Dependent (related) t-test

Dependent (related) t-test

1. looking for a difference

2. interval or ratio

3. repeated measures

4. related/dependent/paired samples/within groups/repeated measures

   • e.g does recall improve after attending 'Hypnotic Memory Training' with Paul McKenna

5. Dependent samples (or paired) t-test typically consists of a sample of matched pairs of similar units, or one group of units that has been tested twice ( a 'repeated measures' t-test

   • A typical example of the repeated measures t-test would be where subjects are tested prior to treatment with a blood-pressure lowering medication

6. parametric

7. If the p-value of the obtained test statistic is less than 0.5, reject the null hypothesis, one tailed test

8. To report an independent T-test in APA format you need to know the degrees of freedom, the t-value and the p-value

9. To report an independent T-test in APA format you need to know the degress of freedom, the t-value and the p-valuie

   • spss always gives you the significance value for a 2 -tailed test so you just divide it by two
   • if you get a value of .000 you just say less than 0.0001
   • The results of the dependent t-test showed that excersise had a significant effect onthe number of pull ups attained (t(24) = 5.196, p<0.001) participants could perform significantly more pull ups after excersise (M =4.2, SD=1.29) than before excersise (M=3.6, SD=1.72
     • we look at the mean!

10. Effect size

    • effect size is a simple way to quantifying the difference between two groups
    • Effect size emphasises the size of the difference rather than confounding this with the sample size. It is particularly valuable for quanitfying the effectiveneess of a particular intervention, relative to some comparison. It allows us to move beyond the simplistic , 'Does it work or not? ' to the far more sophisticated, 'How well does A work in a range of contexts
      • placing the emphasis on the most important aspect of intervention - the size of the effect - it promotes an alternative to it's statistical significance
    • effect size is a measure of the difference between the two sample means
    • Types of effect size
      • pearson r correlation
        • most used correlation
        • it is a measure that we can compare
        • pearson's r correlation can vary in magnitude from -1 to 1
          • -1 indicating a perfect negative relationship
          • 1 indicating a perfect positive relationship
          • 0 indicating no relationship between to variables
        • Cohen gives the following guidelines for the social sciences
          • small effect size, r = 0.1
          • medium effect size, r = 0.3
          • large, r = 0.5
    • Cohen's d
      • Cohen's d is the appropriate effect size measure to use in the context of a t-test on means
        • small effect size - may need a lot of people to actually find this
      • d is defined as the difference between two means divided by the pooled standard deviation or those means
      • the most accepted opinions on how to interpret the resultant effect size is
        • 0.2 small effect 0.5 medium effect and 0.8 large effect
      • If you have a large effect size you wouldn't find to hard to replicate this.