Academic Integrity: tutoring, explanations, and feedback — we don’t complete graded work or submit on a student’s behalf.

(a) Suppose a survey determines the political orientation of 60 men in a certain

ID: 3339784 • Letter: #

Question

(a) Suppose a survey determines the political orientation of 60 men in a certain community: Left Middle Right Men 12 18 30 60 Among these men, calculate the proportion belonging to each political category. Then show that a Chi-squared Test of the null hypothesis of equal proportions leads to its rejection at the = .05 significance level. Conclusion? (b) Suppose the survey also determines the political orientation of 540 women in the same communitv: Left Middle Right Women 108 162 270 540 Among these women, calculate the proportion belonging to each political category. How do these proportions compare with those in (a)? Show that a Chi-squared Test of the null hypothesis of equal proportions Ho: 1eft 1 women-Zyid I women-Right I women leads to its rejection at the = .05 significance level. Conclusion?

Explanation / Answer

Solutions:

Part a

First of all we have to find proportions of men belonging to each political category. Required proportions are given as below:

Left

Middle

Right

Total

Men

12

18

30

60

Proportion

12/60 = 0.2

18/60 = 0.3

30/60 = 0.5

1

Now, we have o use chi square test for testing the given null hypothesis of equal proportions.

Test statistic formula is given as below:

Chi square = [(O – E)^2/E]

Where O is observed frequencies and E is expected frequencies.

Table for calculation of test statistic is given as below:

O

E

(O - E)^2

(O - E)^2/E

Left

12

20

64

3.2

Middle

18

20

4

0.2

Right

30

20

100

5

Total

60

60

8.4

Chi square = [(O – E)^2/E] = 8.4

Degrees of freedom = n - 1 = 3 – 1 = 2

P-value = 0.014995577 (by using Chi square table)

We are given

= 0.05

P-value < = 0.05

So, we reject the null hypothesis of equal proportions.

The proportions of men for left, middle and right political categories are not same.

There is insufficient evidence to conclude that proportions of men for left, middle and right political categories are same.

Part b

Now, we have to perform same test for women. The required proportions and chi square test is given as below:

Left

Middle

Right

Total

Women

108

162

270

540

Proportion

0.2

0.3

0.5

1

O

E

(O - E)^2

(O - E)^2/E

Left

108

180

5184

28.8

Middle

162

180

324

1.8

Right

270

180

8100

45

Total

540

540

75.6

Chi square = [(O – E)^2/E] = 75.6

Degrees of freedom = n - 1 = 3 – 1 = 2

P-value = 0.00 (by using Chi square table)

We are given

= 0.05

P-value < = 0.05

So, we reject the null hypothesis of equal proportions.

The proportions of women for left, middle and right political categories are not same.

There is insufficient evidence to conclude that proportions of women for left, middle and right political categories are same.

Now, we have to check whether the two categorical variables such as gender and political category are independent from each other or not.

We assume 5% level of significance. = 0.05

The test statistic formula for this test is given as below:

Chi square = [(O – E)^2/E]

Where, O is observed frequencies and E is expected frequencies.

Expected frequencies are calculated as below:

E = Row total * Column total / Grand total

For checking the given hypothesis, table for calculations are given as below:

Observed Frequencies (O)

Political category

Gender

Left

Middle

Right

Total

Men

12

18

30

60

Women

108

162

270

540

Total

120

180

300

600

Expected Frequencies (E)

Political category

Gender

Left

Middle

Right

Total

Men

12

18

30

60

Women

108

162

270

540

Total

120

180

300

600

O - E

0

0

0

0

0

0

(O - E)^2/E

0

0

0

0

0

0

Chi square = [(O – E)^2/E] = 0.00

Number of rows = r = 2

Number of columns = c = 3

DF = (r – 1)*(c – 1) = (2 – 1) * (3 – 1) = 1*2 = 2

P-value = 1

P-value > = 0.05

So, we do not reject the null hypothesis.

P-value is 1, which means 100% acceptance at 5% level of significance.

We concluded that there is 100% acceptance that the proportions of men and women are independent from the political categories.

Left

Middle

Right

Total

Men

12

18

30

60

Proportion

12/60 = 0.2

18/60 = 0.3

30/60 = 0.5

1