Evidence For Construct Validity Of Scale Research Articles

Development of a Questionnaire Assessing School Physical Activity Environment

This study was designed to develop the Questionnaire Assessing School Physical Activity Environment (Q--SPACE) based on student perceptions. Twenty-eight items rated on 4-point Likert scales were administered to 244 middle school students in 9 schools. Exploratory factor analysis was used to evaluate the underlying structure of the items and 2 factors were extracted: physical environment (PE) and social environment (SE). Twelve and 8 items loaded saliently on PE (e.g., gym classes available) and SE (e.g., teacher encouragement), respectively. Factor scales had alpha coefficients of .86 (PE) and .81 (SE). One-week test–retest reliabilities for the factor scales of PE and SE were .78 and .72, respectively. Differences in PE scale scores across schools and PE and SE scale scores across student school physical activity behavior (e.g., participation on school teams) provided some evidence of scale construct validity. Overall, Q–SPACE demonstrates acceptable reliability for capturing middle school students' perceptions of school physical activity environment. The factorial validity needs to be assessed with confirmatory factor analysis and invariance testing procedures.

Creation and Validation of a Scale Measuring Perceived Control Over the Institutional Environment

This article describes the development and validation of a 14-item, self-report measure of residents' perception of control over the institutional environment. Two studies of residents from a variety of institutional settings provided evidence of reliability and validity. Evidence of scale construct validity included the finding that residents living in a less restricted setting had higher Perceived Control Scale scores than residents of a more restricted setting. Another set of validity findings showed that the Perceived Control Scale was significantly associated in the expected direction with measures of locus of control, self-esteem, fatigue, vigor, and depression. The prediction that nurses' ratings of hospital patients' health would be related to the patients' scale scores was not confirmed. An intervention study showed that after participating in a workshop designed to increase perceived control over the institutional environment, residents of a retirement facility had significantly higher Perceived Control Scale scores than residents in a control group. The individuals participating in the workshop also had significantly better scores on theoretically related measures of well-being such as depression, vigor, and life satisfaction.

Development and evaluation of a measure of the tendency to be goal oriented

This paper describes the development and evaluation of a 15-item, self-report measure of the tendency to set goals and make plans. A series of studies provided evidence of reliability and validity. The internal consistency of the scale ranged from 0.81 to 0.83; 3-week test-retest reliability was 0.82. Evidence of scale construct validity included significant associations in the expected direction with measures of several theoretically related constructs, including the number of goals one has, impulsivity, need for achievement, instability, hopelessness, depression, and official GPA. Another validity finding indicated that college students receiving psychotherapy had significantly lower goal orientation as measured by the scale than other college students. A treatment study produced evidence that goal orientation and official Grade Point Average (GPA) increased significantly in university counseling center clients who received training in goal setting and planning. The results provide initial evidence that the goal-orientation scale has promise as a measure of a psychologically significant construct.