Evaluating latent class analysis models in qualitative phenotype identification

By Chih-Chien Yang. Computational Statistics & Data Analysis, 2006

601 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/37d8cd9649faea97a16e0d49259b44f4e01493e3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

The Impact of Ignoring a Level of Nesting Structure in Multilevel Growth Mixture Models: A Monte Carlo Study (2010, methodology): “The commonly used and recommended fit index SABIC (Sample–size Adjusted Bayesian Information Criterion) ( Henson et al, 2007 ; Nylund et al, 2007 ; Tofighi & Enders, 2008 ; Yang, 2006 ) was used in determining if the two-class model is the best solution.”
Violence and Externalizing Behavior Among Youth in the United States (2014, background): “Notably, simulation studies indicate that the BIC is typically a superior indicator when compared to other information criterion statistics (Yang, 2006); however, it is a common practice to use a variety of indicators in the identification of the number of classes in LCA, latent profile analysis…”
Stability, change, and implications of students’ motivation profiles: A latent transition analysis (2017, methodology): “…effective (Henson, Reise, & Kim, 2007; Nylund, Asparouhov, & Muthén 2007; Peugh & Fan, 2013; Tein, Coxe, & Cham, 2013; Tofighi & Enders, 2008; Yang, 2006), while the AIC and LMR/ALMR should not be used in the class enumeration process as they respectively tend to over- and under- extract…”
A Multifoci Person-Centered Perspective on Workplace Affective Commitment: A Latent Profile/Factor Mixture Analysis (2011, methodology): “Among the recommended indices (e.g., Nylund et al., 2007; Yang, 2006), two converge on the five-class model (BIC and CAIC: previous simulations show that both tend to underestimate the number of classes when they are wrong) and two converge on the six-class model (ABIC and BLRT: previous…”
Covariates and Mixture Modeling: Results of a Simulation Study Exploring the Impact of Misspecified Effects on Class Enumeration (2016, methodology): “In this study, we use fit indexes that have been shown to perform well for mixture models based on other simulation studies (Li & Hser, 2011; Nylund et al., 2007; Yang, 2006).”
Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis (2013, methodology): “However, statistical power in latent class analyses is understudied; only a handful of studies have examined power or the effect of sample size on selecting the correct number of latent classes (see Nylund, Asparouhov, & Muthén, 2007; Tofighi & Enders, 2006; Yang, 2006).”
Deciding on the Number of Classes in Latent Class Analysis and Growth Mixture Modeling: A Monte Carlo Simulation Study (2007, methodology): “Simulation studies considering LCA models suggest that the adjusted BIC (Sclove, 1987) is superior to other IC statistics ( Yang, 2006 ).”
Disentangling Shape from Level Effects in Person-Centered Analyses: An Illustration Based on University Teachers’ Multidimensional Profiles of Effectiveness (2015, neutral): “…in mixture models (Henson, Reise, & Kim, 2007; McLachlan & Peel, 2000; Nylund, Asparouhov, & Muthén, 2007; Tofighi & Enders, 2008; Tolvanen, 2007; Yang, 2006): (a) the consistent Akaike’s Information Criterion (CAIC), (b) the Bayesian Information Criterion (BIC), (c) the sample-size adjusted…”
Sensitivity and Specificity of Information Criteria (2018, neutral): “Yang (2006) reported that ABIC performed better in general than AIC (whose model selection accuracy never got to 100% regardless of n) or BIC or CAIC (which underfit too often and required large n to be accurate).”
Impact of Misspecifications of the Latent Variance–Covariance and Residual Matrices on the Class Enumeration Accuracy of Growth Mixture Models (2016, background): “Previous research has shown sample size to be critical in the ability to recover the true number of classes present in the data (e.g., Li & Hser, 2011; Nylund, Asparouhov, & Muthén, 2007; Tofighi & Enders, 2007; Yang, 2006).”
Mixed Mode Latent Class Analysis: An Examination of Fit Index Performance for Classification (2015, methodology): “Prior simulation studies have shown that AIC is not a good indicator of the number of classes when used for finite mixture models (e.g., Koehler & Murphee, 1988; Nylund et al., 2007; Yang, 2006).”