Alcohol and school attendance

Topic: Studies, data, and evidence · more questions on this topic
‹ A dose comparison for middle-of-the-night insomnia

A sample of 7874 adolescents is linked to a school registry. Frequent intoxication is recorded as a yes-or-no classification, collapsed from a five-point question. After accounting for age, sex, socio-economic status, and mental health, adolescents who were frequently intoxicated missed more hours of school.

  1. Is this an observational study or a designed experiment?

  2. What values can frequent intoxication take, and what type of variable is it?

  3. What values can hours missed take, and what type of variable is it?

  4. Which variable is the response, and why?

  5. Which variables are explanatory?

  1. Observational: nobody assigned the adolescents to be frequently intoxicated or not. The association with absence, even after adjustment, may be due to variables that were not recorded.

  2. Two values, yes or no, so it is a binary categorical variable. It was collapsed from a five-point ordinal question, which throws away some information.

  3. Hours missed can be any non-negative number, so it is quantitative. Recorded in whole hours it is discrete, and it is likely to be right-skewed, with many students missing little school and a few missing a lot.

  4. Hours missed is the response, because the study asks whether intoxication goes with more missed school.

  5. Frequent intoxication is the explanatory variable of interest. Age, sex, socio-economic status and mental health are also explanatory variables, included to adjust for possible confounding.