Question

A researcher tested gender differences in participants' responses to the item "I find strength and comfort in my religion or spirituality." The results were as follows: t(83) = 1.92, p = .06 What can we conclude?

Answer 1

Our objective here, is to test whether the response "I find strength and comfort in my religion or spirituality." differs among males and females.

Assuming the dependent variable to be continuous, suppose, the response is measured in the form of scores; Let denote the mean scores for males and females respectively. We would like to test the hypothesis at say, 5% level of significance.:

Vs

Running an independent t test, we get the computed test statistic value t(83) = 1.92.

And, since the p-value obtained for the test 0.06 > 0.05 (the chance of obtaining a result as extreme as the one obtained, when the null hypothesis is true, is greater than the fixed significance level; there is a greater chance that H₀ might actually be true), we do not have sufficient evidence to reject H₀. We fail to reject H_0. We may conclude that the response "I find strength and comfort in my religion or spirituality." do not differ significantly among males and females.

*However, if the significance is fixed at 10% level, we find the test result to be significant (p value = 0.06 < 0.10).