Children's Perspectives on Exercise and Physical Activity: Measurement Issues and Concerns
Brustad, Robert J., Journal of School Health
The apparent decline in children's physical fitness  represents an important national concern. Enhanced understanding of children's perspectives on exercise and physical activity could increase knowledge about the role of attitudinal and belief characteristics to children's activity patterns. This article reviews measures of children's perspectives on physical activity, identifies current measurement issues, and suggests future directions in this area of research.
Processes by which children form attitudes and values about involvement in physical activity are likely to differ from those used in developing perspectives on other health-promoting behaviors for at least two reasons. From an early age, children have abundant opportunities to engage in physical activity, games, and sport. Consequently, attitudes toward these activities may be fairly well developed during early childhood and likely will precede awareness of health-related benefits from physical activity. Second, development of physical capacities and skills is a highly valued component of child and adolescent status systems. 
Measurement of children's physical activity-related beliefs also is more complex than measurement of other health-related beliefs and behaviors. Physical activity is, by nature, a multidimensional experience.  A child may associate the terms "physical activity" or "exercise" with personal experiences that have emphasized exertion, competition, play, social interaction, or skill development. Furthermore, children's perspectives will undergo change and refinement as a consequence of experience and maturation.
REVIEW OF MEASUREMENT APPROACHES
Children's perspectives on exercise and physical activity have been assessed only within the past 15 years. Much of this research has been conducted by Schutz and Smoll. [4-6] Their efforts were inspired by Kenyon's  work that identified six factors that attract adults to physical activity: social aspects, health and fitness aspects, vertigo, or physical experiences providing some risk or physical excitement, aesthetic elements such as graceful or beautiful movements, catharsis or release of tension, and ascetic aspects such as activities involving difficult or strenuous training. Kenyon used this information to develop a measure for assessment of adults' attitudes toward physical activity (ATPA). 
The Children's Attitudes Toward Physical Activity (CATPA) Scale, developed by Simon and Smoll  in 1974, was patterned directly after the adult scale. CATPA maintained the same six attitudinal dimensions, modifying only the wording to make it more appropriate for children. Unfortunately, Kenyon's six factors, generated entirely through research with adults, were never validated as representative of children's orientations toward physical activity. Aspects of the adult inventory do not intuitively appear to reflect attractive features of physical activity for children. For example, the ascetic dimension that entails long and arduous training seems unlikely to represent an appealing aspect of physical activity for children.
CATPA has been the most frequently used measure for examining attitudinal characteristics of youngsters regarding physical activity and sport and for studying relationships between attitudes and physical activity-related behaviors. Schutz et al  also devoted considerable effort to identifying CATPA's psychometric properties.
The value of CATPA is limited by two psychometric concerns. The scale's validity can be questioned since the six factors were generated from adults and not children. Second, CATPA was intended to measure enduring or "trait-like" aspects of children's attitudes; yet, its relatively poor test-retest reliability of .60 over a six-week period for elementary school students  suggests that this intention was not realized. Consequently, the authors suggested CATPA be used neither to examine attitude-behavior relationships over time, nor for research involving individual difference variables. …