ANOVA

Question 1

Why can’t we do multiple pairwise t-tests if you have multiple groups to compare?

Answer 1

It’s obvious and too simple, it’s also fishing for a result that results into too much inflation of type I error which is no good.

Question 2

How do you prevent from generating a type 1 error inflation?

Answer 2

Use the F-distribution.

Question 3

What is an F-distribution?

Answer 3

Basically it’s squaring a t-distribution to make all the negatives become positive. It allows us to get over the original problem of an inflated type 1 error.

Question 4

What does ANOVA mean?

Answer 4

Analysis of Variance

Question 5

How is the F-distribution used with ANOVA?

Answer 5

You compare the F-statistic to the distribution.

Question 6

What is an ANOVA?

Answer 6

It involves one or more independent variables with different levels/conditions (groups or measurements) analyzing differences in means (comparing them) and it’s on a continuous dependent variable.
Each different levels or conditions or measurements reflect a different mean related to that particular independent variable but with the same DV

Question 7

What is specific to the IV(s) in ANOVA?

Answer 7

This variable has to be a dichotomous variable.
You can put people into groups based on any
category (gender, handedness) or your
experimental manipulation (instructions versus no
instructions).

Question 8

What is specific to the DV in ANOVA?

Answer 8

The dependent
variable needs to be a continuous variable. Your dependent variable should be the
measurement you took in your study, or what
information you are expecting to see changed
over groups.

Question 9

What are levels?

Answer 9

Levels are the different groups within an IV (e.g. gender)

Question 10

What are conditions?

Answer 10

Conditions are the combinations of different

levels.

Question 11

What are the types of ANOVA?

Answer 11

You can have one-way or two-way or three-way ANOVAs.
The word “way” just means number of
independent variables.
• A one-way ANOVA is a hypothesis test
including one nominal IV (> two levels), and a
scale DV
• A two-way has two IVs.

Question 12

What’s a between/independent ANOVA?

Answer 12

ANOVA for independent measures where participants

are in only one level/condition.

Question 13

What’s a repeated measures/ dependent/ within group ANOVA?

Answer 13

ANOVA for dependent measures where participants
are in all levels/conditions.
Multiple time points - comparing them as many times as we want

Question 14

What’s a mixed designs?

Answer 14

ANOVA for when you have two variables, one
between subjects and one repeated measures.
Two independent variables one is often a repeated measure.

Question 15

If I say i have a 2X4 ANOVA?

Answer 15

There are 2 independent variables and 4 levels.