Effects of different combinations of intensity categories on self-reported exercise

Self-reports of exercise are used extensively in behavioral, social psychological, and epidemiological research (Ainsworth, Montoye, and Leon, 1994; Caspersen, 1997). Schwarz (1999) noted that many characteristics strongly influence self-reports of behavior, including question wording, format, and context. Of particular interest in the present study is the possible effect of providing different combinations of intensity categories (i.e., light/mild, moderate, and vigorous/strenuous) on self-reported exercise. A review of the exercise measurement literature indicates that researcher-developed and published questionnaires have varied in the number of exercise intensity categories they present to respondents. For example, researcher-developed questionnaires have often used only one category of exercise intensity, such as moderate (e.g., Miller, Trost, and Brown, 2002; Wallace, Buckworth, Kirby, and Sherman, 2000) or vigorous (e.g., Owen, Sedgwick, and Davies, 1988; Washburn, Goldfield, Smith, and McKinlay, 1990). Conversely, published questionnaires have typically used multiple intensity categories, such as moderate and vigorous/strenuous (e.g., Blair et al., 1985; Heath, Pate, and Pratt, 1993) or light/mild, moderate, and vigorous/strenuous (e.g., Baecke, Burema, and Frijters 1982; Godin and Shephard, 1985; Myers, Bader, Madhavan, and Froelicher, 2001). It is unknown, however, if providing different combinations of exercise intensity categories has any effect on the amount of exercise reported in a given intensity category or in total.