Lecture 12 - T-tests Flashcards
1
Q
What is the research process in psychology?
A
- ask a research question
- review the existing research
- develop a hypothesis
- collect data to test your hypothesis
2
Q
Categorical design?
A
when your looking at the frequency of occurrence across 2 categories
3
Q
Experimental design?
A
when your manipulating variables to look at differences between groups of participants or conditions
4
Q
Correlational design?
A
when your looking at relationships between 2 continuous variables
5
Q
What are the 2 types of variables?
A
- IV = what you manipulate
- DV = what you measure
- the IV is always nominal in experimental design and the DV is always continuous
- in an experimental design your IV can be:
-> independent (or between subject) measure = different participants are recruited into each of the conditions
-> repeated (within subjects) measures/ design = 1 group of participants that repeatedly take part in the study under different conditions
6
Q
What do t-tests look at at?
A
the difference between the 2 experimental conditions and the variability within the experiential conditions
7
Q
What is a parametric test?
A
- a test that requires data from 1 of the large catalogue of distributions that statistics have described, this is often the assumption that the population data are normally distributed
- a t-test is a parametric test
8
Q
What are the assumptions for a t-test (parametric test)?
A
- All observations should be independent so that each individuals participants data do not influence any other participants data
- Data should be collected at an interval or ratio level
- Data are (roughly) normally distributed
- Amount of variability should be the same for each condition = homogeneity of variance
9
Q
What is the repeated measures test?
A
- A test we use when we want to compare 2 sets of data coming from the same group of participants
- We are looking at the difference between the 2 experimental conditions
10
Q
What is the 1 sample t-test?
A
- use when we only have 1 set of data which we want to compare to a reference value
- the homogeneity of variance is not relevant for 1 sample t-test