PROC LCA: A SAS Procedure for Latent Class Analysis

Latent class analysis (LCA) is a statistical method used to identify a set of discrete, mutually exclusive latent classes of individuals based on their responses to a set of observed categorical variables. In multiple-group LCA, both the measurement part and structural part of the model can vary across groups, and measurement invariance across groups can be empirically tested. LCA with covariates extends the model to include predictors of class membership. In this article, we introduce PROC LCA, a new SAS procedure for conducting LCA, multiple-group LCA, and LCA with covariates. The procedure is demonstrated using data on alcohol use behavior in a national sample of high school seniors.     

The influence of self-efficacy beliefs for student parents attending university

Student parents (i.e. students who have their own dependent children) are a specific subpopulation of adult learners. This study investigated the impact of self-efficacy beliefs on student parents’ perceived capacity to manage multiple roles and their satisfaction with family, school and life. Survey data collected from 398 student parents enroled at four Canadian universities were analysed. Latent variable analysis was conducted using maximum likelihood estimation with robust standard errors using Mplus. Self-efficacy beliefs were found to influence student parents’ perceptions of satisfaction at school, in the family and with life in general. Perceptions of one’s capacity to manage multiple roles (i.e. school–family balance) were found to mediate the relationship between academic self-efficacy and school satisfaction as well as parental self-efficacy and family satisfaction. Furthermore, preliminary evidence is provided of unique subgroups within the student parent population based on children’s ages, partner status and enrolment status (i.e. full/part-time studies).     

Performance of the Grade of Membership Model Under a Variety of Sample Sizes,Group Size Ratios,and Differential Group Response Probabilities for Dichotomous Indicators

Social scientists are frequently interested in identifying latent subgroups within the population, based on a set of observed variables. One of the more common tools for this purpose is latent class analysis (LCA), which models a scenario involving k finite and mutually exclusive classes within the population. An alternative approach to this problem is presented by the grade of membership (GoM) model, in which individuals are assumed to have partial membership in multiple population subgroups. In this respect, it differs from the hard groupings associated with LCA. The current Monte Carlo simulation study extended on prior work on the GoM by investigating its ability to recover underlying subgroups in the population for a variety of sample sizes, latent group size ratios, and differing group response profiles. In addition, this study compared the performance of GoM with that of LCA. Results demonstrated that when the underlying process conforms to the GoM model form, the GoM approach yielded more accurate classification results than did LCA. In addition, it was found that the GoM modeling paradigm yielded accurate results for samples as small as 200, even when latent subgroups were very unequal in size. Implications for practice were discussed.     

What do we know about child abuse and neglect patterns of co-occurrence? A systematic review of profiling studies and recommendations for future research

Latent class (LCA) and latent profile (LPA) analysis represent methodological approaches to identify subgroups of maltreated individuals. Although research examining child abuse and neglect (CAN) profiles is still rare, the application of person-centered techniques to clarify CAN types co-occurrence has substantially increased in recent years. Therefore, the aim of the present study was to provide a summary and critical evaluation of the findings of LCA/LPA child maltreatment research to: (a) systemize the current understanding of patterns of maltreatment across populations and (b) elucidate interactive effects of CAN types on psychosocial functioning. A search in PsychInfo, Eric, PubMed, Scopus, and Science Direct, and Google Scholar was performed. Sixteen studies examining the co-occurrence between child physical abuse, emotional abuse, sexual abuse, neglect, and/or exposure to domestic violence were identified. A critical review of the studies revealed inconsistent findings as to the number of CAN classes, but most research uncovered a poly-victimized and a low abuse group. Further, multiple victimization was associated with most adverse internalizing and externalizing outcomes, especially when sexual abuse was present. Exposure to physical and emotional abuse was frequently reported to lead to behavioural problems. Based on the present study results, we provide a set of recommendations for surpassing the current methodological and conceptual limitations in future research.     

Why Tenth Graders Fail to Finish High School: A Dropout Typology Latent Class Analysis

A large percentage of the students who drop out of K–12 schools in the United States do so at the end of high school, at some point after grade 10. Yet little is known about the differences between types of students who drop out near the end of high school. The purpose of this study is to examine a typology of high school dropouts from a large nationally representative dataset (ELS:2002) using latent class analysis (LCA). We found three significantly different types of dropouts; Quiet, Jaded, and Involved. Based on this typology of three subgroups, we discuss implications for future dropout intervention research, policy, and practice.     

The role of student characteristics in studying micro teaching–learning environments

This study explored the role of student characteristics in studying micro teaching–learning environments. The overarching hypothesis is that teachers teach differently to micro environments in their classrooms. This study is the first of a series exploring the following four questions: (1) What student profiles are identified at the beginning of a school year with respect to cognitive and motivational-affective factors? (2) How do students with different profiles perceive conditions in their learning environment? (3) To what degree do classrooms differ in the composition of student profiles? (4) What are possible consequences for examining micro teaching–learning environments? The study investigated 82 randomly selected high school science classrooms. Student characteristics were assessed at the beginning of the school year. After a video taped teaching unit, students were asked to rate the degree to which they experienced learning conditions as supportive. Latent class analysis (LCA) showed five distinct student profiles that varied along cognitive and motivational-affective dimensions. Multilevel analyses showed effects of student profiles assessed at the beginning of the school year on the students’ perception of learning conditions in a teaching unit 4 months later. To illustrate consequences for examining micro teaching–learning environments, student profiles were linked to video examples. The examples point to the special value of LCA in studying micro teaching–learning environments: they make it possible to focus on the individual student and to investigate the interplay of student characteristics and the learning environment.     

Prevalence,stability, and functionality of achievement goal profiles in mathematics from third to seventh grade

This study examined the longitudinal trajectories of achievement goal profiles in mathematics from third to seventh grade in a sample of 302 German students. Latent profile analyses were conducted separately for each school year and revealed three subgroups of students with distinct goal profiles labeled high multiple goals, moderate multiple goals, and primarily mastery-oriented. Only about one third of the students held the same goal profile across all school years. The amount of students pursuing moderate multiple goals increased over time, which supplements previous findings and theorizing about an ongoing differentiation of achievement goals during early adolescence. There were remarkably few differences in educational outcomes (interest, effort, achievement) between students from distinct goal profile groups. However, high multiple goal students showed the lowest test scores in sixth and seventh grade. Moreover, if students showed low performance during one school year, they were more likely to adopt a high multiple goals profile in the following year. Results are discussed in relation to individual cognitive developments, changes in school environments, and special characteristics of educational systems.     

Conducting Confirmatory Latent Class Analysis Using Mplus

Latent class analysis (LCA) is an increasingly popular tool that researchers can use to identify latent groups in the population underlying a sample of responses to categorical observed variables. LCA is most commonly used in an exploratory fashion whereby no parameters are specified a priori. Although this exploratory approach is reasonable when very little prior research has been conducted in the area under study, it can be very limiting when much is already known about the variables and population. Confirmatory latent class analysis (CLCA) provides researchers with a tool for modeling and testing specific hypotheses about response patterns in the observed variables. CLCA is based on placing specific constraints on the parameters to reflect these hypotheses. The popular and easy-to-use latent variable modeling software package Mplus can be used to conduct a variety of CLCA types using these parameter constraints. This article focuses on the basic principles underlying the use of CLCA, and the Mplus programming code necessary for carrying it out.     

Exploring patterns of Latino/a children's school readiness at kindergarten entry and their relations with Grade 2 achievement

This study contributed to the school readiness literature by taking an intrachild perspective that examined the relations between Latino/a children's school readiness profiles and later academic achievement. Teachers rated the school readiness of 781 Latino/a kindergartners during the first month of school using the Kindergarten Student Entrance Profile (KSEP). Latent class analysis (LCA) examined KSEP profiles across social-emotional, physical, and cognitive domains and identified five distinct school readiness classes that described students’ strengths and weaknesses at kindergarten entry. Among the predictors examined, gender was the only significant difference among the top two readiness classes, with girls less likely to be in the lower of these two classes (OR = 0.38). In addition, children in the bottom three readiness classes were significantly less likely than students in the top readiness class to have preschool experience (ORs = 0.02–0.19) and had significantly lower levels of English proficiency (ORs = 0.51–0.72). Class membership was significantly associated with scores on the Grade 2 California Standards Tests and only the top two readiness classes had reading fluency rates near or above a national benchmark at the end of Grade 2. The variation of early achievement found across readiness classes also showed that considering the pattern of a child's social-emotional and cognitive readiness might enhance school readiness assessment. Implications for integrating universal school readiness screeners within a comprehensive multigating assessment model are also discussed.     

Integrating the light and dark sides of student engagement using person-oriented and situation-specific approaches

This study contributes to the research on student engagement in three ways: 1) by combining questionnaire and situational measures of engagement using the Experience Sampling Method (ESM), 2) by applying a demands-resources model to describe the positive and negative aspects of student engagement, and 3) by adopting a person-oriented approach to describe subgroups of students with different profiles of engagement and burnout symptoms. Two studies were conducted: sample one comprised 255 US high school students (45.5% female, 9th – 12th grade), and sample two 188 Finnish comprehensive and high school students (59.6% female, 9th to 10th grade). Latent profile analyses (LPA) of person-level measures of schoolwork engagement and burnout in the US and Finland revealed four profiles: 1) engaged, 2) engaged-exhausted, 3) moderately burned out (risk for burnout) and 4) burned out. These four groups were identified in both samples, but differed in their prevalence. The groups differed in their state experiences of situational demands, resources and engagement at school. Engagement is not wholly an experience of ‘flourishing’: some students experienced elevated levels of both engagement and burnout. Thus, positive and negative aspects of engagement should be studied and modeled together.     

Impacts of enquiry-based science teaching on achievement gap between high-and-low SES students: findings from PISA 2015

The purpose of this study was to empirically investigate, in a large-scale assessment, the prevalence of enquiry-based teaching, as defined by PISA, implemented in science lessons, and the extent to which it is related to students from different socioeconomic status (SES) family backgrounds in the U.S. The Programme of International Student Assessment (PISA) 2015 data including 5,146 students in the U.S. were used. A Latent Profile Analysis (LPA) was conducted to classify students into different science teaching subgroups. Mixture regression analysis was conducted to examine whether there the association between SES and achievement varied across science teaching subgroups. Four science teaching subgroups were identified, varying in the frequency of and type of teaching practices used. Students’ attitudes, gender, science instruction time, and reading achievement are significant predictors of the subgroup membership. SES-achievement gap also varied across the science teaching subgroups. Discussions for the results and suggestions for future study were presented.     

Identifying maltreatment subgroups with patterns of maltreatment subtype and chronicity: A latent class analysis approach

Maltreatment experiences are complex, and it is difficult to characterize the heterogeneity in types of maltreatment. Subtypes, such as emotional maltreatment, sexual abuse, physical abuse, and neglect commonly co-occur and may persist across development. Therefore, treating individual maltreatment subtypes as independently occurring is not representative of the nature of maltreatment as it occurs in children’s lives. Latent class analysis (LCA) is employed herein to identify subgroups of maltreated children based on commonalities in maltreatment subtype and chronicity. In a sample of 674 low-income urban children, 51.6% of whom experienced officially documented maltreatment, our analyses identified four classes of children, with three distinct classes based on maltreatment subtypes and chronicity, and one group of children who did not experience maltreatment. The largest class of maltreated children identified was the chronic, multi-subtype maltreatment class (57% of maltreated children); a second class was characterized by only neglect in a single developmental period (31% of maltreated children), and the smallest class was characterized by a single subtype of maltreatment (emotional maltreatment, physical, or sexual abuse) occurring in a single developmental period (12% of maltreated children). Characterization of these groups confirms the overlapping nature of maltreatment subtypes. There were notable differences between latent classes on child behavioral and socio-emotional outcomes measured by child self-report and camp counselors report during a one-week summer camp. The largest differences were between the non-maltreated class and the chronic maltreatment class. Children who experienced chronic, multi-subtype maltreatment showed higher levels of externalizing behavior, emotion dysregulation, depression, and anxiety.     

Examining the Validity of GOLD® With 4-Year-Old Dual Language Learners

Research Findings: This study builds on prior work related to the assessment of young dual language learners (DLLs). The purposes of the study were to (a) determine whether latent subgroups of preschool DLLs would replicate those found previously and (b) examine the validity of GOLD^® by Teaching Strategies with empirically derived subgroups. Latent class analysis confirmed previous findings of 3 distinct latent subgroups of DLLs (bilingual children, emergent bilingual children, and heritage language speakers). Results of differential item functioning analysis showed that with few exceptions, GOLD items functioned similarly, which indicates that groups matched on ability were similar in their item scores. The item pertaining to using conventional grammar consistently favored non-DLLs over heritage language speakers. The item pertaining to name writing consistently favored DLLs as a single group, emergent bilingual children, and heritage language speakers. Practice or Policy: Study results provide further support for the heterogeneity of DLLs and the use of GOLD with DLL subgroups. This provides the field with an opportunity to better understand this special population of children and enables teachers to plan with greater precision experiences that contribute to their development and learning.     

CORRELATES OF CONDUCT PROBLEMS AND DEPRESSION COMORBIDITY IN ELEMENTARY SCHOOL BOYS AND GIRLS RECEIVING SPECIAL EDUCATIONAL SERVICES

There is limited empirical research on the correlates of conduct problems (CP) and depression comorbidity during childhood. This study investigated 479 elementary school children (48.2% girls). It compared children with comorbidity to children with CP only, depression only, and control children on individual, academic, social, and family characteristics. The study also analyzed gender differences in the associations between correlates and comorbidity. Multinomial logistic regression results revealed that children with CP and depression had higher levels of anxiety and more school difficulties than children with CP only, more social difficulties and more severe attention deficit and hyperactivity disorder (ADHD) symptoms than children with depression only, and more difficulties in all domains than control children. Girls with CP and depression presented a particularly negative profile, including lower school abilities than CP and control girls, and lower social skills and more severe ADHD symptoms than control girls.  Implications for evaluation and intervention planning are discussed.     

A TYPOLOGY OF CHILD SCHOOL BEHAVIOR: INVESTIGATION USING LATENT PROFILE ANALYSIS AND CLUSTER ANALYSIS

To describe and facilitate the identification of child school behavior patterns, we developed a typology of child school behavior (ages 6–11 years) using the norming data (N = 2,338) for the second edition of the Behavior Assessment System for Children Teacher Rating–Child form). Latent profile analysis was conducted with the entire data set, whereas cluster analysis was conducted within each latent profile. The resulting classification system included three latent profiles: (a) Adequate Adjustment (AA), (b) Mild Adjustment Difficulties (MAD), and (c) Functionally Impaired Adjustment (FIA). The AA latent profile included the (a) Well‐Adapted, (b) Average, and (c) Worry clusters. The clusters identified within the MAD latent profile were (a) Academic Problems, (b) Physical Complaints, (c) Disruptive Behavior Problems, and (d) Internalizing Problems. Finally, the FIA latent profile comprised the (a) Clinical Problems–External and (b) Clinical Problems–Internal clusters.     

Numerical Minority Status in Middle School and Racial/Ethnic Segregation in Academic Classes

This study examined the effects of racial/ethnic segregation (i.e., overrepresentation) in academic classes on belonging, fairness, intergroup attitudes, and achievement across middle school (n = 4,361; M_ageT1 = 11.33 years), and whether effects depended on numerical minority status in school and race/ethnicity. Latent growth curve models revealed that experiencing more segregation than usual predicted less belonging and fairness than usual for all youth in the numerical minority, and greater in-group preference for numerical minority Whites. Academic classroom segregation throughout middle school predicted less steep declines in in-group preference for adolescents in the numerical minority, and declines in achievement for African American numerical minority youth. Results highlight the need to treat the racial/ethnic context as a structural and dynamic construct.     

Inattention,hyperactivity and low parental education in children with conduct problems and poor reading skills

Concerns over the poor reading performance of children with conduct problems are often expressed by their parents and teachers. In order to intervene effectively with these children, it is necessary to know what their difficulties are, both at school and at home. To date, there is still a lot of ambiguity in relation to the specific difficulties children with conduct problems and coexisting poor reading experience in important academic domains and in the family context. The aim of the present study was to compare the problems of attention and hyperactivity in the classroom and parental education in four groups of children: conduct problems only (CP); conduct problems and coexisting poor reading (CP‐PR); poor reading only (PR); and a comparison group (COM). A sample of 2nd Grade Greek children (n = 113) was selected on the basis of the Conners' Teacher Rating Scale‐28 (CTRS‐28), Conduct Problems Scale and the Reading Ability Detection Test. The CTRS‐28, Inattention and Hyperactivity scales and a brief questionnaire on parental education were used to assess attention, hyperactivity and level of parental education, respectively. The results showed that children with CP‐PR were significantly more inattentive and hyperactive than the other groups of children. Their parents were also more likely to have spent fewer years in school than the parents of CP only and COM children. The earlier findings have implications for intervention planning and theoretical understanding of childhood conduct problems.     

Is there a Complex Trauma Experience typology for Australians experiencing extreme social disadvantage and low housing stability?

Traumatic childhood experiences predict many adverse outcomes in adulthood including Complex-PTSD. Understanding complex trauma within socially disadvantaged populations has important implications for policy development and intervention implementation. This paper examined the nature of complex trauma experienced by disadvantaged individuals using a latent class analysis (LCA) approach. Data were collected through the large-scale Journeys Home Study (N = 1682), utilising a representative sample of individuals experiencing low housing stability. Data on adverse childhood experiences, adulthood interpersonal trauma and relevant covariates were collected through interviews at baseline (Wave 1). Latent class analysis (LCA) was conducted to identify distinct classes of childhood trauma history, which included physical assault, neglect, and sexual abuse. Multinomial logistic regression investigated childhood relevant factors associated with class membership such as biological relationship of primary carer at age 14 years and number of times in foster care. Of the total sample (N = 1682), 99% reported traumatic adverse childhood experiences. The most common included witnessing of violence, threat/experience of physical abuse, and sexual assault. LCA identified six distinct childhood trauma history classes including high violence and multiple traumas. Significant covariate differences between classes included: gender, biological relationship of primary carer at age 14 years, and time in foster care. Identification of six distinct childhood trauma history profiles suggests there might be unique treatment implications for individuals living in extreme social disadvantage. Further research is required to examine the relationship between these classes of experience, consequent impact on adulthood engagement, and future transitions though homelessness.     

Profiles of adverse childhood experiences and impulsivity

The influence of adverse childhood experiences (ACEs), which often include childhood exposure to maltreatment and household dysfunction, on health risk behaviors during young adulthood has been widely documented. A vulnerability marker for the increased risky behaviors among young ACEs victims such as impulsivity remains to be explored. The present study investigated how different profiles of ACEs influence impulsivity in young adulthood. Respondents were young people (N = 336; ages 18–25) who were recruited from the community. Latent class analysis (LCA) was used to identify subpopulations of individuals based on varying exposure to 13 types of ACEs, including childhood maltreatment, household dysfunction, and community violence. Four distinct classes emerged: Low ACEs (56%), Household Dysfunction/Community Violence (14%), Emotional ACEs (14%), and High/Multiple ACEs (16%). Multiple regression analyses found that compared to those in the Low ACEs class, young adults in the Emotional ACEs and High/Multiple ACEs respectively, reported increased levels of negative urgency, controlling for sociodemographic characteristics and other impulsivity traits. This result suggests that childhood exposure to multiple ACEs at high levels is particularly related to impulsive self-control in the context of intense negative emotionality.     

Do I belong in my math class? The importance of perceived racial/ethnic context and math course sequence

As careers in science, technology, engineering, and math (STEM) continue to grow, so has attention to Algebra 1 enrollment timing that serves as a critical predictor of STEM success. The present study adds to the literature by examining if Algebra 1 enrollment timing from 8th to 9th grade is related to sense of belonging in math, and whether this association changes as a function of the students’ perceived school and math race/ethnic context. To capture the dynamic nature of these contexts, we examined racial/ethnic incongruence, or the difference in the perceived number of same-race/ethnic peers in math class and the school of Black, White, Latino, and Asian students. Mixed effects linear modeling analyses were conducted on a sample of 2,938 participants (46% males; 54% females) who attended 26 racially/ethnically diverse middle schools and who transitioned to 142 public high schools in California. The results showed that enrolling and successfully passing 8th Algebra was protective for sense of belonging but this association depended on students’ race/ethnicity and the racial/ethnic incongruence between the math class and school context. Findings have important implications for math education, teaching, and policy.