Random Effects Regression Models Research Articles

Effects of School-Based Lifestyle Interventions on Cardiovascular Health in Chinese Children and Adolescents: A Post Hoc Analysis of a National Multicenter Study.

We evaluated the effects of a national school-based lifestyle intervention program against childhood obesity on improving cardiovascular health (CVH), assessed using the recently introduced Life's Essential 8 metric framework. Our post hoc analysis of the national school-based lifestyle intervention targeting diet and physical activity included 94 schools with eligible students aged 7 to 17 years (intervention group n=30 629; control group n=26 581). We assessed CVH metrics by individual scores for 4 health behaviors (diet, physical activity, nicotine exposure, and body mass index) and 3 health factors (blood lipids, glucose, and pressure). These scores were subsequently combined to calculate the CVH score, health behavior score, and health factor score. The primary outcomes were these 3 composite scores and high CVH (CVH score≥80). All analyses applied a likelihood-based random-effects regression modeling following the intention-to-treat. The school-based lifestyle intervention was associated with a 0.89-point increase in the CVH score (95% CI, 0.03-1.74), 14% greater odds of having high CVH (odds ratio, 1.14 [95% CI, 1.01-1.29]), and a 1.35-pointimprovement in the health behavior score (95% CI, 0.32-2.38). According to our subgroup analyses, this intervention had stronger favorable effects on CVH and health behavior scores in primary (grades 1-6) and junior (grades 7-9) schools than in senior high (grades 10-12) schools (P<0.001). This school-based lifestyle intervention improved CVH behaviors in Chinese children and adolescents. The differences in the intervention effect according to school grade imply critical windows for forming healthy lifestyles at younger ages. URL: https://www.clinicaltrials.gov; Unique identifier: NCT02343588.

Age-Related Trajectories of Health Decline Among Immigrants and Natives in Europe: The Effect of Education.

Background: The ability to age healthily is highly dependent on individual characteristics that include gender, social class, a range of biological and contextual factors, and migrant background. Indeed, immigration has changed the demographic composition and social structure of many European countries, generating an increasing interest in how societies, and immigrants in particular, are aging. Research Design: This paper compares the age-related trajectories of health decline in three health measures (activity limitation indicator, self-perceived health, and chronic conditions) among 7,429 immigrants and 81,424 native-born populations aged 50 years old and over using longitudinal data from seven waves of the Survey of Health, Ageing and Retirement in Europe (SHARE) (2004-2020). Analysis: We use descriptive and multivariate models (random-effects regression models) to analyze the association between migration status and each health outcome. We are particularly interested in determining the age-related trajectory of this association across three levels of education, exploring at the same time the effect of the interaction between immigration status and age on health. Results: Our results highlight potential gaps in health between immigrants and native-born people that are particularly large for the low-educated group. In other words, the health decline is more marked for low-educated immigrants compared to native-born populations in particular for activity limitation indicator and self-perceived health, while it is less pronounced with the accumulation of chronic conditions. Conclusions: Our findings should serve to enhance the design of the provision of social services and support and the promotion of equal opportunities.

Factors Influencing Dividend Payout Policy: An Investigation of Food and Allied Companies Listed on the Dhaka Stock Exchange Limited

The main objective of the study is to identify if there is a relationship between a number of company-selected factors such as Profitability, Liquidity, Size, Growth, Leverage and Taxation of Food and Allied Companies Listed in the Dhaka Stock Exchange and the companies’ dividend payout ratio which is used as a proxy of dividend payout policy. A sample of seven companies has been taken from Food and Allied Companies listed in DSE following a purposive sampling technique. To meet the objective of this research, ten-year financial data from 2011-2020 of listed food and allied companies is used and examined to determine the impact of selected variables on dividend payout. The random effect regression model was used as suggested by the result of the Hausman test to test the relationship between dividend determinants and dividend payout. Results indicate that among the independent variables Size and Profitability significantly impact the dividend payout policy but in terms of robustness of the model, Liquidity, Size and profitability significantly impact the dividend payout policy of Food and Allied companies listed in DSE. Taking into consideration these findings, it can be said that firms having higher profitability are most likely to pay more dividends. Firms with higher liquidity, growth, size, leverage and taxation pay fewer dividends and vice versa.

Association of Discrimination Experiences with Rapid Subsequent Changes in Anxiety and Depressive Symptoms in U.S. Adults During the COVID-19 Pandemic.

This study explores how discrimination experiences during the COVID-19 pandemic relate to anxiety and depressive symptoms in U.S. adults. Using a national representative intensive longitudinal survey, the study evaluates rapid subsequent changes in anxiety and depression when individuals undergo heightened discrimination beyond their usual experiences. The study used 23 survey timepoints, primarily with 2-week intervals, from the Understanding America Study (n=8,198). Time-varying and time-lagged associations between discrimination experiences and anxiety and depression were modeled using multi-level logistic random-effect repeated-measures regression models. The results showed that discrimination experiences were associated with moderate-to-severe anxiety and depressive symptoms, as well as more than one comorbid psychological distress symptom (adjusted Odds Ratios [AORs]=1.10 to 1.13). The association remained significant regardless of inter-individual differences in exposure to discrimination. Non-Hispanic Blacks, Asians, and other race/ethnicities exhibited stronger associations between discrimination and psychological distress (AORs=1.63 to 1.93) compared to Hispanic and White respondents (AORs=1.13 to 1.25). Our findings suggest that individuals experience a rapid deterioration in their emotional well-being when subjected to heightened levels of discrimination beyond their typical experiences.

The impact of cybersecurity disclosure on banks’ performance: the moderating role of corporate governance in the MENA region

This study aims to: (1) examine the impact of cybersecurity disclosure on banks’ performance and (2) explore whether the existence of a chief risk officer (CRO), an information technology (IT) committee, and a board of directors (BOD)’ size moderates the association between cybersecurity disclosure and bank performance. The study used manual textual analysis to measure cybersecurity disclosure in a sample of listed banks in the MENA region countries based on data from 2019 to 2021. The data were collected from annual reports and financial statements of banks available at Orbis Bank Focus database. The study employed a random effect regression model to test the hypotheses and discuss the results. The findings show that banks in the MENA region are increasingly interested in disclosing cybersecurity information, where cybersecurity disclosure over the sample years is increasing from 17% in 2019 to 19.6% in 2021. In addition, the results show that cybersecurity disclosure has a positive and significant influence on bank performance. Furthermore, the findings indicate that the presence of a CRO moderates the relationship between cybersecurity disclosure and bank performance. These findings show that depending largely on a bank's CRO to handle complex and dynamic risks can have serious consequences for decision making processes connected to managing cybersecurity risk and disclosure. This paper creates a new research paradigm by focusing on the disclosure of cybersecurity information in the MENA banking sector, where exploring the moderating role of the CRO, IT committee, and board size in enhancing the cybersecurity disclosure-bank performance relationship is lacking. The findings provide practical implications for various stakeholders, where it reveals the current practices of cybersecurity disclosure of banks in the MENA region with the objective of minimizing information asymmetry, maintaining public trust, and identifying potential risks of financial distress. In addition, the results direct the attention of banks and regulators toward the role of CRO in risk governance, particularly in managing cyber risks within the banking industry.

Building Financial Wellness: Randomized Controlled Trial of a Financial Education and Support Intervention.

Many people receiving services for psychiatric disorders live on low incomes, navigate complex financial situations, and have limited economic security. The authors sought to determine whether a financial wellness intervention delivered virtually by peers would increase financial literacy, reduce economic strain, and improve financial competency. One hundred participants receiving services for psychiatric disorders were recruited from community programs and via social media and were randomly assigned (1:1) to receive either an intervention called Building Financial Wellness (N=51) or services as usual (N=49). The intervention involved six virtual 1.5-hour group sessions, followed by three virtual one-on-one booster sessions. Outcomes were assessed with the Consumer Financial Protection Bureau's Financial Well-Being Scale (primary), Financial Strain Questionnaire, Financial Capability Scale, and Financial Self-Sufficiency Scale. Knowledge regarding money management was measured with a financial literacy assessment based on class content. Intent-to-treat longitudinal multivariable analysis included random-effects linear and logistic regression models. Overall, 86% of the study participants either lived in poverty (46%) or met the federal definition of low income (40%). Participants assigned to the intervention had greater improvement over time in financial well-being, capability, strain, self-sufficiency, and literacy. Satisfaction with the intervention was high. Reports from participants receiving booster sessions indicated that ongoing support facilitated their achievement of longer-term financial goals. Provision of financial education and support improved monetary well-being, competence, and literacy. With poverty long considered to be an intractable problem associated with psychiatric disorders, peer-taught financial education focused on financial wellness can help people acquire money management skills and manage economic stressors.

Who’s in? Variance in top management team definition effect on corporate political activity

PurposeCausal relationships between top management team (TMT) characteristics and corporate political activity (CPA) are well established in management literature. Yet, recent research has uncovered that there is little consistency in how researchers operationalize the TMT. Thus, the purpose of this paper is to empirically demonstrate how variation in TMT operationalization can affect causal relationships between TMT characteristics and CPA. Theoretical implications for these findings are discussed.Design/methodology/approachThis study examines how varying the operationalization of the TMT affects the statistical outcomes of random-effects negative binomial regression models that examine how the state-level CPA of the largest 100 firms headquartered in Texas (from 2000 to 2010) is influenced by TMT characteristics.FindingsHow the TMT is operationalized greatly affects the level and significance of study results. These findings inform TMT and CPA research theoretically and statistically.Research limitations/implicationsThis study underscores the importance of definition of TMT in studies where it is a variable.Originality/valueThis study provides novel insights into how TMT operationalization can greatly affect TMT research and theory.

The Possibility of Financial Distress and Its Determinants: Evidence from the Banking Sector of Bangladesh

The banking sector of Bangladesh has shown poor performance for over a decade now. The non-performing loans increased over the years. This prompts the question of whether the banks are capable of maintaining financial stability and continue as a going concern. This is further fueled by the number of financial scandals which took place over the years, such as the Hallmark scandal, irregularities in Islami Bank Bangladesh Ltd., BASIC Bank Limited etc. that further highlights the signs of possible financial distress. These incidents call for an investigation of bank-specific determinants that might have a significant effect in leading a bank towards bankruptcy. This study has used the data of 30 listed commercial banks in Bangladesh, from the year 2014 to 2020. The bankruptcy risk is measured using the Altman Z-score, which has been extensively used as a suitable bankruptcy risk measurement tool. Statistical analysis has been conducted, using random-effects regression model and correlation, to test the relationship and impact of bank-specific factors. Capital adequacy ratio has shown significantly positive impact on the bankruptcy risk, while age has shown significantly negative impact on bankruptcy risk. Non-performing loan has also shown negative relationship with Z-score, which is contrary to the conventional belief.

Evaluating Board Characteristics’ Influence on the Readability of Annual Reports: Insights from the Egyptian Banking Sector

This study aims to examine the impact of board characteristics (BCs) on banks’ annual reports readability (BARR). Further, it examines whether bank size (BS) moderates the association between BC and BARR. The study employs a sample of 208 bank-year observations from both listed and non-listed banks in the Egyptian stock exchange (EGX), utilizing data spanning from 2016 to 2023. The study employs a random-effect regression model to test the hypotheses and discuss the results. The results suggest that BARR has a significant association with board meetings, gender and cultural diversity. Furthermore, BS played a moderating role in determining the association between BCs and BARR, supporting the second hypothesis. The findings show that the BCs and disclosure quality differ for banks of varying sizes. The findings have practical implications for the Egyptian banking sector, highlighting that board structure is critical to transparency and maintaining public trust. Additionally, the results focus policymakers’ attention on standardizing the contents and structure of banks’ annual reports, with the aim of reducing managers’ manipulation of disclosures and reducing the level of information asymmetry between stockholders, as suggested by the agency theory.

Tax Aggressiveness and Earnings Management: Evidence from Listed Commercial Banks in Nigeria

This study examined the extents to which tax aggressiveness affects earnings management of listed commercial banks in Nigeria. Ex-post facto research design and panel data was obtained for twelve (12) listed commercial banks from 2014-2023. Tax aggressiveness was measured using cash effective rate while two earnings management models were employed namely, Chang and Jones models. Data collected were analyzed through descriptive, post-estimation and inferential statistical techniques. The fixed and random effects regression models showed that tax aggressiveness significantly affects earnings management models. In view of the findings, it recommends the needs for management of listed commercial banks to dampen earnings management practices since it is negatively influenced by the level of tax aggressiveness. Furthermore, management of publicly listed commercial banks needs to engage in aggressive tax planning to enhance their performance; however such level of tax aggressive planning should be kept minimal as possible to further discourage tax fraud and evasion.

CSR associations and market value: the moderating role of market competition

Purpose This paper aims to study the impact of consumers’ corporate social responsibility (CSR) associations on corporate financial performance and the moderating role of market competition. Design/methodology/approach The panel data set is analyzed using a random effects regression model. The analyzed data is based on the unique RepZ Responsibility scores published by the global research agency Kantar Millward Brown and contains information about consumer CSR associations. Findings This study reveals CSR associations' positive, lagged, direct impact on firms’ market value. Market competition moderates this relationship in the way that a company’s market value benefits more from consumers' CSR associations when facing high rather than low market competition. Practical implications Consumers' CSR perceptions increase the market value of a company. This effect is intensified when brands are exposed to intense competition, which allows conclusions about CSR as a differentiation strategy to be drawn: To stand out in a competitive market, brands should prioritize improving their CSR associations among consumers to differentiate themselves and increase their market value. Originality/value To the best of the authors’ knowledge, this study is the first to test the effect of consumers’ CSR associations on forward-looking financial performance measures. Moreover, by analyzing the moderating effect of market competition on the relationship between CSR associations and firms' market value, this study provides information about the differentiating power of CSR from a brand perspective using a panel-data analysis.

From commitment to capital: does environmental innovation reduce corporate cash holdings?

ABSTRACT Using 7066 firm-year observations of the US-listed firms for the period of 2008–2021, we examine a channel through which eco-innovation affects Corporate Cash holding. More specifically, we modeled the role of goodwill between eco-innovation and Corporate Cash holding. Using a random effect regression model, we show the economically relevant and statistically significant negative impact of eco-innovation on firm-level cash holdings. We provide evidence that goodwill fully mediates the relationship between eco-innovation and cash holdings. Our analyses further document that the impact of eco-innovation on firm-level cash holdings is more pronounced for the firms that operate in environmentally insensitive industries. The influence of eco-innovation on cash holdings is less for low-size firms relative to the high size and more for highly leveraged firms. The results are robust for alternative measures of eco-innovation, selection biases and potential endogeneity. The findings have imperative implications for stakeholders’ engagement from a social capital perspective.

ChatGPT Performs Worse on USMLE-Style Ethics Questions Compared to Medical Knowledge Questions.

The main objective of this study is to evaluate the ability of the Large Language Model Chat Generative Pre-Trained Transformer (ChatGPT) to accurately answer the United States Medical Licensing Examination (USMLE) board-style medical ethics questions compared to medical knowledge-based questions. This study has the additional objectives of comparing the overall accuracy of GPT-3.5 to GPT-4 and assessing the variability of responses given by each version. Using AMBOSS, a third-party USMLE Step Exam test prep service, we selected one group of 27 medical ethics questions and a second group of 27 medical knowledge questions matched on question difficulty for medical students. We ran 30 trials asking these questions on GPT-3.5 and GPT-4 and recorded the output. A random-effects linear probability regression model evaluated accuracy and a Shannon entropy calculation evaluated response variation. Both versions of ChatGPT demonstrated worse performance on medical ethics questions compared to medical knowledge questions. GPT-4 performed 18% points (p < 0.05) worse on medical ethics questions compared to medical knowledge questions and GPT-3.5 performed 7% points (p = 0.41) worse. GPT-4 outperformed GPT-3.5 by 22% points (p < 0.001) on medical ethics and 33% points (p < 0.001) on medical knowledge. GPT-4 also exhibited an overall lower Shannon entropy for medical ethics and medical knowledge questions (0.21 and 0.11, respectively) than GPT-3.5 (0.59 and 0.55, respectively) which indicates lower variability in response. Both versions of ChatGPT performed more poorly on medical ethics questions compared to medical knowledge questions. GPT-4 significantly outperformed GPT-3.5 on overall accuracy and exhibited a significantly lower response variability in answer choices. This underscores the need for ongoing assessment of ChatGPT versions for medical education.

Statistics in Phonetics

Phonetics is the scientific field concerned with the study of how speech is produced, heard, and perceived. It abounds with data, such as acoustic speech recordings, neuroimaging data, or articulatory data. In this article, we provide an introduction to different areas of phonetics (acoustic phonetics, sociophonetics, speech perception, articulatory phonetics, speech inversion, sound change, and speech technology), an overview of the statistical methods for analyzing their data, and an introduction to the signal processing methods commonly applied to speech recordings. A major transition in the statistical modeling of phonetic data has been the shift from fixed effects to random effects regression models, the modeling of curve data (for instance, via generalized additive mixed models or functional data analysis methods), and the use of Bayesian methods. This shift has been driven in part by the increased focus on large speech corpora in phonetics, which has arisen from machine learning methods such as forced alignment. We conclude by identifying opportunities for future research.

Are social protection and food security accelerators for adolescents to achieve the Global AIDS targets?

Without effective, scalable interventions, we will fail to achieve the Global AIDS Targets of zero AIDS-related deaths, zero HIV transmission and zero discrimination. This study examines associations of social protection and food security among adolescents living with HIV (ALHIV), with three Global AIDS Targets aligned outcomes: antiretroviral treatment (ART) adherence and viral suppression, HIV transmission risk behaviour and enacted stigma. We conducted three study visits over 2014-2018 with 1046 ALHIV in South Africa's Eastern Cape province. Standardized surveys provided information on receipt of government-provided cash transfers and past-week food security, alongside self-reported ART adherence, sexual debut and condom use, and enacted HIV-related stigma. Viral load (VL) data was obtained through data extraction from patient files and linkage with National Health Laboratory Service test results (2014-2020). We used a multivariable random-effects regression model to estimate associations between receiving government cash transfers and food security and three outcomes: ART adherence and viral suppression, delayed sexual debut or consistent condom use and no enacted stigma. We tested moderation by sex and age and fitted disaggregated models for each outcome. Among the 933 ALHIV completing all three study visits, 55% were female, and the mean age was 13.6 years at baseline. Household receipt of a government cash transfer was associated with improvements on all outcomes: ART adherence and viral suppression (aOR 2.03, 95% CI 1.29-3.19), delayed sexual debut or consistent condom use (aOR 1.62, 95% CI 1.16-2.27) and no enacted stigma(aOR 2.33, 95% CI 1.39-3.89). Food security was associated with improvements on all outcomes: ART adherence and viral suppression (aOR 1.73, 95% CI 1.30-2.30), delayed sexual debut or consistent condom use (aOR 1.30, 95% CI 1.03-1.64) and no enacted stigma (aOR 1.91, 95% CI 1.32-2.76). Receiving both cash transfers and food security increased the probability of ART adherence and VL suppression from 36% to 60%; delayed sexual debut or consistent condom use from 67% to 81%; and no enacted stigma from 84% to 96%. Government-provided cash transfers and food security, individually and in combination, are associated with improved outcomes for ALHIV aligned with Global AIDS Targets. They may be important, and underutilized, accelerators for achieving these targets.

Human Resource Accounting and Share Price Performance Evidence from Listed Insurance Firms in Nigeria

This study investigates the impact of Human Resource Accounting (HRA) on share price performance among listed insurance firms in Nigeria. Specifically, the study examines the effects of human capital efficiency, structural capital efficiency, capital employed efficiency, and the value-added intellectual capital coefficient (VAIC) on share price performance. This research adopts ex-post facto and descriptive research designs, utilizing data from 13 insurance firms listed on the Nigerian Exchange Group from 2012 to 2021. Random effect regression model was employed to address statistical issues ensuring robust results, after controlling for multicollinearity and heteroscedasticity. The findings reveal that human capital efficiency and capital employed efficiency significantly enhance share price performance, highlighting the importance of these factors in driving market valuation. Conversely, VAIC shows a significant negative impact on share price performance, suggesting that not all aspects of intellectual capital positively influence market outcomes in the insurance sector. These results contribute to a deeper understanding of how human resource investments affect financial performance, particularly in the context of emerging markets like Nigeria. Based on these findings, this study recommends among others that insurance firms should prioritize investments in human capital development through targeted training and continuous professional development. This approach will enhance employee productivity and innovation, which in turn would positively influence share price performance and contribute to the overall financial health and market competitiveness of the firm.

Subjective Well-Being of Parents and Childless People in Older Age in Germany

The growing number of older childless individuals in Western societies has raised concerns about their subjective well-being. In this study, we scrutinize the subjective well-being of parents and childless individuals aged 60 years and older. We examine subjective well-being as a multi-dimensional construct encompassing life satisfaction, positive affect, and negative affect. Based on the value-of-children approach, we hypothesize that childless individuals experience lower subjective well-being than parents and that the disparity in well-being differs with age, gender, and the quality of the parent-child relationship. For our analysis, we use data from a nationally representative sample of adults aged 60 years and older (n = 10,682) drawn from the German Ageing Survey. Our findings from random effects regression models show that childless individuals report lower life satisfaction, less positive affect, and more negative affect than parents. These associations are larger for men than for women but do not vary significantly across age groups. Furthermore, our results suggest that the benefits of parenthood strongly depend on the quality of the parent-child relationship. Childless people report lower subjective well-being than parents with a close emotional bond with their children, but similar or even higher subjective well-being than parents with an emotionally distant parent-child relationship. In sum, our study indicates that childless people experience, on average, lower well-being across all three dimensions of subjective well-being in older age. However, under specific circumstances, childless individuals may fare better than parents.

Community Interaction, Influence, and Child Maltreatment: Variations by Rural and Urban Residence Status

Effects of community interventions on child maltreatment are interpreted via rational deterrence theory’s assumption of instrumental rationality. However, recent contrasting findings have raised questions about the social exchange currency that regulates human behaviour within the rational deterrence framework. This study contributes to this debate by examining the social currency of influence and its role in child maltreatment prevention. We hypothesized that both individual and community wealth of influence would predict physical abuse severity. Data was obtained from 1,100 nationally representative sample of female mothers in Ghana, and associations were examined using random effects regression models. We found a negative association between physical abuse severity and individual level influence as well as community wealth of influence. Results were only significant in the rural sample. The findings suggest that interventions aiming to increase community commitment to deter and prevent child maltreatment should focus on strengthening the currency of influence in communities.

Blocked Pipes: The Economic Consequence of Skilled Worker Shortages

The thinning of the supply of skilled labor is having major negative implications on the economy. This research explores the “skilled labor gap” by examining the plumbing industry and quantifying the economic value lost through inefficiencies and labor premiums. The analysis utilizes state-level panel data for the years 2010 to 2018, employing both a linear OLS model of pooled data and a random effects regression model to determine the relationship between the number of plumbers and the price of plumbing services (proxied in plumber wages). Price effects are modeled across industries to determine the effects on the overall economy.

Police personnel allocation and homicide clearance

PurposeThis paper applies novel techniques from the field of operations management to examine the allocation of patrol and investigative personnel to identify which is most effective in improving police performance around homicide clearance.Design/methodology/approachA panel sample of homicide clearance rates from the 100 largest US cities between 2000 and 2013 were analyzed in two steps: first, a random-effects regression model was performed to locate influential factors; second, optimum analysis was applied to locate the optimal values that yield maximal homicide clearance.FindingsBoth patrol and investigative personnel levels have a significant impact on homicide clearance. Maximal clearance can be achieved by allocating departmental personnel to investigative roles.Research limitations/implicationsGiven recent trends around “defunding” police and public sector austerity measures, future research should continue to explore the utility of optimum analysis for efficient allocation of policing personnel.Originality/valueThis study provides proof of concept for the use of optimum analysis in policing research.