How to Conduct Market Research Surveys

a person's hand filling out a survey

 Sherri Lynn / Getty Images

Surveys are one of the most common market research tools. Most surveys emphasize either a qualitative or a qualitative design, but both approaches can be used in a single survey. Still trying to choose between qualitative and quantitative research methods?  

Why You Should a Quantitative Approach

At the most basic level, quantitative surveys are designed to take hard, numerical measures of responses to a research question, such as frequency, magnitude, and direction of change. Quantitative research methods are, of necessity, somewhat rigid.

The questions asked of survey respondents must stay the same from person to person, day-to-day, and so on. Any changes to the research script used by the interviewers, or the written questionnaire, or the survey processes could result in data that is not a true measurement of the research question or data that cannot be statistically analyzed.

Why You Should Choose a Qualitative Approach

Qualitative surveys explore the softer measures of research, such as "opinions" that are expressed in an open-ended fashion or the "why or why not" reasons people have for their responses to quantitative questions that do not ask for a reason or rationale.

Interviewers play a particularly important role in qualitative survey research that is administered in person or over the telephone. The way that questions are phrased, the intonation of spoken questions and the rapport between the respondent and the interviewer can have an important impact on data quality and direction.

For surveys covering complex subjects, it is critical that the interviewer understands the topic well in order to ask in-depth questions and probe effectively if respondents develop a reluctance to provide quality answers.

The Wording of Questions Is Important

The quality of the data that a researcher obtains from a survey is, to a large degree, dependent upon the skillful construction and wording of the questions posed. Survey questions must be clear, fairly easy to answer, and relatively short. It is helpful to conduct a pilot survey to see how people respond to the questions. What is clear to one person may be foggy to another. 

Here are some general guidelines for constructing survey questions.

  • Avoid ambiguous words
  • Omit easily misinterpreted words
  • Ask only one question at a time
  • Ensure questions do not "lead" respondents
  • Restrict questions to the research topic
  • Pilot questions before going live

Written or online questionnaires can generally be slightly more complex than face-to-face surveys where all information is exchanged orally. This is particularly true if respondents can easily refer back to earlier questions or navigate an online survey site without losing previous responses. As a general rule, survey respondents resent being asked questions that are deliberately designed to elicit a certain answer from them. Be sure that the questionnaires used for market research cannot be interpreted by respondents to be a sales pitch.

A Brief Review of Survey and Questionnaire Terms

  • Closed-ended questions: Respondents are asked to select from answers provided by the researcher in closed-ended questions. These questions are easy to quantify because of the uniformity of the responses, a factor that makes the question form popular in surveys. 
  • Open-ended questions: Respondents must supply their own answers to open-ended questions, such as those used for qualitative in-depth interviews.
  • Questionnaire: This is the document containing the questions designed to obtain the information that will be analyzed to answer research questions. Questionnaires are used in experiments, field research, surveys, and some forms of observation.
  • Response rate: Of the number of people who agreed to participate in a survey, the response rate is determined by the number of people who returned the survey. Also known as the completion rate. Expressed as a percentage. Response rate = # of survey participants / # in survey sample
  • Return rate: The percentage of questionnaires that have been returned of all the questionnaires sent out or distributed.