MATH 1532 Lecture Notes - Lecture 95: Null Hypothesis, Parametric Statistics, Type I And Type Ii Errors
9 Jul 2017
School
Department
Course
Professor
Document Summary
Note: please do not get discouraged by these errata. Most of the book is well written and correct. Chapter 2, descriptive statistics: page 11: the formula for range, which subtracts the minimum data value from the maximum, then adds one is non-standard. Most statisticians do not add one: page 18: the statement with a skewed distribution, the median will pro- vide a more appropriate indication of the typical score. is too strong. De- pending on the use of the typical score , the median may be more appro- priate that the mean. The statement that outliers are scores in the distribution that are more than. 1. 5 box lengths from the 25th or 75th percentile is misleading. This applies only to the conventional picture of a boxplot. A more general dis- cussion of what it means to be an outlier is much more complex. Spss de nes extreme values as those points that are more than 3 (not 4)
