Meta-analysis data of sex differences in physiological and subjective thermal responses

Data collected from articles studying steady-state thermal conditions without any localised heating or cooling and for which all the environmental and personal conditions, i.e. the mean observed values of the air temperature (TA), mean radiant temperature (TMR), relative humidity (RH), air velocity (VA), participants' clothing insulation in clo and metabolic rate in met were reported. All the studies surveyed the thermal sensation vote on the classical ASHRAE seven-point scale: “Hot” (+3), “Warm” (+2), “Slightly Warm” (+1), “Neutral” (0), “Slightly Cool” (-1), “Cool” (-2), and “Cold” (-3), except for Yang and colleagues, that used an extended nine-point scale. We calculated the Fanger PMV index for each exposure, expressed on the ASHRAE seven-point thermal sensation scale. All the studies involved at-rest activities of an office-type (reading, typing, etc.), so the metabolic rate was always set equal to 1.1 met based on standard tabulated values. When the clothing insulation was not explicitly reported but a description of the clothing was provided, it was estimated using standard tabulated values as well. For each article, we extracted the sample size, mean, and standard deviation of thermal sensation, mean skin temperature, and core body temperature for female and male groups. Data were taken from the main text, tables, and figures. Mean and standard deviation data presented in the articles' figures were manually extracted using the freely available software program Plot Digitizer (http://plotdigitizer. sourceforge.net/). Data from seven studies were directly provided by the authors upon request by email. We used the raw mean differences between female and male thermal sensation, mean skin temperature, and core body temperature data as effect size metrics. In both cases, a difference of 0.0 represents no difference between the sexes. In the different studies, mean skin temperature was calculated using different weighting formulae from different skin locations and core body temperature was measured on various body parts (ear, oesophageal or rectal temperature). Many of the considered studies provide more than one effect size estimate, so their effect sizes are not independent.