Content Validity Of Visual Analog Scales For Exertion And Comfort During Exercise In The Heat

PURPOSE: Subjective, Likert-type scales are ubiquitous in exercise physiology research. However, many argue ordinal data create measurement and evaluation issues with parametric analysis and recall bias. Visual analog scales (VAS) provide an alternative tool to assess subjective measures related to exercise. This study aimed to evaluate the validity of a visual analog scale for session ratings of perceived exertion (SRPE - VSRPE) and ratings of thermal comfort (RTC - VRTC) to discriminate between two exercise intensities in the heat. METHODS: Ten (n = 10; female = 4) participants (aged 24 ± 3 y) completed 20 minutes of walking at 3 METs or 6 METs (69.68 m·min-1 at 0% or 8.5% grade) followed by 10 minutes of seated recovery in counter-balanced, cross-over design within an environmental chamber (35.9 ± 1.9 0C; 34.9 ± 3.1% RH). VAS scales were created using 100 mm lines anchored with two extremes “No Exertion” and “Maximal Exertion” (for VSRPE) and “Very Comfortable” and “Very Uncomfortable” (for VRTC). Participants were instructed to draw a vertical line intersecting at a point that best represented where they fell on the spectrum for exertion and comfort during the protocol. A paired-samples, one-tailed t-Test was used to investigate mean differences between session measures of the 3-MET and 6-MET protocols for VRPE and VRTC. Continuous variables are reported as means ± SD. RESULTS: VSRPE was different after the protocol (exercise + recovery) between 3 METs (27 ± 27 mm) and 6 METs (45 ± 17 mm; p = 0.03, d = 0.75). Similarly, VRTC differed between 3 METs (22 ± 17 mm) and 6 METs (37 ± 25 mm; p = 0.002, d = 0.70). CONCLUSIONS: Both VAS scales were able to demonstrate a sensitivity to detect increased session RPE or session RTC at differing exercise intensities in the heat. Taken together, these data support the use of VRPE and VRTC in applications where researchers or practitioners would like to have continuous data that limits bias in serial measures.