Longitudinal data analysis

Latent transition analysis in organizational psychology: A simplified “how to” guide by using an applied example

Demands for more robust designs in organizational research have led to a steady increase in the number of longitudinal studies in organizational psychology (OP) journals. Similarly, the number and ways to analyze longitudinal data have also increased. In this paper, we adopt a relatively new and promising approach to help researchers analyze their longitudinal data in OP, namely latent transition analysis (LTA). We present a simplified guideline on LTA and discuss its role for OP researchers. Moreover, we demonstrate how organizational scholars can use this method with a practical example. In this example, we investigate (a) if there are qualitatively distinct subgroups of employees based on particular patterns of psychological capital (PsyCap) dimensions (i.e., efficacy, hope, resilience, and optimism), (b) if employees stay in these subgroups or transition to other groups over time, and finally, (c) if leader-member exchange (LMX) is associated with this transition. We use LTA to examine these steps in a German sample (N = 180).

The impact of engaging leadership on employee engagement and team effectiveness: A longitudinal, multi-level study on the mediating role of personal- and team resources

Most research on the effect of leadership behavior on employees' well-being and organizational outcomes is based on leadership frameworks that are not rooted in sound psychological theories of motivation and are limited to either an individual or organizational levels of analysis. The current paper investigates whether individual and team resources explain the impact of engaging leadership on work engagement and team effectiveness, respectively. Data were collected at two time points on N = 1,048 employees nested within 90 work teams. The Multilevel Structural Equation Modeling results revealed that personal resources (i.e., optimism, resiliency, self-efficacy, and flexibility) partially mediated the impact of T1 individual perceptions of engaging leadership on T2 work engagement. Furthermore, joint perceptions of engaging leadership among team members at T1 resulted in greater team effectiveness at T2. This association was fully mediated by team resources (i.e., performance feedback, trust in management, communication, and participation in decision-making). Moreover, team resources had a significant cross-level effect on individual levels of engagement. In practical terms, training and supporting leaders who inspire, strengthen, and connect their subordinates could significantly improve employees' motivation and involvement and enable teams to pursue their common goals successfully.

Unpacking the Relationship Between Organizational Citizenship Behavior and Counterproductive Work Behavior: Moral Licensing and Temporal Focus

Traditionally, scientific- and practitioner-oriented publications tend to categorize employees in groups of either “good” or “bad” employees, thereby omitting that one category of employees might engage in organizational citizenship behavior (OCB-O) and counterproductive work behavior (CWB-O). In this study, we concurrently examine the mediating role of moral credits and credentials, as well as the moderating role of subjective temporal focus. Specifically, we argue that when employees enact OCB-O, they obtain moral credits and credentials, which in turn might make employees more likely to enact CWB-O. Moreover, we argue that the latter relationships depend on an employee’s subjective temporal focus, resulting in an OCB-O—CWB-O relationship that is (1) positive for a past temporal focus, (2) negative for a future temporal focus, and (3) non-significant for a present temporal focus. We examined these hypotheses by means of a multilevel weekly survey study and largely found support for our hypotheses, especially with regard to the role of moral credentials as the mediating mechanism and the aggravating versus attenuating effect of past versus future temporal focus, respectively. We end with a discussion on implications, suggestions for future research, and recommendations for practice.

12-Year longitudinal study linking within-person changes in work and family transitions and workplace injury risk

INTRODUCTION

Despite the rich tradition of research on predictors of workplace injury, most studies rely on cross-sectional, between-person designs. Furthermore, prior research has often overlooked the possibility that factors outside the work domain can influence the occurrence of actual injuries at work. To address these limitations, the current study examined the effects of work and family demands on the occurrence of workplace injury. Drawing on the intuition of the work-home resources model (W-HR), we investigated how within-person level changes in demands and resources from both domains influence work injuries over a 12-year period.

METHOD

We used 12 years of longitudinal data (N = 7,820) to study the long-term within-person changes in work and family domains and to capture the event of low frequency incidence such as workplace injury. Specifically, we conducted multilevel analyses to study the links between within-person change in time and energy resources both in work and family domains and within-person change in the likelihood of experiencing a workplace injury.

RESULTS AND CONCLUSION

The findings showed that within-person changes in work hours, spousal work hours, income and number of children, were significantly associated with changes in the likelihood of experiencing a workplace injury. We conclude with a discussion of implications for theory and future research of workplace injuries. Practical application: The research provided useful insights on the intimate association between work and family domains in the context of safety management.

Good Things Are Not Eternal: How Consideration Leadership and Initiating Structure Influence the Dynamic Nature of Organizational Justice and Extra-Role Behaviors at the Collective Level

The objective of this study was to test the linear and curvilinear influence of initiating structure (IS) and consideration leadership on the variability of extra-role behaviors over time, through the mediating effect of within-store level of distributive (DJ) and procedural justice (PJ). Data from 1,857 employees in 116 business units of a Canadian retailer collected over four waves were analyzed using random coefficient modeling. Our results show that IS and consideration have significant linear and curvilinear effects on within-unit PJ and DJ climates over time, and an indirect influence on the within-unit variability of extra-role behaviors through the mediating effect of DJ and PJ climates. Results underline the complexity of the effects of both leadership dimensions on PJ and DJ climates. While consideration is associated with an increase of both justice climates over time, low and high levels of IS are associated with a decrease in PJ climate over time, and a moderate level of IS was related to an increase of PJ climate. Moreover, the positive effects of leadership behaviors on extra-role behaviors shift completely after a certain time threshold. Surprisingly, while unit-level extra-role behaviors increased over time under high structure leadership, such behaviors decreased under high consideration. Finally, the results showed that the positive influence of PJ and DJ climates on unit-level extra-role behaviors decreases over time, and that low fairness units experience a more dramatic decrease of extra-role behaviors with time.

When Task Conflict Becomes Personal: The Impact of Perceived Team Performance

Although potentially beneficial, task conflict may threaten teams because it often leads to relationship conflict. Prior research has identified a set of interpersonal factors (e.g., team communication, team trust) that help attenuate this association. The purpose of this article is to provide an alternative perspective that focuses on the moderating role of performance-related factors (i.e., perceived team performance). Using social identity theory, we build a model that predicts how task conflict associates with growth in relationship conflict and how perceived team performance influences this association. We test a three-wave longitudinal model by means of random coefficient growth modeling, using data from 60 ongoing teams working in a health care organization. Results provide partial support for our hypotheses. Only when perceived team performance is low, do task conflicts relate with growth in relationship conflict. We conclude that perceived team performance seems to enable teams to uncouple task from relationship conflict.

Differential Reactivity and the Within-person Job Stressor-Satisfaction Relationship

An experience sampling methodology was used to study the direct and conditional within-person relationship between job stressors and job satisfaction. One hundred and one full-time administrative staff completed momentary measures of job stressors and job satisfaction three times a day on six different workdays over a 3-week period (N = 1818 observations). Multilevel random coefficients models were specified, and the results suggest that within-person stressors are negatively related to within-person job satisfaction. These results stand when controlling for the effects of time, demographics, work characteristics, baseline levels of job stressors and satisfaction, and between-person effects of job stressors. Furthermore, consistent with the differential reactivity model, the results suggest that the observed within-person stressors-satisfaction relationship is conditional upon locus of control and positive affect. Copyright © 2015 John Wiley & Sons, Ltd.

Growth Modeling Using Random Coefficient Models: Model Building, Testing, and Illustrations

In this article, the authors illustrate how random coefficient modeling can be used to develop growth models for the analysis of longitudinal data. In contrast to previous discussions of random coefficient models, this article provides step-by-step guidance using a model comparison framework. By approaching the modeling this way, the authors are able to build off a regression foundation and progressively estimate and evaluate more complex models. In the model comparison framework, the article illustrates the value of using likelihood tests to contrast alternative models (rather than the typical reliance on tests of significance involving individual parameters), and it provides code in the open-source language R to allow readers to replicate the results. The article concludes with practical guidelines for estimating growth models.

Project Complexity and Bonding Social Capital in Network Organizations

This study conceptually and empirically explores how project complexity and bonding forms of social capital influence performance outcomes in network organizations. Specifically, we focus on how bonding social capital within network organizations—measured as frequency of collaboration and degree of network coupling—can influence project performance outcomes both (a) directly by facilitating cooperative interaction and (b) contingently by mitigating the transaction costs associated with the management of complex projects. Using longitudinal data on contracted construction jobs to test our hypotheses, we find that project complexity is negatively related to project performance and bonding social capital has both direct and moderating effects. Contrary to expectations, however, we find that the different types of bonding social capital affect project performance uniquely and not always in an improved direction. Our findings suggest a more multifarious relationship than previous social capital research might imply.

Authentic leadership and mindfulness development through action learning

Purpose

– The purpose of this paper is to evaluate a three-year training program based on action learning principles with regard to its effectiveness in fostering authentic leadership (AL) and mindfulness among the participants.

Design/methodology/approach

– Data were obtained using a mixed-method design. Quantitative data were collected using a quasi-experimental sequential cohort design with comparison group, in which 143 participants responded to a self-evaluation questionnaire up to six times over a three-year period. Semi-structured interviews were also conducted with 24 managers.

Findings

– The results indicate that, as participants evolved through the leadership development program, self-reports of AL and mindfulness increased significantly and linearly as determined using repeated measures ANOVA, paired t-tests, and content analysis of interviews.

Practical implications

– The results suggest that a leadership development program based on action learning principles can foster the development of AL and mindfulness. The core elements of action learning (i.e. working on real problems, gaining new insights in a supportive and confrontational environment of one’s peer) appear to be key to bringing about real changes in the behavior of participating managers and maximizing the chances of generating lasting effects.

Originality/value

– This is the first longitudinal study to demonstrate that the development of mindfulness and AL – which calls for internalization of attitudes and behaviors – can be fostered by a leadership development program. The question of whether AL can be developed through planned interventions is paramount for advancing theory and research on AL.

Unpacking team dynamics with growth modeling

In this paper we advocate the use of growth modeling as an approach that is particularly useful for testing and refining existing theory on team dynamics, as well as integrating different theoretical perspectives. Quantitative studies that test team theories have typically included only one or two time points, between-team research designs, and hierarchical regression-based statistical analyses. Such an approach enables exploration of antecedents to explain why some teams are more effective than others at specified points in the team task or lifespan. In contrast, using three or more time points of data and applying growth modeling statistical analyses is atypical, but can allow for informative investigations of team trajectories, or patterns of change within teams. We argue that this approach can facilitate fruitful insights about team dynamics, and we provide guidelines for researchers as to how to investigate such team dynamics using growth modeling.

Treatment-related reductions in PTSD and changes in physical health symptoms in women

This study examined the relationship between change in posttraumatic stress disorder (PTSD) symptoms over the course of PTSD treatment and the association with changes in general physical health symptoms. Both positive health habits (e.g., exercise) and negative (e.g., smoking), were examined to determine if they accounted for the association between changes in PTSD severity over time and changes in physical health. Participants were 150 women seeking treatment for PTSD. Latent growth curve modeling indicated a substantial relationship (R (2) = 34%) between changes in PTSD and changes in physical health that occurred during and shortly following treatment for PTSD. However, there was no evidence to suggest that changes in health behaviors accounted for this relationship. Thus, PTSD treatment can have beneficial effects on self-reported physical health symptoms, even without direct treatment focus on health per se, and is not accounted for by shifts in health behavior.

Intimate partner violence exposure predicts PTSD treatment engagement and outcome in cognitive processing therapy

Intimate partner violence (IPV) exposure was examined as a predictor of treatment engagement (i.e., starting and completing therapy) and treatment outcome in 150 women taking part in a dismantling study of cognitive processing therapy (CPT) for posttraumatic stress disorder (PTSD; Resick et al., 2008). Results indicate that women in a current intimate relationship with recent IPV (i.e., past year) were less likely to begin treatment relative to women who reported past IPV only or no history of IPV. For women who began treatment, IPV exposure was not predictive of whether or not they completed treatment. Among women who began treatment, the frequency of IPV was associated with treatment outcome such that women who experienced more frequent IPV exhibited larger reductions in PTSD and depression symptoms over the course of treatment, but experienced similar levels of PTSD and depression severity at the 6-month follow-up. Findings highlight the importance of targeting treatment engagement among women who report recent IPV and suggest that women who have experienced frequent IPV respond well to CPT treatment in spite of their IPV experiences.

Development of a minimization instrument for allocation of a hospital-level performance improvement intervention to reduce waiting times in Ontario emergency departments

BACKGROUND

Rigorous evaluation of an intervention requires that its allocation be unbiased with respect to confounders; this is especially difficult in complex, system-wide healthcare interventions. We developed a short survey instrument to identify factors for a minimization algorithm for the allocation of a hospital-level intervention to reduce emergency department (ED) waiting times in Ontario, Canada.

METHODS

Potential confounders influencing the intervention's success were identified by literature review, and grouped by healthcare setting specific change stages. An international multi-disciplinary (clinical, administrative, decision maker, management) panel evaluated these factors in a two-stage modified-delphi and nominal group process based on four domains: change readiness, evidence base, face validity, and clarity of definition.

RESULTS

An original set of 33 factors were identified from the literature. The panel reduced the list to 12 in the first round survey. In the second survey, experts scored each factor according to the four domains; summary scores and consensus discussion resulted in the final selection and measurement of four hospital-level factors to be used in the minimization algorithm: improved patient flow as a hospital's leadership priority; physicians' receptiveness to organizational change; efficiency of bed management; and physician incentives supporting the change goal.

CONCLUSION

We developed a simple tool designed to gather data from senior hospital administrators on factors likely to affect the success of a hospital patient flow improvement intervention. A minimization algorithm will ensure balanced allocation of the intervention with respect to these factors in study hospitals.

A Longitudinal Study of the Relationships Among Self-Monitoring, Authentic Leadership, and Perceptions of Leadership

There is little research examining variability in leadership outcomes over time as a function of individual differences. The present study investigates how the extent to which individuals are perceived as leaders varies over time, with self-monitoring and authentic leadership as predictors. Hierarchical linear modeling analyses found that change in the extent to which individuals were perceived to be leaders by others varied across individuals, but this variability was not due to individuals' self-monitoring, authentic leadership, or declining interrater reliability. It was also found that individuals' overestimation of themselves as leaders increased over time. Finally, an initial measure of authentic leadership was developed.

A group randomized trial of critical incident stress debriefing provided to U.S. peacekeepers

In a group randomized trial of critical incident stress debriefing (CISD) with platoons of 952 peacekeepers, CISD was compared with a stress management class (SMC) and survey-only (SO) condition. Multilevel growth curve modeling found that CISD did not differentially hasten recovery compared to the other two conditions. For those soldiers reporting the highest degree of exposure to mission stressors, CISD was minimally associated with lower reports of posttraumatic stress and aggression (vs. SMC), higher perceived organizational support (vs. SO), and more alcohol problems than SMC and SO. Soldiers reported that they liked CISD more than the SMC, and CISD did not cause undue distress.

Evaluation of treatment programs for dual disorder individuals: modeling longitudinal and mediation effects

This study evaluated the effectiveness of the three approaches for treating dual disorder clients who were homeless at intake: integrated assertive community treatment (IACT), assertive community treatment only (ACTO), and standard care (SC). Multilevel Random Coefficient Modeling (MRCM) was used to analyze longitudinal effects and to identify mediators of significant treatment effects. The outcome variables were consumer satisfaction, stable housing, psychiatric symptoms, and substance abuse. The eight mediators were service utilization variables: program contacts, phone contacts, substance abuse contacts, assistance with activities of daily living, transportation assistance, help finding permanent housing, help with emotional problems, and medication assistance. The 191 eligible participants were randomly assigned to one of the three conditions and followed for a period of 30 months. Both ACTO and IACT produced better outcomes than SC on consumer satisfaction and stable housing. There were no differences on any of the outcome variables between ACTO versus IACT when comparing main effects. However, there were several treatment by time interactions. In addition, there were many mediation effects.

Time in Applied Psychology

Against the background of a growing awareness of time among researchers, this articles sets out to assess the current position of time in applied psychological theory and research. A critical analysis of recent publications in The Journal of Applied Psychology and Applied Psychology, An International Review reveals that with some noticeable exceptions, the role of time in human behavior is still largely neglected. It is posited that this state of affairs can be explained by a conceptual bias among researchers, which makes them think in terms of “what is” rather than “what happens.” A proposal is made for an alternative way of conceptualizing, designated as temporalism, which starts from the concept of phenomenon, which is defined dynamically and operationalized with the help of time. A plea is made for a research strategy that comprises the identification of the dynamic features of phenomena, an analysis of their temporal relations, and an assessment of long term stability and changes of temporal parameters. It is argued that such an approach leads to theoretical innovations and to a substantial expansion of possibilities for intervention. This is illustrated by a discussion of three prototypical patterns of development referred to as improvement, bifurcation, and relapse, which underlie a broad range of practical problems known from social and organizational settings. Finally, attention is directed to the role that time plays when psychological knowledge is applied to such practical problems. It is concluded that time should not only be addressed in the study of behavior of other people but also in the complex interactions between psychologists and clients involved in the application of psychology.

The Accuracy of Significance Tests for Slope Variance Components in Multilevel Random Coefficient Models

This study examines the behavior of three tests for significant slope variance in multilevel random coefficient (MRC) models: the Hierarchical Linear Modeling chi-square test, the likelihood ratio test (LRT), and the corrected LRT. Monte Carlo simulations are conducted varying the numbers of groups, group size, and effect size. Results suggest that neither the number of groups nor group size influenced Type I errors. Group size has a stronger effect on power compared with the number of groups. The one-tailed LRT demonstrates the best balance between power and Type I errors. Recommendations for conducting MRC analyses are presented.

Multilevel Methods

The study of multilevel phenomena in organizations involves a complex interplay between methods and statistics on one hand and theory development on the other. In this introduction, the authors provide a short summary of the five articles in this feature topic and use them as a platform to discuss the broad need for work in the two areas of (a) multilevel construct validation and measurement and (b) statistical advances in variance decomposition. Within these two broad frameworks, the authors specifically discuss, first, the need to continue moving beyond notions of isomorphism in developing and testing aggregate-level constructs. Second, they discuss the potential value of using discontinuous growth models to understand transitions in longitudinal studies. Finally, they discuss some of the issues surrounding the ability to decompose variance in multilevel modeling of dichotomous and other nonnormal outcome data.

Postmarketing surveillance study of benazepril in chinese patients with hypertension: An open-label, experimental, epidemiologic study

BACKGROUND

Benazepril hydrochloride is an angiotensin-converting enzyme inhibitor. Previous clinical trials show that antihypertensive treatment with benazepril provides effective blood pressure (BP) control and is generally well tolerated by patients with hypertension. However, the long-term antihypertensive effects and tolerability of benazepril remain to be established in Chinese patients with hypertension.

OBJECTIVE

The aim of this study was to investigate the long-term efficacy and tolerability of benazepril in Chinese patients with essential hypertension.

METHODS

This 36-month, community-based, open-label, postmarketing surveillance study was conducted in the Nanshi District (Shanghai, China). Chinese patients with essential hypertension were to receive 1 or more benazepril tablets PO QD in the morning for 36 months. Data for BP and pulse pressure (PP) were collected at baseline (month 0) and throughout the surveillance period. The rate of patients achieving BP targets (systolic BP [SBP]/diastolic BP [DBP], <140/<90 mm Hg) was determined, as was the rate of decrease in BP. Subanalyses by sex and age group also were conducted.

RESULTS

A total of 1831 patients (1090 men, 741 women; mean [SD] age, 55.8 [10.1] years [range, 35-88 years]) entered the study. After the 36-month treatment period, 75.7% of patients receiving benazepril as prescribed (1289 patients) had achieved the SBP target, 87.4% achieved the DBP target, and 71.5% achieved both targets. After 36 months of treatment, the mean (SD) decreases in SBP, DBP, and PP were 15.1 (0.4) mm Hg, 11.0 (0.3) mm Hg, and 4.2 (0.4) mm Hg, respectively, among compliers. In general, the rate of BP decrease slowed over time. No serious adverse drug reactions (ADRs) were detected during the 36-month follow-up period. All ADRs except cough (19.9%) occurred at a relatively low incidence rate (<3.0%). The cumulative incidence of benazepril related cough was statistically significantly higher in women than in men (23.6% vs 18.8%, respectively; P = 0.007). Of the 1831 patients studied, 1360 patients (74.3%) persisted in taking benazepril and were considered optimally compliant at 36-month follow-up.

CONCLUSION

In this study of Chinese patients with hypertension, benazepril was associated with prolonged, stable efficacy in lowering BP and relatively low incidence of ADRs.