The Basic Experiment

In an effort to help prevent childhood obesity, some researchers have developed behavioral interventions. In

one school program, nutrition researchers wanted to teach children to make healthier food choices in the cafeteria.

They developed a short training program that involved a 20-minute interactive video.

To test the effectiveness of this video training program, the researchers designed a study with an experimental

condition, in which school children in first grade were exposed to the new program, and a control condition, in

which children watched a 20-minute cartoon having nothing to do with nutrition.

A food choice test measured the children's behavior in their real school cafeteria. Each food line presented

four choice points, in which a healthier option (e.g., fruit) competed against a less healthy option (e.g., vanilla

pudding). By closely observing a child as he moved through the cafeteria line, a researcher (a graduate student)

marked down the number of healthy options a child selected. A child could get a score from zero (child made

no healthy choices) to four (child made the healthy choice at all choice points).

Below are four descriptions of variations on the study described above.

a. Identify the design as pretest-posttest, posttest only, or repeated measures. (Just indicate A, B, or C.)

b. Identify the threats to internal validity in each design. Possible threats include history, maturation, testing, instrumentation, attrition, regression to the mean, selection, experimenter expectancy. (Again, indicate A, B, C, or D.)

c. Draw a sketch of each design, using the tic-tac-toe notation discussed Lecture.

Variation 1

Each child's name was put into a computer .Using an online randomizer, students were assigned to the experimental group and to the control group. One day at school, research assistants showed each group their respective videos in separate classrooms. Two days later, the food choice test was conducted. For ease of record keeping, all control group children were tested at lunch first, then all the experimental group children were tested. The graduate student (who was unaware of the group sequence) observed all of the children's food choices. In the beginning, he hid behind a cashier's booth and strained to see the food line clearly. Later on, he decided he could see better by standing nonchalantly near the food line. The mean score for children in the control group was 1.0 and the mean score for children in the experimental group was 3.5. We conclude that the 20-minute interactive video improved children's food choices.

1.The design is ________________.

A)pretest-posttest

B)posttest only

C)repeated measures

2.The largest threat to internal validity for this design is _________________.

A) attrition

B) regression to the mean

C) instrumentation

D) Selection

3.Sketch the design using tic-tac-toe notation. (Be sure to define the notation you use.)