RESEARCH STUDY DESIGN

We thought you’d like to know a little about the traditional steps of a research project. Here goes:

Define Your Objectives.

Why are you doing research? What do you need to find out? How will you use what you learn? These are important questions. If you can’t answer them, you’re not ready to do research.

REMEMBER: If you can’t do something with the results of your research (e.g. make a product change, improve customer service, etc) then all you have is something interesting to read.

Find out what you already know.

Look at your sales data, your customer comment cards, your demographic information. Look at previous studies you may have done. Do research at the library, look at journals, government publications, county or state data, computer data bases, and professional association literature.

NOTE: You can do all of this yourself or ask a market research vendor to do it for you. This type of research is called Secondary Research.

Decide which group or groups of people you want to talk to.

Who has the information, experience or opinions you want to know about? Is it all female customers, a group of specialized physicians or the man-on-the-street?

SUGGESTION: If you haven’t done so yet, this is a good time to bring in your research vendor.

Begin to formulate the types of questions you want to ask.

Only you and your associates know what you need to know. Talk to others in the company. What kinds of questions would they like to have answered? Write down your basic questions.

NOTE: Your research vendor has the skill to take a few roughly worded questions and turn them into gold. Based on the questions you put together, the professionals who work with you on your project can create all of the questions you will need to make this project work for you when the develop the questionnaire?

Develop the questionnaire.

Questions must be unbiased, clearly written, universally understandable and quantifiable. This is where the research company’s experience is invaluable.

SUGGESTION: Please read our section on Questionnaire Design.

Pre-test the questionnaire.

This is where you find out if there are any problems with your survey questionnaire. A small group of people from your target audience is questioned using the survey instrument. If there are any problems, this is where you’ll find them – and have plenty of time to fix them before the official project is launched.

NOTE: Sometimes the rewritten or reformulated questionnaire is pre-tested again.

Implement the survey.

This can take a few days to a few weeks or more, depending on the volume of your survey audience and how easy or difficult your audience is to find.

Create codes or categories that allow you to quantify responses to open-end questions.

If enough people have responded similarly to an open-end question in your survey, you can create a code or category for that response and quantify (by count and percentage) the number of respondents who answered the question in the same way.

Enter the data you have collected.

Create cross-tabulation charts.

Run statistical tests/analyses.

Interpret the data and write a report.