Karen Nylund-Gibson

A Latent Transition Mixture Model Using the Three-Step Specification

The 3-step method for estimating the effects of auxiliary variables (i.e., covariates and distal outcome) in mixture modeling provides a useful way to specify complex mixture models. One of the benefits of this method is that the measurement parameters of the mixture model are not influenced by the auxiliary variable(s). In addition, it allows for models that involve multiple latent class variables to be specified without each part of the model influencing the others. This article describes a unique latent transition analysis model where the measurement models are a latent class analysis model and a growth mixture model. We highlight the application of this model to study kindergarten readiness profiles and link it to elementary students’ reading trajectories. Mplus syntax for the 3-step specification is provided.

Patterns of Community Violence Exposure During Adolescence

This study examined cross-sectional and longitudinal patterns of community violence exposure and malleable predictors of these exposure patterns among a community sample of 543 urban African American early adolescents (45.3% female; mean age: 11.76). In each of grades 6, 7, and 8, latent class analyses revealed two patterns of community violence exposure: high exposure and low exposure. For the majority of participants, experiences with community violence were similar at each grade. Impulsive behavior and depressive symptoms distinguished adolescents in the high and low exposure classes in grade 6. Implications for interventions to prevent community violence exposure are discussed.

Covariates and Mixture Modeling: Results of a Simulation Study Exploring the Impact of Misspecified Effects on Class Enumeration

For some time, there have been differing recommendations about how and when to include covariates in the mixture model building process. Some have advocated the inclusion of covariates after enumeration, whereas others recommend including them early on in the modeling process. These conflicting recommendations have led to inconsistent practices and unease in trusting modeling results. In an attempt to resolve this discord, we conducted a Monte Carlo simulation to examine the impact of covariate exclusion and misspecification of covariate effects on the enumeration process. We considered population and analysis models with both direct and indirect paths from the covariates to the latent class indicators. As expected, misspecified covariate effects most commonly led to the overextraction of classes. Findings suggest that the number of classes could be reliably determined using the unconditional latent class model, thus our recommendation is that class enumeration be done prior to the inclusion of covariates.

Long-Term Stability of Screening for Behavioral and Emotional Risk

The practice of screening students to identify behavioral and emotional risk is gaining momentum, with limited guidance regarding the frequency with which screenings should occur. Screening frequency decisions are influenced by the stability of the constructs assessed and changes in risk status over time. This study investigated the 4-year longitudinal stability of behavioral and emotional risk screening scores among a sample of youth to examine change in risk status over time. Youth (N = 156) completed a self-report screening measure, the Behavioral and Emotional Screening System, at 1-year intervals in the 8th through 11th grades. Categorical and dimensional stability coefficients, as well as transitions across risk status categories, were analyzed. A latent profile analysis was conducted to determine if there were salient and consistent patterns of screening scores over time. Stability coefficients were moderate to large, with stronger coefficients across shorter time intervals. Latent profile analysis pointed to a three-class solution in which classes were generally consistent with risk categories and stable across time. Results showed that the vast majority of students continued to be classified within the same risk category across time points. Implications for practice and future research needs are discussed.

Campus Racial Climate: A Litmus Test for Faculty Satisfaction at Four-Year Colleges and Universities

This article reports on a study that investigated the relationship between three dimensions of campus racial climate and faculty satisfaction with a large, nationally representative faculty sample (N = 29,169), using a multilevel structural equation model. Results indicate that campus racial climate has a large and highly significant effect upon faculty satisfaction at the individual level.

Adolescent gang involvement: The role of individual, family, peer, and school factors in a multilevel perspective

Youth gang involvement is a serious public health challenge as adolescents involved in gangs are more likely than others to engage in violence and aggression. To better understand gang involvement, we examined the role of protective (empathy and parental support) and risk (peer deviance and lack of safety at school) factors, as well as their interactions, in predicting adolescent gang affiliation. The study involved a sample of 26,232 students (53.4% females; mean age = 14.62, SD = 1.69) participating in the California Healthy Kids Survey (CHKS), a survey investigating a wide range of youth health and risk behaviors administered in all California schools every 2 years. Using hierarchical linear modeling (HLM), findings indicated that high levels of empathy and parental support were associated with a lower likelihood of affiliating with a gang. Associating with deviant peers and perceiving the school as unsafe were positively correlated with gang membership. At the school level, lack of safety and type of school (special education, vocational, or alternative school vs. comprehensive schools) were associated with greater probability of gang membership. Empathy mitigated the association between deviant peers and gang membership.

A latent class analysis of past victimization exposures as predictors of college mental health.

Objective: The transition to college offers both new opportunities and challenges, and it is important to identify which students might be at greatest risk for adjustment difficulties. To that end, this study examines how victimization before college affects the transition to college, with particular attention to how victimizations group together and how these groups are differentially associated with depression and anxiety. Method: Participants from 4 universities (n = 1,337; 65.0% women) participated in online surveys of college adjustment in the fall of their first year, with 435 also completing surveys in the spring of their first year. Students reported about victimization before college (i.e., peer victimization, dating violence, attempted/completed rape, and child maltreatment), and about their fall and spring depression and anxiety. Results: Using latent class analysis, results indicated 4 victimization groups existed: Minimal, Peer Victims, Poly (sexual) Victims, and Polyvictims. Broadly, the college students in the 2 polyvictimization groups reported more psychological distress than students in other groups, but there were not significant differences between students in the Minimal or Peer Victim groups. Conclusions: Findings highlight the necessity of support and intervention services on college campuses attending to victimization as an indicator of potential maladjustment among first year college students.

Trauma Symptoms for Men and Women in Substance Abuse Treatment: A Latent Transition Analysis

Differences in trauma symptoms among men and women in two court-involved substance abuse treatment programs were examined using latent transition analysis (LTA). It was hypothesized that women would be more likely to report clinical-level trauma symptoms than would men, but that both groups would show reductions in symptoms over time. Symptom classifications were determined by the LTA. Scores on the Trauma Symptom Inventory (TSI) were obtained on 381 program participants, 112 men and 269 women, at intake and after 6 months in treatment. Three ordered classes were obtained for men and women at each time point: non-clinical (no TSI scales elevated), moderate symptoms (1 or 2 scales elevated) and severe symptoms (all scales elevated). Men were more likely to be represented in the non-clinical class at intake, while women had higher representation in the severe symptoms classification. There was a reduction of trauma symptoms for most men and women, but some groupings had symptoms that remained the same or became worse over time. Using gender and trauma-symptoms to help determine interventions is discussed.

Using latent class analysis to identify participant typologies in a drug treatment court

BACKGROUND Drug treatment courts serve a diverse population of adults. While all have engaged in criminal activities and have substance abuse problems, participants vary in the intensity of their problems as well as related concerns in other domains of functioning which also may require intervention. The purpose of this study was to identify differences among participants, which could have implications for the effectiveness of drug treatment courts. METHODS Latent class analysis (LCA) was used to identify subgroups from a sample of over 1000 adults attending two drug treatment courts in central California. Indicators measuring substance abuse, motivation for treatment, mental health concerns, education, employment, medical concerns, social supports, and demographic characteristics were obtained from the Addiction Severity Index while measures of prior criminal activity and treatment outcomes were obtained from probation; all were entered into the LCA. RESULTS The LCA yielded three groups, which were labeled a Psychological Problems group, an Early Delinquent group, and a Subthreshold Need group. Significant differences in graduation and recidivism rates were found across these groups, with the Early Delinquent group demonstrating the poorest outcomes. CONCLUSIONS This study provides evidence that there are significant differences among subgroups of drug treatment court participants. Implications for alternate treatment approaches based on participant characteristics are discussed.

Trajectories from academic intrinsic motivation to need for cognition and educational attainment

ABSTRACT This long-term longitudinal study addressed the theoretical underpinning of academic intrinsic motivation (AIM) from childhood through adolescence, to need for cognition (NFC) and educational attainment (EA) during adulthood. AIM was measured from 9 to 17 years old, NFC and EA at 29 years old, and IQ at 8 years old. Latent change and growth mixture modeling were utilized. These models complemented each other, revealing that initial motivational status significantly related to both outcomes. Growth mixture modeling elaborated the findings by identifying distinctive subgroups in initial status and developmental change. In contrast to children with initially higher AIM, those starting lower declined resulting in lower NFC and EA. IQ was controlled in these analyses. Findings enhance understanding of trajectories across two decades of development, indicating that students’ early motivation relates to adulthood NFC and EA. Implications for educational practices are discussed.