Interpretational Confounding or Confounded Interpretations of Causal Indicators?

Studying Socioeconomic Status: Conceptual Problems and an Alternative Path Forward

Socioeconomic status (SES; or social class) is considered an important determinant of psychological and life outcomes. Despite this importance, how to appropriately conceive of and measure it remains unsettled. In this article, I argue that SES is, under conventional conceptions of the construct, an unmeasurable construct and present an alternative strategy for studying socioeconomic conditions. I make this argument using several lines of analysis. First, a literature review of 20 years of psychological research on SES reveals that psychologists rarely define SES theoretically (79.6% of articles did not) but call a great number of operationalizations measures of SES (147 in total). Second, current recommendations for studying SES permit contradictory predictions, rendering the recommendations unsatisfactory. Third, the appropriate measurement model for SES inhibits accumulation of results across studies, which makes studying the construct practically impossible. To rectify these issues, I reconceptualize SES as a set of socioeconomic conditions and develop a measurement strategy for studying these conditions. I conclude by considering implications for ongoing research on socioeconomic conditions and for interpreting past research on SES.

The Potential for Interpretational Confounding in Cognitive Diagnosis Models

Binary examinee mastery/nonmastery classifications in cognitive diagnosis models may often be an approximation to proficiencies that are better regarded as continuous. Such misspecification can lead to inconsistencies in the operational definition of "mastery" when binary skills models are assumed. In this paper we demonstrate the potential for an interpretational confounding of the latent skills when truly continuous skills are treated as binary. Using the DINA model as an example, we show how such forms of confounding can be observed through item and/or examinee parameter change when (1) different collections of items (such as representing different test forms) previously calibrated separately are subsequently calibrated together; and (2) when structural restrictions are placed on the relationships among skill attributes (such as the assumption of strictly nonnegative growth over time), among other possibilities. We examine these occurrences in both simulation and real data studies. It is suggested that researchers should regularly attend to the potential for interpretational confounding by studying differences in attribute mastery proportions and/or changes in item parameter (e.g., slip and guess) estimates attributable to skill continuity when the same samples of examinees are administered different test forms, or the same test forms are involved in different calibrations.

The LOHAS (Lifestyle of Health and Sustainability) Scale Development and Validation

LOHAS (Lifestyle of Health and Sustainability) describes an emerging lifestyle that is defined by attention to health, well-being, and environmental sustainability. Discussions of the LOHAS lifestyle have moved faster than any of the research to support it. Originally developed in South Korea, it has been picked up in the U.S. and other cultures worldwide. However, researchers have proceeded as if one scale fits all. The implications of LOHAS can only proceed if there is a reliable and valid measure for LOHAS and empirical evidence that the scale is effective for diverse groups. The current research focuses on the development of a psychometrically reliable and valid scale to measure the multi-dimensional nature of LOHAS. By following generally accepted scale development procedures, a LOHAS scale is created and tested for its reliability, dimensionality, construct, and nomological validity. Finally, theoretical and managerial implications are outlined.

Three Conceptual Impediments to Developing Scale Theory for Formative Scales

Abstract. Bollen and colleagues have advocated the use of formative scales despite the fact that formative scales lack an adequate underlying theory to guide development or validation such as that which underlies reflective scales. Three conceptual impediments impede the development of such theory: the redefinition of measurement restricted to the context of model fitting, the inscrutable notion of conceptual unity, and a systematic conflation of item scores with attributes. Setting aside these impediments opens the door to progress in developing the needed theory to support formative scale use. A broader perspective facilitates consideration of standard scale development concerns as applied to formative scales including scale development, item analysis, reliability, and item bias. While formative scales require a different pattern of emphasis, all five of the traditional sources of validity evidence apply to formative scales. Responsible use of formative scales requires greater attention to developing the requisite underlying theory.

Notes on measurement theory for causal-formative indicators: A reply to Hardin

The current article is a rejoinder to "A Call for Theory to Support the Use of Causal-Formative Indicators: A Commentary on Bollen and Diamantopoulos." Our article comments on the 6 research questions raised by Hardin (2017) in his constructive commentary on our original article (i.e., "In Defense of Causal-Formative Indicators: A Minority Report"). (PsycINFO Database Record

Are reflective models appropriate for very short scales? Proofs of concept of formative models using the Ten-Item Personality Inventory

OBJECTIVE

Because of their length and objective of broad content coverage, very short scales can show limited internal consistency and structural validity. We argue that it is because their objectives may be better aligned with formative investigations than with reflective measurement methods that capitalize on content overlap. As proofs of concept of formative investigations of short scales, we investigate the Ten-Item Personality Inventory (TIPI).

METHOD

In Study 1, we administered the TIPI and the Big Five Inventory (BFI) to 938 adults and fitted a formative Multiple Indicators Multiple Causes model, which consisted of the TIPI items forming five latent variables, which in turn predicted the five BFI scores. These results were replicated in Study 2 on a sample of 759 adults, but this time with the Revised NEO Personality Inventory (NEO-PI-R) as the external criterion.

RESULTS

The models fit the data adequately, and moderate to strong significant effects (.37 < |β| < .69, all ps < .001) of all five latent formative variables on their corresponding BFI and NEO-PI-R scores were observed.

CONCLUSIONS

This study presents a formative approach that we propose to be more consistent with the aims of scales with broad content and short length like the TIPI.

How Should Alcohol Problems Be Conceptualized? Causal Indicators Within the Rutgers Alcohol Problem Index

Alcohol-related problems have traditionally been conceptualized and measured by an effect indicator model. That is, it is generally assumed that observed indicators of alcohol problems are caused by a latent variable. However, there are reasons to think that this construct is more accurately conceptualized as including at least some causal indicators, in which observed indicators cause the latent variable. The present study examined the measurement model of a well-known alcohol consequences questionnaire, the Rutgers Alcohol Problem Index. Participants were 703 students from a large public university in the Northeast mandated to an alcohol intervention. We conducted a zero tetrad test to examine a measurement model consisting solely of effect indicators and a model with both causal and effect indicators. Overall, the results suggested the hybrid model fit the data better than a model with only effect indicators. These findings have implications regarding the theoretical underpinnings of alcohol-related consequences.

In defense of causal-formative indicators: A minority report

Causal-formative indicators directly affect their corresponding latent variable. They run counter to the predominant view that indicators depend on latent variables and are thus often controversial. If present, such indicators have serious implications for factor analysis, reliability theory, item response theory, structural equation models, and most measurement approaches that are based on reflective or effect indicators. Psychological Methods has published a number of influential articles on causal and formative indicators as well as launching the first major backlash against them. This article examines 7 common criticisms of these indicators distilled from the literature: (a) A construct measured with "formative" indicators does not exist independently of its indicators; (b) Such indicators are causes rather than measures; (c) They imply multiple dimensions to a construct and this is a liability; (d) They are assumed to be error-free, which is unrealistic; (e) They are inherently subject to interpretational confounding; (f) They fail proportionality constraints; and (g) Their coefficients should be set in advance and not estimated. We summarize each of these criticisms and point out the flaws in the logic and evidence marshaled in their support. The most common problems are not distinguishing between what we call causal-formative and composite-formative indicators, tautological fallacies, and highlighting issues that are common to all indicators, but presenting them as special problems of causal-formative indicators. We conclude that measurement theory needs (a) to incorporate these types of indicators, and (b) to better understand their similarities to and differences from traditional indicators. (PsycINFO Database Record