Question 8. Are the t tests important to the conclusions of the study? In other words, if the sample were large enough for these differences to be significant with these means and standard deviations, would this change the conclusion?
The t tests are not important to the conclusions of the study. What we want to know is how good the semen index is at discriminating between fertile and infertile donors, not whether the mean level is different. The degree of overlap between the fertile and infertile populations is much more important. As the means and standard deviations suggest that there is much overlap, the conventional semen indices do not discriminate well. The sensitivity and specificity of the semen tests would be more useful.
Back to question.
This page maintained by Martin Bland.
Last updated: 27 July, 2009.
Back to top.