Methods Week 7 - Choosing Statistical Tests (1/2)

z-tests

a statistical test used to determine whether there is a significant
difference between sample and population means

what do you need to do a z-test

the population parameters (know about the pop. ex. IQ is known)

• know the mean and SD

one-sample t-tests

compare the mean of a single sample to a known population
mean to determine if there is a statistically significant difference

• know the mean but not the SD

• finding the diff between the sample and the world

between-group T-test

compare the means of two independent groups to
determine if there is a statistically significant difference between them.

• don’t know the mean or SD

• finding the difference between two groups

paired/within T-test

compare the means of two related groups to determine if
there is a statistically significant difference between them

testing means between the same group

• don’t know the mean or SD

• finding difference in one group between two points in time

One-way ANOVA

compares the means of two or more independent groups based on one independent variable (factor).

• a t-test but can run two or more independent variables

• ex. compare coffee, tea, or water before bed

repeated measures ANOVA

compares the means of one group based on two or more independent variable levels (conditions).

• can compare at more than two different time points

• within subject ANOVA

Factorial ANOVA between subject example

compare the means of two or more independent groups based on two or more independent variables (factor).

ex. comparing independent variable 1: type of drink before bed to independent variable 2: exercise before bed or not

Mixed (factorial) ANOVA

compare the means of two or more independent groups based on one or more between-subject IV, and one or more within-subject IV

requires a mixed design (between subject and within subject design)

how to calculate how many factors you have in factorial AVOVA

multiply the levels you have in the independent variable groups by each other

factorial ANOVA within-subject example

measuring DV in ONE group

why are ANOVA’s good?

good at looking at interactions as well as multiple comparisons

what are interactions

ex. if one level of one iV effects other levels of the other IV differently

ex. does drinking coffee before bed impact young and elderly groups differently 

Eta squared

measures effect size for ANOVA’s

what do you do to measure effect size of

t-tests: cohon’s d

ANOVA’s: Eta squared

what is an ANOVA and how is it different from Bonferri’s correction?

ANOVAs are a way of comparing means across 3+ groups. it reduces the amount of work and correction you have to do BUT it only tells you IF there is a difference at all between the 3+ groups.

To figure out which group had the difference, then you would have to do Bonferri’s correction to find the specific differences. This method is also a correction for addition rule, reducing type 1 error.

which ANOVA type compares the means of 3+ groups with one IV

one-way ANOVA

which ANOVA type compares the means of one group with 3+ IV

repeated measures ANOVA

which ANOVA type compares the means of 2+ independent groups with 2+ IV

factorial ANOVA

which ANOVA type compares the means of 2+ groups with one btwn group and one within group

mixed ANOVA