Week 5-Experimental Designs and Observation Methods Flashcards
Define Between-Subjects Design
■Groups are made up of different people
■Measures difference in performance between subjects (groups).
Give 2 disadvantages to Between-Subjects Design
1.Large no. of participants required
2.Participant characteristics and variables (individual differences).
E.g., Gender, age, personality, IQ, family background, level of education, etc.
Define Confounding Variables
Extraneous variable that influences the results of the study e.g., situational variables
■Have to think of constancy in Conditions
Give 3 examples of situational variables
1.Environment
2.Room temperature
3.Time of day
Give an example of expectancy effects
When someone expects a given result, that expectation unconsciously affects the outcome or report of the expected result
E.g.:
–P’s may expect exercise not to affect their weight gain
–Constancy:keep P’s naïve to the true aim of the study
–Effect not caused by expectancy.
State 3 main between-subjects designs used in balancing and matching participant characteristics/variables
Random allocation design
Matched group design
Natural groups design
Explain Random Allocation Design
–P’s randomly assigned to conditions (groups)
–Controls for participant variables:
Age, gender, occupation, IQ, smokers etc.
–Sample size in all conditions should be similar
Explain Matched Group Design
Matching participants in each group based on a certain characteristic. E.g. Gender, IQ, age, political orientation, ethnicity, smoker/non-smoker, level of education etc.
Define Within-Subjects Design
Repeatedly measuring the same people on the same DV.
Give disadvantages of Within-Subjects Design
1.Boredom/fatigue
2.Practice (Order) effects:
Learning
Fatigue
Habituation
Sensitisation
Contrast
Adaptation
What’s Incomplete Within-Subjects Design?
■Each condition was administered to each participant once.
■Order of administration varied across participants.
■Practice effects balanced across individuals.
What 2 main counter-balancing methods are used to prevent practice effects in incomplete within-subjects design?
1.Do all the possible orders. (levels of IV=number of orders/groups needed 2=2x1=2 3=3x2x1=6 etc.)
2.Selected orders. (Latin square design and each condition occurs once in each position.):
■Each condition precedes and follows each other condition only once.
Explain Complete Within-Subjects Design
■Each condition administered several times - different orders each time.
■Practice effects balanced for each participant.
Give the 2 main counter-balancing methods used in complete within-subjects design
1.Block Randomisation.
2.The ABBA design.
Explain Block Randomisation
■A block consists of all conditions.
■P’s complete the condition several times, each time in a different order:
Block 1 Block 2 Block 3
HFSN - SNFH - FHNS etc.
Every participant does all possible orders.
Explain The ABBA design
■Present one random sequence of conditions (e.g. FSHN).
■Then present the opposite sequence (NHSF).
■Each condition has the same amount of practice effects.
■If there are only 2 levels of the IV then present them in one order and then the other several times
Give an example showing the differences between Incomplete vs. Complete design
■Incomplete: either HIT then No activity or No activity then HIT NOT BOTH
■Complete: HIT then No activity and No activity then HIT. BOTH
Give 5 Limitations of the Within-Subjects Design
■Individual differences (e.g. gender, IQ)
■Time-consuming conditions
■Levels of the IV represent an unfolding time sequence that can’t be undone
■The task (DV) can’t be repeated
■Differential transfer between conditions
Define Differential Transfer
Effects of one condition affect performance in subsequent conditions.
E.g. Test on the same day in each condition:
Neutral Words – Frightening Words = No transfer
Frightening Words – Neutral Words = Transfer
What are 2 ways to avoid the transfer effect?
1.Between-subjects design
2.Within-subjects design with each condition sufficiently spaced
Define Observational Methods
■Study of animals and people within a natural environment.
■Used if practical and moral considerations mean no control of
variables.
Explain Observation Without
Intervention: Naturalistic observation
■Behaviour as it occurs naturally without intervention
■Passive recorder of behaviour
■Can be overt/ covert
Give 3 famous animal observational studies
1.Charles Darwin’s voyage on HMS Beagle.
2.Jane Goodall’s study of chimpanzees.
3.Diane Fossey’s study of mountain gorillas.
Give 4 advantages of Observation Without Intervention
■Behaviour in the real world
■High external validity
■Study complex social situations
■Developing theories.
Give 3 disadvantages of Observation Without Intervention
■Time-consuming/usually longitudinal studies
■Description not causation
■Not useful for investigating specific hypotheses.
Explain Observation with Intervention
■The observer intervenes or manipulates the event/variables in
order to:
–Create a situation which doesn’t occur frequently
–Test the impact of variables on behaviour
–Gain access to a situation/event closed to observation
■Control antecedent events and observe consequent behaviours
■Vary the qualities of a stimulus event to investigate the response
State the 3 kinds of Observation with Intervention
1.Participant Observation
2.Structured Observation
3.Field Experiments
Define Participant Observation
■Observe and participate.
2 kinds of participant observation:
– Undisguised (overt)
– Disguised (covert)
Define Undisguised (Overt) Participant Observation
■Researcher is part of the group.
■In-depth interviews and observation
What are the 3 advantages of Undisguised (Overt) Participant Observation?
■No ethical problems
■Natural setting
■Openly record data
What’s a disadvantage of Undisguised (Overt) Participant Observation?
Observer influence /
Experimenter’s bias
Explain Disguised (Covert) Participant Observation
■Those observed are unaware
■Prevents observer influence E.g. Griffin, J.H. (1961) – Racial prejudice.
What are the 2 advantages of Disguised (Covert) Participant Observation?
■Access to particular social
groups
■Natural setting/ecological validity
What are the 5 disadvantages of Disguised (Covert) Participant Observation?
■Ethical issues
■Problems recording data
■Researcher bias
■Careful training is required to identify events of interest
■Interaction with participants could affect their behaviour
Explain Structured Observation
■Cause an event or set up a situation
■Observe specific behaviour in a
particular setting
■Used by clinical/developmental
psychologists
■No attempt to control other variables
E.g.:
-Bandura, Ross & Ross (1961) - Bobo Doll
-Ainsworth (1978) - Strange Situation
■Data recorded using checklist or coding scheme
■ Mutually exclusive categories
■ Same procedures
Give 3 advantages of structured observations
■Easy for other researchers to replicate - same coding
■Data easier to analyse, less time consuming
■Easier to test larger samples – generalise findings to a population
Give a disadvantage of structured observations
Can introduce demand characteristics – when participants know they are being observed, they can modify their behaviour
Explain Field Experiments
■Controlled experiment in a natural setting
■Manipulates IV – observe the effect on behaviour
■Amount of control
Explain Qualitative Observation Data
–Natural settings, own words.
–In-depth information.
–Notes, videos, audio recordings etc.
Explain Quantitative Observation Data
–Specific, measurable behaviour.
–Checklists, coding systems, scales
etc.
Explain Inter-Observer Reliability
■Consistency in measuring
■Inter-observer/interrater
reliability
■Correlation to check the reliability
What 2 solutions can be used to improve reliability?
1.Clear categories/definitions
2.Training
Observer Influence: Explain Reactivity
Participant modifies their behaviour:
–Socially normative behaviour
Gain approval
–Demand characteristics
Expected behaviour
Explain The Hawthorne Effect Discovered by Landsberger (1950)
■When people change their behaviour simply because
they are being studied
■High vs. low lighting conditions
■Workers’ performance improved whenever lighting conditions changed, even from low to high lighting due to observation
How can we control Reactivity?
■Unobtrusive measurement
■Disguised participant observation
Adaptation:
–Habituation (decreased response to repetitive stimulus, e.g. when treating phobias)
–Desensitisation (e.g. exposure to violence may reduce anxiety)
Indirect measurement:
–Physical traces (examining remnants of past behaviour)
–Archival data (analysing previously collected data)