Using multilevel modeling and mixed methods to make theoretical progress in microfoundations for strategy research

The microfoundations research agenda presents an expanded theoretical perspective because it considers individuals, their characteristics, and their interactions as relevant variables to help us understand firm-level strategic issues. However, microfoundations empirical research faces unique challenges because processes take place at different levels of analysis and these multilevel processes must be considered simultaneously. We describe multilevel modeling and mixed methods as methodological approaches whose use will allow for theoretical advancements. We describe key issues regarding the use of these two types of methods and, more importantly, discuss pressing substantive questions and topics that can be addressed with each of these methodological approaches with the goal of making theoretical advancements regarding the microfoundations research agenda and strategic management studies in general.

Ontological Multilevel Modeling Language

This paper presents ontological multilevel modeling language O2ML, aimed at using with metadata driven information systems. The first part of this paper briefly surveys existing modeling languages and approaches, while the last part proposes a new language to combine their benefits.

Multilevel analysis of hepatitis A infection in children and adolescents: a household survey in the Northeast and Central-west regions of Brazil

Background The objectives were to estimate the prevalence of hepatitis A among children and adolescents from the Northeast and Midwest regions and the Federal District of Brazil and to identify individual-, household- and area-levels factors associated with hepatitis A infection. Methods This population-based survey was conducted in 20042005 and covered individuals aged between 5 and 19 years. A stratified multistage cluster sampling technique with probability proportional to size was used to select 1937 individuals aged between 5 and 19 years living in the Federal capital and in the State capitals of 12 states in the study regions. The sample was stratified according to age (59 and 10- to 19-years-old) and capital within each region. Individual- and household-level data were collected by interview at the home of the individual. Variables related to the area were retrieved from census tract data. The outcome was total antibodies to hepatitis A virus detected using commercial EIA. The age distribution of the susceptible population was estimated using a simple catalytic model. The associations between HAV infection and independent variables were assessed using the odds ratio and corrected for the random design effect and sampling weight. Multilevel analysis was performed by GLLAMM using Stata 9.2. Results The prevalence of hepatitis A infection in the 59 and 1019 age-group was 41.5 and 57.4, respectively for the Northeast, 32.3 and 56.0, respectively for the Midwest and 33.8 and 65.1 for the Federal District. A trend for the prevalence of HAV infection to increase according to age was detected in all sites. By the age of 5, 31.5 of the children had already been infected with HAV in the Northeast region compared with 20.0 in the other sites. By the age of 19 years, seropositivity was 70 in all areas. The curves of susceptible populations differed from one area to another. Multilevel modeling showed that variables relating to different levels of education were associated with HAV infection in all sites. Conclusion The study sites were classified as areas with intermediate endemicity area for hepatitis A infection. Differences in age trends of infection were detected among settings. This multilevel model allowed for quantification of contextual predictors of hepatitis A infection in urban areas.

Adolescents' value of children and their intentions to have children: A cross-cultural and multilevel analysis

This study contributes to research regarding the value of children (VOC) by comparing adolescents' VOC and their intentions to have children across 12 cultures and by exploring the relations between these constructs at the individual and cultural levels using multilevel modeling. A total of 3,348 adolescents from 12 cultures participated in this study. On average, adolescents reported that they intended to have about two children and also reported emotional VOC as being highly important. Adolescents from cultures with a low as compared to a high level of economic development reported a higher importance of the utilitarian-normative VOC. Results of the multilevel analyses showed that the reported emotional VOC was positively related to the number of children adolescents intended to have at the individual level, whereas the utilitarian-normative VOC was not related to adolescents' intention to have children. At the cultural level, the VOC dimensions were only partly related to adolescents' intention to have children. The results are discussed with regard to adolescents' future family orientation and in relation to the VOC approach.

Evaluation of the accuracy of multilevel analysis parameter estimates for intra-cluster correlation in correlated binary data

Many studies in biostatistics deal with binary data. Some of these studies involve correlated observations, which can complicate the analysis of the resulting data. Studies of this kind typically arise when a high degree of commonality exists between test subjects. If there exists a natural hierarchy in the data, multilevel analysis is an appropriate tool for the analysis. Two examples are the measurements on identical twins, or the study of symmetrical organs or appendages such as in the case of ophthalmic studies. Although this type of matching appears ideal for the purposes of comparison, analysis of the resulting data while ignoring the effect of intra-cluster correlation has been shown to produce biased results.^ This paper will explore the use of multilevel modeling of simulated binary data with predetermined levels of correlation. Data will be generated using the Beta-Binomial method with varying degrees of correlation between the lower level observations. The data will be analyzed using the multilevel software package MlwiN (Woodhouse, et al, 1995). Comparisons between the specified intra-cluster correlation of these data and the estimated correlations, using multilevel analysis, will be used to examine the accuracy of this technique in analyzing this type of data. ^

Bullying Involvement and Adolescent Substance Use: A Study of Multilevel Risk and Protective Factors

Bullying, frequent drunkenness, and frequent cannabis use are significant health-risk behaviours among youth. While many studies have demonstrated that bullying involvement may initiate a developmental pathway to both types of frequent substance use, there is a limited understanding of the connection between these behaviours. The presence of risk and protective factors within youths’ relationships and within their neighbourhoods may alter the associations between bullying involvement and both types of frequent substance use. A systemic approach is needed to assess the complex, social environments in which youth are embedded. The current thesis consists of two studies that examined the associations between bullying and both types of frequent substance use within the context of youths’ social environments. In Study 1, multilevel modeling was used to examine the associations between bullying and frequent substance use within the context of individual and neighbourhood risk factors. Our results indicated that the risk factors associated with both frequent drunkenness and frequent cannabis use exist at both levels, with neighbourhoods altering the association of individual risk factors. Moreover, bullying was a unique risk factor associated with both types of frequent substance use, whereas indirect associations were observed for victimization. Study 2 used a similar methodology to examine the association between bullying and both types of frequent substance use within the context of individual and neighbourhood protective factors. Once again, our results indicated that the protective factors associated with both types of frequent substance use exist at multiple levels, and that neighbourhoods altered the association of individual protective factors. Additionally, positive relationship characteristics interacted with the link between bullying and both types of frequent substance use. Together, these findings clarify the nature of the bullying-substance use link and emphasize the need to study adolescent development in context.

Bullying Involvement and Adolescent Substance Use: A Study of Multilevel Risk and Protective Factors

Thesis (Master, Psychology) -- Queen's University, 2016-08-02 13:18:40.65

Intraindividual allometric development of aerobic power in 8- to 16-year-old boys

Purpose: The aims of this study are two-fold: first, to analyze intraindividual allometric development of aerobic power of 73 boys followed at annual intervals from 8 to 16 yr, and second, to relate scaled aerobic power with level of habitual physical activity and biological maturity status. Methods: Peak (V) over dot O-2 (treadmill), height, and body mass were measured. Biological maturity was based on age at peak height velocity (PHV) and level of physical activity was based on five assessments between 11 and 15 yr and at 17 yr. Interindividual and intraindividual allometric coefficients were calculated. Multilevel modeling was applied to verify if maturity status and activity explain a significant proportion of peak (V) over dot O-2 after controlling for other explanatory characteristics. Results: At most age levels, interindividual allometry coefficients for body mass exceed k = 0.750. Intraindividual coefficients of peak (V) over dot O-2 by body mass vary widely and range from k' = 0,555 to k' = 1,178. Late maturing boys have smaller k' coefficients than early maturing boys. Conclusion: Peak (V) over dot O-2 is largely explained by body mass, but activity level and its interaction with maturity status contribute independently to peak (V) over dot O-2 even after adjusting for body mass.

Modelos estadísticos y evaluación: tres estudios en educación

The educational system in Spain is undergoing a reorganization. At present, high-school graduates who want to enroll at a public university must take a set of examinations Pruebas de Aptitud para el Acceso a la Universidad (PAAU). A "new formula" (components, weights, type of exam,...) for university admission is been discussed. The present paper summarizes part of the research done by the author in her PhD. The context for this thesis is the evaluation of large-scale and complex systems of assessment. The main objectives were: to achieve a deep knowledge of the entire university admissions process in Spain, to discover the main sources of uncertainty and topromote empirical research in a continual improvement of the entire process. Focusing in the suitable statistical models and strategies which allow to high-light the imperfections of the system and reduce them, the paper develops, among other approaches, some applications of multilevel modeling.

Another way of thinking about ADHD: the predictive role of early attachment deprivation in adolescents' level of symptoms.

PURPOSE: Attention deficit and hyperactivity disorder (ADHD) is one of the most frequent disorders in childhood and adolescence. Both neurocognitive and environmental factors have been related to ADHD. The current study contributes to the documentation of the predictive relation between early attachment deprivation and ADHD. METHOD: Data were collected from 641 adopted adolescents (53.2 % girls) aged 11-16 years in five countries, using the DSM oriented scale for ADHD of the Child Behavior Checklist (CBCL) (Achenbach and Rescorla, Manual for the ASEBA school-age forms and profiles. University of Vermont, Research Center for Children, Youth and Families, Burlington, 2001). The influence of attachment deprivation on ADHD symptoms was initially tested taking into consideration several key variables that have been reported as influencing ADHD at the adoptee level (age, gender, length of time in the adoptive family, parents' educational level and marital status), and at the level of the country of origin and country of adoption (poverty, quality of health services and values). The analyses were computed using the multilevel modeling technique. RESULTS: The results showed that an increase in the level of ADHD symptoms was predicted by the duration of exposure to early attachment deprivation, estimated from the age of adoption, after controlling for the influence of adoptee and country variables. The effect of the age of adoption was also demonstrated to be specific to the level of ADHD symptoms in comparison to both the externalizing and internalizing behavior scales of the CBCL. CONCLUSION: Deprivation of stable and sensitive care in infancy may have long-lasting consequences for children's development.

Conceptualizing and counteracting the bullying dynamic: Classroom and school contexts matter

Guided by the social-ecological conceptualization of bullying, this thesis examines the implications of classroom and school contexts—that is, students’ shared microsystems—for peer-to-peer bullying and antibullying practices. Included are four original publications, three of which are empirical studies utilizing data from a large Finnish sample of students in the upper grade levels of elementary school. Both self- and peer reports of bullying and victimization are utilized, and the hierarchical nature of the data collected from students nested within school ecologies is accounted for by multilevel modeling techniques. The first objective of the thesis is to simultaneously examine risk factors for victimization at individual, classroom, and school levels (Study I). The second objective is to uncover the individual- and classroom-level working mechanisms of the KiVa antibullying program which has been shown to be effective in reducing bullying problems in Finnish schools (Study II). Thirdly, an overview of the extant literature on classroom- and school-level contributions to bullying and victimization is provided (Study III). Finally, attention is paid to the assessment of victimization and, more specifically, to how the classroom context influences the concordance between self- and peer reports of victimization (Study IV). Findings demonstrate the multiple ways in which contextual factors, and importantly students’ perceptions thereof, contribute to the bullying dynamic and efforts to counteract it. Whereas certain popular beliefs regarding the implications of classroom and school contexts do not receive support, the role of peer contextual factors and the significance of students’ perceptions of teachers’ attitudes toward bullying are highlighted. Directions for future research and school-based antibullying practices are suggested.

L’influence des caractéristiques organisationnelles de la première ligne médicale sur l’auto-soins en maladies chroniques au Québec

La prise en charge efficace des maladies chroniques nécessite une implication accrue des patients dans la gestion de leurs soins et de leur état de santé. Le lien existant entre les caractéristiques organisationnelles de la 1ère ligne et la capacité des patients pour l’auto-soins étant peu documenté, l’objectif de ce mémoire est de mieux comprendre comment l’organisation de la 1ère ligne médicale peut influencer la capacité des personnes pour l’auto-soins d’une maladie chronique. Pour atteindre cet objectif, 776 patients atteints de maladie chronique ont été recrutés dans 33 cliniques ayant participé à une enquête organisationnelle. Ces patients ont répondu à un questionnaire sur l’auto-efficacité pour la gestion d’une maladie chronique à 0, 6 et 12 mois. Des analyses par régression hiérarchique à 3 niveaux ont permis de mesurer les liens entre les variables organisationnelles sélectionnées et la capacité perçue des patients pour l’auto-soins de leur maladie. Les résultats montrent que les caractéristiques organisationnelles ont peu d’effet sur l’auto-efficacité pour l’auto-soins en comparaison aux facteurs individuels, et que le soutien à l’auto-soins est peu présent dans les cliniques participantes sans égard à leurs attributs organisationnels. Bien qu’aucune des caractéristiques sélectionnées n’apparaisse significativement associée à l’auto-soins dans le modèle final, des associations significatives ont été observées en analyses bivariées. Ces résultats suggèrent que toutes les organisations de 1ère ligne doivent fournir un effort explicite pour mieux intégrer l’auto-soins dans les services de santé usuels et que la présence d’éléments organisationnels facilitateurs ne se traduit pas systématiquement par une provision accrue de soutien à l’auto-soins.

Approximation de la distribution a posteriori d'un modèle Gamma-Poisson hiérarchique à effets mixtes

La méthode que nous présentons pour modéliser des données dites de "comptage" ou données de Poisson est basée sur la procédure nommée Modélisation multi-niveau et interactive de la régression de Poisson (PRIMM) développée par Christiansen et Morris (1997). Dans la méthode PRIMM, la régression de Poisson ne comprend que des effets fixes tandis que notre modèle intègre en plus des effets aléatoires. De même que Christiansen et Morris (1997), le modèle étudié consiste à faire de l'inférence basée sur des approximations analytiques des distributions a posteriori des paramètres, évitant ainsi d'utiliser des méthodes computationnelles comme les méthodes de Monte Carlo par chaînes de Markov (MCMC). Les approximations sont basées sur la méthode de Laplace et la théorie asymptotique liée à l'approximation normale pour les lois a posteriori. L'estimation des paramètres de la régression de Poisson est faite par la maximisation de leur densité a posteriori via l'algorithme de Newton-Raphson. Cette étude détermine également les deux premiers moments a posteriori des paramètres de la loi de Poisson dont la distribution a posteriori de chacun d'eux est approximativement une loi gamma. Des applications sur deux exemples de données ont permis de vérifier que ce modèle peut être considéré dans une certaine mesure comme une généralisation de la méthode PRIMM. En effet, le modèle s'applique aussi bien aux données de Poisson non stratifiées qu'aux données stratifiées; et dans ce dernier cas, il comporte non seulement des effets fixes mais aussi des effets aléatoires liés aux strates. Enfin, le modèle est appliqué aux données relatives à plusieurs types d'effets indésirables observés chez les participants d'un essai clinique impliquant un vaccin quadrivalent contre la rougeole, les oreillons, la rub\'eole et la varicelle. La régression de Poisson comprend l'effet fixe correspondant à la variable traitement/contrôle, ainsi que des effets aléatoires liés aux systèmes biologiques du corps humain auxquels sont attribués les effets indésirables considérés.

Déterminants individuels et contextuels de la mortalité des enfants de moins de cinq ans en Afrique au sud du Sahara : analyse comparative des enquêtes démographiques et de santé

La santé des enfants demeure une question prioritaire en Afrique sub-saharienne. Les disparités en matière de mortalité entre pays et au sein des pays persistent et se sont fortement accrues durant la dernière décennie. En dépit de solides arguments théoriques voulant que les variables contextuelles soient des déterminants importants de la santé des enfants, ces facteurs, et particulièrement les influences du contexte local, ont été étudiées beaucoup moins souvent que les caractéristiques individuelles. L'objectif principal de la présente thèse est d’identifier les déterminants individuels et contextuels associés à la mortalité des enfants de moins de 5 ans en Afrique sub-saharienne. L’analyse systématique est basée sur les données les plus récentes des enquêtes démographiques et de santé (DHS/EDS). Deux questions spécifiques sont examinées dans cette thèse. La première évalue la mesure dans la quelle le contexte local affecte la mortalité infanto-juvénile, net des caractéristiques individuelles. La seconde question est consacrée à l’examen de l’effet du faible poids à la naissance sur le risque de décès avant 5 ans. Par rapport à la première question, les analyses multi-niveaux confirment pour plusieurs pays étudiés l’importance simultanée de l’environnement familial et du contexte local de résidence dans l’explication des différences de mortalité infanto-juvénile. Toutefois, par comparaison au contexte familial, l’ampleur de l’effet de l’environnement local paraît assez modeste. Il apparaît donc que le contexte familial reste un puissant déterminant de la mortalité des enfants de moins de 5 ans en Afrique sub-saharienne. Les résultats indiquent en outre que certains attributs du contexte local de résidence influencent le risque de décès des enfants avant 5 ans, au-delà des facteurs individuels dans plusieurs pays. Cette thèse confirme l’effet contextuel de l’éducation sur la mortalité des enfants. Cet effet s’ajoute, dans certains pays, à l'effet positif du niveau individuel d’éducation de la mère sur la survie de l'enfant. Les résultats montrent aussi que le degré d’homogénéité ethnique de la localité influence fortement la probabilité de mourir avant 5 ans dans certains pays. Globalement, les résultats de cette thèse suggèrent que le défi de réduire la mortalité des enfants va au-delà des stratégies visant uniquement les facteurs individuels, et nécessite une meilleure compréhension de l’influence des facteurs contextuels. Par rapport à la deuxième question, les résultats montrent également que les facteurs individuels restent aussi très importants dans l’explication des différences de mortalité des enfants dans plusieurs pays étudiés. Nos résultats indiquent que les différences de mortalité selon le poids à la naissance sont significatives dans tous les pays inclus dans l’analyse. Les enfants nés avec un faible poids (moins de 2500 grammes) courent presque 2 à 4 fois plus de risques de mourir au cours des cinq premières années de vie que les enfants de poids normal, même après correction pour l’hétérogénéité non observée. Ce résultat suggère qu’en plus des mesures visant à réduire la pauvreté et les inégalités de revenus, la réduction de l’incidence du faible poids à la naissance pourrait apporter une contribution majeure aux Objectifs du Millénaire pour le développement; spécialement comme une stratégie efficace pour réduire le niveau de mortalité parmi les enfants de moins de cinq ans.