ABSTRACT
Consider that we have a sample of blood pressure readings from one patient who has kept track of her blood pressure for a few weeks, and we notice that on average she had slightly elevated blood pressure readings. High blood pressure is a risk factor for stroke. But does the average (mean) blood pressure reading tell us everything we might need to know about this patient’s risk of stroke? Rothwell et al. (2010) reported that having a great deal of variation (spread) in blood pressure is another risk factor for stroke. The mean blood pressure is not a measure of variation; it is a measure of the location of the data on the number line. We need other statistics that will measure the spread in the blood pressure readings. We also may be interested in knowing whether some extreme readings are appearing in only one direction (i.e., a few extreme readings that are all high or all low), which may indicate the patient has episodic hypertension or hypotension. In that
Introduction Measures of Location Measures of Spread or
Variability
Measure of Skewness or Departure From
Symmetry What’s Next
case, we would need statistics that measure departure from symmetry. A symmetrical data set would have the same shape below and above the middle score, as we will show in this chapter. A departure from symmetry may be clinically noteworthy and influence other statistics.
