Strategies for Minimizing Total Error

Strategies for minimizing total survey error and sampling strategy Tang Li

We may find 3 kinds of survey errors: the sampling, nonresponse and the coverage error. In order to minimize the total survey error, we will apply the following 3 strategic procedures.
Minimizing Sampling error:
Firstly, we will get a copy of the full GPPI student list from Mrs. Pace, Asistant Dean for Academic and Student Affairs. Then we will randomly select a number of 100 students which will represent our sample. We believe that since there are about 300 students in GPPI, a sample size of 100 is large enough to minimize the sampling error. As an additional step, we will conduct a balance test to make sure that the sample is not statistically different from the total population.

Minimizing Coverage error:
Since we have already got the complete list of students and email addresses, we have covered all the samples, thus the coverage error will be minimized. But we should pay attention to those students who may not use their Georgetown email thus cannot take the survey, although the chance is low, we still need to send follow-up emails to make sure that they all take the survey, if necessary, we should make phone calls to those who do not use the Georgetown email address to complete the survey.

Minimizing Non-reponse
We have three Amazon gift cards as gifts for those who take the survey. In dealing with those people who haven’t complete the survey in a certain amount of time, we would send follow-up emails. If the follow-up emails do not work, we would make phone calls to complete the survey