A Survey of the Situation

Well, the time has come for yet another transition in the course of my research. In my last post, I noted some of the challenges I encountered when collecting contact information, particularly the disproportionate amount of New England contacts I recorded compared to other regions.

Thus, in order to avoid the data from being biased towards any specific region, I chose to stratify my sample into five different regions:  Northeastern States (defined as Connecticut, Delaware, Maine, Massachusetts, Maryland, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont), Southeastern States (defined as Alabama, Arkansas, Florida, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, Tennessee,Virginia, and West Virginia),  Midwestern States (defined as Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, North Dakota, Ohio, South Dakota, and Wisconsin), Southwestern States (defined as Arizona, New Mexico, Oklahoma, and Texas), and Western/Non-Contiguous States (defined as Alaska, California, Colorado, Hawaii, Idaho, Montana,  Nevada, Oregon, Utah, Washington, and Wyoming.) These strata were then selected in proportion to the number of electoral votes each respective region holds within the U.S. Electoral College. Within each strata, contacts were assigned a number and then randomly selected using a random number generator.

Emails including a brief description of the study and a link to the electronic survey were then sent out to a sample of approximately 1700 participants. In order to increase the response rate, emails were individually addressed to the appropriate election administrator and only sent during the five day work week. As of this post, 337 surveys have been started with 284 of those surveys considered usable. The response rate is lower than expected, coming in at approximately 20%.

However, after consulting with my faculty advisor, Dr. deFur, we have deemed that this response rate and number of participants are a respectable base for analysis. Thus, in lieu of sending out a second wave of reminder emails to non-respondents, I am instead going to devote my energy to focus more fully on analyzing the data. As many election administrators have been out of the office for personal vacations, I have decided to leave the survey open until this Friday, July 18. Once the survey is closed, I will then begin calculating my finding’s statistics. I am very excited to be moving on to this phase of the research, and look forward to sharing the findings!


  1. I am curious to know what questions were included in your survey and I can’t wait to see the results! Was your survey sent as a link in the email or were election officers asked to reply back to you with written answers to questions posed in your original email?