Please enable JavaScript.
Coggle requires JavaScript to display documents.
Correlational Analysis (Hypotheses for correlational analysis (Need to…
Correlational Analysis
Correlation
-
It's a statistical technique used to measure or quantify strength of a relationship between 2 variables.
Strengths- Used when would be unethical or impractical to conduct an experiment because correlations don't require manipulation of IV.
Lead to further investigation because they can show the researcher if there's an initial relationship.
Weaknesses- Doesn't establish cause and effect because no manipulation of the IV.
Some variables can't be correlated because don't produce data on a numerical scale.
-
Negative Correlation
High values of one variable associated with low values of the other. As one variable increases the other decreases.
-
Correlation Coefficient
To work out strength of correlation you carry out Spearman's Rho. Giving correlation coefficient, number between -1 and +1 stating how strong correlation is.
If number close to +1 then there's a positive correlation but if close to -1 then negative correlation. If number close to 0 then variables uncorrelated.
E.g. +1 perfect positive correlation, +0.8 strong positive correlation, +0.3 weak positive correlation, -0.4 weak negative correlation, -0.8 strong negative correlation.
Variables
Correlation needs to have 2 variables that need to operationalised.
Intelligence- measured by GCSE score.
Memory- measured by performance on memory test.
-
-
-
Validity of Correlation
Factors affecting validity: Whether tasks measured what they were supposed to.
Realism of task (ecological validity).
Setting (ecological validity).
How generalisable sample is (population validity)
Level of demand characteristics
Improvements- Create realistic tasks reflecting real life. Collect data in realistic settings.
Collect large sample of data.
Avoid researcher bias.