Correlation Flashcards
What is correlation?
The relationship between two DVs
Positive correlation?
As one variable increase, so does the other
Negative correlation?
As one variable increases, the other decreases
Zero correlation?
No consistent relation between variables, very scattered
Strong correlation?
Data points correspond with the relationship, close together
Weak correlation?
Less data points corresponds with the relationship, more outliers, more scattered data
Why is strength and direction important?
Helps with predictions
Computing correlation?
A numerical way to express correlation (more descriptive statistics)
- Use or choice depends on characteristics of variables
- One commonly used by psychologists is Peasron r
Pearson r formula
Summing the relationship of the z-score on the x-axis multiplying with those on the y-axis and multiplying it by the sample size
-No relationship can have an r value of 0
Things to note about r? x3
- Variables to be correlated must be measured on the same individuals
- Variables must be measured on an interval or ratio scale
- r can detect only linear relationships
linear lines
Linear points generally fall on a straight line, can be important when talking about different types of relationships
Curvilinear - Yerkes-Dodson around the curve
Increase in x results in initially an increase in y, then decrease in y
Restricted range
You’ve measured something that isn’t capturing the full extent of a given relationship
Cross-lagged-panel correlation procedure
A way of dealing with the directionality problem to a certain extent - doing a follow up later on to see if the correlation has gotten stronger
Cross-lagged- panel correlation - underlying assumption
if one variables ‘causes’ the other, it should be more strongly related over time
