Direct marketing problems have been intensively reviewed in the marketing literature recently, such as purchase frequency and time, sales profit, and brand choices. However, modeling the customer response, which is an important issue in direct marketing research, remains a significant challenge. This thesis is an empirical study of predicting customer response to direct marketing and applies a Bayesian approach, including the Bayesian Binary Regression (BBR) and the Hierarchical Bayes (HB). Other classical methods, such as Logistic Regression and Latent Class Analysis (LCA), have been conducted for the purpose of comparison. The results of comparing the performance of all these techniques suggest that the Bayesian methods are more appropriate in predicting direct marketing customer responses. Specifically, when customers are analyzed as a whole group, the Bayesian Binary Regression (BBR) has greater predictive accuracy than Logistic Regression. When we consider customer heterogeneity, the Hierarchical Bayes (HB) models, which use demographic and geographic variables for clustering, do not match the performance of Latent Class Analysis (LCA). Further analyses indicate that when latent variables are used for clustering, the Hierarchical Bayes (HB) approach has the highest predictive accuracy.

Indiana University-Purdue University Indianapolis (IUPUI) / Rates of vaping among high school students have increased significantly over the past decade. Prior research has found significant associations between youth vaping and substance use. However, little is known about how vaping is associated with various patterns of polysubstance use and asthma in youth. We aimed to identify how youth are best categorized into classes based on co-occurring vaping and polysubstance use behaviors, how these classes are associated with youth background and demographic characteristics, and if these classes significantly predict asthma outcomes. Our sample consisted of nationally representative data from the 2017 and 2019 waves of the Youth Risk Behavior Survey (N = 28,442). We utilized Latent Class Analysis, multinomial logistic regression analyses, and binary logistic regression analyses to examine relationships between youth vaping, polysubstance use, and asthma. Three latent classes of substance use were identified: Polysubstance Users, Lifetime Alcohol and Vape Users, and Abstainers. Age, gender, grade and race were all significantly associated with odds of membership in the Polysubstance Users class, compared to the Abstainers class. Sexual identity was not associated with class membership. Membership in the Polysubstance Users class was significantly associated with higher odds of asthma, as compared to membership in the other two classes. These findings indicate that recent vaping is associated with high probabilities of recent polysubstance use. They also suggest that youth with high probabilities of vaping and polysubstance use are at significantly high risk for asthma compared to other classes vi of youth users and non-users. We recommend that future youth intervention strategies be tailored differently toward different classes of substance use and vaping. Future research should examine how the classes of vaping and substance use that we identify emerge in youth and what social factors (e.g., peer behavior, parental connectedness, etc.) influence their development.

No description available.

Objectives: To evaluate the effects of an appearance-focused intervention to reduce the risk of skin cancer by decreasing indoor tanning, examine potential heterogeneity in tanning across this time, and correlate the subtypes with predictors collected at baseline. Design: Randomized controlled trial with 379 female college students measured at 6 monthly time points. Main Outcome Measure: Self-reported indoor tanning frequency. Results: The intervention was effective at decreasing tanning over the period between the fall and spring. Longitudinal latent class analysis found 3 patterns of tann1ers among the treatment individuals: abstainers, moderate tanners, and heavy tanners. These classes appeared in both the treatment and control conditions, and the intervention had a harm reduction effect by reducing levels of exposure within the moderate and heavy tanner classes. Participant age and self-reported tanning patterns were found to be predictive of class membership. Conclusions: This research suggests that brief intervention approaches can be effective at reducing risk for skin cancer and illustrates several ways in which these protective effects can be enhanced.

Background For the 1.2 million older adults residing in U.S. nursing homes, little is known about their experience with physical frailty and cognitive impairment, two critical interrelated aging conditions. Methods Minimum Data Set 3.0 was used. Physical frailty was measured by FRAIL-NH and cognitive impairment by Brief Interview for Mental Status and Cognitive Performance Scale. Demographic and clinical characteristics were adjusted accordingly. Aim 1 described the prevalence of physical frailty and cognitive impairment and longitudinally examined the association between two conditions with the non-proportional odds model. Aim 2 used latent class analysis to identify physical frailty subgroups and estimated their association with cognitive impairment using multinomial logistic regression. Aim 3 fitted group-based trajectory models to identify physical frailty trajectories and cognitive impairment trajectories and quantified the association between the two sets of trajectories. Main Results Around 60% of older residents were physically frail and 68% had moderate/severe cognitive impairment, with improvement and worsening observed in both conditions, particularly in the first three months. Older residents with moderate/severe cognitive impairment were consistently and increasingly more likely to be frail. Three physical frailty subgroups were identified at admission. Greater cognitive impairment was associated with higher odds to belong to “severe physical frailty”. Five physical frailty trajectories and three cognitive impairment trajectories were identified over the first six months. One in five older residents were “Consistently Frail” and “Consistently Severe Cognitive Impairment”. Conclusion Findings emphasized the need for care management tailored to the heterogeneous presentations and progression trajectories of physical frailty and cognitive impairment.

No description available.

With a higher number of nocturnal tornado events, I surveyed residents of Alabama, Arkansas, Kentucky, Mississippi, and Tennessee (N = 487 for each sample) to determine the time, in minutes, it took to reach a decision on shelter-seeking. I utilized latent class analysis (LCA) to create class memberships, based on diurnal and nocturnal scenarios, to associate with time intervals. Four actors were identified for each scenario: Tech Users, Typical Actors, Non-Reactors, and Social Actors for the day sample, Tech Users, Typical Actors, Passive Actors, and Non-Reactors for the night sample. Time intervals were created and applied to each class. All class assignments except one, Traditional Actors in the night sample, used more time than allotted in an average tornado warning lead time (~15 minutes). Future studies may be necessary to determine a reduction in time needed for decision-making, such as establishing the most impactful warning sources.

The subject of this thesis is the structure of society and its didactic elaboration. The first few chapters are focused on Czech curriculum documents and three chosen foreign curricula (England, New Zealand, Australia) and provides an analysis of how these documents deal with the subject of the structure of society. The results of these analyses are compared with each other afterwards. The next part of the thesis deals with EU Council recommendations for the development of key student competencies. These competencies are then compared to the approach to student competencies in the Czech curriculum. One chapter is focused on textbooks for upper secondary schools and their approach to the structure of society. This is followed up with a chapter about Czech sociological research going into this subject, the results of these studies are presented here. The last part of this thesis contains teaching suggestions for the education of the structure of society at upper secondary schools. The principles of pedagogical constructivism are explained first, followed up by two activities based on the results of the analyses of curriculum documents and the results of sociological research presented earlier. As emerged from the analyses, foreign curricula, more than ours, focus on the current state of society and...

In this dissertation I develop a Marxian class analysis of corn-producing family farms in the Midwestern United States during the early twentieth century. I theorize the family farm as a complex hybrid of mostly feudal and ancient class structures that has survived through a contradictory combination of strategies that includes the feudal exploitation of farm family members, the cannibalization of neighboring ancient farmers in a vicious hunt for superprofits, and the intervention of state welfare programs. The class-based definition of the family farm yields unique insights into three broad aspects of U.S. agricultural history. First, my analysis highlights the crucial, yet under-recognized role of farm women and children's unpaid labor in subsidizing the family farm. Second I offer a new, class-based perspective on the roots of the twentieth century "miracle of productivity" in U.S. agriculture, the rise of the agribusiness giants that depended on the perpetual, technology-induced crisis of that agriculture, and the implications of government farm programs. Third, this dissertation demonstrates how the unique set of contradictions and circumstances facing family farmers during the early twentieth century, including class exploitation, were connected to concern for their ability to serve the needs of U.S. industrial capitalist development. The argument presented here highlights the significant costs associated with the intensification of exploitation in the transition to industrial agriculture in the U.S. The family farm is implicated in this social theft. Ironically, the same family farm is often held up as the bedrock of American life. Its exalted status as an example of democracy, independence, self-sufficiency, and morality is enabled among other things by the absence of class awareness in U.S. society. When viewed through the lens of class, the hallowed family farm becomes example of one of the most exploitative institutions in the U.S. economy. The myth of its superiority takes on a new significance as one of the important non-economic processes helping to overdetermine the family farm's long survival, while participating in foreclosing truly radical transformations of these institutions to non-exploitative alternatives.

No description available.