# Data analysis 5 and 6 part 2- use slides

version from 2017-10-17 08:57

## Section 1

if our sample mean is the same then just use the SE pooled calculation
is the T test the same equation for even and uneven sample sizes yes
How do you work out degrees of freedom of 2 different sample sizes (n1-1)+(n2-1) = n-2
testing the null hypothesis if t>CVreject Ho BIGGER values are more likely to result in rejection of Ho
testing the null hypothesis in SPSS output-Goes an extra step -Presents the probability (p-value) the null hypothesis is true -Probabilities are always less than 1 -SMALLER values result in rejection of Ho -If alpha = 0.05 it is necessary for the p-value to be less than 0.05 to reject the null hypothesis. Look at Sig (2 tailed) column. OR if the 95% confidence interval overlaps 0.
What does SPSS check for equality of variance. That Group A variance is not significantly different to Group B variance.
p value SPSS >0.05read equal variances assumed
p value SPSS =0.5read equal variances not assumed
What data analysis do you use in method section ( Reporting1 the results of an independent samples t-test)A two-sample independent/ unpaired Student’s t-test assuming equal variances using a pooled standard deviation was performed to test the hypothesis that the mean number of days taken to recover from malaria with Drug A is the same as that for Drug B.”
What data analysis do you use in results section ( Reporting1 the results of an independent samples t-test)An independent samples t-test indicated that the difference between the two drugs was not significant t(13- (degrees of freedom)) = 1.87, p=.084, using a two tailed test. To be more complete: An independent samples t-test indicated that the difference between Drug A (Mean = 23.38, SD=4.18, N=8) was not significantly different from that of Drug B (Mean = 20.00, SD=2.45, N=7), t(13)=1.87 p=.084, using a two tailed test.
how to convert SPSS output to a one tailed tes Simply take the p-value (sig) associated with t from SPSS output and divide by 2 and see if its p<0.05. Then use t-table to work out calculation by hand.
two sample t-test- cautions -Don’t Misuse the two sample t-Test (inappropriately analysed data as two sample t-tests) such as: Comparing Paired Subjects. Comparing to a Known Value. -Preplan One-Tailed t-Tests with strong evidence -Small Sample Sizes Make Normality Difficult to Assess: Outliers can be a problem -Performing Multiple t-Tests Causes Loss of Control of the -Experiment-Wise Significance Level – more in the ANOVA lectures (comparing more than 2 groups)