Oversampling targeted groups

Build A Better Grinnell Survey is currently in its third phase. The first stage was collecting information from self-guided focus group tool kits; the gathered information formed a basis for the development of survey questionnaires. The second phase was launched in June 2012 in the form of a random telephone surveys. We received responses from 436 households (12.4% of the total households in Grinnell). The third phase includes oversampling certain groups such as the high school students, college students and low-income beneficiaries of MICA. 

Why oversample? The random survey in June was done by calling landline numbers (randomly selected) in Grinnell. While we got a good number of responses, about 50% of the respondents are 65 year or older. On a positive note, most of them must have lived in Grinnell long enough to have good comments on how well the city is implementing its services or what it needs to improve. However, the sample is not representative of the population. In fact, only 20% of the Grinnellians are 65 years of age or older (Grinnell Blue Zones Application, 2012).  Hence, we have started oversampling certain underrepresented groups to adjust the distribution of our sample.

How are we oversampling? We are sending out our online surveys (in Survey Monkey) through emails to our target groups such as the high school, middle school and the Grinnell College. In addition, we are interviewing low-income beneficiaries of MICA. 

We plan to have the data collected by the end of October. Our next phase will be data analysis followed by the final phase where we will be reporting our survey results to our stakeholders and the community of Grinnell. The results will likely be published by the end of the year 2012.