Paul D. Bliese’s research while affiliated with University of South Carolina and other places

Intensive longitudinal models (ILMs) allow researchers to analyze nested data collected through frequent measurements—typically 20 or more repeated occasions—over densely spaced durations. Rather than being a single statistical approach, ILMs encompass various models unified by their capability to handle densely collected longitudinal data. We briefly summarize the nature of intensive longitudinal designs and why such designs require the use of ILMs. We then provide a classification typology to help readers understand the features of an ILM they should adopt. This classification typology provides the structure for a narrative review of existing ILM research. We conclude with specific recommendations for using ILMs to enhance theory, design, and analysis. Altogether, ILMs are a fairly straightforward extension of longitudinal models many researchers already use, and so we encourage their application to a broader range of theories and topics.

Objectives Although studies have addressed food insecurity among veterans, few have focused on active-duty soldiers or on variables associated with the military occupational context. We examined the link between marginal food insecurity (defined as anxiety over food sufficiency or shortage of food in the house) among US soldiers and demographic, behavioral health, and unit-related factors. Methods We analyzed survey data from 6343 active-duty soldiers using χ ² tests, generalized linear mixed-effect models, and adjusted odds ratios (AORs) to identify significant differences between soldiers categorized as marginally food insecure versus those who were not. Results In a fully adjusted model taking unit into account, marginal food insecurity was associated with preferring not to report gender (vs reporting being male) (AOR = 1.39; 95% CI, 1.08-1.78), being married/in a relationship (vs being single) (AOR = 1.22; 95% CI, 1.06-1.40), junior enlisted rank (vs noncommissioned officer: AOR = 0.45; 95% CI, 0.37-0.54; and vs officer: AOR = 0.13; 95% CI, 0.09-0.19), less time in unit (vs more time) (AOR = 0.99; 95% CI, 0.99-1.00), screening positive for depression (vs not) (AOR = 2.67; 95% CI, 2.30-3.11), screening positive for hazardous drinking (vs not) (AOR = 1.34; 95% CI, 1.11-1.63), and lack of reported unit-related social support (vs support) (AOR = 0.52; 95% CI, 0.45-0.59). Conclusions In this sample, more than 1 in 5 US soldiers reported marginal food insecurity. In addition to supporting households with financial and food assistance and targeting junior enlisted personnel, policy makers and leaders should prioritize soldiers who are married or in a relationship, who are new to their unit, and who screen positive for depression and hazardous drinking, and they should encourage units to take care of unit members who need support. Policy makers and leaders can use these study results to direct prevention and early intervention initiatives.

Despite the prevalence of research on the consequences of collective turnover (TO), we lack an understanding of how, when, and why changes in the external environment influence collective turnover. The present study extends context emergent turnover and threat-rigidity theories to consider temporal changes in rates of collective turnover brought on by an external disruption. We also conduct variance decomposition to evaluate the relative influence of internal and external factors on collective turnover and examine how changes in the external environment impact relative influences. Finally, we examine the role of collective engagement in explaining patterns of collective turnover over time. Our study is based on a large, geographically dispersed U.S. firm. Findings from a two-phase longitudinal model reveal that rates of collective turnover change over time in ways that are predictable from threat-rigidity theory. Variance decomposition analysis finds that internal store-level factors explain substantially more variance than external factors, but the balance changes in response to an external disruption. We also show that collective engagement can mitigate increases in collective turnover. Results inform theory regarding the relative importance of internal versus external factors in influencing collective turnover and provide a framework for predicting how contextual change in the external environment impacts collective turnover over time.

Dishop (2022) identifies the consensus emergence model (CEM) as a useful tool for future research on emergence but argues that autoregressive models with positive autoregressive effects are an important alternative data-generating mechanism that researchers need to rule out. Here, we acknowledge that alternative data-generating mechanisms are possibility for most, if not all, nonexperimental designs and appreciate Dishop’s attempts to identify cases where the CEM could provide misleading results. However, in a series of independent simulations, we were unable to replicate two of three key analyses, and the results for the third analysis did not support the earlier conclusions. The discrepancies appear to originate from Dishop’s simulation code and what appear to be inconsistent model specifications that neither simulate the models described in the article nor include notable positive autoregressive effects. We contribute to the wider literature by suggesting four key criteria that researchers can apply to evaluate the possibility of alternative data-generating mechanisms: Theory, parameter recovery, fit to real data, and context. Applied to autoregressive effects and emergence data, these criteria reveal that (a) theory in psychology would generally suggest negative instead of positive autoregressive effects for behavior, (b) it is challenging to recover true autoregressive parameters from simulated data, and (c) that real data sets across a number of different contexts show little to no evidence for autoregressive effects. Instead, our analyses suggest that CEM results are congruent with the temporal changes occurring within groups and that autoregressive effects do not lead to spurious CEM results.