Data Mining For Dummies
Book image
Explore Book Buy On Amazon

Surveys are useful for collecting data about almost any aspect of human life. You can only ignore surveys if your profession has nothing to do with people, like say, astrophysics. Then again, astrophysicists need people to fund their research and want people to visit planetariums, so they might need surveys, too! Here are examples of the varied uses of surveys:

  • Government: Assess the economic, physical, and mental state of people and businesses to support government activity. Nearly 200 national statistical agencies exist around the world, all using survey research to better understand their people. States, counties, and cities conduct local surveys for the same purposes.

  • Psychology: Study mental health and the working of the human mind.

  • Sociology and political science: Understand public attitudes regarding current issues.

  • Public health: Find out how people care for themselves, what choices they make regarding health options, and why.

  • Marketing and advertising: Measure brand awareness, product preferences, and other factors that impact purchasing behavior.

  • Advocacy and campaign management: Identify characteristics of supporters and detractors, and test campaign messaging options.

  • Media: Obtain information about public attitudes to include in reports and predict election outcomes.

  • Customer service: Assess customer satisfaction, and identify problems and possible solutions.

Developing questions

A good survey question should be

  • Specific: Deal with just one idea.

  • Narrow: Limited to a specific time frame, location, or whatever scope is appropriate for your needs.

  • Neutral: The wording should not lead the respondent to an answer.

  • Clear: Easily understood by anyone who might take the survey.

Often, it makes sense to offer options for responses. Response options should be

  • Simple: So that all respondents can understand them.

  • Consistent: All response options should have the same structure.

  • Complete: The full range of options must be covered.

  • Distinct: Options must not overlap.

These last two items, covering all possibilities and not overlapping, are often neglected. The result is confusion and frustration for respondents, and flawed data for you. So, put care into developing your survey questionnaires.

Conducting surveys

Now that you’ve developed your survey questions and assembled them into a questionnaire, you can reach out to people for answers. You will have to select one or more channels for reaching your respondents:

  • Face to face: An interviewer meets personally with the respondent, asks questions, and records responses. This method is often used for complex surveys, such as medical surveys or surveys where respondents are reluctant to respond, as with some government surveys.

  • Paper or kiosk: A respondent is given a form (or simply picks up a form left in a convenient spot) or is directed to an electronic kiosk to take the survey. Often used for customer service surveys.

  • Mail: The survey is mailed to the respondent, who fills it out and returns it by mail.

  • Telephone: An interviewer calls the respondent, asks the questions, and records the responses.

  • Internet: Respondents are recruited and take the survey online.

Your choice among these options depends on many factors. Your desired respondents may prefer some channels over others. Some cost more than others: A face-to-face interview in the respondent’s home costs far more than presenting the same questions online. And the time required to complete your survey varies with the way that you conduct it.

About This Article

This article is from the book:

About the book author:

Meta S. Brown helps organizations use practical data analysis to solve everyday business problems. A hands-on data miner who has tackled projects with up to $900 million at stake, she is a recognized expert in cutting-edge business analytics.

This article can be found in the category: