Small N Research Designs Flashcards
1
Q
Small n research
A
- idiographic/morphogenic
- each subject is a different experiment
- uses within subject design
- present data of individual subjects
- and comparisons between subjects
- reliability assessed by replication
2
Q
Where is this approach useful
A
- when there is assumption of minimal biological variability
- much of neuroscience
- psychophysics
- cognitive and behavioural neuroscience
- human clinical neuropsychology/neurology
- clinical psychology
- animal learning and cognition
3
Q
When is this approach useful
A
- when there are constraints: availability, convenience
- when there is no need for generalization
- assumption of low biological variability
- interest in a small layer of population
- need to understand the process in time
- detailed comparison of subjects
4
Q
Key principals of small n research
A
- consistent level: magnitude of treatment effects
- consistent trend: unidirectional changes
- stability: consistent level or consistent trend
- search of temporal patterns: time series analyses
- search of patio-temporal patterns: sequential analysis of behaviour. Serial configuration of event and actions in time
5
Q
Can it work?
A
- principle of experimental research: replication
- works due to:
- large number of observations per subjects: fluctuations average out
- factors that can contribute to variability are controlled in a strict manner
- focus on powerful variables with a clear effect: effect sizes
6
Q
Problems
A
- similar problems for a within subject design
- irreversible changes in behaviours of subjects induced by experimental procedures
- weak effects of IV on DV
- unstable DV despite strong experimental control
- uncontrolled variables
- extraneous variables
7
Q
Types of designs
A
- baseline designs: typically what research refers to when mentioning single subject
- dynamic: good for moment by moment changes in behaviour
- discrete trial design: good for a focus on performance of individual subjects (psychophysics)
8
Q
Baseline designs
A
- issue to remember: reliability- would replication produce similar results
- the behaviour is analyzed both on between and within experimental treatments data
- no averaging of data
- baseline phase: behavioural baseline needs to be established until a stability criterion is reached
- experimental phase: expose subject to treatment until criterion reached
- ABA designs: baseline, treatment, baseline etc
- intra-subject replication: subjects are their own controls
- internal validity
- inter-subject replication: testing more than 1 (3-6) subjects
- external validity (not just applicable to 1 individual)
9
Q
Baseline design issues
A
- stability criterion: choice is subjective, or based on previous research/pilot study
- removes transitional data
- fixed time or trials: give a time-limit to reach criterion
- uncontrolled variability/variation: unstable, drifting baselines caused by extraneous variables
- irreversible baselines
10
Q
Contrasting approaches in small n research
A
- group approach assumes that if experimental controls fail to reduce uncontrolled variation, then statistical methods should be used to control it
- the single-subject approach assumes that if experimental controls fail to reduce uncontrolled variation, then one should identify extraneous variables responsible for it and control them
11
Q
Replications
A
- exact/direct: part of single subject procedures
- systematic replication: above and beyond that procedure
12
Q
Problems with baselines
A
- drifting baselines: slow, systematic changes
- unrecoverable baselines: reversal doesnt work due to carryover effects
- unequal baselines: individual differences in baseline levels
- inappropriate baseline levels: low baseline is good unless data has a floor effect, and high baseline is hood unless data has Cellini good effect
13
Q
Single-subject baseline designs
A
- single factor: single independent variable. AB (necessary if irreversible changes are present), ABA, or ABAB
- multifactor: 2 or more independent variables. Eg. Melatonin and bright light
- multiple baseline: several independent variables. Eg. Polygraphs
14
Q
Dynamic designs
A
- transitional processes and behaviours are crucial and measured
- excellent for measuring behavioural/process dynamics
- also useful for continuous variations in independent variable
- dependent variable is typically continuous
- often called time series designs
- eg. Hand writing behaviour and stress/anxiety
- IV: stress condition, then non-stress condition
- eg. Galvanic skin response and stress/anxiety
- IV: questions sequentially presented
15
Q
Discrete trials designs
A
- subjects receive each treatment condition many times
- 1 trial = 1 data point
- extraneous variables strictly controlled
- randomized or counterbalance presentation of treatment to account for order effects
- inter-subject variation often analyzed: comparisons between subjects
