Common Method Biases in Behavioral Research: A Critical Review of the

Literature and Recommended Remedies

Philip M. Podsakoff, Scott B. MacKenzie, Jeong-Yeon Lee, Nathan P. Podsakoff

Journal of Applied Psychology, 2003, Vol 88, No.5, 879-903

Advanced Business Research MethodIntructor : Prof. Feng-Hui Huang

Agung D. BuchdadiDA21G201

1. Introduction2. Extent of Bias Caused by Common Method Variance3. Potential Sources of Common Method Biases4. Processes Through Which Method Biases Influence

Respondent Behavior5. Techniques for Controlling Common Method Biases6. Comparison of Statistical Remedies for Common

Method Biases7. Recommendations for Controlling Method Biases in

Research Settings8. Some Additional Considerations9. Conclusions

Contents

Interest in the problem of method biases has a long history in the behavioral sciences.

The purpose of this article is to examine the extent to which method biases influence behavioral research results, identify potential sources of method biases, discuss the cognitive processes through which method biases influence responses to measures, evaluate the many different procedural and statistical techniques that can be used to control method biases, and provide recommendations for how to select appropriate procedural and statistical remedies for different types of research settings.

Introduction

Cote and Buckley (1987) found that approximately one quarter (26.3%) of the variance in a typical research measure might be due to systematic sources of measurement error like common method biases.

However, they also found that the amount of variance attributable to method biases varied considerably by discipline and by the type of construct being investigated.

For example, Cote and Buckley (1987) found that, on average, method variance was lowest in the field of marketing (15.8%) and highest in the field of education (30.5%).

They also found that typical job performance measures contained an average of 22.5% method variance, whereas attitude measures contain an average of 40.7%.

It is also noted that the CMV on controlled research is lower than CMV on uncontrolled research

Extent of Bias Caused by Common Method Variance

Cote and Buckley (1987) indicate that CMV can inflate or deflate the relationship between construct, as shown in table 1

Extent of Bias Caused by Common Method Variance

Method effects produced by a common source or rater

Potential Sources of Common Method Biases

Method effect produced by items context

Potential Sources of Common Method Biases

Method effect produced by measurement context

Potential Sources of Common Method Biases

Processes Through Which Method Biases Influence Respondent Behavior

Procedural Remedies1. Obtain measures of the predictor and

criterion variables from different sources2. Temporal, proximal, psychological, or

methodological separation of measurementa. Reducing biases in retrieval stageb. Reducing respondent ability to re”use” previous

answerc. Reducing biases in the report stage

3. Protecting respondent anonymity and reducing evaluation apprehension

Techniques for Controlling Common Method Biases

4. Counterbalancing question order5. Improving scale items

a. Define ambiguous or unfamiliar termsb. Avoid vague concepts and provide examples when

such concepts must be used; c. Keep questions simple, specific, and concise; d. Avoid double-barreled questions; e. Decompose questions relating to more than one

possibility into simpler, more focused questions; f. Avoid complicated syntax. g. eliminate item social desirability and demand

characteristics

Techniques for Controlling Common Method Biases

Statistical Remedies1. Herman’s single-factor test

Techniques for Controlling Common Method Biases

Description Example

Include all items from all of the construct in the study into a factor analysis to determine whether the majority of the variance can be accounted for by one general factor

Statistical Remedies2. Partial corellation procedure

Techniques for Controlling Common Method Biases

Description Example

Partialling out {social desirability or affectivity; an unrelated “marker variable”; a general method factor} as a surrogate for method variance

Statistical Remedies3. Controlling for the effect of a directly

measured latent method factor

Techniques for Controlling Common Method Biases

Statistical remedies4. Controlling for the effects of an

unmeasured latent method factor

Techniques for Controlling Common Method Biases

5. Multiple method factorsa. CFA of MTMM model

Techniques for Controlling Common Method Biases

5. Multiple method factorsb. Correlated uniqueness model

Techniques for Controlling Common Method Biases

5. Multiple method factorsc. Direct product model

Techniques for Controlling Common Method Biases

Comparison of Statistical Remedies for Common Method Biases

Comparison of Statistical Remedies for Common Method Biases

Recommendations for Controlling Method Biases in Research Settings

Controlling for Method Variance in Experimental Research Examining Mediated Effects

Method biases contribute to observed relationship between the mediator and the dependent measure in this research as it is usually obtained from the same object at the same point

Using single-common-method-factor approach

Some Additional Considerations

Controlling for Method Variance in Studies Using Formative Constructs

Since in formative construct the characteristic of indicators are different from those in reflective construct, To controll method bias, the researcher must be more careful in designing the research and the procedural controlls, then, are the best way in this matters.

Some Additional Considerations

This paper provides the process to controll method bias in conducting research in behavior. The process starts by assesing the research setting to identify the potential sources of bias and then implementing both procedural and statistical methods of control.

Conclusions