Athletic testing 101 part two

     In part one we discussed the validity of athletic tests and underlying issues of validity. On the other side of the coin in terms of testing is reliability. Reliability is simply the repeatability of a test, or the consistency built around the test. Consider the NFL combine and the 40-yard dash. The 40-yard dash is conducted indoors with electronic and manual timers. If you can remember the 40 passes validity of maximum speed (if you're not familiar with the term validity in terms of testing, click on the link and check it out). It is also a repeatable test since it is consistent for each player. Now, consider if the skill players performed their 40 indoors and the lineman conducted their test outdoors. I have never been to Indy during February but I'm willing to bet that it is cold and possibly snowy. That would undoubtably throw off the results creating an unreliable test. The results would mean nothing. It is possible to have a reliable test and it is not valid. In the example of the combine if the 40 was changed to the 200m run then it would be reliable for speed but not valid for football. For those unfamiliar with American football the entire field is 100 yards, so a speed test that doubles the entire length of the field would not be valid. A baseline measurement of reliable tests is called test-retest reliability that of course is measured statistically. The history of the 40 tells us that 4.3 speed is really fast. We know that because the test has been proven over the years. Now if the electronic timer breaks during the 40 then that would be typical error of measurement. The test itself is valid and reliable but equipment does fail. We used an electronic timer for our two-mile run and in some cases, it froze up. We also had handheld timers just in the event that the timer does freeze up.

    Intrasubject reliability pertains to the athlete themselves. Are the results consistent? If a player runs a 4.4 then runs a 4.9 then that would be an example of intrasubject reliability. Interrater reliability is how raters are consistent and agreeable with the results. A trained, experienced coach watches the athletes before the start of the 40 and is consistent with actions for the start of the 40. It is widely accepted that the coach for the 40 will be consistent with the application of the rules. If that coach hypes up certain players for the 40 then ho hum for other would interfere with the results of the test. Intrarater would be the lack of consistency of the coach for the 40-yard dash. If the coach allows different start techniques for different athletes, then that would be a violation of intrarater variability. Or if the coach for the 225-bench press allows one athlete to perform half reps as opposed to full reps that would be another example of intrarater variability.

    Point being that if you're going to have athletes perform athletic tests then ensure that they are valid for the sport and are reliable.