1 Single-Case Research Designs PS1006 Lecture 6 Sam Cromie
Dec 21, 2015
1
Single-Case Research Designs
PS1006 Lecture 6
Sam Cromie
2
Group Vs Single Case Designs
Group designs:• Average performance
of a group• Comparing average
performance between groups
• Group variability • Statistical significance
Single case designs:• Actual performance of
an individual• Comparing individual
performance in different conditions
• Individual variability• Clinical significance
3
Examples of SC Resesarch• Ebbinghaus (1885) - participant and experimenter -
first systematic evidence of forgetting over time • Freud’s psychoanalytical case studies• Behaviour Analysis – SC experiments with
pigeons, monkeys, humans• Psychophysics, study of expert performance e.g.,
chess players & musicians, • Oliver Sacks - ‘The Man Who mistook His Wife for
a Hat’
4
Single case studies - characteristics
• Intensive description and analysis of single individual.
• Data obtained through: naturalistic observation, interviews, psychological tests, experimental measurement
• May describe the application and results of a particular treatment.
5
Case Studies (CS) vs Single Case Experiments (SCE)
• Exploratory• Qualitative• May generate
hypotheses for experimental research
• Experimental manipulation
• IVs and DVs• Operational definition• Measurement• Hypotheses• 6+ participants
6
Advantages of Case studies
• Provide new ideas and hypotheses – Open the way for discoveries based on
other methodologies – Provides opportunity to develop new clinical
techniques
• Try out new clinical techniques the utility of which may only become apparent in specific cases
7
• Provides chance to study rare phenomena – Infrequent occurrences can only be examined
through intensive study e.g., • Feral children - ‘The wild boy of Aveyron’ - Victor -
Lived alone in woods from ages of 5-11/12• ‘The forbidden experiment’ - Genie
– Such cases do not offer definitive answers rather ‘obliges us to reflect on how to live with these unsolved questions’ - (Shattuck, 1994, p182)
Advantages
8
• Can support or challenge scientific theories– Falsificationism
• Genie was found at 13 never having learned to communicate due to lack of human interaction.
• Lenneberg proposed critical period of language development = 2-puberty
• Lenneberg theory could be tested by determining whether Genie could now acquire language.
• Genie showed some language development but was never completely normal
• Lenneberg’s theory at the very least should be modified after the evidence ‘provided’ by Genie
Advantages
9
• Can support or challenge scientific theories– Atkinson and Shiffrin’s multi-storey theory of
memory gets considerable support from patients who show specific breakdowns in one part of the memory system.
– H.M. could have conversation and remember events for short periods of time but could not form new memories.
Advantages
10
– Individual is more than can be represented by the collection of average values on various dimensions.
– Has the ability to reveal various nuances and subtleties of behaviour that a group approach may miss.
Advantages
11
Difficulties of Case studies• Difficulty of drawing cause-effect
conclusions – Illnesses can subside spontaneously – Other aspects of the patient-therapist
relationship may have an impact– Genie - a Doctor’s examination at 14mnths lead
to the comment that she was possibly retarded - there is no way of concluding that Genie’s disposition was a product of the poor environment which she inhabited.
12
• Biases– In interpretation– Data collection. Archival records or information based
on self-reports are particularly vulnerable.
• Lack of generalisation – Difficult to generalise from case to case. – Except where it is assumed that the underlying
physiological/behavioural systems are shared e.g. psychophysics assumes that for example visual systems are based on a shared physiological makeup
Difficulties of Case studies
13
• Public often considers personal testimony as measure of efficacy – In 1980’s Laetrile, made from apricot pits,
supposed to be beneficial to the treatment of cancer.
• By using Laetrile instead of traditional therapies, many patients may have postponed valid courses of treatment and thus contributed to the spread of their cancer.
Difficulties of Case studies
14
Single-Case Experimental Designs• Core tool of Behaviour Analysis (Skinner) • Manipulating single IVs and measuring behavioural
change in one individual• Baseline compared with intervention phases• Assumes the behavioural principles are universal,
across individuals and organisms• Graphical depiction of results • Impact of results visually rather then statistically
determined – “if the difference is not obvious it is not significant”
15
Basic SC experimental designs
• A= Baseline; B= Intervention 1; C= Intervention 2• AB – weak design• ABA – reversal procedures adds to predictive power, but
ends up with baseline• ABAB – even more predictive power and ends up with
intervention• ABC, ABAC, ABCD, etc.• Problems:
– ethics of withdrawal, – intervention may have non-reversible effects – collateral
reinforcement, verbal behaviour
16
AB designJack
0
5
10
15
20
25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
No. of Days
Inap
pro
pri
ate
Vo
caliz
atio
ns/
Hit
tin
g
Series1
BaselineToken Economy/ DRO.
A B
17
ABAB Design Experiment
18
An ABC design experiment
Junior Infants
0
20
40
60
80
100
120
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58
No of Sessions
Off
Tas
k P
erce
nt
Series1
Baseline Training Token Economy
19
Multiple baseline designs
• An independent variable is sequentially applied to at least two dependent variables
• Multiple baseline – Across behaviours– Across settings– Across participants
• Avoid problems of reversals while still demonstrating the impact of the independent variable
20
Multiple baseline design
21
Changing criterion designs
• Progressively change the level of the target behaviour required for reinforcement
• Track changes relative to the criterion
22
Changing Criterion Design
23
Readings
Concise Overview:
• Leslie & O’Reilly Behaviour Analysis: foundations and applications to psychology 1999. Chapter 8
More detail:
• Cooper,J; Heron,T; Heward,B; Applied Behaviour Analysis 2007 2nd Edition