ChiSquare Tests GoodnessofFit Test for Independence Test for

  • Slides: 9
Download presentation
Chi-Square Tests Goodness-of-Fit Test for Independence Test for Equality of Several Proportions

Chi-Square Tests Goodness-of-Fit Test for Independence Test for Equality of Several Proportions

Chi-square Test Statistic p p Reject Ho if The degrees of freedom depends on

Chi-square Test Statistic p p Reject Ho if The degrees of freedom depends on the application Chi-square Tests : PLBautista

Goodness-of-fit Tests We wish to test if the data follows a specified distribution. p

Goodness-of-fit Tests We wish to test if the data follows a specified distribution. p Ho: The data follows the specified distribution. p H 1: The data does not follow the specified distribution. p df=k-1, where k is the number of cells p Chi-square Tests : PLBautista

Example p Consider the experiment of tossing a sixsided die. We wish to test

Example p Consider the experiment of tossing a sixsided die. We wish to test if the die is balanced (that is, that the distribution of our data is uniform). We then toss the coin 120 times and obtain the following data: p Test if the die is balanced using a 0. 05 level of significance. Chi-square Tests : PLBautista

Test for Independence We wish to test if two variables are independent of one

Test for Independence We wish to test if two variables are independent of one another p Ho: Variable A and Variable B are independent. p H 1: Variable A and Variable B are not independent. p df = (r-1)(c-1), where r is the number of rows and c is the number of columns p Chi-square Tests : PLBautista

Example p Suppose we wish to test if the presence or absence of hypertension

Example p Suppose we wish to test if the presence or absence of hypertension is independent of one’s smoking habits. We obtain the following data from a sample of 180 individuals: p Test if the presence or absence of hypertension is independent of a person’s smoking habits using a 0. 05 level of significance. Chi-square Tests : PLBautista

Test for Equality of Several Proportions We wish to test if the proportion of

Test for Equality of Several Proportions We wish to test if the proportion of success is equal for more than two populations p Ho: p 1=p 2=…=pk p H 1: at least one pair is not equal p Like the test for independence with r=2 p df = k-1, where k is the number of populations p Chi-square Tests : PLBautista

Example p In a shop study, a set of data was collected to determine

Example p In a shop study, a set of data was collected to determine whether or not the proportion of defectives produced by workers was the same for the day, evening, or night shift worked. The following data were collected: p Use a 0. 025 level of significance to determine if the proportion of defectives is the same for all three shifts. Chi-square Tests : PLBautista

Exercise p In a study to estimate the proportion of wives who regularly watch

Exercise p In a study to estimate the proportion of wives who regularly watch soap operas, it is found that 48 of 200 wives in Denver, 29 of 150 wives in Phoenix, and 35 out of 150 wives in Rochester watch at least one soap opera. Use a 0. 05 level of significance to test the hypothesis that there is no difference between the true proportions of wives who watch soap operas in these three cities. Chi-square Tests : PLBautista