FROM THE EDITOR

TERRY R. RAKES, Decision Line Editor Virginia Polytechnic Institute and State University

The Institute has always recognized that doctoral students are the lifeblood of the organization, and has developed programs to encourage student participation. The Doctoral Student Consortium at the annual meeting, and the graduate student workshops held annually by several of the regions have helped to involve graduate students in the Institute early in their careers. DECISION LINE routinely runs announcements of activities that might interest our students, and has topical columns relevant to students and other members. However, in the past there has not been a substantial forum within DECISION LINE devoted entirely to doctoral student issues. I am happy to say that, starting with the next issue, we will be instituting a permanent column devoted to student affairs. Robert T. Sumichrast, Virginia Tech, will serve as the feature editor of this new column. Bob is the 1994-95 chair of the Doctoral Student Affairs Committee, and is very familiar with issues that are important to students.

Our goals for this new column are to discuss topics that will capture the interest of students, and to disseminate information in a timely fashion. Topics such as placement, interviewing, industry versus academic careers, choosing a dissertation topic, sources of graduate student funding from fellowships and grants, and many more are possible subjects of future columns.

In the next issue's inaugural column we plan to present a discussion by F. Robert Jacobs of Indiana University, the 1995 Doctoral Student Consortium coordinator, that should clarify issues and prepare students for the consortium in Boston. If you have a particular topic that you would like to see addressed, or if you are a student or faculty member who is interested in writing for the column, I am sure Bob Sumichrast will be anxious to talk with you (703-231-4535, fax: 703-231-4487).

Occasionally, I will print an article in this column that might appeal to a large number of our members. At the annual meeting in Hawaii, a session on questionnaires and response rates was presented, prompting one of the attendees to go home and write a piece that might interest many of you, including doctoral students whose dissertation research includes surveys. As an administrator, Jim Pope, dean of the Grove College of Business at Shippensburg University, receives many surveys and has definite criteria for deciding which to answer. I hope his comments will assist those of you involved in survey research.

IMPROVING SURVEY RESPONSE RATES

by James A. Pope, Shippensburg University

I have two surveys sitting on my desk right now. I will throw both of them in the trash can. This will be frustrating to those conducting the surveys, but they violated some of the rules I use to determine if I will fill out their forms and return them. As a dean, I get 20 to 30 surveys a year. I am writing this short piece to explain how to get me to return one. Perhaps the points I make may be generalized to other types of respondents.

The first thing I look at is the mailing label. If the people sending out the survey have used the AACSB mailing list, I expect them to use AACSB labels. If, on the other hand, they have taken the AACSB directory and copied it into their own data bases, I throw away the surveys. The caveat on AACSB labels specifically notes they are copyrighted and says "One-time use only. Conversion to computerized files or mailing lists strictly prohibited." Using the mailing list without paying is a serious ethical (and legal) violation. Besides, the labels only cost $30 for a full set. Particularly aggravating are those surveys that come addressed to deans who have not been here for several years. AACSB makes, on average, 10 to 15 changes per week to its member data base, so using the directory without paying for the labels guarantees that you will use a lot of bad addresses. Pay the money and do it right.

The cover letter is your next opportunity to make an impression. Mostly I ignore them because they all say the same thing. They try to tell me how significant the research based on the surveys will be and how it will be an important addition to the body of knowledge. There must be a form letter somewhere that people use. In reality, most of the surveys come from assistant professors seeking tenure and a promotion. Any significance of the research is purely incidental. When I do get an honest letter, it certainly catches my attention.

Are you paying the return postage or do you expect me to pay it? Expecting me to pay the postage will not disqualify your survey, but it puts up a barrier. Some people mailing out surveys do not even bother to include a return envelope or label!

Finally, we get to the survey itself. Cover letters abound with statements such as "only a few minutes of your time" or "at most ten minutes of your time." These estimates are generally either plain fraudulent or include only the time to fill in the blanks and not to gather the information. If I estimate that it will take me more than ten minutes total to fill out a survey, I don't do it. My estimate is generally based on a quick glance at the number of pages, the density of questions on each page, and on the kind of information requested.

Some surveys ask excruciating detail, such as the number of part-time female students who transfered from community colleges by race for the past five years per semester. Others want to know minute detail about our courses, including their contents, who takes them, and who teaches them. We have the data available, but I am not about to take the time personally to do the research; nor will I waste the time of those who work with me. Keep it short and simple. If that does not fit the needs of your research, find some other way to get the data.

Finally, I try to discern some point to the survey. Many surveys remind me of drift net fishing. They try to scoop up all the data available in the college. Research should have an hypothesis or a set of hypotheses which the data are used to test. If your only objective is to scoop up a huge mass of data and present it in tables and charts produced with a fancy piece of computer software, then forget it. A well-designed survey that is short and to the point (i.e., that has a point) can be rewarding for both the researcher and the subject. I have filled out some surveys that I have genuinely enjoyed completing. To my dismay, however, that only happens once or twice per year.

Researchers who send out surveys with all the problems I have talked about still get responses. It may not be a good response rate, but that has seldom deterred a presentation or an attempted publication. I have not seen a survey that has tried to measure the true non-response bias. I sometimes get multiple follow-up pleas, but I have never had anyone ask why I did not answer in the first place. Sometimes I will return a survey with a note as to why I did not complete it, but only if the return envelope is postage paid. The point is that if you send out a long complicated survey, the only responses will be from those who have a lot of time on their hands. If you think to measure the demographic characteristics of those who do not return your survey, you still do not know why they chose not to respond.

Survey research can be valuable if one does it right. Part of doing it right is designing a survey that is short, to the point, and one that the respondent will want to return. Pay attention to your subject and your experimental design and you will be pleased with your response rate.

JAMES A. POPE is the dean of the John L. Grove College of Business at Shippensburg University. He received his degrees from the College of Wooster, Northwestern University and the University of North Carolina. He has served the Southeast Region of DSI for ten years as an officer and is completing a two-year-term as a vice president of DSI. Presently he is a member of the APICS E&R Foundation Board of Directors.