Research Methods Flashcards
(12 cards)
RESEARCH ISSUES
- Extraneous variables are ‘nuisance’ variables that make it harder to control the effect. Can be controlled
- Confounding variables change systematically with the IV. Cannot be sure if an effect is caused by the CV or by the IV.
- Demand characteristics refer to any cue that could reveal the aims of the investigation and influence pp behaviour
- Investigator effects refer to any effects of the investigator’s behaviour on the outcome
- Randomisation - using chance to decide aspects of the investigation to control bias
- Standardisation - ensuring the same formalised procedures are used, as any differences would become EVs
PILOT STUDIES
- A small-scale ‘trial run’ of an investigation to test procedures
- Use of control groups (independent groups design) or control conditions (repeated measures design) are used as a baseline
- Single blind = the pp doesn’t know the aims of the experiment
- Double blind = the pp and the researcher don’t know the aims of the study
INDEPENDENT GROUPS
- One group does condition A, the other does condition B
- Random allocation to experimental groups
✓ = No order effects, only tested once
✓ = Less likely to guess aims
SAMPLING
RANDOM SAMPLE - Every person in the target population has equal chance of being selected. Lottery method.
Potentially unbiased, controlling CV/EV, BUT time consuming and pps may refuse to take part.
SYSTEMATIC SAMPLE - Pps selected using a set pattern, e.g. list in alphabetic pattern, with every nth person selected from the list.
Unbiased due to random selection, but takes much time and effort, might as well use random sampling.
STRATIFIED SAMPLE - Sample reflects proportions of subgroups (strata) within the population. Relative percentages of groups (e.g. gender, age) are represented in the sample group.
Representative of the target population w/ good generalisability, but cannot fully represent.
OPPORTUNITY SAMPLE - People who are the most available, asking nearby people.
Most convenient and easy method, but it’s unrepresentative of the whole population and not generalisable.
VOLUNTEER SAMPLE - People select themselves by e.g. responding to an advert in a newspaper.
Willing participants that are more likely to engage, but volunteer bias, more likely to respond to cues
OBSERVATION
Naturalistic vs Controlled
Naturalistic - takes place where target behaviour would normally occur
Controlled - Some control and manipulation of CV/EV
Covert vs Overt
Covert - Pps are unaware they are being studied
Overt - Aware of being studied
Participant observation vs non-participant
Participant observation - Researcher becomes part of the group they’re studying
Non-Participant observation - Researcher remains separate
Event sampling vs Time Sampling
CONTENT ANALYSIS
SELF REPORT
Questionnaires:
- Avoid jargon, avoid double-barrelled questions, avoid leading questions
- Closed questions vs Open questions
Interviews:
- Structured interview vs Unstructured vs Semi-structured
- Interview schedule, a set list of Qs that can help reduce observer bias
- Quiet room
- Building of rapport
- Remind of ethical considerations
DATA
Qualitative = non-numerical data expressed in words
Qualitative = numerical data
Primary data = first hand data collected for the purpose of the investigation
Secondary data = collected by someone other than the one doing the research, e.g. other psychologists
Meta-Analysis = involves combining data from a large number of studies
DISTRIBUTIONS
Negative skew = leaning right, mean dragged to the left
Positive skew = leaning left, mean dragged to the right
CORRELATION
- The closer the coefficient is to 1 (+1 or -1), the stronger the relationship is between the variables.
- BUT, even if weak, can still be statistically significant
- +1 is perfect positive correlation /
- -1 is perfect negative correlation \
- +.5 = -.5
