Correlations Flashcards
What is a correlation?
A mathematical technique in which a researcher investigates an association between two variables, called co-variables. Always ordinal or interval data.
What is a positive correlation?
As one co-variable increases so does the other. For example, the number of people in a room and noise tend to be positively correlated.
What is a negative correlation?
As one co-variable increases the other decreases. For example, the number of people in a room and amount of personal space tend to be negatively correlated.
What is zero correlation?
When there is no relationship between the co-variables. For example, the association between the number of people in a room in Manchester and the total daily rainfall in Peru is likely to be zero.
What are co-variables?
The variables investigated within a correlation, for example, height and weight. They are not referred to as the independent and dependent variables becuase a correlation investigates the association between the variables, rather than trying to show a cuase-and-effect relationship where one variable “depended” on the other.
What is are curvilinear/non-linear relationships?
A curvilinear relationship is a type of relationship between two variables that has a pattern of correspondence or association between the two variables that change as the values of the variables change (increase or decrease).
What are correlation coefficients?
A number between -1 and +1 that represents the direction and strength of a relationship between co-variables.
Why do we use correlations and what statistical test would we use for them?
Correlations are used in the assessment of reliability and validity. The coefficient must be above +.8 for a study to be considered reliable and/or valid. The statistical test would be Spearman’s or Pearson’s.
What are 5 strengths of correlations?
- Correlations can be done when it would be
unethical to manipulate the variables - Correlation can be done when it would be
impractical to manipulate the variables - If a correlation is not significant, you can probably
rule out a casual (cause and effect) relationship - If it is found to be significant, then further
investigation (e.g. An experiment) can be justified - Like with experiments, procedures can be
repeated to confirm findings
What are 3 weaknesses of correlation?
- People often misinterpret correlations and
assume that a cuase and effect relationship has
been found, but this is not possible - Unknown variables (intervening variables) may
explain why the co-variables being studied are
linked - As with experiement, they ay lack validity e.g. The
way they measure a variable isn’t realistic or the
sample may not be generalisable