Research Methods Flashcards
qualitative data
non-numerical
quantitative
numerical data
quantitative - advantage
easy to analyse
quantitative - disadvantage
lacks detail
qualitative - advantage
rich in detail
qualitative - disadvantage
hard to analyse
What distinguishes experimental methods from others?
an IV, DV and cause and effect
IV
deliberately manipulated by investigator
DV
measured by investigator
Participant variables
any characteristics or traits of the participant that might unfairly influence the results e.g. age, gender
Investigator effects
any cues from investigator that encourage a certain behaviour in participants e.g. facial expressions
environmental variables
any aspects of the environment or situation that might unfairly influence the results e.g. time of day, light
Demand characteristics
any cues that reveal the aims of the study to the participants
- if they become aware, may alter behaviour to help or hinder the experiment
-may just alter behaviour without intention
Standardisation
putting controls into place
-ensures all aspects are the same for every participant
-consistent standard
-environment, procedures, instructions
Randomisation
Ensures all choices are randomly selected by chance
-rather than determined by researcher
confounding variables
Any extraneous variable that is not controlled and spoils the results
Internal validity
Whether or not the research measured what it intended to
-EVs are controlled
External validity
whether results can be generalised to real life outside world
-representative
ecological validity
Real life setting
Population validity
Generalisable to other people in target population
Temporal validity
Generalisable to modern world
reliability
consistency
lab experiments
establish a cause and effect relationship between IV and DV in a controlled environment using standardised procedures
lab advantages
+high control
+easy to replicate
lab disadvantages
-demand characteristics
-low ecological validity
Field experiments
Establish cause and effect relationship between IV and DV conducted outside lab in real life environments
field advantages
+high ecological validity
+low demand characteristics
field disadvantages
-low degree of control
-difficult to replicate
natural experiments
establish a cause and effect relationship between IV and DV where IV varies naturally and is not deliberately manipulated by investigator
natural advantages
+high ecological validity
+provides opportunity for research that may not otherwise take place - e.g. unethical (age at which child is adopted) or impractical (intro of TV to society)
natural disadvantages
-random allocation to conditions is not possible - condition participants are in also occurs naturally which may be biased e.g. all people in one condition may be more intelligent or friendly
-difficult to replicate
quasi experiments
Establish a cause and effect relationship between IV and DV where IV doesn’t vary at all because the condition already exists e.g. gender, age
quasi advantages
+high ecological validity
+provides opportunity for research that may not otherwise take place - allows research where varying IV isn’t possible
quasi disadvantages
-random allocation not possible
-difficult to replicate
aims
General statement about what the researcher intends to study
Hypothesis
Precise and testable statement that states relationship between variables
Nom directional hypothesis
States that there will be an effect but doesn’t state which way that effect will go
Directional hypothesis
States which way effect will go
Pilot study
Small scale trial run
- checks research works as intended to
- no extraneous variables
Why pilot study?
1) do participants understand instructions
2)are materials and timings appropriate
3)ask a few participants about their experience
4)have variables been operationalised sufficiently
Population
Large group of individuals who share specific characteristics that a researcher is interested in studying
Sample
Smaller group that is representative of population
Bias in terms of sampling
Under or over representing certain groups within sample
Generalisation in terms of sampling
extent to which sample can be applied to population
Opportunity sampling
Anyone who is willing and available
E.g. approach on street
Opportunity advantage
+quick, convenient, economical
Opportunity disadvantages
-biased and unrepresentative
Volunteer sampling
Individuals put themselves forward
E.g. adverts in newspaper or notice boards
Volunteer advantage
+quick, convenient, economical
Volunteer disadvantage
-biased and unrepresentative
Systematic sampling
Every nth member of target population
E.g. school register
Systematic advantage
+avoids researcher bias
Systematic disadvantage
-not guaranteed to be representative e.g, could all be male - not everyone has equal chance of being selected
Random sampling
Everyone in target population has equal chance of being selected
E.g. names in hat
Random advantage
+avoids researcher bias
Random disadvantage
-not guaranteed to be representative
Stratified sampling
Sample that reflects the proportions of people in different subgroups within population
E.g. if 15% are under 18
1) identify subgroup
2) work out proportions
3) participants from each subgroup chosen randomly
Stratified advantage
+highly representative - names from hat also focus on subgroups
Reduces researcher bias
Stratified disadvantage
-time consuming and inconvenient - lots of necessary planning
Experimental design
Way in which participants are allocated to different conditions
Repeated measures
All participants take part in all conditions
Repeated measures advantages
+no participant variables
+requires half as many participants
Repeated measures disadvantages
-suffers from order effects
-high demand characteristics
Independent groups
Participants are placed into separate groups randomly and each complete one of the conditions
Independent groups advantages
+no order effects
+low demand characteristics
Independent groups disadvantages
-low degree of control over participant variables
-requires twice as many participants
Matched pairs
Different participants used for different conditions, however, participants have been matched into pairs based on important characteristics. One of each pairs takes condition A and the other takes condition B
Matched pairs advantages
+no order effects
+low demand characteristics
Matched pairs disadvantages
-less control over participant variables than repeated measures
-requires twice as many participants
Random allocation
-controls investigator bias in independent groups
-assigning all participants a number, putting numbers onto slips of paper and slips of paper into hat
-numbers are drawn so that every other number goes in condition A
Counterbalancing
-controls impact of order effects in repeated measures
-half the group does A then B, other half does B then A
-order effects are distributed evenly
Doesn’t eliminate completely
Ethical issues
-deception
-lack of informed consent
-lack of protection from harm
-lack of right to withdraw
-lack of confidentiality
Deception
Participants are deliberately misled about the true nature of the study
-informed consent cannot be gained
-participant may become distrustful of psychology studies
How to deal with deception
Debrief
Retrospective consent
Lack of informed consent
For informed consent to be given, true aims of study must be revealed before giving participants choice to agree or refuse
How to deal with lack of informed consent
Presumptive consent
Retrospective consent
Lack of protection from harm
Ensure participants come to no more harm than they would in everyday life
-both psychological (e.g. embarrassment, anxiety) and physical harm
How to deal with lack of protection from harm
Keep within ethical code
Abandon study immediately if participants are harmed
Lack of right to withdraw
Must be made aware before, during and after study that it is their right to be able to leave
How to deal with lack of right to withdraw
Retrospective withdrawal
Lack of confidentiality
The right to keep data anonymous
How to deal with confidentiality
Pseudonyms
Anonymise and don’t publish anything identifiable
Observational techniques
Involves watching and recording behaviour
Non experimental
Naturalistic
Natural situation where researcher doesn’t influence in any way
E.g. infants in nursery
Naturalistic advantages
+no demand characteristics
+high ecological validity
Naturalistic disadvantages
-difficult to replicate
-low control over variables
Controlled observation
Environment has been controlled and regulated by researcher
E.g. observing infants in pre planned situations
Controlled advantages
+easy to replicate
+high control over variables
Controlled disadvantages
-low ecological validity
-high demand characteristics
Overt observations
Participant is aware they are being observed
Overt advantage
+no ethical issues
Overt disadvantage
-high demand characteristics
Covert observations
Participants are not aware that their behaviour is being watched
Covert advantage
+no demand characteristics
Covert disadvantage
-lack of informed consent
Non-participant observations
Researcher remains separate from people they are studying and records behaviour in more objective manner
E.g. female researcher observing boys at school
Non-participant advantage
+no investigator effects
Non-participant disadvantage
-no first hand insight
Participant observations
Observer becomes part of group that they are observing
Watches behaviour from within social situation
E.g. joining criminal gang
Participant observation advantage
+first hand insight
Participant observation disadvantage
-investigator effects
Observer bias
Observer’s expectations influence what researcher sees or hears or even the data they record
E.g. expecting boys to be more aggressive than girls might lead a researcher to spend more time looking for aggression in the boys
Event sampling
Continuously watching and counting number of times event occurs
Time sampling
Recording behaviour at specific time intervals
Event sampling advantage
+doesn’t overlook important behaviour
Event sampling disadvantage
-increases observer bias
Time sampling advantage
+reduces observer bias
Time sampling disadvantage
-overlooks important behaviour
Self report techniques
Questionnaire
Interview
Questionnaire
Written questions on a topic to assess participants thoughts, feelings, opinions
Always predetermined and structured
Open and closed questions
Closed questions
Fixed number of responses
Quantitative
Open questions
No fixed answer
Qualitative
Interviews
Asking participants questions face to face
Structured and unstructured
Open and closed questions
Questionnaire advantages
+no investigator effects
+easy to gather large sample
Questionnaire disadvantages
-can’t clarify ambiguous questions
-can’t observe behaviour
Interview advantages
+can clarify ambiguous questions
+can observe behaviour
Interview disadvantage
-investigator effects
-difficult to gather large sample
Structured interviews
Predetermined set of questions in fixed order
Standardised
Unstructured interviews
Free flowing conversation with no set questions
General aim of topic discussed
Interviewee is encouraged to expand and elaborate
Structured advantages
+easy to replicate
+less investigator effects
Structured interview disadvantages
-less detail and flexibility to explore
-less trust and rapport to reduce social desirability bias
Unstructured interview advantages
+more detail and flexibility to explore
+more trust and rapport to reduce social desirability bias
Unstructured interview disadvantages
-difficult to replicate
-investigator effects
Semi structured interviews
List of predetermined questions but also freedom to explore and ask follow up questions
Correlation
analyses strength and direction of a relationship between two co variables
-neither are manipulated so no cause and effect
Co-variables
variables in a correlation that are being analysed
Correlations are plotted on a….
Scattergram
Positive Correlation
As one co-variable increases, the other increases
Negative correlation
As one co-variable increases, the other decreases
Zero correlation
No relationship
Correlation co-efficient
number that represents the strength and direction of the relationship between two co-variables
-1 = perfect negative
0 = no correlation
+1 = perfect positive
Correlation - advantages
+easy to analyse - quantitive
+further research - identifies link that opens up new lines of research
Correlation - disadvantage
- no cause and effect
-3rd variable problem - could be due to a third variable - link between stress and illness could be lack of sleep caused by stress causing illness
Primary data
Original data directly observed and collected specifically for the purpose of the investigation
Secondary data
Collected by someone other than the person conducting the research- data that already exists e.g. journals, books, websites
Meta analysis
Combining results from a number of studies
Primary data advantages
+quality control
+ensure research fits objectives
Primary data disadvantages
-time consuming and expensive
-statistical significance not known
Secondary data advantages
+quick and cheap
+statistical significance known
Secondary data disadvantages
-no quality control
-can’t ensure research fits objectives
Meta-analysis advantage
+easy to gather on large scale
Meta-analysis disadvantage
-may suffer file drawer effect - process of selecting them can be open to bias - leave out studies that don’t suit hypothesis
Mean
Add up all scores and divide by total number of scores
Median
Middle score
Mode
Most common score
Mean advantage
+Sensitive measure
Mean disadvantage
-skewed by extreme measures
Median and mode advantage
+not skewed by extreme scores
Median and mode disadvantage
-not a sensitive measure
Range
Highest subtract lowest
Standard deviation
Measures spread around the mean
-large = widely spread
-small = clustered
Range advantage
+easy to calculate
Range disadvantage
-not sensitive
SD advantage
+sensitive
SD disadvantage
-difficult to measure
Bar charts
Discrete data - categories
Space between bars - not continuous
Histograms
Bars touch - continuous data
Line graph
Continuous joined by line
Scattergram
Continuous data
Do not depict
Normal distribution
-mean median and mode are exact mid point
-distribution is symmetrical
-dispersion on either side is consistent
68.26% of people lie within 1 SD
95.44% of people lie within 2SD
Skewed distribution
Not symmetrically spread
Positive to left
Negative to right
Mode remains at midpoint
Median
Then mode
Peer review
Assessment of work by independent experts that ensures high quality
Peer review aims
1) validate quality
2) suggest changes or improvements
3) allocate funding
4) validity
Peer review strength
+stops flawed and fraudulent research
Peer review disadvantage
-slows down publication - anonymous so could use as opportunity to
-growing use of internet means less peer review
Implications for economy - memory
Cognitive interview - Kohnken
Implications for economy - attachment
Bowlbys theory of maternal deprivation
Implications for economy - approaches and psychopathology
Mental health - £22.5 billion
Implications for economy - social influence
Behavioural insight team