Construct validity

Construct (engl. Construct validity ) referred to in the multivariate statistical one aspect of validity and is present when the measurement of a construct is not corrupted by systematic errors or by other constructs. To a certain extent, it is about the question of whether the indicators chosen to measure the construct behave in a way that allows them to be understood collectively as an “intellectual whole”.

target

The construct validation pursues the goal of analyzing a test psychologically and is thus in contrast to the criteria- related validity . The focus of construct validity is on the theoretical clarification of what a test measures. For example, an intelligence test or an ability test records assumed properties or constructs.

This is a derived complex unit that cannot be directly grasped operationally. Due to the lack of operational availability, construct validation is associated with considerable difficulties and great technical and economic effort. Thereby theory and empiricism are in an interactive relationship to one another.

The test and the construct also interact with one another, so that the test can change the construct and the construct can influence the structure of the test. Construct validity is therefore anchored to a much higher degree than criteria-related and logical validity in basic research on personality psychology.

history

The concepts of convergence and discriminant validity as sub-aspects of construct validity were introduced by Campbell and Fiske (1959). Since then, further aspects of the construct validity have been suggested.

Finding

Construct validity is given if content validity , convergent validity , discriminant and nomological validity can be determined and methods distortion ( common-method bias ) can be excluded. While content validity closes the gap between a conceptual-theoretical construct and its measurement using a scale made up of indicators , convergence and discriminant validity are often determined using objective, statistically measurable key figures. Content, convergence and discriminant validity can also be improved with the help of jurors. Such a subjective procedure, in which jurors are to assign indicators noted on index cards, on the one hand to self-chosen and self-named categories (and therefore to constructs) and on the other hand to predefined categories, was presented by Moore and Benbasat to improve construct validity.

criticism

The mere consideration of convergence and discriminant validity to determine construct validity is mainly criticized by John R. Rossiter , who argues that the construct validity must be achieved independently of other constructs. He emphasizes the importance of content validity and even equates it with construct validity. Measures to improve discriminant and convergence validity can lead to indicators being removed and the statistically measurable properties of the measurement models being improved, while at the same time the measurement models move away from the semantic content of their constructs. The indicators circle a construct, so to speak, and the one-sided removal of indicators to improve statistical key figures may remove the construct from the measurement. John G. Wacker (2004) emphasizes the importance of formal conceptual definitions as the most important step before performing any traditional statistical validity test. He describes such definitions as a necessary condition for construct validity, while statistical tests are sufficient conditions. Overall, it can be stated that in the past, measures to define a construct and, in particular, to improve the content validity were often not given the necessary attention, while indicators to improve purely objective statistical quality criteria such as Cronbach's alpha or the model quality of a structural equation model were often premature at the expense of construct validity have been deleted.

Web links

Handbook of Management Scales , contains a set of scales for measuring constructs from business research. The construct validity of these was often clarified. (engl.)

swell

↑ Lienert, Gustav A. (1998): Test setup and test analysis. [Study edition]. 6th edition Weinheim: Beltz Psychologie-Verl.-Union.
↑ Campell, DT; Fiske, DW (1959): Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, Vol. 56, pp. 81-105.
^ Moore, Gary C .; Benbasat, Izak (1991): Development of an Instrument to Measure the Perceptions of Adopting an Information Technology Innovation. Information Systems Research, Vol. 2, No. 3, pp. 192-222.
↑ Rossiter, John R. (2008): Content Validity of Measures of Abstract Constructs in Management and Organizational Research. British Journal of Management, Vol. 19, pp. 380-388.
^ Wacker, John G. (2004): A theory of formal conceptual definitions: developing theory-building measurement instruments. Journal of Operations Management Volume, Vol. 22, No. 6, pp. 629-650.

[1] Lienert, Gustav A. (1998): Test setup and test analysis. [Study edition]. 6th edition Weinheim: Beltz Psychologie-Verl.-Union.

[2] Campell, DT; Fiske, DW (1959): Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, Vol. 56, pp. 81-105.

[3] Moore, Gary C .; Benbasat, Izak (1991): Development of an Instrument to Measure the Perceptions of Adopting an Information Technology Innovation. Information Systems Research, Vol. 2, No. 3, pp. 192-222.

[4] Rossiter, John R. (2008): Content Validity of Measures of Abstract Constructs in Management and Organizational Research. British Journal of Management, Vol. 19, pp. 380-388.

[5] Wacker, John G. (2004): A theory of formal conceptual definitions: developing theory-building measurement instruments. Journal of Operations Management Volume, Vol. 22, No. 6, pp. 629-650.