Understanding Linkages Among Mixture Models

A practice-based study of relational virtues and alliance correspondence in psychodynamic psychotherapy

OBJECTIVE

Treatment outcome monitoring typically emphasizes pathology. In contrast, we responded to the need to establish psychodynamic psychotherapy as evidence-based by modeling changes in gratitude and forgiveness.

METHOD

We utilized a practice-based research design involving non-manualized outpatient treatment. We employed a longitudinal mixture modeling approach to evaluate treatment effectiveness. We did so by testing the theorized role for relational virtues (i.e., gratitude, forgiveness) as signs of progress in psychodynamic treatment, with relational virtues referring to the application of character strengths to specific situations. We modeled clients' self-reported level on the virtues as a joint process over five time points, and examined the influence of early treatment alliance correspondence on patterns of change using a sample of outpatient clients (N = 185; Mage = 40.12; 60% female; 74.1% White).

RESULTS

A 3-class solution best fit the data, with one class exhibiting growth in gratitude and forgiveness, improved symptoms, and a greater likelihood of symptom improvement relative to well-being gains. Alliance correspondence predicted the classes of change patterns, with greater similarity between clinicians' and clients' perceptions about the alliance predicting greater likelihood of belonging to the subgroup showing highest levels of virtues and well-being, lowest symptoms, and improved well-being.

CONCLUSION

Clinical implications involve monitoring gratitude and forgiveness as signs of progress and navigating the dialectic between implicit alliance processes and explicit virtue interventions. The former involves nurturing a strong alliance and repairing ruptures, whereas the latter involves direct in-session conversation and/or the practice of virtue interventions in and/or outside of session.

Promoting Adolescents' Heritage Cultural Identity Development: Exploring the Role of Autonomy and Relatedness Satisfaction in School-Based Interventions

Given the significance of heritage cultural identity for optimal adolescent development, it is imperative to investigate factors influencing the efficacy of interventions aimed at promoting heritage cultural identity. Using latent profile cluster analysis and multinomial logistic regressions, this longitudinal study examined how autonomy and relatedness need satisfaction at school (1) related to heritage cultural identity development trajectories, and (2) moderated effects of a school-based intervention. The study included N = 198 adolescents (Mage = 12.86 years, SDage = 0.75, 52% female, 41% immigrant descent, 49% intervention group) in Germany. Teacher-student relationships played an important role in facilitating intervention effects on identity development trajectories, emphasizing the importance of the relational context when implementing school-based interventions to promote heritage cultural identity development.

Sign-tracking and goal-tracking in humans: Utilising eye-tracking in clinical and non-clinical populations

BACKGROUND

In Pavlovian conditioning, learned behaviour varies according to the perceived value of environmental cues. For goal-trackers (GT), the cue merely predicts a reward, whilst for sign-trackers (ST), the cue holds incentive value. The sign-tracking/goal-tracking model is well-validated in animals, but translational work is lacking. Despite the model's relevance to several conditions, including attention deficit hyperactivity disorder (ADHD), we are unaware of any studies that have examined the model in clinical populations.

METHODS

The current study used an eye-tracking Pavlovian conditioning paradigm to identify ST and GT in non-clinical (N = 54) and ADHD (N = 57) participants. Eye movements were recorded whilst performing the task. Dwell time was measured for two areas of interest: sign (i.e., cue) and goal (i.e., reward), and an eye-gaze index (EGI) was computed based on the dwell time sign-to-goal ratio. Higher EGI values indicate sign-tracking behaviour. ST and GT were determined using median and tertiary split approaches in both samples.

RESULTS

Despite greater propensity for sign-tracking in those with ADHD, there was no significant difference between groups. The oculomotor conditioned response was reward-specific (CS+) and present, at least partly, from the start of the task indicating dispositional and learned components. There were no differences in externalising behaviours between ST and GT for either sample.

CONCLUSIONS

Sign-tracking is associated with CS+ trials only. There may be both dispositional and learned components to sign-tracking, potentially more common in those with ADHD. This holds translational potential for understanding individual differences in reward-learning.

Parent-adolescent discrepancies in positive parenting and adolescent problem behaviors in Chinese families

Based on the discrepancy-maladaptive hypothesis and general strain theory, in this study, we examined two key aspects: first, the mediating role of self-control in the relationship between parent-adolescent discrepancies in positive parenting and adolescent internalizing and externalizing problems, and second, variations in problem behavior among subgroups with different parent-adolescent dyads reporting patterns. The participants were 349 intact Chinese families, with parents as the primary caregivers and teenagers aged 15-18 years who are attending secondary vocational schools. The results revealed that adolescents generally perceived lower levels of caring and behavioral control than parents. Compared to behavioral control, discrepancies in perceived levels of caring had more significant predicting levels of internalizing and externalizing problems, and the relationship between discrepancies of caring and internalizing and externalizing problems was mediated by self-control. Latent profile analysis revealed three parent-adolescent responding patterns (subgroups); compared to the other subgroups, only the subgroup characterized by adolescents perceiving lower caring and behavioral control than parents exhibited higher levels of internalizing and externalizing problems. The findings of this study provide insights on how parent-adolescent discrepancies may lead to adolescent problem behaviors and highlight the importance of self-control as a mediating mechanism.

Analyzing Latent Burnout Profiles in a Sample of Spanish Nursing and Psychology Undergraduates

There is abundant literature suggesting that university students in helping professions experience high levels of stress, leading to an increased risk of developing burnout. The objective of this study was to identify burnout profiles in a sample of 1162 Spanish nursing and psychology undergraduates using latent profile analysis, a person-oriented statistical method that can identify hidden homogenous subgroups within a heterogeneous population. We expected to replicate in university students the five-profile structure (burnout, overextended, disengaged, ineffective, and engagement) proposed by Leiter and Maslach using the burnout dimensions (emotional exhaustion, cynicism, and inefficacy) as indicators. The results showed that burnout, overextended, and engagement profiles were adequately replicated. Given that levels of inefficacy and cynicism were medium to low, the ineffective and disengaged profiles somewhat deviated from those identified by Leiter and Maslach. We found differences between the five latent profiles in several psychological variables, such as depression and anxiety. These results suggest that psychosocial factors (e.g., workload) are significant among students and may adversely impact their health, leading to psychosomatic and emotional disorders. Hence, designing effective interventions to prevent health problems associated with burnout seems advisable, considering the specific burnout profile that a student exhibits.

Stressful life events and resilience in individuals with and without a history of eating disorders: a latent class analysis

BACKGROUND

Eating disorders (EDs) are associated with a range of stressful life events, but few have investigated protective factors that may affect these associations. The current study used mixture modelling to describe typologies in life stress exposure and availability of protective resources in individuals with and without eating disorders (EDs).

METHODS

A case - control sample (n = 916) completed measures of stressful life events, resilience protective factors, emotion regulation, and symptoms of EDs, depression and anxiety. We conducted latent class analyses to identify subgroups of stress exposure and profile analyses of emotional regulation and resilience. The resulting two latent variables were combined to explore effects on ED status and symptomatology, depression, and anxiety as distal outcome variables.

RESULTS

We identified four classes of stressful life events (generally low, some abuse/bullying, sexual/emotional assaults, and high adversity). For protective resources, we identified six profiles that ranged from low to higher levels of protection with variations in social/family resources. The latent protection variable contributed more strongly to the distal outcomes than the latent stress variable, but did not moderate the latent stress and distal outcome variable relationships. Profiles characterized by lower protective resources included higher proportions of individuals with a lifetime ED, and were associated with higher scores on all symptom measures.

CONCLUSIONS

Intra- and interpersonal protective resources were strongly associated with lifetime EDs and current mental health symptom burden after accounting for stressful event exposure, suggesting protective factors may be useful to target in the clinical treatment of patients with ED.

Exploring reading profiles of rural school students

This study investigates the reading profiles of rural Grade 5 and 6 students (N = 262), a sample with a high proportion of English language learners. We administered a battery of reading and cognitive assessments to classify students' reading profiles and evaluate if performance on cognitive measures predicted membership in particular profiles. Data were analyzed using latent profile analysis. Latent profile analysis showed four distinct reading profiles in our sample: students with severe reading disabilities (< 2%), students at high risk of reading disability (14%), students at some-risk of reading disability (46%), and students who are typical readers (38%). Lower performance on cognitive measures was associated with group membership in the severe reading profile group compared to the group of students at some-risk of reading failure. In contrast, higher performance on cognitive measures was associated with group membership in the typical reader group compared to students at some-risk of reading failure. In keeping with the findings from past studies documenting reader profiles, we found heterogeneity in the reading profiles of rural upper-elementary grade students. We discuss the need for multicomponent interventions that target all areas of reading with some flexibility in the dosage of each reading component dependent on the reader profiles established prior to intervention.

A Cautionary Note about Having the Right Mixture Model but Classifying the Wrong People

Mixture models can be used for explanation or individual prediction and classification. In practice, researchers are often tempted to make the class membership manifest by classifying cases according to their class of maximum posterior probability and using the "observed" class membership directly or as a variable in follow-up analyses to predict distal outcomes. This study revisits the issue of correct class assignment in latent profile analysis by providing an example where the number of classes is known (3-classes), sampling variability is eliminated, and precise estimates of classification indices are provided. This pseudo-population study design assumes the data-generating mechanism is known and provides a "best-case" scenario for evaluating correct class assignment. We use a variety of classification indices and graphical displays to show that correct classification may be poor despite relatively high entropy and overall correct class assignment metrics (e.g., percent correct). Our study serves as a reminder of the risks associated with trying to make latent class memberships manifest.

Burnout of emergency nurses in a South African context: the role of job demands and resources, and capabilities

Emergency nurses are prone to burnout due to the nature of their profession and working environment, potentially putting their sustainable employability at risk and so too the care provided by and success of emergency departments. Psychological research has predominantly focused on samples drawn from western, educated, industrialized, rich, and democratic (WEIRD) societies, concerning a small part of the world population. Consequently, this study investigated emergency nurses' burnout in a non-WEIRD society and assessed the role of job demands-resources and work capabilities on their burnout levels. A total of 204 emergency nurses in a South African context participated in a cross-sectional survey. The Job Demands-Resources Scale, the Capability Set for Work Questionnaire, and the Burnout Assessment Tool-Short Form were administered. Using and developing knowledge and skills and building and maintaining meaningful relationships were the strongest work capabilities of emergency nurses. In contrast, earning a good income, involvement in important decisions, and contributing to something valuable were the weakest capabilities. Latent class analysis resulted in three capability sets: a robust capability set, an inadequate capability set, and a weak capability set. Regarding job resources, emergency nurses with a robust capability set reported better relationships with their supervisors and higher job autonomy than the inadequate and weak capability sets. In addition, emergency nurses with a robust capability set reported better co-worker relationships and better access to good equipment than those with a weak capability set. Nurses with an inadequate capability set experienced significantly more challenging job demands than the other two sets. Finally, nurses with a weak capability set (compared to the robust capability set) experienced significantly higher levels of exhaustion and mental distance. Improving emergency nurses' job resources (especially relationships with co-workers and supervisors, job autonomy, and equipment sufficiency) would increase their capabilities, decreasing their burnout levels, especially exhaustion and mental distance.

Latent profile analyses of disordered eating behaviors and nonsuicidal self-injury among Vietnamese adolescents

OBJECTIVE

Comorbidity of disordered eating (DE) behaviors and nonsuicidal self-injury (NSSI) is linked to increased functional impairment. The present study identified subtypes of DE and NSSI comorbidity in a non-Western, low- and middle-income country where there has been particularly little research in this area.

METHOD

Latent profile analyses (LPA) were conducted to identify patterns of subgroup comorbidity in self-reported DE behaviors and NSSI behaviors, in a sample of Vietnamese high-school students (N = 1451, 51% female). Parallel-process LPA was used to predict NSSI subgroup membership from DE subgroup membership.

RESULTS

A seven-class LPA model was identified for DE: (1) Low frequencies of all DE behaviors; (2) Frequent levels of all DE behaviors; (3) Frequent fasting and purging behaviors; (4) Frequent binge eating only; (5) Moderate binge eating; (6) Moderate fasting and purging behaviors; (7) Exercise and fasting. A two-class (high versus low) model was identified for NSSI. Odds of membership in the high-NSSI class were significantly increased for all DE classes except the class characterized by moderate binge eating. Odds of belonging to the high-NSSI class were highest for those in DE classes involving purging behaviors. Males predominated in classes characterized by high behavior dysregulation.

DISCUSSION

DE and NSSI represent important clinical concerns among Vietnamese youth in urban contexts, and males may show particular vulnerability for certain risk behaviors. The association of classes of DE behaviors to NSSI was particularly strong for classes that involved purging, highlighting the importance of this specific behavior for understanding comorbidity of DE and NSSI.

PUBLIC SIGNIFICANCE

Disordered eating and nonsuicidal self-injury (NSSI; hurting oneself without intending to die) commonly co-occur. There has been less research on disordered eating and NSSI among individuals from non-Western or low/middle-income countries. We examined patterns of these behaviors in a school-based sample of Vietnamese adolescents. We found that patterns of disordered eating distinguished by purging showed particular links to NSSI, which could have implications for screening and treatment.

The longitudinal transition of the moral character latent profile of elementary school students and predictive factor verification in Korea

This study investigated the longitudinal transition of the moral character profiles of elementary school students and their predictive factors. Latent profile, latent transition, and logistic regression analyses were conducted using data from 2132 elementary school students in the fourth and sixth grades, who responded to both the first and third surveys in 2017 and 2019, respectively. The results were as follows: First, the latent groups of moral character profiles of elementary school students were classified into three types each year: the "low," "intermediate," and "high" moral character types. Second, the highest transition pattern was observed in the transition from the "intermediate" moral character type to the "high" type. Third, "happiness" was the most powerful predictor affecting the latent group transition of moral character, followed by "parental relationship," "teacher-student relationship," and "volunteer activities.". The findings suggest that the government needs to organize and provide effective educational methods considering parental relationship, student-teacher relations, and "volunteer activities" to improve the morality of elementary school students in the future.

Moore EW, Hull DM. Finding latent groups in observed data: A primer on latent profile analysis in Mplus for applied researchers

The present guide provides a practical guide to conducting latent profile analysis (LPA) in the Mplus software system. This guide is intended for researchers familiar with some latent variable modeling but not LPA specifically. A general procedure for conducting LPA is provided in six steps: (a) data inspection, (b) iterative evaluation of models, (c) model fit and interpretability, (d) investigation of patterns of profiles in a retained model, (e) covariate analysis, and (f) presentation of results. A worked example is provided with syntax and results to exemplify the steps.

Exploring Science Career Interest: Latent Profile Analysis of High School Occupational Preferences for Science

The present study evaluates high school juniors and seniors ( n = 295) to explore their preference for science as indicated by science motivation, attitude, academic experience, and interest. Latent profile analysis was used to model profiles of preferences for science with a person-centered approach. Then, the impact of self-concept variables was explored and four profiles of science interest were identified. Gender differences were of particular interest due to concerns noted in the literature, and some gender differences were identified in the present study. Covariate analysis indicated vocabulary ability and personality as significantly different for students in the high science interest profile. Implications of these results and future research directions are discussed.

The Impact of Item Misspecification and Dichotomization on Class and Parameter Recovery in LCA of Count Data

Mixture analysis of count data has become increasingly popular among researchers of substance use, behavioral analysis, and program evaluation. However, this increase in popularity seems to have occurred along with adoption of some conventions in model specification based on arbitrary heuristics that may impact the validity of results. Findings from a systematic review of recent drug and alcohol publications suggested count variables are often dichotomized or misspecified as continuous normal indicators in mixture analysis. Prior research suggests that misspecifying skewed distributions of continuous indicators in mixture analysis introduces bias, though the consequences of this practice when applied to count indicators has not been studied. The present work describes results from a simulation study examining bias in mixture recovery when count indicators are dichotomized (median split; presence vs. absence), ordinalized, or the distribution is misspecified (continuous normal; incorrect count distribution). All distributional misspecifications and methods of categorizing resulted in greater bias in parameter estimates and recovery of class membership relative to specifying the true distribution, though dichotomization appeared to improve class enumeration accuracy relative to all other specifications. Overall, results demonstrate the importance of accurately modeling count indicators in mixture analysis, as misspecification and categorizing data can distort study outcomes.

Extensions of Auxiliary Variable Approaches for the Investigation of Mediation, Moderation, and Conditional Effects in Mixture Models

Person-centered analyses and mixture models, such as latent profile analyses (LPA), are becoming increasingly common in the organizational literature. However, common usage of LPA rarely extends to the estimation of moderation, conditional effects, and mediation within a single model. This can affect the accuracy of parameter estimates, and it interferes with development and investigation of complex theories. The current study provides an overview of systematic approaches that allows researchers to investigate models involving moderation, conditional effects on outcomes, and mediation. Using M plus, we offer an accessible method of testing complex statistical models that are auxiliary to the focal mixture model. We provide syntax for typical moderation, conditional effects, and mediation hypotheses, and we provide a detailed explanation of the procedures. We demonstrate these procedures with applications involving the five-factor model (FFM) of personality and several additional variables that comprise complex auxiliary statistical models. The pedagogical approach offered by this research will facilitate future theoretical developments and empirical advancements in the use of person-centered analyses.

Multilevel Latent Profile Analysis With Covariates

Latent profile analysis (LPA) is a person-centered method commonly used in organizational research to identify homogeneous subpopulations of employees within a heterogeneous population. However, in the case of nested data structures, such as employees nested in work departments, multilevel techniques are needed. Multilevel LPA (MLPA) enables adequate modeling of subpopulations in hierarchical data sets. MLPA enables investigation of variability in the proportions of Level 1 profiles across Level 2 units, and of Level 2 latent classes based on the proportions of Level 1 latent profiles and Level 1 ratings, and the extent to which covariates drawn from the different hierarchical levels of the data affect the probability of a membership of a particular profile. We demonstrate the use of MLPA by investigating job characteristics profiles based on the job-demand-control-support (JDCS) model using data from 1,958 university employees clustered in 78 work departments. The implications of the results for organizational research are discussed, together with several issues related to the potential of MLPA for wider application.

Longitudinal profiles of mental well-being as correlates of successful aging in middle age

This study analyzed the multidimensional (including emotional, psychological, and social well-being) profiles of mental well-being and their links to various indicators of successful aging (SA; including diseases, cognitive and physical function, and engagement with life). The analyses were based on the Finnish Jyväskylä Longitudinal Study of Personality and Social Development, where the age-cohort participants have been followed from age 8 to 50. Data on 335 participants collected in mid-adulthood were analyzed. Applying Latent Profile Analysis and measures of life satisfaction and psychological well-being at ages 36, 42, and 50 and social well-being at ages 42 and 50, four longitudinal well-being profiles were extracted: high (29% of the participants), relatively high (47%), moderate (22%), and low (3%). ANCOVAs (controlling for gender and education) revealed that the participants in the high, relatively high, and moderate well-being profiles had more satisfying relationships, better labor market success, and fewer diseases at age 50 than those in the low well-being profile. Fewer inter-profile differences were observed in physical or cognitive function. Favorable profiles of mental well-being are related to SA in mid-adulthood. Future studies should investigate the causal relations between well-being and SA.

Capturing Context-Related Change in Emotional Dynamics via Fixed Moderated Time Series Analysis

Much of recent affect research relies on intensive longitudinal studies to assess daily emotional experiences. The resulting data are analyzed with dynamic models to capture regulatory processes involved in emotional functioning. Daily contexts, however, are commonly ignored. This may not only result in biased parameter estimates and wrong conclusions, but also ignores the opportunity to investigate contextual effects on emotional dynamics. With fixed moderated time series analysis, we present an approach that resolves this problem by estimating context-dependent change in dynamic parameters in single-subject time series models. The approach examines parameter changes of known shape and thus addresses the problem of observed intra-individual heterogeneity (e.g., changes in emotional dynamics due to observed changes in daily stress). In comparison to existing approaches to unobserved heterogeneity, model estimation is facilitated and different forms of change can readily be accommodated. We demonstrate the approach's viability given relatively short time series by means of a simulation study. In addition, we present an empirical application, targeting the joint dynamics of affect and stress and how these co-vary with daily events. We discuss potentials and limitations of the approach and close with an outlook on the broader implications for understanding emotional adaption and development.

Outcome Trajectories among Homeless Individuals with Mental Disorders in a Multisite Randomised Controlled Trial of Housing First

PURPOSE

Housing First (HF) has been shown to improve housing stability, on average, for formerly homeless adults with mental illness. However, little is known about patterns of change and characteristics that predict different outcome trajectories over time. This article reports on latent trajectories of housing stability among 2140 participants (84% followed 24 months) of a multisite randomised controlled trial of HF.

METHODS

Data were analyzed using generalised growth mixture modeling for the total cohort. Predictor variables were chosen based on the original program logic model and detailed reviews of other qualitative and quantitative findings. Treatment group assignment and level of need at baseline were included in the model.

RESULTS

In total, 73% of HF participants and 43% of treatment-as-usual (TAU) participants were in stable housing after 24 months of follow-up. Six trajectories of housing stability were identified for each of the HF and TAU groups. Variables that distinguished different trajectories included gender, age, prior month income, Aboriginal status, total time homeless, previous hospitalizations, overall health, psychiatric symptoms, and comorbidity, while others such as education, diagnosis, and substance use problems did not.

CONCLUSION

While the observed patterns and their predictors are of interest for further research and general service planning, no set of variables is yet known that can accurately predict the likelihood of particular individuals benefiting from HF programs at the outset.

A Latent Transition Analysis Model for Latent-State-Dependent Nonignorable Missingness

Psychologists often use latent transition analysis (LTA) to investigate state-to-state change in discrete latent constructs involving delinquent or risky behaviors. In this setting, latent-state-dependent nonignorable missingness is a potential concern. For some longitudinal models (e.g., growth models), a large literature has addressed extensions to accommodate nonignorable missingness. In contrast, little research has addressed how to extend the LTA to accommodate nonignorable missingness. Here we present a shared parameter LTA that can reduce bias due to latent-state-dependent nonignorable missingness: a parallel-process missing-not-at-random (MNAR-PP) LTA. The MNAR-PP LTA allows outcome process parameters to be interpreted as in the conventional LTA, which facilitates sensitivity analyses assessing changes in estimates between LTA and MNAR-PP LTA. In a sensitivity analysis for our empirical example, previous and current membership in high-delinquency states predicted adolescents' membership in missingness states that had high nonresponse probabilities for some or all items. A conventional LTA overestimated the proportion of adolescents ending up in a low-delinquency state, compared to an MNAR-PP LTA.

Regime Switching Modeling of Substance Use: Time-Varying and Second-Order Markov Models and Individual Probability Plots

A linear latent growth curve mixture model with regime switching is extended in 2 ways. Previously, the matrix of first-order Markov switching probabilities was specified to be time-invariant, regardless of the pair of occasions being considered. The first extension, time-varying transitions, specifies different Markov transition matrices between each pair of occasions. The second extension is second-order time-invariant Markov transition probabilities, such that the probability of switching depends on the states at the 2 previous occasions. The models are implemented using the R package OpenMx, which facilitates data handling, parallel computation, and further model development. It also enables the extraction and display of relative likelihoods for every individual in the sample. The models are illustrated with previously published data on alcohol use observed on 4 occasions as part of the National Longitudinal Survey of Youth, and demonstrate improved fit to the data.

Models for the Detection of Deviations from the Expected Processing Strategy in Completing the Items of Cognitive Measures

This paper presents confirmatory factor models with fixed factor loadings that enable the identification of deviations from the expected processing strategy. The instructions usually define the expected processing strategy to a considerable degree. Simplification is a deviation from instructions that is likely to occur in complex cognitive measures. Since simplification impairs the validity of the measure, its identification is important. Models representing simplicity and instruction-based processing strategies were considered in investigating the data of 345 participants obtained by a working memory measure in order to find out whether and how the use of these strategies influences model-data fit. As expected, the consideration of simplicity strategies improved the model-data fit achieved for the instruction-based strategy.

Handling Missing Covariates in Conditional Mixture Models Under Missing at Random Assumptions

Mixture modeling is a popular method that accounts for unobserved population heterogeneity using multiple latent classes that differ in response patterns. Psychologists use conditional mixture models to incorporate covariates into between-class and/or within-class regressions. Although psychologists often have missing covariate data, conditional mixtures are currently fit with a conditional likelihood, treating covariates as fixed and fully observed. Under this exogenous-x approach, missing covariates are handled primarily via listwise deletion. This sacrifices efficiency and does not allow missingness to depend on observed outcomes. Here we describe a modified joint likelihood approach that (a) allows inference about parameters of the exogenous-x conditional mixture even with nonnormal covariates, unlike a conventional multivariate mixture; (b) retains all cases under missing at random assumptions; (c) yields lower bias and higher efficiency than the exogenous-x approach under a variety of conditions with missing covariates; and (d) is straightforward to implement in available commercial software. The proposed approach is illustrated with an empirical analysis predicting membership in latent classes of conduct problems. Recommendations for practice are discussed.