HPED 2030 Lecture Notes - Lecture 10: Frequency Distribution, Null Hypothesis, Chi-Squared Distribution
Document Summary
Assumptions underlying parametric tests can"t be met. Used to compare how well an observed frequency distribution fits an expected distribution. Population parameters do not need to be known. Can be used for two or more groups. Negatives: less powerful, more difficult to reject ho when ho is false, more difficult to avoid a type i error. Appropriate use & assumptions for use of chi-square 2. 3. data must be frequency count (not percentages/mean/etc) observations independent of one another (one category) expected and observed frequencies have to equal one another the sample size must be adequate. If frequency is less than 5 in any cell, chi. Frequencies expected in the sample if the sample data represent the population. Degrees of freedom df = r-1 ( 2 test) Random sample of 80 students from sunder high. 7 students from sunder high have been diagnosed as alcoholics.