Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Transcript of Correlations
2. How to explain a correlation between two or more variables.
3. How to evaluate correlations
4. How to graphically present a correlation Research methods such as observations, interviews, questionnaires and case studies can all produce CORRELATIONS. positive correlations show an increase in both variables, so they go uphill What is a correlation? "The measurement of a relationship between two or more variables". The measured variables are called 'co-variables'. Look carefully at the word... CORRELATION Correlational Hypotheses Previously you have learned about 'experimental hypotheses' and 'null hypotheses'. However, you can't call the hypothesis for a 'correlation' an 'experimental' one can you? Because there is no IV or DV in any other method except an experiment.
So - in a correlational piece of research you will call it an 'alternative hypothesis' and it will predict the expected relationship between two variables.
For example, "There will be a significant correlation between age and beauty" STRENGTH You should explain a correlation in terms of its This is called the 'correlation coefficient' and it's just a number Imagine a number line from zero to one.
One is a STRONG correlation, Zero is no correlation. 0.5 is moderate. DIRECTION There are POSITIVE correlations. For example... "As people get older they become more attractive" This is POSITIVE because as one variable increases, the other also increases (as age increases, attractiveness increases too) There are NEGATIVE correlations. For example... "As people get older they become LESS attractive" This is NEGATIVE because as one variable increases, the other DECREASES (as age increases, attractiveness decreases) Scattergrams how to graphically present correlational data negative correlations show an increase in one variable, with a decrease in the other, so they go downhill Data that has 'no correlation' is random and forms no pattern EVALUATION Strengths Correlations can be used when it would be unethical to manipulate variables in an experiment
If a significant correlation is found then further investigations can be justified Weaknesses We cannot establish 'cause and effect' from a correlation, only a relationship between variables. If we were to report a causal relationship between two variables it could lead to public misunderstanding
There may be other variables that can explain why the co-variables being studied are linked. This would reduce the validity of the research Negative correlations are shown with a negative sign (e.g. r= -0.5) Positive correlations are shown with a positive sign (e.g. r= +0.5) If the data has a strong correlation coefficient (r=1) then all the data points will lie on the line of best fit.
As the correlation gets weaker, from 1 towards 0, then the data points get wider spread around the line
We describe the correlation with both the coefficient of its strength AND it's direction. For example: r=-0.7 (moderately strong, negative correlation) Sometimes the positive sign (+) is left out. If there is no sign, it is a positive correlation. Remember that POSITIVE and NEGATIVE only refers to the direction of the relationship between co-variables!