Theory
Hypothesis
Operationalisation of variables
Selection of participants
Experimental design/ survey studies
Data collection
Data analysis
Findings
1 of 19
Type I error
False positive.
Stating there is an effect when there isn't.
2 of 19
Type II error
False negative.
Stating there is not an effect, when there is one.
3 of 19
Alpha number
0.05
4 of 19
What does Alpha represent?
The probability of a type I error
5 of 19
Beta figure
0.20
6 of 19
What does beta represent?
Probability of a type II error
7 of 19
What is effect size?
gives an indication of how big a relationship is.
8 of 19
what do effect sizes attempt to control for?
Type I error
9 of 19
What is power analysis?
They tell us the strength a test would need to find an effect
10 of 19
An attempt to control type II errors
Power analysis
11 of 19
A priori
a type of power analysis that is conducted before data collection. used to determine the sample size.
12 of 19
Post hoc
A type of power analysis conducted after data collection and analysis.
13 of 19
Parametric assumptions are...
Normal distribution.
Homogeneity of variance.
Ratio/Interval data.
Equal size groups.
14 of 19
Non-parametric assumptions/data
used when data do not meet the required assumptions.
Ordinal data
Smaller/unequal group sizes.
15 of 19
Four core principles of open science
1- honesty in all aspects of research
2- Accountability in the conduct of research
3- Professional courtsey and fairness in working with others
4- Good stewardship of research on behalf of others.
16 of 19
what is the replication crisis? (2005)
Concerns regarding the reliability of research findings in literature- only 36% of studies can be replicated and produce a significant effect as reported in original study.
17 of 19
What is P Hacking?
When you set out data analysis plan in advance and stick to it.
18 of 19
What is it called when you set your hypothesis out in advance and don't change it?
Harking
19 of 19
Other cards in this set
Card 2
Front
Type I error
Back
False positive.
Stating there is an effect when there isn't.
Comments
No comments have yet been made