Correlations Flashcards
What is correlational analysis?
Testing a hypothesis using an association that is found between two variables.
What cannot be done with a correlational analysis?
You cannot create cause and effect, you can only state that there is a strong correlation between the variables.
What is used in correlational analysis instead of IV and DV. Give examples.
Co-variables instead of independent and dependent variables - variables 1 (V1) and variable 2 (V2).
E.g. TV violence and aggressive behaviour.
E.g. Ice-cream sales and sunglasses.
E.g. Jealousy and stalker behaviour.
What are the 3 types of correlations?
Positive correlations.
Negative correlations.
Zero correlation.
What are positive correlations?
An increase in one variable leads to an increase in another variable.
What are negative correlations?
As one variable increases, the other decreases.
What are zero correlations?
There is no correlation.
Measuring a correlation is done in two ways. What are they?
Scatter graphs.
Correlation coefficients - the numerical representation of the strength direction of the relationships between two variables.
What must be done to variables in correlations? Why?
Because these are statistical methods using quantitative data, you need to operationalise your variables.
Example: Is there a correlation between education and intelligence?
State the non-directional hypothesis.
State the directional hypothesis.
State the null hypothesis.
Non-directional hypothesis: There will be a correlation between time spent in education and IQ.
Directional hypothesis: As the number of years spent in education increases, so will the scores on an IQ test.
Null hypothesis: There will be no correlation between the number of years spent in education and scores of an IQ test.
It is known that throughout the year, murder rates and ice cream sales are highly positively correlated. That is, as murder rates rise, so does the sale of ice cream.
What could other explanations be? Confounding variables?
Murders cause people to buy ice-cream.
People who buy ice-creams cause people to murder or get murdered.
Confounding variables such as temperature.
Murders increase in the summer, not due to ice-cream, but due to hotter climates.
State two strengths of using correlational analysis.
Allows researchers to analyse situations that could not be manipulated experimentally.
Can produce reasonably definitive information about causal relationships if there is no correlation between two co-variables.
Can collect a great amount of data quickly (a lot is done via self-report - quantitative questionnaires).
Easy and quick to analyse.
Allows us to see a relationship between two variables (even if it’s not cause – effect).
State two limitations of using correlational analysis.
Cannot establish cause and effect.
Researcher cannot manipulate variables .
Confounding Variables other than the ones you are measuring could have an effect.
Ethical issues – often study controversial/sensitive issues – need to be aware of social sensitivity. E.g. correlations between institutionalisation and IQ - stigmatise children grown up in institutions - stereotypes.
