Correlation= data analysis technique measuring the RELATIONSHIP between 2 variables to see if a trend exists between them (can't establish cause and effect). Can be gathered through:
The reason it cannot establish cause and effect is because it doesn't involve any manipulation of the variables. However this makes it a USEFUL technique when variables can't be manipulated for PRACTICAL or ETHICAL reasons.
Correlation shows the DIRECTION and STRENGTH of the RELATIONSHIP between 2 variables. Scatter graphs are used to display the relationship between the 2 variables. This allows EASY analysis of the DIRECTION (positive or negative) and STRENGTH (strong or weak) of the relationship. The closer the plotted results are, the STRONGER the relationship. The more spread out the plotted results are, the WEAKER the relationship.
- self-report
- observation
- psychological measures
The reason it cannot establish cause and effect is because it doesn't involve any manipulation of the variables. However this makes it a USEFUL technique when variables can't be manipulated for PRACTICAL or ETHICAL reasons.
Correlation shows the DIRECTION and STRENGTH of the RELATIONSHIP between 2 variables. Scatter graphs are used to display the relationship between the 2 variables. This allows EASY analysis of the DIRECTION (positive or negative) and STRENGTH (strong or weak) of the relationship. The closer the plotted results are, the STRONGER the relationship. The more spread out the plotted results are, the WEAKER the relationship.
Correlational hypothesis= predict the relationship between variables.
1-tailed:E.g 'There will be a POSITIVE RELATIONSHIP between the number of hours spent revising and score out of 20 on the Psychology test'
|
2-tailed:
E.g 'There will be a SIGNIFICANT RELATIONSHIP between the number of hours spent revising and score out of 20 on the Psychology test' |
Null correlational hypothesis= suggests there will not be a relationship between the variables. E.g 'There will be NO RELATIONSHIP between the number of hours spent revising and score out of 20 on the Psychology test'
Co-variables= the 2 variables used to investigate whether there is a relationship with them in a correlation. E.g score out of 20 on a Psychology test and number of hours spent revising.
Correlation coefficient= shows how closely linked variables are, by a number ranging from -1 to 1; which indicates what type of correlation it is (positive or negative). They allow a quantification of the STRENGTH of a relationship. For e.g we know a correlation with a coefficient value of 0.9 is stronger than 0.3. Coefficients from a smaller sample are not as meaningful as larger sample sizes and may not be as significant if a large coefficient value is found from the correlation.
- The closer to -1 or 1 means the variables are MORE CLOSELY linked
- -1 to 0 value =negative correlation
- 0 to 1 value = positive correlation
- A value closer to 0 = weaker correlation/no correlation
Evaluation of Correlation:
Strengths:
|
Weakness:
|