Questionnaire Design for Scale Development in Social Sciences: Use of Exploratory (SPSS) and Confirmatory (EQS) Factor Analysis
Keywords:
questionnaire design, scale development, social sciences, exploratory factor analysis, confirmatory factor analysis, eqsSynopsis
Research is a permanent activity in the cycle of analysis, field testing, and synthesis. It requires keen discernment from the researcher to evaluate the sources of information and resources utilized. Moreover, gaining recognition and credibility in scientific research results depends on the strategy that best clarifies and justifies the measurement technique used. Various strategies can be proposed to develop and refine measurements; however, their true impact depends on the type of scientific phenomenon being measured, both with directly observable variables and their underlying relationships.
Thus, based on the focus and scope of your research concerning the broad range of disciplines within social sciences, you should center your interest on the formulation, design, development, and validation of constructs, both of direct and underlying relationships, to theoretically explain reality.
Particularly, this latter type of relationships involves constructs that, representing abstractions, can only be evaluated indirectly. Indirect evaluation implies the design and use of multiple elements (in our case, indicators) that measure the construct, i.e., form the measurement scale. Therefore, starting from a specific research problem and characterizing the factors, variables, and indicators that best describe it in the form of constructs, the main objective of this document is to propose and design the relationships of these factors, variables, and indicators to discuss how to measure and validate them through the creation of scales.
To achieve this, the reader is provided with:
BACKGROUND. RESEARCH AND SOURCES OF INFORMATION. This section offers brief backgrounds on the research idea, its approaches and scopes, problem statement, objectives, and how to build the theoretical framework. Recommendations are given for better data capture. There is a special emphasis on describing the sources of information, how to handle them according to the type of research, and various characteristics and conditions of instruments such as panel designs, interviews, questionnaires, etc.
CHAPTER 1. THE IMPORTANCE OF MEASUREMENTS IN RELATION TO SOCIAL SCIENCES This chapter justifies and emphasizes the importance of scale development theory by explaining key and basic concepts such as dimensionality, reliability, and validity. The importance of the four-stage approach (construct; content; exploratory factor analysis and confirmatory factor analysis) in developing a scale that involves these three concepts is highlighted.
CHAPTER 2. DIMENSIONALITY. This chapter clarifies how this stage defines the scope of the construct and the final model, to understand two basic techniques in all social science research: exploratory factor analysis and confirmatory factor analysis.
CHAPTER 3. RELIABILITY. This section explains the importance of the concept and the different types of reliability, the implication of internal consistency, reliability, and composite variance in scale design.
CHAPTER 4. VALIDITY. Here, the impact of the validity of the instrument to be designed is discussed, highlighting different types, with an emphasis on the importance of discriminant and nomological validity for confirmatory factor analysis.
CHAPTER 5. DEFINING THE CONSTRUCT AND ITS CONTENT. This section begins with the four-stage approach, providing extensive recommendations on the importance of designing a broad and sufficient theoretical framework that supports scale design through the creation of indicators.
CHAPTER 6. DESIGNING THE SCALE. EXPLORATORY FACTOR ANALYSIS (Stage 3). This technique allows the first reductions of variables and their indicators in the construct and final model. It is explained through problems based on the use of the statistical software SPSS for better understanding.
CHAPTER 7. FINALIZING THE SCALE. CONFIRMATORY FACTOR ANALYSIS AND STRUCTURAL EQUATION MODELING. This is the last design stage and accounts for the use of structural analysis software EQS 6.2. The objective is to provide a clear and concise demonstration of how to input and interpret the results provided by this tool to confirm the model and identify underlying relationships that the theoretical framework did not consider.
Finally, it should be noted that scale design based on constructs that ultimately define a model is the foundation of scientific research. These processes are more iterative than linear. This means that, rather than following linear or consecutive steps and activities as in any logical and sequential process, the suggested scale development process tends to be iterative and continuous, with all scale creation procedures being restarted. This is because conscientious researchers learn from their efforts and mistakes in the field of social sciences, making it necessary to revise, including early stages like defining factors, variables, and indicators at the conceptual level, as well as redefining the dimensionality, reliability, and validity of the construct
Downloads
References
Mejía-Trejo, J., Sánchez-Gutiérrez, J., y Vázquez-Ávila, G. (2014). Culture of Competitiveness in the New Global Context. In J. Sánchez-Gutiérrez, J. Vargas-Barraza, y J. Mejía-Trejo, Modeling MLQ5X for Innovation and Value Creation(Vol. 1, pp. 1-18). Guadalajara, Jalisco: Universidad de Guadalajara.
Messick, S. (1993). Validity. In R. L. Linn (Ed.), Educational measurement (2nd ed., pp. 105-146). Phoenix, AZ: American Council on Education and the Oryx Press.
Mick, D. G. (1996). Are studies of dark side variables confounded by socially desir-able responding? The case of materialism. Journal of Consumer Research, 23(September), 106-119.
Miles, J. yShevlin, M. (1998). Effects of sample size, model specification and factor loadings on the GFI in confirmatory factor analysis. Personality and Individual Differences, 25, 85-90.
Mulaik, S. A. (1990), Blurring the Distinction Between Component Analysis and Cornrnon Factor Analysis. Multivariate Behavioral Research 25: 53-59.
Mulaik, S. A., James, L. R., Van Alstine, J., Bennet, N., Lind, S. y Stilwell, C. D. (1989). Evaluation of goodness-of-fit indices for structural equation models. Psychological Bulletin, 105(3), 430-45.
Mulaik, S. A., y McDonald R. P. (1978), The Effect of Additional Variables on Factor Indeterminancy in Models with a Single Cornrnon Factor. Psychometrika 43: 177-92.
Nederhof, A. (1985). Methods of coping with social desirability bias: A review. European Journal of Social Psychology, 15(July-September), 263-280.
Netemeyer, R.G., Bearden, W.O, Sharma, S. (2003) Scaling Procedures.Issues and Applications.USA: Sage Publications
Netemeyer, R. G., Pullig, C., y Bearden, W. O. (2002). Observations on some key psychometric properties of paper-and-pencil measures. In A. G. Woodside y E. M. Moore (Eds.), Essays by distinguished marketing scholars of the Society for Marketing Advances (pp. 115-138). Amsterdam: JAI.
Netemeyer, R. G., Boles, J. S., y McMurrian, R. C. (1996). Development and valida-tion of Work-Family Conflict and Family-Work Conflict scales. Journal of Applied Psychology, 81(4), 400-410.
Netemeyer, R. G., Durvasula, S., y Lichtenstein, D. R. (1991). A cross-national assessment of the reliability and validity of the CETSCALE. Journal of Marketing Research, 28(August), 320-327.
Neuberg, S. L., West, S. G., Thompson, M. M., y Judice, T. N. (1997). On dimensionality, discriminant validity, and the role of psychometric analyses in personality theory and measurement: Reply to Kruglanski et al.’s (1997) defense of the Need for Closure Scale. Journal of Personality and Social Psychology, 73, pp.1017-1029. Nevo, B. (1985). Face validity revisited. Journal of Educational Measurement, 22, 287-293.
Downloads
Published
Categories
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.