Hierarchical Regression Models Research Articles

Cognitive influences on biosecurity measure compliance during a global pandemic

IntroductionDuring the first years of the pandemic, COVID-19 forced governments worldwide to take drastic measures to reduce the spread of the virus. Some of these measures included mandatory confinements, constant use of masks, and social distancing. Despite these measures being mandatory in many countries and the abundance of evidence on their effectiveness at slowing the spread of the virus, many people failed to comply with them.MethodsThis research explored the role of cognitive factors in predicting compliance with COVID-19 safety measures across two separate studies. Building on earlier work demonstrating the relevance of cognitive processes in health behaviour, this study aimed to identify key predictors of adherence to safety guidelines during the pandemic. Utilising hierarchical regression models, we investigated the influence of age, sex, cognitive control, cognitive flexibility (Study 1), working memory, psychological health, and beliefs about COVID-19 (Study 2) on compliance to biosafety measures.ResultsDemographic variables and cognitive control were significant predictors of compliance in both studies. However, cognitive flexibility and working memory did not improve the models’ predictive capacities. In Study 2, integrating measures of psychological health and beliefs regarding COVID-19 severity significantly improved the model. Further, interaction effects between age and other variables also enhanced the predictive value.DiscussionThe findings emphasise the significant role cognitive control, age, psychological health, and perceptions about COVID-19 play in shaping compliance behaviour, highlighting avenues for targeted interventions to improve public health outcomes during a pandemic.

Leading Stressors and Coping Strategies Associated With Maternal Physical and Mental Health During the Extended Postpartum Period.

A critical gap exists in understanding stressors and coping that affect women's health beyond 6 weeks postpartum. Using new stressor and coping scales tailored to postpartum women, we examined the relationship of postpartum-specific stressors and coping to women's physical and mental health between 2 to 22 months after childbirth. A total of 361 women of diverse race, ethnicity, and functional abilities recruited through clinical and online methods completed online surveys that included Sources of Stress-Revised subscales, such as overload, changes after pregnancy, and low support resources; Postpartum Coping Scale subscales, such as self-regulation, self-care, and health promotion; Patient-Reported Outcomes Measurement Information System Global Health, covering physical and mental health dimensions; and social demographic items. Analyses included hierarchical linear regression models adjusted for social factors. Education and employment were the only social factors associated with physical and mental health, respectively. After adjusting for social factors, overload (P <.001) and coping through health promotion (P =.020) were the only additional variables associated with physical health. After adjusting for social factors, overload (P <.001) and low support resources (P =.002) and coping through self-care (P =.036) were the only additional variables associated with mental health. Thus, being overloaded was the key stressor associated with decreases in physical and mental health. Health promotion was associated with increases in physical health, and self-care was associated with increases in mental health. These findings point to directions for health care and community interventions to promote health for postpartum women under stress. Strengths of our study include application of stress and coping scales tailored to postpartum women, whereas a limitation is use of a cross-sectional design.

The Benefits of Acute Exercise for Children's Cognition are Associated with Trait Anxiety.

Children's anxiety is associated with decreased cognitive performance. One well-established behavioral intervention to transiently improve cognitive performance in children is acute aerobic exercise (AAE). Thus far, however, it is unclear whether the benefits of AAE on cognition vary based on individual differences in children's anxiety level. Therefore, we investigated whether trait anxiety levels mediate the effects of AAE on cognitive outcomes in preadolescent children. Ninety-five preadolescent children (9-11 years, 41 females) underwent two experimental interventions in a random, crossover, and counterbalanced design: an exercise protocol (20 minutes of walking at 65-75% of HRpeak) and a non-exercise control (seated rest). Behavioral indices of cognitive performance (accuracy, reaction times, response time variability) were assessed before and after each intervention using a modified flanker task. The effects of each intervention on cognitive performance were calculated as pre- to post-intervention differences (effect scores). These scores were first correlated with children's personal characteristics: anxiety (STAIC), sex, age, BMI, IQ, and aerobic fitness. Significant correlations guided subsequent hierarchical regression models, which specifically tested for associations between the effects scores and anxiety levels while controlling for remaining relevant covariates. Regression analysis revealed that anxiety was a significant predictor of the effects that AAE and rest had on cognitive performance. Specifically, higher anxiety scores were associated with greater exercise-induced cognitive benefits (increased accuracy, decreased response time variability; p's < 0.05) and greater rest-induced cognitive impairments (decreased accuracy, increased response time variability; p's < 0.01). These findings suggest that children reporting higher trait anxiety might experience greater cognitive benefits from a single bout of AAE.

The effect of employees’ health benefits on the performance of commercial state corporations in Kenya

This study sought to assess the effect of employees’ health benefits on the performance of Commercial State Corporations in Kenya. Commercial state corporations are established by the government to offer efficient and effective citizen service delivery whilst surrendering back investment income at the end of each financial year. Despite this, most of them have performed dismally over the years, questioning their justification for continued existence. A survey research design was used to conduct this study, targeting 612 managers of various cadres in the 68 commercial state corporations in Kenya. The Yamane sampling formula was used to arrive at a sample size of 242 managers in these corporations. Data collected through structured questionnaires was analysed using both inferential and descriptive statistics. Multiple regression models and Hierarchical regression were used to test hypotheses for moderation. The outcomes of the hierarchical regression model on all the variables revealed that employees' health benefits (β= 0.138, p= 0.000, p&lt;0.05) had a significant positive relationship with the performance of commercial state corporations. These findings contribute novel insights into the influence of employee compensation on the performance of commercial state corporations. The study concluded that employee compensation affects the performance of commercial state corporations and that remuneration policy guidelines strengthen the positive linkage between employee compensation and performance outcomes. The study recommended that commercial state corporations should give higher priority to employee’s health benefits and Remuneration policy guidelines as they are crucial in enhancing their performance.

Eveningness chronotype influences social functioning by deteriorating depressive symptoms in remitted patients with schizophrenia

ABSTRACT Sleep and circadian rhythm disruption (SCRD) is common in schizophrenia patients, who also typically experience impaired social functioning. While various factors influence social functioning in schizophrenia, the specific impact of sleep and circadian rhythm disruption remains unclear. This study aimed to investigate the connection between chronotype and social functioning in remitted schizophrenia patients, examining the mediating roles of depression and sleep quality. The study included 185 patients diagnosed with schizophrenia based on DSM-5 criteria. After categorizing the patients into morningness, eveningness, or intermediate chronotypes using the Morningness-Eveningness Questionnaire(MEQ), they were assessed with the Positive and Negative Syndrome Scale(PANSS), Calgary Depression Scale for Schizophrenia(CDSS), Personal and Social Performance Scale(PSPS) and Pittsburgh Sleep Quality Index(PSQI). The eveningness chronotype group showed higher CDSS and PSQI scores and lower PWBS and PSPS-Total scores than the other groups (p < 0.05). A hierarchical linear regression model assessed MEQ, PSQI, and CDSS scores’ effects on PSPS total scores. MEQ scores’ significance diminished when CDSS scores were included. Eveningness chronotype, particularly with increased depressive symptoms, negatively impacts social functioning in remitted schizophrenia patients.These findings contribute to the understudied area of chronotype in schizophrenia and its impact on social functioning, including its interaction with sleep..

Lower-Body Power, Body Composition, Speed, and Agility Performance among Youth Soccer Players.

Speed and agility have been described as crucial abilities for soccer players. The purpose of this study was to analyze, in detail, the variance in speed and agility tasks explained by lower-body power after controlling for age and body composition. The participants were 96 male soccer players aged 16.1 ± 1.6 years. Body composition (stature, body fat percentage-BF%, body mass, and fat-free mass-FFM), lower-body power (countermovement jump-CMJ, and squat jump-SJ), speed (5-, 10-, and 35 m sprints), and agility (t-test) were assessed. Among body composition parameters, BF% presented the highest number of significant relationships with speed and agility, with the strength of correlations ranging from small (5 m sprint, r = 0.25) to large (35 m sprint, r = 0.52). The strongest correlation coefficient emerged between FFM and the 35 m sprint (r = -0.65). Significant correlations were found between vertical jump performance and the 35 m sprint (CMJ: r = -0.68; SJ: r = -0.69), followed by the t-test (CMJ: r = -0.35; SJ: r = -0.47). The hierarchical multiple regression model could explain 22% to 67% of the variance observed in agility scores and speed. BF% remained the most statistically significant negative predictor of all regression models. The CMJ remained a statistically significant positive predictor of the 35 m sprint (β = -0.581, p ≤ 0.01) after controlling for age and body composition. Integrating programs targeting lower-body power might be important to enhance speed and agility performance in youth soccer. On the other hand, future research based on multidisciplinary approaches to investigate the effects of nutritional strategies in reducing or preventing gains in BF% is still needed, which remained a significant predictor of sprint and agility performance in the final models.

English A cross cultural investigation on Smart Home Technology:

Some individuals may feel threatened by the digital age and suffer from technophobia, whereas technophilic individuals embrace new technologies. Relying on a study of 243 participants aged 18-65, this study explored the roles of technophobia, technophilia, trust in technology, and digital competence, and how they interact within the SHT paradigm. Technophobia positively correlated with age and negatively with ownership and most digital skills. Males outscored females in overall competence, especially in creative digital skills and technophilia. Spanish participants scored higher than English participants on operational digital skills, creative digital skills, and technophilia-dependency. A hierarchical regression model with technophobia as the dependent variable showed that operational, navigational, and mobile digital skills contributed negatively to technophobia. Technophilia-enthusiasm negatively predicted technophobia, whereas technophilia-reputation did so positively. Culture and gender were not significant predictors of technophobia in this context. There is a need to provide digital skills for all sections of society to increase self-efficacy and counterbalance any feelings of anxiety and technophobia that may manifest in individuals who feel left behind in the modern digital world. The onus is for SHT manufacturers to design products that require basic digital skills and are custom-made for different populations cohorts within society.

An exploratory study of the relationship between pulse transit time and blood pressure based on causal inference

Aim: We propose to examine the causal relationship between the noninvasive features represented by pulse transit time (PTT) and blood pressure (BP), with the aim of mitigating the impact of confounding factor(s) and thus improving the performance of BP estimation. Methods: We identified the causal graph of BP and the important noninvasive features extracted from electrocardiogram (ECG) and photoplethysmogram (PPG) via fast causal inference (FCI) algorithm, with the orientations of the edges in the causal graph being determined by the causal generative neural networks (CGNN) algorithm. With the knowledge obtained from the causal graph, we further used hierarchical regression model to estimate BP, and validated the proposed method on 17 subjects. Results: We found that the obtained causal graph was almost consistent with the prior knowledge, and heart rate (HR) was one of the main confounding factors of PTT and BP. Incorporating HR into the hierarchical regression model to eliminate its confounding effect on the PTT-based BP estimation, the mean value of SBP and DBP estimation was improved by 1.27 and 1.89 mmHg, respectively, and the mean absolute difference (MAD) was improved by 2.28 and 3.60 mmHg, respectively. Conclusion: Causal inference-based method has the potential to clarify the causal relationship between BP and related wearable noninvasive features, which can further shed light on developing new methods for cuffless BP with acceptable accuracy.

The impact of social and emotional support on serious psychological distress among people with functional disabilities and type 2 diabetes.

To examine the impact of social and emotional support on serious psychological distress (SPD) in individuals with type 2 diabetes (T2DM) and functional disabilities after controlling for socio-demographic factors and health status. Findings provide information for healthcare providers to enhance patients' psychological well-being. Data from 529 adults were obtained from the 2021 National Health Interview Survey, including those who had T2DM and also reported significant difficulty or inability to perform an activity in any of the following domains: seeing, hearing, walking or climbing stairs, communicating, remembering or concentrating, or practicing self-care. Descriptive analysis and a hierarchical regression model of SPD were used. The mean age of participants was 67.88 years old, and the mean duration of diabetes diagnosis was 16.88 years. Notably, 12.5% of individuals reported SPD. A decreased likelihood of reporting SPD occurrence was associated with older age (odds ratio (OR) = 0.95), a longer duration of T2DM diagnosis (OR = 0.97), having at least a high school education (OR = 0.54), and receiving social and emotional support. Social and emotional support likely mitigates psychological distress, suggesting that social and emotional support resources should be enhanced, especially among individuals who are younger and those more recently diagnosed with T2DM.

COVID-19 vaccine fatigue in Scotland: how do the trends in attrition rates for the second and third doses differ by age, sex, and council area?

Abstract Vaccine fatigue is the propensity for individuals to start but not finish a vaccination program with several doses, which means they are less protected. Our study examines vaccine fatigue in the COVID-19 vaccination programme in Scotland. We model the attrition rates across multiple doses with a hierarchical binomial logistic regression model with correlated age- and area-specific random effects with conditional autoregressive priors, where inference uses integrated nested Laplace approximations. We find attrition rates smoothly decrease with increasing age, are much higher for the third dose than the second, and are generally higher for males than females.

Changing US Support for Public Health Data Use Through Pandemic and Political Turmoil.

Recent legislative privacy efforts have not included special provisions for public health data use. Although past studies documented support for public health data use, several global events in 2020 have raised awareness and concern about privacy and data use. This study aims to understand whether the events of 2020 affected US privacy preferences on secondary uses of identifiable data, focusing on public health and research uses. We deployed two online surveys-in February and November 2020-on data privacy attitudes and preferences using a choice-based-conjoint analysis. Participants received different data-use scenario pairs-varied by the type of data, user, and purpose-and selected scenarios based on their comfort. A hierarchical Bayes regression model simulated population preferences. There were 1,373 responses. There was no statistically significant difference in the population's data preferences between February and November, each showing the highest comfort with population health and research data activities and the lowest with profit-driven activities. Most subgroups' data preferences were comparable with the population's preferences, except African Americans who showed significant decreases in comfort with population health and research. Despite world-changing events, including a pandemic, we found bipartisan public support for using identifiable data for public health and research. The decreasing support among African Americans could relate to the increased awareness of systemic racism, its harms, and persistent disparities. The US population's preferences support including legal provisions that permit public health and research data use in US laws, which are currently lacking specific public health use permissions.

Drinking While Stressed and Drinking to Cope Differentially Relate to Mental Health.

Excessive alcohol use, as well as drinking to manage distress, are known to undermine mental health. The current study examined the unique associations of simply consuming alcohol while stressed, versus using alcohol to cope with distress, with mental health during the early stages of COVID-19. Participants (N = 264) reported their alcohol use and use of alcohol/substances to cope with stress daily for 22 days and completed measures of mental health at baseline and every 7 days thereafter. Hierarchical regression models were tested to predict drinking while stressed, drinking to cope, and mental health. At baseline, drinking while stressed was not significantly associated with mental health when coping motives were accounted for. However, drinking to cope was significantly associated with greater stress, anxiety, depression, and loneliness. Drinking while stressed was significantly predicted by baseline stress and loneliness, while drinking to cope was significantly predicted by baseline anxiety. Drinking while stressed was not a significant predictor of change in mental health when coping motives were accounted for. Drinking to cope was a significant predictor of increases in depression but not of change in stress, anxiety, or loneliness. The motivation to manage distress appears to be driving the negative effects of alcohol use on mental health.

Elder mistreatment and psychological resilience: the moderating role of social support

ABSTRACT This study examines the possible consequence of elder mistreatment on resilience and whether social support moderates this posited association, using a rural sample of 432 community-dwelling older Chinese adults aged 60 to 79 years. Elder abuse included verbal abuse, physical abuse, or financial exploitation. Social support was measured by The Multidimensional Scale of Perceived Social Support (MSPSS). Resilience was represented by a seven-item scale. Hierarchical regression models indicated that mistreatment is significantly related to low levels of resilience when confounding factors are adjusted. A significant interaction term (abuse × social support) was also detected. Mistreated respondents who reported higher levels of social support were less likely to experience low resilience compared to those with lower levels of social support. Social support buffers against the undesirable effect of mistreatment on resilience, especially for those who were abused.

Trends and projections of age-appropriate vaccination coverage in 41 low- and middle- income countries in Asia and Sub-Saharan Africa, 2000-2030.

Routine immunization programs have focused on increasing vaccination coverage, which is equally important for decreasing vaccine-preventable diseases (VPDs), particularly in low- and lower-middle-income countries (LMICs). We estimated the trends and projections of age-appropriate vaccination coverage at the regional and national levels, as well as place of residence and wealth index in LMICs. In total, 174 nationally representative household surveys from 2000 to 2020 from 41 LMICs were included in this study. Bayesian hierarchical regression models were used to estimate trends and projections of age-appropriate vaccination. The trend in coverage of age-appropriate Bacillus Calmette-Guérin (BCG), third dose of diphtheria, tetanus, and pertussis (DTP3), third dose of polio (polio3), and measles-containing vaccine (MCV) increased rapidly from 2000 to 2020 in LMICs. Findings indicate substantial increases at the regional and national levels, and by area of residence and socioeconomic status between 2000 and 2030. The largest rise was observed in East Africa, followed by South and Southeast Asia. However, out of the 41 countries, only 10 countries are estimated to achieve 90% coverage of the BCG vaccine by 2030, five of DTP3, three of polio3, and none of MCV. Additionally, by 2030, wider pro-urban and -rich inequalities are expected in several African countries. Significant progress in age-appropriate vaccination coverage has been made in LMICs from 2000 to 2020. Despite this, projections show many countries will not meet the 2030 coverage goals, with persistent urban-rural and socioeconomic disparities. Therefore, LMICs must prioritize underperforming areas and reduce inequalities through stronger health systems and increased community engagement to ensure high coverage and equitable vaccine access.

Comparing the Diagnostic and Statistical Manual of Mental Disorders, fifth edition, personality disorder models scored from the same interview.

The alternative model of personality disorders (AMPD) traits were designed to maintain continuity with the Section II personality disorder (PD) diagnoses by retaining the same clinical information. Whether the AMPD traits achieve this is not well established. Prior work testing incremental validity of AMPD traits and Section II diagnoses is limited by the fact each model was measured by a different instrument or rater, making it unclear whether discrepancies are due to the constructs or methods. Here, we compare the incremental validity of AMPD traits versus Section II PDs assessed by the same instrument and rater. Participants (N = 311, 50% received past-year mental health treatment) completed a clinical interview, baseline self-reports, and 14-day ambulatory assessment protocol. Interviewers rated AMPD domains, facets, and Section II criteria from the same interview (Structured Interview for DSM-IV Personality). We used hierarchical regression models to evaluate the variance predicted in 17 clinically relevant cross-sectional and momentary variables by the AMPD traits and Section II PDs. Incremental R² showed that Section II PDs account for little variance in outcomes over and above the AMPD domains/facets, whereas the AMPD facets were generally more predictive of outcomes than the Section II PDs. Results add novel evidence that dimensional PD traits-not a particular assessment method-are equivalent or superior to PD categories for predicting social, emotional, and behavioral functioning. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Approximate inferences for Bayesian hierarchical generalised linear regression models

SummaryGeneralised linear mixed regression models are fundamental in statistics. Modelling random effects that are shared by individuals allows for correlation among those individuals. There are many methods and statistical packages available for analysing data using these models. Most require some form of numerical or analytic approximation because the likelihood function generally involves intractable integrals over the latents. The Bayesian approach avoids this issue by iteratively sampling the full conditional distributions for various blocks of parameters and latent random effects. Depending on the choice of the prior, some full conditionals are recognisable while others are not. In this paper we develop a novel normal approximation for the random effects full conditional, establish its asymptotic correctness and evaluate how well it performs. We make the case for hierarchical binomial and Poisson regression models with canonical link functions, for hierarchical gamma regression models with log link and for other cases. We also develop what we term a sufficient reduction (SR) approach to the Markov Chain Monte Carlo algorithm that allows for making inferences about all model parameters by replacing the full conditional for the latent variables with a considerably reduced dimensional function of the latents. We expect that this approximation could be quite useful in situations where there are a very large number of latent effects, which may be occurring in an increasingly ‘Big Data’ world. In the sequel, we compare our methods with INLA, which is a particularly popular method and which has been shown to be excellent in terms of speed and accuracy across a variety of settings. Our methods appear to be comparable to theirs in terms of accuracy, while INLA was faster, for the settings we considered. In addition, we note that our methods and those of others that involve Gibbs sampling trivially handle parameters that are functions of multiple parameters, while INLA approximations do not. Our primary illustration is for a three‐level hierarchical binomial regression model for data on health outcomes for patients who are clustered within physicians who are clustered within particular hospitals or hospital systems.

Contextualizing school discipline: Examining the role of general peer and teacher discrimination at the individual- and school-level on individual suspension.

Past research has linked peer and teacher discrimination to risk factors for school discipline, but few studies have examined whether peer and teacher discrimination have a direct impact on school discipline. This study examines the effects of general peer and teacher discrimination at the individual- and school-level on school suspension using nationally representative, secondary data on almost 12,000 youth across 131 schools. Hierarchical logistic regression models indicated that general teacher discrimination at the individual- and school-level-but not general peer discrimination-increased the odds of receiving school suspension. Findings suggest that general discrimination by direct learning instructors and teachers representing the broader school culture can shape student conduct. Reducing school discipline thus falls on teachers, staff, principals, and learners.

Using Artificial Intelligence Algorithms to Predict Self-Reported Problem Gambling Among Online Casino Gamblers from Different Countries Using Account-Based Player Data

AbstractThe prevalence of online gambling and the potential for related harm necessitate predictive models for early detection of problem gambling. The present study expands upon prior research by incorporating a cross-country approach to predict self-reported problem gambling using player-tracking data in an online casino setting. Utilizing a secondary dataset comprising 1743 British, Canadian, and Spanish online casino gamblers (39% female; mean age = 42.4 years; 27.4% scoring 8 + on the Problem Gambling Severity Index), the present study examined the association between demographic, behavioral, and monetary intensity variables with self-reported problem gambling, employing a hierarchical logistic regression model. The study also tested the efficacy of five different machine learning models to predict self-reported problem gambling among online casino gamblers from different countries. The findings indicated that behavioral variables, such as taking self-exclusions, frequent in-session monetary depositing, and account depletion, were paramount in predicting self-reported problem gambling over monetary intensity variables. The study also demonstrated that while machine learning models can effectively predict problem gambling across different countries without country-specific training data, incorporating such data improved the overall model performance. This suggests that specific behavioral patterns are universal, yet nuanced differences across countries exist that can improve prediction models.

Satisfaction with life and its predictive factors in a cohort of fathers 24-months postpartum

AbstractSatisfaction with life is considered an essential indicator of quality of life and has implications not only for parents but also for the general health of society. It is relevant to know the factors that could explain differences in parents’ satisfaction with life. The objectives of this study were to analyse the degree of satisfaction with life of a cohort of fathers 13–24 months after the birth of their child and to analyse the predictive capacity of sociodemographic, physical, psychological, and social factors on their long-term life satisfaction. This was a longitudinal study with follow-up at 6–12 months and at 13–24 months (n = 152 fathers). The Fatigue Assessment Scale, Athens Insomnia Scale, Parental Stress Scale and short version of the Dyadic Adjustment Scale were completed online at 6–12 months. Finally, Satisfaction with Life Scale was assessed between 13 and 24 months after birth. The mean score of life satisfaction of the participants 13–24 months after the birth of their child was 18.72 (SD = 3.71) and was negatively correlated with fatigue, insomnia, and stress (p &lt; 0.01), and positively correlated with dyadic adjustment (p &lt; 0.01). The final hierarchical regression model showed that educational levels and insomnia were the main predictive variables for life satisfaction during the second-year postpartum. This work has important implications for clinical practice because it allows health professionals to understand the factors influencing satisfaction with life and health among fathers and to plan more effective antenatal and postnatal care.

The nexus of employees’ in-role and extra-role behaviour and customer service: the moderating role of gender

PurposeThis paper investigates the influence of employees’ extra-role and in-role behaviours on customer service alongside the moderating role of gender.Design/methodology/approachThis paper employs the theory of behavioural intentions, cross-sectional survey design and quantitative approach to collect the data from 426 purposively sampled workers and customers of oil marketing companies. The data were analysed using descriptive statistics, correlation and the hierarchical regression model in SPSS.FindingsThe results indicate that employees’ extra-role behaviour has a significant positive effect on customer service while employees’ in-role behaviour has no significant effect on customer service. It is also established that gender of staff can significantly moderate the relationship between extra-role behaviour and customer service such that the behaviour of female staff has greater effect on customer service than their male counterparts. However, the gender of staff has no moderating effect on the relationship between in-role behaviour and customer service.Practical implicationsThe findings imply that female staff should be allowed to directly engage customers more often than male staff to promote superior customer service. Managers should continuously improve upon the behaviour of employees through orientations, workshops and mentoring. Behaviour stimuli such as awards, appreciations and recognition for best workers would have to be encouraged to induce employees to act beyond their prescribed-roles.Originality/valueThis study is the first to investigate how staff behaviours (in-role and extra-role) impact customer service, with gender of the employees as a moderator. This paper contributes to literature by empirically confirming the differential influence of employees’ extra role and in-role behaviours on customer service and the effectiveness of gender as a moderator on the relationship between extra-role behaviour and customer service from a developing country perspective and an industry where there is dearth of research.