Independent Sample T-test
What are the null hypothesis in an Independent samples t-test in words?
(Ho: M1 = M2 or M1- M2 = 0 ) IN THE POPULATION there is no difference between mean 1 and mean 2.
Steps of conducting an independent samples t-test
1) list hypothesis (alternative and null) 2) calculate t- score 3) Normality check: descriptive stats (KS test, central tendency etc) 4) IF normality assumed run independent sample t-test 5) check whether homogenity of variance exists by looking at the levenes test. 6) If equal variances assumed use "equal variances" section of t-test. If no equal variances use the "unequal variances assumed portion." 7) report normality 8)results (t = ___, df = ____, p ____) and in the context of the hypothesis test. Don't forget "IN THE POPULATION" 9) Calculate Cohen's d. and determine if small, med, or large effect.
What are the assumptions for the Independent samples t-test?
1) Scores of the dependent variable are normally distributed in the population 2) Independence between groups - faith in researcher 3) random assignment - faith in researcher 4) Homogeneity of variances
What is an example of an independent Samples T- test?
Comparing independent treatment and control groups on mean level of depression. OR Comparing women and men (who are not connected to each other - complete strangers) on mean levels if self esteem.
What can the alternative hypothesis be for a one tailed test in statistical notion?
Ha: M1 - M2 > 0 OR Ha: M1 - M2 < 0
What can be the alternative hypothesis for an Independent samples t- test in statistical notation?
Ha: M1 does not equal M2 OR Ha: M1 - M2 does not = 0
What can be the different null hypothesis in an Independent samples t-test in statistical notation?
Ho: M1 = M2 OR Ho: M1 - M2 = 0
What is the purpose of Independent sample t-test?
It is used when you have two samples that are not related to one another and you want to see of there is a difference between the means of the two populations form which the samples are drawn.
What are the alternative hypothesis in an Independent samples t-test (one- tailed) in words?
One tailed: IN THE POPULATION , the difference between the mean of group one and the mean of group two is greater/less than zero (depending on the directionality). i.e in the population there is a difference between the mean annual salary of dental hygienist and the mean annual salary of lawyers.
What are the alternative hypotheses in an Independent samples t-test (two-tailed) in words?
Two tailed: IN THE POPULATION, there is no difference between mean one and mean two.
What test helps us determine whether we have homogeneity of variance within our two independent groups?
Use the Leven's test that is automatically produced when you run an indpendent samples t-test. IF p-value is significant, ( p < 0,05) we cannot use the independent sample t-test andnormality assumptions have been violated. If P> 0.05 than we can conclude that homogeneity of variances exists.