Question

A sociologist will conduct a two-sample tt-test for a difference in means to investigate whether there is a significant difference, on average, between the salaries of people with bachelorâs degrees and people with masterâs degrees. From a random sample of 32 people with a bachelorâs degree, the average salary was $55,000 with standard deviation $3,500. From a random sample of 28 people with a masterâs degree, the average salary was $58,000 with a standard deviation of $4,000. With a null hypothesis of no difference in the means, what is the test statistic for the appropriate test to investigate whether there is a difference in population means (masterâs degree minus bachelorâs degree)?

Answer 1

1) H0: μ₁=μ₂ against the claim Ha: μ₁ ≠μ₂

2)The test statistic is

t= x1`-x2`/ Sp √ ( 1/n₁ + 1/n₂)

where

Sp= s₁² ( n1-1) + s₂²(n2-1) /(n1-1) + (n2-1)

3) D.f= n1+n2-2= 32+28-2= 58

Explanation:

The null hypothesis is that there is no difference in population means (masterâs degree minus bachelorâs degree)

The claim is that there there is a difference in population means (masterâs degree minus bachelorâs degree)

1) H0: μ₁=μ₂ against the claim Ha: μ₁ ≠μ₂

2)The test statistic is

t= x1`-x2`/ Sp √ ( 1/n₁ + 1/n₂)

where

Sp= s₁² ( n1-1) + s₂²(n2-1) /(n1-1) + (n2-1)

Putting the values

Sp= 12,250,000( 31) + 16,000,000(27)/ 31+27

Sp= 379,750,000- 432,000,000/58

Sp= -52,250,000/58

Sp= -900,862.1

Now Putting the values in the t- test

t= x1`-x2`/ Sp √ ( 1/n₁ + 1/n₂)

t= 55,000- 58,000/ -900,862.1 √1/32+ 1/28)

t= -3000/ -233,120.21

t= 0.012868

3) D.f= n1+n2-2= 32+28-2= 58

for 2 tailed test the critical value of t is obtained by

t≥ t∝/2 ( d.f)=t≥ t0.025 (58)= ± 2.0017

4) Since the calculated value of t= 0.012868 does not fall in the critical region t≥ t∝/2 ( d.f)= ± 2.0017 we conclude that H0 is true and accept the null hypothesis.