Cover Image

Not for Sale

View/Hide Left Panel
Click for next page ( 136

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 135
Surveys of Area Residents 135 advance based on the available budget, assumptions about the importance of the results, and the risk of making wrong decisions based on the findings. Customarily in opinion research, the confidence level is fixed at 95%. The confidence interval or margin of error is then stipulated based on the factors outlined in the previous paragraph. For public opinion research, the margin of error is most often fixed at 5 percentage points, leading to a sample size of 400 (rounded up from 384). A margin of 3 percentage points requiring a sample size of 1,000 is also common. Larger sample sizes are utilized for high-risk projects and when sub- group analysis will be conducted. Commercial market researchers frequently use a 6 percentage point margin of error and thus a sample size of 300. Selection of sample sizes and the associated errors and confidence intervals are discussed in detail in Section 3.4, and further examples for deter- mining the required sample size are provided in Appendix B. 8.4 Questionnaire Wording and Length Experience suggests that surveys of the general public will start to experience unacceptable rates of refusals and terminations (people who quit in the middle of the interview) when the interview goes past about 10 minutes. Cooperation rates are highest when interviews are five minutes or less, although it is admittedly difficult to craft such a short survey on most topics. A detailed discussion of question wording issues can be found in Section 4.3. There are nuances in how questions are worded across methods (e.g., the phrase "the following list" works well for self-completed questionnaires but sounds nonsensical over the telephone), but the fundamental principles are the same for all methods. A sample questionnaire for area residents is provided in Appendix J. 8.5 Measures to Obtain Adequate Response As pointed out in previous sections, obtaining an adequate response is key to a survey's success. A low response rate leads to questionable results. Important aspects in conducting a telephone survey include the following: A centralized facility where interviewers are closely supervised and regularly monitored, ideally during every shift. A thorough interviewer training program. A comprehensive and proactive approach to coaching interviewers whose skills need improve- ment or who are doing something wrong during their interviews. Generally, data quality will be superior if a computer-assisted telephone interviewing (CATI) system is used, although this is not necessarily the case. However, complicated questionnaires with many skips and branches should always be done on CATI systems; they are simply too difficult for interviewers to follow correctly on paper. Finally, there is the issue of languages. There are some communities in which interviewing needs to be conducted in a language other than English. However, even in areas with high percentages of people who speak other languages, many of these people also speak English, and some would pre- fer to be interviewed in English to show they are trying to learn, even if the interview takes longer as a result. It is therefore wise to ask the call center what their experience is when questionnaires are translated; often the translation is not worth the time and cost. It is also important to note that interviewing in other languages requires a written translation of the questionnaire in addition to bilingual interviewers. If interviewers are simply asked to translate