Premium
Distinguishing Constructs from Variables in Designing Research
Author(s) -
Calder Bobby J.,
Brendl C. Miguel,
Tybout Alice M.,
Sternthal Brian
Publication year - 2021
Publication title -
journal of consumer psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 4.433
H-Index - 110
eISSN - 1532-7663
pISSN - 1057-7408
DOI - 10.1002/jcpy.1204
Subject(s) - conflation , construct (python library) , explanatory power , psychology , inference , matching (statistics) , variable (mathematics) , mediation , perspective (graphical) , variables , causal inference , exploit , computer science , econometrics , epistemology , artificial intelligence , statistics , machine learning , mathematics , sociology , mathematical analysis , social science , philosophy , computer security , programming language
Current research practices often conflate theoretical constructs and explanatory hypotheses with variables and predicted effects, to the detriment of research progress. This has led to the use of procedures such as manipulation checks, mediation analysis, and boundary conditions predicated on the idea that matching constructs to variables is necessary to validate that a theory corresponds to an effect . An alternative perspective, Inference to the Best Explanation (IBE), calls for designing research to exploit the power of distinguishing constructs from variables, hypotheses from predictions, and theory from effects. IBE calls for stating hypotheses (Hs) about construct‐to‐construct relationships and, separately, the predictions (Ps) about variable‐to‐variable effects that are explained by the hypotheses. In addition, articles should include disparate effects, a single explanation covering all studies, and a discussion of the use of the research in specific problem contexts. The application of IBE is illustrated with research investigating when judgments are based on a feeling about the ease of information retrieval versus the information content itself.