DASAR – DASAR RISET
Disajikan oleh: Dr. Werner R. Murhadi Jurusan Manajemen Fakultas Bisnis & Ekonomi Universitas Surabaya (UBAYA) Dr. Werner R. Murhadi
Filsafat Ringkas 1. Epistemologi (yunani: episcmc = knowledge/pengetahuan dan logos = teori) yang pertama kali digunakan oleh J.F. Ferrier
(1854) --------- teori tentang pengetahuan atau the nature of knowledge 2. Ontologi (yunani: on = being/wujud atau apa dan logos = teori) -
--------- teori tentang perwujudan/empiris atau the nature of reality 3. Axiologi (yunani: axios = nilai dan logos = teori) ---------- teori
kemanfaatan 4. Metodologi --------- teori tentang cara/metode atau the nature of how one comes to know Dr. Werner R. Murhadi
METODE PENELITIAN PADA DASARNYA MERUPAKAN CARA ILMIAH UNTUK MENDAPATKAN INFORMASI DENGAN TUJUAN DAN KEGUNAAN TERTENTU CARA ILMIAH DIDASARKAN PADA CIRI-CIRI KEILMUAN:
RASIONAL EMPIRIS SISTEMATIS Dr. Werner R. Murhadi
The Hallmarks of Scientific Research 1. 2. 3. 4. 5. 6. 7. 8.
Purposiveness Rigor Testability Replicability Precision and confidence Objectivity Generalizability Parsimony
Dr. Werner R. Murhadi
Types of research From the viewpoint of
Application
Pure research
Applied research
Type of information sought
Objectives
Descriptive research
Exploratory research
Correlational research
Explanatory research
Dr. Werner R. Murhadi
Quantitative research
Qualitative research
Klasifikasi Riset: How research is used? Basic
Applied
Research is intrinsically satisfying and judgments are by other sociologist
Research is part of a job and is judged by sponsors who are outside the discipline of sociology
Research problem and subject are selected with a great deal of freedom
Research problem are “narrowly constrained” to the demand of employers or sponsors
Research is judged by absolute norm of scientific rigor, & the highest standard of scholarship are sought
The rigor & standard of scholarship depend on the use of result. Research can be “quick and dirty” or may match high scientific standard
The primary concern is with the internal logic and rigor of research design
The primary concern is with the ability to generalize findings to areas of interest to sponsor
The driving goal is to contribute to basic, theoretical knowledge
The driving goal is to have practical payoffs or uses for results
Success come when result appear in a scholarly journal & have an impact on others in the scientific community
Success come when results are used by sponsor in decision making
Dr. Werner R. Murhadi
Klasifikasi Riset: Purposes of the study Research design
Exploratory Research
Conclusive Research
Descriptive
Cross sectional
Single Cross Sectional
Causal
Longitudinal
Multiple Cross Sectional
Dr. Werner R. Murhadi
Perbandingan Desain Dasar Riset Exploratory
Descriptive
Causal
Objective
Discover idea and insight
Describe market characteristic or function
Determine cause and effect relationship
Characteristics
Flexible Versatile Often the front end of total research design
Marked by the prior formulation of specific hypothesis Preplanned and structured design
Manipulation of one or more independent variables Control of other mediating variables
Methods
Experts survey Pilot survey Secondary data
Secondary data Survey Panels Observational & Other data
Experiments
Qualitative research
Dr. Werner R. Murhadi
Perbedaan Qualitative dan Quantitative Qualitative Research
Quantitative Research
Objective
To gain a qualitative understanding of the underlying reasons and motivations
To quantify the data and generalize the results from the sample to the populations of interest
Sample
Small number of nonrepresentative cases
Large number of representative cases
Data Collection
Unstructured
Structured
Data Analysis
Nonstatistical
Statistical
Outcome
Develop an initial understanding
Recommend a final course of action
Dr. Werner R. Murhadi
Exploratory research design Klasifikasi prosedur riset kualitatif Qualitative Research Procedures
Direct (Nondisguised)
Focus Group
Nondirect (Disguised)
Depth Interview
Projective Techniques
Association Techniques
Completion Techniques
Dr. Werner R. Murhadi
Construction Techniques
Expressive Techniques
Descriptive research design Survey dan Observasi Survey Method
Telephone Interview
Personal Interview
Mail Interview
Electronic Interview
Observation Method
Personal Observation Mechanical Observation
Audit
Dr. Werner R. Murhadi
Content Analysis
Trace Analysis
Causal research design Experimen Experimental Design
Pre-Experimental
True-Experimental
One Shot Case Study
Pretest-posttest Control Group
Quasi or Field Exp.
Time Series One Group Pretest-posttest
Post test only Control Group
Static Group
Solomon Four Group
Multiple Time Series
Dr. Werner R. Murhadi
Statistical
Randomized Blocks Latin Square Factorial
Sumber Error dalam Penelitian Total Error
Random Sampling Error, occur because the particular sample selected is an imperfect representation of the population of interest
Researcher Error
Surrogate Information Error Measurement Error Population Definition Error Sampling Frame error Data Analysis error
Nonsampling Error
Response Error
Interviewer Error
Respondent Selection Error Questioning Error Recording Error Cheating error
Dr. Werner R. Murhadi
Nonresponse Error
Respondent Error
Inability Error Unwillingness Error
Klasifikasi Teknik Sampling Sampling technique
Nonprobability
Convenience/ Haphazard/ Accidental
Judgmental/ Purposive
Simple Random
Probability
Quota
Snowball
Systematic
Stratified
Proportionate
Dr. Werner R. Murhadi
Cluster
Disproportionate
Others
Kapan mempergunakan Nonprobability sampling VS Probability sampling?
Factors
Nature of researc
Conditions Favoring The use of Nonprobability
Probability
Exploratory
Conclusive
Relative magnitude of Non sampling error are sampling and non sampling larger errors
Sampling error are large
Variability in the population
Homogeneous (Low)
Heterogeneous (High)
Statistical considerations
Unfavorable
Favorable
Operational consideration
Favorable
Unfavorable
Dr. Werner R. Murhadi
Sinonim: Independent dan Dependent Independent Variable
Dependent Variable
Explanatory
Explained
Regressor
Regressand
Exogenous
Endogenous
Covariate (Metric) Factor (Nonmetric-katagorical)
Outcome
Predictor
Predictand
Predictor
Criterion
Stimulus
Response
Antecedent
Consequence
Control Variable
Controlled Variable
Predicted From
Predicted To
Presumed Cause
Presumed Effect
Manipulated
Dr. Werner R. Murhadi
Measured Outcome
JENIS VARIABEL BERDASARKAN KEDUDUKANNYA DI DALAM HIPOTESIS Antecedent
Intervening/ Mediating
Moderating Independent
Dependent
Intraneous Extraneous
Concomitant
CONFOUNDING
Dr. Werner R. Murhadi
Control
Keterangan: Variabel Intraneous merupakan variabel yg tercakup dalam riset.
V. Extraneous merupakan variabel yang tidak tercakup dalam riset, dimana sifatnya benalu yang dapat menggerogoti informasi. Variabel Intervening merupakan variabel yang bersifat menjadi
perantara (mediating) dari hubungan variabel bebas ke variabel dependen. Contoh:* Kualitas Produk Sales Profit dalam hal ini sales merupakan jembatan/pengantara. Artinya tidak akan ada profit bila tidak ada sales. * Pendidikan
Gaji
Dr. Werner R. Murhadi
Kinerja
Keterangan: Variabel Moderator merupakan variabel yg bersifat
memperkuat/memperlemah pengaruh variabel bebas terhadap var. dependen. Cth: Kualitas Produk
Sales
Promosi Dimana ada/tidak promosi, sales tetap ada, namun banyak/sedikitnya tergantung pada promosi yg dilakukan. V. Confounding (Pembaur) merupakan var. yg tidak tercakup dalam hipotesisi, akan tetapi muncul dalam riset dan berpengaruh terhadap var. dependen. Pengaruhnya mencapuri atau berbaur dengan
variabel bebas Dr. Werner R. Murhadi
Keterangan: Variabel Control (Kendali) merupakan variabel pembaur yang
pengaruhnya dapat dikendalikan. Pengendalian dapat dilakukan dengan cara blocking dengan membuat kriteria ekslusi-inklusi, yaitu dengan mengeluarkan objek yang tidak memenuhi kriteria (ekslusi) dan mengambil objek yang memenuhi kriteria (inklusi). Variabel Concomitant (Penyerta) merupakan variabel pembaur yang tidak dapat dikendalikan, sehingga tetap menyertai dalam proses riset. Konsekuensinya: data tersebut harus diamati. Pengaruh baurnya dihilangkan pada tahap analisis data.
Dr. Werner R. Murhadi
Beberapa Alat Multivariat Var.Dependent Skala
Var.Independent
Jumlah
Skala
Alat Jumlah
metrik
1
Metrik Nonmetrik
>1 >1
Regresi
metrik
1
Nonmetrik
>1
ANOVA
metrik
>1
Nonmetrik
>1
MANOVA
metrik
1
Metrik (covariate) Nonmetrik (Factor)
>1 >1
ANCOVA
Nonmetrik
1
Metrik
>1
Discriminant anaylsis
Nonmetrik or Metrik
1
Nonmetrik
>1
Conjoint Analysis
Nonmetrik Metrik
>1
Nonmetrik Metrik
>1
Canonical Correlation
Metrik *
>1
Nonmetrik Metrik
>1
Structural Equation Model
Dr. Werner R. Murhadi
Teknik Multivariate (Hair et.al., 2006) What type of relationship Dependence
Interdependence
How many variable are being predicted
Multiple Dependent & independence
Several Dependent
What’s scale of the Dependent variable?
SEM
Metric
Non-Metric
Is the structure of relationship among:
One Dependent
What’s scale of the Dependent variable?
Metric
Variables
Factor Analysis
CFA
Case/Resp.
Cluster Analysis
Object
How are the Attribute Measure?
Non-Metric Metric
What’s scale of the Independent variable?
Metric
Canonical Correl. An.
Non-Metric
Non-Metric
Non-Metric
Canonical with Dummy var. Regression & Conjoint An.
MDA & Linear Prob.
MANOVA Dr. Werner R. Murhadi
MDS
Correspondence Analysis
Validity dan Reliability Reliability means dependability or consistency. It suggests that the same thing is repeated or recur under the identical or very similar conditions. Validity is the extent to which a test measures what we actually wish to measure. Summary of measurement reliability and validity type Reliability (Consistency)
Validity (True Measure)
Stability – over time
Face – in the judgment of others
Representative – across subgroups
Content – capture the entire meaning
Equivalence – across indicators
Criterion – agree with external sources: * Concurrent – agree with a pre-existing measure * Predictive – agree with future behavior Construct – multiple indicator are consistent * Convergent – alike one are similar * Discriminant – different ones differ Dr. Werner R. Murhadi
Other uses of the terms Reliable and Valid • Internal Validity : there are no error internal to the design of the research project. • External Validity : it is the ability to generalize finding from a specific setting and small group to a broad range of setting and people. • Statistical Validity : the correct statistical procedure is chosen and its assumptions are fully met.
Dr. Werner R. Murhadi
Relationship between Validity and Reliability
Low Reliability & Low Validity
High reliability but Low validity
Dr. Werner R. Murhadi
High reliability & High validity
Summary of Validity Estimate Type
What is Measured
Methods
Content
Degree to which the content of the items adequately represents the universe of all relevant items under study
• Judgmental • Panel evaluation with content validity ratio
Criterion – Related
Degree to which the predictor is adequate in capturing the relevant aspect of the criterion. Description of the present; criterion data are available at the same time as predictor scores Prediction of the future; criterion data are measured after the passage of time
• Correlation
Answer the question, “what accounts for the variance in the measure?”; attempts to identify the underlying construct being measured and determine how well the test represent it.
• Judgmental • Correlation of proposed test with established one • Convergent – discriminant techniques • Factor analysis • Multitrait – multimethod analysis
Concurrent
Predictive Construct
Dr. Werner R. Murhadi
• Correlation
• Correlation
Summary of Reliability Estimate Type
Coefficient
What is Measured
Methods
Test-Retest
Stability
Reliability of a test or instrument inferred from examinee scores; same test is administered twice to same subject over an interval of less than six months
•Correlation
Parallel Form
Equivalence
Degree to which alternative forms of the same measure produce same of similar result; administered simultaneously or with a delay. Inter rater estimates of the similarity of judges observations or scores.
•Correlation
Split Half, KR20, Cronbach’s Alpha
Internal Consistency
Degree to which instrument items are homogeneous and reflect the same underlying construct
•Specialized correlations formula
Dr. Werner R. Murhadi
Aliran Riset (Neuman, 2003, p.91) Positivism
Interpretive
Critical
Feminist
Postmodern
Reason for research
To discover natural laws so people can predict and control
To understand & describe meaningful social action
To smash myths & empower people to change society radically
To smash myths & empower people to advance value of nurturing other & equality
To express the subjective self, to be playful, & to entertain & stimulate other people
Nature of social reality
Stable preexisting patterns or order that can be discovered
Fluid definitions of a situation created by human interaction
Conflict filled & governed by hidden underlying structures
Conflict filled, structured power relations that keep many people oppressed
Chaotic & fluid without any real patterns or master plan
Nature of human beings
Self interested & rational individual who are shaped by external forces
Social being who create meaning & who constantly make sense of their world
Creative, adaptive people with unrealized potential, trapped by illusion & exploitation
Creative, gendered being with unrealized potential who are often trapped by unseen forces
Creative, dynamic beings with unrealized potential
Role of common sense
Clearly distinct from & less valid than science
Powerful every day theories used by ordinary people
False beliefs that hide power & objective condition
False beliefs that hide power & objective conditions
The essence of social reality that is superior to scientific or bureaucratic form of reasoning
Theory looks like
A logical deductive system of interconnected definition, axiom, & law
A description of how a groups meaning system is generated & sustained
A critique that reveals true condition & help people see the way to a better world
A critique that reveals true conditions & help people see the way to a better world
A performance or work of artistic expression that can amuse, shock or stimulate others
An explanatio n that is true
Is logically connected to laws & based on fact
Resonates or feels right to those who are being studied
Supplies people with tools needed to change the world
Supplies ideas/tools to help liberate people from oppressive relations
No one explanation is more true; all are true for those who accept them
Good evidence
Is based on precise observations that other can repeat
Is embedded in the context of fluid social interactions
Is informed by a theory that unveils illusions
Is informed by theory that unveils illusion
Has aesthetic properties & resonates with people inner feeling/emotions
Place for values
Science is value free, & value have no place except when choosing a topic
Value are an integral part of social life; no group values are wrong, only different.
All science must begin with a value position, some position are right, some are wrong
Value are essential to research, & feminist ones are clearly preferred
Value are integral to research, but all value positions are equal.
Dr. Werner R. Murhadi