Research methods

Research methods

1. Hypotheses

   • Null
     • There will be no significant difference in (DV) in (IV 1) and (IV 2)
   • Alternative
     • There will be a significantly (higher/lower DV) in (IV 1) than in (IV 2)

2. Experiment types

   • Natural
     • 1- May happen very rarely 2- participant differences can affect validity of experiment
     • 1- High ecological validity
   • Field
     • 1- Researcher may lose control over EVs since it's a real life setting 2- Ethical issues (do they know its a study?)
     • 1- More realistic than lab experiments 2- some control over EVs since procedures can be standardised
   • Lab
     • 1- low ecological validity 2- demand characteristics
     • 1- low extraneous variables

3. Variables

   • Extraneous variable (EV)
     • anything that changes other than the IV
     • can jeopardise the validity of an experiment
   • Dependant variable (DV)
     • the results from the experiment
   • Independant Variable(IV)
     • what is changed in the experiment (usually only two categories)
     • needs to be operationalised (given quantities

4. Experiment designs

   • Independent groups (No person does both IVs)
     • order effects is no problem
     • Participant variables affect results
       • random allocation can be used to fix this
   • Repeated measures (everyone does both conditions of IV)
     • Order effects (participants might get practice from the first condition they were in)
       • Counter-balancing used to fix this
     • 1- Participant variables are removed 2- less participants needed
   • Matched pairs (people of similar abilites are paired together, and one half of the pair does IV 1 while the other does IV 2)
     • 1- Time consuming 2- not always exact
     • 1- No order effects 2- almost no participant variables

5. Sampling methods

   • Random sampling
     • - Time consuming - May not reflect target population
     • No bias
   • Opportunity sampling
     • Does not reflect target population
     • Easy to do
   • Systematic sampling
     • representativeness not guaranteed
     • Avoids bias
   • Stratified sampling
     • Most representative of all sampling methods
     • Very time consuming

6. Ethics

   • Informed consent
   • Deception
   • Protection from harm
   • Privacy
   • Confidentiality
   • Right to withdraw
   • Debriefing can be used to deal with
     • Informed consent, privacy, and confidentiality (ask for permission to use participant results)
     • Deception (inform of anything that they weren't told during the study eg: aim of study)
     • Protection from psychological harm (telling them that the way they acted was completely normal)

7. Data

   • Types
     • Qualitative (words, open questions)
     • Quantitative (numbers, statistics)
   • Collection
     • Interviews
       • Structured (prepared Qs) Unstructured (no prepared Qs) Semi-structured (some are prepared, some not)
       • - difficult to analyse - people may be uncomfortable
       • - gives an extensive amount of information - can be used if someone cannot read/write
     • Questionnaire
       • Open Qs
       • Closed Qs
       • - may have leading questions - social desirability bias (people answering what they think they should be answering) - misunderstandings
       • - Can get a lot of information quicky - easier to analyse
   • Primary & secondary data

8. Case Studies

   • usually collects qualitative data, and take place over a long period of time
   • can study rare occurrences
   • - concerns unique people or events cases - cases are subjective (different researchers may get different readings)

9. Statistics

   • Range (biggest-smallest)
   • Median (middle number when all numbers are placed in ascending order)
   • Mean (add everything together, and divide by how many groups there are)
   • Mode (class group that has the most)