Latent classes from complex assessments

2021 ◽  
Author(s):  
Jake McMullen ◽  
Drew H Bailey
Keyword(s):  
Predictive Value ◽  
Test Scores ◽  
Latent Variable ◽  
Latent Class ◽  
Achievement Tests ◽  
Developmental Theory ◽  
Face Validity ◽  
Latent Classes ◽  
Practical Applications ◽  
Standardized Achievement Tests

Latent variable mixture models are commonly used to examine patterns of students' knowledge. These models, including Latent Class Analysis (LCA), have proven valuable for uncovering qualitative variation in students' knowledge that is hidden by traditional variable-centered approaches, particularly when testing a particular cognitive or developmental theory. However, it is far less clear that these models, when applied to broader measures of student knowledge, have practical applications, such as identifying meaningful and actionable knowledge patterns on standardized achievement tests. In the present study, we probe the practical effectiveness of LCA for identifying valid patterns of students' knowledge on broadly defined achievement tests that provide added predictive value beyond overall scores and other known indicators of success. We examined the performance of 3481 fifth-grade students from a mid-sized school district in the western United States on two benchmark assessments of their mathematics achievement during the school year. Latent classes extracted from pass-fail scores on specific standards measured by these assessments were then used to predict students' end-of-year performance on a statewide-standardized mathematics assessment. Latent classes generally showed face validity and identified qualitatively different knowledge patterns. The predictive value of class membership for end-of-year test scores was greatly reduced when adjusting for overall benchmark scores and very small after also adjusting for additional pre-existing differences among students. These results suggest that, although LCA might improve the interpretability of achievement test scores, their predictive value is largely redundant with overall scores. These results are tentative; we encourage replication with different kinds of data, especially with finer-grained measures.

scholarly journals The Q-Matrix Anchored Mixture Rasch Model

2021 ◽  
Vol 12 ◽  
Author(s):  
Ming-Chi Tseng ◽  
Wen-Chung Wang
Keyword(s):  
Rasch Model ◽  
Latent Variable ◽  
Latent Class ◽  
Parameter Estimates ◽  
Model Parameters ◽  
Latent Classes ◽  
Modeling Framework ◽  
Class Invariant ◽  
Common Scale ◽  
Q Matrix

Mixture item response theory (IRT) models include a mixture of latent subpopulations such that there are qualitative differences between subgroups but within each subpopulation the measure model based on a continuous latent variable holds. Under this modeling framework, students can be characterized by both their location on a continuous latent variable and by their latent class membership according to Students’ responses. It is important to identify anchor items for constructing a common scale between latent classes beforehand under the mixture IRT framework. Then, all model parameters across latent classes can be estimated on the common scale. In the study, we proposed Q-matrix anchored mixture Rasch model (QAMRM), including a Q-matrix and the traditional mixture Rasch model. The Q-matrix in QAMRM can use class invariant items to place all model parameter estimates from different latent classes on a common scale regardless of the ability distribution. A simulation study was conducted, and it was found that the estimated parameters of the QAMRM recovered fairly well. A real dataset from the Certificate of Proficiency in English was analyzed with the QAMRM, LCDM. It was found the QAMRM outperformed the LCDM in terms of model fit indices.

Examining Associations Between Race, Gender, Alcohol Use, School Performance, and Patterns of Bullying in the School Context: A Latent Class Analysis

2021 ◽  
pp. 088626052199912
Author(s):  
Valdemir Ferreira-Junior ◽  
Juliana Y. Valente ◽  
Zila M. Sanchez
Keyword(s):  
Alcohol Use ◽  
Latent Class Analysis ◽  
School Performance ◽  
Latent Class ◽  
School Context ◽  
Bullying Victimization ◽  
Latent Classes ◽  
Class Analysis ◽  
Seventh Grade ◽  
Sociodemographic Variables

Although many studies addressed bullying occurrence and its associations, they often use individual variables constructed from few items that probably are inadequate to evaluate bullying severity and type. We aimed to identify involvement patterns in bullying victimization and perpetration, and its association with alcohol use, school performance, and sociodemographic variables. Baseline assessment of a randomized controlled trial were used and a latent class analysis was conducted to identify bullying patterns among 1,742 fifth-grade and 2,316 seventh-grade students from 30 public schools in São Paulo, Brazil. Data were collected using an anonymous self-reported, audio-guided questionnaire completed by the participants on smartphones. Multinomial logistic regressions were performed to verify how covariant variables affected bullying latent classes. Both grades presented the same four latent classes: low bullying, moderate bullying victimization, high bullying victimization, and high bullying victimization and perpetration. Alcohol use was associated with all bullying classes in both grades, with odds ratio up to 5.36 (95% CI 3.05; 10.38) among fifth graders from the high bullying victimization and perpetration class. Poor school performance was also strongly associated with this class (aOR = 10.12, 95%CI = 4.19; 24.41). Black/brown 5th graders were 3.35 times more likely to fit into the high bullying victimization class (95% CI 1.34; 8.37). Lack of evidence for association of sociodemographic variables and bullying latent class among seventh-grade students was found. Bullying and alcohol use are highly harmful behaviors that must be prevented. However, prevention programs should consider how racial and gender issues are influencing the way students experience violence.

scholarly journals Co-occurrence subgroups of child sexual abuse, health risk behaviors and their associations among secondary school students in China

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Yan Zhang ◽  
Xiayun Zuo ◽  
Yanyan Mao ◽  
Qiguo Lian ◽  
Shan Luo ◽  
...  
Keyword(s):  
Sexual Abuse ◽  
Child Sexual Abuse ◽  
Secondary School ◽  
Health Risk ◽  
Risk Behaviors ◽  
Latent Variable ◽  
Secondary School Students ◽  
Latent Class ◽  
Computer Assisted ◽  
School Students

Abstract Background Little is known on the co-occurrence and heterogeneity of child sexual abuse (CSA) or health risk behavior (HRB) prevalence nor the associations among the victims. Objectives To detect the prevalence and subgroups of adolescents reporting CSAs or HRBs, and to examine the association between the subgroups. Methods Participants were secondary school students in a national survey in China (N = 8746). Self-reported CSA and HRB experiences were collected through a computer assisted questionnaire. Prevalence and confidence intervals were calculated. Multigroup latent class analysis (LCA) was used to examine latent subgroups of CSA and HRB. Dual latent class regression analysis was used to examine the association between CSA and HRB classes. Results A total of 8746 students participated in our study. The prevalence of having ever experienced any of the reported seven CSA items was 12.9%. The preferred LCA model consisted of a three-class CSA latent variable, i.e. “Low CSAs”(95.7% of the total respondents), “Verbal or exhibitionism CSAs”(3.3%), and “high multiple CSAs” (1.1%); and a three-class HRB latent variable, i.e. “Low HRBs”(70.5%), “externalizing HRBs” (20.7%), and “internalizing HRBs” (8.7%). Students in the “Verbal or exhibitionism CSAs” or “high multiple CSAs” classes had higher probabilities of being in “externalizing HRBs” or “internalizing HRBs” classes. The probabilities were higher in “high multiple CSAs” class(male externalizing OR 4.05, 95%CI 1.71–9.57; internalizing OR 11.77, 95%CI 4.76–29.13; female externalizing OR 4.97, 95%CI 1.99–12.44; internalizing OR 9.87, 95%CI 3.71–26.25) than those in “Verbal or exhibitionism CSA”(male externalizing OR 2.51, 95%CI 1.50–4.20; internalizing OR 3.08, 95%CI 1.48–6.40; female externalizing OR 2.53, 95%CI 1.63–3.95; internalizing OR 6.05, 95%CI 3.73–9.80). Conclusions Prevalence of CSA items varies. Non-contact CSAs are the most common forms of child sexual abuse among Chinese school students. There are different latent class co-occurrence patterns of CSA items or HRB items among the respondents. CSA experiences are in association with HRB experiences and the associations between latent classes are dose-responded. Multi-victimization has more significantly negative effects. The results could help identify high-risk subgroups and promote more nuanced interventions addressing adverse experiences and risk behaviors among at-risk adolescents.

Risk Factors Explaining Military Deaths From Suicide, 2008–2017: A Latent Class Analysis

2021 ◽  
pp. 0095327X2110469
Author(s):  
Scott D. Landes ◽  
Janet M. Wilmoth ◽  
Andrew S. London ◽  
Ann T. Landes
Keyword(s):  
Mental Health ◽  
Risk Factors ◽  
Latent Class Analysis ◽  
Military Service ◽  
Latent Class ◽  
Service Providers ◽  
Service Members ◽  
Latent Classes ◽  
Class Analysis ◽  
Military Service Members

Military suicide prevention efforts would benefit from population-based research documenting patterns in risk factors among service members who die from suicide. We use latent class analysis to analyze patterns in identified risk factors among the population of 2660 active-duty military service members that the Department of Defense Suicide Event Report (DoDSER) system indicates died by suicide between 2008 and 2017. The largest of five empirically derived latent classes was primarily characterized by the dissolution of an intimate relationship in the past year. Relationship dissolution was common in the other four latent classes, but those classes were also characterized by job, administrative, or legal problems, or mental health factors. Distinct demographic and military-status differences were apparent across the latent classes. Results point to the need to increase awareness among mental health service providers and others that suicide among military service members often involves a constellation of potentially interrelated risk factors.

Modelling Museum Visitors’ Perception and Experience of Authenticity: Examining Heterogeneity with a Finite Mixture Model

2021 ◽  
Vol 56 ◽  
pp. 0-0
Author(s):  
Claudia Bauer-Krösbacher ◽  
Josef Mazanec
Keyword(s):  
Latent Variable ◽  
Structural Equation ◽  
Latent Class ◽  
Structural Model ◽  
Unobserved Heterogeneity ◽  
Model Fitting ◽  
Equation Model ◽  
Finite Mixture ◽  
Museum Visitors ◽  
Specific Strength

Purpose. In this study, the authors explore the role of museum visitors’ perceptions and experiences of authenticity. They introduce several variants of authenticity experience and analyse how they are intertwined and feed visitor satisfaction. Method. The authors apply a multi-step model fitting and validation procedure including inferred causation methods and finite mixture modelling to verify whether the visitors’ perceptions of authenticity are subject to unobserved heterogeneity. They elaborate an Authenticity Model that demonstrates out-of-sample validity and generalisability by being exposed to new data for another cultural attraction in another city. Then, they address the heterogeneity hypothesis and evaluate it for the case study with the larger sample. Findings. In both application cases, the Sisi museum in Vienna and the Guinness Storehouse in Dublin, the empirical results support the assumed cause-effect sequence, translating high quality information display—from traditional and multimedia sources—into Perceived Authenticity and its experiential consequences such as Depth and Satisfaction. Accounting for unobserved heterogeneity detects three latent classes with segment-specific strength of relationships within the structural model. Research and conclusions limitations. The combined latent-class, structural-equation model needs validation with another sample that would have to be larger than the available Guinness database. Future studies will have to complement the purely data-driven search for heterogeneity with theory-guided reasoning about potential causes of diversity in the strength of the structural relationships. Practical implications. Cultural heritage sites are among the attractions most typical of city tourism. History tends to materialise in the artefacts accumulated by the population among the urban agglomerations, and museums are the natural places for preserving exhibits of cultural value. Authenticity must be considered an important quality assessment criterion for many visitors, whereby, the distinction between object authenticity and existential authenticity is crucial. Originality. In addition to making substantive contributions to authenticity theory, the authors also extend previous research in terms of methodological effort. Authenticity research, so far, has neither exploited inferred causation methods nor combined latent variable modelling with detecting unobserved heterogeneity. Type of paper: Research article.

Cluster Analysis with General Latent Class Model

2011 ◽  
pp. 225-230
Author(s):  
Dingxi Qiu ◽  
Edward C. Malthouse
Keyword(s):  
Cluster Analysis ◽  
Statistical Models ◽  
Latent Variable ◽  
Latent Class ◽  
Latent Class Model ◽  
Teaching Style ◽  
Teaching Styles ◽  
Binary Responses ◽  
Cluster Membership ◽  
Wide Range

Cluster analysis is a set of statistical models and algorithms that attempt to find “natural groupings” of sampling units (e.g., customers, survey respondents, plant or animal species) based on measurements. The observable measurements are sometimes called manifest variables and cluster membership is called a latent variable. It is assumed that each sampling unit comes from one of K clusters or classes, but the cluster identifier cannot be observed directly and can only be inferred from the manifest variables. See Bartholomew and Knott (1999) and Everitt, Landau and Leese (2001) for a broader survey of existing methods for cluster analysis. Many applications in science, engineering, social science, and industry require grouping observations into “types.” Identifying typologies is challenging, especially when the responses (manifest variables) are categorical. The classical approach to cluster analysis on those data is to apply the latent class analysis (LCA) methodology, where the manifest variables are assumed to be independent conditional on the cluster identity. For example, Aitkin, Anderson and Hinde (1981) classified 468 teachers into clusters according to their binary responses to 38 teaching style questions. This basic assumption in classical LCA is often violated and seems to have been made out of convenience rather than it being reasonable for a wide range of situations. For example, in the teaching styles study two questions are “Do you usually allow your pupils to move around the classroom?” and “Do you usually allow your pupils to talk to one another?” These questions are mostly likely correlated even within a class.

scholarly journals Methods to Account for Uncertainty in Latent Class Assignments When Using Latent Classes as Predictors in Regression Models, with Application to Acculturation Strategy Measures

Epidemiology ◽  
2020 ◽  
Vol 31 (2) ◽  
pp. 194-204 ◽  
Author(s):  
Michael R. Elliott ◽  
Zhangchen Zhao ◽  
Bhramar Mukherjee ◽  
Alka Kanaya ◽  
Belinda L. Needham
Keyword(s):  
Regression Models ◽  
Latent Class ◽  
Latent Classes ◽  
Acculturation Strategy

scholarly journals A MULTIDIMENSIONAL MODEL FOR POLYPHARMACY MEASUREMENT IN OLDER ADULTS: EVIDENCE FROM THE HEALTH RETIREMENT STUDY

2019 ◽  
Vol 3 (Supplement_1) ◽  
pp. S707-S707
Author(s):  
Rebecca Bendayan ◽  
Ewan Carr ◽  
Alex D Federman ◽  
Richard J Dobson
Keyword(s):  
Older Adults ◽  
Latent Class ◽  
Optimal Number ◽  
Model Fit ◽  
Latent Classes ◽  
Chi Square ◽  
Chronic Health ◽  
Adverse Health Outcomes ◽  
Traditional Research ◽  
Retirement Study

Abstract Polypharmacy is associated with increased health care costs and adverse health outcomes. Traditional research on polypharmacy uses dichotomous measures which overlook its multidimensional nature. We propose a new approach to grouping older adults based on the number and type of medications taken as well as other indicators of polypharmacy. Data was extracted from 1328 respondents of the 2007 Prescription Drug Survey (a sub-study of the Health Retirement Study) who were between 50 and 70 years old and taking ≥1 medication each month. Latent class analysis was carried out with the optimal number of classes assessed based on relative model fit (AIC, adjusted BIC) and interpretability. Latent classes were formed based on the number of medications, drug types, duration of medication intake, side effects, and presence of chronic health conditions. A four-class model was selected based on model fit and interpretability of the solutions. Although there was some overlap when we compared our model with standard cut-offs for polypharmacy (i.e., ‘high polypharmacy’ classes were more likely to take 5+ and 9+ medications), chi-square tests showed significant differences between our latent classes and cut-offs based on 5+ [X2 = 894; p<0.001] and 9+ medications [X2 = 398; p<0.001]. Among individuals taking <5 medications, our model differentiated two distinct types of ‘low polypharmacy’ based on the types of drugs reported. Our proposal to incorporate a multidimensional assessment of polypharmacy considers the wider context of medication use and chronic health in older age, moving beyond crude medication counts.

Sign in / Sign up

Export Citation Format

Share Document