Research Methods

Research Methods

1. Sample

   • Target population - group interested in studying
   • Sample - the participants
   • if the sample represents the target population then it is generalisable

2. Data

   • Types of data
     • Quantitative - numbers
     • Qualitative - words
   • Levels of data
     • Nominal - Tally/total number of times something happens
     • Ordinal - Rank order, from smallest to largest
     • Interval - Precise measurement (universal)
   • Primary - collected by researcher
   • Secondary - gathered by someone else

3. Reliability

   • How consistent a measuring device is
   • Internal - can you replicate the study?
     • Split-half method - compare results of one half with the other. if positive correlation its is assumed to be consistent
     • Inter-rater reliability - if different observers get the same results
   • External - have the participants got a consistent result?
     • Test-retest method - compare results taken on two different occasions

4. Validity

   • How accurate a study is
   • Internal - Does it measure what its meant to measure?
     • Face validity - does it measure what its supposed too
     • Construct validity - is the measuring device appropriate to the concept
     • Concurrent validity - comparing it with a well established test
     • Criterion validity - how well it predicts future performances
   • External - can the results be generalised to more people?
     • Population validity - can the results be generalised to the wider population
     • Ecological validity - resembles what will happen in everyday life

5. Ethics

   • Respect
     • Consent
     • Withdrawal
     • Confidentiality
   • Competence
   • Responsibility
     • Debriefing
     • Protection from harm
   • Integrity
     • Deception

6. Sampling methods

   • Self-selecting = people volunteer to take part
     • not biased and got consent
     • too many or not enough people
   • Opportunity = readily available at a certain time
     • quick and convenient
     • biased sample as from one place
   • Random = selecting names from a hat
     • Generalisable and unbiased
     • Impractical for a large target population
   • Snowball = participants contact friends and family to take part
     • cheap - no advertising
     • biased as they all know each other so will be similar