Evaluating latent class analysis models in qualitative phenotype identification

Yang, Chih-Chien

doi:10.1016/j.csda.2004.11.004

Cited by 474 publications

(360 citation statements)

References 18 publications

Supporting

Mentioning

346

Contrasting

Unclassified

Order By: Relevance

“…Nested models were compared to determine the number of classes that provided the best model fit. The fit of models was compared by the bootstrap of the Lo-MendellRubin Likelihood Ratio Test (Lo et al, 2001), the Bayesian Information Criterion (BIC; Yang, 2006) and entropy. Class membership was decided based on the (maximum) probability of membership in each class.…”

Section: Discussionmentioning

confidence: 99%

The profile of psychiatric symptoms exacerbated by methamphetamine use

McKetin

Dawe

Burns

et al. 2016

Drug and Alcohol Dependence

View full text Add to dashboard Cite

Methods: 164 methamphetamine users, who did not meet DSM-IV criteria for a lifetime primary psychotic disorder, were followed monthly for one year to assess the relationship between days of methamphetamine use and symptom severity on the 24-item Brief Psychiatric Rating Scale.Exacerbation of psychiatric symptoms with methamphetamine use was quantified using random coefficient models. The dimensions of symptom exacerbation were examined using principal axis factoring and a latent profile analysis. Results:Symptoms exacerbated by methamphetamine loaded on three factors: positive psychotic symptoms (suspiciousness, unusual thought content, hallucinations, bizarre behavior); affective symptoms (depression, suicidality, guilt, hostility, somatic concern, self-neglect); and psychomotor symptoms (tension, excitement, distractibility, motor hyperactivity). Methamphetamine use did not significantly increase negative symptoms. Vulnerability to positive psychotic and affective symptom exacerbation was shared by 28% of participants, and this vulnerability aligned with a past year DSM-IV diagnosis of substance-induced psychosis (38% vs. 22%, χ 2 (df 1) = 3.66, p = 0.056). Conclusion:Methamphetamine use produced a symptom profile comprised of positive psychotic and affective symptoms, which aligned with a diagnosis of substance-induced psychosis, with no evidence of a negative syndrome.

show abstract

Section: Discussionmentioning

confidence: 99%

The profile of psychiatric symptoms exacerbated by methamphetamine use

McKetin

Dawe

Burns

et al. 2016

Drug and Alcohol Dependence

View full text Add to dashboard Cite

show abstract

“…Different criteria were used to determine the best model (Nylund et al, 2007). First, the Bayesian Information Criterion (BIC; Henson, Reise, & Kim, 2007) and the sample-size adjusted BIC (SSA-BIC; Yang, 2006) were inspected, with lower values indicating better model fit. Second, the bootstrap likelihood ratio test (BLRT; Nylund et al, 2007) was used to compare the fit of two competing models.…”

Section: Discussionmentioning

confidence: 99%

Youth Sport Coaches’ Well-Being Across the Season: The Psychological Costs and Benefits of Giving Empowering and Disempowering Sports Coaching to Athletes

Solstad

Ivarsson

Haug

et al. 2018

International Sport Coaching Journal

View full text Add to dashboard Cite

The purpose of this study was to investigate the associations between giving empowering and disempowering sports coaching to young athletes and coaches' well-being across the season. The sample comprised 169 Norwegian youth football (i.e., European soccer) coaches with a mean age of 41.99 (SD = 6.32). Moreover, we were interested in examining heterogeneous groups of coaches showing variability in their self-reporting of empowering and disempowering behaviors towards their athletes. Thus, a person-centered approach was used. The latent profile analysis revealed three distinct profiles and the association between these profiles and coaches' well-being was in line with the outlined hypotheses. Specifically, coaches who gave higher levels of empowering and lower levels of disempowering sports coaching to their athletes at the beginning of the season also reported higher levels of well-being at the end of the season. The results indicate that there exists an intrinsic value as to why coaches should give empowering sports coaching, as opposed to disempowering sports coaching, to their athletes; namely, these actions may be advantageous in terms of improving their own well-being. In practical terms, future coach education may take advantage of these findings by providing coaches another reason for coaching in an empowering manner.

show abstract

“…The best-fit fourclass solution selected for each offending stage presented high classification accuracy based on posterior probabilities 5 , confirming their stability and relevance. The assigned label and probability of membership for each encounter site and victim release site class, as well as the item-response probabilities for endorsing each item of the class, are shown in Tables 3 and 4 tends to underestimate the number of latent classes when limited sample sizes and/or large numbers of parameters are engaged (Yang, 2006). 5 Average assignment probabilities based on posterior probabilities for the four model solution ranged from .973 (.564-.993; neighborhood site profile), to .912 (.413-.999; shopping center site profile) for the victim encounter LCA solution and from .999 (.999-1.00; unfamiliar site profile), to .940 (.536-1.00; shopping center site profile) for the victim release LCA solution.…”

Section: Identification Of Latent Subgroups Of Victim Encounter and Vmentioning

confidence: 99%

Consistency in crime site selection: An investigation of crime sites used by serial sex offenders across crime series

Deslauriers‐Varin¹,

Beauregard²

2014

Journal of Criminal Justice

View full text Add to dashboard Cite

Purpose: Knowing sites used by serial sex offenders to commit their crimes is highly beneficial for criminal investigations. However, environmental choices of serial sex offenders remain unclear to this date. Considering the challenges these offenders pose to law enforcement, the study aims to identify sites serial sex offenders use to encounter and release their victims and investigate their stability across crime series. Methods: The study uses latent class analysis (LCA) to identify victim encounter and release sites used by 72 serial sex offenders having committed 361 sex offenses. Additional LCA are performed to investigate the stability of these offense environments across offenders' crime series. Results: Distinct profiles of crime sites that are recurrent across crime series are found, suggesting that serial sex offenders present a limited diversity of victim encounter and victim release sites. Encounter sites representative of longer crime series are also identified. Specifically, the use of sites known to "attract" potential victims decreases over series and offenders become more risk-taking in regard of sites used to encounter their victims. Conclusions: The study identifies patterns of site-selection for the victim encounter and release in cases of serial crimes. Implications for crime linkage and police investigations strategies are discussed.

show abstract

Evaluating latent class analysis models in qualitative phenotype identification

Cited by 474 publications

References 18 publications

The profile of psychiatric symptoms exacerbated by methamphetamine use

The profile of psychiatric symptoms exacerbated by methamphetamine use

Youth Sport Coaches’ Well-Being Across the Season: The Psychological Costs and Benefits of Giving Empowering and Disempowering Sports Coaching to Athletes

Consistency in crime site selection: An investigation of crime sites used by serial sex offenders across crime series

Contact Info

Product

Resources

About