Please enable JavaScript.

Coggle requires JavaScript to display documents.

Research Methods Psychology - correlational studies, dispersion, measures…

- - - - allows predictions to be made
      - allows quantification of relationships - shown by correlation coefficients - e.g 0.9= 90whilst- shows a very string positive correlation -0.1 = 10% - shows a very weak negative correlation
      - correlations dont need manipulation of variables and so can be used when carrying out an experiment that may be unethical
    - - BUT there may be a quantification problem- as coefficients that may appear low (+0.25) may actually be significant when statistically tested. while correlations that deem high like +75 are not always statistically significant.
      - Because they arent done under controlled conditions- dont show causality - so cannot say that one variable causes another
      - Other variables may influence the co variables- extraneous variables - no may not be due to one variable being directly related to another.
      - Correlations only work for linear relationships
- - - - NOT AFFECTED BY EXTREME CORES LIKE OUTLIERS
      - USUALLY EASIER TO CALCULATE THAN THE MEAN
      - MEDIAN CAN BE USED WITH ORDIANL DATA UNLIKE THE MEAN
    - - NOT AS SENSITIVE AS THE MEAN- AS NOT ALL SOCRES ARE USED IN THE CALCULATION.
      - CAN BE UNREPRESENTATIVE IN A SAMLL SET OF DATA
  - - - MOST ACCURATE MEASURE OF CENTRAL TENDENCY- as it uses the INTERVAL LEVEL OF MEASUREMENT - where units of measure of equal size like seconds in time
      - it uses all data in its calculation
    - - LESS USEFUL IF SOME SCORES ARE SKEWED- LIKE IF THERE ARE SOME LARGE OR SOME SMALL SCORES
      - MEAN SOCRES MAY NOT BE ONE OF THE ACTUAL SCORES IN THE SET OF DATA
  - - - LESS PRONE TO DISTORTION BY EXTREME VALUES
      - SOMETIMES MAKES MORE SENSE THAN THE MEAN- E.GAVERAGE NUMBER OF CHILDREN IN FAILY BETTER SHWON AS 2 (MODE) THAN 2.4 (THE MEAN)
    - - CAN BE MORE THAN ONE MODE IN A SET OF DATA
      - DOESNT USE ALL THE SCORES
      - CAN BE NO MODE AT ALL/ ALL THE VALUES CAN BE THE MODE
- - - - fairly easy and quick to work out
      - takes full account of extreme values
    - - Can be distorted by extreme freak values
      - does not show whether data are clustered or spread evenly around the mean- e.g the range for 2 sets of data could be the same even though the data is very different
  - - - more sensitive dispersion measure than the range since all scores are used in the calculation.
      - allows interpretation of individual scores
    - - more complicated to work out
      - it is less meaningful if the data are not normally distributed
- - - - look to see if most scores are clustered around the mean
      - calculate the mean, mode and median to see if they are similar.
      - plot the data on a histogram to see if it forms a bell shaped curve