url

docs/pkgdown.yml

@@ -4,7 +4,7 @@ pkgdown_sha: ~
 articles:
   glca: glca.html
   model-description: model-description.html
-last_built: 2024-08-27T12:54Z
+last_built: 2024-08-27T14:20Z
 urls:
   reference: http://kim0sun.github.io/glca/reference
   article: http://kim0sun.github.io/glca/articles

vignettes/glca.Rmd

@@ -393,7 +393,7 @@ The estimated coefficients for `SEX` and their odds ratios show that females are
 
 ### Tobacco smoking behavior
 
-The dataset `nyts18` comprises five dichotomized manifest items on the life-time experience of several types of tobacco including cigarettes (`ECIGT`), cigars (`ECIGAR`), chewing tobacco/snuff/or dip (`ESLT`), electronic cigarettes (`EELCIGT`), and hookah or water pipe (`EHOOKAH`) taken from the *National Youth Tobacco Survey 2018* (*NYTS 2018*, <https://www.cdc.gov/tobacco/data_statistics/surveys/nyts/>). The sample considered in this study includes 1,743 non-Hispanic white students from 45 schools. The number of sampled students from each school is in the range of 30 to 50. The school membership can be identified by `SCH_ID` and each school is classified as either middle or high school (`SCH_LEV`). (`?nyts18`)
+The dataset `nyts18` comprises five dichotomized manifest items on the life-time experience of several types of tobacco including cigarettes (`ECIGT`), cigars (`ECIGAR`), chewing tobacco/snuff/or dip (`ESLT`), electronic cigarettes (`EELCIGT`), and hookah or water pipe (`EHOOKAH`) taken from the *National Youth Tobacco Survey 2018* (*NYTS 2018*, <https://www.cdc.gov/tobacco/>). The sample considered in this study includes 1,743 non-Hispanic white students from 45 schools. The number of sampled students from each school is in the range of 30 to 50. The school membership can be identified by `SCH_ID` and each school is classified as either middle or high school (`SCH_LEV`). (`?nyts18`)
 
 According to socioecological models, patterns of adolescent tobacco smoking are best understood as embedded within social contexts. These social contexts can be either proximal in terms of individuals and peer groups or more distal in terms of schools and community. Socioecological models suggest that students within the same school often share common socio-economic status (SES) and cultural characteristics that may cause different tobacco smoking patterns compared to students attending other schools. However, reflecting school (group) effect in an mgLCR is less likely to provide a meaningful summary because there are too many groups, that is, 45 schools in the `nyts18` dataset. In this case, npLCR would be more appropriate model to investigate group difference in terms of a small number of latent clusters of schools.