RESEARCH PART 2

• correlation

  mathematical technique where a researcher investigates an association between two variables
• co-variables

  variables investigated within a correlation, not IV or DV as investigating correlation rather than cause and effect
• primary data

  original data that has been collected specifically for the purpose of the investigation by the researcher
• strengths of primary data

  'fit for purpose' - designed to target the information that the researcher requires
• limitation of primary data

  requires time and effort from the researcher
• examples of primary data

  data gathered by conducting an experiment, questionnaire, interview
• secondary data

  data that has been collected by someone else, already existing data
• examples of secondary data

  data in journal articles, book, websites, stats
• strength of secondary data

  requires minimal efforts as it is cheap and easy to access
• limitation of secondary data

  variation in quality and accuracy (outdated or incomplete), and content may not match researcher's needs
• qualitative data

  data expressed in words and take the form of written descriptions
• examples of qualitative data

  interview transcript, diary extracts, notes
• strengths of qualitative data

  rich and detailed, more meaningful insight
• limits of qualitative data - difficult to analyse

  cannot be summarised statistically, patterns and comparisons hard to identify
• limit of qualitative data - conclusions

  conclusions may be biased as could be subjective interpretations from researcher who could have preconceptions
• quantitative data

  data expressed numerically and can be analysed through graphs
• examples of quantitative data

  data in individual scores
• strength of quantitative data

  more objective and less bias
• strength of quantitative data

  simple to analyse so easy comparisons
• limit of quantitative data

  may not represent real life as it little detail
• meta-analysis

  research that uses secondary data
• meta-analysis is the process in which

  data from a large number of studies, which have the same research questions and methods, are combined
• quantitative meta-analysis involves

  the effect size
• effect size

  dependent variable of meta-analysis
• what does the effect size tell us

  overall statistic measure of the relationship between variables across a number of studies
• advantages of meta-analysis

  allows us to view data with confidence

  results can be generalised across larger populations
• disadvantages of meta-analysis

  publication bias - (file drawer problem)

  -> researcher may not select all relevant studies

  -> choosing to leave out the studies with negative results

  -> data may be biased as only represents some relevant data and perhaps incorrect conclusions
• correlation

  technique we use to determine the relationship between 2 co-variables
• positive correlation

  • as one co-variable increases, the other co-variable increases
  • (as the number of people in a room increases, the noise levels increase too)
• negative correlation

  • as one co-variable increases, the other co-variable decreases
  • (as the number of people in a room increases, the personal space decreases)
• no correlation

  • there is no relationship between 2 co-variables
  • the amount of caffeine someone drinks and the dogs someone sees on the street
• strength of a correlation

  allows quantification of relationships as they show the strength of the relationship between variables
• strength of a correlation

  there is no manipulation of behaviour required so can be a quick and ethical method of data collection and analysis
• strength of a correlation

  correlations provide a starting point for the research as they show possible patterns between variables before researchers commit to an experiment
• limit of a correlation

  cannot infer cause and effect so cannot assume one variable causes the other
• limit of correlation

  extraneous variables may influence the correlation as other variables may influence both measured variables
• limit of correlation

  only works for linear relationships as the relationships between some variables may be curvilinear
• directional correlational hypothesis

  there will be a positive/negative correlation between variable 1 and variable 2
• non-directional correlational hypothesis

  there will be a correlation between variable 1 and variable 2
• null correlational hypothesis

  there will be no correlation between variable 1 and variable 2