Data analysis
Returned questionnaires were coded and the information was entered into a computer using a standard statistical software package—Statistical Analysis System (SAS). Frequency distribution and cross-tabulations were calculated for the coded data, and responses to open-ended questions were categorized and summarized.
Sampling size, missing data, and reporting items
This study collected information on both visitor groups and individual group members. Thus, the sample size ("N") varies from figure to figure. For example, while Figure 1 shows information for 516 visitor groups, Figure 3 presents data for 1,625 individuals. A note above each graph or table specifies the information illustrated.
Occasionally, a respondent may not have answered all of the questions, or may have answered some incorrectly. Unanswered questions result in missing data and cause the number in the sample to vary from figure to figure. For example, although Joshua Tree NP visitors returned 525 questionnaires, Figure 1 shows data for only 516 respondents.
Questions answered incorrectly due to carelessness, misunderstanding directions, and so forth turn up in the data as reporting errors. These create small data inconsistencies.
Limitations
Like all surveys, this study has limitations that should be considered when interpreting the results.
- It is not possible to know whether visitor responses reflect actual behavior. This disadvantage applies to all such studies and is reduced by having visitors fill out the questionnaire soon after they visit the park.
- The data reflect visitor use patterns of visitors to the selected sites during the study period of April 3-9, 2004. The results do not necessarily apply to visitors during other times of the year.
- Caution is advised when interpreting any data with a sample size of less than 30, as the results may be unreliable. Whenever the sample size is less than 30, the word "CAUTION!" is included in the graph, figure, or table.
3
