Purpose
The purpose of this study was to develop a scale to measure Outdoor Recreation Self-Efficacy (ORSE), and to assess the reliability of the overall scale and sub-scales. Bandura defines self-efficacy as, "beliefs in one's capabilities to execute courses of action required to produce given attainments¨ (1997, p. 3). Self-efficacy theory is salient to a variety of recreation domains. Items in the ORSE scale represent concepts central to self-efficacy and were developed with outdoor recreation activities in mind.
Methods
Originally, during phase one, the ORSE scale had 19 items, and was part of a 5-page survey to measure the impact of the Becoming an Outdoors Woman (BOW) program. The survey was sent to 2,500 women in Ohio; 546 were returned (21.8%). Of 508 usable surveys, 300 women (55%) had attended BOW and 246 (45%) had not. About 600 women attended BOW between 2000-2005, representing a 50% response rate.
Analysis/Results
Kaiser-Meyer-Olkin (KMO) and Bartlett's tests were conducted to check for normality and sampling adequacy. Bartlett's test [X2 = 9069.03 (df=171), p = .000], and KMO test results (.95) indicated these data were "marvelous" for conducting a factor analysis. Exploratory factor analysis revealed two sub-scales. A 10-item Enjoyment/Accomplishment subscale accounted for 57.19% of explained variance and a 9-item Skills/Competence subscale accounted for 11.75%, explaining a total of 68.94% of the total variance. The overall reliability was quite high (alpha=.95) and for each of the subscales (Enjoyment/Accomplishment alpha=.95; Skill/Competence alpha =.92). During phase two of the study, two items were deleted because wording seemed confusing. The item "empowered¨ was added. This revised 18-item scale was administered to students enrolled in recreation courses at a Midwestern university at two times, one week apart. The overall reliability for Time1 (alpha=.96) and Time2 (alpha=.95) provides further validation of the ORSE scale with a different population with regard to gender and age. The test-retest reliability (r=.94) was also clearly established indicating a high degree of internal consistency.
Conclusions
The ORSE scale provided reliable results for mostly middle-aged Caucasian women, and for male and female college-age students (M=20.8), also mostly Caucasian. Future studies will focus on: (a) groups and settings more diverse in terms of ethnicity, race, and social class; (b) providing evidence for its validity by examining the relationship between the ORSE and other theoretical constructs and scales; and (c) assessing one outdoor recreation activity at a time, such as rock-climbing or kayaking.