Causal Indicators in Psychometrics

Psychometric evaluation of an adapted version of the perceived stress scale for ecological momentary assessment research

“…Responses are recorded on a 4‐point Likert‐type scale from strongly agree to strongly disagree. In this study, provocations were averaged to provide a composite measure of level of provocation experienced (as opposed to used within a latent variable measurement model), as the items are designed for formative rather than reflective measurement (Murray & Booth, 2018).…”

Development of a new measure of encounters with health care barriers: The Commonly Experienced Health Care Barriers Index.

“…Notably, items assessing common barriers are likely to be causal indicators. Most scales use effect indicators, where responses are caused by an underlying latent variable and all items loading on a single latent variable are intercorrelated and theoretically interchangeable (Bollen & Bauldry, 2011; Murray & Booth, 2018). In contrast, causal indicators are expected to describe causes of the latent variable, rather than effects.…”

“…In contrast to the central role of barriers in Andersen’s (1995) model, this assessment of barriers was only trivially associated with health care utilization problems, potentially due to this omission. Murray and Booth (2018) recommended overinclusiveness during the item generation stage, which could be accomplished by drawing from theory, reviewing previous literature, and conducting qualitative research to identify a pool of potential barriers to include on the scale.…”

“…At the same time, it is crucial to ensure that only necessary items are included, both to ensure the definition of the construct is not affected by irrelevant items (Murray & Booth, 2018) and to ensure brevity. Maximum brevity is especially important for scales designed for use in health care settings because in such settings, feasibility is directly linked to scale length (Miller et al, 2003).…”

“…To determine if a final set of items are indeed functioning as causal indicators, Murray and Booth (2018) recommended a process that begins with simply reviewing the items to confirm that they meet five checks regarding theoretical assumptions of causal indicators and then conducting an empirical tetrad analysis. The five conceptual checks include determining if there is theoretical reason to assume (a) that each indicator causes the construct, (b) that a change in each indicator will produce a change construct, (c) that the construct is not a common cause of the indicators, (d) that there is potential for the indicators to have different causes, and (e) that the indicators are not interchangeable.…”