Search results

Although the use of data from different levels is very common in international marketing research, the practice of employing multi‐level analysis techniques is relatively new. The paper aims to provide an application of a specific case of multi‐level modelling – where the dependent variable is dichotomous, which is often the case in marketing research (e.g. whether a consumer buys the brand or not, whether he/she is aware of the brand or not, etc.)

A hierarchical generalized linear model is employed.

Since this is a technical paper, the authors would like to emphasize the process rather than the empirical findings. In summary, the paper: provides a brief theoretical overview of Hierarchical Linear Modeling and Hierarchical Generalized Linear Modeling; illustrates the application of the method using the domains of consumers within countries and a dichotomous dependent variable; focuses on interpretation of log‐odds results; and concludes with practical issues and research implications.

The main value of this research is to demonstrate how to employ multi‐level models when the dependent variable is dichotomous. Multi‐level techniques are quite new in international marketing research, although nested data structures are relatively common in our field. This is a technical paper that guides the researchers as to how to apply and interpret the results when modeling such data with a dichotomous dependent variable.

Purpose – This chapter provides an overview of multilevel modeling with a focus on the application of hierarchical linear modeling (HLM) in international management research.

Findings – The key topics covered include an introduction to hierarchical linear models, how to apply appropriate hierarchical linear models to address different types of international management research questions, and six methodological issues concerning international management research with a multilevel analysis.

Originality/value – The overview of HLM and its relevance for international management research facilitates researchers to apply this powerful analytical strategy in their future research.

Prior studies of criminal sentencing have largely focused on individual-level predictors of sentencing outcomes. The purpose of this paper is to examine the effects of a variety of theoretically derived community measurements of social threat and disadvantage on the criminal sentencing of convicted felons. This analysis permits an evaluation of whether legal ideals such as equality before the law and policy goals of equal treatment for like offenders are achieved.

The study examines data of individuals sentenced in the state of Florida and community level measurements of racial and ethnic threat and community disadvantage. Hierarchical generalized linear model is used to analyze the effect of these measures on the dichotomous in/out imprisonment variable, and standard hierarchical linear regression analysis is used to model the continuous dependent variable of sentence length.

The results provide support for the racial threat perspective though not for ethnic threat nor community disadvantage. The findings and their implications are discussed in terms of theory, research and policy.

Racial disparity in criminal justice practices is receiving increasing public and policy attention, as evidenced by the growing Black Lives Matter movement. Regarding sentencing, racial disparity remains a major research and policy question. While the current research and theoretical literature on sentencing is not conclusive, it is clear that race matters. As a result, racial disparity in sentencing needs to be a priority in subsequent “transitional criminology” efforts between researchers and policy makers to identify, explain and ultimately predict exactly how race impacts sentencing, and how to reduce it as a consideration from sentencing.

This study contributes to a growing body of literature that examines the social context of punishments by using several community level measurements of threat and disadvantage, while modeling the two-step sentencing outcome of imprisonment and sentence length.

The purpose of this study is to examine the influence of factors affecting students’ choice of accounting as a study major in Hong Kong.

Multinomial logistic regression and Hierarchical Generalized Linear Modeling (HGLM) are used to analyze the survey data for the level one and level two data, which is the first time such an approach has been used in the literature. Twenty semi‐structured interviews are conducted.

Results reveal that parental influence has the highest explanatory power among all three groups of students (i.e. accounting major, accounting minor, and other majors). This finding reflects the inculcation in students of the Confucian cultural norms ingrained in Chinese societies. Intrinsic interest is the second most important influencing factor reported by students. Two factors, “intrinsic interest” (attitude towards the behavior) and “parental influence” (subjective norm) in the multinomial logistic regressions and HGLM make significant contributions to predicting a student's decision to major in accounting for the level one (undergraduate) and level two model (master's students). The results support the theory of reasoned action (TRA) model. “First accounting course,” “high school accounting,” and “financial rewards” are not found to be factors influencing the decision to major in accounting, in contrast to the findings of the literature prior to the Enron scandal. Females are more likely to choose accounting as a major or minor than are males in the level one and level two models.

The findings indicate that Hong Kong universities and the Hong Kong Institute of Certified Public Accountants (HKICPA) need to promote accounting to improve the current negative image of the profession to attract more bright students, who traditionally might choose a non‐accounting major.

The findings of this study extend the application of the TRA model to the accounting profession. Besides, results provide insights for employers to better understand the mindsets of potential accounting graduates.

Grounded in theory of planned behavior, this study explores how national social capital, as an important but untested element of institutional environment, influences the effect of perceived entrepreneurial ability on entrepreneurial intentions.

Using data of more than 95,000 individuals in 33 countries from Adult Population Survey (APS) conducted by Global Entrepreneurship Monitor (GEM), combined with data collected by World Values Survey (WVS), this study employs hierarchical linear modeling (HLM) to assess the moderation effect of national social capital on the relationship between perceived entrepreneurial ability of individuals nested within different countries and entrepreneurial intentions.

The results reveal that national social capital enhances the positive impact of perceived entrepreneurial ability on the intent of individuals to start a new business. Notably, this study finds all three components of national social capital – generalized trust, breadth of formal organizational memberships and civic engagement– strengthen the influence of perceived entrepreneurial ability on entrepreneurial intentions.

As the main contribution, this research illustrates the combined effects of social environment (i.e. national social capital) and individuals' evaluations regarding their ability to start a business (i.e. perceived entrepreneurial ability) on entrepreneurial intentions. This study highlights that national social capital complements perceived entrepreneurial ability by helping identify favorable start-up conditions and increasing the tendency of people to act on those conditions.

Following social disorganization theory and the process-based model of policing, the present study aims to examine how characteristics of communities are meaningful in shaping citizens' willingness to work with the police in urban China.

Based on survey data from Shanghai, China, the study adopts a generalized hierarchical linear modeling (GHLM) approach to examine the effects of both individual- and neighborhood-level predictors on the outcome, while taking into consideration the unobserved additional neighborhood-level variations.

Results suggest potential need of the process-based model to be modified in a Chinese context in that while police presence matters, measures on legitimacy are nonsignificant in shaping willingness to help the police, after controlling for neighborhood characteristics. More importantly, collective efficacy at the neighborhood level is related positively to residents' willingness to work with the police. Constant attention should be paid to the promotion of a collaborative and collectively caring environment, which can contribute to better coordination with the police, and greater willingness to be part of the larger cause of public safety.

The paper advances the scholarship with multi-level modeling and the role of communities in building stronger relationship with the police.

The purpose of this paper is to provide the first large data-set regression analysis to test Kerlin’s (2013) macro-institutional social enterprise framework in relation to the country social enterprise models that flow from it. Kerlin (2013) offers a conceptual framework for country social enterprise models that allows countries to retain their unique understanding of social enterprise and better understand the factors influencing its development.

This paper draws on the theory of historical institutionalism and multiple global datasets to test formal hypotheses on the relationship between macro-institutional factors and the size of the social enterprise sector across countries. Social enterprise data were obtained from the 2009 Global Entrepreneurship Monitor dataset. Hypotheses were tested using logistic hierarchical linear modeling.

Results provide support for the framework at a significant level. Nearly half of the variance in the size of the social enterprise sector can be attributed to countries-level factors. We also find that the size of the social enterprise sector varies by economic competitiveness rank, size of the welfare state and collectivist cultural orientation.

The countries included in this study are not representative of the global landscape. Researchers are encouraged to test the framework with a more representative sample of countries, including those in the Global South.

The paper includes implications for policy makers and researchers seeking to facilitate cross-regional dialogue, the transfer and support of social enterprises and research.

This paper fulfills an identified need to advance the field of social enterprise by quantitatively testing established frameworks.

Police factor in extra‐legal as well as legal context in their decision to arrest a suspect. The purpose of this paper is to examine the effect of extra‐legal factors at both situational and neighborhood levels.

Using hierarchical generalized linear modeling, over 9,000 domestic violence cases across 421 census tracts in Houston, Texas were examined. Situational information was derived from police reports, and neighborhood factors were measured by population characteristics drawn from the US Census Bureau. The model also controls for spatial autocorrelation of arrest rates between census tracts in the estimation of officer's arrest decision.

At the neighborhood level, concentrated disadvantage and immigration concentration had positive effect on the odds of arrest. At the situational level, the time of day, day of the week, premise type, and gender and racial relations between suspect and complainant, along with offense type and weapons use, had significant impact.

The scant literature has not yet provided an affirmative set of extra‐legal factors affecting police arrest decision. The paper's findings may contribute to the literature and suggest the need for guidelines concerning officer discretion exercised in the line of duty.

The majority of the previous research on hot spots policing focuses on a single set of relatively small selected experimental areas. However, given limited resources, most law enforcement agencies dispatch hot spots intervention units to several areas on a rotation basis. The purpose of this paper is to examine policing activities in hot spots to determine if the various types of crimes were affected when deployment was applied on a rotation basis.

This study uses data from the Dallas Police Department. The differential influence of police activities, including stops, citations, and arrests, are observed against a number of aggregate crime measures (i.e. violent, property, nuisance offenses, and total index crimes). The impact of police activities have been observed for their immediate and lagged effects during the following week to measure residual deterrence effects.

It was found that the DPD's Disruption Unit's hot spots policing immediately affected violent crimes, nuisance offenses, and total index crimes, while there were no residual effects of hot spots policing. The Disruption Unit was engaged in policing activities that include motor vehicle and pedestrian stops, issuing citations, and making arrests. Among these activities, the number of police stops was the most significant factor for the reduction in violent crime and nuisance offenses.

The researchers use a patrol sector as a unit of analysis in order to compare the influence of various types of police activities on crime across a broader area. Future research should consider using an intermediate geographic unit of analysis (e.g. patrol beat).

The paper examines the differential influence of policing activities on different types of crime around hot spots when deployment was applied on a rotation basis. Both immediate and lagged effects were investigated to find residual deterrence effects of hot spots policing.

Business victimization is a serious and pervasive issue within the USA. According to recent estimates, roughly 2,058,194 businesses are victimized each year. Of those, approximately 33 percent of business victimization cases are solved. Taken together, it is important for research to examine factors that influence business victimization clearance. The purpose of this paper is to examine how broken windows enforcement, social disorganization, community and police organizational factors influence business robbery clearance using data from Houston, Texas over a two-year period from 2010 to 2012.

Using a hierarchical linear modeling strategy, the current study found no effect of broken windows tactics, social disorganization elements and various organizational, and community characteristics on business robbery clearance.

Significant effects were found for a number of incident and offense characteristics including gang involvement, business type, type of weapon used in the crime, the number of business entities in an area, and racial populous.

To date, few studies have examined factors that influence clearance rates for business robberies. Thus, the current study adds to and extends upon the literature in theoretically relevant ways by exploring how broken windows policing, social disorganization and various community/police organizational variables influence business robbery clearances in a large city.