PS211 - Exam 2 - One-sample t-test

when do you use a t-test as opposed to a z-test

• z-test

  • pop SD known

• t-test

  • pop SD not known

  • sample mean and sample SD known

assumptions for t-test

• DV measures interval/ratio

• pop. of raw scores forms normal distribution

• pop mean is known, pop SD not known

steps to performing t-test

• calculate SEM

• calculate t-obt

• calculate df

• compare t-obt to t-crit

• reject/retain null

• calculate effect size

• calculate confidence interval

degrees of freedom

• n-1

• number scores free to vary in sample

  • higher df = more normal distribution

  • smaller df, t-crit>z-crit for same area

distribution for t-test

family of t-distributions

when to reject null

t-obt>t-crit

cohen’s d

describes size of difference between means expressed in SD units

small effect size

0.2

medium effect size

0.5

large effect size

0.8

eta squared

proportion of variability in DV that’s accounted for by IV

confidence interval

highest mean diff and lowest mean diff we expect for pop mean and sample mean