Multilevel Structural Equation Modeling Research Articles

Revisiting happiness at work within the hospitality sector: revealing the power of ethical leadership and group diversity

PurposeIndividual happiness at the workplace allows hospitality organizations to foster an optimistic work environment and endorse the well-being of their employees, which is essential for achieving long-term success. While ethical leadership has been verified as an interpreter of happiness, however, enhancing employee happiness at work remains a gap. The current study seeks to fill this gap by utilizing a multilevel model of the direct and indirect relationship between ethical leadership and happiness at work and also the mediation of group diversity in this relationship.Design/methodology/approachThe study involved a cross-sectional study with a quantitative methodology. The authors collected data from 36 Jordanian hospitality teams. A total number of 960 participants were selected using the convenience sampling method. The model of the study was validated by multilevel structural equation modeling, and the hypothesized relationships were tested using SPSS 26.FindingsResults showed that ethical leadership directly promoted happiness and indirectly through group diversity. Moreover, group diversity was found to mediate this relationship.Practical implicationsFindings propose that leaders within hospitality organizations can attain happiness at work by showing more ethical leadership. Practitioners within the hospitality industry likewise ought to recognize that promoting altruistic attitudes among followers is capable of guiding ethical leadership into happiness at work. Moreover, building a developmental culture is crucial for hospitality teams to reinforce the impact of ethical leadership on happiness at work.Originality/valueThe current study magnifies the leadership-happiness research by unlocking both the direct as well as indirect links, the mediation effect, between ethical leadership and happiness at work.

The social convoy for depression among the older adults in rural China: A multilevel structural equation modeling analysis using a national sample

Rural older adults experience significant mental and physical health challenges. Social convoy theory offers insights into this by assuming functional limitations as antecedent factors of depression through multiple social-related levels. The combined panel data from 2011, 2013, 2015, and 2018 of the China Health and Retirement Longitudinal Study (1,354 individuals with 5,416 responses) was utilized to address the above assumption. The multilevel structural equation modeling was employed to examine whether social activity participation and parent-child interaction mediate the relationship between functional limitations and depressive symptoms at multiple levels. The results indicated that the mediation effects existed only at the between-person level. Specifically, older adults with more functional limitations participated in less social activity, leading to more depressive symptoms, but engaged in more parent-child interaction, reducing depressive symptoms. At the within-person level, depressive symptoms in a given older adult were correlated with functional limitations. In years with greater limitations, older adults participated less in both social activity and parent-child interaction. Interventions should be tailored to older adults’ functional limitations, focusing on social activities for those with typically greater limitations, intergenerational programs for those with fewer limitations, and a combination of both for those experiencing significant increases in limitations.

The double‐edged sword effect of reading self‐concept on math achievement: Additional insights for making dimensional comparisons

AbstractA common belief underlying the dimensional comparison theory is that reading self‐concept has a positive relation to reading achievement but a negative relation to math achievement. However, this belief is incomplete and potentially misleading as supporting evidence is usually derived from studies without simultaneously considering the mediated effect of reading self‐concept through reading achievement. The current study examined the direct and indirect effects of reading self‐concept on math achievement through reading achievement, using PISA 2018 data from 501,766 students across 74 countries/territories. Multilevel structural equation modeling was performed to control for confounding variances from the school level and above. The results showed that reading self‐concept had a direct and negative effect on math achievement, while also exerting a positive indirect effect on math achievement through reading achievement. Overall, the indirect effect outperformed the negative direct effect on math achievement. We concluded that reading self‐concept exerted a double‐edged effect on math achievement. These results should be able to shed new light on the contrasting comparison effect of self‐concept on academic achievement.

“It's all about trust!” a multilevel model of the effect of servant leadership on firefighters' group task performance, adaptivity and emotional exhaustion

AbstractHow and why does servant leaders' behavior influence both performance (individual and collective) and emotional exhaustion within dynamic and extreme environments such as those of firefighters? We develop and test a multilevel model that integrates the principles of servant leadership with social exchange theory to explore how servant leadership positively influences collective task performance and how it strengthens adaptivity at the individual level and reduces emotional exhaustion. Our four‐wave and three‐source study sample comprised 303 firefighters nested in 45 fire stations. The results of multilevel structural equation model (MSEM) analyses indicate that at the individual level, servant leadership significantly predicts high adaptivity and low emotional exhaustion through the mediating influence of firefighters' felt trust and the trust climate. The implications of our results for theory and practice are discussed.

High School Curricular Rigor and Cognitive Function among White Older Adults.

Most research on the strong relationship between education and cognitive aging has focused on years of schooling. Using data from the Wisconsin Longitudinal Study-a sample of White persons born in 1939-we explored whether greater curricular rigor in high school was also associated with better cognitive function in later life. We estimated multilevel structural equation models in data from 2,749 participants who attended 308 Wisconsin high schools, graduating in 1957. Independent of academic ability and performance and school-level financial and material resources, a more rigorous high school curriculum was associated with significantly better global cognitive functioning in 2020, when most participants were 81 years old. There was also a significant mediation via eventual degree attainment. The mediation was moderated such that men and participants from high socioeconomic status families benefited most from a rigorous curriculum. We discuss implications for modern educational policy.

Daily manifestations of psychopathology in response to stress.

Psychological functioning is shaped by how people navigate their environment. Accordingly, psychopathology is often caused and maintained by patterns of responding to the environment that do not meet situational demands. In particular, psychopathology is often expressed in an inflexible or intense manner of coping with stressful situations. Prior research on psychopathology and daily life stress is limited by an overreliance on negative affect reactivity, which neglects the myriad responses that can create problems in a person's life. In this study, we assessed a broad range of daily manifestations of psychopathology to examine daily psychopathology-stress associations. We conceptualized individual differences in functioning as psychopathology traits and daily fluctuations in the interrelated thoughts, behaviors, and emotions as psychopathology states, with traits and states corresponding to the same domains of functioning (i.e., antagonism, detachment, disinhibition, negative affectivity, anankastia, psychoticism). Data have been taken from two samples enriched for psychopathology (N = 112, N = 294 participants) who completed daily assessments of stressors and psychopathological states (n = 9,201, n = 4,292 days). We used multilevel structural equation models to examine average, within-person associations between stressors and psychopathological states and correlations between psychopathological traits and stress responses. Results showed that (a) most people experience increases in psychopathological states when stressed and (b) psychopathological traits relate to more consistent and stronger increases in psychopathological states. Our study suggests that psychopathology reflects how people cope with stressful situations, and what distinguishes people with high-trait psychopathology from those who experience typical upticks in psychopathology when stressed is the consistency and extremity of their responses. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Personality Traits and Ensemble Identification: A Multilevel Structural Equation Modeling Approach

Personality Traits and Ensemble Identification: A Multilevel Structural Equation Modeling Approach

From trait affect to situated performance of change-oriented organizational citizenship behavior: the pathway through situational construal and emotion utilization

Prior studies indicate that fostering change-oriented organizational citizenship behavior (OCB-CH) among employees, wherein they actively contribute to the collective welfare of modern organizations, correlates positively with their positive affect. However, the precise mechanisms linking trait positive affect with OCB-CH remain unclear. This study used aggregated 5-day daily diary data from a sample of 236 full-time employees in Greece to examine positive affectivity differences in the characteristics of situational experiences and how these differences relate to employee OCB-CH via the use of emotions to facilitate performance. Employing multilevel structural equation modeling, the findings revealed a positive association between trait positive affectivity and the situation characteristics of duty, intellect, and positivity, from the DIAMONDS taxonomy. Furthermore, it was determined that perceptions of situational characteristics and emotional utilization sequentially mediated the relationship between positive affectivity and OCB-CH. This study enhances our understanding of how employee trait positive affectivity influences the manifestation of positive change in the form of OCB-CH within organizations.

Urban green space, perceived everyday discrimination and mental health among rural-to-urban migrants: a multilevel analysis in Wuhan, China

BackgroundWhile the mental health benefits of urban green spaces (UGS) are increasingly recognized, less is known about how these relationships vary for socially marginalized groups. This study investigates the association between UGS and mental health among rural-to-urban migrants in Wuhan, China, examining the roles of the quality and quantity of UGS and the intermediary function of perceived everyday discrimination.MethodsWe used Multilevel Structural Equation Modeling to analyze data from a social survey, integrating with park-related social media ratings, street view imagery, and geospatial datasets to characterize UGS features and contextual factors, therefore verifying our hypotheses.ResultsBoth the quality and quantity of UGS significantly influence migrants’ mental health, with quantity demonstrating a stronger overall correlation, challenging common assumptions. Notably, social media scores of parks, reflecting positive user experiences, were found to improve mental health. However, the relationship with UGS quantity was nuanced: higher park density and green view index were positively associated with mental health, while increased park area proportion demonstrated the opposite effect. Furthermore, perceived discrimination emerged as a critical socio-psychological factor and operated spatial heterogeneity. In inner-city areas, neighborhoods characterized by plaza-type parks and high park density were associated with reduced perceived discrimination among migrants, showing active social functions of UGS. However, larger park areas are paradoxically correlated with increased discrimination experiences and poorer mental health. Interestingly, this mediatory effect of perceived discrimination was less pronounced in inner-suburban areas. These findings suggest a nuanced role of UGS in the lives of migrants. While certain aspects of UGS quantity, such as plentiful smaller parks, can facilitate social inclusion and improve mental health, others, like overlarge parks, may unintentionally contribute to feelings of marginalization and negatively impact mental health.ConclusionOur findings highlight the crucial need for context-sensitive green space planning that balances quality and quantity while mitigating discriminatory experiences to improve the mental health of rural-to-urban migrants.

Emotion-specific recognition biases and how they relate to emotion-specific recognition accuracy, family and child demographic factors, and social behaviour

ABSTRACT The errors young children make when recognising others’ emotions may be systematic over-identification biases and may partially explain the challenges some have socially. These biases and associations may be differential by emotion. In a sample of 871 ethnically and racially diverse preschool-aged children (i.e. 33–68 months; 49% Hispanic/Latine, 52% Children of Colour), emotion recognition was assessed, and scores for accuracy and bias were calculated by emotion (i.e. anger, sad, happy, calm, and fear). Child and family characteristics and teacher-reported social behaviour were also collected. Multilevel structural equation modelling revealed emotion-specific recognition accuracies varied between 36 and 65% whereas biases varied between 4 and 13%. Anger was the strongest bias followed by sad, happy, fear, and calm, in contrast to the pattern for accuracy – happy, sad, angry, fear, and calm. More variance was explained in emotion-specific recognition accuracies by child and family characteristics – 7–38% – than biases – 3–7%. Negatively-valanced emotion recognition biases associated with positively-valanced accuracies, and positively-valued emotion recognition biases associated with negatively-valued accuracies. Biases did not have meaningful associations with social behaviour. This study highlights that children’s emotion recognition errors may partially be systematic, but future studies are needed to understand the underlying cognitive mechanisms.

Teacher self-efficacy and teaching quality: A three-wave longitudinal investigation.

Self-efficacy beliefs have cyclical nature as they enhance performance and performance, in turn, influences subsequent self-efficacy beliefs. Likewise, teacher self-efficacy is proposed to shape teaching quality which, in turn, informs future teacher self-efficacy beliefs. To examine these associations, longitudinal studies are needed but are still sparse. Therefore, the present research employed a three-wave longitudinal design to examine the predictive effects of teacher self-efficacy on teaching quality as well as the predictive effects of teaching quality on future teacher self-efficacy by using data from large samples of secondary school teachers (N = 1030) and their students (N = 17,381). Teachers self-reported their efficacy for student engagement, efficacy for instructional strategies and efficacy for classroom management whereas students rated the teaching quality (i.e., cognitive activation, classroom management, and student support) of their teachers. The results of the multilevel structural equation modelling showed that all three dimensions of teacher self-efficacy predicted teaching quality but teaching quality, in turn, predicted only teacher efficacy for student engagement. These results suggest that efforts in raising teacher self-efficacy may show fruitful in raising overall teaching effectiveness.

Inclusive leadership and employee workplace well-being: the role of vigor and supervisor developmental feedback

PurposeEmployees within organizations actively pursue and maintain their workplace well-being. Although there are current studies that have examined the linking inclusive leadership to employee workplace well-being, the underlying intrinsic link between the two remains unclear. On the basis of self-determination theory, this research examined the relationship between inclusive leadership, vigor, supervisor developmental feedback, and workplace well-being.Patients and methodsData were collected from 61 teams that totaled 342 full-time employees through a 3-stage questionnaire. Multilevel Structural Equation Modeling and Monte Carlo simulations were conducted on data for hypothesis testing.ResultsInclusive leadership positively correlated with employee workplace well-being. Employee vigor mediates the link between inclusive leadership and employee workplace well-being. Supervisor developmental feedback moderated the relationship between inclusive leadership and employee vigor.ConclusionsInclusive leadership can improve employees’ workplace well-being by stimulating their vigor. Therefore, managers need to be able to effectively meet employees’ basic needs to stimulate their vigor. Based on meeting employees’ basic needs, managers can also provide targeted developmental feedback to meet employees’ growth needs, which can more effectively promote employees’ vigor and workplace well-being.

Health enhancement through activity travel participation and physical activity intensity

IntroductionPast studies investigate the impact of health parameters on activity and transport options. However, limited research has been done on how activity-travel participation influences health outcomes. Physical activity has the potential to improve health characteristics; however, its intensity (duration and frequency) and optimal mode are yet to be determined. Therefore, the current study aims to use physical activity intensity (PAI) as an intermediate variable to study the correlation among daily activities, transport mode choice, and health parameters to pursue sustainable transportation and improve health parameters. MethodsIn September 2013, a multidimensional three-week household survey was used to collect the data containing 191 households and 732 individuals representing 0.029% of the total population of Bandung, Indonesia. SPSS was used for the data normalization and descriptive analysis, whereas R was used for the multilevel regression and structural equation modeling (SEM). Four models were developed between endogenous, mediator, and exogenous variables. ResultsA significant correlation was found between daily activities, transport mode choice, and health parameters with exceptionally acceptable R2 and p-values. The appraised results indicate that strenuous PAI is 2.1% and 0.3% positively correlates with physical health (PH) and social health (SH), showing that it mediates the correlation among transport-related daily activities and health parameters. High-income households are dependent on motorized transport which is 0.1% and 0.2% negatively associated with PH and SH. Whereas, active transport is 2.0% and 0.2% and public transport is 0.2% and 1.5% positively associated with PH and SH. Replacing short car trips with a sustainable transportation system enhances health parameters and reduces GHG emissions from the transportation sector. ConclusionThe current study concluded that the association was mediated by transport-related PAI which enhances health parameters. Policymakers and practitioners promote transport-related physical activity which may help to enhance health parameters and provide a sustainable transportation system.

Methods in Causal Inference Part 2: Interaction, Mediation, and Time-Varying Treatments

Abstract The analysis of ‘moderation’, ‘interaction’, ‘mediation’, and ‘longitudinal growth’ is widespread in the human sciences, yet subject to confusion. To clarify these concepts, it is essential to state causal estimands, which requires specifying counterfactual contrasts for a target population on an appropriate scale. Once causal estimands are defined, we must consider their identification. I employ causal directed acyclic graphs (causal DAGs) and Single World Intervention Graphs to elucidate identification workflows. I show that when multiple treatments exist, common methods for statistical inference, such as multi-level regressions and statistical structural equation models, cannot typically recover the causal quantities we seek. By properly framing and addressing causal questions of interaction, mediation, and time-varying treatments, we can expose the limitations of popular methods and guide researchers to a clearer understanding of the phenomena that animate our interests.

Tests of moderation of the big-fish-little-pond effect across students with and without special educational needs in regular versus special education schools

BackgroundInclusive education has become increasingly popular based on the assumption that it has multiple benefits for students with special educational needs (SEN). However, contradictions remain regarding the widespread nature of these benefits, particularly when it comes to academic motivation. AimsIn this large-scale cross-sectional study, we relied on the Big-Fish-Little-Pond Effect (BFLPE) to assess the links between inclusive education and students’ academic self-concepts and learning interests in the mathematics and verbal domains. SampleThe sample consisted of 21,219 German elementary school children in Grade 4 who were enrolled in three groups: students without SEN attending regular schools (n1 = 19,069), students with SEN attending regular schools (n2 = 933), and students with SEN attending special education schools (n3 = 1214). MethodsDoubly latent multi-group multilevel structural equation models and tests of latent interaction were performed to test the BFLPE. ResultsResults supported the BFLPE for both outcomes and domains. However, no BFLPE was identified for learning interest in the verbal domain among students with SEN attending special education schools, although the size of this effect did not differ significantly from that observed among students without SEN. In regular schools, the BFLPE was almost two times stronger among students with SEN than among their peers without SEN. ConclusionsResults support the generalizability of the BFLPE to students with SEN, while casting doubts on the motivational benefits of inclusive education for these students. Interventions targeted at attenuating the BFLPE should thus be tailored for both regular and special education schools.

Daily sibling interactions among latino children in middle childhood: An examination of familism values and sibling roles

Siblings are children’s most frequent daily companions in middle childhood and these relationships are theorized to be dynamic in nature but are rarely studied in a way that captures their day-to-day variation. Guided by cultural-ecological and relationship perspectives, this study examined the role of older siblings’ familism values and caring roles as predictors of average levels of and day-to-day variation in siblings’ positive and negative exchanges across seven days and tested moderation by child sex. Participants were older ( M = 10.6 years; SD = 0.39; 46.6% female) and younger siblings ( M = 8.2 years; SD = 1.07; 47.2% female) in 307 Latino sibling dyads recruited from 11 elementary schools in the southwestern U.S. Data were collected via home visits with parents, and in afterschool sessions and seven nightly phone calls with siblings. Analyses were conducted using a random variability model in a multilevel structural equation modeling approach. Findings revealed consistent associations between older siblings’ familism values and sibling roles and both average levels of and intraindividual variation in older siblings’ positivity and negativity. For younger siblings, some findings were moderated by child sex with more consistent links to younger sisters’ positivity and negativity than younger brothers’. Discussion highlights how older siblings’ cultural values and roles shape both overall levels of and day-to-day fluctuations in siblings’ exchanges and offers future directions of research.

The Cumulative Impact of Recurrent Experiences of Intimate Partner Violence on Emotion Dysregulation: A Longitudinal Investigation.

Intimate partner violence (IPV) disproportionately affects women and has been linked to a range of negative psychological outcomes, including emotion dysregulation. Although IPV is often recurrent and tends to escalate in severity over time, few studies have examined the potential effects of recurrent experiences of IPV on emotion dysregulation. The current study employed a longitudinal design to examine the cumulative impact of recurrent experiences of IPV on emotion dysregulation. Participants were 491 young adult community women (Mage = 21.74; 61.3% White, 35% African American/Black) who completed self-report measures of IPV and emotion dysregulation every 4 months for 28 months. Multilevel structural equation modeling techniques revealed that, at the within-person level, increases in cumulative IPV were associated with increases in global emotion dysregulation across the 28 months. In particular, the accumulation of IPV experiences over time was associated with increases in difficulties controlling impulsive behaviors when distressed and accessing effective emotion regulation strategies. At the between-person level, greater average cumulative IPV was associated with greater mean levels of global emotion dysregulation, as well as four specific dimensions of emotion dysregulation, including difficulties accepting emotional responses, accessing effective emotion regulation strategies, and both engaging in goal-directed behaviors and controlling impulsive behaviors when distressed. These findings suggest the potential cumulative impact of IPV on both global and more specific dimensions of emotion dysregulation. Results also highlight emotion dysregulation as a potentially important construct to assess and target in interventions for women experiencing recurrent IPV.

A Brief Measure of Classroom Climate for Elementary School Students

Since classroom climate is inherently a group level phenomenon, we argue that measures shown to be valid and reliable in multi-level settings are needed. We propose a brief, 5-item measure of classroom climate and demonstrate its reliability and validity in multi-level structural equation models. The measure demonstrated good reliability and we also show the measure’s relationship to a key student outcome, social emotional learning (SEL) knowledge. In classrooms with higher climate levels, fourth grade students are more likely to have higher SEL knowledge.

Socioemotional classroom environments mediating teacher expectation effects: A multilevel structural equation modeling approach

In higher education, the pathways through which teacher expectations could contribute to students’ academic achievement are an area of limited empirical investigation. In this study, we investigated the mediating role of the socioemotional classroom environment in the relations between university teachers’ early expectations and their students’ later achievement. The participants were 176 teachers and their 6,506 first-year undergraduate students from public universities in China. Data from teachers’ initial expectations, students’ prior and year-end achievement on standardized tests, and students’ perceptions of socioemotional factors within the classroom were collected. As students were nested in classes, multilevel structural equation modeling was employed for data analysis. The results showed that, at the individual level, with students’ prior achievement controlled, the extent of the teacher’s differential treatment, the quality of the teacher-student relationships, and the quality of the peer relationships in turn were statistically significantly related to the university teachers’ initial expectations, which ultimately was associated with the individual students’ year-end academic achievement. Further, the university teachers’ expectations for their individual students played a role in the quality of the peer relationships which subsequently contributed to students’ later academic achievements. Moreover, the university classes whose teachers tended on average to have higher expectations tended to achieve at higher levels. However, we did not identify mediating paths via socioemotional factors from university teacher expectations to student achievement at the class level.

Should I stay (home) or should I go (party)? Examination of drinking as a mediator of the relationship between alcohol-related social media content and adherence to COVID-19 recommendations among college students.

The COVID-19 pandemic impacted young people's drinking. Yet, despite social gatherings posing a known risk for increased contraction, some college students still congregated to drink. Furthermore, some students posted about these drinking events to their public and/or private social media feeds. It is crucial to understand the relationships between posting alcohol-related content (ARC), drinking and adherence to COVID-19 recommendations because they may have contributed to the spread of the virus. The current interval contingent, 14-day diary study assessed students' (N = 129) public and private ARC; their drinking; percentage of time wearing a mask and social distancing; and number of people they socialised with during drinking events (total crowd size) each day. Multilevel structural equation modelling was employed to examine relationships between ARC posting, drinking and behavioural outcomes. On days in which students posted both public and private ARC, they tended to drink more and in turn, reported attending gatherings with larger crowds. Curiously, on days in which students drank more, they indicated more mask wearing and social distancing, and on days in which students posted more private ARC, they also reported larger crowds. Finally, an indirect effect of drinking on the links between public and private ARC and mask wearing, social distancing and total crowd sizes emerged. Findings revealed students continued to post and drink socially despite the risks, which may have encouraged others within their networks to engage in similar risky behaviours. Future public health crises should balance young people's need for social connection with risk mitigation efforts.