Cover Image

Not for Sale

View/Hide Left Panel
Click for next page ( 140

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 139
Surveys of Area Businesses 139 9.3 Sampling Methodology The first step in determining the sampling methodology will be to define the population of interest. Is it all businesses in the area, a particular industry segment, or just businesses that make use of the airport? The sampling method and sample size will depend on the survey method used. Using the Internet method in a small city, it may be possible to survey all businesses for which email addresses are available, but generally only a sample will be surveyed. The low response rates for Internet and mail surveys need to be taken into account when determining the sample size. For telephone or in-person interviews, the budget will usually only permit a survey of a pro- portion of area businesses. Given the wide range of businesses and their varying air travel requirements, businesses should be grouped and a stratified sample of businesses selected to sur- vey. Stratification should be based, if possible, on the type of business and number of local employees or some other measure of the size of the business. Unfortunately, data on businesses from business organizations do not always include a measure of the size of the business. If pos- sible, categories of businesses should be chosen so that they are comparable to categories used by other organizations or government departments (e.g., the North American Industry Classi- fication System codes of the U.S. Census Bureau) to allow comparison and sharing of data. Where stratified sampling is used, the proportion of businesses in each stratum (group) will gen- erally not be the same for all strata. As discussed in Section 3.3, the sampling fraction should be greater in strata with higher variation in the variables of interest within the stratum. For example, if interested in the numbers of air trips by employees of a business, the variation in the numbers of trips per employee will vary by type of business. Variation for consulting firms is likely to be much greater than for retail firms. Thus, for determining the total air trips by employees in the area, the proportion of consulting companies in the area surveyed should be higher than the proportion of retail firms. (Appendix B provides information on determining the sample size for each stratum.) Alternatively, sample sizes for each stratum could be set so that the level of accuracy in the variable of interest is similar in each stratum. For a categorical variable using random sampling within a stratum, Table 3-3 could be used to determine the sample size for each stratum. With strata based on the size of the business, there will be few very large businesses and many very small businesses and so a much higher proportion of the very large businesses will be required than of the very small businesses to obtain a similar level of accuracy.30 9.4 Questionnaire Wording and Length Questionnaires for surveys of area businesses should be kept short to maximize response rates. The time required to complete the questionnaire should not exceed about 20 minutes for busi- nesses where air travel is very important to the business and about 10 minutes in other cases. The questionnaire should be structured so that businesses that rarely or never use air service will only answer the first few questions. For Internet surveys, these questions could even be included in the email message so that businesses can respond with this information very easily. The questionnaire format, structure, and clarity, and the use of pre-tests are important con- siderations, as discussed in Chapter 2. To keep the questionnaire as short as possible, it is very important to match the questions to the purpose and goals of the survey, eliminating "nice to know" questions and focusing only on "need to know" information. 30For example, if there are 10,000 businesses with fewer than 10 employees and 50 with more than 500 employees, to obtain similar accuracy in the two groups for a categorical question, 370 of the 10,000 very small businesses (3.7%) and 45 of the 50 large businesses (90%) would have to be surveyed to give an accuracy of 5 percentage points.