Premium
Eight issues in studying organizational constructs: Tackling a complex problem
Author(s) -
Ree Malcolm James,
Carretta Thomas R.
Publication year - 2018
Publication title -
international journal of selection and assessment
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.812
H-Index - 61
eISSN - 1468-2389
pISSN - 0965-075X
DOI - 10.1111/ijsa.12219
Subject(s) - nomological network , psychology , variance (accounting) , reliability (semiconductor) , construct (python library) , interpretation (philosophy) , sample (material) , latent variable , organizational behavior , industrial and organizational psychology , social psychology , management science , structural equation modeling , applied psychology , computer science , artificial intelligence , machine learning , power (physics) , chemistry , physics , accounting , chromatography , quantum mechanics , economics , business , programming language
Much of the industrial/organizational psychology research literature involves abstractions called constructs. Frequently, new constructs are proposed and measures are developed, revised, and evaluated in studies for modeling or predicting organizational criteria. There are many hazards in these studies that lead to incorrect conclusions and decisions. This is especially applicable to the introduction and use of new constructs. We discuss several methodological issues that affect the interpretation of results from theoretical and applied studies and provide examples where construct confusion occurred due to failure to use best practices. A paradigm is proposed to increase the efficacy of studies regardless of which model of constructs has been chosen. Consideration is given to sample characteristics, range restriction, reliability of scores, criterion development, identifying latent sources of variance, content validity, comparing regression equations, and defining the nomological network.