Author Question: A researcher wishes to see whether there is any difference in the average weight gains of athletes ... (Read 54 times)

tiffannnnyyyyyy · **on:** Jul 24, 2018

A researcher wishes to see whether there is any difference in the average weight gains of athletes following one of three special diets (Diet 1, Diet 2, and Diet 3). The initial study design was to randomly assign the athletes to one of the three diet groups and place them on the diet for six weeks. The weight gain (in pounds) at the end of the six weeks would be recorded for each athlete. Suppose that 80 of the athletes are men and 20 of the athletes are women. In a randomized design it is possible that a diet (treatment) could have very few or no female athletes assigned to it. Suggest a way to improve on the design for the next study so as to control for the possible effects of gender.

Question 2

A researcher wishes to see whether there is any difference in the average weight gains of athletes following one of three special diets (Diet 1, Diet 2, and Diet 3). The initial study design was to randomly assign the athletes to one of the three diet groups and place them on the diet for six weeks. The weight gain (in pounds) at the end of the six weeks would be recorded for each athlete. Gender was also recorded for this study. The results showed that the amount by which the weight gains for the three diets differed was affected by whether the athlete was a male or female. What is the statistical name for the variable gender in this case?

choc0chan · **Reply #1 on:** Jul 24, 2018

Answer to Question 1

A block design could be used with gender as the blocking variable. The female athletes would be randomly assigned to the three diets and then the male athletes would be randomly assigned to the three diets.

Answer to Question 2

Gender is an explanatory variable that interacts with the principle explanatory variable (type of diet) in its relationship with the response variable (weight gain). The variable gender is an interacting variable.