Correlations Flashcards
What is correlation?
Plotted in a scatter gram to tell us the strength and direction of a relationship between two or more co-variables
What are the three types of correlation?
Positive, negative or zero
What is positive correlation?
Two variables rise and fall together
What is negative correlation?
One variable rises and the other one falls
What is zero correlation?
There is no relationship between the variables
What are co variables?
They occur naturally and are measured for the purpose of research
What are experimental variables?
The IV is created/manipulated by the researcher and the DV is measured
How are correlations represented?
By correlation coefficients
What does correlation investigate?
The relationship between two variables
What do experiments investigate?
The difference and if the IV has caused a change in the DV
What is a very strong correlation?
+/- 0.8 to 1
What is strong correlation?
+/- 0.6 to 0.8
What is moderate correlation?
+/- 0.4 to 0.6
What is weak correlation?
+/- 0.2 to 0.4
What is negligible/no correlation?
+/- 0.2 to 0
What is a directional correlation hypothesis
States whether the correlation is positive or negative e.g, there is a ____________ correlation between (co-variable 1) and (co-variable 2)
What is a non-directional correlation hypothesis?
Simply states if there is a correlation, e.g, there is a correlation between (co-variable 1) and (co-variable 2)
What are strengths of correlation?
Useful preliminary tool for research, quick and economical to carry out, less time consuming than experiments and more ethical than experiments as data already exists
What are limitations of correlations?
Cannot demonstrate cause and effect, third variable problem and correlation may miss curvilinear relationships
Why can’t correlations not measure curvilinear relationships?
They only measure linear relationships so if correlation coefficient is 0, curvilinear or no relationship could be present
What is the issues with intervening variables?
No IV is manipulated so cannot draw causal conclusions
What is an example of an intervening variable?
Caffeine and anxiety. If a person has a high stress job, this can lead to anxiety and lead to caffeine due to tiredness so caffeine ≠ anxiety
How are correlations more ethical than experiments?
An IV is not manipulated as this could be unethical