When you need structured information from a large number of people, surveys are the way to go. They allow you to fairly quickly collect data from many people in different locations. They can be used to inform project direction and focus of effort and to build empirical support for a hypothesis.

Surveys usually employ short, simple questions. They generally offer a series of answers from which participants choose. For example, a survey could ask respondents how strongly they agree or disagree with statements about the potential impact of a new project. Or you might offer multiple-choice questions that ask people to choose among alternative responses.

Structured questionnaires. Surveys may be self-administered questionnaires that are distributed by hand, mail, fax, or e-mail. They can also be administered face-to-face or over the phone. Regardless of what form they take, surveys are designed to gather information from a representative group of people.

Methods for getting information from a defined population. Sampling is an important part of survey planning and design. For example, if you want to learn what people over 65 think about your new statewide initiative, you could try to find and ask every person in this age group. However, this is not likely to be logistically or financially feasible. Instead, you will probably ask a random sample of the population of seniors. If your sample is drawn with statistical precision (a topic much too complicated for this note), you can then generalize the results from this smaller group to the entire population of people in that age group.

Sometimes the population you want to know about is accessible and small enough to be surveyed directly. For example, you may want the opinions of all the people in your bureau about their usage of email. In this case, you would send a survey to every person -- a 100% sample.

Gathering opinions and demographics. Surveys allow you to assess user, client, or stakeholder opinions and evaluations. You can also collect demographic descriptions of these groups.

Supporting qualitative analysis. You can obtain quantifiable results for statistical analysis of such subjective material as attitudes, opinions, and values.

Low cost data collection. Surveys require little time, cost, and commitment from the participants.

Keeping respondents anonymous. Like some other methods of information gathering, surveys allow you to preserve the anonymity of respondents. You're more likely to increase the number of responses when people know their opinions are anonymous.

Careful design and field testing. The questions must be carefully designed and field-tested. You must know what questions to ask and how to word them. Even slightly imprecise wording in the question or the answer choices can give results very different from the ones you seek.

The cost of design and analysis. While the cost of a survey is low for participants, a good survey design, execution, and analysis may be expensive. Questions must be carefully designed and the answers thoughtfully analyzed, both of which take expertise and time.

No chance to discuss answers. Surveys provide little or no opportunity to discuss, elaborate, or explain answers. Respondents who do not understand a question may answer improperly or not at all.

Potential problems with the sample. Biases or flaws in a sample can make it difficult or inappropriate to generalize about the larger population.

Absence of sampling. If you don't use statistical sampling, you can't generalize your findings beyond the group you actually contacted. This is fine when the group is relatively small and well known. For example, if you want to know what every person on your staff thinks about the new proposal, you can ask them all, and report your results without worrying about statistical validity.

American Association for Public Opinion Research (AAPOR). "Best Practices for Survey and Public Opinion Research." http://www.aapor.org/default.asp?page=survey_methods/standards_and_best_practices/ best_practices_for_survey_and_public_opinion_research [Retrieved June 9, 2003]

Council of American Survey Research Organizations (CASRO). "CASRO Guidelines for Survey Research Quality." http://www.casro.org/guidelines.cfm [Retrieved June 9, 2003].

Babbie, Earl R. (1973) Survey Research Methods. Belmont, CA: Wadsworth.

The Survey System had an overview for beginners of survey and questionnaire design at http://www.surveysystem.com/sdesign.htm [Retrieved July 7, 2003]