Buy Coursework: Coursework Writing- determining “equal variances not assumed”
In order to determine “equal variances not assumed”, researchers must once again look at the Levene test (Warner, 2013). If the population and/or group being tested displayed skewed correlations and different variances, then the researchers would deem “equal variances not assumed” (Warner, 2013). Another time that researchers may decide that “equal variances not assumed” would be if the f value on the Levene test was very small (Warner, 2013). If it was f value did not show statistical significance, then the test would once again be reported as “equal variances not assumed” (Warner, 2013). This would also be true if the p-value was determined to be small (ie. p<.05 or p<.01) (Warner, 2013).
There are two different versions of the t-test on the SPSS printout. One version states, “equal variances assumed” while the other reports “equal variances not assumed”. In order to determine which version of the t-test should be utilized, the Levene test should be used (Warner, 2013). Using the Levene test to determine which t-test should be used is the best way to decipher whether or not “equal variance assumed” or “equal variance not assumed”. If a researcher looked at the Levene test and saw that there was no severe and/or significant violation, then the researcher would be able to determine that “equal variance assumed” (Warner, 2013). When there is “equal variance assumed”, it is important to remember that there is always a risk for Type 1 error (Warner, 2013). In order for researchers to determine “equal variance assumed”, they must also make sure that the population is not small and unequal (Warner, 2013). If there was a tiny, uneven population and/or group being measured, then the risk of Type 1 error becomes higher
Equal variance assumed is when in a sample the two independent samples are assumed that are drawn for populations that are identical in their variances, in this t test it is assume equal variances (Warner, 2013). Equal variances not assumed is when two independent samples are assumed that are drawn from populations that are unequal in variances, in this case it is not assume equal variances (Warner, 2013). In the SPSS output, we have both equal variances, to determine which variance is going to be used we need to rely on what the Levene’s test indicates. For example, if the Levene’s test indicates that the variance is equal in both groups then the equal variances assumed is appropriate, but if indicates that both groups are not equal then the equal variances not assume would be appropriate. When utilizing equal variances are assumed the researcher utilizes the calculation of pooled variances (Warner, 2013). Instead when equal variances cannot be assumed, researcher would use calculation un-pooled variances and a correction to the degrees of freedom.
Validate the above statemments with clear examples
