Tips on Survey Development

Before You Conduct a Survey

All research not originating in DAIR (including interviews, surveys, and questionnaires) involving humans as subjects must be reviewed by the Institutional Review Board (IRB). Provisional approval may be granted by the IRB as needed during the design of a project or preparation of a proposal. Full approval must be sought as soon as feasible, and must be obtained before the involvement of human subjects in the project begins.

Please see American University in Cairo Research Policies for further information.

Additionally, if your survey will be conducted outside of the AUC community, further approval may need to be sought from Egyptian authorities. Please see CAPMAS Procedures for Conducting Field Research.

Important Resources

 

Advantages of Web Surveys

  • Savings in printing, postage, data entry
  • No data entry errors from hand entry
  • Shortened timeframe to administer surveys (three weeks with web surveys, vs. six or more weeks with paper surveys)
  • Easier to provide skip patterns or survey sections customized to different respondent populations
  • Immediate access to data for analysis
  • Can easily link to background data (i.e. gender, grades and rank). 

 Background Questions to Ask Yourself

  • What is the purpose of the survey, and how will we use the survey data?
  • Who are we surveying?
  • Who are the audiences for the survey results?
  • What kind of support do we need for the survey effort?
  • What survey methodology will we use?
  • What is the project budget? Are funds available to pay incentives to potential respondents?
  • What is the time frame for this survey project (i.e. administration, time for responses, analysis, etc.)

Tips on Defining Survey Content

  • Identify themes that are of importance to the sponsor/department and other internal decision-makers. Obtain early and timely confirmation/approval from the Institutional Review Board.
  • Write questions using the assistance of experts (Office of Data Analytics and Institutional Research) to be sure that questions will yield meaningful results that will inform policy setting, changes to programs, etc. Avoid using jargon, shorthand, and technical terms unless you are certain this language will make sense to all survey participants.
  • Have other individuals review survey questions to ensure they represent the target audience, and to make certain the survey questions are appropriate for the times and the culture.
  • Revise questions as needed. Eliminate redundant or unnecessary words/questions to meet goals for survey length. Surveys with 30 or fewer questions (about 25 to 30 minutes) are most likely to receive a larger response rate.

Writing the Questions

The goal of writing a question is to develop a query that every potential respondent will interpret the same way, be able to respond accurately, and be willing to answer.

Types of Questions:

• Do not be vague – be very specific.
• Use vocabulary appropriate to audience.
• Avoid objectionable questions/language.
• Avoid hypothetical questions.
• Avoid leading questions.
• Ask only what you need.
• Equal number of positive/negative answer choices (in appropriate order).
• Range within answer choices should be concise/tight.
• Primary answer vs. all that apply.
• Limit open-ended questions.
• Don’t over-survey.

Questionnaire Design:

• Keep it short – 30 questions is best.
• Instructions – clear, to the point, and where they need to be.
• Use headings/division titles where appropriate.
• Number questions consecutively; keep answers responses structured similarly.
• Demographics – where to place?
• Take question order very seriously: general to specific.
• Avoid lots of skipping/go to questions.
• Open-ended Questions – place at end.
• Status bars/ page numbers.
• Print on both sides of paper (you decide, depending on length).
• Never landscape page format, always portrait.
• Avoid strange paper folding.
• Consider font type and size.
• Check spelling and grammar.

Source: AIR Professional File, No. 102, Winter 2006.

Additional Resources for Designing and Conducting Surveys

Best Practices for Survey and Public Opinion Research from the American Association for Public Opinion Research.

Conducting Primary Research: Tips on getting started, creating unbiased questions, and even collecting and analyzing data. (From Purdue University)

Essential Steps for Web Surveys: A guide to designing, administering and utilizing web surveys for university decision-making from the Association for Institutional Research

Sample question formats for web-based surveys

Sample Size calculator by Rasoft, Inc.

Sample Size calculator by Creative Research Systems

The Question Bank: Social surveys online

STATA Tutorial from the Carolina Population Center at the University of North Carolina at Chapel Hill

Survey Monkey: Online survey website that generates and distributes web-based surveys, in addition to providing basic analysis

Tips for Running a Survey on the Web by Jag Patel, of MIT's survey research center