Q1:
Consider a healthy vending program within a school or workplace. This type of program involves the inclusion and marketing of only “healthy” snacks and drinks within the school/workplace vending machines. The goal of this program is to improve the nutritional quality of the snacks/drinks consumed by the students and/or employees. You are asked to evaluate the effectiveness of this program.
Describe the conditions under which the following designs would be most appropriate.
Also, for each design, Identify and explain ONE (1) limitation/weakness with respect to internal or external validity
A) Post-program only with retrospective pre-program measure design (i.e., pre-and post-program outcome level is measured at the same time)
B) Post-program only with non-equivalent control group design (e.g., using another school/workplace as a comparison)
Q2: Suppose you wanted to gather a sample of 100 NYU School of Public Health students/faculty/staff to assess whether a healthy vending program indeed improved the nutritional quality of the snacks/drinks.
Answer each of the following:
• A) Name one of the probability sampling frames, and briefly describe how you might go about selecting your sample based on this sampling frame (4pts)
• B) Name one of the non-probability sampling frames, and briefly describe how you might go about selecting your sample based on this sampling frame (4 pts)
• C) Why might you choose the non-probability sampling approach over the probability sampling approach (3pts)
• D) What is one limitation of the non-probability sampling approach? (3pts)
Q3:
Consider a program that distributes bed nets to communities to help prevent malaria incidence.
For each of the following conditions, briefly describe one potential threat to INTERNAL validity, and make a recommendation for how the outcome evaluation could be re-designed in order to address or minimize this threat.
A) A simple one group pretest-posttest design
Condition: the subjects who agree to be observed as part of the evaluation tend to be more highly educated than the typical individual of the wider target population. (4pts)
B) A simple one group pretest-posttest design
Condition: the evaluation of this program begins in July and ends in September. Every year, over the course of this time period, malaria rates throughout the country tend to drop significantly simply due to the change in seasons. (4pts)
C) A non-equivalent control group pre and post design
Condition: the experimental group is more highly educated (on average) than the control group. (4pts)
Q4:
A) Describe ONE conclusion you can make about the findings on attendance and/or homework return rates. (3pts)
B) Then, speculate about how this conclusion from a process evaluation might be helpful when eventually interpreting results from an outcome evaluation. (3pts)
