The Importance of Measure in Psychology

Developing adequate measures is essential for the advancement of psychology as a science. Without the ability to adequately measure intended constructs, it is difficult for scientists to conduct experiments, form theories, or improve interventions. How do we know if a measure is adequate? Construct validity is the extent to which a measure actually measures what it is intended to measure. For example, an IQ test with high construct validity would actually measure intelligence and not something else.

In a large online sample and a college sample, our research group examined the construct validity of multiple forms of affective judgments of physical activity. Affective judgments of physical activity are the overall pleasure, enjoyment, and feelings associated with the experience of exercise or physical activity. Affective judgments have received recent attention because they are strong predictors of the amount of physical activity an individual engages in, and interventions targeting affective judgments appear to be a promising approach to increasing physical activity. We found that the majority of physical activity measures have substantial construct validity limitations. First, although the scales had adequate measurement consistency across items, most measures did not provide consistent results from one test occasion to the next. Second, these measures lacked convergent validity (i.e., the correlations among these measures were lower than would be expected if these measures were measuring the same construct). Third, they failed to demonstrate discriminant validity from some other exercise-related measures. However, the measures had adequate criterion validity (i.e., they predicted physical activity). Of note, scales with stronger psychometric properties demonstrated a stronger association with physical activity.

There are important implications from this study. Overall, affective judgment measures, and particularly those with stronger psychometric properties, are able to predict physical activity, suggesting affective judgments are an important construct to study. Focusing on improving psychometric properties of affective judgment measures could increase the predictive validity of affective judgment measures. However, it appears that the various affective judgment of physical activity measures have substantial construct validity limitations, suggesting that future research should focus on improving these measures. The lack of reliability limits our ability to know whether any changes observed in research are due to actual changes or a lack of precision in the measures. The lack of convergent validity suggests that future research needs to clarify whether enjoyment, motivation, affective attitudes, and affective associations are distinct constructs. This is important because if it is not clear regarding which construct we are measuring, we won’t know which construct might be changing.

Overall, these findings suggest that there should be a future focus on developing better affective judgment measures. If we have better measures of affective judgments, we will be in a better position to understand why people have the affective judgments they do and how to effectively intervene to increase physical activity. A focus on construct validity will allow us to identify the many possible dimensions of affective judgments of physical activity (e.g., enjoyment vs. satisfaction) and will allow us to identify when affective judgments change. A focus on construct validity can also improve interventions by allowing us to target specific affective judgment dimensions as well as allow us to more accurately evaluate which affective judgment dimensions are changing.

Chmielewski, M., Sala, M., Tang, R., & Baldwin, A. (2016). Examining the construct validity of affective judgments of physical activity measures. Psychological Assessment, 28(9), 1128-1141

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *