The NES Child Rearing Scale: A Detailed Empirical Examination of Its Measurement Properties
13 Pages Posted: 10 Oct 2017 Last revised: 27 Nov 2019
Date Written: October 10, 2017
The objective of this paper is to provide a more accurate assessment of the measurement properties of the NES child rearing scale. In the literature so far, the usual go-to evaluation procedures, Cronbach’s alpha and confirmatory factor analysis, have been used. It is argued in the paper that these procedures yield misleading results of the measurement properties of the scale because the nature of the scores yielded by an application of the child rearing scale are not suited to an application of these procedures. Another objective is to arrive at a more valid characterization of the meaning of the scores yielded by the child rearing scale. It is argued that deeming the scores as representing authoritarianism is not supported. Rather it is argued that a more meaningful characterization is “inner-directed autonomy” versus “outer-directed conformity”. These ideas are then applied to a reinterpretation of the differences of the scores on the child rearing scale between whites and racial and ethnic minorities. The methods employed are simple measures of association and a straight-forward scaling analysis. The results suggest that Cronbach’s alpha underestimates the reliability of the scores yielded by the child rearing scale and that confirmatory factor analysis is not an appropriate procedure for assessing the homogeneity of the items on the scale. Also, a detailed analysis of the scores points to the conclusion that the considerate/well-behaved item needs to be rethought. Huge differences in the scores between whites and racial and ethnic minorities are observed. Instead of basing an explanation of these differences on authoritarianism, an explanation is based on the survival value of conformity for members of ethnic and racial minorities that are the object of discrimination. The conclusion is that the child rearing scale is a worthy addition to the armamentarium of social science measurement procedures with a wide range of potential applications.
The data analyzed here were gathered by the ANES at the University of Michigan in their 2016 National Election Study. The author bears all responsibility for the uses and interpretations of these data.
Suggested Citation: Suggested Citation