Guidance for Establishing Cut-Points for Youth Fitness Tests
As mentioned above, setting cut-points for evaluating results of health-related fitness tests entails a number of challenges, one of which is the lack of appropriate data. Ideally, cut-points are established from data on performance on a specific fitness test and health outcomes in a broad-based youth population. More often, however, these data are not available; instead, there may be enough data on an association with health in the adult population but only growing evidence from studies in small samples of the youth population. In yet a different scenario, only growing evidence for an association between a specific test and health exists, with no data coming from broad populations. Until data in broad youth populations are generated so cut-points can be derived, cut-points should be referred to as interim criterion-referenced cut-points (or interim cut-points). Various approaches can be used to select these interim criterion-referenced cut-points, depending on the available data. The committee’s guidance on these approaches is presented in this section. Often a number of field tests are used simultaneously to measure the same fitness component. The reader is referred to Zhu et al. (2010) and Jackson (1989) for information on setting cut-points for multiple tests of a single component.
Several considerations apply in interpreting the results of fitness tests. For health-related fitness testing in youth, the key interest is not only whether a test taker is “fit enough” to be free of potential health risks but also whether the test taker is “fit enough for the future.” In addition, because the key outcome of interest of the criterion-referenced approach to evaluating test results is classification (e.g., being at risk of a health outcome versus not being at risk), the accuracy of the classification is key. Further, regardless of how well the related cut-point is established, it will be possible to misclassify individuals. There are two kinds of misclassification: (1) when a fit test taker is misclassified as unfit and (2) when an unfit test taker is misclassified as fit. The committee considered the first of these to be more problematic because it would result in a greater likelihood of recommending an exercise intervention to youth who do not need it, thereby depleting already limited resources that should be used for youth who need them the most. To minimize the effects of misclassification, cut-points need to be validated or cross-validated using additional measures and samples.
Finally, whether cut-points should be established differently for various subpopulations must be examined and determined empirically. As discussed earlier, while age and gender often have been taken into consideration in setting cut-points, many other factors, such as race/ethnicity, maturation status, and disability, have not been considered.
Once cut-points have been established for a specific test and age/gender group, they should be used in interpreting test results and communicating