Part I: A friendly introduction to latent class analysis

Aflaki, Kayvan; Vigod, Simone N.; Ray, Joel G.

doi:10.1016/j.jclinepi.2022.05.008

Cited by 23 publications

(15 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A three-class model was selected, as it had the lowest CAIC and BIC and entropy was ≥.80. 21,22 The frequency in each class is determined from the average probability of class membership for each of the health behaviors (Figure A1). 23,24 For frequent users, we expect that individuals have almost all the eHealth behaviors, and therefore a probability closer to 1 is expected.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Latent class analysis of eHealth behaviors among adults with epilepsy

et al. 2022

View full text Add to dashboard Cite

analyzed. We assessed the proportions of use of five domains of eHealth in those with epilepsy: looked up health information on the internet, filled a prescription on the internet, scheduled a medical appointment on the internet, communicated with a health care provider via email, and used chat groups to learn about health topics. Multivariate logistic regressions were conducted to identify factors associated with any eHealth behaviors among those with active epilepsy. Latent class analysis was performed to identify underlying patterns of eHealth activity. Survey participants were classified into three discrete classes: (1) frequent, (2) infrequent, and (3) nonusers of eHealth. Multinomial logistic regression was performed to identify factors associated with frequency of eHealth use.Results: There were 1770 adults with epilepsy, of whom 65.87% had at least one eHealth behavior in the prior year. By domain, 62.61% looked up health information on the internet, 15.81% filled a prescription on the internet, 14.95% scheduled a medical appointment on the internet, 17.20% communicated with a health care provider via email, and 8.27% used chat groups to learn about health topics. Among those with active epilepsy, female sex, more frequent computer usage, and internet usage were associated with any eHealth behavior. Female sex and frequent computer use were associated with frequent eHealth use as compared to nonusers.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The LCA model is selected using the Akaike information criterion (AIC), Bayesian information criterion (BIC), consistent AIC (CAIC), adjusted BIC, and entropy. A three‐class model was selected, as it had the lowest CAIC and BIC and entropy was ≥.80 21,22 …”

Section: Methodsmentioning

confidence: 99%

Latent class analysis of eHealth behaviors among adults with epilepsy

et al. 2022

View full text Add to dashboard Cite

show abstract

“…For the triangulation of our clustering analysis, we also applied a Latent Class Analysis (LCA) to identify subgroups within a population based on the patterns of observed variables (Aflaki et al, 2022). Like KAMILA, LCA has been shown to perform well in clustering the mixed type of data (Preud'Homme et al, 2021).…”

Section: Cluster Analysismentioning

confidence: 99%

Modularization of the front‐end logistics services in e‐fulfillment

Yurt

Sorkun²,

Hsuan

2023

J of Business Logistics

View full text Add to dashboard Cite

This study exploits service modularity in front‐end logistics services in e‐fulfillment, from a customer‐centric approach, particularly in order management, delivery, and return. Through an online survey of UK customers, the service priorities of 494 respondents via AHP (Analytic Hierarchical Process) were analyzed. Extracting customers' service priorities, ordering behavior, and demographic information as input data, the clustering algorithm KAMILA (KAy‐means for MIxed LArge data sets) was further applied. The three identified customer clusters (multichannel shoppers, infrequent shoppers, and online fans) provide preliminary evidence on how commonality and variability aspects of service modularity in front‐end logistics services can optimize the number of service options and their performance levels. Therefore, our study, building on value co‐creation and modularity, proposes a systematic way of exploiting service modularity for the customer segmentation process that addresses heterogeneous customer preferences cost‐efficiently and uncomplicatedly. Furthermore, we provide a framework for the governance of front‐end logistics services, guiding outsourcing decisions. Accordingly, it reveals the implications of customer priorities and service decomposition logic choices on value creation. Finally, the propositions formulated aim to develop theoretical foundations for explaining how the heterogeneity in customer priorities for logistics services can be managed with modularity, creating value both for customers and retailers.

show abstract

“…Another method employed in research on survey response behavior is latent class analysis (LCA) used to classify patterns of historical response. LCA classi es individuals into meaningful but unmeasured groups by identifying underlying patterns across a set of indicator variables using a probability model [10]. Cernat and Sakshaug [11] used LCA to classify participants enrolled in the Understanding Society Innovation Panel based on survey response over ve waves and examined the relationship between the historical survey response with later survey response to Wave 10.…”

Section: Introductionmentioning

confidence: 99%

Enhancing targeted outreach for longitudinal surveys: predictive analytics for participant response in the Millennium Cohort Study

Barkho,

Carnes,

Kolaja

et al. 2024

Preprint

View full text Add to dashboard Cite

Background: The Millennium Cohort Study is a prospective cohort study designed to examine the long-term effects of military service. The study collects self-reported data from surveys administered every 3–5 years to military personnel and veterans. Participant nonresponse to follow-up surveys presents a potential threat to the validity and generalizability of study findings. In recent years, predictive analytics has emerged as a promising tool to identify predictors of nonresponse. Methods: Here, we present a method that leverages machine learning techniques to develop a high-skill classifier to predict participant response to Millennium Cohort Study follow-up surveys. Using a temporal cross-validation method, six supervised algorithms, each using differing learning strategies, were employed to predict response to the 2021 follow-up survey. Using latent class analysis (LCA), we classified participants based on historical survey response and compared prediction performance with and without this variable. Feature analysis was subsequently conducted on the best performing model to identify the most important factors. Results: The LCA variable consisted of six distinct classes: consistent web responders, consistent paper responders, mode switchers, early and late consistent nonresponders, and inconsistent responders. Notably, the LCA classes were found to be statistically significant across various military and demographic characteristics. Including the LCA variable in the machine learning analysis, all six algorithms performed comparably to one another. However, without the LCA variable, random forest outperformed the benchmark regression model by 9.3% in area under the curve (AUC) of the receiver operating characteristic (ROC) curve and 7.7% in AUC of precision-recall. Moreover, inclusion of the LCA variable increased the ROC AUC for all algorithms by 10% or more, except for random forest and interaction forest, which improved by approximately 5% in ROC AUC. Feature analysis indicated LCA was the most important predictor in the model. Conclusions: Our findings highlight the importance of historical response patterns to significantly improve prediction performance of participant response to follow-up surveys. Machine learning algorithms can be especially valuable when historical data are not available. Implementing these methods in longitudinal studies can enhance outreach efforts by strategically targeting participants, ultimately boosting survey response rates and mitigating nonresponse.

show abstract

Part I: A friendly introduction to latent class analysis

Cited by 23 publications

References 5 publications

Latent class analysis of eHealth behaviors among adults with epilepsy

Latent class analysis of eHealth behaviors among adults with epilepsy

Modularization of the front‐end logistics services in e‐fulfillment

Enhancing targeted outreach for longitudinal surveys: predictive analytics for participant response in the Millennium Cohort Study

Contact Info

Product

Resources

About