No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis.
Abstract
Differentiates among indices of agreement, reliability, and non-independence in organizational multilevel research and provides an overview of what each measure reveals about the group-level properties of one's data. Particular attention is devoted to the concept of non-independence and to the relationship between one form of the intraclass correlation coefficient and eta-squared. The 2nd purpose of this chapter is to extend the discussion of agreement, reliability, and non-independence from an exclusive focus on validating one's measurement model to a broader focus that includes testing substantive models and detecting emergent phenomena. In the discussion of bottom-up processes, particular attention is paid to a category that the author defines as "fuzzy" composition processes. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
... Note. To examine how much of the total variability in the outcome variables could be attributed to variability between the clusters in our study (i.e., classrooms and students within classrooms), we calculated ICC1 values (Bliese, 2000). For instance, regarding inclusion expectations, 2% of the total variance was attributed to differences between classrooms, indicating that children within one classroom responded differently compared to children in another classroom. ...
... In addition to the ICC1 values, we also calculated an ICC2 value for the shared classroom norm variable that was aggregated at the classroom level. The calculation of ICC2 is important to assess the degree to which aggregating students' perceptions form a reliable classroom aggregate (Bliese, 2000). ...
... Specifically, the design included repeated measures (Level 1) from students (Level 2) within classrooms (Level 3). For correlations among the main study variables at Level 1, see Supplementary Materials 4. To determine the significance of controlling for between-level variance, we examined the variability in the outcome variables across the three levels of analysis (Bliese, 2000). Therefore, we fitted two intercept-only models for each outcome variable; these models included one model with a fixed intercept and one with a random intercept, which allowed the outcome variable to vary between classrooms and students. ...
... The first aim of the study was to determine the extent to which therapists' perceptions of organizational climate were shared within supervisory teams and unique from other supervisory teams. Agreement indices capture the extent to which scores generated by different respondents are equivalent in terms of absolute value (e.g., both Therapist A and Therapist B from Supervisory Team 1 rate climate as a "3" on a five-point Likert scale; Bliese, 2000). Although historically the r wg(j) index has been used to characterize agreement for multi-item scales, it has several limitations (Brown & Hauenstein, 2005). ...
... First, r wg(j) uses a uniform null distribution (i.e., assumes all Likert scale points are equally likely to be selected) for comparison when estimating agreement. This is problematic because it does not account for rater bias (e.g., tending to use the higher end of the scale because of social desirability; Bliese, 2000). In addition, the r wg(j) index is influenced by sample size and number of points on the Likert scale (Brown & Hauenstein, 2005). ...
... For this reason, it is important to contextualize agreement indices by pairing them with reliability indices. Reliability indices capture relative consistency among respondents (e.g., In Supervisory Team 1, Therapist A rates items one through three as "2", "3", "4" and Therapist B rates items one through three as "3", "4", and "5"; Bliese, 2000). Two types of interclass correlation coefficients (ICC (1) and ICC (2)) were calculated using multilevel models (therapists nested within supervisors) with random intercepts and no predictors for each climate subscale (i.e., Mission, Cohesion, Autonomy, Stress, and Change) and total score. ...
Positive organizational climate — employee perceptions of their work environment and the impact of this environment on well-being and functioning — is associated with desirable organizational and client-level outcomes in mental health organizations. Clinical supervisors are well-positioned to impact organizational climate, as they serve as intermediaries between higher-level administrators who drive the policies and procedures and the therapists impacted by such decisions. This cross-sectional study examined the role of clinical supervisors as drivers of therapist perceptions of organizational climate within supervisory teams. Specifically, the present study investigated: (1) shared perceptions of organizational climate among therapists on the same supervisory team; (2) predictors of therapist climate perceptions. Eighty-six therapists were supervised by 22 supervisors. Indices of interrater agreement and interrater reliability of therapists on the same supervisory team were examined to determine shared or distinct perceptions of organizational climate. Multi-level models were used to examine whether supervisor attitudes towards evidence-based practices and therapist perceptions of supervisor communication predicted perceived organizational climate. Results showed perceptions of organizational cohesion and autonomy were shared among therapists on the same supervisory team and distinct from therapists on different supervisory teams. Therapist perceptions of their supervisor’s communication was positively associated with perceptions of organizational cohesion and autonomy. These findings align with emerging evidence that middle managers shape their employees’ experience of their work environment through communication strategies. These findings also point to the potential for intervening at lower organizational levels to improve overall organizational climate.
... To justify the appropriateness of aggregating the measures to the group level, we calculated within-group agreement r wg(j) (James et al., 1984) and intraclass correlations ICC(1) and ICC(2) as indicators of interrater reliability and group-mean reliability (Bliese, 2000), respectively. The average r wg(j) values were .92 ...
... and .50, respectively, indicating the appropriateness of data aggregation (Bliese, 2000;Fleiss, 2011). Although the ICC(2) of TA and TIP fell below the recommended cutoff, this can be explained by the small group size. ...
Purpose
The team collective voice (TCV) concept has received limited attention in existing research despite its potential significance, particularly within Eastern cultural contexts. Hence, this paper aims to extend the employee voice literature by conducting a comprehensive exploration of the role of TCV as opposed to individual voice (IV) in promoting team innovative performance (TIP).
Design/methodology/approach
The theoretical model is tested with survey data collected among ICT employees who work in large telecommunication organizations in Vietnam.
Findings
TCV was used more by participants than IV. Both IV and TCV exhibited noteworthy influences on TIP explained by the role of routine changes, with TCV serving as a full mediator for the relationship between IV and TIP. In addition, team autonomy (TA) also significantly enhanced TIP.
Originality/value
By measuring TCV as shared and consensus team voice, this study makes significant contributions to employee voice literature. It not only confirms the prevalence but also offers insightful explanations for why Vietnamese employees tend to favor TCV over IV. Additionally, the findings enrich the innovation literature by presenting a novel perspective on how employee voice can enhance TIP through the introduction of exogenous routine changes. It introduces a unique perspective on employee voice as a form of a search routine, distinct from Nelson and Winter's (1982) conceptualization of search, generating changes in lower-order routines, ultimately contributing to enhanced team innovation and performance.
... Before conducting regression analyses, we calculated R wg(j) [85] and intra-class correlations of ICC(1) and ICC(2) [86] to test whether the constructs (team abusive supervision, team proactive behavior, and team building) assessed by project members were appropriate to aggregate to the team level. The R wg(j) value was used to determine the interrater agreement [85]. ...
... The R wg(j) value was used to determine the interrater agreement [85]. The intra-class correlations of ICC(1) value were calculated to examine the proportion of variance in ratings attributable to team membership, and ICC(2) was used to assess the reliability of team mean difference [86]. For team abusive supervision, the median R wg(j) is 0.93, and the average R wg(j) is 0.92; ICC(1) and ICC(2) are 0.30 and 0.63, respectively. ...
While considerable attention has been devoted to positive leadership patterns in the realm of project management, the dark side of leadership has rarely been studied within project teams. To address this gap, we focus on abusive supervision in project teams and develop a team-level moderated mediation model to examine whether, how, and when abusive supervision influences project outcomes by drawing from the Proactive Motivation Theory. Survey data were collected from 132 project teams containing 132 project managers and 392 project members using a multi-source time-lagged survey design. Our findings reveal significant negative relationships between abusive supervision and both project performance and project team creativity. Furthermore, we found that a team’s proactive behavior plays a mediating role in these relationships. More importantly, our study identifies that team building mitigates the direct negative impact of abusive supervision on proactive behavior and the indirect effects of abusive supervision on project performance and project team creativity. These findings provide valuable theoretical and managerial implications for abusive supervision and project management scholars and practitioners.
... Collective efficacy is an individual-level measure, but it is a construct at the classroom level and must therefore be tested for data aggregation. In multilevel analysis, established indicators of data aggregation tests include the within-group agreement index, R wg , and the within-group reliability indices, intraclass correlation coefficient, ICC 1 and ICC 2 (Bliese, 2000). R wg assesses within-group agreement by measuring the consensus between ratings given by students in the same classroom. ...
... ICC 1 and ICC 2 evaluate the between-classroom variance to examine the reliability of classroom-level constructs obtained through aggregated student-level ratings. According to Bliese (2000), R wg > 0.70, ICC 1 > 0.05, and ICC 2 > 0.70 are indicative criteria for good data aggregation. The mean R wg is 0.95, greater than 0.70. ...
Few studies have examined the process of the relationship between social support and bystander behavior, and previous research has primarily focused on individual- or group-level correlates of bystander behavior without a full understanding of these correlates and their interactions. To address this research gap, the present study examined whether social support and moral disengagement at the individual level, as well as collective efficacy at the classroom level, were associated with bystander behavior in school bullying situations. Questionnaires were administered to 1,310 elementary school students in Grades 4–6 (Mage = 10.97 ± 0.98 years) from 61 classes in the Zhejiang Province of China. Multilevel moderated mediation analyses revealed that higher levels of social support were directly related to more defender behavior and less reinforcer and outsider behavior, and social support was also related to bystander behavior through the indirect role of moral disengagement. In classrooms with higher levels of collective efficacy to stop peer bullying, students tended to show more defender behavior and less outsider behavior. Higher levels of classroom collective efficacy did not moderate the relationship between moral disengagement and defender behavior, although it did inhibit the reinforcer and outsider behaviors of those with high levels of moral disengagement. This study sheds light on the relationship between multiple factors and bystander behavior at both the individual and classroom levels and contributes to the understanding of school bullying in the context of peer group interactions.
... We used interrater agreement (r wg (j); James et al., 1984) and intraclass correlation coefficients (ICC(1) and ICC(2); Bliese, 2000) to evaluate the suitability of aggregating individual scores of group cognitive diversity, group time pressure, social feedback-seeking climate, group recovery opportunities, and group viability. We computed r wg values for our variables and obtained median values of 0.76 for group cognitive diversity, 0.75 for group time pressure, 0.87 for social feedback-seeking climate, 0.78 for recovery opportunities, and 0.85 for group viability. ...
... We also calculated ICC(1) and ICC(2). ICC(1) values for all of our variables exceeded accepted cutoff values (ICC(1) > 0.06 (Bliese, 2000) which indicated sufficient levels of between-group variability. The F ratios associated with the ICC(1) value were statistically significant. ...
Our paper aims to advance our understanding of group non-purposeful ostracism at work by studying the causes and consequences of the non-purposeful ostracism and the conditions under which non-purposeful ostracism increases. Using network analysis on the ostracism data collected in two waves from 535 members across 73 groups, we found that group cognitive diversity and group time pressure were positively related to group non-purposeful ostracism, which in turn was negatively related to a group’s effectiveness. Group non-purposeful ostracism mediated group cognitive diversity and group time pressure in relation to group effectiveness. Based on the job demands-resources model, we hypothesized and found that recovery opportunities and social feedback climate, respectively, moderated/buffered group time pressure and group cognitive diversity in relation to group non-purposeful ostracism. Theoretical and practical contributions are discussed.
... also respectively, representing strong interrater agreement (LeBreton & Senter, 2008). Although the ICC(2) is below the recommended cutoff (.60), this value does not prevent aggregation if aggregation is theoretically justified and r wg is high (Bliese, 2000;Chen et al., 2004). ...
... .82 and .90, also respectively, representing moderate-strong (.67) and strong interrater agreement (LeBreton & Senter, 2008). Although the ICC(2) is below the recommended cutoff (.60), this value does not prevent aggregation if aggregation is theoretically justified and r wg is high (Bliese, 2000;Chen et al., 2004). ...
... We aggregated each team member's rating of abusive supervision to the team level and treated it as a team-level construct (Chan, 1998). To examine the appropriateness of aggregating abusive supervision at the team level, we calculated within-group interrater reliability using the within-group interrater agreement measure (i.e., Rwg) and the intra-class correlation (ICC) coefficients (Bliese, 2000;James et al., 1984). We followed James et al.'s (1984) method to calculate Rwg(j) and identified 13 teams with negative Rwg(j) values. ...
... The ICC(1) value was greater than the recommended value of 0.059 (Cohen, 1988), which indicated high agreement on experienced abusive supervision among team members within a team. The ICC(2) value was relatively low, which can be explained by the relatively small number of responses per group (average = 4.36; Bliese, 2000). However, the aggregation of abusive supervision is still appropriate, as abusive supervision can be theoretically examined as a group-level construct (e.g., Liu et al., 2012), and other within-group agreement indices (e.g., high Rwg(j) and ICC(1) values) also support the appropriateness of aggregation (Chen & Bliese, 2002). ...
Although prior research has documented a divergent relationship between leader Machiavellianism and abusive supervision, it fails to uncover the underlying mechanisms of this relationship. Drawing from trait activation theory as the overarching theory, we develop and test a dual-path model to examine how and when leader Machiavellianism leads to abusive supervision. Specifically, we theorize leader perceived threat to hierarchy (power-threatening process) and perceived power dependence on subordinations (power-sustaining process) as two parallel mechanisms through which leader Machiavellianism affects abusive supervision. We further identify leader position power as a boundary factor that influences the power-threatening and power-sustaining processes. Using multi-wave, multi-source data collected from 175 supervisors and their 763 subordinates, we found that Machiavellian leaders were more likely to perceive high threats from subordinates to the existing hierarchy, though this threat perception was not significantly associated with abusive supervision. Additionally, Machiavellian leaders were also more likely to perceive high power dependence on subordinates, which in turn reduced their abusive supervision. We further found that leader position power strengthened the positive effect of leader Machiavellianism on leader perceived threat to hierarchy, but did not weaken the positive effect of leader Machiavellianism on leader perceived power dependence on subordinates. The implications of our findings are discussed.
... first, engagement conceptualized and operationalized at the individual level is functionally isomorphic to collective engagement at the department level. Therefore, to ensure that engagement was acceptable for aggregation at the department level based on Bliese(144), we assessed the ICCs of the levels. The ICC1 value is .276, ...
... ICC(1) values indicate the proportion of between-group variance to total variance attributed to group membership, whereas ICC(2) values measure the reliability of average team perceptions (Klein and Kozlowski, 2000). We expected ICC(1) scores different from zero, with values close to 0.20 interpreted as high scores (Bliese, 2000). Glick (1985) proposed that ICC(2) values over 0.60 reflect high scores. ...
Family doctor teams, serving as health gatekeepers, are extensively advocated in China. Their composition, comprising a heterogeneous mix of professionals, contributes to a more comprehensive service, but also poses challenges. Consequently, scholarly interest has arisen in comprehending how these compositions, known as faultlines, influence team dynamics and outcomes. However, there is a lack of comprehensive exploration into how faultlines influence team members’ communication processes and knowledge sharing. This study aims to provide insights into the associations between faultlines in primary care teams and team performance, specifically exploring how knowledge sharing may mediate these effects, with the goal of revealing key insights to optimize contracted family doctor services. Survey data from 291 family doctor teams in China was utilized to test hypotheses, revealing a negative association between (social-category and information-based) faultlines and knowledge sharing. Team knowledge sharing acts as a mediator in the relationship between these faultlines and team performance. Our findings advance faultlines theory and emphasize the mediating role of knowledge sharing in elucidating the interplay between faultlines and team performance. These insights are crucial for fostering collaboration, managing faultlines, and enhancing healthcare team performance.
... first, engagement conceptualized and operationalized at the individual level is functionally isomorphic to collective engagement at the department level. Therefore, to ensure that engagement was acceptable for aggregation at the department level based on Bliese(144), we assessed the ICCs of the levels. The ICC1 value is .276, ...
Perceived HPWSs (PS-HPWSs) are increasingly indicated for understanding the relationship between organizational HPWSs and employee-level outcomes. However, the mediating effect of perceived HPWSs is inconsistent and unclear. Hence, relying on integrated HPWS strength and signal theories, we tested whether PS-HPWSs positively mediated the relationship between MR-HPWSs and engagement with data collected from multiple and multilevel sources of 102 department heads and 360 employees of three Ethiopian public higher education institutions from June 03/2023 to September 10/2023. We also tested the positive moderation impact of procedural justice between MR-HPWSs and PS-HPWSs and its conditional effect on engagement using the MLmed Beta macro SPSS package. Therefore, procedural justice moderates the mediation of PS-HPWSs between manager-rated HPWSs and employee engagement. Hence, this study will address the inconsistency of PS-HPWSs between manager-rated HPWSs through the moderation of procedural justice (one of nine features of HPWS strength); this in turn has an ample effect on engagement. Accordingly, further research should include one or more of the nine HPWS features as moderators of the mediating effect of PS-HPWSs.
... The ICC value in our dataset was 0.139, indicating that 13.9% of the total variance resided at the country level for each social entrepreneurial activity. This is within the normal range for management research and indicates that country-level variance is significant, justifying the application of multilevel regression (Bliese, 2000;Hox et al., 2017). We tested for multicollinearity using the variance correlation estimator (VCE) and found that the highest correlation between any two variables was 0.002. ...
Scholars tend to evaluate the effects of cultural factors on social entrepreneurial activity based on either cultural values or cultural practices. However, societal inconsistencies between values and practices have the potential to create uncertainty in expected entrepreneurial behaviors. In this paper, we operationalize cultural dissonance as the gap between cultural values and cultural practices and draw on role congruity theory to theorize and test how cultural dissonance influences engagement in social entrepreneurship. Using data from the Global Entrepreneurship Monitor, the Global Leadership and Organizational Behavior Effectiveness, and the World Bank datasets (N = 23,828), we show that cultural dissonance can either impede or encourage social entrepreneurial activity and that female entrepreneurs are less sensitive to these effects than male entrepreneurs.
... and ICC(2) = .69 in the current study. These results satisfy the criteria for within-team agreement (LeBreton & Senter, 2007) and inter-member reliability (Bliese, 2000), providing support for aggregating grateful climate scores to the team level. Participants rated each item on a scale ranging from 1 (strongly disagree) to 6 (strongly agree). ...
... Martončik, 2019) ale zároveň je aj postupom ako je možné agregovať dáta pri viacúrovňovom výskume (napr. Bliese, 2000;Chan, 1998;Kozlowski a Klein, 2000;LeBreton a Senter, 2008). ...
Interrater agreement is one way to establish reliability (and also validity) in social science research. The traditionally preferred method of measuring interrater agreement is the descriptive approach owing to its simplicity. This approach is also associated with a number of different agreement indices, which makes it difficult to select the right index. This article summarises theoretical definition on the prevailing approach used to measure interrater agreement (in both quantitative and qualitative research). From a practical point of view, the article focuses on the possibilities of measuring agreement by using percent agreement, the kappa coefficient, and the AC1 coefficient. A more detailed description of the indices explains how to define, calculate, and interpret them and the problems associated with their use. The indices are then discussed in comparison. Although
underestimated and criticised, percent agreement may be a good indicator of interrater agreement. Several paradoxes accompany the use of the kappa coefficient, which is only possible under certain conditions. The appropriate alternative to it is the AC1 coefficient. The article concludes with a summary of recommendations for improving the quantification of interrater agreement.
... and leader level (r wg(j) = .67; Bliese (2000). Prior to the analysis, we z-transformed (using the STANDARDIZE command in Mplus) all variables except for the outcomes to ease the presentation of the cross-level interactions. ...
... One might argue that the data in this study could be analyzed using a two-level or multilevel analysis due to medical students' repeated responses regarding their burnout and compassion satisfaction longitudinally with respect to their demographics, personal quality, and contexts. Since the intraclass correlation coefficients for medical students' burnout and compassion satisfaction (0.003 and 0.004, respectively) in individual specialty rotations were smaller than 0.05, the minimum cut-off point for performing multilevel analysis (33,34), multilevel effects were ignored in this dataset. ...
Introduction
In medical education, the clerkship phase is a demanding period during which medical students learn to navigate the responsibilities of medical school and clinical medicine. Grit, a personal quality regarded as a non-cognitive trait, refers to perseverance and passion; specifically, it represents the ability to endure hardship and work industriously toward a goal. Most studies analysed grit as a single concept and few studies have investigated the effect of grit on the well-being of medical students through the whole-specialty training (i.e. surgical and non-surgical specialty rotations) required in clinical clerkships. Therefore, this study investigated whether associations exist between medical students’ grit, measured by the two subconstructs of perseverance and passion, and their well-being during clerkships in surgical and non-surgical specialty units.
Methods
This one-year prospective web-based questionnaire study enrolled fifth-year medical students at a tertiary medical centre in central Taiwan between September 2017 and July 2018 in their first-year clerkship. The students’ sex, age, and grit were measured at the start of their clerkship. Routine surveys were conducted over one year to assess burnout and compassion satisfaction for students’ well-being, and the training specialty characteristics of the surgical and non-surgical specialty departments were recorded. This study included 92 medical students and 1,055 survey responses from individual specialty rotations. Descriptive, univariate and multivariate analyses were performed.
Results
Our results revealed that medical students’ perseverance, as part of grit, was related to lower burnout and higher compassion satisfaction during clerkships, but not the subconstruct of passion. Moreover, the positive trait of perseverance measured in our study had greater explanatory power for compassion satisfaction than for burnout. Furthermore, the results revealed that older medical students suffered from less burnout than their younger counterparts, and that male medical students expressed higher compassion satisfaction than their female counterparts.
Discussion
Perseverance, as a subconstruct of grit, is a positive personal quality for medical students’ clerkships, and methods driving the cultivation of perseverance in medical education should be considered. In addition, even though positive traits such as perseverance equipped medical students for compassion satisfaction, additional factors attributed to medical students’ burnout must be identified.
... ICC coefficient measures the extent to which the importance of clustering regarding a specific variable is defined in terms of variance (Leyland et al., 2020). Table 4 provides the results of Muthén's ICC that suggest sufficient variations in variance proportion between cluster variations required for multilevel analysis (Bliese, 2000). ...
Knowledge hiding has emerged as a real challenge in today's knowledge economies, curbing an organization's ability to attain sustainable growth and adversely affecting individuals and organizational culture, performance, and retaining talent. This study explores the effect of perceived negative workplace gossip (PNWG) on supervisors' knowledge-hiding (SKH) behavior and how supervisors' knowledge-hiding behavior trickles down into the behaviors of subordinates. Based on the conservation of the resource and social cognition as a theoretical lens, this study examines the relationship between PNWG, SKH, and subordinates' knowledge-hiding from co-workers (SKHC) with a mediating role of subordinates' moral disengagement (MD) using a multilevel approach. Data were collected from 108 supervisors and 492 subordinates of employees of small and medium enterprises (SMEs). Mplus software was used to analyze the multilevel relationships. The results of this study reveal that PNWG influences SKH; SKH is positively associated with SKHC; finally, we found that MD mediates the relationship between SKH and SKHC. This study brings new insights by uncovering the relationships between the studied variables using a multilevel approach and the underexplored trickle-down effects of knowledge-hiding in SMEs. This study has important theoretical implications and practical suggestions for organizations and human resources managers to develop policies and procedures to overcome this issue and make organizations and individuals more productive and valuable.
... collected questionnaire data from individuals about their LMX perceptions with their leaders and about their individual levels of psychological strain. As all LMX constructs and psychological strain showed notable shares of variance at the team level (see Table 2), the use of multilevel modeling seems justified (Bliese, 2000). Following recommendations, we centered our predictor variable on the team mean and added the team mean on Level 2 (see Enders & Tofighi, 2007). ...
Due to social comparison in the team, leader-member exchange (LMX) differentiation has been shown to play a crucial role for explaining follower outcomes. LMX differentiation may be beneficial for some followers, but also detrimental for others, when perceived as unfair. In order to understand the impact of LMX differentiation on work-related outcomes, LMX differentiation should be described by three properties (relative LMX position, LMX variability, and team-level LMX). The aim of the study is to focus on LMX differentiation and test relationships between different properties of LMX and followers’ psychological strain. In sum, 75 teams with 322 followers answered an online survey about their leaders’ LMX and their individual levels of psychological strain. Results of multilevel modeling showed that the relative LMX position was negatively related to psychological strain on the individual level. This relation was moderated by LMX variability and team-level LMX. However, we did not find significant relations between team-level LMX and team-level psychological strain. The present study extends previous research by looking at relations between LMX and psychological strain on an individual follower level as well as on the team level. Our results suggest that the relative position of LMX within a team seems to play a crucial role for individual follower psychological strain. Results imply that leaders should aim to improve LMX relationships within their team and to reduce differences between team members. Leadership trainings should impart knowledge and skills to improve LMX quality and include elements about LMX differentiation, its origins, as well as consequences.
... TikTok, Weibo, LinkedIn) to support relationship building, communication and sales and marketing (Iankova et al., 2019). To measure TMT SMU, we reviewed existing scales of teams' SMU (Ding et al., 2019) (Bliese, 2000;James et al., 1984). ...
Firms are increasingly applying multiple information technologies (ITs) to pursue corporate entrepreneurship. However, how different ITs can be leveraged collectively to achieve corporate entrepreneurship remains largely underexplored. Drawing upon the knowledge‐based view, we develop an integrative theoretical model to delineate how three types of commonly deployed IT (i.e. organisational enterprise systems use, organisational social media use and top management team social media use) can be used collectively in the knowledge acquisition and utilisation processes to achieve corporate entrepreneurship. Specifically, we posit that organisational IT uses (i.e. organisational enterprise systems use and organisational social media use) enhance openness to external knowledge in the knowledge acquisition process, which in turn serves as the mediation mechanism to channel organisational IT uses into corporate entrepreneurship. Moreover, top management team IT use (i.e. top management team social media use) shapes the effect of openness to external knowledge on corporate entrepreneurship in the knowledge utilisation process. The empirical analysis using multi‐respondent survey data from 1252 managers in 313 firms provides strong support for the proposed model. This study contributes to the information systems research on how IT influences corporate entrepreneurship by distinguishing the roles of organisational enterprise systems use, organisational social media use and top management team social media use in enhancing corporate entrepreneurship through firms' openness to external knowledge.
... Finally, once the fieldwork was completed, usable data were obtained from 176 CEOs and 393 team supervisors (input was obtained from at least two team supervisors, and the mean team supervisor evaluations per hotel was 2.23). Supervisor evaluations were averaged and aggregated to form a single evaluation of CEO servant leadership for each hotel (n = 176 hotels), once all the established criteria had been met to ensure proper data aggregation (Bliese, 2000;James et al., 1984). In our case, the rwg index (James et al., 1984), which measures the level of equality of response among supervisors, was above 0.70 (rwg = 0.75), and the intra-class correlation indices -ICC(1), ICC(2)-were above 0.5 (ICC(1) = 0.57, ICC(2) = 0.78), and so there was a strong intragroup agreement and a high level of reliability of the data. ...
Firm innovation is a characteristic of firms with a robust entrepreneurial orientation that wish to have a strong competitive position in their sector. Drawing on the theories of servant leadership and upper echelons, this study aims to highlight the mechanisms through which CEO servant leadership enhances firm innovation. This paper examines (1) knowledge absorptive capacity as a mediator in the CEO servant leadership–firm innovation relationship, and (2) the moderating role of social capital (internal and external) in this relationship. Original data from 176 large and medium-sized hotels in the Canary and Balearic Islands (Spain) were used to test the hypotheses, and a partial least squares (PLS) analysis using Smart PLS 4 revealed that CEO servant leadership was positive for firm innovation, which was, in turn, mediated by firm knowledge absorptive capacity. Furthermore, the relationship between CEO servant leadership and firm knowledge absorptive capacity was found to grow stronger as internal social capital increased, while external social capital strengthened the positive link between firm knowledge absorptive capacity and firm innovation. New light is therefore shed on how and when CEO servant leadership is most effective in fostering firm innovation. Additionally, our firm-level findings contribute to the servant leadership literature, which is dominated by individual and group-level research, and expand the mounting evidence for the positive impact of CEO servant leadership on firm-level outcomes.
... This pattern indicates that there was substantial within-group agreement in ratings of leader empathy, meeting commonly accepted criteria for justifying aggregation (cf. Bliese, 2000;LeBreton & Senter, 2008;LeBreton et al., 2023). ...
This article provides an empirical examination of the popular belief that empathy has become more important to leadership since the onset of the COVID-19 pandemic. Three research questions were addressed using data for over 5,000 U.S.-based managers and executives in for-profit businesses rated between January 2019 and September 30, 2022. The questions concerned changes since 2019 in the prevalence of observed empathetic behavior, including the possibility of leaders demonstrating too much empathy, and the effects of leader empathy on leader effectiveness, employee engagement, and team and business-unit productivity. The results were consistent with claims that empathy has become more important, particularly in terms of a stronger relationship with employee engagement since the onset of the pandemic. The costs of too much empathy in terms of reduced engagement were also less severe since the pandemic. The relationship between leader empathy and productivity also changed: Less empathetic leaders drove productivity higher in the prepandemic sample but have undermined productivity since the 1st year of the pandemic. Also since that 1st year, leaders who were excessively empathetic undermined productivity as well and to a greater degree than leaders who lacked empathy. The pattern of results points to an increasingly difficult balancing act for empathetic leaders in optimizing both organizational culture and business results.
... indicate a small effect, ICC1 values of .10 a medium effect, and ICC1 values of .25 a large effect (LeBreton & Senter, 2008), both ICC1 values can be considered as medium effects. Based on the ICC1 and the number of units, it is possible to calculate the ICC2 (Bliese, 2000). Values above .70 ...
... Organizational researchers in the past have utilized the intraclass correlation, Type 1 (ICC1) to evaluate the degree to which groups have shared perceptions (e.g., O'Neill & Allen, 2014). The ICC1 can be defined as a function of the group (σ τ 2 ) and residual variance (σ e 2 ) from an intercept-only multilevel mixed-effects model (Bliese, 2000), ICC1 = σ τ 2 /(σ τ 2 + σ e 2 ). In principle, the ICC1 could be estimated each time period to reflect patterns of consensus such that a pattern of increasingly large ICC1 values over time would be indicative of consensus and a pattern of decreasing ICC1 values would be indicative of dissensus. ...
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.
... Intraclass correlation coefficients (see Table 3) indicated a notable share of variance at the team and organizational level for all constructs, which justifies the aggregation approach to form the collective constructs (e.g. Bliese, 2000). ...
Drawing on the conservation of resources (COR) theory, we analyze multilevel restructuring impact perceptions and their association with emotional exhaustion. We explore whether distributive justice at the individual (perceived distributive justice), team, and organizational levels (distributive justice climate) moderates the association of restructuring impact and emotional exhaustion. In total, 1523 employees, nested in 166 teams and 26 organizations, participated in our study. To test our hypotheses, we used Bayesian multilevel modelling. We found positive associations between restructuring impact and emotional exhaustion on all three organizational levels. Distributive justice at the team level cross-level moderated the relationship between restructuring impact and emotional exhaustion at the individual level. Our study contributes a multilevel understanding of restructuring impact, possible consequences for emotional exhaustion, and the moderating role of distributive justice. Furthermore, it substantiates COR theory’s caravan passageway notion, which as yet lacks empirical support.
... and Rwg(j) = .94) (Bliese, 2000;James et al., 1984). ...
Relationship‐oriented human resource management (HRM) contributes to organizational functioning by promoting employee relationships, coordination, and cooperation. We propose unit‐level positive affective climate as a motivational mechanism through which relationship‐oriented HR systems can positively influence unit service quality and relationships with beneficiaries, and prevent individual emotional exhaustion. Moreover, we propose collective occupational calling serves as an alternative motivational source that can substitute for the positive effects of positive affective climate. In analyzing a sample composed of 742 nurses from 48 nursing units of two hospitals in China with data collection at three time points before and after the COVID‐19 outbreak, our results supported the substituting effects of unit‐level collective occupational calling on positive affective climate stemmed from relationship‐oriented HR systems. Our study connects strategic HRM and motivation research by shedding light on an affective mechanism from relationship‐oriented HR systems and the contingencies involving employees' various sources of motivation, such as occupational calling. We further discuss theoretical and practical implications of the research.
... Moreover, to arrive at valid conclusions, it is important to measure and analyze retirement-related variables, as well as their potential antecedents and consequences, in correspondence with their intended conceptual level (Chan, 1998). For example, data on a shared "organizational retirement climate" can be collected from individual workers, but the data need to be aggregated to the organization level based on adequate values of aggregation indices, such as intraclass correlation coefficients (Bliese, 2000). Importantly, the expanded multilevel perspective on retirement does not comprehensively specify all relevant retirement-related variables that must be considered theoretically and empirically at each level. ...
Retirement is an inherently multilevel phenomenon and, therefore, research on retirement should consider factors at the macro level (i.e., retirement as an institution in different countries), meso level (i.e., organizational retirement policies, practices, and cultures), and micro level (i.e., the individual retirement process including individuals’ behaviors and experiences related to the retirement transition). The first goal of this chapter is to take stock of four existing conceptual multilevel frameworks of retirement. Second, this chapter reviews empirical studies that have explicitly used a multilevel approach to examine cross-level main and moderation effects to predict micro level retirement outcomes, with a focus on studies published within the last decade. Third, drawing from, integrating, and advancing this literature, this chapter outlines an expanded multilevel perspective on retirement. This expanded perspective entails the conceptualization of (a) antecedents and consequences of retirement variables at the macro, meso, and micro levels, (b) between-unit differences and within-unit changes at each level, and (c) cross-level main and moderation effects of retirement variables at different levels. Finally, the chapter concludes with directions for future research adopting an expanded multilevel perspective on retirement.
... We used the remaining three practices from the MQ as workplace-level measures. To ensure interrater reliability when aggregating employee-to workplace-level variables, we calculated three statistics: two inter-rater reliability indices (Intra-Class Correlation, ICC (1), ICC (2)) (Bliese, 2000) and one inter-rate agreement index (R wg ) (James, 1982). ...
Existing research examines the impact of human resource (HR) practices on employee wellbeing by considering each practice in isolation or multiple practices as a bundle, focusing on linear associations. Drawing on the too‐much‐of‐a‐good‐thing (TMGT) meta‐theory, we examine possible nonlinear effects of Ability‐Motivation‐Opportunity (AMO) sub‐bundles on job satisfaction and job stress. We, also, examine boundary conditions on whether and how the nature of the identified curvilinear associations varies across employees in high‐, medium‐, and low‐skilled occupations. Using data from the Workplace Employment Relations Study (WERS2011), we uncover an inverse U‐shaped association between motivation‐enhancing (ME) practices and job satisfaction and a U‐shaped association between opportunity‐enhancing (OE) practices and job stress. No evidence of a curvilinear ability‐enhancing (AE) practices‐wellbeing association emerges. Additionally, occupational differences in skills levels moderate the curvilinear ME practices‐stress association. Likewise, occupational skills differences moderate the associations between OE practices and job satisfaction, and work stress. There is no suggestion that occupational differences moderate the AE practices‐wellbeing association. These findings underline the contingent nature of the TMGT effect and call for a more nuanced investigation of the HR‐wellbeing association.
... They are also used to argue for or against applying multilevel model specifications (e.g. intra-class correlation [ICC]) (Bliese, 2000;Bliese et al., 2018;Boyer & Verma, 2000). This illustrates the predominant use of agreement (and interrater reliability) measures for methodological reasons focusing on data and the robustness of methods, rather than on theory development and testing (Loignon et al., 2019). ...
We start from the observation that agreement metrics are mainly used in group and organization research for methodological reasons only. Therefore, we think there are many missed opportunities to study within-group agreement and consensus as core theoretical concepts in group and organization management. We muse in this article about opportunities to rely on agreement metrics for centerpiece research questions, beyond mere methodological argumentations about data selection and technical model specifications. We structure our considerations in three guiding questions for scholars who are interested in multi-level group and organization research. To initiate a good and constructive scientific debate on this – and thus deliver what is expected from a GOMusing article – we hope for a fair amount of disagreement, but not too much, in the group and organization research community.
... Although the respondents are nested within 102 supervisors or 102 organizations, we tested the hypotheses at the individual level. To do so, we followed Bliese's (2000) recommendations and estimated ICC (1) values for adaptive-selling behavior and salesperson resilience. The ICC (1) values for adaptive-selling behavior and salesperson resilience were 0.007 (ns) and 0.01 (ns), respectively, showing that non-independence was not a problem in our data. ...
This study investigates the role of spiritual leadership in shaping salesperson adaptive-selling behavior, a crucial factor for enhancing customer satisfaction and loyalty. Despite its significance, research on the leadership-adaptive selling link remains limited. Grounded in the conservation of resources theory, this study predicts that spiritual leadership is positively associated with adaptive selling. The present work also examines the mediating and moderating roles of salesperson trait resilience and salesperson political skill, respectively. Our results based on data from 505 supervisor-salesperson dyads corroborated our predictions. Specifically, the results reveal that spiritual leadership positively impacts adaptive selling directly and via resilience. Further, our findings illustrate that salespeople high on political skill benefit more from spiritual leadership to enhance their resilience and adaptive selling. This study provides actionable insights for managers to foster their salespeople's resilience and promote adaptive selling to improve organizational effectiveness.
... Where rY1,Y1 and rY2,Y2 represent reliabilities for mean levels of men and women, respectively. These were estimated as ICC(2) based on the variance partition coefficient calculated with equation S4 following procedures presented by Bliese (2000). ...
Studies advancing the hypothesis of a “gender-equality paradox” have found that societies with more gender equality demonstrate larger gender differences across a range of phenomena. In doing so, they rely on that practice of predicting an algebraic difference score—calculated from mean scores for men and women across a set of countries—with an index of gender equality, or some related concept. We argue that direct difference score predictions of this type are impossible to interpret because very different combinations of constituents—mean scores of men and women and properties of these means—can produce identical direct difference score predictions. We reanalyzed three large cross-cultural datasets with fifteen variables from three different domains—attitudes towards science and technology, economic preferences, and personality traits—to showcase our method of deconstructing difference score predictions and to investigate to what extent the rhetoric of the gender-equality paradox describes a real phenomenon. The results were highly heterogeneous. For some characteristics, men’s and women’s country-level means varied identically as a function of country-level gender equality (no paradox). For other characteristics, there were differences in how men’s and women’s means varied. Whether these differences could be described in the rhetoric of the paradox varied. More pertinent is the necessity of deconstructing difference score predictions into their constituent components before attempting to answer questions regarding a paradox. It is in the terminology of these components and their properties that future hypotheses should be tested.
... Max = .51) (Bliese, 2000). The Cronbach's alphas across the five waves were .93, ...
How can teams sustain their team voice behavior over time? Drawing from meta-perceptions and the collective action perspective, we propose a reciprocal relationship between collective perception of voice importance for leaders (i.e., the extent to which team members infer or pre-judge leaders’ importance assessment of team voice) and team voice. Using five-wave longitudinal field data from 554 individual team members of 71 teams, our results supported such a reciprocal relationship over time. Our findings shed new light on how team members’ inference or pre-judgement of their team leader’s attitudes towards team voice affects subsequent team voice. In turn, the team voice reversely shapes the subsequent collective perception of voice importance for leaders. Our findings provide theoretical and practical implications for how team members can sustain team voice.
... Work group supervisors at each hotel rated their general manager using 14-item scale designed by Ehrhart (2004). A firm-level measure was created by averaging the ratings as supported by within-firm agreement (rwg = 0.75, SD =.15; James et al., 1984), ICC (1) = 0.57, and ICC (2) = 0.78 (Bliese, 1998(Bliese, , 2000. Additionally, analysis of variance indicated significant between-firm differences in the average level of SL among general managers (F = 8.63, p < .01). ...
Studies advancing the hypothesis of a “gender-equality paradox” have found that societies with more gender equality demonstrate larger gender differences across a range of phenomena. In doing so, they rely on that practice of predicting an algebraic difference score—calculated from mean scores for men and women across a set of countries—with an index of gender equality or some related concept. We argue that direct difference score predictions of this type are impossible to interpret because very different combinations of constituents—mean scores of men and women and properties of these means—can produce identical direct difference score predictions. We reanalyzed three large cross-cultural data sets with 15 variables from three different domains—attitudes toward science and technology, economic preferences, and personality traits—to showcase our method of deconstructing difference score predictions and to investigate to what extent the rhetoric of the gender-equality paradox describes a real phenomenon. The results were highly heterogeneous. For some characteristics, men’s and women’s country-level means varied identically as a function of country-level gender equality (no paradox). For other characteristics, there were differences in how men’s and women’s means varied. Whether these differences could be described in the rhetoric of the paradox varied. More pertinent is the necessity of deconstructing difference score predictions into their constituent components before attempting to answer questions regarding a paradox. It is in the terminology of these components and their properties that future hypotheses should be tested.
This chapter introduces the framework and causal model of organizational culture, organizational climate, knowledge management, and job performance related to business process orientation. It argues that dimensions of organizational culture, organizational climate, and knowledge management have mediated positive effect on job performance. Knowledge management positively mediates the relationships between organizational culture and job performance and between organizational climate and job performance. Organizational culture is positively related to organizational climate. Furthermore, the author hopes that understanding the theoretical constructs of organizational culture, organizational climate, knowledge management, and job performance through the use of the framework and causal model will not only inform researchers of a better design for studying organizational culture, organizational climate, knowledge management, and job performance, but also assist in the understanding of intricate relationships between different factors.
Purpose
Although employee overqualification is a common occurrence in the workplace, most research has focused on overqualification at the individual level rather than at the team level. Drawing on social cognitive theory, this study aimed to uncover how leaders' perception of team overqualification influenced their cognition and follow-up behavior.
Design/methodology/approach
We performed two studies to test our model. In Study 1, we conducted an experiment to examine the causal relationship between leaders' perception of team overqualification and leadership self-efficacy. In Study 2, a two-wave field study was conducted to test the overall model based on a sample obtained from a steel company in China.
Findings
We found that leaders' perception of team overqualification reduced leadership self-efficacy, which in turn hindered leaders' empowering behavior. In addition, leaders' social face consciousness strengthened the negative relationship between leaders' perception of team overqualification and leadership self-efficacy, such that the relationship was more negative when leaders' social face consciousness was high rather than low.
Originality/value
Our study contributes to the literature on employee overqualification and its effects on leaders through investigation at the team level to show how leaders respond to overqualified teams.
Based on conservation of resources theory and the work–home resources model, this study examines how and when narcissistic leadership influences employees’ change-oriented organizational citizenship behavior. A total of 363 employees from 61 teams across numerous enterprises based in central China were surveyed using a questionnaire. The study hypotheses were tested using structural equation modeling and Monte Carlo simulation analysis. The findings revealed that narcissistic leadership results in the development of a negative team climate, termed “team chaxu climate,” which, in turn, hinders employees’ change-oriented organizational citizenship behavior. Furthermore, this study explored the moderating role of leaders’ family affective support in the relationship between narcissistic leadership and team chaxu climate. This study contributes to our understanding of the relationship between narcissistic leadership and employee organizational citizenship behavior and empirically validates the work–home resources model.
This study examines the role of marital beliefs, marital virtues, and neighborhood cohesion in marital satisfaction among 386 participants in Addis Ababa, Ethiopia. The study used a descriptive correlational study design. Participants were selected using multistage random sampling techniques, and data were collected using a questionnaire. Multilevel modeling was employed to analyze the data. Findings indicated that all marital beliefs, marital virtues, and neighborhood cohesion were positively and significantly associated with marital satisfaction. The findings from multilevel modeling showed that marital virtues and neighborhood cohesion significantly predicted marital satisfaction at both individual and neighborhood levels, whereas marital beliefs predicted marital satisfaction only at the individual level. Besides, marital virtues moderated the association between shared neighborhood cohesion and marital satisfaction, such that participants who reported lower marital virtues had a lower level of marital satisfaction when their neighbors experienced lower shared neighborhood cohesion. The interactive effect of shared marital virtues and marital beliefs can also influence marital satisfaction when participants reported low levels of marital belief and from a neighbor of fewer marital virtues; they tend to report low levels of marital satisfaction. The study concludes that the participants' positive beliefs about marriage, marital virtues, and cohesive relationships with their neighbors positively influence their marital satisfaction. Marriage practitioners, policymakers, and community workers might benefit from this study. Limitations and future directions are also discussed.
Engaging in learning with colleagues is paramount for teachers to enhance students’ learning. To be effective and sustainable, it requires engagement both by individual teachers as well as whole teams. Surprisingly however, research examining learning activity as a characteristic of individuals in teams and a property of teams, as well as their antecedents on both levels, is still scarce. This study examined how perceptions of transformational leadership, as mediated by goal interdependence and collective-efficacy, facilitate teachers’ engagement in information sharing, and whether effects are similar for individuals and teams. Questionnaire data were analyzed using multilevel structural equation modelling. Results showed that individual engagement in information sharing and individuals’ perceptions of goal interdependence were directly impacted by transformational leadership practices. Information sharing of teams, and collective efficacy, were impacted by consistent perceptions of goal interdependence. Teachers shape their learning amongst themselves in teams, with individual support from their leader.
This study addresses an understudied research area in current voice research—the employee voice gap, defined as the extent to which employees perceived actual voice falls short of their demand for voice. Drawing on social exchange theory, we propose that the employee voice gap can be a significant reason for turnover intention, and such a relationship is mediated by employees’ perceptions of working conditions and industrial relations climate. We also argue that these relationships exist at the firm level and that the aggregated voice gaps are associated with organizational turnover rates. We provide empirical evidence for the relationships using the 2019 China Employer-Employee Matched Survey data of 4602 employees from 301 firms. The findings of our study point to similar voice gap patterns to those found in the North American context, indicating the universal effects of workplace voice gaps. In addition, our analysis reveals that having more voice than needed is not associated with these outcomes. Our study enriches the employee voice literature in general and in the Chinese context specifically through an interdisciplinary lens.
In the most comprehensive population surveys, mental health is only broadly captured through questionnaires asking about “mentally unhealthy days” or feelings of “sadness.” Further, population mental health estimates are predominantly consolidated to yearly estimates at the state level, which is considerably coarser than the best estimates of physical health. Through the large-scale analysis of social media, robust estimation of population mental health is feasible at finer resolutions. In this study, we created a pipeline that used ~1 billion Tweets from 2 million geo-located users to estimate mental health levels and changes for depression and anxiety, the two leading mental health conditions. Language-based mental health assessments (LBMHAs) had substantially higher levels of reliability across space and time than available survey measures. This work presents reliable assessments of depression and anxiety down to the county-weeks level. Where surveys were available, we found moderate to strong associations between the LBMHAs and survey scores for multiple levels of granularity, from the national level down to weekly county measurements (fixed effects β = 0.34 to 1.82 ; p < 0.001). LBMHAs demonstrated temporal validity, showing clear absolute increases after a list of major societal events (+23% absolute change for depression assessments). LBMHAs showed improved external validity, evidenced by stronger correlations with measures of health and socioeconomic status than population surveys. This study shows that the careful aggregation of social media data yields spatiotemporal estimates of population mental health that exceed the granularity achievable by existing population surveys, and does so with generally greater reliability and validity.
Purpose
The rise of communication software has changed our work style. The objectives of this study are: (1) to explore the effect of supervisors making after-hours work requests using communication software (SWRUCS) on employees’ job stress, quality of life and (2) to examine the moderating effect of personality traits and the cross-level contextual effect of social support.
Design/methodology/approach
A questionnaire survey was conducted to obtain information from 357 employees.
Findings
The results suggested that SWRUCS exacerbated job stress, which negatively impacted on quality of life and well-being. Moreover, different personality traits can either increase or decrease the positive or negative effect of SWRUCS on job stress. This study also revealed that social support can reduce employees’ job stress in a cross-level fashion. Furthermore, social support, especially organizational and supervisory support, can decrease the negative effect of job stress on employees’ quality of life and well-being.
Originality/value
Theoretically, this study has broadened the research scope of the organizational application of communication software, and practically, this study has demonstrated the reason why organizations should provide social support and select employees with suitable personality traits.
Employees in high-risk occupations like the military are often provided resilience training as a way to improve mental health and performance. This training typically reflects a one-size-fits-all model, even though employees likely differ in their readiness to receive resilience training. Borrowing from the readiness to change literature, the present study examined whether employees could be categorized in terms of their readiness to receive resilience training and whether this categorization was related to perceptions of the utility of resilience training, as well as self-reported resilience and mental health symptoms. Data were collected with an anonymous survey of 1,751 U.S. soldiers in a brigade combat team. Survey items assessed readiness for resilience training, self-reported resilience, mental health symptoms, and perceptions of unit-based resilience training. Following a factor analysis that identified three categories underlying readiness for resilience training (pre-contemplation, contemplation, and action), a finite mixture analysis resulted in the identification of four classes: receptive (71%), resistant (16%), engaged (9%), and disconnected (4%). In a sub-set of the sample (n = 1054) who reported participating in unit-based resilience training, those in the engaged class reported the most positive evaluations of the program. Relative to the other three classes, soldiers in the engaged class also reported the highest level of resilience and fewest mental health symptoms. Thus, those least receptive to resilience training may have been those who needed it most. These results can be used to tailor resilience interventions by matching intervention approach to the individual’s level of readiness to receive the training.
ResearchGate has not been able to resolve any references for this publication.