Differential Effects of Time Constraints on Athletic Behavior and Survey Reports of Athletic Behavior

ABSTRACT

Time constraints—having more to do than time to do it—can prevent us from doing everything we need and want to do. When lack of time constrains our behavior, the behavior linked to voluntary role identities (like being a member of a softball team or bowling league) may be the first to be cut. Although, as many of these role identities are socially desirable, survey respondents may still claim to have performed them. Thus, this study examines role behavior and its measurement in the face of time constraints. The athlete identity is examined as a potential casualty of the time crunch, used because it is a common, typically voluntary identity that has a relatively standard set of role behaviors (e.g., participating in sporting events, like games or matches, as an athlete). Situational constraints, namely, a lack of time, are brought into a model based on the structure posited in identity theory (Stryker [1980] 2003) to help explain variation between self-reported and actual role behavior. Thus, the current study examines the extent to which feeling pressed for time reduces actual athletic activity but fails to result in a concomitant reduction in self-reported athletic activity.

Notes

¹Time and money are, of course, not the only constraints. Other constraints are possible, such as losing the ability to perform an identity given the constraints of illness or injury (McCall 2016).

²Notably, the cost of survey responding can be negative as most academic surveys provide monetary incentive payments to thank or at least partially compensate respondents for their time.

³Such a gender-based effect has been found elsewhere for other normative behavior (see Brenner 2014).

⁴The analytic sample is limited to those 18 years of age or older given the different, and potentially obligatory, nature of sports participation in secondary schools.

⁵Sports were coded from respondents’ verbatim reports, including (Canadian gridiron) football, field hockey, baseball or softball, soccer, volleyball, hockey, basketball, racquet sports (e.g., squash, tennis), golf, miniature golf, water sports (e.g., swimming, polo), snow sports (e.g., downhill and cross-country skiing), ice sports (e.g., skating, curling), bowling, roller sports (e.g., in-line skating), fighting sports (e.g., boxing, fencing), cycling, and other sports. These may include community-based, school-organized, and other amateur leagues and professional sports.

⁶These six items are a subset of a larger series of 10 questions. The four excluded questions focused more specifically on the main outcome of the time crunch—stress about the lack of time—rather than the respondent’s perception of the time crunch itself.

⁷The interaction of constraints and salience on role behavior was also tested. As it is negligible in size and not statistically significant, it is not included in the presented models.

⁸Allison (1999) strongly suggested two tests of the hypothesis of equal residual variation between group models before using this test to compare logits across groups. Neither of these tests reach conventional levels of statistical significance—likelihood ratio (χ² = 2.06, p = .15) and Wald (χ² = 1.64, p = .20). Thus, the null hypothesis of equal residual variation is not rejected and the comparison of logits is deemed appropriate. The potential perils of this approach is discussed in the Discussion section of this article.

⁹Male pronouns are used here given the findings of this study.

Additional information

Notes on contributors

Philip S. Brenner

Philip S. Brenner is Assistant Professor of Sociology, Senior Research Fellow in the Center for Survey Research at the University of Massachusetts Boston. His research examines social desirability bias and other measurement errors in interviewer- and self-administered surveys.