Laserfiche WebLink
MI MINN2SOTA <br />In the example on page three, t-test results <br />would not be used because the <br />underpayment ratio is above 80%, but let's <br />assume we needed to check these results. <br />First, we would find 16 in the DF column <br />and then look across to find the value of t at <br />1.746. Since our t-test number is-3.732, <br />well below the value of t on the table, these <br />results would show that on average, females <br />are not underpaid compared to males. <br />T-Test Table <br />(5% Significance) <br />DF <br />Value of <br />DF Value of t <br />DF <br />Value of t <br />1 <br />6.314 <br />12 1.782 <br />23 <br />1.714 <br />2 <br />2.920 <br />13 1.771 <br />24 <br />1.711 <br />3 <br />2.353 <br />14 1.761 <br />25 <br />1.708 <br />4 <br />2.132 <br />15 1.753 <br />26 <br />1.706 <br />5 <br />2.015 <br />16 1.746 <br />27 <br />1.703 <br />6 <br />1.943 <br />17 1.740 <br />28 <br />1.701 <br />7 <br />1.895 <br />18 1.734 <br />29 <br />1.699 <br />8 <br />1.860 <br />19 1.729 <br />30 <br />1.697 <br />9 <br />1.833 <br />20 1.725 <br />40 <br />1.684 <br />10 <br />1.812 <br />21 1.721 <br />60 <br />1.671 <br />11 <br />1.796 <br />22 1.717 <br />120 <br />1.658 <br />Infinity <br />1.645 <br />While the entire method for calculating t-test <br />results cannot be explained here, it is a <br />commonly accepted mathematical technique <br />for measuring statistical significance. The <br />formula is fairly complex, but basically it <br />factors in predicted pay, the dollar <br />difference from predicted pay and the <br />number of employees. The DF number is <br />the total number of employees in male or <br />female dominated classes only, minus two. <br />8. Average Dollar Amount Male Classes are <br />Above or Below Predicted Pay <br />In the example on page three, the maximum <br />monthly salary for male classes, on average, <br />is $2 above predicted pay. <br />9. Average Dollar Amount Female Classes <br />are Above or Below Predicted Pay <br />In the example on page three, the maximum <br />monthly salary for female classes, on <br />average, is $75 above predicted pay. <br />10. Salary Range Test <br />This number must be either 0% or 80% or <br />more to pass this test. In the example on <br />page three, 105.71% is passing. <br />Jurisdictions not passing this test can pass it <br />by reducing the number of years it takes for <br />female classes to reach maximum salaries, <br />increasing the number of years for males to <br />reach maximum salaries, or some <br />combination of both. A result of 0% would <br />mean that either there are no male classes <br />with an established number of years to move <br />through a salary range, no female classes <br />with an established number of years to move <br />through a salary range, or both. A <br />description of how the salary range test is <br />calculated is on page 18. <br />11. Exceptional Service Pay Test <br />This number must be either 0% or 80% or <br />more to pass this test. In the example on <br />page three, 50% is not passing. Jurisdictions <br />not passing this test can pass it by either <br />increasing the number of female classes that <br />receive exceptional service pay, decreasing <br />the number of male classes that receive <br />exceptional service pay, or some <br />combination of both. A result of 0% could <br />mean that fewer than 20% of male classes <br />receive exceptional service pay or that no <br />female classes receive exceptional service <br />pay. A description of how the exceptional <br />service pay test is calculated is on page 19. <br />Guide to Understanding Pay Equity Compliance Tests - 10/16 Page 5 <br />