Problem No.
4(Chi-square test for testing independence of attributes)
A sample survey of PSTU graduates employed in various organization gave the
following results.
Utilization of educational qualification in present job
Type of organization
Fully utilized Partially utilized Very little or not at all
Education and research 112 19 6
Extension 88 27 1
Others 27 40 26
Test whether there is any association between the utilization of education with
the type of employer organization.
Solution to the Problem no.4
Here, 𝐻0 : There is no association between the utilization of education with the
type of employer organization
𝐻1 : There is association between the utilization of education with the
type of employer organization
2
𝑂𝑖𝑗
The required test statistic is 𝜒 2 = σ( )-N with (m-1)(n-1) d.f.
𝐸𝑖𝑗
So, we need the following table to get the expected cell frequencies.
Type of organization Utilization of educational qualification in present job
Total
Fully utilized Partially utilized Very little or not at all
Education and research 90 34 13 137
Extension 76 29 11 116
. Others 61 23 9 93
Total 227 86 33 346
227∗137 86∗137 33∗137
𝐸112 = = 90, 𝐸19 = = 34, 𝐸6 = = 13,
346 346 346
227∗116 86∗116 33∗116
𝐸88 = = 76, 𝐸27 = = 29, 𝐸1 = = 11,
346 346 346
227∗93 86∗93 33∗93
𝐸27 = = 61, 𝐸40 = = 23, 𝐸26 = = 9.
346 346 346
2
𝑂𝑖𝑗
The required test statistic is 𝜒 2 = σ( )-N with (m-1)(n-1) d.f.
𝐸𝑖𝑗
1122 192 62 882 272 12 272 402 262
= + + + + + + + + − 346
90 34 13 76 29 11 61 23 9
𝑤𝑖𝑡ℎ 3 − 1 3 − 1 𝑑. 𝑓.
= 90.52 with 4 d.f.
2
𝑂𝑖𝑗
The required test statistic is 𝜒 2 = σ( )-N with (m-1)(n-1) d.f.
𝐸𝑖𝑗
1122 192 62 882 272 12 272 402 262
= + + + + + + + + − 346
90 34 13 76 29 11 61 23 9
𝑤𝑖𝑡ℎ 3 − 1 3 − 1 𝑑. 𝑓.
= 90.52 with 4 d.f.
Comment: The critical value of 𝜒 2 at 1% level of significance with 4 d.f. is
13.28, the calculated value is greater than the critical value. Hence, we
reject the null hypothesis.