Question

Sex	Sports	Popularity	Grades
Boys	60	50	117
Girls	30	91	130

Sports

Popularity

Grades

Proportion of boys who selected each category.	0.2643171806	0.2202643172	0.5154185022
Proportion of girls who selected each category	0.1195219124	0.3625498008	0.5179282869

1)if goals were contingent on district type. Was there any difference in the goals of students from rural, urban, or suburban areas?

Create an observed frequency table using the raw data and calculate Χ². Χ² =

Hint: I counted up the observed frequencies for the Sex X Goals contingency table and gave you a table that summarized the results. Now you will need to do this to make a District X Goals contingency table. Use the data in the 'Raw Data' worksheet to make your table. Most spreadsheet programs will tell you the total number of cells selected at the bottom or in a pop-up box near the cursor. If you sort all the data by district and then use this feature it will be much faster than actually counting every cell by hand. If you can't find it then you can also use the =COUNT() function and select all the cells for a district. If you run into any trouble please let me know.

2)How many degrees of freedom do you have for this test?

3)Based on our results, we should:

		reject the null hypothesis; goals were contingent on district type
		fail to reject the null hypothesis; goals were not contingent on district type

4)Chase and Dummer claimed that approximately 1/3 of their sample came from each district type. Let's use a goodness-of-fit test to see if there was a significant difference in the number of children from each district. Based on our results we can say:

Hint 1: your observed frequencies for this chi-square should be your row or column totals (depending on how you organized your table) for the last test

Hint 2: for goodness of fit you have to use the bottom table on the website

		There was no significant difference in the number of children from each district.
		There was a significant difference, but we can't tell which district was significant different from other districts.
		There was significantly more children from the Urban district than from any other district.
		There was significantly less children from the Rural district than any other district.

5)

If we have multiple IVs and we want to know if one IV is contingent on another IV then the Chi-square variant we need to do is called:

		within-subjects
		goodness-of-fit
		contingency-tables
		none of the above

Answer 1

Rows: C1   Columns: Worksheet columns

	sports	popularity	grades	All
Boys	60	50	117	227
Expected count	42.74	66.96	117.30
Girls	30	91	130	251
Expected count	47.26	74.04	129.70
All	90	141	247	478

Cell Contents
      Count
      Expected count

Chi-Square Test

	Chi-Square	DF	P-Value
Pearson	21.455	2	0.000
Likelihood Ratio	21.769	2	0.000

1.

2. Df = (No. of rows-1)*(No. of columns -1) = 2*1 =2

3. reject the null hypothesis; goals were contingent on district type

Reason = P-value is less than 0.05