Skip to main content
Fig. 1 | BMC Medical Research Methodology

Fig. 1

From: Utilizing distributional analytics and electronic records to assess timeliness of inpatient blood glucose monitoring in non-critical care wards

Fig. 1

P-values of Kolmogorov-Smirnov tests. The top panel displayed the boxplots of –log10(p-values) from the one-sample Kolmogorov-Smirnov (1S-KS), the middle panel displayed the minimum (min), median and maximum (max) daily number of blood glucose (BG) measurements over the 2-month period, and bottom panel displayed the boxplot of p-value from the two-sample Kolmogorov-Smirnov (2S-KS) tests. The boxplots were grouped together according to the medical specialties of the wards, where Orthopedic represented Orthopedic surgery, Cardio represented Cardiology, O&G represented Obstetrics and Gynaecology. Boxplots in white and dark gray shade corresponded to wards with mean or median power ≥ 90 % and <90 % respectively. Within each specialty, wards were ranked by the proportions of significant p-values and when there were ties the median p-values were used. The light gray region in the top and bottom panels corresponded to the region where the p-values were >0.05 and <0.05 respectively

Back to article page