Distribution questions test what percent lies within a range and how skew moves the mean.
Normal curve
About 68 percent of values lie within one standard deviation, 95 percent within two (more exactly 1.96) and 99.7 percent within three. A z-score is the number of standard deviations from the mean. In a normal distribution, mean equals median equals mode.
Skew
In a positive (right) skew, the tail is to the right and mean is greater than median greater than mode. In a negative (left) skew, the mean is pulled toward the tail and is lower than the median. The median is least affected by outliers.
Standard error
Standard error of the mean equals standard deviation divided by the square root of n. A 95 percent confidence interval is the mean plus or minus 1.96 standard errors. Larger samples shrink the standard error but not the standard deviation.
Keep going in biostatistics, epidemiology and ethics
- Screening and prevention: incidence, prevalence and prevention levels
- Survival analysis: Kaplan-Meier curves and hazard ratios
- Patient safety and quality improvement: errors, RCA and handoffs
- Medical law: malpractice, reporting duties and advance directives