Quantitative Research: Using Closed- and Open-end Questions to Yield Quality Research

This is the second of three eTips! on quantitative market research. October's issue described quantitative research in general terms, specifically its differences from qualitative methods and its value in predicting the effect of marketing decisions on a particular population. Next month will explore the different types of closed-end questions in detail.

As established in the October eTip!, quantitative market research is used to statistically estimate or project a population's behavior in a given situation. Quantitative research varies from the qualitative variety because it is evaluative and not idea-generating in nature. When considering or conducting quantitative research, one should understand the basics behind the two types of questions used in its design: closed and open-end.

Closed-end questions assign a numeric or distinct value to each response, and no responses are accepted outside of those parameters. They are typically used in quantitative research because they allow an efficient collection of data from a large, representative audience. The most commonly types of closed-end questions include: rating scale, dichotomous (e.g. yes or no), paired and forced preference, numeric, and list. Because each of these tools force the participant to choose among pre-established choices, the resulting data is very specific and lends itself to accurate measurement and analysis and, if implemented, smart business decisions.

Open-end questions allow a participant to provide a textual response that captures more "explanatory" information. They are used in qualitative research because they allow the researcher to deeply probe participants and narrow topics, but they can also be used in quantitative research if used appropriately. Because "true" open-end questions are entirely textual responses that are recorded verbatim, they may require data collection personnel to clarify vague responses and probe for additional information. To avoid that additional effort, open-end questions are often designed with pre-coded responses that are anticipated by the researcher. Participants are not, however, presented with these pre-coded responses, so that unforeseen responses may be captured and included in the analysis as well.

Relying exclusively on closed-end questions has the potential to reduce the overall usefulness of a carefully planned quantitative project. It is often appropriate to gather information that cannot be captured by a numerical response. Also, when paired with closed-end questions, the open-end variety can identify the "why" behind a specific rating or preference. For instance, a company may discover that most individuals have retained a certain promotional or advertising message, but unless they know "why" this has occurred, the information could be of little value.

In summary, when combined properly, closed-end and open-end questions should provide a comprehensive picture of the situation at hand and will ultimately allow companies to make better strategic decisions for the future. When choosing a qualified market research supplier, these factors should always be addressed.