disapproval could be the belief that counterfeits are still of inferior quality, not being authentic, the potential loss of face (Gentry et al., 2006), etc. 11
Many researchers have investigated the influence of subjective norm on counterfeiting (Ang. et al., 2001; Penz and Stöttinger, 2005; De Matos et al., 2007; Penz et al., 2009). However, results are mixed depending on the context in which the research is undertaken. As already mentioned above, previous research suggests investigating the counterfeiting phenomenon in a country-specific and product category-specific way (Tom et al., 1998; Wee et al., 1995; Veloutsou and Bian, 2008). The current research is undertaken in a Belgian context and is solely focussing on CLFI. For this reason the SN construct is integrated in this research as well. Based on the discussion above we hypothesize the following: H2a: A person’s normative susceptibility will have a negative impact on one’s attitude toward purchasing CLFI.
H2b: A person’s normative susceptibility will have a negative impact on one’s purchase intention for CLFI.
4.1.3. Perceived behavioural control Ajzen (1991) defines ‘perceived behavioural control’ (PBC) as “the perceived ease or difficulty of performing the behaviour in question”. In turn, the PBC construct is determined by control beliefs which represent the underlying dimensions of PBC. Ajzen (1991) defines this dimension as: “control beliefs are a set of beliefs that deal with the presence or absence of requisite resources and opportunities”. Meng-Hsiang et al. (2006) define PBC as ‘perceived behavioural control reflects one’s perceptions of the availability of resources or opportunities necessary for performing a behaviour’. As such, ‘perceived availability’ can be seen as a part of PBC. The higher one’s perceived availability, the higher one’s perceived ease of acquisition and the higher one’s perceived behavioural control. Ajzen (1991) also suggests PBC is highly accurate in predicting ‘intentions to perform behaviours of different kinds’. In addition, Mannetti et al. (2002) found PBC is positively contributing to the prediction of purchase intentions. Counterfeits are becoming more and more widely available in different price and quality levels (OECD, 2008). As a consequence, the perceived ease of acquisition and thus
12
perceived behavioural control are expected to rise. Chaudry and Stumpf (2010) found ‘availability’ and the ‘ease of obtaining’ counterfeits is indeed a positive contributor toward consumer complicity in an Australian and U.S. context. Therefore we postulate the following: H3: The higher the perceived availability of CLFI is, the higher the purchase intention of consumers.
4.1.4. Perceived risk It is widely accepted that perceived risk is a key construct in examining consumer behaviour (Mitchell, 1999; Taylor, 1974). In the case of counterfeit purchase behaviour, Veloutsou and Bian (2008) suggest that the experienced level of overall perceived risk is moderate. Additionally, De Matos et al. (2007) suggest that perceived risk is the most important variable to predict consumer attitude toward counterfeits.
Bian and
Moutinhou (2009) found evidence that perceived risk is a factor that negatively influences the purchase intention of counterfeits. In conclusion the importance of the perceived risk construct is proven. Important to notice is that in earlier research on noncounterfeit related perceived risk, it is suggested that the concept of risk should be examined in a purchase-specific manner (Taylor, 1974). Despite the fact perceived risk is defined in many different ways over the past decades (Mitchell, 1999; Stone and Gronhaug, 1993), we will use the definition proposed by Cunningham (1967): “perceived risk is the amount that would be lost (i.e. that which is at stake) if the consequences of an act were not favourable, and the individual’s subjective feeling of certainty that the consequences will be unfavourable”. Previous research suggests a person’s overall perceived risk consists out of six different risk dimensions: financial, performance, psychological, physical, social and time risk (Stone and Gronhaug, 1993). Veloutsou and Bian (2008) identified the same six risk dimensions in the context of purchasing counterfeits. Below one can find short definitions of each risk, related to the counterfeiting context:
13
• Social risk can be defined as the negative outcome one attaches to ‘being caught by significant others’ when possessing or purchasing a counterfeit. • Time risk is seen as the time one might lose in the search for a counterfeit article. • Financial risk involves the potential loss of money when buying a counterfeit. • Physical risk is the possibility of bodily harm. • Performance risk refers to the chance of malfunctioning of the knockoff bought. • In explaining psychological risk the author prefers to refer to an equivalent, but more explicit definition given by Dholakia (2001, adopted from Perugini and Bagozzi, 1999): “psychological risk refers to the experience of anxiety or psychological discomfort arising from anticipated postbehavioral affective reactions such as worry and regret from purchasing and using the product”.
4.1.5. Perceived harm and actor-proximity Empirical and academic evidence show that counterfeiting is a harming and thus risky business (Pollinger, 2008 and Lewis, 2009). In our research, we would like to differentiate between two types of harm caused by the purchase of counterfeit goods: personal and societal harm. This classification is based on actor-proximity, i.e. ‘the proximity away from the perceived harm to oneself or to family, rather than to overall society’ (Thompson et al., 2005). With personal harm, the individual itself is affected by purchasing counterfeit products. With societal harm, counterfeit trade affects the society as a whole and thus actor-proximity is less. Related to the concept of actor-proximity, Casola et al. (2008) found that ‘respondents saw participation (in buying counterfeit goods) as less acceptable (…) when the victim was seen as an individual rather than as society or an organisation.’ Thompson et al. (2005) posit that ‘the average consumer does not consider overall societal issues when faced with the option of purchasing an original or a counterfeit product.’ Therefore the author expects perceived personal harm to have a greater negative impact on one’s attitude toward purchasing CLFI than perceived societal harm.
14
4.1.6. Perceived Personal harm Veloutsou et al. (2008) indicated that western consumers rated physical risk, performance risk and financial risk as the most important types of risks in the context of non-deceptive counterfeiting. As already mentioned above, they also found time risk, psychological risk and social risk contributed to the overall risk level. Informing consumers about the personal consequences these risks might cause can be used as negative cues to be sent to the potential consumers of counterfeit goods. As such, this information is expected to have an impact on perceived personal harm, as the individual itself might become affected. Important for this research is that Chakraborty et al. (1997) suggest that negative cues about counterfeiting lead to lower purchase intentions. To give some examples of the most frequently mentioned personal consequences that are linked to these types of risk in a CLFI related context, the author refers to Gessler (2009) and Phillips (2005): • CLFI can cause heavy skin rash (cfr. physical risk). • CLFI do not ‘function’ as promised, e.g. colours fade away very easily (cfr. performance risk). • There exists a high probability one pays too much for what the CLFI is actually worth, considering its objective quality (cfr. financial risk).
4.1.7. Perceived Societal harm In assessing societal harm the author would like to refer to Lewis et al. (2009), the OECD report (2008), the BASCAP reports (2009), Gessler (2009) and Pollinger (2008). The most frequently mentioned consequences counterfeiting has on society are: • Funding of international crime. Counterfeiting is even called ‘the crime of the 21st century’ as terrorists can get ‘easy’ money without risking a lot. The counterfeiting activity seems extremely attractive to terrorists compared to other illegal activities like drug smuggling and human trafficking.
15
• Job losses. Counterfeiting is held accountable for job losses at a large scale in legitimate companies and their subsidiaries. • Loss of sales. Legitimate owners suffer a direct loss of sales. • Loss of taxes. The government suffer significant losses of tax revenues due to unpaid sales tax, unpaid business taxes on production and sale of counterfeit goods and lost customs duties. • Decreased innovation. One argues that counterfeiting causes a disincentive for innovation by legitimate owners as they have to fund future R&D projects with fewer sales. This view is also supported by Miss Hagenaers (Trade Mark and Design Attorney, 16 Feb 2010, Antwerp). • Child and forced labour. The production of counterfeit articles is not in line with current labour legislations. Consequently, child and forced labour are rigorously present. Informing consumers about the societal consequences of counterfeiting can be used as negative cues to be sent to potential consumers of counterfeit goods. As Chakraborty et al. (1997) suggest, negative cues about counterfeiting lead to lower purchase intentions of counterfeit products. In addition, Penz et al. (2009) found that ‘informing consumers about the business practices of counterfeiters may change their attitudes toward the phenomenon’. Based on this literature review we postulate the following: H4: Messages provoking perceived personal harm will affect the attitude towards CLFI in a more negative way than messages provoking perceived societal harm as actorproximity is closer in the case of personal harm.
4.1.8. Fashion consciousness Fashion consciousness can be defined as the degree to which one finds it important to be perceived as a fashionable individual or the degree to which an individual keeps his ‘styling’ and thus variety of new fashion items up-to-date. As we are investigating consumers’ purchase intention for luxury fashion-related counterfeits, the results of Wee
16
et al. (1995) are of major importance to our research. They reveal that ‘for fashionrelated counterfeit products, purchase intention is determined by the similarities in appearance, quality and image projected by the counterfeit version compared to the originals.’ In addition, one must bear in mind that the quality and thus the ‘similaritypotential’ of counterfeits are skyrocketing (OECD, 2008; Gentry et al., 2006). Gentry et al. (2006) also found that consumers consider counterfeits as a relatively cheap way to keep up with the latest fashion trends. Taking into account the above results of previous research, one might expect it is not only the lower price that influences consumers’ purchase intention of fashion-related counterfeit items, but also their fashion consciousness, as these CLFI make luxury fashion available for a much broader public than does the original. Therefore we postulate the following: H5: Fashion consciousness has a positive impact on purchase intentions of CLFI.
4.1.9. Past behaviour In a recent investigation toward the role of past (buying) behaviour and its effect on future (buying) behaviour in a non-counterfeit related context, Smith et al. (2008) found self-reported past (buying) behaviour was a strong predictor of self-reported purchase intentions. Also Ouelette and Wood (1998) suggest past behaviour has a significant influence on intentions and therefore on actual behaviour. In addition they state that the frequency of performing certain behaviour has a direct impact on future behaviour. Ajzen (1991) already suggested the salient influence of past experience on perceived behavioural control and thus on purchase intentions. In the context of counterfeit buying behaviour, Yoo and Lee (2009) found past behaviour had a significant positive influence on predicting purchase intention of luxury fashion counterfeits in a South Korean context. However, the main limitation for their study is that the findings may have very limited ‘generalizability’. As they suggest themselves, ‘it would be more meaningful if the same findings hold consistent in different types of consumers, in different regions and in different cultures’. To help improving the generalizability potential of the ‘past behaviour’ construct in the area of consumer behaviour and counterfeiting, we will investigate the impact it has on intention to
17
purchase counterfeit luxury goods in an individualistic Western country (Belgium). Based on the above discussion, we hypothesize the following: H6: Past behaviour has a positive influence on purchase intention of CLFI.
4.1.10.
Price/quality inference
Price is often used as an extrinsic cue to signal product quality. ‘Extrinsic attributes are not considered to be product-specific and can therefore serve as general indicators of quality across all types of products’ (Zeithaml, 1988). Nevertheless, many authors have argued the accuracy of this reasoning. For example, Gerstner (1985) found that for many products, the relationship between price and quality is rather weak and thus price is found a poor predictor of quality. In their meta-analysis of the price/perceived quality relationship, Völckner and Hoffmann (2007) conclude that ‘the price effect on perceived quality has decreased’. However, Lichtenstein and Burton (1989) found that one has to consider the price/quality relationship in a product-specific way. For this reason we integrate the price/quality relationship in our research on counterfeiting. How do Belgian consumers evaluate the price/quality relationship of counterfeits? To which extent do consumers believe the price of a CLFI is a predictor for its quality and what is the impact of this belief on their attitudes toward purchasing CLFI? Phau et al. (2008) found price/quality inference to be a significant predictor for attitude toward counterfeiting. However, in practice one must remain critical in assessing this price quality relationship as Gentry et al. (2006) suggest ‘the seller’s willingness to negotiate price for a CLFI may be the real cue for quality, rather than the initial price itself’’. H7: One’s price/quality inference rating has an impact on one’s attitude toward purchasing CLFI.
4.1.11.
Perceived quality
Another determinant of major importance in examining consumer behaviour in the context of counterfeit goods is perceived quality. This construct can be defined as ‘a
18
consumer’s judgment about a product’s overall excellence and superiority’ (Zeithaml, 1988). Wee et al. (1995) state that consumers’ intention to purchase counterfeits is dominated by perceived quality. Prendergast et al. (2002) also prove the importance of perceived product quality when consumers buy CLFI. Although counterfeits are often seen as low quality copies of the real product, there is an upward trend towards high quality counterfeits (Hilton et al., 2004; OECD, 2008). ‘Supercopies’ are on the rise. Over the years, counterfeits have enjoyed increased quality levels due to the widely available, cheap and easy accessible new production technologies. (Gessler, 2009; Alcock, 2003). Gentry et al. (2006) even state that the ability to distinguish counterfeits from genuine items becomes less important as the quality of many counterfeits is more and more approximating the quality of the real product. In fact, consumers report that there might not be any noticeable difference in perceived quality at all (Tom et al., 1998). The following questions will be investigated in this context: “What is consumers’ perception of quality in case of CLFI? Does it influence their attitude toward purchasing counterfeits and therefore has an impact on their purchase intentions for such articles? “
4.2. Extrinsic Determinants
4.2.1. Price ‘Low price’ is frequently mentioned as one of the major product attributes affecting purchase intention of counterfeit goods (Penz et al., 2009; Large, 2009; Albers-Miller, 1999; Chaudry, 2008; Tom et al., 1998; Ang et al., 2001; Bloch et al., 1993; Large, 2009). The price of a counterfeit luxury fashion item is mostly set as a fraction of the price of the matching genuine item (Penz and Stöttinger, 2005; Ang et al., 2001; Tom et al., 1998). One can find practical examples of counterfeiters’ low-pricing practices just by googling the word ‘replica’. Several sites will be displayed, offering cheap fakes. It is this low-pricing strategy that attracts consumer’s attention on purchasing a counterfeit. However, because of the rise of the so-called ‘supercopies’, prices of these high quality counterfeits have risen too. It is interesting to investigate whether consumers use real 19
price and quality cues of CLFI as an indicator of perceived quality, as is sometimes the case with genuine items (Völckner and Hofmann, 2007). To the authors’ best knowledge, no research has yet been conducted that integrates the possibility of consumers encountering relatively high prices for counterfeits. Based on the discussion above on perceived quality and the price construct reasoning, the following hypotheses are postulated: H8: Messages containing a high price or a high quality cue have a positive impact on the perceived quality of a CLFI.
H9a: The higher the perceived quality of a CLFI, the more positive one’s attitude towards purchasing counterfeits.
H9b: The higher the perceived quality of a CLFI is, the higher one’s purchase intention for CLFI.
20
4.3. Conceptual model for intention to purchase CLFI Figure 3 gives a graphical representation of the research design. A dotted line implies a comparison is made between the connected constructs. A solid line evaluates the direct relationship between the connected constructs. Figure 3: Conceptual model for intention to purchase counterfeit luxury fashion items
21
5. Summary of the research focus It is many times suggested in the anti-counterfeiting literature that consumers must become aware of the negative consequences bound to counterfeit trade (Lewis, 2009; Thompson et al., 2005; Berman, 2008). We will investigate the impact of messages that attract consumers’ attention on these negative consequences, both on individual and societal level. In fact we will assess if there exists a difference in attitude one holds toward purchasing CLFI, depending on which type of message (personal harm or societal harm message) one is being exposed to. The relationship between attitude and purchase intention is investigated as well. We will consider the effect of price and quality messages on perceived quality because many authors (Gentry et al., 2006; Gessler, 2009) suggest ‘retail’ prices and actual quality of CLFI are rising. The author integrates the higher availability of counterfeits by assessing the impact of perceived behavioural control on purchase intention. Because we find ourselves in a fashion-related context, we assess the impact of one’s fashion consciousness on purchase intention. Subjective norms are often reported to be significant influencers of attitude and purchase intention. Therefore, one’s normative susceptibility is taken into account as well. Finally, we assess the impact of past behaviour on the intention to purchase CLFI.
22
6. Pre-test In our main research, each participant will see one message stimulus that contains information about a product being counterfeited, the price and the quality level of the product in question. Above all, they are either shown a personal harm or a societal harm message. Different messages were constructed to manipulate these variables correctly. A pre-test was set up to test one’s perception of the constructs being manipulated. The reader can find the questionnaire used for pre-testing in Annex 1.1. All statistical analyses concerning the pre-test results are conducted with SPSS® 15 and can be found in Annex 1.2. The pre-test has been conducted with seventeen people of different ages. Ten of them were women (58.82%), while seven of them were men (41.17%).
6.1. Products and brands Two of the most frequently counterfeited luxury fashion items are watches and handbags (Brandhome, 2008; OECD, 2008; BASCAP, 2009a). We investigated the importance men attach to a (genuine) watch and the importance women attach to a (genuine) purse on a 7-point Likert scale, ranging from ‘totally not important’ to ‘totally important’. Results show no significant difference in importance attached by men and women to these products (t(17)=1.385, p>0,05) (Mmen=5.57, SD=1,72; Mwomen=6.40, SD=0,70) Thus, a watch for a man is equally important as a purse for a woman. For this reason we can use these two products as product stimuli in our main research. In the message, brand names of luxury fashion brands were mentioned. To make sure we used the brands that are most likely to be purchased when counterfeited, respondents were asked to answer the question from which brand(s) they should ever consider buying a counterfeit version, if it were easily available at a (for them) acceptable price and quality level. Respondents were given a list with widely counterfeited brands, but they also had the opportunity to write some down that were not in this list. Results indicate that Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex and Calvin Klein are the most preferred brands. 23
6.2. Perceived personal harm To be able to manipulate perceived personal harm correctly in our main research, we confronted respondents with messages demonstrating the risks consumers take when purchasing counterfeits. To find out which risks were perceived as having the highest impact on perceived personal harm, participants were asked to indicate to which extent each of the six risk dimensions of Veloutsou and Bian (2008) (cfr. infra) were perceived capable of inflicting personal harm on a 7-point Likert scale ranging from ‘I don’t think this is worse for myself at all’ to ‘I think this is definitely worse for myself’. Results show a pretty high mean for each dimension, ranging from 3,88 for psychological risk to 6,12 for physical risk. Table 1 shows that, except for psychological risk, there is found no significant difference in the perceived personal harm men and women associate with each of the different risk dimensions separately. Thus, only for psychological risk, men and women indicate it is personally harming them differently. Table 1: Perceived personal harm per risk type & gender differences RISK TYPE
t(17)
Physical Performance Financial Social Time Psychological
1.248 0.885 0.515 0.656 0.603 2.546
p-value Mmen SDmen Mwomen SDwomen Moverall >0.05 >0.05 >0.05 >0.05 >0.05 <0.05
5.57 5.00 4.86 3.86 4.00 2.57
2.149 1.639 0.690 2.34 1.826 1.718
6.50 5.60 5.10 4.50 4.50 4.80
0.85 1.174 1.10 1.716 1.581 1.814
6.12 5.35 5.00 4.23 3.88 4.29
SDoverall 1.536 1.367 0.935 1.954 2.058 1.649
We selected the three risks that were perceived as having the most impact (i.e. having the highest average score) on perceived personal harm to use in the message stimuli of our main research: physical risk (M=6,12; SD=1,54), performance risk (M=5,35; SD=1,37) and financial risk (M=5,00; SD=0,94).
6.3. Perceived societal harm To be able to manipulate perceived societal harm correctly in our main research, we confronted respondents with messages demonstrating the consequences society undergoes because of the existence of counterfeit trade. To find out which consequences were perceived as having the highest impact on society, participants
24
were asked to indicate to which extent each of the six most frequently mentioned consequences proposed by Lewis et al. (2009), the OECD report (2008), the BASCAP reports (2009), Gessler (2009) and Pollinger (2008) is perceived capable of harming society as a whole on a 7-point Likert scale ranging from ‘I don’t think this is worse for society at all’ to ‘I think this is definitely worse for society’. Table 2 shows that, for each of the consequences separately, no significant difference is found in the perceived societal harm men and women associate with it. Table 2: Perceived societal harm per societal consequence & gender differences SOCIETAL CONSEQUENCE Funding of international crime Loss of sales for authentic company
t(17)
p-value
Mmen
SDmen
Mwomen
SDwomen
Moverall
SDoverall
1.552
>0.05
5.29
2.059
6.40
0.843
5.94
1.519
0.943
>0.05
4.43
1.718
5.20
1.619
4.88
1.654
Loss of government taxes
-0.667
>0.05
4.14
1.952
3.50
1.958
3.76
1.921
Job loss
1.419
>0.05
5.29
1.38
6.10
0.994
5.76
1.200
Disincentive for innovation
0.911
>0.05
4.71
1.113
5.30
1.418
5.06
1.298
Child & forced labour
2.045
>0.05
5.29
2.059
6.70
0.675
6.12
1.536
We selected the three consequences that were perceived as having the most impact (i.e. having the highest average score) on perceived societal harm to be used in the message stimuli of our main research: child and forced labour (M=6.12; SD=1.536), job losses (M=5.76; SD=1.200) and funding of international crime (M=5.94; SD=1.519).
6.4. Price, quality, perceived harm and message credibility Respondents in the pre-test were shown a list with six items about personal harm and six items about societal harm, numbered from 1 to 12. Next, they were asked to indicate on a 7-point Likert scale (ranging from ‘not at all credible’ to ‘very credible’) the overall credibility of all harm items at once, in combination with the information about a counterfeit product of either moderate or high quality and that is either priced low or moderate. Results show that all combinations are found to be credible as there is no mean score smaller than 5.29 (SD= 1.105). Table 3 shows that overall message credibility for each of the situations separately did not differ between men and women.
25
Table 3: Message credibility per price and quality situation & gender differences t(17)
p-value
Mmen
SDmen
Mwomen
SDwomen
Moverall
SDoverall
High quality, moderate price
-1.345
>0.05
5.71
0.488
5.00
1.33
5.29
1.105
High quality, low price
-0.191
>0.05
6.00
0.577
5.90
1.287
5.94
1.029
Moderate quality, moderate price
0.881
>0.05
5.29
1.496
5.80
0.919
5.59
1.176
Moderate quality, low price
0.875
>0.05
5.71
1.380
6.20
0.919
6.00
1.118
SITUATION
As mentioned above, all items were shown at once in one list. Because the researcher expected some items to be less credible than others in certain situations, participants were given the opportunity to write down the number(s) of the item(s) (ranging from 1 to 12) they found less credible in combination with the above-mentioned information of different price (low and moderate) and quality (moderate and high) levels. Many respondents indicated that especially messages about ‘physical risk’ (item no. 1 in the list), ‘performance risk’ (item no. 2) and ‘financial risk’ (item no. 3) could not be combined with the ‘high quality’ condition (see annex 1.2). Therefore, in our main research, we will not use stimuli that present moderate or high quality. Instead we will use ‘low’ and ‘moderate’ quality stimuli.
26
7. Main research The different stimuli used in the main research can be found in Annex 2. The full questionnaire can be found in Annex 3.1. All scales used can be found in ‘Marketing Scales Handbook: A Compilation of Multi-item Measures – Vol. 3’ (Bruner II, James and Hensel, 2001). All statistical analyses concerning the results of the main research are conducted with SPSS® 15 and can be found in Annex 3.2.
7.1. Methodology
7.1.1. Data collection A self-administered online questionnaire was created using the Qualtrics™ software. About four hundred people, both students and non-students, were emailed the website address and were kindly asked to take part in the survey. In addition, a group was created on the social networking site Facebook© where people could subscribe to take part in this research. It was stressed that participation was totally voluntary, free of obligations and totally anonymous. The researcher attached the possibility of winning a price as compensation for one’s participation. Demographic details were requested purely for statistical use. Participants who took part in the competition were also asked to leave their e-mail address. This was only used to inform the winners. Respondents were given about 3 weeks to complete the questionnaire. Reminders were sent via email and Facebook© statuses about the progress of the research were updated regularly. Four hundred and ten surveys were completed, but ninety-one were rejected due to incomplete information. In this way, 319 responses were used in our final analysis.
7.1.2. Sample Many studies investigating the determinants of counterfeit purchase intention collect data with students only. However, this research meets the proposition of Yoo and Lee (2008) to also include financially independent respondents of a greater age range. By 27
doing so, one may obtain a greater generalizability of the results. 319 Belgians (130 male and 189 female) ranging in age from 16 to 60 years (M=26.88; SD=11.46) filled out the online survey. The majority (72.8 per cent) of the respondents were between 16 and 25 years old. 36.1 per cent of respondents reported they have an own income at their disposal. Table 4 presents a detailed demographic profile of all respondents. Respondents were randomly assigned to one of the eight conditions of our 2 (perceived harm/message type: personal versus societal) by 2 (price: low versus moderate) by 2 (quality: low versus moderate) between-subjects experimental design. Table 4: Demographic profile of respondents Demographics
N
Per cent
GENDER Female Male
189 130
59.2 40.8
AGE 16-25 26-35 36 and above
230 23 63
72.1 7.2 19.7
INCOME Yes No
115 203
36.1 63.6
GROUP Student Working Jobless Retired Other
218 88 2 3 7
68.3 27.6 0.6 0.9 2.2
7.1.3. Stimuli The researcher briefly describes the construction of the eight different experimental stimuli. The first factor of the between-subjects design, personal harm versus societal harm (i.e. message type) concerns the fact whether the harm message indicates that society as a whole or that you yourself can be affected in a negative way by the purchase of counterfeits. Consequently, four messages stress the fact that there are negative societal consequences (cf. Pre-test) bound to counterfeit trade, while the other four messages stress certain risks that may personally harm (cf. Pre-test) purchasers of 28
CLFI. The second factor (i.e. price) of the message deals with the fact that one has the opportunity to buy a ‘low priced’ or a ‘moderate priced’ CLFI. The third factor (i.e. quality) of the communications deals with the fact that CLFI of top brands can be either of ‘low’ or ‘moderate quality’. The eight different scenarios can be found in Annex 2.
7.1.4. Measures Respondents indicate their purchase intention for a CLFI on a seven-point Likert scale based on the one developed by Sweeney, Soutar and Johnson (1999). The two items are: ‘Are you planning to buy a counterfeit luxury fashion item in the near future’, and ‘How big is the chance you will buy a counterfeit luxury fashion item in the near future’. The responses to these two items are averaged to form an overall purchase intention (Pearson correlation= 0.873). The researcher opted to take into account two variables that indicate differences among individuals concerning the impact of the messages they are exposed to. First, the respondent’s level of message involvement is measured by means of Cox and Cox’ (1991) six-item scale (Cronbach’s alpha= 0.80).
This high reliability allows us to
compute an average score. Second, message credibility is measured using a single item ‘The information shown above is credible’. The items of the two variables were measured using a 7-point Likert scale ranging from ‘I totally don’t agree’ to ‘I totally agree’. The attitude toward purchasing a CLFI is assessed using a seven-item (e.g. good-bad, rewarding-punishing, not harmful-harmful) seven-point semantic differential scale (Cronbach’s alpha= 0.872) based on the one developed by Ajzen and Fishbein (1980). In fact, these items allow the researcher to gain insights in what respondents think about the act of purchasing CLFI. Because of the high reliability, an average score can be calculated. Past purchase behaviour is operationalized using a single question asking whether the respondent has ever bought a counterfeit luxury fashion item before. Answers were
29
‘yes’ or ‘no’. In addition, the researcher asks to report how often per year one buys such items (M=0.463; SD=1.141). Respondents are asked to indicate their perceived quality level of a CLFI on a five-item (e.g. ‘The workmanship of this product would be’, ‘The likelihood that the product would be reliable is’) seven-point semantic differential scale (low-high) based on the one developed by Petroshius and Monroe (1987). Reliability was high (Cronbach’s alpha= 0.760) and thus an average score across the five items is calculated. One’s price-quality inference rating is assessed using a four-item (e.g. ‘The higher the price of the CLFI, the better the quality’) seven-point Likert scale ranging from ‘totally not agree’ to ‘totally agree’. Reliability was again high (Cronbach’s alpha= 0.806) and therefore an average score is computed. The scale used is adopted from Gotlieb and Sarel (1991). Subjective norm and thus, in the context of this research, normative susceptibility is measured using a four-item (e.g. My friends think I should not buy counterfeit luxury fashion items) seven-point Likert scale (Cronbach’s alpha= 0.794) ranging from ‘I totally don’t agree’ to ‘I totally agree’. The scale used is based on the ‘Normative Interpersonal Influence Scale’ developed by Bearden et al. (1989). The researcher used a slightly adapted version of Lumpkin and Darden’s (1982) scale to measure fashion consciousness. A four-item (e.g. I buy in different shops to obtain a great variety in my clothing) seven-point Likert scale (Cronbach’s alpha= 0.903) ranging from ‘I totally don’t agree’ to ‘I totally agree’ was developed. Item scores were averaged to form an overall fashion consciousness score. A single item asking how easy one finds it to buy a CLFI in the future six months, measures perceived behavioural control. Linked to the concept of availability, the respondent is asked to indicate where (flea markets, internet, on holidays, other) he or she would buy a CLFI if one would buy such thing at all.
30
7.2. Results
7.2.1. Manipulation check Respondents are randomly assigned to one of the two perceived harm conditions. Consequently, respondents in condition 1 are confronted with a ‘personal harm’ message, while respondents in condition 2 are confronted with a ‘societal harm’ message. Before analyzing the data, we need to make sure respondents correctly perceived these messages as causing harm to oneself in condition 1 or to society in condition 2. The researcher asks questions concerning the manipulation check at the end of the questionnaire (see Annex 3.1). In this way the least interference between the manipulation check and the actual manipulation is expected. Participants are asked in two separate questions to indicate on a seven-point Likert scale to which extent each of the risk types (physical, performance and financial risk) and each of the societal consequences (child and forced labour, job losses and funding of international crime) selected in the pre-test are capable of causing respectively personal harm (ranging from ‘I don’t think this is worse for me at all’ to ‘I definitely think this is worse for me’) and societal harm (ranging from ‘I don’t think this is worse for society as a whole’ to ‘I definitely think this is worse for society as a whole’). Before comparing mean scores, it is interesting to investigate whether the different items of the manipulation check questions assessing perceived personal harm indeed measure the same construct (in casu perceived personal harm). The same goes for the items intended to measure perceived societal harm. In order to check whether the three different risk types indeed measure perceived personal harm and the three different societal consequences indeed measure perceived societal harm, Cronbach’s alphas are calculated. Results in Table 5 suggest the different items of the two perceived harm constructs are internally consistent as Cronbach’s alphas are high. In both cases item-total correlation is acceptable because none of the items has a low correlation with the construct measured. Therefore the researcher concludes the different items mentioned are indeed capable of provoking personal and societal harm. Item scores are averaged for the both constructs separately in order to create for each respondent an overall score
31
for the manipulation check of perceived personal harm and an overall score for the manipulation check of perceived societal harm. The next question to answer is whether respondents also perceive it this way and to which extent they do so. Table 5: Cronbach's alpha analysis for perceived societal harm and perceived personal harm Perceived Societal Harm Cronbach's alpha= 0.790 ITEM Societal harm: terrorist funding Societal harm: job losses Societal harm: child labour
Item-total correlation Alpha if Item Deleted 0,615 0,738 0,664 0,688 0,634 0,724
Perceived Personal Harm Cronbach's Alpha= 0.760 ITEM Personal harm: physical risk Personal harm: performance risk Personal harm: financial risk
Item-total correlation Alpha if Item Deleted 0,537 0,737 0,667 0,588 0,571 0,700
First we split our file based on the message type (group 1: personal harm message; group 2: societal harm message) one has been exposed to. Next we conduct two OneSample T-tests to test whether the mean score in the manipulation check for perceived personal harm and the mean score in the manipulation check for perceived societal harm is significantly different from the scale’s middle point (i.e. 4 on a 7-point Likert scale). Results in Table 6 show that respondents who are exposed to a personal harm message indeed perceive the risks mentioned as personally harming them because the mean score is significantly different (in casu higher) from the scale’s middle point. In the same way, respondents who are exposed to a societal harm message indeed perceive the consequences mentioned as causing harm to society. Table 6: Manipulation check for perceived harm
T-value
N
Manipulation check Mean
SD
Personal Harm
14.286**
155
5.277
1.113
Societal Harm
31.202**
163
6.123
0.869
Message Type
Test Value= 4 **: significant at the 0,01 level (2-tailed)
32
7.2.2. Descriptive statistics Table 7 quantitatively summarizes the measured variables in our data set and thus gives an average idea of the investigated consumer behaviour concepts in the counterfeiting context. As mentioned above, all variables except past purchase behaviour are measured on 7-point Likert scales. Table 7: Descriptive statistics for our data set
Message Credibility Message Involvement Purchase Intention Attitude Perceived Quality Price Quality Inference Subjective Norm Fashion Consciousness Past Purchase Behaviour Perceived Behavioural Control Valid N (listwise)
N 319 319 319 319 319 318 318 318 319 315 315
Min 2 1 1 1 1 1 1 1 0 1
Max 7 7 7 5,86 5,75 6,25 7 7 1 7
Mean 5,49 4,7837 2,2382 3,3108 2,2978 3 4,0299 4,2948 0,4013 4,62
SD 1,0030 0,9379 1,5635 1,0574 0,9192 1,1202 1,1389 1,4632 0,4909 1,9780
Please note that across the entire sample respondents hold a somewhat negative attitude toward purchasing CLFI. The low dispersion (SD=1.0574) indicates respondents’ opinions are very much alike. Purchase intentions are low. However, one must notice a relatively high dispersion (SD=1.5635) in this case. Respondents perceive CLFI as low quality products. In addition, they are quite unanimous about this matter as dispersion is pretty low (SD=0.9192). Respondents’ price quality inference rating is low too. This means they do not see a higher (lower) price as an indicator for higher (lower) quality. Again they very much hold the same opinion about this matter (SD=1.202). Perceived availability and thus PBC of CLFI is fairly high. This means consumers find it relatively easy to acquire a CLFI. However, one must take into account the very high dispersion (SD=1.9780). In this context respondents indicate that if they would purchase a CLFI, they would do this in the first place when being on holiday. Concerning past purchase behaviour, 40.13% of respondents indicated to have purchased a CLFI before.
33
7.2.3. Message credibility and message involvement In order to identify if there exist individual differences in message credibility depending on which of the eight messages or conditions (see Annex 2) one has been exposed to, the researcher conducts an Analysis of variance test (see Annex 3.2). Results show us there is no significant effect of ‘conditions’ on message credibility (F=0.928; p>0.05). All messages are perceived as being equally credible. If we consider the high mean score for message credibility (M=5.49; SD=1.003), we can conclude respondents perceived these different messages as highly credible. In order to identify whether there exist individual differences in message involvement depending on which message type (personal harm or societal harm) one has been exposed to, the researcher conducts an independent samples T-test. Results show there indeed exists a significant difference (t(319)=-5.208; p<0.05) in message involvement between respondents who are exposed to a personal harm message (M=4.52; SD=0.91) and respondents who are exposed to a societal harm message (M=5.04; SD=0.89). The latter group is thus more involved.
7.2.4. Development of regression models for attitude toward purchasing CLFI and for intention to purchase CLFI Because there have been mixed results in past research concerning the impact of socio-demographics on the attitude toward buying counterfeits (Yoo and Lee, 2009), we chose to include these characteristics in our regression model as independent control variables. Therefore, the researcher opted to conduct hierarchical multiple regression analyses (University of Texas, 2009) for attitude toward purchasing CLFI and purchase intention for CLFI. Two separate models are developed. There are two stages in the development of each model. In a first stage the independent variables that we want to control for are entered into the regression. In the second stage, the independent variables whose relationship we want to examine after the controls are entered. A statistical test of the change in R² from the first stage is used to evaluate the importance of the variables entered in the second stage. For each model, the standardised beta coefficients are reported because they can be mutually compared.
34
It is important to notice that all underlying assumptions (Van Kenhove et al., 2008) to perform a linear regression analysis have been verified in every model. Regression model 1 for attitude toward purchasing CLFI satisfies all underlying assumptions (standardized residuals are distributed normally, tolerance indicates no multicollinearity problems, etc.). Regression model 2 for purchase intention of CLFI satisfies all conditions, except the normal distribution of standardized residuals. Thus, results for this model have to be interpreted with caution.
7.2.4.1. Regression model for attitude toward purchasing CLFI
In order to test hypotheses H2a, H4, H7 and H9a a multiple hierarchical regression is conducted to analyse the effects of the independent variables (message type, perceived quality, normative susceptibility and price quality inference) and the control variables (age, income, group, gender) on the dependent variable attitude towards purchasing counterfeits. Results generated are shown in Table 8. It is important to mention that ‘message type’ is a dummy variable, coded 0 and 1. Respondents assigned to the personal harm condition receive the value 0 for this variable. Respondents assigned to the societal harm condition receive the value 1 for message type. Table 8: Regression model of predictors for attitude towards purchasing CLFI Model 1a: Control Variables
Model 1b: All variables
Stand. Beta
T-value
Stand. Beta
T-value
Age Income Group Gender Message Type (Perceived Harm) Perceived Quality Price Quality Inference Subjective Norm
-0.118 -0.046 -0.190 -0.036
-1.363 -0.550 -2.419* -0.646
-0.104 -0.103 -0.117 -0.005 -0.181 0.322 0.075 -0.461
-1.598 -1.640 -1.979* -0.126 -4.367** -7.566** 1.768 -10.903**
R² Adjusted R² F-value
0.102 0.089 8.237**
0.508 0.494 37.031**
R² change F change
0.102 8.237**
0.406 59.231**
Dependent variable: Attitude *: significant at 0.05 level **: significant at 0.01 level
35
F-values for both models show there exists indeed a statistically significant relationship between the set of all independent variables and the dependent variable (attitude). Thus, we can conclude there exists a good fit between the data and the assumed regression model. F-change statistics for Model 1b indicate that the addition of the independent variables (message type, perceived quality, price quality inference and subjective norm) to the control variables indeed improves the relationship with the dependent variable (attitude toward purchasing CLFI) significantly. The adjusted R2 in our final Model 1b indicates that 49.4% of the variation in ‘attitude toward purchasing CLFI’ is explained by the variation in the four independent variables and the four control variables. The independent variables ‘perceived quality’, ‘subjective norm’ and ‘message type’, are found to be significant predictors of attitudes towards purchasing CLFI as their regression coefficients appear to be significantly different from zero. This is not the case for one’s price quality inference rating. One’s price quality inference rating is not found to be a significant predictor of attitude towards purchasing CLFI. Hence, Hypothesis 7 is rejected. Perceived quality is positively influencing the attitude toward purchasing CLFI. If one perceives the quality of a counterfeit being low (high), one’s attitude will be less (more) positive. Hence, Hypothesis 9a is supported. One’s normative susceptibility (subjective norm) is negatively influencing one’s attitude toward purchasing CLFI. The more susceptible one is for the importance significant others (e.g. friends, relatives) attach to not buying CLFI, the less positive one’s attitude toward purchasing CLFI. Therefore Hypothesis 2a is supported. Message type is negatively influencing one’s attitude toward purchasing CLFI. As such, the fact that one is exposed to a message containing information about the societal consequences associated with purchasing counterfeits instead of a message containing information about risks that may cause personal harm, has a negative influence on one’s attitude toward purchasing counterfeits. Hence, Hypothesis 4 is rejected. Table 9 36
confirms this result as there is found a significant difference in attitude toward purchasing CLFI between respondents confronted with a personal harm message (higher attitude score) and respondents confronted with a societal harm message (lower attitude score). Table 9: Attitude and message type differences
Attitude
T-value
N
M personal harm message
SDphm
M societal harm message
SDshm
2.465*
319
3.459
1.021
3.169
1.075
*: significant at 0.05 level
Regression results also show a significant effect of the control variable ‘group’. Thus, respondents’ attitudes towards purchasing CLFI differ depending on the group they belong to. However, further analysis is needed to examine this effect in depth as this ‘group’ variable is an ordinal one and can therefore not be interpreted properly in this regression. Analysis of variance confirms there is indeed a significant effect (F=8.630; p<0.05) of ‘group’ on attitude. Post hoc analysis using the Tamhane criterion for significance indicates that students (M=3.51; SD=0.93) hold a more positive attitude towards purchasing CLFI than respondents who reported to be working (M=2.93; SD=1.15).
7.2.4.2. Regression model for purchase intention of CLFI
In order to test hypotheses H1, H2b, H3, H5, H6 and H9b a multiple hierarchical regression is conducted to analyse the effects of the independent variables (attitude, normative susceptibility, perceived behavioural control, past behaviour and perceived quality) and the control variables (age, income, group, gender) on the dependent variable purchase intention for CLFI. Results generated are shown in Table 10. It is important to mention that ‘past purchase behaviour’ is a dummy variable, coded 0 and 1. Respondents indicating they have not purchased a CLFI before receive the value
37
0 for this variable. Respondents indicating they have purchased a CLFI before receive the value 1 for this variable. Table 10: Regression model for predictors of purchase intention for CLFI Model 2a: Control Variables
Model 2b: All variables
Stand. Beta
T-Value
Stand. Beta
T-value
Age Income Group Gender Fashion Consciousness Subjective Norm Past Purchase Behaviour Availability (PBC) Attitude Perceived Quality
-0.045 -0.115 -0.052 0.030
-0.492 -1.381 -0.644 0.513
0.021 -0.031 -0.036 0.004 0.121 -0.100 0.331 0.050 0.358 0.185
0.301 -0.494 -0.586 -0.093 2.525* -2.083* 7.241** 1,118 6.502** 3.842**
R² Adjusted R² F-value
0.041 0.028 3.103
0.486 0.468 26.873**
R² change F change
0.037 2.786*
0.449 41.380**
Dependent variable: Purchase Intention *: significant at 0.05 level **: significant at 0.01 level
The F-value for model 2b shows there is indeed a statistically significant relationship between the set of all independent variables and the dependent variable (attitude). Thus, we can again conclude there exists a good fit between the data and the assumed regression model. F-change statistics for Model 2b indicate that the addition of the independent variables (attitude, normative susceptibility, perceived behavioural control, fashion consciousness, past purchase behaviour and perceived quality) to the control variables indeed improves the relationship with the dependent variable (purchase intention for CLFI) significantly.
38
The adjusted R2 in our final Model 1b indicates that 46.8% of the variation in ‘purchase intention for CLFI’ is explained by the variation in the six independent variables and the four control variables. The independent variables ‘fashion consciousness’, ‘subjective norm’, ‘past purchase behaviour’, ‘attitude (towards purchasing CLFI)’ and ‘perceived quality’ are found to be significant predictors of purchase intention for CLFI as their regression coefficients appear to be significantly different from zero. This is neither the case for perceived behavioural control nor for any of the control variables incorporated in the model. Perceived behavioural control (availability) is not found to be a significant predictor of purchase intention for CLFI. Hence, Hypothesis 3 is rejected. Fashion consciousness is positively influencing one’s intention to purchase CLFI. The more one is concerned with e.g. being in fashion with current fashion styles, the higher one’s purchase intention for CLFI. Hence, Hypothesis 5 is supported. Normative susceptibility (subjective norm) is negatively influencing one’s intention to purchase CLFI. A similar reasoning as in predicting attitude can be adopted here: the more importance significant others attach to not buying CLFI, the lower one’s purchase intention for CLFI. As such, Hypothesis 2b is supported. Past purchase behaviour emerged to have a significant positive relationship toward purchasing intention for CLFI. Respondents who bought a CLFI before have a higher purchase intention than those who did not. Hence, Hypothesis 6 is supported. Attitude toward purchasing CLFI has a positive impact on purchase intention for CLFI. Respondents who indicate to hold a more positive attitude toward purchasing CLFI have higher purchase intentions. This result is in support of Hypothesis 1. Perceived quality is positively influencing purchase intention for CLFI. The greater the extent to which one perceives CLFI to have a great workmanship, to be sustainable or to be reliable, the higher one’s purchase intention for CLFI. Hence, Hypothesis 9b is supported. 39
7.2.4.3. Mediation
The author uses the four Baron and Kenny (1986) steps to assess the role of the ‘attitude’ variable as a mediation variable in the relationship between its determinants and the variable ‘purchase intention’. We will illustrate the procedure for ‘perceived quality’. Perceived quality is the independent variable (X) in this case. Attitude is the mediating variable (M). Purchase intention is the dependent variable (Y). In the first step we use purchase intention as a dependent variable in the regression equation and perceived quality as the independent variable. The model is significant (F=49.516; p<0,05). The same goes for the beta coefficient of perceived quality (t(318)=7.037; p<0.05). Thus, step one is passed. In the second step we use attitude as a dependent variable in the regression equation and perceived quality as the independent variable. The model is significant (F=58.260; p<0.05). The same goes for the beta coefficient of perceived quality (t(318)=7.633; p<0.05). Thus, step two is passed. In the third step we use purchase intention as dependent variable in our multiple regression. Perceived quality and attitude are used as independent variables. The model is significant (F=65.932; p<0.05). The same goes for the beta coefficient of attitude (t(318)=8.447; p<0.05). Thus, step three is passed. In step four we consider the beta coefficient for perceived quality. In this case, it is found to be significant as well (t(318)=3.821; p<0.05). Because of the fact both beta coefficients are significant in our third regression equation, attitude is classified as a partial mediator. Controlling for attitude, the effect of ‘perceived quality’ on ‘purchase intention’ is thus reduced, but is still significant from zero. The same steps are conducted for the ‘subjective norm’ variable. However, in the third equation only the beta coefficient for attitude is found to be significant (t(318)= 8.865; p<0.05). ‘Subjective norm’ is not found to be a significant predictor (t(318)=-1.790; p>0.05). Therefore, in this case attitude is classified as a complete mediator. Controlling for attitude, ‘subjective norm’ no longer affects ‘purchase intention’. ‘Price quality inference’ and ‘message type’ do not even pass step one. Therefore no mediation effect of attitude is present.
40
7.2.5. The influence of price and quality messages on perceived quality To investigate the effect of our price and quality manipulations on perceived quality, two independent samples T-tests were conducted. Results are shown in Table 11. It is important to mention that we made use of dummy variables in this context. As such, respondents can be divided in two groups: group 0 and group 1. ‘Price message’ is a dummy variable in which ‘0’ stands for the ‘low price’ condition and ‘1’ stands for the ‘moderate price’ condition. ‘Quality message’ is a dummy variable in which ‘0’ stands for the ‘low quality’ condition and ‘1’ stands for the ‘moderate quality’ condition. Table 11: Perceived quality and price & quality condition differences Mean Mean low moderate T-value p-value price SD lpc SD mpc price condition condition Perceived Quality
-0.916
>0.05
2.248
0.87
2.343
0.96
Mean Mean low moderate T-value p-value quality SD lqc SD mqc quality condition condition Perceived Quality
-0.520
>0.05
2.270
0.902
2.323
0.936
There is found no significant difference in perceived quality between respondents who were exposed to a ‘low price’ message and those exposed to a ‘’moderate price’ message. The same goes for the quality messages: no significant difference is found between respondents exposed to a ‘low quality’ message and those exposed to a ‘moderate quality’ message. Hence, Hypothesis 8 is rejected.
41
7.2.6. Summary Research results are summarised in Table 12. Table 12: Summary of research results Hypothesis H1 H2a H2b H3 H4 H5 H6 H7 H8 H9a H9b
Dependent Variable Attitude Purchase Intention Normative Susceptibility Attitude Normative Susceptibility Purchase Intention PBC Purchase Intention Message Type Attitude Fashion Consciousness Purchase Intention Past Behaviour Purchase Intention Price Quality Inference Attitude Price and quality message Perceived Quality Perceived Quality Attitude Perceived Quality Purchase Intention Independent Variable
Result Supported Supported Supported Not Supported Not Supported Supported Supported Not Supported Not Supported Supported Supported
42
8. Special topic: explorative questions about counterfeiting and its harming effects on society, businesses and individuals. In exploring consumers’ insights and opinions concerning the topic of counterfeit trade in a Belgian context, the researcher opted to include some explorative questions in the questionnaire. Answers were given on a 7-point Likert scale ranging from ‘I totally don’t agree’ to ‘I totally agree’. Statistical analyses can be found in Annex 4. Is the selling of counterfeit luxury fashion items an illegal business practice? Consumers seem to be well aware of the illegal character of selling counterfeit goods. Over 318 respondents, a mean score of 6.06 (SD=1.448) is obtained. There are no significant differences between men and women (t(318)=1.168; p>0.05) (Mmen=6.18, SD=1.355; Mwomen=5.98, SD=1.507). In addition there are no differences depending on the group (cf. demographic profile) respondents belong to (F=1.288; p>0.05). Do you think the selling of counterfeit luxury fashion items should be penalized? Again a high mean score is obtained (M=5.65; SD=1.392) indicating that consumers tend to agree there should be some kind of penalty bound to selling CLFI. No differences (t(318)=1.890; p>0.05) (Mmen=5.83, SD=1.415; Mwomen=5.53, SD=1.366) exist between the opinion men and women hold toward this issue. However, an ANOVA test shows there are significant differences (F=5.098; p<0.05) depending on the group respondents belong to. Post hoc analysis using the Tamhane criterion for significance indicates that students (M=5.44; SD=1.384) hold a milder opinion about penalizing sellers than respondents who reported to be working (M=6.02; SD=1.356). Do you think conscious buying of counterfeit luxury fashion items should be penalized? The general opinion is quite neutral as the mean score for this question was 3.90 (SD= 1.873). No significant difference (t(318)=1.066; p>0.05) (Mmen=4.03, SD=1.996; Mwomen=3.80, SD=1.782) is found in the opinion men and women hold toward this question. Again, there are significant (F=4.522; p<0.05) differences between students and respondents who reported to be working. Post hoc analysis using the Tamhane
43
criterion for significance indicates working respondents (M=4.45; SD=2.11) indicate more than students (M=3.61; SD=1.72) that conscious buying of CLFI should be penalized. Do you think the average consumer is sufficiently aware of personal and societal consequences of buying counterfeit luxury fashion items? Respondents indicate that they are not sufficiently aware of these consequences, as the mean score is rather low (M=2.53; SD=1.61). Men and women share the same opinion on this matter (t(318)=1.015; p>0.05) (Mmen=2.64, SD=1.60; Mwomen=2.45, SD=1.62). Above all, it does not matter to which demographic group one belongs (F=1.952; p>0.05), because there are no differences between their opinions about the awareness level concerning the consequences bound to purchasing CLFI. Do you think people would buy less counterfeit luxury fashion items if they were better aware of the personal and societal consequences bound to purchasing CLFI? A mean score of 4.85 (SD=1.634) shows that respondents slightly indicate awareness building could influence buying behaviour. Men and women again share the same opinion on this matter (t(318)= -1.707; p>0.05) (Mmen=4.66, SD=1.692; Mwomen=4.98, SD=1.585). No group differences are found (F=2.924; p>0.05). How high should be the chance of arrest incentivizing respondents not to buy counterfeit luxury fashion items anymore? Figure 4 gives a graphical overview of the frequency of the replies. Figure 4: Chance of arrest incentivizing respondents not to buy CLFI anymore
44
9. Discussion and implications The linkage between attitudes and purchase intention is reconfirmed again, but this time in a counterfeit-related (fashion) context. Belgian consumers that hold a (un)favourable attitude toward purchasing CLFI will also have (weaker) stronger purchase intentions for CLFI. In addition, attitude is found to be the best predictor for purchase intention of all variables investigated. Important to mention is the fact that Penz and Stöttinger (2005) confirm intentions to purchase CLFI are good predictors for actual purchase behaviour. Normative susceptibility (i.e. the extent to which one is susceptible to normative social influences, cf. infra) is a factor of major importance in predicting attitudes toward purchasing CLFI and in a smaller degree in predicting purchase intentions for CLFI. In fact, the more consumers perceive a normative pressure from significant others on their attitude towards purchasing CLFI, the more negative their actual attitude will be. The same reasoning can be applied for purchase intention. Taking into account the low overall attitude score, the author puts forward two possible explanations for the importance of the normative susceptibility construct. First, engaging in counterfeit trade is an illicit practice and can thus be seen as a criminal activity or consumer misbehaviour (Penz and Stöttinger, 2008). In this context, Tyler (2006) states “people are reluctant to commit criminal acts for which their family and friends would sanction them”. Studies (Tonglet, 2001; Albers-Miller, 1999) about other forms of consumer misbehaviour (e.g. shoplifting) indicate family and friends indeed play an important role in determining one’s attitude toward performing the illegal behaviour. So, if there exists a norm in one’s social group (group norms are defined as ‘regularities in attitudes and behavior that characterize a social group and differentiate it from other social groups’, Hogg and Reid, 2006) not to take part in any illegal activity and one’s normative susceptibility for this is high, attitude toward purchasing CLFI will be negative and purchase intention for CLFI will be low. Second, in the context of fashion consumption O’Cass and Frost (2002) state normative social influence is particularly important as ‘it taps impression creation, approval and achieving a sense of belonging.’ In addition they suggest ‘status products may be used for image portrayal to provide entry into certain groups or to fit into different situations’. Yoo and Lee (2009) show that counterfeit luxury fashion items do not succeed in fulfilling this status role of genuine luxury items. So, if there exists a norm in one’s social group not to buy CLFI e.g. because they are not able 45
to project the same status as genuine items, and one’s normative susceptibility for this is high, attitude will be negative and purchase intention will be low. Business practisers and governments aiming at reducing counterfeit trade should take into account the important role of normative social influences on attitude and purchase intention. In the example given above anti-counterfeiting communications should stress the importance significant others attach to buying genuine items and this in different buying situations, e.g. being on holiday. Considering the nature of the products (i.e. fashion items) used in this research, it is not surprising that one’s fashion consciousness has a (minor) influence on purchase intention for CLFI. The more respondents are keeping their ‘styling’ and variety of new fashion items up-to-date, the higher their purchase intention for CLFI. However, this might have to do with the fact that CLFI are often seen as risk-free (in terms of a small monetary loss) trial versions of the real product (Yoo and Lee, 2009). Consumers may in this way first want to find out if the luxury fashion item is indeed fashionable enough to spend a lot of money on. In this context it can be useful to investigate possible interaction effects between one’s fashion consciousness and his or her normative susceptibility. There might exist a norm in one’s social group not to buy CLFI. But will this norm hold in all situations? Is the buying of CLFI seen as something negative if it serves as a product trial? If not, this might be an indication for genuine item manufacturers to introduce some sort of cheap product trials themselves. In this way they could counter counterfeiting not only by investing loads of precious cash flow in anti-counterfeiting campaigns, but also by creaming-off counterfeiters’ ‘customers’. Informing consumers about the societal consequences linked to counterfeit trade is affecting attitude toward purchasing CLFI in a more negative way than informing them about the personal harm (in terms of risk) counterfeits possibly entail. This conclusion is not in line with our expectations. As such, in this context the theory based on actorproximity does not hold. The author gives several possible explanations for this. First, it could be due to the product category investigated (i.e. fashion items) that respondents see more severe harm in societal consequences than in personal consequences. After all, Large (2009) suggests CLFI cannot be classified as safety-critical counterfeits. Second, respondents reported to be more involved with the information received when exposed to a societal harm message rather than to a personal harm message. This 46
higher involvement might affect one’s message processing intensity and can therefore have a greater and more enduring impact on attitude (Maheswaran and Meyer-Levy, 1990). Third, it could be due to the fact consumers think to be fully aware of the personal risks bound to purchasing CLFI, but are astonished when informed about the societal consequences the buying of counterfeits entails. For example, consumers buying a counterfeit watch may accept or expect the fact that it will not be as durable (cf. performance risk) as a genuine item because of the lower price they pay for it. However, they might not be aware of the much broader consequences bound to counterfeit trade such as terrorist funding activities, job losses in authentic manufacturing companies and their subsidiaries etc. Educating consumers about the illicit business practices of counterfeiters and stressing the enormous societal impact the buying of counterfeits has can be used in demand reducing campaigns (BASCAP, 2009). As such, this research is complementary to the one conducted by Chakraborty et al (1997). Their findings indicated that sending negative cues about the personal harm (cf. risk dimensions) counterfeits cause, indeed affects purchase intentions to knowingly buy counterfeits. Our research indicates that informing potential customers about societal harm even has a larger impact on attitudes toward purchasing counterfeits than informing them about personal harm. Taking into account on the one hand that respondents in our sample indicate not to be sufficiently aware of personal and societal consequences bound to counterfeiting and that they on the other hand indicate awareness-building concerning these harms could influence consumers’ purchase behaviour, this academic research can serve as the basis for the development of an awareness-building campaign aimed at reducing the purchasing of non-deceptive counterfeits. Considering the great impact normative social influences have on attitude and purchase intentions, these awareness campaigns could serve a dual role. First, making consumers aware of the different types of consequences associated with counterfeit trade could influence their attitude and purchase intention in a direct way. Second, developing greater social awareness might create ‘new’ norms in social groups and can therefore have an effect on attitude through one’s normative susceptibility. Despite the fact that consumers indeed perceive counterfeits to be easily available, it does not influence their purchase intention for CLFI. This can be seen as good news for genuine item manufacturers. This is especially true if one combines our result with the finding of Nia and Zaichkowsky (2000) that ‘the majority of respondents disagreed the 47
availability of counterfeits negatively affects their purchase intentions for original brands’. Consequently, this reasoning indicates one must remain critical in assessing the intensity of the direct impact counterfeits have on sales of original brand manufacturers. This might explain why the ‘societal harm’-argument of counterfeits stealing sales from original brands was perceived as being the second less severe societal consequence in our pre-test. A study conducted by the company Brand home (2008) revealed that fifty percent of the Flemish people who took part in their survey, has already bought a counterfeit. Figures in our sample approximate this number as forty percent of all respondents indicated to have purchased a counterfeit item before. These results become increasingly important if one considers past behaviour is the second largest influencer of purchase intentions for CLFI. If one reported to have bought a CLFI, his or her purchase intention was significantly higher compared to respondents who did not. This finding is congruent with the research conducted by Yoo and Lee (2009). Therefore, our research increases the generalizability of the influence past behaviour has on purchase intention for counterfeits. This can be an indication for anti-counterfeiting campaigns to target in first instance those who have purchased a counterfeit before. Answers to the explorative questions about the sanctioning of counterfeit trade provide support for the research conducted by Cordel et al. (1996) and Ang et al. (2001). Also Belgian consumers obviously hold a double standard (i.e. ‘a situation in which consumers do not hold themselves to the same principles as their counterparty in the transaction’; Cordell et al. (1996)) concerning the illegal character of counterfeit trade. They agreed on the fact that the selling of counterfeits is illegal and should be penalized. On the contrary, they hold a much more neutral opinion toward penalizing conscious buyers of counterfeits. This might be due to the absence of clear-cut, crossborder international criminal penalties for the purchasing of non-deceptive counterfeits (Yoo and Lee, 2009). Confronting respondents with messages containing different price (low and moderate) and quality (low and moderate) levels did not have an impact on their perceived quality of CLFI. The author proposes two explanations. First, it could be that our manipulation of price and quality did not succeed. After all, respondents were neither presented real 48
objects nor concrete prices. Second, we can compare this finding with one’s low overall price quality inference rating. Respondents do not seem to use price as an indicator of quality in the case of CLFI. In this view our manipulation of price did succeed.
49
10. Research limitations and recommendations This research investigated the impact of message type on one’s attitude toward purchasing CLFI. There were only two conditions: respondents were shown a personal harm message or a societal harm message. We did not integrate a control group who did not receive any message at all. For this reason it is impossible to compare the absolute impact of these messages on one’s attitude. We could only make a comparison between the impact of personal harm cueing versus the impact of societal harm cueing on attitude. Therefore the author suggests further research introducing a control group. Considering the fact consumers are obviously not indifferent toward informing them about the societal consequences counterfeit trade causes, further academic research is advised in this area. First, one has to investigate the generalizability of these results on a European or even a worldwide level. The findings could provide the academic basis for integrating societal harm messages in awareness-building campaigns that aim to reduce non-deceptive counterfeit purchase behaviour. Second, it should be investigated if the messages aimed at informing consumers about the different harms associated with counterfeit trade, would have to be formulated as gain-framed or loss-framed messages. This idea is based on the finding that ‘consumers respond differently to product risk depending both on the nature of this risk and the framing of the persuasive message’ (Cox et al., 2006). Studies investigating attitudes toward purchasing CLFI should take into account the difference in attitude students and working class respondents hold towards the phenomenon. The use of students only in a research sample may cause limited generalizability of results. Many authors (Tom et al., 1998; Wee et al., 1995) suggest that the counterfeiting phenomenon needs to be examined on an industry and product category-specific basis. For example, there seems a significant difference in the non-price determinants of counterfeit purchase intention for functional products and fashion-related items. Above all, Large (2009) suggests fashion counterfeiting should be investigated separately from 50
other product categories as it concerns non-safety critical counterfeits. Therefore, determinants of attitude and purchase intention that are investigated for CLFI may not be generalizable to other product categories. Future research could assess the validity of the determinants proven important in this context for other product categories as well. Research has also found various cultural and country-specific differences in the volitional purchase of counterfeit luxury goods (Koklic and Vida, 2009; Santos and Ribeiro, 2006; Gentry et al., 2006; Veloutsou and Bian, 2008). As this research has only been conducted with Belgians, further research is needed in other countries and cultures to gain greater generalizability of the results mentioned in this investigation. Hilton et al. (2004) and Chaudry and Zimmerman (2008) suggest that also the ethical perspective toward IPR and counterfeiting is a possible contributor for purchase intention of fashion counterfeits. In this context, Maldonado and Hume (2005) indicate that ‘consumer ethics’ and ‘ethical relativism’ play an important role in the evaluation of buying counterfeits. As there were no ethical considerations taken into account in this research it might be useful to integrate these in future explorations of the purchase intention for CLFI.
51
REFERENCES Ajzen, I. (1991), “The Theory of Planned Behavior,” Organizational Behavior and Human Decision Processes, Vol. 50, No. 2, 179-211 Ajzen, I. and Martin Fishbein (1980), Understanding Attitudes and Predicting Social Behavior. Englewood Cliffs, NJ: Prentice-Hall, Inc. Albers-Miller, N. D. (1999), “Consumer Misbehavior: Why People Buy Illicit Goods,” Journal of Consumer Marketing, Vol. 16, No. 3, 273-287 Alcock, J., Chen, P., Chung H. M. and Hodsen S. (2003), “Counterfeiting: Tricks and Trends,” Journal of Brand Management, Vol. 11, No. 2, 133-136 Ang, S.W., Peng Sim Cheng, Elison A. C. Lim and Siok Kuan Tambyah (2001), “Spot The Difference: Consumer Responses Towards Counterfeits,” Journal of Consumer Marketing, Vol. 18, No. 3, 219-235 Aqueveque C. (2006), “Extrinsic Cues and Perceived Risk: the Influence of Consumption Situation,” Journal of Consumer Marketing, Vol. 23, No. 5, 237-247 Armitage, C. J. and Mark Conner (2001), “Efficacy of The Theory of Planned Behaviour: A Meta-Analytic Review,” British Journal of Social Psychology, Vol. 40, 471-499 Baron, R. M. and David A. Kenny (1986), “The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategical and statistical considerations,” Journal of Personality and Social Psychology, Vol. 51, 1173-1182 BASCAP (2009a), “Research Report on Consumer Attitudes and Perceptions on Counterfeiting and Piracy,” URL: http://www.iccwbo.org/bascap (21/02/2010) BASCAP (2009b), The Impact of Counterfeiting on Governments and Consumers, Frontier Economics Ltd: London, URL: http://osiris.iccwbo.org/uploadedFiles/BASCAP/Pages/OECD-FullReport.pdf (16/11/2009) Bearden, W. O., Richard G. Netemeyer and Jesse E. Teel (1989), “Measurement of Consumer Susceptibility to Interpersonal Influence,” The Journal of Consumer Research, Vol. 15, No. 4, 473-481 Berman, B. (2008), “Strategies to Detect and Reduce Counterfeiting Activity,” Business Horizons, Vol. 51, 191-199 Bian X. and Luiz Moutinho (2009), “An investigation of determinants of counterfeit purchase consideration,” Journal of Business Research, Vol. 62, 368-378 Bloch, P. H., Bush, R. F. and Campbell, L. (1993), “Consumer accomplices’ in product counterfeiting: a demand-side investigation,” Journal of Consumer Marketing, Vol. 10, No. 4, 27-36 Bosworth, D. (2006), “Counterfeiting and Piracy: The State of The Art,” Working paper, Oxford
52
Brandhome (2008), “1 op 2 Vlamingen Koopt Fake: Resultaten FAKE Onderzoek Verrassend,” URL: http://www.brandhome.com/files/PERS/Uitkomsten_FAKE_onderzoek.pdf (19/02/2010) Bruner II G. C., Karen E. James and Paul J. Hensel (2001), “Marketing Scales Handbook: A Compilation of Multi-item Measures – Vol. 3,” American Marketing Association, Chicago, IL Bullock, K., Rezina Chowdhury and Polly Hollings (2009), “Research Report 16: Public Concerns About Organised Crime”, URL: http://www.homeoffice.gov.uk/rds/pdfs09/horr16c.pdf (18/01/2010) Bush R. F., Peter H. Bloch and Scott Dawson (1989), “Remedies for Product Counterfeiting,” Business Horizons, Vol. 32, No. 1, 59-65 BusinessWeek (2007), “Faking Out The Fakers,” URL: http://www.fakesareneverinfashion.com/content/pdf/fakes_arresting_businessweek.pdf (15/04/2010) Caprara, G. Vittorio, Claudio Barbaranelli and Gianluigi Guido (2001), “Brand Personality: How to Make the Metaphor Fit?,” Journal of Economic Psychology, Vol. 22, No. 3, 377-395 Casola, L., Simon Kemp and Alexander Mackenzie (2008), “Consumer Decisions in the Black Market for Stolen or Counterfeit Goods,” Journal of Economic Psychology, Vol. 30, No. 2, 162-171 Chakraborty, G., Anthony Allred, Ajay Singh Sukhdial and Terry Bristol (1997), “Use of Negative Cues to Reduce Demand for Counterfeit Products,” Advances in Consumer Research, Vol. 24, 345-349 Chaudry, P. and Alan Zimmerman (2008) The Economics of Counterfeit Trade: Governments, Consumers, Pirates and Intellectual Property Rights. Berlin: Springer. Chaudry, P. and Stephen A. Stumpf (2010), “Country Matters: Executives Weigh in on The Causes and Counter Measures of Counterfeit Trade,” Business Horizons, Vol. 53, No. 3, 305-314 Cho S. and Jessie Chen-Yu (2009), “Effects of Experiences and Brand-Self Image Congruity on Perceived Risk and Purchase Intention in Apparel Online Shopping Context,” International Textile and Apparel Association Inc. – ITAA Proceedings No. 66 Commuri, S. (2009), “The Impact of Counterfeiting on Genuine-Item Consumers’ Brand Relationships,” Journal of Marketing, Vol. 73, No. 3, 86-98 Cordell, V. V., Nittaya Wongtada and Robert L. Kieschnick Jr. (1996), “Counterfeit Purchase Intentions: The Role of Lawfulness Attitudes and Product Traits as Determinants,” Journal of Business Research, Vol. 35, No. 1, 41-53 Cox, A.D., Dena Cox and Gregory Zimet (2006), “Understanding Consumer Responses to Product Risk Information,” Journal of Marketing, Vol. 70, No. 1, 79-91 Cox, D. and Anthony D. Cox (1991), “Communicating the Consequences of Early Detection: The Role of Evidence and Framing,” Journal of Marketing, Vol. 65, July, 91-103
53
Cronley, Maria L., Steven S. Posavac, Tracy Meyer, Frank R. Kardes and James J. Kellaris (2005), “A Selective Hypothesis Testing Perspective on Price-Quality Inference and Inference-Based Choice,” Journal of Consumer Psychology, Vol. 15, No. 2, 159-169 Cunningham, S. M. (1967), “The Major Dimensions of Perceived Risk,” in Cox, D. F. (Ed.), Risk Taking and Information Handling in Consumer Behavior, Graduate School of Business Administration, Harvard University Press, Boston, MA, 82-108 David A. Kenny (2009), “Mediation,” URL: http://davidakenny.net/cm/mediate.htm#WIM (15/05/2010) De Bock, T., Iris Vermeir, Mario Pandelaere and Patrick Van Kenhove (2010), “Exploring The Impact of Fear Appeals on The Prevention of Shoplifting,” 1-21 De Matos, C. A., Cristiana Trindade IItuassu and Carlos Alberto Vargas Rossi (2007), “Consumer Attitudes Toward Counterfeits: A Review and Extension,” Journal of Consumer Marketing, Vol. 24, No. 1, 36-47 Deutsch, M. and H. B. Gerard (1955), “A Study of Normative and Informational Social Influences Upon Individual Judgement,” Journal of Abnormal and Social Psychology, Vol. 51, 629-636 Dholakia U. M. (2001), “A Motivational Process Model of Product Involvement and Consumer Risk Perception,” European Journal of Marketing, Vol. 35, No. 11/12, 1340-1360 Dowling, G.R. and Richard Stealin (1994), “A Model of Perceived Risk and Intended Riskhandling Activities,” The Journal of Consumer Research, Vol. 21, No. 1, 119-134 Eisend, M. and Pakize Schuchert-Güler (2006), “Explaining Counterfeit Purchases: A Review and Preview,” Academy of Marketing Science Review, Vol. 10, No. 12, 1-25 European Commission (2008), “Report on EU Customs Enforcement of Intellectual Property Rights: Results at the European Border,” URL: http://ec.europa.eu/taxation_customs/resources/documents/customs/customs_controls/co unterfeit_piracy/statistics/2009_statistics_for_2008_full_report_en.pdf (24/02/2010) Gentry J. W., Sanjay Putrevu and Clifford J Schultz II (2006), “The Effects of Counterfeiting on Consumer Search,” Journal of Consumer Behaviour, Vol. 5, No. 3, 245-256 Gentry, James W., Sanjay Putrevu, Clifford Schultz II and Suraj Commuri (2001), “How Now Ralph Lauren? The Separation of Brand and Product in a Counterfeit Culture,” Advances in Consumer Research, Vol. 28, 258-265 Gessler, C. (2009), “Counterfeiting in The Luxury Industry: The True Costs of Counterfeit Goods,” VDM Verlag Dr. Müller Aktiengesellschaft & Co. KG, Germany. Gotlieb, J. B. and Dan Sarel (1991), “Comparative Advertising Effectiveness: The Role of Involvement and Source Credibility,” JA, Vol. 20, No. 1, 38-45 Green, R.T. and Tasman Smith (2002), “Executive Insights: Countering Brand Counterfeiters,” Journal of International Marketing, Vol. 10, No. 4, 89-106 Grossman, G. M. and Carl Shapiro (1988), “Counterfeit-product Trade,” American Economic Review, Vol. 78, No. 1, 59-75
54
Heffes, Ellen M. (2008), “Fending Off Pirates,” Financial Executive, Vol. 24, 40-42. Hilton, B., Chong Ju Choi and Stephen Chen (2004), “The Ethics of Counterfeiting in the Fashion Industry: Quality, Credence and Profit Issues,” Journal of Business Ethics, Vol. 55, No. 4, 345-354 Hogg, Michael A. and Scott A. Reid (2006), “Social Identity, Self-Categorization, and the Communication of Group Norms,” Communication Theory, Vol. 16, 7-30 Interview with Miss Hagenaers A., Trade Mark & Design Attorney, 16 Feb 2010, Antwerp. Jenner, T. and Artun, E. (2005), “Determinantendes Erwerbs gefälschter Markenprodukte: Ergebnisse einer empirischen Untersuchung,” Der Markt, Vol. 44, No. ¾, 142-150 JISC Legal (2008), “Intellectual Property Rights – Overview,” URL: http://www.jisclegal.ac.uk/Portals/12/Documents/PDFs/IPROverview.pdf (03/05/2010) Kapoor A. and Chinmaya Kulshrestha (2008), “Fashion Fetish,” Monash Business Review, Vol. 4, No. 1, 8-9 Kos Koklic, M. and Irena O. Vida (2009) “A cross cultural comparison of the determinants of consumer willingness to purchase non-deceptive counterfeit products.” In: Seawright, K. (Ed.), Smith, S. (Ed.). Proceedings of the Fourteenth Cross-Cultural Research Conference, Puerto Vallarta, December 13-16, 2009. Brigham Young University-Hawaii, 2009. Large, J. (2009), “Consuming Counterfeits: Exploring Assumptions about Fashion Counterfeiting,” Papers from the British Criminology Conference – postgraduate paper, Vol. 9, 3-20 Lewis, K. (2009), “The Fake and the Fatal: The Consequences of Counterfeits,” The Park Place Economist, Vol. 17, No. 1, 47-58 Lichtenstein, D. R. and Scott Burton (1989), “The Relationship Between Perceived and objective Price-Quality,” Journal of Marketing Research, Vol. XXVI, 429-443 Maheswaran, D. and Joan Meyer-Levy (1990), “The Influence of Message-Framing and Issue Involvement,” Journal of Marketing Research, Vol. 27, No. 3, 361-367 Maldonado C. and Evelyne C. Hume (2005), “Attitudes toward Counterfeit Products: an Ethical Perspective,” Journal of Legal, Ethical and Regulatory Issues, Vol. 8, No. 2, 105-117 Mannetti L., Antonio Pierro and Stefano Livi (2002), “Explaining Consumer Conduct: From Planned to Self-Expressive Behavior,” Journal of Applied Social Psychology, Vol. 32, No. 7, 1431-1451 Meng-Hsiang, H., Chia-Hui Yen, Chao-Min Chiu and Chun-Ming Chan (2006), “A Longitudinal Investigation of Continued Online Shopping Behaviour: An Extension of The Theory of Planned Behaviour,” Int. J. Human-Computer Studies, Vol. 64, 889-904 Mitchell, V. W. (1999), “Consumer Perceived Risk: Conceptualisations and Models,” European Journal of Marketing, Vol. 33, No. ½, 163-195
55
Mitra D. and Peter N. Golder (2006), “How Does Objective Quality Affect Perceived Quality? Short-term Effects, Long-term Effects, and Assymmetries,” Marketing Science, Vol. 25, No. 3, 230-247 Nia, A. And Judith Lynne Zaichkowsky (2000), “Do Counterfeits Devalue the Ownership of Luxury Brands?,” Journal of Product and Brand Management, Vol. 9, No. 7, 485-497 Nielsen (2008), “Consumers and Designer Brands: A Global Nielsen Report,” URL: http://tw.nielsen.com/site/documents/GlobalNielsenLuxuryBrandsMay08.pdf O’ Cass, A. (2003), “Fashion Clothing Consumption: Antecedents and Consequences of Fashion Clothing Involvement,” European Journal of Marketing, Vol. 38, No. 7, 869-882 O’ Cass, A. and Craig Julian (2001), “Fashion Clothing Consumption: Studying the Effects of Materialistic Values, Self-Image/Product-Image Congruency Relationships, Gender and Age on Fashion Clothing Involvement,” published in the proceedings of the 2001 Australia New Zealand Marketing Academy Conference (ANZMAC) O’ Cass, A. and Hmily Frost (2002), “Status Consciousness and Fashion Consumption,” published in the proceedings of the 2002 Australia New Zealand Marketing Academy Conference (ANZMAC) OECD (2008), The Economic Impact of Counterfeiting and Piracy Organisation for Economic Co-operation and Development (2008), The Economic Impact of Counterfeiting and Piracy. Paris, URL: http://www.sourceoecd.org/industrytrade/9789264045514 (13/11/2009) Ouelette, J. A. and Wendy Wood (1998), “Habit and Intention in Everyday Life: The Multiple processes by Which Past Behavior Predicts Future Behavior,’ Psychological Bulletin, Vol. 124, No. 1, 54-74 Penz, E. and Barbara Stöttinger (2005), “Forget the ‘Real’ Thing-Take the Copy! An Explanatory Model for the Volitional Purchase of Counterfeit Products,” Advances in Consumer Research, Vol. 32, 568-575 Penz, E. and Barbara Stöttinger (2008), “Original Brands and Counterfeit Brands-Do They Have Anything in Common?,” Journal of Consumer Behaviour, Vol. 7, No. 2, 146-163 Penz, E., Bodo B. Schlegelmilch and Barbara Stöttinger (2009), “Voluntary Purchase of Counterfeit Products: Empirical Evidence From Four Countries,” Journal of International Consumer Marketing, Vol. 21, No. 1, 67-84 Perugini, M. and Richard P. Bagozzi (1999), “The Role of Desires and Anticipated Emotions in Goal-Directed Behaviors: Expanding and Deepening The Theory of Planned Behavior,” Working Paper, The University of Michigan, Ann Arbor, MI. Phau, I. and Min Teah (2009), “Devil Wears (counterfeit) Prada: A study of Antecedents and Outcomes of Attitudes Towards Counterfeits of Luxury Brands,” Journal of Consumer Marketing, Vol. 26, No. 1, 15-27 Phau, I., Min Teah and Agnes Lee (2009), “Targeting Buyers of Counterfeits of Luxury Brands: A Study on Attitudes of Singaporean Consumers,” Journal of Targeting, Measurement and Analysis for Marketing, Vol. 17, No. 1, 3-15
56
Phillips, T. (2005), “Knockoff: The Deadly Trade in Counterfeit Goods,” Kogan Page: United Kingdom Pollinger, Z. A. (2008), “Counterfeit Goods and Their Potential Financing of International Terrorism,” The Michigan Journal of Business, Vol. 1, No. 1, 85-102 Prendergast, G., Leung H. Chuen and Ian Phau (2002), “Understanding Consumer Demand for Non-deceptive Pirated Brands,” Marketing and Intelligence Planning, Vol. 20, No. 7, 405416 Qian, Y. (2008), “Impacts of Entry By Counterfeiters,” The Quarterly Journal of Economics, Vol. November, 1577-1609 Rodwell, S., Van Eeckhout, P., Reid, A. and Walendowski, J. (2007) Study: Effects of counterfeiting on EU SMEs and a review of various public and private IPR enforcement initiatives and resources, Framework Contract B3/ENTR/04/093- FC-Lot 6. Santos, J. Freitas and J. Cadima Ribiero (2006), “An exploratory Study of The Relationship Between Counterfeiting and Culture,” Polytechnical Studies Review, Vol. 3, No. 5/6, 227243 Smith, J. R., Deborah J. Terry, Antony S. R. Manstead, Winnifred R. Louis, Diana Kotterman and Jacqueline Wolfs (2008), “The Attitude-Behavior Relationship in Consumer Conduct: The Role of Norms, Past Behavior and Self-Identity,” The Journal of Social Psychology, Vol. 148, No. 3, 311-333 Staake, T., Frédéric Thiesse and Elgar Fleisch (2009), “The Emergence of Counterfeit Trade: A Literature Review,” European Journal Of Marketing, Vol. 43, No. ¾, 320-349 Stone R. N. and Kjell Gronhaug (1993), “Perceived Risk: Further Considerations for The Marketing Discipline,” European Journal of Marketing, Vol. 27, No. 3, 39-50 Swami, V., Thomas Chamorro-Premuzic and Adrian Furnham (2009), “Faking It: Personality and Individual Difference Predictors of Willingness to Buy Counterfeit Goods,” The Journal of Socio-Economics, Vol. 38, No. 5, 820-825 Sweeney, J.C., Geoffrey N. Soutar and Lester W. Johnson (1999), “The Role of Perceived Risk in the Quality-Value Relationship: A Study in a Retail Environment,” Journal of Retailing, Vol. 75, No. 1, 77-105 Taylor, J. W. (1974), “The Role of Risk in Consumer Behavior,” The Journal of Marketing, Vol. 38, No. 2, 54-60 Thompson, J. C., Allen D. Engle and Judith Winters Spain (2005), “Counterfeit Products and Actor Proximity: An Exploratory Multidimensional Study Design.” New York: 12th Annual International Conference Promoting Business Ethics. October 25-28, 2005. Tom, G., Barbara Garibaldi, Yvette Zeng and Julie Pilcher (1998), “Consumer Demand for Counterfeit Goods,” Psychology and Marketing, Vol. 15, No. 5, 405-421 Tonglet M. (2001), “Consumer Misbehaviour: An Exploratory Study of Shoplifting,” Journal of Consumer Behaviour, Vol. 1, No. 4, 336-354
57
Trott, P. And A. Hoecht (2007) “Product Counterfeiting, Non-consensual Acquisition of Technology and New Product Development: An Innovation Perspective,” European Journal of Innovation Management, Vol. 10, No. 1, 126-143 Tuyls, P. and Lejla Batina (2006), “RFID-tags for Anti-Counterfeiting,” In D. Pointcheval, editor, Topics in Cryptology - CT-RSA 2006, The Cryptographers’ Track at the RSA Conference 2006, Lecture Notes in Computer Science, San Jose, California, USA, February 2006. Springer-Verlag. Tyler, Tom R. (2006), “Why people obey the law,” Princeton University Press: Princeton, New Jersey UNICRI (2009), “Counterfeiting: A Global Spread, A Global Threat,” Trens Organ Crim (Springer), Vol. 12, 59-77 University of Texas (2009), “Solving homework problems in data analysis II,” Lecture notes, URL: http://www.utexas.edu/courses/schwab/sw388r7/SolvingProblems/SolvingHomeworkProbl ems.htm (12/05/2010) Vagg, J. and Justine Harris (2000) “False Profits: Why Product Counterfeiting Is Increasing,” European Journal on Criminal Policy and Research, Vol. 8, 107-115 Van Kenhove P. and Patrick De Pelsmacker (2006), “Marktonderzoek: Methoden en Toepassingen,” Pearson Education Benelux Van Kenhove P., Patrick De Pelsmacker, Katrien Wijnen and Wim Janssens (2008), “Marketing Research with SPSS,” Pearson Education Limited: England Veloutsou, C. and Xuemei Bian (2008), “A Cross-national Examination of Consumer Perceived Risk in The Context of Non-deceptive Counterfeit Brands,” Journal of Consumer Behaviour, Vol. 7, No. 1, 3-20 Vida, I. (2007), “Determinants of Consumer Wiilingness to Purchase Non-Deceptive Counterfeit Products,” Managing Global Transitions, Vol. 5, No. 3, 253-270 Vlaev, I., Nick Chater, Rich Lewis and Greg Davies (2009), “Reason-based Judgments: Using Reasons to Decouple Perceived Price-Quality Correlation,” Journal of Economic Psychology, Vol. 30, No. 5, 721-731 Völckner, F. and Julian Hofmann (2007), “The Price-perceived Quality Relationship: A Metaanalytic Review and Assessment of Its Determinants,” Market Lett, Vol. 18, 181-196 Wakefield, K. L. and J. Jeffrey Inman (2003), “Situational Price Sensitivity: The Role of Consumption Occasion, Social Context and Income,” Journal of Retailing, Vol. 79, 199212 Ward, S. F. (2007), “Knockoffs Landing on Retail Shelves,” ABA Journal, Vol. 93, No.2 Wee, C.H., Soo-Jiuan Tan and Kim-Hong Cheok (1995), “Non-price Determinants of Intention to Purchase Counterfeit Goods,” International Marketing Review, Vol. 12, No. 6, 19-46 Wilcox, K., Hyeong Min Kim and Sankar Sen (2009), “Why Do Consumers Buy Counterfeit Luxury Brands?,” Journal of Marketing Research, Vol. 46, No. 2, 247-259
58
Yoo, B. and Seung-Hee Lee (2009), “A Review of The Determinants of Counterfeiting and Piracy and The Proposition for Further Research,” The Korean Journal of Policy Studies, Vol. 24, No. 1, 1-38 Yoo, B. and Seung-Hee Lee (2009), “Buy Genuine Luxury Fashion Products or Counterfeits?,” Advances in Consumer Research, Vol. 36, 280-286 Zeitmahl, V. A. (1988), “Consumer Perceptions of Price, Quality, and Value: A Means-End Model and Synthesis of Evidence,” Journal of Marketing, Vol. 52, No. 3, 2-22
59
ANNEX 1.1
ANNEX 1.1: Questionnaire used for pre-test OPENINGSBOODSCHAP
Hallo! Mijn naam is Dennis De Cat. Ik studeer Toegepaste Economische Wetenschappen aan de Universiteit Gent. Momenteel ben ik bezig aan mijn thesisonderzoek. Uiteraard kan ik dit niet alleen. Daarom wordt jouw hulp enorm op prijs gesteld. Indien je deelneemt aan dit onderzoek, maak je bovendien kans op enkele leuke prijzen: Je hebt de keuze uit: - een Fnac-bon ter waarde van €25 - een handtas van het merk BOO! Bij de verschillende vragen in deze enquête zijn er geen juiste of foute antwoorden. Ik ben enkel geïnteresseerd in jouw persoonlijke mening over het ondervraagde onderwerp. Gelieve deze vragenlijst dan ook individueel en zo waarheidsgetrouw mogelijk in te vullen. Deze vragenlijst is gegarandeerd honderd procent anoniem. Als u wenst deel te nemen aan de wedstrijd, volstaat het de wedstrijdvraag te beantwoorden én uw emailadres in te vullen. Geen paniek! Dit e-mailadres zal enkel gebruikt worden voor het selecteren en het contacteren van de winnaars en dus zeker niet voor andere doeleinden. Klik op ‘>>’ als je wilt deelnemen aan het onderzoek. Het invullen van deze korte vragenlijst zal ongeveer 15 minuten van jouw tijd in beslag nemen. Gelieve uw browser te maximaliseren voor een optimale weergavegrootte. AANDACHTSTREKKER
Bedankt om deel te nemen aan dit onderzoek. U bent een zeer grote hulp voor mij! Ik zou u willen vragen deze vragenlijst zo aandachtig, zorgvuldig en waarheidsgetrouw mogelijk in te vullen. Dit is van zeer groot belang voor de resultaten van het onderzoek, aangezien er slecht weinig respondenten gecontacteerd zullen worden. De antwoorden die u geeft, zijn voor mij dus van zeer groot belang. Klik op '>>' om het onderzoek te starten. SOCIO-DEMOGRAFISCH
60
ANNEX 1.1 Wat is uw geslacht? -‐
Man
-‐
Vrouw
PRODUCT RELEVANCE
MAN: In welke mate betekent een horloge iets voor jou? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’) VROUW: In welke mate betekent een handtas iets voor jou? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’) BOODSCHAP MET DEFINITIE NAMAAKGOEDEREN
Dit onderzoek gaat over namaakproducten. Gelieve onderstaande definitie van 'namaakproducten' zorgvuldig en aandachtig te lezen. "Namaakproducten zijn imitaties of reproducties van merkgoederen. De merknaam of herkenbare elementen, zoals het logo of de verpakking, werden zonder toelating gebruikt." BRAND RELEVANCE
Van welk(e) merk(en) zou je overwegen ooit een namaakproduct te kopen in de veronderstelling dat dit makkelijk verkrijgbaar is voor een voor jou aanvaardbare prijs en kwaliteit? Je mag meerdere merken aanduiden. • Lacoste
• Diesel
• Calvin Klein
• Louis Vuitton
• DKNY
• Tommy Hilfiger
• Ralph Lauren
• Rolex
• Tag Heuer
• Bulgari
61
ANNEX 1.1 • Longchamp
• Dolce & Gabbana
• Gucci
• Versace
• Armani
• Prada
• Andere: … PERSONAL HARM
MAN: Voor mannen wordt de boodschap getoond met het product ‘horloge’. VROUW: Voor vrouwen wordt de boodschap getoond met het product ‘handtas’.
In welke mate vind je dat jij als persoon schade berokkend wordt in de volgende situaties (7-punt Likert schaal gaande van ‘Ik vind dit helemaal niet erg voor mezelf’ tot ‘Ik vind dit zeer erg voor mezelf’): 1. Een namaak(horloge/handtas) bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. 2. Een namaak(horloge/handtas) is vervaardigd uit materiaal dat snel verkleurt en functioneert mogelijks niet zoals het hoort. 3. Een namaak(horloge/handtas) is mogelijks zijn prijs niet waard. 4. Anderen zouden kunnen zien dat je een namaak(horloge/handtas) draagt. 5. Het feit dat je een namaak(horloge/handtas) zou kopen, geeft je een ongemakkelijk gevoel. 6. Het aanschaffen van een namaak(horloge/handtas) kan je veel tijd kosten (bv. lang zoeken op internet, lang onderhandelen met verkoper, etc.) SOCIETAL HARM
62
ANNEX 1.1 In welke mate vind je dat de maatschappij schade berokkend wordt in de volgende gevallen? (7-punt Likert schaal gaande van ‘Ik vind dit helemaal niet erg voor de maatschappij’ tot ‘ Ik vind dit zeer erg voor de maatschappij’) 1. Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties. 2. Door het kopen van namaakgoederen verliezen de authentieke bedrijven (wiens producten dus nagemaakt worden) hun inkomsten. 3. Door het kopen van namaakgoederen misloopt de staat belastingsinkomsten. 4. De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven én hun toeleveringsbedrijven. 5. Uit onderzoek blijkt dat de namaakindustrie fungeert als een rem op de ontwikkeling van nieuwe, innovatieve producten door de originele producenten. 6. De productie van namaakgoederen verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij. PRICE AND QUALITY
MAN: Voor mannen worden de vragen gesteld met het product ‘horloge’. VROUW: Voor vrouwen worden de vragen gesteld met het product ‘handtas’. Gelieve volgende vragen zo waarheidsgetrouw mogelijk in te vullen. U kan gewoon een bedrag noteren. Het €-teken hoef je niet toe te voegen. -‐
Welke prijs bent u bereid maximaal te betalen voor een namaak(horloge/handtas) van hoge kwaliteit van een luxemerk dat je zou willen hebben? 63
ANNEX 1.1 -‐
Welke prijs bent u bereid maximaal te betalen voor een namaak(horloge/handtas) van gemiddelde kwaliteit van een luxemerk dat je zou willen hebben?
-‐
Wat vindt u een gemiddelde prijs voor een namaak(horloge/handtas) van hoge kwaliteit van een luxemerk dat je zou willen hebben?
-‐
Wat vindt u een gemiddelde prijs voor een namaak(horloge/handtas) van gemiddelde kwaliteit van een luxemerk dat je zou willen hebben?
-‐
Wat vindt u een lage prijs voor een namaak(horloge/handtas) van hoge kwaliteit van een luxemerk dat je zou willen hebben?
-‐
Wat vindt u een lage prijs voor een namaak(horloge/handtas) van gemiddelde kwaliteit van een luxemerk dat je zou willen hebben?
-‐
Welke prijs zou u betalen voor diezelfde horloge/handtas van het echte merk?
STIMULI PRE-TEST
Hieronder krijg je verschillende boodschappen over namaakproducten te zien. Lees deze aandachtig en antwoord daarna op onderstaande vragen. BOODSCHAPPEN: 1. Een namaakproduct bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. 2. Een namaakproduct is vervaardigd uit materiaal dat snel verkleurt en functioneert mogelijks niet zoals het hoort. 3. Een namaakproduct is mogelijks zijn prijs niet waard. 4. Anderen zouden kunnen zien dat je een namaakproduct draagt. 5. Het feit dat je een namaakproduct zou kopen, geeft je een ongemakkelijk gevoel. 6. Het aanschaffen van een namaakproduct kan je veel tijd kosten (bv. lang zoeken op internet, lang onderhandelen met verkoper, etc.) 7. Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties.
64
ANNEX 1.1 8. Door het kopen van namaakgoederen verliezen de authentieke bedrijven (wiens producten dus nagemaakt worden) hun inkomsten. 9. Door het kopen van namaakgoederen misloopt de staat belastingsinkomsten. 10. De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven én hun toeleveringsbedrijven. 11. Uit onderzoek blijkt dat de namaakindustrie fungeert als een rem op de ontwikkeling van nieuwe, innovatieve producten door de originele producenten. 12. De productie van namaakgoederen verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij. VRAGEN (7-punt Likert schaal gaande van ‘helemaal niet geloofwaardig’ tot ‘zeer geloofwaardig’): -‐
Zou het geloofwaardig zijn voor een namaakproduct van hoge kwaliteit en gemiddelde prijs om één van bovenstaande boodschappen te krijgen?
-‐
Zou het geloofwaardig zijn voor een namaakproduct van hoge kwaliteit en lage prijs om één van bovenstaande boodschappen te krijgen?
-‐
Zou het geloofwaardig zijn voor een namaakproduct van gemiddelde kwaliteit en gemiddelde prijs om één van bovenstaande boodschappen te krijgen?
-‐
Zou het geloofwaardig zijn voor een namaakproduct van gemiddelde kwaliteit en lage prijs om één van bovenstaande boodschappen te krijgen?
OPMERKING: Bij elke vraag werd ook gevraagd aan te duiden (met nummers van 1 tot 12) welke boodschappen niet geloofwaardig lijken in deze situatie. PERCEIVED QUALITY
MAN: Voor mannen wordt de boodschap getoond met het product ‘horloge’. VROUW: Voor vrouwen wordt de boodschap getoond met het product ‘handtas’.
65
ANNEX 1.1 Gelieve hieronder aan te duiden in welke mate jij denkt dat een namaakhorloge van een luxemerk dat jij zou willen, volgende eigenschappen bezit: (7-punt Likert schaal gaande van ‘heel laag’ tot ‘heel hoog’) 1. De waarschijnlijkheid dat dit product kwalitatief betrouwbaar is, is... 2. De graad van vakmanschap waarmee dit product vervaardigd is, zal waarschijnlijk ... zijn 3. Het product zou duurzaam kunnen lijken. (7-punt Likert schaal gaande van ‘helemaal niet akkoord’ tot ‘helemaal akkoord’) 4. De waarschijnlijkheid dat dit product feilloos is, is... 5. Dit product zou van ... kwaliteit moeten zijn. WEDSTRIJDVRAAG
Wens je deel te nemen aan de wedstrijd die verbonden is aan dit onderzoek? Ter herinnering: Er zijn volgende prijzen te winnen: - een Fnac-bon t.w.v. €25 - een handtas van het merk BOO! --------- indien men kiest om deel te nemen, krijgt men de volgende vragen -------Wat is het meest gekochte luxe modemerk ter wereld? -‐
Calvin Klein
-‐
Diesel
-‐
Ralph Lauren
-‐
Chanel
Hoeveel procent van de deelnemers zal deze vraag correct beantwoorden? Naar welke prijs gaat jouw voorkeur? -‐
Fnac-bon t.w.v. €25
-‐
Een handtas van het merk BOO!
66
ANNEX 1.1 Wat is je emailadres waarop we je kunnen contacteren indien je een prijs gewonnen hebt? BEDANKING VOOR DEELNAME
Je hebt met succes de vragenlijst beëindigd. Nogmaals bedankt voor de deelname aan dit onderzoek! Klik op '>>' om uw antwoorden op te slagen. Daarna kan u uw browser sluiten.
67
ANNEX 1.2
ANNEX 1.2: Statistical analyses of pre-test 1. Products and brands 1.1 Product relevance
1.2 Brands
68
ANNEX 1.2
2. Personal harm Descriptives N
Minimum
Maximum
Mean
Std. Deviation
PersonalHarmPhysicalRisk
17
1,00
7,00
6,1176
1,53632
PersonalHarmPerformanceRisk
17
2,00
7,00
5,3529
1,36662
PersonalHarmFinancialRisk
17
3,00
7,00
5,0000
0,93541
PersonalHarmSocialRisk
17
1,00
7,00
4,2353
1,95350
PersonalHarmPsychologicalRisk
17
1,00
7,00
3,8824
2,05798
PersonalHarmTimeRisk
17
1,00
7,00
4,2941
1,64942
Valid N (listwise)
17
Group Statistics
N
Mean
Std. Deviation
Std. Error Mean
woman
10
6,5000
0,84984
0,26874
man
7
5,5714
2,14920
0,81232
woman
10
5,6000
1,17379
0,37118
man
7
5,0000
1,63299
0,61721
woman
10
5,1000
1,10050
0,34801
man
7
4,8571
0,69007
0,26082
woman
10
4,5000
1,71594
0,54263
man
7
3,8571
2,34013
0,88448
woman
10
4,5000
1,58114
0,50000
man
7
4,0000
1,82574
0,69007
woman
10
4,8000
1,81353
0,57349
man
7
2,5714
1,71825
0,64944
Gender PersonalHarmPhysicalRisk
PersonalHarmPerformanceRisk
PersonalHarmFinancialRisk PersonalHarmSocialRisk PersonalHarmTimeRisk PersonalHarmPsychologicalRisk
69
ANNEX 1.2
3. Societal harm Descriptives N
Minimum
Maximum
Mean
Std. Deviation
Societal harm terrorism
17
1
7
5,94
1,519
Societal harm authentic co
17
1
7
4,88
1,654
Societal harm taxes
17
1
6
3,76
1,921
Societal harm job loss
17
3
7
5,76
1,200
Societal harm innovation
17
2
7
5,06
1,298
Societal harm child labour
17
1
7
6,12
1,536
Valid N (listwise)
17
Group Statistics
Societal harm terrorism Societal harm authentic co Societal harm taxes Societal harm job loss Societal harm innovation Societal harm child labour
Gender
N
Mean
Std. Deviation
Std. Error Mean
woman
10
6,40
,843
,267
man
7
5,29
2,059
,778
woman man woman
10 7 10
5,20 4,43 3,50
1,619 1,718 1,958
,512 ,649 ,619
man
7
4,14
1,952
,738
woman
10
6,10
,994
,314
man
7
5,29
1,380
,522
woman
10
5,30
1,418
,448
man
7
4,71
1,113
,421
woman
10
6,70
,675
,213
man
7
5,29
2,059
,778
70
ANNEX 1.2
4.
Price, quality, perceived harm and message credibility Descriptives N
Minimum
Maximum
Mean
Std. Deviation
CREDIBILITY high quality moderate price
17
3
7
5,29
1,105
CREDIBILITY high quality low price
17
3
7
5,94
1,029
CREDIBILITY moderate quality moderate price
17
2
7
5,59
1,176
CREDIBILITY moderate quality low price
17
3
7
6,00
1,118
Valid N (listwise)
17
Group Statistics
Gender
N
Mean
Std. Deviation
Std. Error Mean
CREDIBILITY high quality moderate price
woman
10
5,00
1,333
,422
man
7
5,71
,488
,184
CREDIBILITY high quality low price
woman
10
5,90
1,287
,407
man
7
6,00
,577
,218
CREDIBILITY moderate quality moderate price
woman
10
5,80
,919
,291
man
7
5,29
1,496
,565
CREDIBILITY moderate quality low price
woman
10
6,20
,919
,291
man
7
5,71
1,380
,522
71
ANNEX 1.2
5. Credibility
72
ANNEX 2
ANNEX 2: Scenarios (stimuli) used in the main research 1. LOW price, LOW quality, PERSONAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex,... te kopen. Dit product heeft een lage prijs, maar de kwaliteit lijkt van een laag niveau. Uit recent onderzoek (2010) blijkt dat er enkele problemen verbonden zijn aan het kopen van namaakproducten: - Een namaakproduct bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. - Een namaakproduct is vaak vervaardigd uit materiaal dat snel verkleurt én functioneert mogelijkerwijze niet zoals het hoort. - Een namaakproduct is mogelijkerwijze zijn prijs niet waard. 2. LOW price, MODERATE quality, PERSONAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een lage prijs en de kwaliteit lijkt van een gemiddeld niveau.
Uit recent onderzoek (2010) blijkt dat er enkele problemen verbonden zijn aan het kopen van namaakproducten: - Een namaakproduct bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. - Een namaakproduct is vaak vervaardigd uit materiaal dat snel verkleurt én functioneert mogelijkerwijze niet zoals het hoort. - Een namaakproduct is mogelijkerwijze zijn prijs niet waard. 3. MODERATE price, LOW quality, PERSONAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een gemiddelde prijs, maar de kwaliteit lijkt van 73
ANNEX 2 een laag niveau. Uit recent onderzoek (2010) blijkt dat er enkele problemen verbonden zijn aan het kopen van namaakproducten: - Een namaakproduct bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. - Een namaakproduct is vaak vervaardigd uit materiaal dat snel verkleurt én functioneert mogelijkerwijze niet zoals het hoort. - Een namaakproduct is mogelijkerwijze zijn prijs niet waard. 4. MODERATE price, MODERATE quality, PERSONAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een gemiddelde prijs en de kwaliteit lijkt van een gemiddeld niveau. Uit recent onderzoek (2010) blijkt dat er enkele problemen verbonden zijn aan het kopen van namaakproducten: - Een namaakproduct bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. - Een namaakproduct is vaak vervaardigd uit materiaal dat snel verkleurt én functioneert mogelijkerwijze niet zoals het hoort. - Een namaakproduct is mogelijkerwijze zijn prijs niet waard. 5. LOW price, LOW quality, SOCIETAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een lage prijs, maar de kwaliteit lijkt van een laag niveau. Uit recent onderzoek (2010) blijkt dat het kopen van namaakproducten kan leiden tot maatschappelijke problemen: - Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en het verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties.
74
ANNEX 2
- De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven (diegenen die het originele product maken) én hun toeleveringsbedrijven. - De productie van namaakproducten verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij. 6. LOW price, MODERATE quality, SOCIETAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een lage prijs en de kwaliteit lijkt van gemiddeld niveau. Uit recent onderzoek (2010) blijkt dat het kopen van namaakproducten kan leiden tot maatschappelijke problemen: - Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en het verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties. - De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven (diegenen die het originele product maken) én hun toeleveringsbedrijven. - De productie van namaakproducten verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij. 7. MODERATE price, LOW quality, SOCIETAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een gemiddelde prijs, maar de kwaliteit lijkt van een laag niveau. Uit recent onderzoek (2010) blijkt dat het kopen van namaakproducten kan leiden tot maatschappelijke problemen: - Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en het verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties.
75
ANNEX 2 - De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven (diegenen die het originele product maken) én hun toeleveringsbedrijven. - De productie van namaakproducten verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij. 8. MODERATE price, MODERATE quality, SOCIETAL harm Stel je voor dat je de kans krijgt een namaak luxeproduct (bv. handtas of horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex, Calvin Klein,... te kopen. Dit product heeft een gemiddelde prijs en de kwaliteit lijkt van een gemiddeld niveau. Uit recent onderzoek (2010) blijkt dat het kopen van namaakproducten kan leiden tot maatschappelijke problemen: - Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en het verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties. - De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven (diegenen die het originele product maken) én hun toeleveringsbedrijven. - De productie van namaakproducten verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij.
76
ANNEX 3.1
ANNEX 3.1: Questionnaire used in the main research OPENINGSBOODSCHAP
Hallo! Mijn naam is Dennis De Cat. Ik studeer Toegepaste Economische Wetenschappen aan de Universiteit Gent. Momenteel ben ik bezig aan mijn thesisonderzoek. Uiteraard kan ik dit niet alleen. Daarom wordt jouw hulp enorm op prijs gesteld. Indien je deelneemt aan dit onderzoek, maak je bovendien kans op enkele leuke prijzen: Je hebt de keuze uit: - een Fnac-bon ter waarde van €25 - een handtas van het merk BOO! Bij de verschillende vragen in deze enquête zijn er geen juiste of foute antwoorden. Ik ben enkel geïnteresseerd in jouw persoonlijke mening over het ondervraagde onderwerp. Gelieve deze vragenlijst dan ook individueel en zo waarheidsgetrouw mogelijk in te vullen. Deze vragenlijst is gegarandeerd honderd procent anoniem. Als u wenst deel te nemen aan de wedstrijd, volstaat het de wedstrijdvraag te beantwoorden én uw emailadres in te vullen. Geen paniek! Dit e-mailadres zal enkel gebruikt worden voor het selecteren en het contacteren van de winnaars en dus zeker niet voor andere doeleinden. Klik op ‘>>’ als je wilt deelnemen aan het onderzoek. Het invullen van deze korte vragenlijst zal ongeveer 15 minuten van jouw tijd in beslag nemen. Gelieve uw browser te maximaliseren voor een optimale weergavegrootte. AANDACHTSTREKKER EN BEDANKING
Bedankt om deel te nemen aan dit onderzoek. U bent een zeer grote hulp voor mij! Ik zou u willen vragen deze vragenlijst zo aandachtig, zorgvuldig en waarheidsgetrouw mogelijk in te vullen. Dit is van zeer groot belang voor de resultaten van het onderzoek, aangezien er slechts een beperkt aantal respondenten gecontacteerd zullen worden. De antwoorden die u geeft, zijn voor mij dus van zeer groot belang. Klik op '>>' om het onderzoek te starten. GESLACHT 77
ANNEX 3.1
Wat is uw geslacht? -‐
Man
-‐
Vrouw
BOODSCHAP MET DEFINITIE NAMAAKGOEDEREN
Dit onderzoek gaat over namaakproducten. Gelieve onderstaande definitie van 'namaakproducten' zorgvuldig en aandachtig te lezen. "Namaakproducten zijn imitaties of reproducties van merkgoederen. De merknaam of herkenbare elementen, zoals het logo of de verpakking, werden zonder toelating gebruikt." STIMULI
Respondenten worden ‘at random’ toegewezen aan 1 van de 8 condities. Men kan de verschillende scenario’s vinden in Annex 2. MESSAGE INVOLVEMENT
Gelieve hieronder aan te duiden in welke mate je akkoord gaat met volgende uitspraken omtrent de informatie die je net te zien kreeg. (7-punt Likert schaal gaande van ‘helemaal niet akkoord’ tot ‘helemaal akkoord’) 1. De informatie deed me nadenken. 2. Ik voelde me betrokken bij wat de informatie te vertellen had. 3. De informatie was interessant. 4. De informatie bracht me op bepaalde ideeën en gedachten. 5. Ik voelde sterke emoties en gevoelens bij het lezen van de informatie. 6. Ik vond de informatie relevant. AD CREDIBILITY
78
ANNEX 3.1 In welke mate ga je akkoord met onderstaande uitspraak m.b.t. de informatie die je net te zien kreeg? (7-punt Likert schaal gaande van ‘helemaal niet akkoord’ tot ‘helemaal akkoord’) -‐
De informatie is geloofwaardig.
PURCHASE INTENTION
1. Ben je van plan om in de toekomst een namaak luxeproduct (bv. handtas of horloge) te kopen van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex,... ? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’) 2. Hoe groot is de kans dat je in de toekomst een namaak luxeproduct (bv. handtas of horloge) zal kopen van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex,... ? (7-punt Likert schaal gaande van ‘heel klein’ tot ‘heel groot’) ATTITUDE TOWARD BUYING COUNTERFEIT LUXURY FASHION ITEMS
Wat vind jij over het kopen van namaak luxeproducten (bv. handtas, horloge) van één van de topmerken Diesel, Ralph Lauren, Gucci, Armani, Delvaux, Rolex,... ? (7-punt semantische differentiaal) 1. Slecht – goed 2. Onverstandig – verstandig 3. Schadelijk – onschadelijk 4. Onveilig – veilig 5. Onaangenaam – aangenaam 6. Onbevredigend – bevredigend 7. Bestraffend – belonend PAST PURCHASE BEHAVIOUR
79
ANNEX 3.1 Hebt u ooit al bewust een namaakproduct (eender welke categorie van producten) gekocht? -‐
ja
-‐
nee
Hebt u ooit al een namaak luxeproduct gekocht? (bv. kleding, horloge, handtas, juwelen, accessories) -‐
ja
-‐
nee
Hoeveel keer per jaar koopt u gemiddeld een namaak luxeproduct? (invullen in cijfers) PERCEIVED QUALITY
Gelieve hieronder aan te duiden in welke mate jij denkt dat een namaakhorloge (bv. handtas, horloge) van een luxemerk dat jij zou willen, volgende eigenschappen bezit: (7-punt semantische differentiaal schaal: laag-hoog) 1. De waarschijnlijkheid dat dit product kwalitatief betrouwbaar is, is... 2. De graad van vakmanschap waarmee dit product vervaardigd is, zal waarschijnlijk ... zijn 3. Het product zou duurzaam kunnen lijken. (7-punt Likert schaal gaande van ‘helemaal niet akkoord’ tot ‘helemaal akkoord’) 4. De waarschijnlijkheid dat dit product feilloos is, is... 5. Dit product zou van ... kwaliteit moeten zijn. PRICE QUALITY INFERENCE
Geef aan in welke mate je akkoord bent met volgende uitspraken. Ter informatie: een namaak luxeproduct kan bijvoorbeeld een handtas of een horloge zijn. (7-punt Likert schaal gaande van ‘helemaal niet akkoord’ tot ‘helemaal akkoord’) 1. Hoe hoger de prijs van een namaak luxeproduct, hoe hoger de kwaliteit.
80
ANNEX 3.1 2. Hoe meer men betaalt voor een namaak luxeproduct, hoe beter de kwaliteit. 3. Een namaak luxeproduct dat meer kost, verzekert betere prestaties. 4. Namaak luxeproducten die weinig kosten zijn van lage kwaliteit. SUBJECTIVE NORM
Gelieve aan te duiden in welke mate je het eens bent met onderstaande uitspraken. (7punt Likert schaal gaande van ‘helemaal oneens’ tot ‘helemaal eens’) 1. Mijn vrienden vinden het best dat ik geen namaak luxeproducten (bv. handtas, horloge) koop. 2. Mijn familie vindt dat ik geen namaak luxeproducten (bv. handtas, horloge) mag kopen. 3. Mensen die mijn beslissingen beïnvloeden vinden dat ik geen namaak luxeproducten (bv. handtas, horloge) mag kopen. 4. De maatschappij verwacht dat ik geen namaak luxeproducten (bv. handtas, horloge) koop. FASHION CONSCIOUSNESS
Gelieve hieronder aan te duiden in welke mate je akkoord gaat met volgende uitspraken omtrent jouw modebewustzijn. (7-punt Likert schaal gaande van ‘helemaal niet akkoord’ tot ‘helemaal akkoord’) 1. Ik heb meestal één of meerdere outfits van de nieuwste mode. 2. Ik houd mijn kleerkast up-to-date met de nieuwste modetrens. 3. Modieuze en aantrekkelijke 'styling' is zeer belangrijk voor mij. 4. Ik koop in verschillende winkels en kies verschillende merken om zoveel mogelijk variëteit te bekomen. AVAILABILITY (PBC)
81
ANNEX 3.1 In welke mate vind je dat namaak luxeproducten (bv. handtas, horloge) makkelijk beschikbaar zijn? (7-punt Likert schaal gaande van ‘helemaal niet gemakkelijk beschikbaar’ tot ‘heel gemakkelijk beschikbaar’) Als u een namaak luxeproduct (bv. handtas, horloge) zou kopen, waar zou u dit dan doen? -‐
Internet
-‐
Markten
-‐
Vakantiebestemmingen
-‐
Andere: …
Hoe makkelijk is het voor jou om binnen de 6 maand een namaak luxeproduct (bv. handtas, horloge) te kopen? (7-punt Likert schaal gaande van ‘helemaal niet gemakkelijk’ tot ‘heel gemakkelijk’) PRODUCT INVOLVEMENT
MAN: In welke mate betekent een horloge iets voor jou? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’) VROUW: In welke mate betekent een handtas iets voor jou? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’) MANIPULATIECHECK PERSONAL HARM
In welke mate vind je dat jij als persoon schade berokkend wordt in de volgende situaties (7-punt Likert schaal gaande van ‘Ik vind dit helemaal niet erg voor mezelf’ tot ‘Ik vind dit zeer erg voor mezelf’):
82
ANNEX 3.1 1. Een namaak(horloge/handtas) bevat mogelijks gevaarlijke stoffen waardoor je uitslag kan krijgen. 2. Een namaak(horloge/handtas) is vervaardigd uit materiaal dat snel verkleurt en functioneert mogelijks niet zoals het hoort. 3. Een namaak(horloge/handtas) is mogelijks zijn prijs niet waard. MANIPULATIECHECK SOCIETAL HARM
In welke mate vind je dat de maatschappij schade berokkend wordt in de volgende gevallen? (7-punt Likert schaal gaande van ‘Ik vind dit helemaal niet erg voor de maatschappij’ tot ‘ Ik vind dit zeer erg voor de maatschappij’) 1. Uit wetenschappelijk én empirisch onderzoek blijkt dat er een zeer sterke link is tussen de namaakindustrie en de fondsenverzameling van terroristische organisaties. Er wordt dus beweerd dat het produceren en verkopen van namaakgoederen dient als dekmantel voor de financiering van terroristische organisaties. 2. De namaakindustrie is verantwoordelijk voor jobverliezen in grote aantallen in de authentieke bedrijven én hun toeleveringsbedrijven. 3. De productie van namaakgoederen verloopt niet conform de strikte regels met betrekking tot de arbeidswetgeving. Kinderarbeid viert hoogtij. EXPLORATIEVE VRAGEN -‐
Is het verkopen van namaak luxeproducten (bv. handtas, horloge) illegaal? (7punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’)
-‐
Vindt u dat het verkopen van namaak luxeproducten (bv. handtas, horloge) bestraft moet worden? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’)
-‐
Vindt u dat het bewust kopen van namaak luxeproducten (bv. handtas, horloge) bestraft moet worden? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’)
83
ANNEX 3.1 -‐
Vindt u dat de doorsnee consument voldoende op de hoogte is van de gevolgen (persoonlijk én maatschappelijk) van het kopen en verkopen van namaak luxeproducten (bv. handtas, horloge)? (7-punt Likert schaal gaande van ‘helemaal niet’ tot ‘helemaal wel’)
-‐
Denkt u dat mensen minder namaak luxeproducten (bv. handtas, horloge) gaan kopen als ze beter op de hoogte zijn van de persoonlijke en maatschappelijke gevolgen van het kopen en verkopen van namaakproducten? (7-punt Likert schaal gaande van ‘zeker niet’ tot ‘zeker wel’)
-‐
Hoe hoog moet de pakkans zijn opdat u geen namaak luxeproducten (bv. handtas, horloge) zou kopen? o kleiner dan 20 % o tussen 20 en 40 % o tussen 40 en 60 % o tussen 60 en 80 % o groter dan 80 %
SOCIO-DEMOGRAFISCHE GEGEVENS
Wat is uw leeftijd? (getal in cijfers, bv. 30) Beschikt u over een eigen inkomen? -‐
ja
-‐
nee
Tot welke van onderstaande groepen behoort u? -‐
student
-‐
werkend
-‐
werkloos
-‐
gepensioneerd
-‐
andere: …
Wat is uw nationaliteit?
84
ANNEX 3.1 WEDSTRIJDVRAAG
Wens je deel te nemen aan de wedstrijd die verbonden is aan dit onderzoek? Ter herinnering: Er zijn volgende prijzen te winnen: - een Fnac-bon t.w.v. €25 - een handtas van het merk BOO! --------- indien men kiest om deel te nemen, krijgt men de volgende vragen -------Wat is het meest gekochte luxe modemerk ter wereld? -‐
Calvin Klein
-‐
Diesel
-‐
Ralph Lauren
-‐
Chanel
Hoeveel procent van de deelnemers zal deze vraag correct beantwoorden? Naar welke prijs gaat jouw voorkeur? -‐
Fnac-bon t.w.v. €25
-‐
Een handtas van het merk BOO!
Wat is je emailadres waarop we je kunnen contacteren indien je een prijs gewonnen hebt? BEDANKING VOOR DEELNAME
Je hebt met succes de vragenlijst beëindigd. Nogmaals bedankt voor de deelname aan dit onderzoek! Klik op '>>' om uw antwoorden op te slagen. Daarna kan u uw browser sluiten.
85
ANNEX 3.2
ANNEX 3.2: Statistical analyses of main research
1. Manipulation check Message Type: Perceived Harm
Personal harm
Societal harm
N
Mean
Std. Deviation
Std. Error Mean
Societal harm manipulatiecheck
155
6,1183
,81273
,06528
Personal harm manipulatiecheck
155
5,2774
1,11323
,08942
Societal harm manipulatiecheck
163
6,1227
,86856
,06803
Personal harm manipulatiecheck
163
5,4376
1,06993
,08380
Message Type: Perceived Harm
Personal harm
Societal harm
Test Value = 4
Societal harm manipulatiecheck Personal harm manipulatiecheck Societal harm manipulatiecheck Personal harm manipulatiecheck
t
df
Sig. (2-tailed)
Mean Difference
95% Confidence Interval of the Difference
Lower
Upper
Lower
Upper
Lower
Upper
32,449
154
,000
2,11828
1,9893
2,2472
14,286
154
,000
1,27742
1,1008
1,4541
31,202
162
,000
2,12270
1,9884
2,2570
17,155
162
,000
1,43763
1,2721
1,6031
Remark: even if we did not average the item scores in the manipulation check for each construct, all means were significantly different from the Test Value (4).
86
ANNEX 3.2
87
ANNEX 3.2
2. ANOVA and Post Hoc test to evaluate possible differences in Ad credibility
Multiple Comparisons Dependent Variable: adcredibility
Bonferroni
(I) condities van 1 tot 8
(J) condities van 1 tot 8
1,00
2,00 3,00
-,006 ,077
,230 ,227
1,000 1,000
-,73 -,64
,72 ,79
4,00
,030
,224
1,000
-,68
,74
5,00
,089
,229
1,000
-,63
,81
6,00
,431
,229
1,000
-,29
1,15
7,00
,310
,232
1,000
-,42
1,04
8,00
,125
,213
1,000
-,55
,80
1,00
,006
,230
1,000
-,72
,73
3,00
,083 ,036
,230 ,228
1,000 1,000
-,64 -,68
,81 ,75
2,00
4,00
Mean Difference (I-J)
Std. Error
95% Confidence Interval
Sig.
Upper Bound
Lower Bound
88
ANNEX 3.2
3,00
5,00
,095
,232
1,000
-,63
,83
6,00
,437
,232
1,000
-,29
1,17
7,00
,316
,235
1,000
-,42
1,06
8,00
,131
,217
1,000
-,55
,81
1,00
-,077
,227
1,000
-,79
,64
2,00
-,083
,230
1,000
-,81
,64
4,00
-,047
,224
1,000
-,75
,66
5,00
,012 ,354 ,233
,229 ,229 ,232
1,000 1,000 1,000
-,71 -,37 -,50
,73 1,07 ,96
8,00
,048
,213
1,000
-,62
,72
1,00
-,030
,224
1,000
-,74
,68
2,00
-,036
,228
1,000
-,75
,68
3,00
,047
,224
1,000
-,66
,75
5,00
,059
,226
1,000
-,65
,77
6,00
,401
,226
1,000
-,31
1,11
7,00 8,00 1,00
,280 ,095 -,089
,229 ,210 ,229
1,000 1,000 1,000
-,44 -,57 -,81
1,00 ,76 ,63
2,00
-,095
,232
1,000
-,83
,63
3,00
-,012
,229
1,000
-,73
,71
4,00
-,059
,226
1,000
-,77
,65
6,00
,342
,230
1,000
-,38
1,07
7,00
,221
,233
1,000
-,51
,96
8,00
,036
,215
1,000
-,64
,71
1,00 2,00 3,00
-,431 -,437 -,354
,229 ,232 ,229
1,000 1,000 1,000
-1,15 -1,17 -1,07
,29 ,29 ,37
4,00
-,401
,226
1,000
-1,11
,31
5,00
-,342
,230
1,000
-1,07
,38
7,00
-,121
,233
1,000
-,86
,61
8,00
-,306
,215
1,000
-,98
,37
1,00
-,310
,232
1,000
-1,04
,42
2,00
-,316
,235
1,000
-1,06
,42
3,00
-,233 -,280 -,221
,232 ,229 ,233
1,000 1,000 1,000
-,96 -1,00 -,96
,50 ,44 ,51
6,00
,121
,233
1,000
-,61
,86
8,00
-,185
,218
1,000
-,87
,50
1,00
-,125
,213
1,000
-,80
,55
2,00
-,131
,217
1,000
-,81
,55
3,00
-,048
,213
1,000
-,72
,62
4,00
-,095
,210
1,000
-,76
,57
5,00 6,00 7,00
-,036 ,306 ,185
,215 ,215 ,218
1,000 1,000 1,000
-,71 -,37 -,50
,64 ,98 ,87
2,00
-,006
,193
1,000
-,63
,62
3,00
,077
,196
1,000
-,56
,71
4,00
,030
,226
1,000
-,70
,76
5,00
,089
,226
1,000
-,65
,82
6,00
,431
,203
,654
-,23
1,09
7,00
,310
,203
,980
-,35
,97
6,00 7,00 4,00
5,00
6,00
7,00
4,00 5,00
8,00
Tamhane
1,00
89
ANNEX 3.2 8,00 2,00
3,00
4,00
5,00
6,00
7,00
8,00
1,00 3,00 4,00
,125 ,006 ,083 ,036
,176 ,193 ,217 ,244
1,000 1,000 1,000 1,000
-,44 -,62 -,62 -,75
,69 ,63 ,79 ,83
5,00
,095
,244
1,000
-,70
,89
6,00
,437
,223
,789
-,29
1,16
7,00
,316
,223
,993
-,41
1,04
8,00
,131
,200
1,000
-,51
,78
1,00
-,077
,196
1,000
-,71
,56
2,00
-,083
,217
1,000
-,79
,62
4,00 5,00 6,00
-,047 ,012 ,354
,246 ,246 ,226
1,000 1,000 ,973
-,84 -,79 -,37
,75 ,81 1,08
7,00
,233
,225
1,000
-,50
,96
8,00
,048
,202
1,000
-,60
,70
1,00
-,030
,226
1,000
-,76
,70
2,00
-,036
,244
1,000
-,83
,75
3,00
,047
,246
1,000
-,75
,84
5,00
,059
,270
1,000
-,81
,93
6,00 7,00 8,00
,401 ,280 ,095
,252 ,252 ,231
,968 1,000 1,000
-,41 -,53 -,65
1,21 1,09 ,84
1,00
-,089
,226
1,000
-,82
,65
2,00
-,095
,244
1,000
-,89
,70
3,00
-,012
,246
1,000
-,81
,79
4,00
-,059
,270
1,000
-,93
,81
6,00
,342
,252
,996
-,47
1,16
7,00
,221
,252
1,000
-,59
1,04
8,00 1,00 2,00
,036 -,431 -,437
,231 ,203 ,223
1,000 ,654 ,789
-,71 -1,09 -1,16
,79 ,23 ,29
3,00
-,354
,226
,973
-1,08
,37
4,00
-,401
,252
,968
-1,21
,41
5,00
-,342
,252
,996
-1,16
,47
7,00
-,121
,231
1,000
-,87
,63
8,00
-,306
,209
,988
-,98
,37
1,00
-,310
,203
,980
-,97
,35
2,00 3,00 4,00
-,316 -,233 -,280
,223 ,225 ,252
,993 1,000 1,000
-1,04 -,96 -1,09
,41 ,50 ,53
5,00
-,221
,252
1,000
-1,04
,59
6,00
,121
,231
1,000
-,63
,87
8,00
-,185
,208
1,000
-,86
,49
1,00
-,125
,176
1,000
-,69
,44
2,00
-,131
,200
1,000
-,78
,51
3,00
-,048
,202
1,000
-,70
,60
4,00 5,00 6,00
-,095 -,036 ,306
,231 ,231 ,209
1,000 1,000 ,988
-,84 -,79 -,37
,65 ,71 ,98
7,00
,185
,208
1,000
-,49
,86
90
ANNEX 3.2
3. T-test for identifying differences in Message involvement according to message type Group Statistics
message involvement
Message Type: Perceived Harm personal harm
N
societal harm
Mean
Std. Deviation
Std. Error Mean
156
4,5150
,91314
,07311
163
5,0409
,89045
,06975
4. T-test for evaluating differences in attitude depending on the message one has been exposed to. Group Statistics
attitude
Message Type: Perceived Harm personal harm societal harm
N
Mean
Std. Deviation
Std. Error Mean
156
3,4588
1,02091
,08174
163
3,1691
1,07530
,08422
91
ANNEX 3.2
5. ANOVA and Post Hoc test for evaluating differences in attitude depending on the group one finds him/herself in.
Multiple Comparisons Dependent Variable: attitude
Bonferroni
(I) Groep Student
Werkend
(J) Groep Werkend Werkloos
Std. Error ,12766 ,71803
Sig. ,000 1,000
Upper Bound ,2203 -1,5915
Lower Bound ,9422 2,4683
Gepensioneerd
2,12888(*)
,58760
,003
,4677
3,7901
Andere:
,79554
,38814
,412
-,3018
1,8928
Student
-,58126(*)
,12766
,000
-,9422
-,2203
Gepensioneerd Andere:
-,14286 1,54762 ,21429
,72283 ,59346 ,39696
1,000 ,095 1,000
-2,1864 -,1301 -,9079
1,9006 3,2254 1,3365
Student
-,43840
,71803
1,000
-2,4683
1,5915
Werkend
,14286
,72283
1,000
-1,9006
2,1864
1,69048
,92275
,679
-,9182
4,2991
Student Werkend
,35714 -2,12888(*) -1,54762
,81046 ,58760 ,59346
1,000 ,003 ,095
-1,9341 -3,7901 -3,2254
2,6484 -,4677 ,1301
Werkloos
-1,69048
,92275
,679
-4,2991
,9182
Andere:
-1,33333
,69753
,569
-3,3053
,6386
Werkloos
Werkloos
Gepensioneerd Andere: Gepensioneerd
95% Confidence Interval
Mean Difference (I-J) ,58126(*) ,43840
92
ANNEX 3.2 Andere:
Student
-,79554
,38814
,412
-1,8928
,3018
Werkend
-,21429 -,35714 1,33333
,39696 ,81046 ,69753
1,000 1,000 ,569
-1,3365 -2,6484 -,6386
,9079 1,9341 3,3053
Werkloos Gepensioneerd Tamhane
Student
Werkend
,58126(*)
,13796
,000
,1886
,9739
Werkloos
,43840
1,21594
1,000
-147,1139
147,9907
2,12888
,21701
,054
-,0683
4,3261
,79554 -,58126(*) -,14286 1,54762(*)
,53645 ,13796 1,22046 ,24105
,874 ,000 1,000 ,041
-1,4781 -,9739 -138,5450 ,0879
3,0692 -,1886 138,2593 3,0073
Gepensioneerd Andere: Werkend
Werkloos
Student Werkloos Gepensioneerd Andere:
,21429
,54662
1,000
-2,0256
2,4542
Student
-,43840
1,21594
1,000
-147,9907
147,1139
Werkend
,14286
1,22046
1,000
-138,2593
138,5450
1,69048 ,35714 -2,12888
1,23190 1,32599 ,21701
,993 1,000 ,054
-116,9428 -43,7911 -4,3261
120,3238 44,5054 ,0683
Werkend
-1,54762(*)
,24105
,041
-3,0073
-,0879
Werkloos
-1,69048
1,23190
,993
-120,3238
116,9428
Andere:
-1,33333
,57171
,403
-3,5722
,9055
Student Werkend Werkloos
-,79554 -,21429 -,35714
,53645 ,54662 1,32599
,874 1,000 1,000
-3,0692 -2,4542 -44,5054
1,4781 2,0256 43,7911
Gepensioneerd
1,33333
,57171
,403
-,9055
3,5722
Gepensioneerd Gepensioneerd
Andere:
Andere: Student
* The mean difference is significant at the .05 level.
93
ANNEX 4
ANNEX 4: Explorative questions An example of the statistical analyses performed to assess the answers to these questions is showed below. Other questions are analysed analogous. Descriptive Statistics N
Minimum
Maximum
Mean
Std. Deviation
exploratief: verkopen illegaal?
318
1
7
6,06
1,448
exploratief: verkopen bestraft?
318
1
7
5,65
1,392
exploratief: bewust kopen bestraft?
318
1
7
3,90
1,873
exploratief: consumentkentgevolgen?
318
1
7
2,53
1,610
exploratief: mindernamaakalsgevolgge kend?
318
1
7
4,85
1,634
exploratief: hoogte pakkans? (1 tot 5)
318
1
5
2,75
1,373
Valid N (listwise)
318
Is the selling of counterfeit luxury fashion items an illegal business practice? Group Statistics
exploratief: verkopen illegaal?
Wat is uw geslacht? Man Vrouw
N
Mean
Std. Deviation
Std. Error Mean
130
6,18
1,355
,119
188
5,98
1,507
,110
94
ANNEX 4
Test of Homogeneity of Variances exploratief: verkopen illegaal? Levene Statistic
df1
2,506
df2 4
Sig. 313
,042
ANOVA exploratief: verkopen illegaal? Sum of Squares Between Groups
df
Mean Square
F
10,768
4
2,692
Within Groups
653,974
313
2,089
Total
664,742
317
Sig.
1,288
,274
Multiple Comparisons Dependent Variable: exploratief: verkopen illegaal? 95% Confidence Interval
Bonferroni
(I) Groep Student
Werkend
Mean Difference (IJ) -,223 -,995
Std. Error ,183 1,027
Sig. 1,000 1,000
Upper Bound -,74 -3,90
Lower Bound ,29 1,91
Gepensioneerd
1,338
,840
1,000
-1,04
3,71
Andere:
-,138
,555
1,000
-1,71
1,43
Student
,223
,183
1,000
-,29
,74
-,773 1,561 ,084
1,034 ,849 ,568
1,000 ,669 1,000
-3,69 -,84 -1,52
2,15 3,96 1,69
,995
1,027
1,000
-1,91
3,90
(J) Groep Werkend Werkloos
Werkloos Gepensioneerd Andere: Werkloos
Student Werkend
,773
1,034
1,000
-2,15
3,69
2,333
1,320
,780
-1,40
6,06
Student Werkend
,857 -1,338 -1,561
1,159 ,840 ,849
1,000 1,000 ,669
-2,42 -3,71 -3,96
4,13 1,04 ,84
Werkloos
-2,333
1,320
,780
-6,06
1,40
Andere:
-1,476
,997
1,000
-4,30
1,34
Student
,138
,555
1,000
-1,43
1,71
Werkend
-,084
,568
1,000
-1,69
1,52
Gepensioneerd Andere: Gepensioneerd
Andere:
95
ANNEX 4 Werkloos Gepensioneerd Tamhane
Student
-,857 1,476
1,159 ,997
1,000 1,000
Werkend
-,223
,175
,899
-,72
,27
Werkloos
-,995(*)
,096
,000
-1,27
-,72
1,338
1,858
1,000
-24,27
26,94
-,138 ,223 -,773(*) 1,561
,863 ,175 ,146 1,862
1,000 ,899 ,000 ,999
-3,80 -,27 -1,19 -23,74
3,52 ,72 -,35 26,86
Andere:
,084
,870
1,000
-3,55
3,72
Student
,995(*)
,096
,000
,72
1,27
Werkend
,773(*)
,146
,000
,35
1,19
Gepensioneerd Andere: Student
2,333 ,857 -1,338
1,856 ,857 1,858
,983 ,988 1,000
-23,51 -2,82 -26,94
28,18 4,54 24,27
Werkend
-1,561
1,862
,999
-26,86
23,74
Werkloos
-2,333
1,856
,983
-28,18
23,51
Andere:
-1,476
2,044
,999
-17,22
14,27
Student Werkend Werkloos
,138 -,084 -,857
,863 ,870 ,857
1,000 1,000 ,988
-3,52 -3,72 -4,54
3,80 3,55 2,82
Gepensioneerd
1,476
2,044
,999
-14,27
17,22
Gepensioneerd Andere: Werkend
Werkloos
Gepensioneerd
Andere:
Student Werkloos Gepensioneerd
-4,13 -1,34
2,42 4,30
* The mean difference is significant at the .05 level.
96
