Structural Equation Models of Latent Interactions: Evaluation of Alternative Estimation Strategies and Indicator Construction.

Abstract

Interactions between (multiple indicator) latent variables are rarely used because of implementation complexity and competing strategies. Based on 4 simulation studies, the traditional constrained approach performed more poorly than did 3 new approaches--unconstrained, generalized appended product indicator, and quasi-maximum-likelihood (QML). The authors' new unconstrained approach was easiest to apply. All 4 approaches were relatively unbiased for normally distributed indicators, but the constrained and QML approaches were more biased for nonnormal data; the size and direction of the bias varied with the distribution but not with the sample size. QML had more power, but this advantage was qualified by consistently higher Type I error rates. The authors also compared general strategies for defining product indicators to represent the latent interaction factor.

Do Norwegian Sami and non-indigenous individuals understand questions about mental health similarly? A SAMINOR 2 study

“…The fit of the factor model was evaluated as the degree of model discrepancy (RMSEA-root mean square error of approximation) and degree of relative fit (CFI-comparative fit index; and NNFI-non normed fit index). RMSEA values below < .06 are preferable [31], while values for CFI/NNFI should minimally pass > .90 [32] or preferably > .95 [33]. Absolute model fit (i.e.…”

How and When Socially Entrepreneurial Nonprofit Organizations Benefit From Adopting Social Alliance Management Routines to Manage Social Alliances?

“…It is impossible to follow both sets of guidelines. Nevertheless, we followed Marsh et al (2004Marsh et al ( , 2006 recommendations and the approach adopted by Homburg et al (2014) when facing this situation. These infer that using all information available (guideline 1) should lend more weight (i.e., using each item at least once) to enable us to create product indicators by matching one item from either the SA relationship-building motive or the SA benefits-exploiting motive, with five items from the SA management routines.…”

“…For moderation analysis, we followed Kline's (2005) approach to estimate all of the latent variables and their interactions in the same structural equation model. To create latent interaction, we followed Marsh et al (2004Marsh et al ( , 2006 guidelines for forming product indicators: (1) use all of the information and (2) do not reuse any of the information. We measured SA management routines through 10 items, but both the SA relationship-building motive and the SA benefits-exploiting motive only had two items each.…”

The role of ambiguity tolerance in career decision making

“…The latent variable moderation test was conducted with the unconstrained approach (Marsh, Wen, & Hau, 2004), in which the latent interaction terms were indicated by the multiplication products of the indicators for the latent interacted variables and the simple slops of the interaction terms were tested to see if there were any moderation effects.…”