Adolescents are getting less sleep that ever before, and this can have serious behavioral repercussions. The article “Dose-Dependent Associations Between Sleep Duration and Unsafe Behaviors Among US High-School Students” (JAMA Pediatr. 2018: 1187–1189) reported a large-scale study of American teenagers. The investigators fit a simple logistic regression model with the response variable y = 1 if a teenager had driven drunk in the last 30 days (and 0 otherwise), and x = typical number of hours of sleep per night. Information in the article suggests

a. Test whether sleep has an effect on the likelihood of driving drunk among American teenagers, at the .05 significance level.
b. Calculate a 95% confidence interval for e^β₁ .
c. Interpret the confidence interval from part (b) in terms of a one-hour decrease in sleep.

Transcribed Image Text:

P₁ = -.1998 and Sp₁ = .0986.