Statistics Notes: Detecting skewness from summary information

The measurement of airway resistance by the interrupter technique (R_int) needs standardization. Should measurements be made be during the expiratory or inspiratory phase of tidal breathing? In reported studies, the measurement of R_int has been calculated as the median or mean of a small number of values, is there an important difference? Subjects were 2.5–5.0 yrs (median 4.0 yrs) who had previous respiratory symptoms. The R_int in expiration (R_intE) and inspiration (R_intI) pre and postsalbutamol, the coefficient of variation (CV) of values contributing to measurements, and bronchodilator responsiveness (BDR) in both phases were compared. Measurements using median and mean were compared. R_intE was higher than R_intI by 4% (p<0.01). The CV of values making up R_intE and R_intI, and BDR measured in expiration and inspiration were similar. The median difference between means and medians of values making up measurements was 0.6% (range −6–11%). R_intE has been shown to be consistently greater then R_intI but the difference in this study is small. It is suggested that one or the other is chosen as the standard. In the present data the mean of a set of values contributing to a measurement was not significantly different from the median. However, the use of the median has been recommended since it is less affected by possible outlying values such as might be included by fully automated equipment.