Which of the following plans would be the best way to collect data to determine if listening to music while studying for a social studies exam has an effect on exam performance?
(A) Randomly select a sample of students enrolled in a social studies class. After an exam in that class, ask each of these students the following two questions: (1) Did you listen to music while studying for the exam? and (2) What score did you get on the exam?
(B) Survey every student enrolled in a social studies class. After an exam in that class, ask each of these students the following two questions: (1) Did you listen to music while studying for the exam? and (2) What score did you get on the exam?
(C) Use a group of student volunteers who are enrolled in a social studies class. Randomly assign each student to either the music group or the no music group. Students in the music group will listen to music while studying for an upcoming exam and those in the no music group will not listen to music while studying for this exam. Record the exam score for each student.
(D) Use a group of students who are all enrolled in a social studies class. Have each student volunteer for either the music group or the no music group. Students in the music group will listen to music while studying for an upcoming exam and those in the no music group will not listen to music while studying for this exam. Record the exam score for each student.
Standards
S-IC.3: Recognize the purposes of and di"erences among sample surveys, experiments, and observational studies; explain how randomization relates to each.
Correct answer and commentary
Student performance
The correct answer to this question is Option (C). To determine whether listening to music causes an effect on exam performance, the study design must involve random assignment of students to groups. There is variability among the student volunteers in terms of intelligence, study habits, etc., but random assignment tends to “balance out” these differences and minimize the variability. If the experiment begins with two similar groups created by random assignment, a difference in exam performance can be attributed to study conditions (music or no music). If a study does not begin with random assignment, a difference in exam performance may be due to pre-existing differences between the two groups. Option (C) correctly explains how randomization relates to this experiment.
All of the other answer choices describe study designs in which students decide for themselves whether to listen to music while studying for the exam. The conclusions from the studies described in Options (A), (B), and (D) would all be limited, because the students who choose to listen to music while studying may be different from students who choose to study in silence. For instance, what if students who study in silence tend to be more serious about their schoolwork in general? These students may perform better on the exam for reasons (e.g. number of hours spent studying) that have nothing to do with listening to music. Thus, Options (A), (B), and (D) do not allow you to determine if listening to music has an effect on exam performance.