Correlations

• Tips


  -don't use the following terms: 'effects', 'difference', 'compared to'

  -only collect quantitative data (by operationalising the variables)

  -measuring co-variables (there is no IV/DV)
• What are correlations


  -a statistical technique

  -involves measuring + looking for a relationship between 2 variables (co-variables) - a correlation refers to a measure of how strongly the variables are related to each other (not really a method of collecting data but a way of analysing data already collected)

  -cause and effect cannot be established
• What is a positive correlation


  relationship between variables where as one increases, the other increases e.g as temp increases, the number of ice creams sold increases
• What is a negative correlation


  relationship between variables where as one variable increases the other decreases e.g as temp increases, the number of hot water bottles bought decreases
• What is no correlation


  when two variables are said to be uncorrelated and have no relationship
• How are correlations displayed


  on scatter graphs

  the axis should be fully operationalised and labelled to indicate what co-variable they represent and how it was measured

  
• What can cause the relationship between 2 co-variables


  a third variable can affect the relationship between the co-variables e.g as the number of ice creams bought increases, the number of drownings increases -> the other variable that causes this relationship is hot weather
• What is a correlation co-efficient 


  -tell us about the strength + direction of the relationship/correlation between 2 variables

  -calculated using Spearman's Rho analysis

  -refers to a scale between -1 and +1. +1 = perfect positive correlation and -1 = perfect negative correlation. 0 = no correlation
• Operationalise the variables: the older I get, the worse my memory gets


  V1 = age in years

  V2 = score on a word recall test out of 20
• One tailed correlational hypothesis

  • suggests there will be a relationship between 2 co-variables, with a direction (e.g. positive/negative)

  e.g. There will be a negative correlation between the number of hours on average students spend on social networking sites per day and their GCSE English grade from 1-9, as a the number of hours spent on social networking sites increases the GCSE English grade decreases. (must operationalise)
• Two tailed correlation hypothesis

  • suggests there will be a relationship between 2 co-variables, with no direction

  e.g. There will be a significant relationship between the number of hours spent on social networking sites per day and GCSE English grades from 1-9.
• Null correlational hypothesis

  • suggests there will be no relationship/correlation between the 2 variables - 'any relationship will be due to chance'

  e.g. There will be no significant relationship between the number of hours on average students spend on social networking sites per day and their English GCSE grade from 1-9. Any relationship will be due to chance.
• Strengths of conducting correlations


  +able to identify relationships between variables w/o having to manipulate the behaviour so can be used in situations where an experiment would be unethical or impractical/it would be unethical or impractical to manipulate the behv/IV = upholds the rep of psychology

  +starting point of research, often used to suggest ideas for future experimental research = can create hypotheses for future experiments

  +use of quantitative data increases objectivity (validity/science) + replicability (reliability)
• Weaknesses of conducting correlations 


  -no manipulation means you can't establish cause and effect = correlations can only show a relationship between variables, therefore future research is needed to establish cause and effect

  -other unknown variable(s) may be responsible for the relationship between the co-variables being measured = lowers IV as we aren't truly measuring what may have led to the correlation, limits ability to use the research