Types of Sampling Procedures

Event sampling= recording the number of times a certain behaviour occurs in a group

Time sampling = recording behaviours in a given time frame e.g. recording every 30 seconds

Correlational analysis

Statistical analysis based on a relationship between co-variables (naturally occurring variables). Correlations measure the degree in which they co-vary.

The range of correlations – correlation co-efficient – ranges from -1 to +1 depending on the strength and direction of the relationship.

  • + Used when experiment is unethical/impractical
  • + Used to suggest areas worthy of further study
  • + Establish the strength and direction of the relationship between variables
  • – Cannot be used to establish causal relationship
  • – Only useful if looking for linear relationship

Context analysis: identifying themes or codes within data and counting the frequency within the data set

Procedure of content analysis

  • All data needs to be read/looked at
  • Researchers need to come up with operationalised categories/themes
  • Should be done by 2 researchers (inter observer reliability)
  • Conduct a pilot study on a small sample to test appropriateness of codes
  • Rewrite codes/themes if necessary
  • Analyse all data by counting up number of times each code/theme occurs

Case studies

In depth investigation of one person or small group

  • + High levels of validity as in-depth and give insight
  • + allow researchers to study events that they could not practically/ethically manipulate
  • – small samples = difficult to generalise
  • – researcher becomes too involved = lose their objectivity and may misinterpret or influence outcomes
  • – difficult to establish cause and effect as often after the event