Data on Obesity from the Muscatine Coronary Risk Factor Study.

The data are from the Muscatine Coronary Risk Factor (MCRF) study, a longitudinal survey of school-age children in Muscatine, Iowa. The MCRF study had the goal of examining the development and persistence of risk factors for coronary disease in children. In the MCRF study, weight and height measurements of five cohorts of children, initially aged 5-7, 7-9, 9-11, 11-13, and 13-15 years, were obtained biennially from 1977 to 1981. Data were collected on 4856 boys and girls. On the basis of a comparison of their weight to age-gender specific norms, children were classified as obese or not obese.

muscatine

Format

A dataframe with 14568 rows and 7 variables:

id: identifier of child.
gender: gender of child
base_age: baseline age
age: current age
occasion: identifier of occasion of recording
obese: 'yes' or 'no'
numobese: obese in numerical form: 1 corresponds to 'yes' and 0 corresponds to 'no'.

Source

https://content.sph.harvard.edu/fitzmaur/ala2e/muscatine.txt

Woolson, R.F. and Clarke, W.R. (1984). Analysis of categorical incompletel longitudinal data. Journal of the Royal Statistical Society, Series A, 147, 87-99.

Examples

muscatine$cage <- muscatine$age - 12                                         
muscatine$cage2 <- muscatine$cage^2                                          
                                                                        
f1 <- numobese ~ gender                                                 
f2 <- numobese ~ gender + cage + cage2 +                                
    gender:cage + gender:cage2                                          
                                                                        
gee1 <- geeglm(formula = f1, id = id,                                   
               waves = occasion, data = muscatine, family = binomial(),      
               corstr = "independence")                                 
                                                                        
gee2 <- geeglm(formula = f2, id = id,                                   
               waves = occasion, data = muscatine, family = binomial(),      
               corstr = "independence")                                 
                                                                        
tidy(gee1)                                                              
#> # A tibble: 2 × 5
#>   term        estimate std.error statistic p.value
#>   <chr>          <dbl>     <dbl>     <dbl>   <dbl>
#> 1 (Intercept)   -1.36     0.0462    873.    0     
#> 2 genderF        0.128    0.0648      3.90  0.0482
tidy(gee2)                                                              
#> # A tibble: 6 × 5
#>   term          estimate std.error statistic    p.value
#>   <chr>            <dbl>     <dbl>     <dbl>      <dbl>
#> 1 (Intercept)   -1.21      0.0546   494.     0         
#> 2 genderF        0.0962    0.0771     1.56   0.212     
#> 3 cage           0.0324    0.0153     4.48   0.0343    
#> 4 cage2         -0.0183    0.00403   20.7    0.00000537
#> 5 genderF:cage  -0.00427   0.0214     0.0399 0.842     
#> 6 genderF:cage2  0.00372   0.00564    0.436  0.509     
QIC(gee1)
#>       QIC      QICu Quasi Lik       CIC    params      QICC 
#> 10242.124 10239.167 -5117.584     3.478     2.000 10242.126 
QIC(gee2)
#>       QIC      QICu Quasi Lik       CIC    params      QICC 
#> 10202.748 10197.678 -5092.839     8.535     6.000 10202.766