Test validity

Test validity is the extent to which a test measures what it is supposed to measure.^{1Schoenfeld and Snarr (eds.), NSCA’s Essentials of Personal Training, 3rd ed., p. 192.} A 50-yard sprint can’t validly measure aerobic capacity, for example: the test must be long and intense enough to rely mainly on aerobic energy.^{1Schoenfeld and Snarr (eds.), NSCA’s Essentials of Personal Training, 3rd ed., p. 192.} The NSCA describes BMI as a relevant indicator of overweight in a fairly sedentary group, but not in athletes with extra lean muscle and low body fat.^{1Schoenfeld and Snarr (eds.), NSCA’s Essentials of Personal Training, 3rd ed., p. 192.}

The NSCA’s strength and conditioning text treats construct validity as the overall measure, supported by face, content and criterion-referenced validity.^{2Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 267.} Content validity is an expert judgment that a test covers all relevant abilities; a soccer test battery should include sprinting, agility, endurance and kicking power.^{2Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 267.} Criterion-referenced validity describes how closely scores track another measure of the same ability, such as a new body fat device compared with DXA.^{2Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 267.}

In practice

• Reliability doesn’t establish validity.^{3Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 268.} The 60 m dash and the 1.5-mile run are both reliable, but only the run is a valid field test of aerobic fitness.^{3Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 268.}
• The NSCA recommends a high correlation between a field test and the lab gold standard.^{2Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 267.} A field test can be preferable when it takes less time, equipment, expense or expertise.^{2Haff and Triplett (eds.), Essentials of Strength Training and Conditioning, 4th ed., p. 267.}
• Submaximal tests are statistically compared with maximal tests, but their results don’t precisely match the maximal test or each other.^{1Schoenfeld and Snarr (eds.), NSCA’s Essentials of Personal Training, 3rd ed., p. 192.} The Rockport walk equation has a standard error of 5.0 mL/kg/min.^{4Ozemek et al., ACSM’s Guidelines for Exercise Testing and Prescription, 12th ed., p. 217.}
• ACSM no longer recommends the sit-and-reach for hamstring or low back flexibility, citing questionable validity.^{5Ozemek et al., ACSM’s Guidelines for Exercise Testing and Prescription, 12th ed., p. 251.}
• ACSM requires health-related fitness tests to be both reliable and valid.^{6Ozemek et al., ACSM’s Guidelines for Exercise Testing and Prescription, 12th ed., p. 166.}