Latent Class Approach Research Articles

Latent class approach to classify e-scooter non-users: A comparative study of Helsinki and Tokyo

Recent years have seen a rapid deployment of shared electric scooters (e-scooters), across hundreds of cities worldwide. Despite the global diffusion, most urban residents are still not using e-scooters, and they often bear the negative consequences of e- scooter deployment. This comparative study investigates the attitudes and characteristics of the e-scooter non-users in Helsinki, Finland and Tokyo, Japan using latent class clustering analysis. Our findings reveal five classes of e-scooter non-users, ranging from very negative perspectives to the least negative perspectives towards e-scooter deployment. Non-users also have a diverse set of socio-demographics background, including variation across age, gender, and income. In addition, non-users mostly do not feel the necessity of using an e-scooter as well as lack competences and have safety concerns. Safety concerns are dominant among those with more negative perceptions, while classes with the least negative perceptions and low income tend to not use e-scooters due to high rental costs. Understanding the different categories and perspectives of non-users, alongside with understanding users, has clear smart mobility policy and management implications. On the one hand, it should help with understanding the effectiveness of different interventions stemming from e-scooter operators and public policy actors. On the other hand, this study’s findings contribute to understanding public acceptability of e-scooter technology and associated policy and management actions. The study also opens further pathways for both comparative studies and for understanding different groups of users and non-users of smart mobility technologies.

Ideal Family Size and Reproductive Orientations: An Exploration of Change Over Time in the United States

Drawing on more than 30 years of nationally representative microdata from the General Social Survey, this article comprehensively updates recent trends in ideal family size in the United States. It first documents stability in ideal family sizes between the mid-1980s and 2018, even in the face of a recent fertility decline. Next, the study adopts a latent class approach that identifies typologies of “reproductive orientations,” defined as multidimensional mental models of reproduction encompassing ideal family size, attitudes toward reproductive labor, and views on reproduction contexts. The findings indicate three distinct classes of reproductive orientations: Progressive Familialists, Conservative Familialists, and Blended Egalitarians. Further analyses suggest that the prevalence of these classes has changed over time and that class membership is associated with distinct patterns of childbearing and marriage. These findings deepen contemporary understandings of ideal family size in the United States and have broader implications for how demographers conceptualize and measure fertility preferences across diverse contexts.

Instrumental variable estimation with observed and unobserved heterogeneity of the treatment and instrument effect: a latent class approach

Instrumental variable estimation with observed and unobserved heterogeneity of the treatment and instrument effect: a latent class approach

Certified Community Behavioral Health Clinic Services for Clients With Co-occurring Disorders: A Latent Class Approach.

Certified community behavioral health clinics (CCBHCs) are designed to provide comprehensive care for individuals with co-occurring mental and substance use disorders. The authors classified outpatient mental health treatment facilities on the basis of provision of services for clients with co-occurring disorders and assessed whether CCBHCs differed from other outpatient mental health facilities in services provided. The authors used latent class analysis to identify distinct services for clients with co-occurring disorders in 5,692 outpatient mental health facilities in the 2021 National Substance Use and Mental Health Services Survey. Nine indicators were included: treatment for clients with substance or alcohol use disorder co-occurring with serious mental illness or serious emotional disturbance, specialized programs or groups for such clients, medication-assisted treatment (MAT) for alcohol use disorder, MAT for opioid use disorder, detoxification, individual counseling, group counseling, case management, and 12-step groups. A multinomial logistic regression was used to estimate whether CCBHCs were associated with any identified classes after analyses controlled for facility characteristics. A four-class solution provided a model with the best fit, comprising comprehensive services (23.4%), case management services (17.7%), counseling and self-help services (58.6%), and professional services (4.3%). Regressing class membership on facility type and covariates, the authors found that compared with community mental health clinics (CMHCs), CCBHCs were more likely to belong to the comprehensive services class than to the case management services, counseling and self-help services, and professional services classes. CCBHCs were more likely than other outpatient programs to offer comprehensive care, and CCBHC status of a CMHC facilitated enhanced service provisions.

Exploring preferences for contractual terms in a scenario of ecological transition for the agri-food sector: a latent class approach

Governance mechanisms along the agri-food supply chains are increasingly important in a scenario of ecological transition. Under the conceptual and analytical lens of the Transaction Cost Economics, we explored farmers’ preferences towards a variety of clauses usually adopted in production contracts. To this purpose, a discrete choice experiment among 190 durum wheat producers in Italy was conducted. Results from a latent class model showed that producers were mainly interested in fixed prices formula and to join shared rules of production but revealed little or no interest for compelling sustainable cultivation techniques and the provision of technical assistance. However, these preferences are heterogeneous across farmers and vary depending on their level of education and previous use of contractual arrangements, with relevant implications for contract design and management. Governance mechanisms along the agri-food supply chains are increasingly important in a scenario of ecological transition. Under the conceptual and analytical lens of the Neo Institutional Economics, we explored farmers’ preferences towards a variety of clauses usually adopted in production contracts. To this purpose, a discrete choice experiment among 190 durum wheat producers in Italy was conducted. Results from a latent class model showed that that producers were mainly interested in fixed prices formula and to join shared rules of production but revealed little or no interest for compelling sustainable cultivation techniques and the provision of technical assistance. However, these preferences are heterogenous across farmers and vary depending on their level of education and previous use of contractual arrangements, with relevant implications for contract design and management.

Predictors of substance use disorder treatment and mutual support group participation among Black women across the criminal legal spectrum: A latent class approach

BackgroundThe purpose of this study includes: 1) identifying classes of substance-related needs among Black women, and 2) examining the effect of substance-related need classes and culturally-relevant factors on Black women’s use of substance use disorder (SUD) treatment and mutual support groups. MethodsAs part of a longitudinal cohort study, Black women were recruited in prison nearing release, on probation, and in the community without involvement in the criminal legal system (CLS, n=565) and followed-up at 18-months. We conducted a baseline latent class analysis of substance-related needs among Black women. Logistic regression models adjusted for culturally-relevant factors to predict the use of treatment and frequency of mutual support group participation over 18-months among Black women who use drugs. ResultsFour classes by level of needs were found: low, daily marijuana use, high mental health, and high comorbidity. During the 18-month follow-up, women characterized by the high comorbidity need class and with higher scores of religious well-being were more likely to frequently participate in mutual support groups. Non-CLS-involved women were less likely to engage with both treatment and mutual support groups than women from the prison sample at 18-months. ConclusionsThis study highlights four distinct classes of substance-related needs among Black women, highlighting the complex patterns of behavior and within-racial group differences among Black women. Black women with high comorbidity needs were more likely to participate in mutual support groups, but the latent classes did not predict SUD treatment indicating other non-medical and social contextual need factors may be at play.

Trajectories of relationship and sexual satisfaction over 2 years in the Covid-19 pandemic: A latent class analysis.

Previous research on the impact of the Covid-19 pandemic on romantic relationships has mainly concentrated on short-term effects and average trends of change. This study aimed to explore different trajectories of relationship and sexual satisfaction from April 2020 to March 2022. Including a cross-national sample of 2859 individuals, a latent class approach was applied to identify subgroups of sexual and relationship satisfaction trajectories. Participants' satisfaction levels, attachment style, psychological symptoms, life satisfaction, and living arrangements were taken into account. Three latent classes were identified for both relationship and sexual satisfaction: (1) a group with high satisfaction levels but a slightly declining trajectory, (2) a group with low satisfaction levels but an ascending trajectory, and (3) a fluctuating group. Living apart, having children, psychological symptoms, low life satisfaction, and avoidant attachment characterized the two groups with low or fluctuating trajectories. The findings suggest that there were different trajectories of relationship and sexual satisfaction during the pandemic, which can be predicted by individual dispositions. Even 2 years after the pandemic, its effects had not disappeared.

The Association Between Adverse Childhood Experiences and Preterm Delivery: A Latent Class Approach.

Background: Racial and socioeconomic status (SES) disparities in preterm delivery (PTD) have existed in the United States for decades. Disproportionate maternal exposures to adverse childhood experiences (ACEs) may increase the risk for adverse birth outcomes. Moreover, racial and SES disparities exist in the prevalence of ACEs, underscoring the need for research that examines whether ACEs contribute to racial and SES disparities in PTD. Methods: We examined the relationship between ACEs and PTD in a longitudinal sample of N = 3,884 women from the National Longitudinal Study of Adolescent to Adult Health (1994-2018). We applied latent class analysis to (1) identify subgroups of women characterized by patterns of ACE occurrence; (2) estimate the association between latent class membership (LCM) and PTD, and (3) examine whether race and SES influence LCM or the association between LCM and PTD. Results: Two latent classes were identified, with women in the high ACEs class characterized by a higher probability of emotional abuse, physical abuse, sexual abuse, and foster care placement compared with the low ACEs class, but neither class was associated with PTD. Race and SES did not predict LCM. Conclusions: Our findings suggest that ACEs may not impact PTD risk in previously hypothesized ways. Future research should assess the impact of ACEs on the probability of having live birth pregnancies as well as the role of potential protective factors in mitigating the impact of ACEs on PTD.

Profiles of Death Penalty Attitudes: A Latent Class Approach

ABSTRACT Objectives This study examines the heterogeneity in the demographic, normative, and experiential features of those who support, oppose, and are undecided on capital punishment to determine the complexity of characteristics that exist beyond the “averages.” This study also evaluates macro-level changes in death penalty supporter, opposer, and undecided sub-types in the U.S. over time. Methods Drawing upon national trend data from the General Social Survey (1972–2016), a series of Latent Class Analyses (LCA) are employed to identify the number, size, and compositional profile of emergent subtypes of supporters, opposers, and undecideds on the death penalty. We also examine variation in the size and relative distribution of these latent classes over five decades. Results Substantial heterogeneity among death penalty supporters and opposers was found, yielding nine and seven distinct sub-types of each, respectively. Two unique sub-types of undecideds were also uncovered. Conclusion Variation in the composition and size of these latent classes over time provides important insights into the variation in death penalty support over the past 50 years. This is the first study of which we are aware to employ an LCA to national trend data of public opinion toward the death penalty across multiple decades in the United States.

Comparing youth and adult homicide victimization in Finland 2002–2018: A latent class approach

Comparing youth and adult homicide victimization in Finland 2002–2018: A latent class approach

Abstract C012: Integration of social variables resolves subclasses of tumor microenvironment

Abstract Intro: Recent work has identified subclasses of the tumor microenvironment, with most work emphasizing a robust adaptive immune response associated with ER-negative and/or basal-like tumors. Since adaptive immune responses may influence response to chemotherapy and may be targetable by immunotherapy, understanding population-level heterogeneity in immune response is important. Methods: We leveraged the Carolina Breast Cancer Study (CBCS), a large population-based study that oversampled young (≤50 years) and Black women with invasive breast cancer (BC). We performed NanoString gene expression profiling on tissue from 1,329 BC patients, including 704 (53%) Black and 625 (47%) non-Black women, and assayed genes representative of at least 10 individual immune cell types from both adaptive and innate arms of the immune system, homologous recombination deficiency signature and TP53 mutation status. Using a latent class approach that included variables for self-reported race, triple negative status, individual measures of socioeconomic status and community-level measures of advantage and disadvantage, we identified latent classes that integrated tumor microenvironment with tumor biology, individual and community level exposures under a Cells-to-Society conceptual model of breast cancer outcomes. Results: Latent Class Analysis identified 6 integrated classes of the tumor microenvironment, further subdividing a previously identified and validated adaptive immune class into two subgroups. The two adaptive classes were distinguished by different expression levels of B-cells, CD8+ T cells, cytotoxic cells and PD-L1, and differing frequency of basal-like subtype (Adaptive Class 1 RFD [95% CI]: 19.0 [11.9, 26.2]; Adaptive Class 2 RFD [95% CI]: 32.8 [23.3,42.3]) and presence of genomic instability gene expression subtypes (Adaptive Class-1 RFD [95% CI]: 22.1 [14.0, 30.1]; Adaptive Class-2 RFD [95% CI]: 37.5 [28.1,46.2]) relative to the immune-quiet tumors. Discussion: Social variables have complex effects on tumor biology and microenvironment, but latent class models and other model-based approaches may help to integrate biological data with other social factors to identify susceptible subgroups and to optimize precision medicine in diverse populations. Citation Format: Alina M. Hamilton, Kara McCormack, Marc A. Emerson, Melissa A. Troester, Terry Hyslop. Integration of social variables resolves subclasses of tumor microenvironment [abstract]. In: Proceedings of the 16th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2023 Sep 29-Oct 2;Orlando, FL. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2023;32(12 Suppl):Abstract nr C012.

Modeling package delivery acceptance in Crowdshipping systems by Public Transportation Passengers: A latent class approach

The subject of this research is to model the probability of accepting package delivery tasks in Crowd-Shipping (CS) systems focusing on shipping packages by Public Transportation (PT) passengers. Real-world implementation of such a PT-based CS initiative requires a proper grasp of the behavior of crowd shippers, package receivers, and senders, among which crowdshippers are at the center of attention. Given survey data collected from 2,208PT passengers in the Sydney metropolitan area, this study constructs a model for package delivery task acceptance among PT passengers who express a willingness to participate as crowdshippers. A Latent Class (LC) choice model is developed and estimated to explore different classes of participant characteristics and preferences for accepting CS delivery tasks, under different levels of offered incentive, package weight, and required detour distance at the destination. As a key finding, the model identifies three classes, namely “leisurely”, “avid”, and “skeptical” users with class sizes of 19%, 53%, and 28%, respectively. Leisurely users tend to be majorly employed with relatively higher ages and lower PT trip frequency per month. This class shows a potential desire to take CS tasks as long as they are not logistically burdensome. Avid users, forming the largest class of the population, are mostly younger PT passengers who are less sensitive to task difficulty and might accept even CS tasks with minimum incentives (around AU$3). Skeptical users, displaying high PT trip frequency per month and high income, are unwilling to accept the CS tasks unless get compensated well.

Estimating Costs Associated with Disease Model States Using Generalized Linear Models: A Tutorial.

Estimates of costs associated with disease states are required to inform decision analytic disease models to evaluate interventions that modify disease trajectory. Increasingly, decision analytic models are developed using patient-level data with a focus on heterogeneity between patients, and there is a demand for costs informing such models to reflect individual patient costs. Statistical models of health care costs need to recognize the specific features of costs data which typically include a large number of zero observations for non-users, and a skewed and heavy right-hand tailed distribution due to a small number of heavy healthcare users. Different methods are available for modelling costs, such as generalized linear models (GLMs), extended estimating equations and latent class approaches. While there are tutorials addressing approaches to decision modelling, there is no practical guidance on the cost estimation to inform such models. Therefore, this tutorial aims to provide a general guidance on estimating healthcare costs associated with disease states in decision analytic models. Specifically, we present a step-by-step guide to how individual participant data can be used to estimate costs over discrete periods for participants with particular characteristics, based on the GLM framework. We focus on the practical aspects of cost modelling from the conceptualization of the research question to the derivation of costs for an individual in particular disease states. We provide a practical example with step-by-step R code illustrating the process of modelling the hospital costs associated with disease states for a cardiovascular disease model.

Do bipolar disorder symptom profiles matter for suicide risk? A latent class approach to investigating differences in suicidal desire and acquired capability.

Depressive and mixed symptoms in bipolar disorder (BD) have been linked to higher suicide risk. Based on Klonsky and May's three-step theory and Joiner's Interpersonal Psychological Theory of Suicide, we hypothesized that patients diagnosed with BD who reported severe levels of depressive symptoms and mixed depressive and manic symptoms would also report higher levels of suicidal desire and acquired capability of suicide, as well as suicidal thoughts and behaviors. The sample included 177 outpatients diagnosed with BD. Latent class analysis was conducted to replicate the identified groups of a previously conducted study using a smaller but overlapping dataset. Between-class and pairwise analyses with measures of suicidal desire and acquired capability were conducted. As expected, the classes characterized by severe depressive symptoms and mixed symptoms reported higher levels of suicidal desire. However, the results regarding acquired capability were less consistent. Given the overall elevated suicide risk of BD and the consistent relationship between depressive symptoms and other strong correlates of suicide, clinicians who work with patients diagnosed with BD should closely monitor changes in their depressive symptoms.

A Longitudinal Study of Plasma Glycated Albumin across Pregnancy and Associations with Maternal Characteristics and Cardiometabolic Biomarkers.

Glycated albumin (GA) has recently been proposed as a screening marker for diabetes among non-pregnant individuals. However, data on GA during pregnancy are sparse and lacking among women of diverse race/ethnicity. We investigated longitudinal concentrations of GA among multiracial pregnant women in the National Institute of Child Health and Human Development (NICHD) Fetal Growth Studies-Singletons. We quantified GA and cardiometabolic biomarkers using longitudinal plasma samples collected at 10 to 14, 15 to 26 (fasting), 23 to 31, and 33 to 39 gestational weeks from 214 pregnant women without gestational diabetes. We examined the distribution of GA across pregnancy and its association with participants' characteristics including race/ethnicity, pre-pregnancy body mass index (ppBMI), and selected cardiometabolic biomarkers. GA trajectories were estimated using a latent class approach. Medians (interquartile range) of GA concentrations were 12.1% (10.6%-13.4%), 12.5% (10.7%-13.8%), 12.4% (10.9%-13.5%), and 11.5% (10.4%-12.5%) at 10 to 14, 15 to 26, 23 to 31, and 33 to 39 weeks, respectively. There were no significant differences in the pattern among different race/ethnic groups (P > 0.53). A minority of women exhibited a GA trajectory characterized by a high concentration of GA at 15 to 26 weeks. GA concentrations were inversely related to ppBMI and plasma low-density lipoprotein and triglyceride concentrations, but were not significantly related to hemoglobin A1c, fasting insulin, or glucose over pregnancy. In this study of individuals who were normoglycemic before pregnancy, plasma GA concentrations stayed relatively constant over pregnancy, decreasing only in late pregnancy. GA concentrations were inversely related to ppBMI and suboptimal lipid profiles, but did not appear to be a sensitive marker for glucose metabolism in pregnancy.

Modelling the composition of household portfolios: A latent class approach

AbstractWe explore portfolio allocation in Great Britain by introducing a latent class modelling approach using household panel data based on a nationally representative sample of the population, namely the Wealth and Assets Survey. The latent class aspect of the model splits households into four groups, from lowest‐wealth and least‐diversified through to highest‐wealth and most‐diversified, which serves to unveil a more detailed picture of the determinants of portfolio diversification than existing econometric approaches. A pattern of class heterogeneity is revealed that conventional econometric models are unable to identify because the statistical significance and the direction of the effect of some explanatory variables vary across the groups. For example, the effect of labour income on the number of financial assets held influences the level of diversification for the two middle classes, whereas no effect is found for households with the lowest or the highest levels of diversification. Noticeable differences in the magnitude of the effects of pension wealth and occupation are also revealed across the four classes. Such findings demonstrate the importance of accounting for latent heterogeneity when modelling financial behaviour. Ultimately, treating the population as a single homogeneous group may lead to biased parameter estimates, whereby policy based on such models could be inappropriate or erroneous.

Customer preferences heterogeneity toward avocado: a latent class approach based on the best–worst scaling choice modeling

Consumption of the avocado fruit and its availability in the retail market has increased in recent decades and with it the desire to learn more about the market and consumer choices. This research aims to explore the consumers' preference heterogeneity toward avocado fruit in Italy assessing their personal eating orientation and socio-demographic factors. To achieve this purpose, the answers of 817 Italian consumer of avocado were collected using a structured questionnaire shared online at national level. A survey based on the best–worst method was conducted to assess the declared preferences of individuals toward a set of intrinsic, extrinsic and credence attributes of avocado, as well as a latent class analysis of subject preferences indexes was applied to identify different clusters of individuals. The distribution of preferences indexes was investigated among latent classes, which were regressed on the socio-demographic variables covariates using multinomial logistic regression. The results suggest that the desirable attributes for choosing avocado were taste, consistency (ready to eat) and affordable price. In addition, five consumer profiles in the Italian context were defined (named Hedonistic, Avocado fruit lovers, Sustainability sensitive, Value for money and Health sensitive) and characterized in terms of preferences and socio-demographic features. The study addresses a topic that has already been explored, but with an unexploited approach (latent class analysis combined with best–worst choice modeling). Therefore, the results help fill the gap in the existing literature by enriching it with a study that characterizes avocado consumers' preferences considering their heterogeneity in terms of perception and socio-demographic characteristics.

Students' Academic Habitus and Its Relation to Family Capital: A Latent Class Approach to Inequalities among Secondary School Students1

Habitus plays a crucial part in Bourdieu's theory of sociocultural reproduction for understanding the persistence of inequalities in the education system. According to Bourdieu, students from homes that are heavily equipped with cultural capital develop a specific kind of habitus, that is, modes of perceiving, thinking, and acting, remarkably well‐adjusted to the expectations of teachers and educational institutions. However, research has rarely tried to measure what we refer to as students' academic habitus to highlight the different types of habitus that students might express toward school. Drawing on data from secondary students in Luxembourg, we employ a latent class approach to operationalize, measure, and explore students' academic habitus. Our investigation comprises three main steps: First, we develop a multifaceted understanding of students' habitus integrating diverse dispositions toward school and learning; second, we identify different academic habitus types: the habitus of excellence, the habitus of goodwill and loyalty, and the habitus of disengagement. Third, we examine how the three habitus types relate to different axes of inequality: socioeconomic status, cultural capital, family employment, gender, and immigrant background. Our typology of habitus bridges the qualitative literature on habitus with existing quantitative operationalizations. The findings show that students with a habitus of excellence are likely to hail from families with favorable parental employment and high cultural involvement.

Socioeconomic, demographic and geographic determinants of food consumption in Mexico.

What people eat affects public health and human wellbeing, agricultural production, and environmental sustainability. This paper explores the heterogeneity of food consumption patterns in an ecologically and culturally diverse country. Using a latent class approach (which creates clusters of individuals with homogeneous characteristics), we analyse a food questionnaire (from the National Health and Nutrition Survey) applied across Mexico. We identify four clusters of food consumption (staple, prudent, high meat and low fruit) and find that belonging to these clusters is determined by socioeconomic, demographic (age, sex) and geographic (region, urban/rural) characteristics. Maize and pulses tend to constitute a larger proportion of the diet of poor, rural populations living in the south, while urban populations eat more varied foods, including ingredients whose production systems tend to exert more pressure on natural resources (for instance, meat). Despite the importance given in the literature to the Mexican gastronomy and its diverse traditional regional diets, we find that only 6% of the population adopts a food consumption pattern resembling the traditional Mexican diet. Instead, most of the Mexican population has a food consumption pattern resembling a western diet, which is problematic in terms of public health and environmental sustainability.

An Analysis of the DIF Sources of ABIDE Mathematics Self-Efficacy Scale by means of a Latent Class Approach

This study aims to identify the sources of Differential Item Functioning (DIF) using the Mixture Ordinal Logistic Regression (Mixture OLR) method, a contemporary approach for detecting DIF. To analyze mathematics self-efficacy, data from a scale comprising 9 items were obtained from 5000 8th-grade students as part of the ABIDE-2016 project. The study compared the presence and extent of DIF by gender using two methods and examined the sources of DIF for items displaying DIF with Mixture OLR. The OLR analysis revealed that five items exhibited DIF at level A, but no DIF was observed with Mixture OLR. Furthermore, it was found that the magnitude of DIF (B) for an item showing DIF at level A changed due to Mixture OLR. The results indicate that the homogeneity of the data affects both the number of items displaying DIF and the magnitude of DIF. Three items did not exhibit significant DIF according to both methods. One significant finding in the study highlights the moderating effect of latent class on item 8, where DIF was observed. However, the source of DIF was not related to gender but rather stemmed from different ecological variables. An analysis of latent class characteristics revealed that students with significant DIF effects had lower absenteeism and fewer siblings. Additionally, students in this class had greater access to books at home and participated in more out-of-school mathematics courses. Surprisingly, these students were found to engage less in social activities. Various factors can influence how students respond to test items, potentially leading to DIF. These factors may include cultural background, gender, social environment, school, teacher, family interest/attitude toward the child, and home climate. Therefore, when developing and administering tests, it is crucial to test for data homogeneity and consider the impact of these variables, in addition to gender, to identify any sources of DIF in test items.