A study of obesity risk in children in a head start program used a food score calculated from a 45-question food survey to predict body mass index (BMI) percentile in these children 18 months after the initial survey. The study enrolled 130 children. The researchers used a linear regression model for the prediction of BMI percentile. Which of the following is not a condition for linear regression?
A)BMI percentile varies according to a Normal distribution for a fixed food score.
B)The variation of BMI percentile around the regression line is the same for all food scores.
C)Mean BMI percentile increases for all children over the 18-month period.
D)Mean BMI percentile varies linearly with the food score.

Question

Quizplus · Accepted Answer

Linear regression assumes that the relationship between the predictor variable (food score) and the outcome variable (BMI percentile) is linear and that the variation of the outcome variable around the regression line is constant for all values of the predictor variable. Additionally, it assumes that the outcome variable follows a Normal distribution for each value of the predictor variable. However, linear regression does not require that the mean of the outcome variable changes over time, as suggested by choice C. Therefore, choice C is not a condition for linear regression.

A Study of Obesity Risk in Children in a Head