Doctorate in Educational and Child Psychology
Cara Southworth
Case Study 1: An Evidence-Based Practice Review Report
Theme: Interventions for children with Special Educational Needs
From a parent’s perspective, is cognitive-behavioural treatment an effective
intervention for children with Attention Deficit Hyperactivity Disorder?
Summary
Cognitive-behavioural treatment is a theoretical and evidence-based intervention that
addresses dysfunctional thoughts and attributions, maladaptive behaviours and
cognitive processes through goal-orientated and systematic procedures in either
group or individual therapy sessions. Cognitive-behavioural treatment aims to teach
children with Attention Deficit Hyperactivity Disorder (ADHD) strategies to help them
increase their self-control and problem-solving abilities, through modelling, role
playing and self-instruction.
effectiveness
regarding
Cognitive-behavioural treatment has shown mixed
ADHD
behaviours,
yet
still
remains
a
consistent
recommendation of the National Collaborating Centre of Mental Health (NICE)
guidelines when treating children and young people with ADHD. In light of these
discordant views, combined with the fact that some research is now dated, a
reconsideration of the area is necessary. In particular, this review will focus on the
parental perspective. Five studies were evaluated in this review. All five studies
found that the intervention had a positive effect on ADHD behaviours from the
parent’s point of view. Those that followed up at later points also reported
encouraging results. A number of issues including methodological quality and
generalisability are raised, but overall it is recommended that cognitive-behavioural
treatment continue to be considered appropriate for treating children with ADHD.
1
Doctorate in Educational and Child Psychology
Cara Southworth
Introduction
Essential elements of cognitive-behavioural treatment
The cognitive-behavioural model highlights the central role of thoughts and
attributions in understanding an individual’s emotional and behavioural life. The core
techniques involve helping individuals to identify patterns of thinking that interfere
with their optimal functioning (Levine & Anshel, 2011). No single set of methods
defines cognitive-behavioural treatment. However, treatment is governed by two
overarching main themes: the conviction that cognitive processes influence emotion,
motivation and behaviour and the use of cognitive and behaviour-change techniques
in a pragmatic (hypothesis-testing) manner (Butcher, Mineka, & Hooley, 2010). The
paradigm also includes an emphasis on education, arguing that giving people
information about their diagnosis encourages the development of independent
problem-solving skills (Flanagan & Miller, 2010).
Cognitive-behavioural treatment in relation to children with ADHD
Due to some limited effects of psychostimulant medications on decreasing the
cardinal symptoms of ADHD (Hechtman, Weiss, & Perlman, 1984; Gittelman &
Kanner, 1986) researchers explored alternative treatment approaches such as
cognitive-behavioural treatment. Carried out by a therapist, cognitive-behavioural
treatment sessions attempt to teach children strategies to help them increase their
self-control and problem-solving abilities, through modelling, role playing and selfinstruction (Kendall & Braswell, 1985; Kendall, Padever, & Zupan, 1980) to: (1)
define the nature of the problem; (2) reflect on all the possible solutions; and (3)
choose one solution and evaluate its outcome (Kendall & Braswell, 1985).
2
Doctorate in Educational and Child Psychology
Cara Southworth
Most current models of ADHD emphasize the deficiencies in the executive
functioning skills of behavioural inhibition and self-regulation (Barkley, 2006). These
deficiencies in turn manifest themselves in overt behaviours, such as sustaining
attention to academic tasks and inhibiting excessive motor activity. Cognitivebehavioural proponents recognise the goodness of fit between these highlighted
deficits in children with ADHD and the self-regulatory focus of cognitive-behavioural
treatment (Hinshaw & Melnick, 1992). Relatedly, the proponents of cognitivebehavioural approach contend that the maintenance of treatment gains can be
achieved only through teaching a generalised set of cognitive mediational (self-talk)
skills that children with ADHD can internalise.
Basis in psychological theory
The cognitive-behavioural perspective can take a variety of forms and draws on a
range of psychological theories. For example, Bandura’s theory of self-efficacy - the
view that one can achieve desired goals (1986, 1997) and that the beliefs individuals
hold about their capabilities have a strong influence on the ways in which they
behave (Usher & Pajares, 2008) - is an early example of a cognitive-behavioural
perspective. He posited that cognitive-behavioural treatments work in large part by
improving self-efficacy.
Today, the cognitive-behavioural paradigms often follow the model proposed by
Beck (1964), which emphasises the central role of thoughts and attributions in
understanding an individual’s emotional and behavioural life. Beck’s model makes
the assumption that problems result from biased processing of external events or
internal stimuli. These biases distort the way that person makes sense of the
experiences he or she has in the world, leading to cognitive errors. Beck (2005)
argues that underlying these biases is a relatively stable set of schemas that contain
3
Doctorate in Educational and Child Psychology
Cara Southworth
dysfunctional beliefs. When the schemas become activated they bias information
processing. The central focus in treatment is therefore to alter distorted and
maladaptive cognitions and their underlying schemas; this is achieved through
making individuals aware of and exploring the connections between thoughts and
emotional responses and later learning and practicing strategies to better deal with
difficult external events or internal stimuli. In the case of ADHD for example, the
cognitive-behavioural approach helps pupils to understand links between thoughts,
feelings and behaviours and that these may result in unhelpful, inappropriate or
maladaptive consequences. The therapy also explores learning to change these
thoughts, feelings and behaviours to produce more desirable outcomes (NICE,
2009).
Rationale
ADHD is repeatedly singled out as a challenge for teachers, parents and
psychologists, because of the behavioural characteristics commonly demonstrated
by children with this disorder (Langberg, Froelich, Loren, Martin, & Epstein, 2008).
Because ADHD is associated with poor peer relations and negative self-image
(Horn, Ialongo, Greenberg, Packard, & Smith-Winberry, 1990), children with ADHD
are at greater risk as adolescents and adults of several social and emotional
problems, including substance abuse and depression (Gray, Riggs, Min, MikulichGilbertson, Bandyopadhyay, & Winhusen, 2011; Tamm, Trello-Rishel, Riggs,
Nakonezny, Acosta, Bailey, & Winhusen, 2013), and the frequency of academic and
career problems is significantly higher in these children compared with the normal
population (Jensen, Mrazek, Knapp, Steinberg, Pfeffer, Schowalter, & Shapiro, 1997;
Springer, Phillips, Phillips, Cannady, & Kerst-Harris, 1992).
4
Doctorate in Educational and Child Psychology
Cara Southworth
Findings on the effectiveness of cognitive-behavioural interventions on ADHD
behaviours have been mixed. Literature demonstrates the limitations of cognitivebehavioural interventions in directly targeting central ADHD impairments (Abikoff,
1987; Abikoff, 1991). Indeed, despite some early claims of success (e.g. Cameron &
Robinson, 1980), systematic investigations aimed at comparing the benefits of
cognitively based interventions with stimulant medications have demonstrated the
superiority of the latter (Abikoff, Ganeles, Reiter, Blum, Foley, & Klein, 1988).
However, other studies show that cognitive-behaviour techniques are effective in
moderating impulsivity (Kendall & Braswell, 1982). A recent review by MunozSolomando, Kendall, & Whittington (2008) also suggests that cognitive-behavioural
interventions can have beneficial effects delivered in absence of medication or as
adjunct to continued routine medication for children with ADHD. Furthermore, current
NICE guidelines retain their support for cognitive-behavioural interventions for
children with ADHD; they consider group-based cognitive-behavioural interventions
to be both effective with children with ADHD and cost efficient. In light of these
discordant views, combined with the fact that many reviews are now dated, a
reconsideration of the area is necessary.
The parent’s role in supporting the child is crucial, yet literature highlights that
parents of children with ADHD often struggle to manage their child’s problems,
suffering from stress and exhaustion (Green, McGinnity, Meltzer, Ford, & Goodman,
2005). Further, recent legislation has highlighted the importance of involving parents
in treatment options and the treatment process, making the parent’s perspective of
particular relevance to current practice.
As such, this review will focus on the
parent’s perspective.
5
Doctorate in Educational and Child Psychology
Cara Southworth
Review Question
From a parent’s perspective, is cognitive-behavioural treatment an effective
intervention for children with Attention Deficit Hyperactivity Disorder?
Critical Review of Evidence Base
Literature Search
The current review was carried out in December 2013. PsychInfo, ERIC and Medline
(Ovid) databases were searched using the search terms:

cognitive-behavioural* or cognitive-behavioral* or cognitive behavioural* or
cognitive behavioral* or CB*

AND attention deficit hyperactivity disorder OR ADHD

AND adolescen* OR child* OR youth*
The use of both UK and US written English was used to ensure relevant studies
were not overlooked. These search terms, in the abstract, gave a total of 1,054
results (531, 334, and 189 respectively for each database) some of which
overlapped between databases. Of these 1,044 were rejected by title or abstract, ten
remained to inspect in full. Five were excluded based on the inclusion criteria in
Table 1; please see Appendix A and C for the specific reasons for exclusion, as well
as a rationale for each inclusion criterion. Five studies remained from this process.
Ancestral searches were carried out on these five studies to further ensure that
relevant papers were not overlooked. Four studies were highlighted from the title
and inspected in full. All four were excluded based on the inclusion criteria (please
see Appendix D). The full literature search is illustrated in a flow diagram (see
Appendix B). The five studies included in this review can be viewed in Table 2 and
are summarised in Appendix E.
6
Doctorate in Educational and Child Psychology
Cara Southworth
Table 1
Inclusion and Exclusion Criteria
1. Intervention
Inclusion Criteria
Exclusion Criteria
A cognitive-behavioural approach
targeted at ADHD delivered to child.
Not a cognitive-behavioural
approach targeted at ADHD or
delivered to child.
Ability to isolate cognitivebehavioural treatment effect on
dependent measures from other
interventions that do not target
ADHD or are not delivered to child.
2. Variable
A variable measuring ADHD
symptoms from the parent
perspective
Unable to isolate cognitivebehavioural treatment effect on
dependent measures from other
interventions that do not target
ADHD or are not delivered to child.
No variable measuring ADHD
symptoms from the parent
perspective
3. Participant Age
19 and under
Over 19
4. Participant Diagnosis
A diagnosis of ADHD, diagnosed by
DSM-III-R – DSM 5 or European
equivalent (e.g. ICD-10).
No diagnosis of ADHD, diagnosed
by DSM-III-R – DSM 5 or European
equivalent (e.g. ICD-10).
5. Participant Other
Can be on medication for ADHD
On medication for other disorders
No diagnosis of other considerable
and/or distinct disorders
Diagnosis of other considerable
disorders.
6. Type of study
Empirical study containing primary
data.
A review, meta-analyses, pilot
studies or use secondary data.
7. Publication type
Published in a peer reviewed
journal.
Unpublished work or grey literature
such as personal websites or nonpeer reviewed journals.
8. Language
Full article available in English
Full article not available in English
9. Date
Must be published from 1987 to
December 2013
Published before 1987 or after
December 2013
7
Doctorate in Educational and Child Psychology
Cara Southworth
Table 2
Studies Eligible for the Review
Studies Included in the Review
1. Abdollahian, E. Mokhber, N., Balaghi, A., & Moharrari, F. (2013). The effectiveness of
cognitive-behavioural play therapy on the symptoms of attention-deficit/hyperactivity disorder
in children aged 7-9 years. ADHD Attention Deficit and Hyperactivity Disorders, 5 (1), 41 – 46.
2. Fehlings, D. L., Roberts, W., Humphries, T., & Dawe, G. (1991). Attention deficit hyperactivity
disorder: Does cognitive behavioral therapy improve home behavior? Developmental and
Behavioral Pediatrics, 12 (4), 223 – 228.
3. Froelich, J., Doepfner, M., & Lehmkuhl, G. (2002). Effects of combined cognitive behavioural
treatment with parent management training in ADHD. Behavioural and Cognitive
Psychotherapy, 30, 111 – 115.
4. Horn, W. F., Ialongo, N., Greenberg, G., Packard, T., & Smith- Winberry, C. (1990). Additive
effects of behavioral parent training and self-control therapy with attention deficit hyperactivity
disordered children. Journal of Clinical Child Psychology, 19 (2), 98 – 110.
5. Miranda, A., & Jesús Presentación, M. (2000). Efficacy of cognitive-behavioral therapy in the
treatment of children with ADHD, with and without aggressiveness. Psychology in the
Schools, 37 (2), 169 – 182.
Weight of evidence
Gough’s (2007) guidance and a coding protocol (adapted from Kratochwill, 2003 see appendix G) were used to assess the relevance of each paper to the review
question. The articles were coded only for the measures that were relevant to
answering the review question. The weighting given to each study included in this
review is shown in Table 3. The outcomes examined in this review are child
behavioural outcomes associated with ADHD, reported by parents.
8
Doctorate in Educational and Child Psychology
Cara Southworth
Table 3
Weight of Evidence (WoE)
Study
WoE A: Quality
of Methodology
WoE B:
Relevance of
Methodology
WoE C:
Relevance of
Evidence to the
Review
Question
WoE D: Overall
Weight of
Evidence
Abdollahian et al.
(2013)
Medium
Medium
Medium
Medium
Fehlings et al.
(1991)
Medium
High
High
High
Froelich et al.
(2002)
Low
Low
Medium
Low
Horn et al. (1990)
Low
High
Medium
Medium
Miranda and
Jesús
Presentación
(2000)
Low
Low
Low
Low
Participants
The sample sizes ranged from 18 participants (Froelich, Doepfner, & Lehmkuhl,
2002) to 60 (Horn et al., 1990). However, due in part to portions of study samples
having to be excluded (Horn et al., 1990; Miranda & Jesús Presentación, 2000);
none of the studies included in this review achieved a ‘sufficiently large N’ (at alpha
level .05 and a medium effect size). This suggests that the number of participants in
each condition is not necessarily sufficient to yield enough statistical power for
detecting effects of cognitive-behavioural treatment on symptoms of children with
ADHD. Caution is therefore needed when drawing conclusions from study findings.
But in light of the specialised population utilised in all the studies, it is not surprising
that sample sizes did not reach above 30 on average. The ages of the participants
ranged from 6 to 13 years old. Two of the studies did not state the gender of the
participants (see Table 4), and of the remaining studies 81.7% of the sample were
9
Doctorate in Educational and Child Psychology
Cara Southworth
males and 18.3% were females. This difference supports the documented
disproportion of males with ADHD compared to females (Gaub & Carlson, 1997).
All participants included in this review had a diagnosis of ADHD (diagnosed from the
DSM-III-R or subsequent DSM manuals, or European equivalent). Types of ADHD
were not controlled for in this review. Three studies contained a sample of children
diagnosed with ADHD alone (in some studies absence of other disorders was made
explicit, in others it was not), Abdollahian, Mokhber, Balaghi, & Moharrari’s (2013)
sample contained only participants on medication for ADHD and Froelich et al.’s
(2002) sample contained two participants with learning difficulties and eight on
medication for ADHD. Miranda and Jesús Presentación (2000) included children with
aggressiveness but this part of the sample was not looked at in this review. Children
in this study were also receiving anger management therapy (AM) for their ADHD
symptoms.
As Table 4 shows, no trend was found between studies that contained a sample that
included children with learning difficulties and children on medication, and those that
did not. The WoE C took into account the sample characteristics; in order to achieve
a high weighting the sample of interest had to be typically developing children with
the appropriate diagnosis of ADHD who were not receiving additional interventions
aimed at ADHD. Table 4 shows that two of the studies successfully adhered to this
criterion. Fehlings, Roberts, Humphries, & Dawe (1991) had a highly relevant sample
and obtained typically large effect sizes between pre and post-test phases (on
significant findings). Horn et al. (1990) similarly had a relevant sample, but found
only a medium effect size over time from the Child Behaviour Checklist (CBCL)
parent questionnaire.
10
Doctorate in Educational and Child Psychology
Cara Southworth
Table 4
Key Features of the Review Studies
Study
Participants
Research
Design &
Intervention
Details
Outcome
Measuresª
Post
Intervention
Effect Sizesᵇ
Abdollahian et al.
(2013)
7-9, ADHD
(all on
medication)
Quasi-experimental
with control/
CB play therapy
Rutter rating of ADHD
symptoms
0.8
Large
Medium
Fehlings et al. (1991)
7-13 (all m),
ADHD
diagnosis
only
RCT/ CBT
WWAS assessing
hyperactivity
0.98
Large
High
SCRS assessing
impulsivity (n/s)
0.77
Medium
BPC-AP assessing
inattention (n/s)
0.65
Medium
Yale Children’s
Inventory assessing
core ADHD symptoms
u
Yale Children’s
Inventory assessing
conduct problems
u
Home Situations
Questionnaire for
conflict at home
u
Froelich et al. (2002)
6-12 (17/18
m), ADHD
diagnosis,( 2
learning
difficulties, 8
medication)
Pre- & Post- test
(no control)/ CBT +
focus on
generalisation skills
Individual Problem List
to assess individually
relevant problem
behaviour
Horn et al. (1990)
Miranda and Jesús
Presentación (2000)
7 – 11 ½ 9
(42m, 18f),
ADHD
diagnosis
only
9 – 12 ADHD
diagnosis,
receiving AM
as well
RCT/ CB selfcontrol therapy
Pre- & Post- test
(no control)/
CB self-control (+
AM)
11
WoE D
Low
u
Homework Problem
Checklist
u
CBCL CB SC (Total
Problems scale)
0.62
Medium
CBCL CB SC
(Externalising scale)
0.58
Medium
CBCL CB SC
(hyperactivity scale)
(n/s)
0.17
Small
Assessment of
hyperactivity on DSMIII-R (non-aggressive
ADHD)
u
Assessment of
antisocial behaviour on
EPC (non-aggressive
ADHD)
u
Assessment of school
problems on EPC (non-
u
Medium
Low
Doctorate in Educational and Child Psychology
Study
Participants
Research
Design &
Intervention
Details
Outcome
Measuresª
Cara Southworth
Post
Intervention
Effect Sizesᵇ
WoE D
aggressive AD
Assessment of
internalisation on EPC
(n/s)
u
Assessment of
psychopath disorders
on EPC
u
Note.ª n/s refers to a non-significant outcome; ᵇ ‘u’ is marked when effect size is unable to be calculated.
The studies included in this review took place in five different countries: Iran,
Canada, Germany, USA and Spain. As studies needed participants with a specific
condition, they did not take measures to ensure participants came from a diverse
range of socio-economic backgrounds, although some studies did report information
on the socio-economic status of their participants. When this information was
provided it indicated that the vast majority of participants were white and resident to
the country of testing. In addition, the samples were small and gender balance was
skewed. Generalisations to non-white ethnic populations and females are therefore
limited. There is no clear reason why benefits of a cognitive-behavioural approach
should be limited to white and male populations, however. Indeed, Abdollahian et al.
(2013), Horn et al. (1990) and Miranda and Jesús Presentación (2000) were able to
appropriately deliver cognitive-behavioural treatment to children of both sexes and of
varying socio-economic levels and ethnicities. Finally, although two of the studies
were carried out in English speaking countries (US and Canada (Toronto), none of
the papers reviewed were done in the UK. Caution should be taken therefore before
committing resources to implement cognitive-behavioural treatment to support
children with ADHD in the UK.
12
Doctorate in Educational and Child Psychology
Cara Southworth
Research Design
Fehlings et al. (1991) and Horn et al. (1990) employed a randomised controlled trial,
both with active control groups (alternative treatments: supportive therapy and parent
behavioural training respectively). Abdollahian et al. (2013) employed a quasiexperimental study design, and Miranda and Jesús Presentación (2000) and
Froelich et al. (2002) employed a pre-test-post-test design, though only Abdollahian
et al. (2013) set up a control group comparison. All but Abdollahian et al. (2013) and
Froelich et al. (2002) had additional testing at a follow-up time (ranging from 2 to 8
months).
Individual participants were the unit of randomisation in the randomised controlled
trials, however for the other three studies participants were either matched
(Abdollahian et al., 2013) and then allocated to a treatment group, or the allocation
procedure was not made explicit: Miranda and Jesús Presentación (2000) gave little
information as to how participants were allocated to each condition and did not have
an appropriate control group, and Froelich et al. (2002) did not have a control group.
The results of these studies are therefore more likely to be due, in part, to
confounding extraneous variables. It is possible that effects observed were due to
children being in an intervention group of any sort rather than being involved in
cognitive-behavioural treatment. As a result, they obtained a ‘high’ in methodological
quality or relevance, or a ‘high’ overall (WoE D).
13
Doctorate in Educational and Child Psychology
Cara Southworth
Intervention
Table 5
Comparison of Cognitive-Behavioural Interventions
Study
Delivery/Length
/Duration
Procedure
Target Behaviours
Abdollahian et al. (2013)
Therapist - group
Cognitive-behavioural play
sessions conducted with
different thematic activities.
Attention, working memory,
maintenance, impulsivity,
self-control and
organisation.
Direct instruction on CB
strategies: Problem solving
broken down to five step
process.
Home behaviour
Predominantly included
training in self-instructional,
self-monitoring and problemsolving skills. With a focus on
generalisation of skills learnt.
Core symptoms of ADHD
and conduct problems.
Instruction in the self-control
strategies adapted from Camp
and Bash (1981), Kendall and
Braswell (1985) and
Meichenbaum (1977). Each
child was taught a “Problem
Solving Plan”.
ADHD cardinal features (e.g.
externalising and
hyperactivity).
Applied Kendall et al.’s “Stop
and Think” program (1980)
with small modifications.
Improve concentration and
reflection.
Eight sessions twice weekly
45 min
Fehlings et al. (1991)
Therapist – individual & family
Twelve sessions twice weekly
& eight sessions at home
every 2 weeks
60 min & 2 hours
Froelich et al. (2002)
Therapist - individual
Six lessons one per week
60 min
Horn et al. (1990)
Therapist – group
12 sessions weekly
90 min
Miranda and Jesús
Presentación (2000)
Therapist – group
22 session over 3 months
60 min
The studies varied in regard to the cognitive-behavioural treatment they employed
(see Table 5). Four of the studies however, implemented cognitive-behavioural
treatment that instructed children in similar strategies. These included: problemsolving, self-instructional training, self-monitoring and self-control therapy. The
studies used similar methods of training, such as modelling, role playing and guided
practice (Fehlings et al., 1991; Horn et al., 1990; Miranda & Jesús Presentación,
2000).
Several
also
incorporated
problem-solving
14
of
both
academic
and
Doctorate in Educational and Child Psychology
Cara Southworth
interpersonal problems. Froelich et al. (2002) also reported discussing individually
relevant academic or social problems. Home assignments and token reinforcement
systems were used in all four of the studies mentioned. Parents were taught about
CBT in Fehlings et al.’s (1991) and Froelich et al.’s (2002) studies. Finally, particular
efforts to facilitate generalisation of the new skills taught were made by Froelich et al.
(2002). Froelich et al. achieved significant findings across all measures and the
authors felt that emphasis on generalisation of newly acquired skills played a key
part in achieving this. Abdollahian et al. (2013), in contrast to the other studies,
implemented cognitive-behavioural play therapy. Sessions conducted varied in
thematic activities, but the target behaviours were similar, including attention,
impulsivity and self-control.
Horn et al. (1990) and Fehlings et al. (1991) reported steps to ensure faithfulness to
the intervention. It was noted that all therapists were extensively trained, treatment
manuals were developed, supervision was provided on a weekly basis and all
treatment sessions were audiotaped. In the other three studies there was missing
information about intervention fidelity. The interventions were carried out by
therapists in all five of the studies. Further details were provided only by two of the
studies: Horn et al. (1990) reported that treatment groups were co-led by two
advanced level graduate students carrying out clinical psychology training, and eight
therapists then provided the treatment in pairs; Miranda and Jesús Presentación
(2000) reported that the interventions were carried out by therapists with PhDs in
psychology who were experts in the ‘Stop and Think’ programme. Taken alongside
missing integrity to interventions information, concern is warranted over the
consistency of intervention delivery. All the above highlights that there is a lack of
conformity between interventions across the five studies within this review; as a
15
Doctorate in Educational and Child Psychology
Cara Southworth
result it is difficult to establish specific aspects of the cognitive-behavioural treatment
that could be considered particularly effective and care is required when drawing
conclusions.
Outcome Measures
All of the studies included in this review used questionnaires to gather information
from parents. Every study used different questionnaires to measure ADHD
symptoms and related behaviour however. Froelich et al. (2002) used an
abbreviated German version of the Yale Children’s Inventory to measure parents
rating of core ADHD symptoms. Recent studies have found the reliability coefficient
for this scale to be high (.94 for internal consistency for the total score). These
authors also calculated high reliability coefficients for the other rating scales used in
their studies (>.90). Abdollahian et al. (2013) reported that the Rutter Parental
Questionnaire used in their study has been found to have an internal validity and
reliability of .74 (Rutter, 1976). Finally, Miranda and Jesús Presentación (2000)
reported an alpha coefficient of .88 and a test-retest correlation of .79. However,
Miranda and Jesús Presentación (2000) neglected to provide information about
reliability and validity on the second measure they used (adapted DSM-III-R), and
the remaining two studies neglected to mention it entirely. As a result, reported
effects on these measures should be interpreted with caution.
However, many of the measures used in the remaining two studies that neglected to
provide information about reliability and validity are widely used, for example the
CBCL and the Self-Control Rating Scale (SCRS), and reports of their reliability and
validity elsewhere are satisfactory (Barkley, 1990; Chen, Faraone, Biederman, &
Tsuang, 1994; Kendall & Wilcox, 1979). Although interparent interrater reliability was
reported to be good in the Werry Weiss Activity Scale (Barkley, 1988), no information
16
Doctorate in Educational and Child Psychology
Cara Southworth
regarding the test-retest reliability or internal consistency was located and no
information was found for the Behaviour Problem Checklist-Attention Problem
Subscale (BPC-AP). Further, all measures included in this study were parent reports
via rating scales. Parental information was the only outcome of interest in this
review, but it should be considered that the lack of multi-method data collection in
the studies that employed only one questionnaire can lead to a single dimension
conceptualisation of outcomes.
Findings
In all five of the studies, the effectiveness of cognitive-behavioural treatment on
ADHD symptoms in children with ADHD, as measured by parents, was determined
by the interactive effect between groups (intervention vs. control) and/or over time
(pre vs. post intervention) (refer to Appendix E). Significant interactions were found
in all studies, but not across all interactions and measures (See Table 4 for post
intervention interactions). All five papers found significant interactions, on some or all
measures, over time in the experimental group: Abdollahian et al. (2013) and
Fehlings et al. (1991) with large effect sizes, and Horn et al. (1990) with medium
effect sizes. Fehlings et al. (1991), Horn et al. (1990) and Miranda and Jesús
Presentación (2000) also reported significant differences at follow-up times. Nonsignificant findings were found over time in the control groups in two out of the three
studies that had appropriate control groups. Horn et al. (1990) did find a significant
difference in the control group over time, but this is likely due to the active
‘alternative’ treatment in this control group (parental behavioural training).
Table 6 aims to break down the outcomes looked at by the five studies into four
categories: total (ADHD) problems, inattentiveness, hyperactivity and impulsiveness
(& conduct). These categories were developed to reflect the outcomes measured in
17
Doctorate in Educational and Child Psychology
Cara Southworth
the five studies included in this review and the three specific behaviours commonly
used as behavioural components of ADHD. As can be seen in the table, the studies
that reported total problems as their outcomes (Abdollahian et al., 2013; Froelich et
al., 2002; Horn et al., 1990) had medium to large effect sizes, where effect sizes
could be calculated. Unfortunately, although many outcomes were reported, only a
selection of effect sizes could be calculated. Fehlings et al. (1991) found a significant
interaction between cognitive-behavioural treatment and hyperactivity over time with
a large effect size. Finally, Horn et al. (1990) reported a significant interaction
between externalisation and cognitive-behavioural treatment over time but with only
a medium effect size.
18
Doctorate in Educational and Child Psychology
Cara Southworth
Table 6
Comparison of ADHD Behavioural Symptoms
Outcome measure
Significanceª
Effect sizeᵇ
(post
intervention)
WoE D
Abdollahian et al. (2013)
ADHD symptoms
s
0.80
Medium
Froelich et al. (2002)
Core ADHD
symptoms
s
u
Low
Horn et al. (1990)
Total Problems
s
0.62
Medium
n/s
0.65
High
Study
Total behaviours
Specific to one of the three core ADHD behaviours:
Inattentiveness
Fehlings et al. (1991)
Inattention
Hyperactivity
Fehlings et al. (1991)
Hyperactivity
s
0.98
High
Horn et al. (1990)
Hyperactivity
n/s
0.17
Medium
Miranda and Jesús Presentación (2000)
Hyperactivity
s
u
Low
n/s
0.77
High
Impulsiveness (& conduct)
Fehlings et al. (1991)
Impulsivity
Froelich et al. (2002)
Conduct Problems
s
u
Low
Froelich et al. (2002)
Conflict at home
s
u
Low
Froelich et al. (2002)
Homework Problems
s
u
Low
Froelich et al. (2002)
Individually relevant
problem behaviour
s
u
Low
Horn et al. (1990)
Externalising
s
0.58
Medium
Miranda and Jesús Presentación (2000)
Anti-social behaviour
s
u
Low
Miranda and Jesús Presentación (2000)
School Problems
s
u
Low
Miranda and Jesús Presentación (2000)
Internalisation
n/s
u
Low
Miranda and Jesús Presentación (2000)
Psychopath Disorder
s
u
Low
Note.ª n/s refers to a non-significant outcome, s refers to a significant outcome; ᵇ ‘u’ is marked when effect size was unable
to be calculated.
19
Doctorate in Educational and Child Psychology
Cara Southworth
Conclusions and Recommendations
Conclusions
Although grounded in theory, cognitive-behavioural treatment has shown mixed
effectiveness for children with ADHD (Abikoff, 1987; Abikoff, 1991). Cognitivebehavioural interventions continue to be recommended by the NICE, however. This
systematic review aimed to evaluate the effectiveness of cognitive-behavioural
treatments on the behaviours of children with ADHD from the perspective of the
parent.
Two of the studies achieved overall weightings of medium quality, with one receiving
high (Fehlings et al., 1991) and two receiving low (Miranda & Jesús Presentación,
2000; Froelich et al., 2002). Fehlings et al. received a high due to both the relevance
of the methodology and to the review question. Many of the specific weightings
within studies were medium, but Miranda and Jesús Presentación received a low on
all three weightings. Further, Froelich et al. (2002) received a low for both
methodological relevance and methodological quality, and Horn et al. (1990) for
methodological quality.
All the studies included in this review reported that cognitive-behavioural intervention
had some positive effect on ADHD behaviours from the parent’s point of view. When
weight of evidence rating is taken into account, measures assessing overall ADHD
problem behaviours appeared to find the most convincing results, with medium to
large effect sizes. Fehlings et al.’s highly controlled study found a large effect on
hyperactivity, and other effects were calculated, including externalisation (medium).
It should be noted that only a few effect sizes were able to be calculated in this
review, which places restrictions on possible conclusions and recommendations.
20
Doctorate in Educational and Child Psychology
Cara Southworth
This review indicates that parents report that cognitive-behavioural treatment has a
positive impact on the ADHD related behaviours of their children. However, in light of
the limitations noted, conclusions should be interpreted carefully. The limited
methodological quality and relevance of the studies included in this review and the
mixed findings of the studies with sound evidence suggest that evidence is limited.
Recommendations
Based on the findings of this review the recommendations are as follows:
1. Although many of the studies used controlled group experimental designs or
took some appropriate steps to ensure some methodological quality, research
is needed into the effectiveness of cognitive-behavioural treatment on pupils
with ADHD using robust and tested research methods such as group
experimental designs.
2. Four out of the five studies were conducted over 10 years ago. Current
research is needed to be able to generalise findings to the current population.
3. All of the studies were conducted outside of the UK, which creates limitations
for generalisability. Future research is needed on using cognitive-behavioural
interventions with children with ADHD in the UK.
4. Cognitive-behavioural interventions can be considered an effective treatment
for children with ADHD from the parent’s point of view. It should therefore
continue to be a recommended treatment by NICE guidelines and by
educational psychologists, especially when considering NICE’s conclusion
that it is cost effective. However, considering the mixed reliability of the effects
reported and the varying impact on specific ADHD behaviours other
treatments should be considered: cognitive-behavioural treatment does not
necessarily produce the largest reductions in ADHD behaviours.
21
Doctorate in Educational and Child Psychology
Cara Southworth
5. Although excluded in this review, in light of the high frequency of comorbidity
of ADHD with other disorders, future studies should explore cognitivebehavioural treatment effects on children with ADHD as well as other distinct
disorders. It would be of interest to explore if differing diagnoses effects the
impact of cognitive-behavioural treatment, for example by comparing children
with ADHD alone and children with ADHD and an anxiety disorder
comparison.
22
Doctorate in Educational and Child Psychology
Cara Southworth
References
Abdollahian, E. Mokhber, N., Balaghi, A., & Moharrari, F. (2013). The effectiveness
of cognitive-behavioural play therapy on the symptoms of attentiondeficit/hyperactivity disorder in children aged 7-9 years. ADHD Attention Deficit
and Hyperactivity Disorders, 5(1), 41 – 46.
Abikoff, H. (1987). An evaluation of cognitive behaviour therapy for hyperactive
children. In B.B. Lahey & A.E. Kazdin (Eds.), Advances in clinical child
psychology (pp. 171-216). New York: Plenum.
Abikoff, H. (1991). Cognitive training in ADHD children: Less to it than meets the
eye. Journal of Learning Disabilities, 24, 205-209.
Abikoff, H., Ganeles, D., Reiter, G., Blum, C., Foley, C., & Klein, R. G. (1988).
Cognitive training in academically deficient ADHD boys receiving stimulant
medication. Journal of Abnormal Child Psychology, 16, 411-432.
Bandura, A. (1986). Social foundations for thought and action: A social
cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W.H.
Freeman/Times Books/Henry Holt & Co.
Barkley, R. A. (1988). Attention deficit disorder with hyperactivity. In E.J. Mash & L.J.
23
Doctorate in Educational and Child Psychology
Cara Southworth
Terdal (Eds.), Behavioral assessment of childhood disorders (pp. 66-104).
New York: Guildford.
Barkley, R. A. (1990). Attention deficit hyperactivity disorder: A handbook for
diagnosis and treatment. New York: Guildford.
Barkley, R. A. (2006). Attention-deficit hyperactivity disorder: A handbook for
diagnosis and treatment (3rd ed.). New York: Guilford.
Beck, A. T. (1964). Thinking and depression II: Theory and therapy. Archives of
General Psychiatry, 10, 561 – 571.
Beck, A. T. (2005). The current state of cognitive therapy: A 40-year retrospective.
Archives of General Psychiatry, 62, 953-959.
Butcher, J. N., Mineka, S., & Hooley, J. M. (2010). Abnormal Psychology. Boston:
Allyn & Bacon.
Cameron, M. L., & Robinson, V.M.J. (1980). Effects of cognitive training on
academic and on-task behavior of hyperactive children. Journal of Abnormal
Child Psychology, 8, 405-419.
Chen, W. J., Faraone, S. V., Biederman, J., & Tsuang, M. T. (1994). Diagnostic
accuracy of the Child Behavior Checklist scales for attention-deficit
hyperactivty disorder: A reciever-operating characteristic analysis. Journal of
24
Doctorate in Educational and Child Psychology
Cara Southworth
Consulting and Clinical Psychology, 62(5), 1017-1025.
Fehlings, D. L., Roberts, W., Humphries, T., & Dawe, G. (1991). Attention deficit
hyperactivity disorder: Does cognitive behavioral therapy improve home
behavior? Developmental and Behavioral Pediatrics, 12(4), 223 – 228.
Flanagan, R., & Miller, J. A. (2010). Specialty competencies in school psychology.
New York: Oxford University Press.
Froelich, J., Doepfner, M., & Lehmkuhl, G. (2002). Effects of combined cognitive
behavioural treatment with parent management training in ADHD. Behavioural
and Cognitive Psychotherapy, 30, 111 – 115.
Gaub, M., & Carlson, C. L. (1997). Gender differences in ADHD: A meta-analysis
and critical review. Journal of the American Academy of Child & Adolescent
Psychiatry, 36(8), 1036 -1045.
Gittelman, R., & Kanner, A. (1986). Psychopharmacology. In H. Quay & J. Werry
(Eds.), Psychopathological disorders of childhood (pp. 455-495). New
York: Guildford.
Gough, D. (2007). Weight of evidence: A framework for the appraisal of the quality
and relevance of evidence. In J. Furlong & A. Oancea (Eds.), Research
Papers in Education, 22(2), 213-228.
25
Doctorate in Educational and Child Psychology
Cara Southworth
Gray, K. M., Riggs, P. D., Min, S.J., Mikulich-Gilbertson, S. K., Bandyopadhyay, D.,
& Winhusen, T. (2011). Cigarette and cannabis use trajectories among
adolescents in treatment for attention-deficit/hyperactivity disorder and
substance use disorders. Drug and Alcohol Dependence, 117(2-3), 242–247.
Green, H., McGinnity, A., Meltzer, H., Ford, T., & Goodman, R. (2005). Mental
Health of Children and Young People in Great Britain, 2004. A survey carried
out by the Office for National Statistics on behalf of the Department of Health
and the Scottish Executive. Basingstoke: Palgrave Macmillan.
Hechtman, L., Weiss, G., & Perlman, T. (1984). Young adult outcome of hyperactive
children who received long-term stimulant treatment. Journal of the American
Academy of Child Psychiatry, 23, 261- 269.
Hinshaw, S. P., & Melnick, S. (1992). Self-management therapies and attentiondeficit hyperactivity disorder: Reinforced self-evaluation and anger control
interventions. Behavior Modification, 16(2), 253-273.
Horner, R. H., Carr, E. G., Halle, J., Mcgee, G., Odom, S., & Wolery, M. (2005). The
use of single-subject research to identify evidence-based practice in special
education. Exceptional Children, 71, 165-179.
Horn, W. F., Ialongo, N., Greenberg, G., Packard, T., & Smith- Winberry, C. (1990).
26
Doctorate in Educational and Child Psychology
Cara Southworth
Additive effects of behavioral parent training and self-control therapy with
attention deficit hyperactivity disordered children. Journal of Clinical Child
Psychology, 19(2), 98 – 110.
Jensen, P.S., Mrazek, D., Knapp, P. K., Steinberg, L., Pfeffer, C., Schowalter, J., &
Shapiro, T. (1997). Evolution and revolution in child psychiatry: ADHD as a
disorder of adaptation. Journal of the American Academy of Child &
Adolescent Psychiatry, 36(12), 1672-1681.
Kendall, P.C., & Braswell, L. (1982). Cognitive-behavioral self-control therapy for
children: A comprehensive analysis. Journal of Consulting and Clinical
Psychology, 50, 672-689.
Kendall, P.C., & Braswell, L. (1985). Cognitive-behavioral therapy for impulsive
children. New York: Guildford.
Kendall, P.C., Padever, W., & Zupan, B. (1980). Developing self-control in children:
A manual of cognitive-behavioral strategies. Minneapolis: University of
Minnesota.
Kendall, P. C., & Wilcox, L. E. (1979). Self-control in children: Development of a
rating scale. Journal of Consulting and Clinical Psychology, 47(6), 1020-1029.
Kratochwill, T. R., & Steele Shernoff, E. (2004). Evidence-Based Practice: Promoting
27
Doctorate in Educational and Child Psychology
Cara Southworth
Evidence-Based Interventions in School Psychology. School Psychology
Review, 33, 34-48.
Langberg, J. M., Froehlich, T. E., Loren, R. E., Martin, J. E., & Epstein, J. N. (2008).
Assessing children with ADHD in primary care settings. Expert Review of
Neurotherapeutics, 8, 627 – 641.
Levine, E. S., & Anshel, D. J. (2011). “Nothing Works!” A case study using cognitivebehavioral interventions to engage parents, educators, and children in the
management of attention-deficit/hyperactivity disorder. Psychology in the
Schools, 48(3), 297 – 306.
Miranda, A., & Jesús Presentación, M. (2000). Efficacy of cognitive-behavioral
therapy in the treatment of children with ADHD, with and without
aggressiveness. Psychology in the Schools, 37(2), 169 – 182.
Muñoz-Solomando, A., Kendall, T., & Whittington, C. J. (2008). Cognitive
behavioural therapy for children and adolescents. Current Opinion in
Psychiatry, 21(4), 332–7.
NICE (2009). Attention deficit hyperactivity disorder: Diagnosis and management of
ADHD in children young people and adults. London, National Institute for
Health and Clinical Excellence.
28
Doctorate in Educational and Child Psychology
Cara Southworth
Rutter, M. (1976). A children’s behavior questionnaire for completion by teacher.
Journal of Child Psychology and Psychiatry, 8, 1-11.
Springer, J. F., Phillips, J. L., Phillips, L., Cannady, L. P., & Kerst-Harris, E. (1992).
CODA: A creative therapy program for children in families affected by abuse
of alcohol or drugs. Journal of Community Psychology, 75, 55-74.
Tamm, L., Trello-Rishel, K., Riggs, P., Nakonezny, P. A., Acosta, M., Bailey, G., &
Winhusen, T. (2013). Predictors of treatment response in adolescents with
comorbid substance use disorder and attention-deficit/hyperactivity disorder.
Journal of Substance Abuse Treatment, 44(2), 224–30.
Usher, E. L., & Pajares, F. (2008). Sources of self-efficacy in school: Critical review
of the literature and future directions. Review of Educational Research,
78(4), 751–796.
29
Doctorate in Educational and Child Psychology
Cara Southworth
Appendices
Appendix A: Rationale for Inclusion and Exclusion Criteria
1. Inclusion of papers that maintain fidelity to a cognitive-behavioural approach
and that also target attention deficit hyperactivity disorder ensures that the
review question is answered accurately. Ability to isolate cognitivebehavioural
treatment
impact
on
dependent
measures
from
other
interventions that do not target ADHD and/or are not delivered to child or
young person helps to control for extraneous variables. Due to the nature of
ADHD treatment it was not always possible to isolate cognitive-behavioural
treatment from other interventions that target ADHD that are delivered to the
child or young person, such as ADHD medication. This was however taken
into account and affected the weighting of evidence accordingly.
2. The study must measure ADHD symptoms from the parent perspective as this
is the focus of the review.
3. Participants must not be over 19 as when this review was carried out the SEN
Code of Practice stated that a child with special educational needs or
disabilities could be supported for their whole school career, therefore up until
the age of 19.
4. A diagnosis of Attention Deficit Hyperactivity Disorder (ADHD), diagnosed by
DSM-III-R – DSM 5 or European equivalent (e.g. ICD-10) was required as it
was in the DSM-III-R that Attention Deficit Hyperactivity Disorder was refined
and the ADHD label introduced. The ADHD label has only remained stable
since this point. This ensures that the participants included within the review
have undergone the same diagnosis process and have been measured
against the same criteria.
30
Doctorate in Educational and Child Psychology
Cara Southworth
5. Participants on medication for ADHD were accepted within this review as it is
a very common occurrence that children with ADHD are treated with some
form of medication (e.g. methylphenidate based medication). Further, the Nice
Guidelines states that although medication should not be considered as the
first-line of treatment it should be considered for those with more severe
symptoms (NICE 2009). However, participants on medication for other
disorders and participants diagnosed with other distinct disorders were
excluded to help control for extraneous variables.
6. The study must have primary data in order to analyse effect sizes of the
outcomes and to ensure original synthesise of study findings. This review is
investigating impact and so descriptive guides will not be relevant.
7. The publication must be a peer reviewed journal as it should be assessed
against a set of criteria measuring its quality. There is less chance of fictitious
results or large reporting errors when the study has been reviewed.
8. The study must be available in English as the reviewer cannot read other
languages adequately.
9. The study must be published before January 2014 as this was the date of the
database search and must be published after 1987 because this was the year
the DSM-III-R was introduced. See Criteria 5 for further details about the
DSM-III-R.
31
Doctorate in Educational and Child Psychology
Cara Southworth
Appendix B: Flow Diagram of full literature search
Papers identified from search terms (PsychInfo,
ERIC, and Medline)
(n – 1,054)
Excluded based on:
Title and/or Abstract
(n – 1,044)
Full articles assessed
(n - 10)
Excluded based on:
Inclusion Criteria (Criteria 1, 2 and 6)
(n - 5)
Five studies remained from database search
(n - 5)
Ancestral searches conducted on 5 remaining
studies from database search
(n – 193)
Excluded based on:
Title
(n - 189)
Full articles assessed (from ancestral search)
(n - 4)
All Excluded based on:
Inclusion Criteria (Criteria: 1, 6, 7, and 8)
(n - 4)
Five studies obtained from database search
included in synthesis
(n - 5)
32
Doctorate in Educational and Child Psychology
Cara Southworth
Appendix C: Studies excluded from systematic database search
Reference
Muñoz-Solomando, A., Kendall, T., & Whittington, C. J.
(2008). Cognitive behavioural therapy for children and
adolescents. Current Opinion in Psychiatry, 21(4),
332–7.
Reason for excluding
Criteria 6: Is not an empirical
study containing primary data.
Bloomquist, M. L., August, G. J., & Ostrander, R. (1991).
Effects of a school-based cognitive-behavioral
intervention for ADHD children. Journal of Abnormal
Child Psychology, 19(5), 591–605.
Criteria 2: Does not contain a
variable measuring ADHD
symptoms from the parent
perspective
Hinshaw, S. P., & Melnick, S. (1992). Self-management
therapies and attention-deficit hyperactivity disorder:
reinforced self-evaluation and anger control
interventions. Behavior Modification, 16(2), 253–273.
Criteria 6: Is not an empirical
study containing primary data.
Mirnasab, M. M. (2011). Case report effects of selfmonitoring technique on inattentive behaviors of
students with attention deficit hyperactivity disorder, 5,
84–86.
Criteria 2: Does not contain a
variable measuring ADHD
symptoms from the parent
perspective
Levine, E. S., & Anshel, D. J. (2011). “Nothing Works!” A
case study using cognitive-behavioral interventions to
engage parents, educators, and children in the
management of attention-deficit/hyperactivity disorder.
Psychology in the Schools, 48(3), 297 – 306.
Criteria 1: Unable to isolate CB
treatment effect from other
interventions not delivered to
child.
Appendix D: Studies excluded from ancestral search
Reference
Abikoff, H. (1991). Cognitive training in ADHD children:
Less to it than meets the eye. Journal Learning Disabilities,
24, 205-209.
Blinn, E, L. (2000). Efficacy of play therapy on problem
behaviour behaviors of a child with attention deficit
hyperactivity disorders. Retrieved for the World Wide Web.
http://www.Altavista.com
Reason for excluding
Criteria 6: Is not an empirical
study containing primary data
Criteria 7: Unpublished work or
grey literature such as personal
websites or non-peer reviewed
journals
Hansen, S., Miessler, K., Ovens, R. (2000). Kids together: A Criteria 1: Not cognitivebehavioural treatment
group play therapy model for children with ADHD
symptomology. Journal of Child and Adolescent Group
Therapy, 10, 191-21.1
Janatian, S., Nouri, A., Shafti, A., Molvai H., & Samavatyan,
H. (2008). Effectiveness of play therapy on the bases of
cognition behavioural approach of symptoms of attention
deficit/hyperactivity disorder (ADHD) among primary school
male students aged 9-11. Journal of Research in
Behavioural Sciences, 6 (2), 110-112.
33
Criteria 8: Full article not
available in English
Doctorate in Educational and Child Psychology
Cara Southworth
Appendix E: Summary of Studies
Study and
Aims
Sample
Design
Intervention
Relevant Measures
Results in relation to
relevant
measures/sample
Abdollahian et
al. (2013)
N = 30
Quasi-experimental
study.
Setting: Mashhad, Iran
Parents asked to fill in
Rutter Parental
Questionnaire.
(only parent measures reported here)
Aimed to study
the efficacy of
cognitivebehavioural
play therapy on
the symptoms
of ADHD in
children aged 79 years.
Fehlings et al.
(1991).
To determine
CBT’s
effectiveness in
improving home
behavior in
children with
ADHD.
Age: 7-9 (66.7% of
EG and 73.3% of
CG being under age
of 8)
Identification:
Consecutive
referrals to the child
and adolescent
psychiatric clinic of
the Ebne-Sina
university hospital
during a specified 4month period + then
a diagnosis of
ADHD made by two
board certified child
and adolescent
psychiatrists. All
participants were on
medication
N = 26 (all m) (1
excluded from data
analyses)
Age: 7-13
Identification:
Diagnosis of ADHD
(DSM-III-R), a rating
of 15+ on parent
Conners 10 item
scale, 150+ on self-
15 children assigned to
EG and 15 to CG.
Matched in confounding
variables (birth order,
parental occupations
&educ. level, school
marks)
Parents and teachers
blind to group allocation.
CB treatment: cognitivebehavioural play therapy
sessions, conducted with
different thematic activities: don’t
drop it; throw the balls; touch it;
be careful and throw the rings;
note and remember; me and my
shadow; work for a longer time;
tell me act..tell me; running
inside a maze and talking.
Delivery: Therapist
Pre-tests and post-tests
obtained for both groups.
EG: Eight 45-min
sessions of play therapy
on a twice-weekly basis
(in gps of 5)
CG: Eight shame play
therapy sessions (carried
out in same room for
same time & therapeutic
conditions matched).
Target behaviours: aimed to
improve sustained attention,
working memory, maintenance,
impulsivity, self-control and
organisation in children with
ADHD.
Randomised controlled
trial.
Setting: Toronto, Ontario,
Canada
13 children assigned to
CBT EG and 13 to CG
(supportive therapy).
CB treatment: received direct
instruction on CB strategies,
reinforced by token contingency
reward system. Problem solving
broken down to five step
process (1) defining the problem
(2) setting a goal (3) generating
workable problem-solving
strategies (4) choosing a
Pre-tests and post-tests
and 5month post-tests
obtained for both groups.
EG: 12 60-min individual
34
The Rutter Parental
Questionnaire: comprises
of 18 items, including some
related to the health and
habits of children. Rutter
reported internal validity
and reliability of .74 for the
questionnaire. The
correlation between the
responses of the fathers
and mothers was .64, and
the internal consistency of
the test was .79 (Rutter
1976).
Parent measures: Self
Control Rating Scale
(SCRS) (a measure of
impulsivity); Revised
Behavior Problem
Checklist-Attention Problem
Subscale (BPC-AP
subscale) (a measure of
inattention) and the
Modified Werry Weiss
Activity Scale (a measure of
the child’s activity level in
The results of the paired t test for the
scores of both the pre- and post-test
questionnaire for parents showed that:
there was a significant different in the
experimental group (p=.0007, effect size =
0.8); hyperactivity scores decreased after
play therapy treatment, indicating that this
stimulus had a reducing effect. No
significant difference pre- and post- was
found in the control group (p=.116).
A t test also showed that the ADHD scores
for the experimental group were
significantly lower than those for the control
group (p=.000).
One-two way ANOVA with post hoc t tests
was computed for each dependent
variable. Werry Weiss Activity Scale
revealed a statistically significant treatment
effect favouring CBT (p<.03). A post hoc t
test revealed that there was a significant
difference for the pre- and post- treatment
comparison (p<.020 and the pre and
5month post comparison (p<.01).
Paired t test assessing differences in CG
did not reach significance for pre and post
or pre and 5month.
Doctorate in Educational and Child Psychology
Study and
Aims
Cara Southworth
Sample
Design
Intervention
Relevant Measures
control rating scale,
score of 85+ on
WISC-R. Excluded
if on stimulant
medication or if met
criteria for conduct
disorder or other
major psychiatric
disorders.
sessions, between child
and behavioral therapist,
twice weekly, and 8 2hour sessions once every
2 weeks with the family in
their home.
CG: received same
amount of exposure to
therapist and tasks.
solution (5) evaluating the
outcome with self-reinforcement.
Child taught these using
modelling, role playing, selfinstructional training, cue cards,
homework assignments and
behavioural techniques.
Problems included academic
and then interpersonal ones,
with focus on home behavior.
Parents taught about CBT and
how to encourage it.
the home).
Results in relation to
relevant
measures/sample
No sig. differences for SCRS and BPC-AP
EG.
Delivery: Therapist
Target Behaviours: home
behaviour
Froelich et al.
(2002)
To demonstrate
the
effectiveness of
cognitive
behavioural
treatment (CBT)
with a focus on
academic skills
and conduct
problems (and
to increase
parents’
educational
skills in
managing
aggressive and
oppositional
behaviour in a
subsequent
parent training
N= 18 (17m, 1f) 16
cognitive
functioning normal,
2 with LD. 8 on
ADHD medication.
Age: 6-12
Identification:
Referred from the
Child and
Adolescent
Psychiatric
Outpatient unit of
the University Clinic
Cologne. Diagnosis
corresponded to
Hyperkinetic
Conduct disorder
(ADHD Europe)
(ICD-10 F 90.1).
Pre-, post-test study
design without a control
group.
4 weeks baseline, 6
weeks CBT (+GEN) (6
weeks PMT – not part of
this review)
Treatment effects
measured at end of each
week via rating scales.
CB treatment
incorporated six one-hour
lessons for each pupil.
One per week.
Setting: Cologne, Germany
CB treatment: Predominantly
included training in selfinstructional, self-monitoring and
problem-solving skills. Tasks
included attractive games e.g.
puzzles, picture sequences.
Transfer to self-instructional
problem-solving strategies on
homework or school problems to
behavioural problems of
inattention and impulsivity was
initiated. Social-conflict
situations occurring at home or
school were introduced.
Individually relevant academic or
social problems were discussed
during the sessions and
homework assignments given to
facilitate generalisation. Children
received a phone call a week to
support treatment
35
Parents rated core
symptoms of ADHD on an
abbreviated German
version (15 items) of the
Yale Children’s inventory.
Auth. calculated a total
internal consistency for the
total score of .94.
Cronbach’s Alpha for each
symptom subscale ranged
between .80-.88.
Conflict situations at home
were measured in the
Home Situations
Questionnaire, adapted
German version (16 items,
9-point scale). Internal
consistency calculated
alpha > 90). Homework
problems assessed via
modified Homework
Problem Checklist (Alpha
>.90) and Individual
Multivariate analyses of variance
(MANOVA) with one within subject factor
were performed. No statistical differences
at baseline. During CBT (+ GEN) phase
ADHD core symptoms and symptoms
related to conduct disorder significantly
decreased according to parent ratings
(p<.001 core symptoms; p<.0001 conduct
problems).
Plus, the following were also significant
different to baseline after CB treatment:
individually relevant behaviour problems
(p<.004), conflict situations at home
(p<.001) and homework problems
(p<.001).
Doctorate in Educational and Child Psychology
Study and
Aims
Sample
Design
(PMT).
Cara Southworth
Intervention
Relevant Measures
generalization. Parents and
teachers received regular written
hand-outs of treatment content
and continuously asked to
prompt and to reinforce
appropriate behaviour of the
child at home or school.
Problem List to measure
effectiveness of treatment
on individually relevant
problem behaviour.
Results in relation to
relevant
measures/sample
Delivery: Therapist
Target Behaviours: core
symptoms of ADHD and conduct
problems.
Horn et al.
(1990)
To assess the
separate and
combined
effects of
behavioural
parent training
and cognitivebehavioural
self-control
therapy with 42
elementary
school age
children
diagnosed as
having ADHD.
N= 42 (34m, 8f) +
18 non-ADHD for
CG (8m, 10f).
Age: 7 years, 0
months -11 years, 6
months.
Identification:
Referred to a
university-based
psychological clinic
for treatment of
chronic
inattentiveness,
impulsivity, and
over activity.
Diagnosis of ADHD
from DSM-III-R.
Additional inclusion
criteria: not
receiving
medication for
ADHD, absence of
psychoses and
intellectual
difficulties. The
Randomised controlled
trial.
Participants randomly
assigned to PT, CB SC,
or PT + SC, and the nonADHD control group. 14
families in each group.
CB SC and PT (and nonADHD CG) compared in
study and this review.
Dependent measures
were administered at
three points: pre-test,
post-test, and 8-month
follow-up.
CB SC EG: treatment
(therapy group split into
two groups of 7 children)
groups met for 12
weekly- 90-min sessions.
Non-ADHD CG: no
details.
PT group: Parents
received training which
Setting: US
CB Treatment: CB self-control
therapy consisted of instruction
in the self-control strategies
adapted from Camp and Bash
(1981), Kendall and Braswell
(1985) and Meichenbaum
(1977). Each child was taught a
“Problem Solving Plan” that
incorporated following steps: Am
I having a problem? Take a
deep breath and think “calm and
relax”; What is my problem?;
How many solutions can I think
of?; How good is each solution?;
Pick the best solution and try it:
How did my solution work?.
Deep muscle relaxation, in
conjunction with imaginal
rehearsal, was incorporated into
the problem-solving training.
Training was comprised of
didactic presentations,
modelling, guided practice and
role plays. Role play
incorporated both academic and
36
Parents asked to complete
The Child Behavior
Checklist (CBCL; Total
Problems, Externalising
and Hyperactivity scales).
The Treatment x Period MANOVAs carried
out on parent reports did not yield
significant Treatment x Period interactions.
Significant period effects were found for the
MANOVAs performed on parent reports
(p<.001). Subsequent Univariate repeatedmeasures ANOVAs yielded significant pretest-to-post-test period effects for each of
the parent CBCL report scales.
Pre-test to post-test: CBCL total problems
(p<.001), CBCL Externalizing (p<.001),
CBCL Hyperactivity (p<.001).
Pre-test to follow up: CBCL total problems
(p<.01), CBCL Externalizing (p<.01), CBCL
Hyperactivity (p<.01).
Within-group t tests from pre-test to posttest and from pre-test to follow-up for the
parent outcome variables were then carried
out on treatment conditions. Parent reports
in CB SC group resulted in significant
results in all except CBCL hyperactivity
pre-test to post-test and pre-test to followup.
CBCL total: pre-test to post-test (p<.01);
pre-test to follow-up (p<.01)
CBCL Externalizing: pre-test to post-test
(p<.01); pre-test to follow-up (p<.001).
Doctorate in Educational and Child Psychology
Study and
Aims
Cara Southworth
Sample
Design
Intervention
Relevant Measures
Peabody
Vocabulary test
used to screen for
participant’s
intellectual status.
focused on teaching
parents to apply social
learning principles to the
management of their
child’s behaviour.
interpersonal problems. Token
reinforcement system was used
for behavioural control of the
children in group sessions.
Results in relation to
relevant
measures/sample
However, PT also achieved significant
results across the scales, including
hyperactivity.
CB SC was compared with non-ADHD to
see if ADHD subjects were within the
normal limits at post-test or follow-up. In
parent reports ADHD remained significantly
different from non-ADHD subjects at posttest and follow-up.
Delivery: Therapist
Target behaviours: ADHD
cardinal features (e.g.
externalising and hyperactivity).
CB SC final sessions were videotaped to
obtain a measure of how well each of the
children had learned the six steps of the
Problem-solving plan. Analysis indicated
nearly perfect performance of the six steps.
Miranda and
Jesús
Presentación
(2000)
N= 32 total (16 nonaggressive, 16
aggressive sample
excluded)
To show the
efficacy of
cognitivebehavioral selfcontrol therapy
of children with
ADHD (with and
without
aggressiveness
).
Age: 9 – 12
Identification:
Diagnosed by the
school
psychologists as
having chronic
problems of
inattention,
impulsivity, and
over activity.
Teachers
interviewed to
corroborate. Then
had to meet this
criteria: score 15 +
on Conners rating
Scale, score 16 or
higher on DSM-III-
Pre-, post-test (+follow
up) study design without
a control group.
Participants divided into
two groups 16 with and
16 without
aggressiveness.
Participants spilt into 4
groups: ADHD with CB
SC, ADHD with CB SC +
AM, (Excluded:
ADHD&aggressiveness
with CB SC,
ADHD&aggressiveness
with CB SC + AM.)
Dependent measures
taken at three points: pretreatment, post-treatment
and 2-months follow-up.
CB SC EG: 22 sessions
over 3 months, on a
Setting: Castellón, Spain
CB treatment: Applied Kendall
et al.’s “Stop and Think”
program (1980) with small
modifications. Includes selfinstruction, modelling, and
behavioural contingencies, to
help solve various kinds of
problems. Children trained to
use self-instructional strategies
through the process of problem
solving in order to consider
possible course of action, to
reflect on potential outcomes
and to make decisions about
options. The problem solving
process was broken down into 5
statements: Problem definition,
problem approach, focusing
attention, selecting an answer,
self-evaluation/selfreinforcement. The program
looked at interpersonal problem-
37
Parents were given: A
hyperactivity questionnaire
adapted from the DSM-III-R
and the Scale of
Behavioural Problems
(EPC).
The Hyperactivity
questionnaire: 14 items
corresponding to 14 criteria
described in the Diagnostic
and Statistical Manual of
Mental Disorders (APA,
1987) for the diagnosis of
ADHD. Items reflect the
behaviours observable in
inattention, impulsivity and
over activity.
The EPC: gives information
on the child’s character and
actions in different
environments for the
parent’s point of view. This
MANOVA carried out using information
obtained from the parents of the two
groups with eight non-aggressive ADHD
children showed significant differences at
the three times of testing: DSM-III-R
(p≤.001); Antisocial behaviour (p≤.13);
School problems (p≤.004); and
psychopathological disorders p≤.033).
The scores in all these variables
significantly diminished in the post-test and
maintenance phases, compared to the
scores obtained in the pre-test phase. No
differences found either between the
treatment groups or in the analysis of
between-group and intragroup interaction.
It was not possible to separate CB SC &
CB SC + AM for significant effects over
time.
Doctorate in Educational and Child Psychology
Study and
Aims
Cara Southworth
Sample
Design
Intervention
Relevant Measures
R, 1+ persistence of
symptoms, IQ score
80+ of WISC,
absence of
psychosis or any
gross neurological,
sensory or motor
impairment. Iowa
Scale for Teachers
screened for
aggressiveness.
twice-weekly basis and
lastly approx. one-hour
each, with four groups in
the morning and four in
the afternoon. Sessions
took place outside of
school hours.
solving, looked at hypothetical
problems and later real-life
problems. Programme used
token system, social- and selfreinforcement. Response cost
also used. Generalisation
encouraged.
scale has an alpha
coefficient of .88 and a testretest correlation of .79 in
various subscales. It also
has adequate level of
construct and criterial
validity.
Delivery: Therapist.
Target behaviours: improve
concentration and reflection.
38
Results in relation to
relevant
measures/sample
Doctorate in Educational and Child Psychology
Cara Southworth
Appendix F: Weight of Evidence
Weight of Evidence A. Considers the reliability and validity of measures (A.1), any
control measures used (A.2), and the intervention measures (A.3).

In order to gain ‘High’ weighting for methodological quality the study must
have an average weighting of 3.

In order to gain ‘Medium’ weighting for methodological quality the study must
have an average weighting of 2.

In order to gain ‘Low’ weighting for methodological quality the study must
have an average weighting of 1.
Weight of Evidence B. Methodological Relevance considers whether the research
design was suitable for evaluating the effectiveness of using cognitive-behavioural
treatment to improve ADHD symptoms (B.1).

In order to gain ‘High’ weighting for methodological relevance the study must
be randomized with a control group.

In order to gain ‘Medium’ weighting for methodological relevance the study
may be a non-randomised group design with some control measures.

In order to gain ‘Low’ weighting for methodological relevance the study may
be a non-randomised group design with no control measures.
Weight of Evidence C. Relevance to the review question considers whether the
design was suitable for evaluating the effectiveness of using cognitive-behavioural
treatment to improve ADHD symptoms for children with attention deficit hyperactivity
disorder (C.1).

In order to gain ‘High’ weighting for relevance the study must use cognitivebehavioural treatment to target specific behaviours associated with ADHD.
39
Doctorate in Educational and Child Psychology
Cara Southworth
Must measure the target behaviour before and after intervention (follow-up if
possible). The sample of interest must be children with an appropriate
diagnosis of ADHD, comparison group also with a diagnosis of ADHD.
Sample must be typically developing children with the exception of the ADHD
diagnosis, and must not be receiving additional interventions aimed at ADHD
(e.g. medication/anger management). ADHD symptoms must be measured by
the parent.

In order to gain ‘Medium’ weighting for relevance the study must use
cognitive-behavioural treatment to target specific behaviours associated with
ADHD. Must measure the target behaviour before and after intervention
(follow-up if possible). The sample of interest must be children with an
appropriate diagnosis of ADHD. Sample can include children with other
disorders in addition to the diagnosis of ADHD, and can be receiving
medication for ADHD. ADHD symptoms must be measured by the parent.

In order to gain ‘Low’ weighting for relevance the study must use cognitivebehavioural treatment to target specific behaviours associated with ADHD.
Must measure the target behaviour before and after intervention (follow-up if
possible). The sample of interest must be children with an appropriate
diagnosis of ADHD. Sample can include children receiving medication for
ADHD or additional non-medication interventions for ADHD. ADHD symptoms
must be measured by the parent.
Weight of Evidence D. Calculated as an average of the weight of evidence A, B and
C.
40
Doctorate in Educational and Child Psychology
Cara Southworth
Effect Sizes. The effect sizes are listed at the end of the coding protocols. Effect
sizes have either been quoted directly from the studies or they have been calculated.
If calculated, the effect size was calculated using means, standard deviations and ns
with either the following formula –
(1) Cohen’s d from means and standard deviations:
(2) Hedge’s g from means, standard deviations and ns:
Due to the small samples in the studies included in this review, Hedge’s g was used
where possible as it incorporates an adjustment which removes the bias of Cohen’s
d.
The effect size rating (small, medium or large) is listed in Table 7, the rationale for
these ratings can be found in Cohen (1988).
Table 7
Effect size interpretation
Type of effect size
Small
Cohen’s d/Hedge’s g
0.20
Medium
0.50
41
Large
0.80
Doctorate in Educational and Child Psychology
Cara Southworth
Appendix G: Coding Protocol
A coding protocol adapted from Kratochwill (2003) and Gough (2007) was used to
evaluate the studies in this review.
Full Reference: Abdollahian, E. Mokhber, N., Balaghi, A., & Moharrari, F. (2013).
The effectiveness of cognitive-behavioural play therapy on the symptoms of
attention-deficit/hyperactivity disorder in children aged 7-9 years. ADHD Attention
Deficit and Hyperactivity Disorders, 5(1), 41 – 46.
Weight of Evidence A
A.1. Quality of Measures
Check
Reliability and Validity
Multi-method
Multi-source (as only interested in parent)
n/a
(Relevant) Outcomes 1: The Rutter Parental Questionnaire
Reliability - Rutter reported internal validity and reliability of .74 for the questionnaire.
The correlation between the responses of the fathers and mothers was .64, and the
internal consistency of the test was .79 (Rutter 1976)
Validity - above
Outcome 2:
Reliability =
Validity =
Overall Rating for measurement
(3) Strong evidence: case for validity, reliability above 0.85 (in majority or primary
outcomes), multiple methods and multiple sources
(2) Promising evidence: reliability above 0.70, multiple methods or multiple sources
(1) Weak evidence: reliability above 0.50, multiple methods or multiple sources –

Judgement: Reliability reported is high, but limitations elsewhere.
(0) None of the above
A.2 Quality of Control Measure
Check
Control Measure
Type of comparison group
Typical intervention
Attention placebo
Intervention element placebo (intervention minus active ingredient linked to change) –

Sham CB play therapy sessions
Alternative intervention
Pharmacotherapy
No intervention
Wait list/delayed intervention
42
Doctorate in Educational and Child Psychology
Cara Southworth
Minimal contact
Unable to identify type of comparison
Overall judgement of type of comparison group
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)

Unable to identify comparison group
Counterbalancing of change agent (e.g. teacher or therapist)
By change agent
Statistical
Other
Not reported/none – not reported, but control group attended same room where play

therapy took place for same amount of time, who administered sessions is
unknown.
Group equivalence established
Random assignment
Posthoc matched set

Statistical matching
Post hoc test for group equivalence
Equivalent mortality
Low attrition (less than 20% for post)

Low attrition (less than 30% for follow up)
Intent to intervene analysis carried out (include findings):
Overall Rating for Control Measure
(3) Strong evidence: at least one active comparison group, group equivalence must be
established (preferably random assignment), counterbalancing, low attrition at post and
follow up (if applicable)
(2) Promising evidence: at least one ‘no intervention’ comparison group. Plus at least two

of: group equivalence must be established (preferably random assignment), or
counterbalancing or low attrition at post and follow up (if applicable)
(1) Weak evidence: at least one comparison group. Plus at least one of: group equivalence
must be established (preferably random assignment), or counterbalancing or low attrition
at post and follow up (if applicable)
(0) no efforts made to control
A.3 Quality of Intervention
Group Experimental Design
Primary outcomes in appropriate unit of analysis
Primary outcomes familywise/experimenter wise error rate controlled when applicable
Primary outcomes have sufficiently large N
At least 75% of primary outcomes significant

50-74% of primary outcomes significant
25-49% of primary outcomes significant
Less than 25% of primary outcomes significant
Overall Rating for Intervention
(3) Strong evidence: Significant outcomes for at least 75%. Plus appropriateness of
analysis, familywise/experimenter wise error controlled and sufficiently large N.
(2) Promising evidence: Significant outcomes for 50-74%. Plus appropriateness of

analysis, familywise/experimenter wise error controlled and sufficiently large N. –
Judgement: over 50-74% but not sufficiently large N
(1) Weak evidence: Significant outcomes for 24-49%. Plus appropriateness of analysis,
familywise/experimenter wise error controlled and sufficiently large N.
(0) No/Limited evidence: Less than 25% significant outcomes and none of the above
Overall Rating for WoE A (1+2+2) /3 = 1.66

43
Doctorate in Educational and Child Psychology

Cara Southworth
(3) High
(2) Medium
(1) Low
Weight of Evidence B
B.1 Design
Check
Single Case Design
Check Group Experimental Design
Type of Design:
Within –Series: Simple phase change
Non-randomised design
Within – Series: Complex phase
Non-randomised block design (between

change
participants)
Between – Series: Comparing two
Non-randomised block design (within
interventions
participants)
Between – Series: Comparing
Non-randomised hierarchical design
interventions with no interventions
Combined – Series: Multiple baseline
Optional coding for quasi-experimental
across participants
design
Combined – Series: Multiple baseline
across behaviours
Combined – Series: Multiple baseline
across settings
Combined – Series: Multiple probe
Mixed single participant design (tick
two of the above)
Mixed single participant and group
design (tick one of the above and one
in the right hand column)
Other (specify):
If randomisation used:
Completely randomised
Randomised block design (between participants e.g. matched classrooms)
Randomised block design (within participants)
Randomised hierarchical design (nested treatments)
Units of randomisation:
Individual
Classroom
School
Other (specify):
Overall confidence of judgement on how participants were assigned:
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)
N/A
Unknown/unable to code
Overall Rating for WoE B
(3) High
(2) Medium

(1) Low
44
Doctorate in Educational and Child Psychology
Cara Southworth
Weight of Evidence C
C.1 Sample
N
Consecutive referrals to the child and adolescent psychiatric clinic of the Ebne-Sina
university hospital during a specified 4-month period + then a diagnosis of ADHD made by
two board certified child and adolescent psychiatrists
30
Sample size at start of study
Current exposure to alternative intervention
Prior exposure to alternative intervention
No historical exposure to alternative intervention
Single Case Design Participant
N
additional information:
15
15
P1
P2
P3
Overall Rating for WoE C
(3) High
(2) Medium

(1) Low
Group Experimental Design
Intervention Group size
Control Group size
Attrition (drop out)
Weight of Evidence D
Overall Rating for WoE D (2+2+2) /3 = 2
(3) High
(2) Medium

(1) Low
Effect sizes
Relevant
Outcomes
Outcome 1.
Parent
questionnaire
(Rutter) to assess
ADHD symptoms
in intervention
group (pre-and
post-test).
Type of effect size
Effect size
Comment
Cohen’s d
0.80
Explicitly stated.
Unable to convert to Hedge’s g.
Outcome 2.
Parent
questionnaire
(Rutter)
intervention vs.
control group posttest.
Unable to calculate effect size
45
Doctorate in Educational and Child Psychology
Cara Southworth
Full Reference: Fehlings, D. L., Roberts, W., Humphries, T., & Dawe, G. (1991).
Attention deficit hyperactivity disorder: Does cognitive behavioral therapy improve
home behavior? Developmental and Behavioral Pediatrics, 12(4), 223 – 228.
Weight of Evidence A
A.1 Quality of Measures
Check
Reliability and Validity
Multi-method
Multi-source (parents only)
(Relevant) Outcomes 1: Self Control Rating Scale (SCRS)
Reliability Validity (Relevant) Outcome 2: Revised Behaviour Problem Checklist-Attention Problem Subscale
(BPC-AP)
Reliability =
Validity =
(Relevant) Outcome 3: Modified Werry Weiss Activity Scale
Reliability =
Validity =
Overall Rating for measurement
(3) Strong evidence: case for validity, reliability above 0.85 (in majority or primary
outcomes), multiple methods and multiple sources
(2) Promising evidence: reliability above 0.70, multiple methods or multiple sources
(1) Weak evidence: reliability above 0.50, multiple methods or multiple sources

Judgement: References show reliability of methods used but no statement of this in
study or with the study’s population. Multiple methods.
(0) None of the above

n/a
A.2 Quality of Control Measure
Check
Control Measure
Type of comparison group
Typical intervention
Attention placebo
Intervention element placebo (intervention minus active ingredient linked to change)
Alternative intervention – Supportive Therapy

Pharmacotherapy
No intervention
Wait list/delayed intervention
Minimal contact
Unable to identify type of comparison
Overall judgement of type of comparison group
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)

Unable to identify comparison group
Counterbalancing of change agent (e.g. teacher or therapist)
46
Doctorate in Educational and Child Psychology
Cara Southworth

By change agent – Both groups received equal amount of exposure to same therapist
and tasks.
Statistical
Other
Not reported/none
Group equivalence established
Random assignment

Posthoc matched set
Statistical matching
Post hoc test for group equivalence

Equivalent mortality
Low attrition (less than 20% for post)

Low attrition (less than 30% for follow up)

Intent to intervene analysis carried out (include findings):
Overall Rating for Control Measure
(3) Strong evidence: at least one active comparison group, group equivalence must be

established (preferably random assignment), counterbalancing, low attrition at post and
follow up (if applicable)
(2) Promising evidence: at least one ‘no intervention’ comparison group. Plus at least two
of: group equivalence must be established (preferably random assignment), or
counterbalancing or low attrition at post and follow up (if applicable)
(1) Weak evidence: at least one comparison group. Plus at least one of: group equivalence
must be established (preferably random assignment), or counterbalancing or low attrition
at post and follow up (if applicable)
(0) no efforts made to control
A.3 Quality of Intervention
Group Experimental Design
Primary outcomes in appropriate unit of analysis
Primary outcomes familywise/experimenter wise error rate controlled when applicable
Primary outcomes have sufficiently large N
At least 75% of primary outcomes significant
50-74% of primary outcomes significant
25-49% of primary outcomes significant

Less than 25% of primary outcomes significant
Overall Rating for Intervention
(3) Strong evidence: Significant outcomes for at least 75%. Plus appropriateness of
analysis, familywise/experimenter wise error controlled and sufficiently large N.
(2) Promising evidence: Significant outcomes for 50-74%. Plus appropriateness of
analysis, familywise/experimenter wise error controlled and sufficiently large N.
(1) Weak evidence: Significant outcomes for 24-49%. Plus appropriateness of analysis,

familywise/experimenter wise error controlled and sufficiently large N. – Judgement: no
familywise control, but appropriate unit of analysis and 33.3% primary outcome
significance.
(0) No/Limited evidence: Less than 25% significant outcomes and none of the above

Overall Rating for WoE A (1+3+1) /3 = 1.66
(3) High
(2) Medium

(1) Low
47
Doctorate in Educational and Child Psychology
Cara Southworth
Weight of Evidence B
B.1 Design
Check
Single Case Design
Check Group Experimental Design
Type of Design:
Within –Series: Simple phase change
Non-randomised design
Within – Series: Complex phase
Non-randomised block design (between
change
participants)
Between – Series: Comparing two
Non-randomised block design (within
interventions
participants)
Between – Series: Comparing
Non-randomised hierarchical design
interventions with no interventions
Combined – Series: Multiple baseline
Optional coding for quasi-experimental
across participants
design
Combined – Series: Multiple baseline
across behaviours
Combined – Series: Multiple baseline
across settings
Combined – Series: Multiple probe
Mixed single participant design (tick
two of the above)
Mixed single participant and group
design (tick one of the above and one
in the right hand column)
Other (specify):
If randomisation used:
Completely randomised
Randomised block design (between participants e.g. matched classrooms)

Randomised block design (within participants)
Randomised hierarchical design (nested treatments)
Units of randomisation:
Individual

Classroom
School
Other (specify):
Overall confidence of judgement on how participants were assigned:
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)

Very high (explicitly stated)
N/A
Unknown/unable to code
Overall Rating for WoE B
(3) High

(2) Medium
(1) Low
48
Doctorate in Educational and Child Psychology
Cara Southworth
Weight of Evidence C
C.1 Sample
N
Diagnosis of ADHD (DSM-III-R), a rating of 15+ on parent Conners 10 item scale, 150+ on selfcontrol rating scale, score of 85+ on WISC-R. Excluded if on stimulant medication or if met
criteria for conduct disorder or other major psychiatric disorders.
26
Sample size at start of study
Current exposure to alternative intervention
Prior exposure to alternative intervention
No historical exposure to alternative intervention
Single Case Design Participant
N
additional information:
13
13
2
P1
P2
P3
Group Experimental Design
Intervention Group size
Control Group size
Attrition (drop out) – 1 from supportive
therapy group, 1 family from
supportive therapy group at 5-month
follow-up.
Overall Rating for WoE C
(3) High

(2) Medium
(1) Low
Weight of Evidence D
Overall Rating for WoE D (2+3+3)/3 = 2.66
(3) High

(2) Medium
(1) Low
Effect Sizes
Relevant
Outcomes
Type of effect size
Effect size
Cohen’s d(Adjusted to Hedge’s g)
0.77
Outcome 1b.
SCRS CBT pre-tofollow-up
Cohen’s d(Adjusted to Hedge’s g)
1.47
Outcome 1c.
SCRS CBT vs.
Control post-test
Cohen’s d(Adjusted to Hedge’s g)
-0.27
Outcome 1d.
SCRS CBT vs.
Control follow-up
Cohen’s d(Adjusted to Hedge’s g)
-0.28
Outcome 2a.
WWAS CBT pre-topost-test
Cohen’s d (Adjusted to Hedge’s g)
0.98
Outcome 1a.
SCRS
CBT pre-to-posttest
49
Doctorate in Educational and Child Psychology
Relevant
Outcomes
Cara Southworth
Type of effect size
Effect size
Outcome 2b.
WWAS CBT Pre-tofollow-up
Cohen’s d (Adjusted to Hedge’s g)
1.10
Outcome 2c.
WWAS CBT vs.
Control post-test
Cohen’s d(Adjusted to Hedge’s g)
0.72
Outcome 2d.
WWAS CBT vs.
Control follow-up
Cohen’s d(Adjusted to Hedge’s g)
-0.49
Outcome 3a. BPCAP CBT pre-topost-test
Cohen’s d(Adjusted to Hedge’s g)
0.65
Outcome 3b. BPCAP CBT pre-tofollow-up
Cohen’s d(Adjusted to Hedge’s g)
1.00
Outcome 3c. BPCAP CBT vs. Control
post-test.
Cohen’s d(Adjusted to Hedge’s g)
-0.48
Cohen’s d(Adjusted to Hedge’s g)
-0.27
Outcome 3d. BPCAP CBT vs. Control
follow-up
Full Reference: Froelich, J., Doepfner, M., & Lehmkuhl, G. (2002). Effects of
combined cognitive behavioural treatment with parent management training in
ADHD. Behavioural and Cognitive Psychotherapy, 30, 111 – 115.
Weight of Evidence A
A.1 Quality of Measures
Check
Reliability and Validity
Multi-method – Parent questionnaires: Adapted Yale Children’s Inventory, Home

Situations Questionnaire, Homework Problem Checklist & Individual Problem
Checklist
Multi-source
n/a
Outcomes 1: Adapted Yale Children’s Inventory
Reliability - Auth. calculated a total internal consistency for the total score of .94. Cronbach’s

Alpha for each symptom subscale ranged between .80-.88.
Validity –
Outcome 2: Home Situations Questionnaire
Reliability = Internal consistency calculated alpha > 90

Validity =
Outcome 3: Homework Problem Checklist
Reliability = Alpha >.90
50
Doctorate in Educational and Child Psychology
Cara Southworth
Validity =
Overall Rating for measurement
(3) Strong evidence: case for validity, reliability above 0.85 (in majority or primary
outcomes), multiple methods and multiple sources
(2) Promising evidence: reliability above 0.70, multiple methods or multiple sources

(1) Weak evidence: reliability above 0.50, multiple methods or multiple sources
(0) None of the above
A.2 Quality of Control Measure
Check
Control Measure – no control group
Type of comparison group
Typical intervention
Attention placebo
Intervention element placebo (intervention minus active ingredient linked to change)
Alternative intervention Pharmacotherapy
No intervention
Wait list/delayed intervention
Minimal contact
Unable to identify type of comparison
Overall judgement of type of comparison group
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)
Unable to identify comparison group
Counterbalancing of change agent (e.g. teacher or therapist)
By change agent
Statistical
Other
Not reported/none
Group equivalence established
Random assignment
Posthoc matched set
Statistical matching
Post hoc test for group equivalence
Equivalent mortality - unknown
Low attrition (less than 20% for post)
Low attrition (less than 30% for follow up)
Intent to intervene analysis carried out (include findings):
Overall Rating for Control Measure
(3) Strong evidence: at least one active comparison group, group equivalence must be
established (preferably random assignment), counterbalancing, low attrition at post and
follow up (if applicable)
(2) Promising evidence: at least one ‘no intervention’ comparison group. Plus at least two
of: group equivalence must be established (preferably random assignment), or
counterbalancing or low attrition at post and follow up (if applicable)
(1) Weak evidence: at least one comparison group. Plus at least one of: group equivalence
must be established (preferably random assignment), or counterbalancing or low attrition
at post and follow up (if applicable)
(0) no efforts made to control

51
Doctorate in Educational and Child Psychology
Cara Southworth
A.3 Quality of Intervention
Group Experimental Design –
Primary outcomes in appropriate unit of analysis
Primary outcomes familywise/experimenter wise error rate controlled when applicable
Primary outcomes have sufficiently large N
At least 75% of primary outcomes significant

50-74% of primary outcomes significant
25-49% of primary outcomes significant
Less than 25% of primary outcomes significant
Overall Rating for Intervention
(3) Strong evidence: Significant outcomes for at least 75%. Plus appropriateness of
analysis, familywise/experimenter wise error controlled and sufficiently large N.
(2) Promising evidence: Significant outcomes for 50-74%. Plus appropriateness of

analysis, familywise/experimenter wise error controlled and sufficiently large N. –
Judgement due to lack of sufficient N but good significant outcomes.
(1) Weak evidence: Significant outcomes for 24-49%. Plus appropriateness of analysis,
familywise/experimenter wise error controlled and sufficiently large N.
(0) No/Limited evidence: Less than 25% significant outcomes and none of the above


Overall Rating for WoE A (2+0+2)/3 = 1.33
(3) High
(2) Medium
(1) Low

Weight of Evidence B
B.1 Design
Check
Single Subject Design
Check Group Experimental Design
Type of Design:
Within –Series: Simple phase change
Non-randomised design
Within – Series: Complex phase
Non-randomised block design (between
change
participants)
Between – Series: Comparing two
Non-randomised block design (within
interventions
participants)
Between – Series: Comparing
Non-randomised hierarchical design
interventions with no interventions
Combined – Series: Multiple baseline
Optional coding for quasi-experimental

across participants
design
A pre-test- post-test design – no control
Combined – Series: Multiple baseline
group
across behaviours
Combined – Series: Multiple baseline
across settings
Combined – Series: Multiple probe
Mixed single participant design (tick
two of the above)
Mixed single participant and group
design (tick one of the above and one
in the right hand column)
Other (specify):
If randomisation used:
Completely randomised
Randomised block design (between participants e.g. matched classrooms)
Randomised block design (within participants)
Randomised hierarchical design (nested treatments)
52
Doctorate in Educational and Child Psychology
Cara Southworth
Units of randomisation:
Individual
Classroom
School
Other (specify):
Overall confidence of judgement on how participants were assigned:
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)
N/A
Unknown/unable to code
Overall Rating for WoE B
(3) High
(2) Medium
(1) Low

Weight of Evidence C
C.1 Sample
N
Referred from the Child and Adolescent Psychiatric Outpatient unit of the University Clinic
Cologne. Diagnosis corresponded to Hyperkinetic Conduct disorder (ADHD Europe) (ICD-10 F
90.1). 16 cognitive functioning normal, 2 with LD. 8 on ADHD medication.
18
8
Sample size at start of study
Current exposure to alternative intervention - medication
Prior exposure to alternative intervention
No historical exposure to alternative intervention
Single Case Design Participant
N
Group Experimental Design
additional information:
18
All go through baseline then CBT . 2
LD
Attrition (drop out)
Overall Rating for WoE C
(3) High
(2) Medium

(1) Low
Weight of Evidence D
Overall Rating for WoE D (1+1+2)/3 = 1.33
(3) High
(2) Medium
(1) Low

53
Doctorate in Educational and Child Psychology
Cara Southworth
Effect Sizes
Relevant
Outcomes
Type of effect size
Effect size
Outcome 1. ADHD
core symptoms pre-topost-test
Comment
Unable to compute
effect size
Outcome 2. Conduct
Problems pre-to-posttest
Unable to compute
effect size
Outcome 3. Home
Situations pre-to-posttest
Unable to compute
effect size
Outcome 4. Individual
Problem List pre-topost-test
Unable to compute
effect size
Outcome 5. Homework
Problem Checklist preto-post-test
Unable to compute
effect size
Full Reference: Horn, W. F., Ialongo, N., Greenberg, G., Packard, T., & SmithWinberry, C. (1990). Additive effects of behavioral parent training and self-control
therapy with attention deficit hyperactivity disordered children. Journal of Clinical
Child Psychology, 19(2), 98 – 110.
Weight of Evidence A
A.1 Quality of Measures
Check
Reliability and Validity
Multi-method –
Multi-source –
n/a
Outcomes 1: (Parent) – The Child Behaviour Checklist
Reliability Validity –
Outcome 2:
Reliability =
Validity =
Outcome 3:
Reliability =
Validity =
Overall Rating for measurement
(3) Strong evidence: case for validity, reliability above 0.85 (in majority or primary
outcomes), multiple methods and multiple sources
(2) Promising evidence: reliability above 0.70, multiple methods or multiple sources
(1) Weak evidence: reliability above 0.50, multiple methods or multiple sources
54
Doctorate in Educational and Child Psychology

Cara Southworth
(0) None of the above – Judgement: Only one measure for parents and no reliability
or validity noted
A.2 Quality of Control Measure
Check
Control Measure – several comparison groups
Type of comparison group
Typical intervention
Attention placebo
Intervention element placebo (intervention minus active ingredient linked to change)
Alternative intervention - PT

Pharmacotherapy
No intervention
Wait list/delayed intervention
Minimal contact
Unable to identify type of comparison – Non-ADHD control group – no details

Overall judgement of type of comparison group
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)

Unable to identify comparison group
Counterbalancing of change agent (e.g. teacher or therapist)
By change agent
Statistical
Other
Not reported/none

Group equivalence established
Random assignment

Posthoc matched set
Statistical matching
Post hoc test for group equivalence

Equivalent mortality
Low attrition (less than 20% for post)

Low attrition (less than 30% for follow up)

Intent to intervene analysis carried out (include findings):
Overall Rating for Control Measure
(3) Strong evidence: at least one active comparison group, group equivalence must be
established (preferably random assignment), counterbalancing, low attrition at post and
follow up (if applicable)
(2) Promising evidence: at least one ‘no intervention’ comparison group. Plus at least two

of: group equivalence must be established (preferably random assignment), or
counterbalancing or low attrition at post and follow up (if applicable)
(1) Weak evidence: at least one comparison group. Plus at least one of: group equivalence
must be established (preferably random assignment), or counterbalancing or low attrition
at post and follow up (if applicable)
(0) no efforts made to control
55
Doctorate in Educational and Child Psychology
Cara Southworth
A.3 Quality of Intervention
Group Experimental Design
Primary outcomes in appropriate unit of analysis
Primary outcomes familywise/experimenter wise error rate controlled when applicable
Primary outcomes have sufficiently large N
At least 75% of primary outcomes significant
50-74% of primary outcomes significant

25-49% of primary outcomes significant
Less than 25% of primary outcomes significant
Overall Rating for Intervention
(3) Strong evidence: Significant outcomes for at least 75%. Plus appropriateness of
analysis, familywise/experimenter wise error controlled and sufficiently large N.
(2) Promising evidence: Significant outcomes for 50-74%. Plus appropriateness of

analysis, familywise/experimenter wise error controlled and sufficiently large N.
(1) Weak evidence: Significant outcomes for 24-49%. Plus appropriateness of analysis,
familywise/experimenter wise error controlled and sufficiently large N.
(0) No/Limited evidence: Less than 25% significant outcomes and none of the above
Overall Rating for WoE A (0+2+2)/3 = 1.33
(3) High
(2) Medium
(1) Low




Weight of Evidence B
B.1 Design
Check
Single Case Design
Check Group Experimental Design
Type of Design:
Within –Series: Simple phase change
Non-randomised design
Within – Series: Complex phase
Non-randomised block design (between
change
participants)
Between – Series: Comparing two
Non-randomised block design (within
interventions
participants)
Between – Series: Comparing
Non-randomised hierarchical design
interventions with no interventions
Combined – Series: Multiple baseline
Optional coding for quasi-experimental
across participants
design
Combined – Series: Multiple baseline
across behaviours
Combined – Series: Multiple baseline
across settings
Combined – Series: Multiple probe
Mixed single participant design (tick
two of the above)
Mixed single participant and group
design (tick one of the above and one
in the right hand column)
Other (specify):
If randomisation used:
Completely randomised
Randomised block design (between participants e.g. matched classrooms)

Randomised block design (within participants)
Randomised hierarchical design (nested treatments)
Units of randomisation:
Individual

56
Doctorate in Educational and Child Psychology
Cara Southworth
Classroom
School
Other (specify):
Overall confidence of judgement on how participants were assigned:
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)

N/A
Unknown/unable to code
Overall Rating for WoE B
(3) High

(2) Medium
(1) Low
Weight of Evidence C
C.1 Sample
N
Referred to a university-based psychological clinic for treatment of chronic inattentiveness,
impulsivity, and over activity. Diagnosis of ADHD from DSM-III-R. Additional inclusion criteria:
not receiving medication for ADHD, absence of psychoses and intellectual difficulties. The
Peabody Vocabulary test used to screen for participant’s intellectual status. Control group
non-ADHD children.
60
Sample size at start of study
Current exposure to alternative intervention
Prior exposure to alternative intervention
No historical exposure to alternative intervention
Single Case Design Participant
N
additional information:
14
14
(14
18
P1
P2
P3
Overall Rating for WoE C
(3) High
(2) Medium

(1) Low
Weight of Evidence D
Overall Rating for WoE D (1+3+2) /3 = 2
(3) High
(2) Medium

(1) Low
57
Group Experimental Design
CB SC
PT
PT + SC) – excluded
Control
Attrition (drop out)
Doctorate in Educational and Child Psychology
Cara Southworth
Effect Sizes
Relevant
Outcomes
Type of effect size
Effect size
Outcome 1a. CBCL
(Total problems) CB
SC pre-to-post-test
Cohen’s d (Adjusted to Hedge’s g)
0.62
Outcome 1b. CBCL
(Total Problems) CB
SC pre-to-follow- up
Cohen’s d (Adjusted to Hedge’s g)
0.65
Cohen’s d (Adjusted to Hedge’s g)
-0.3
Cohen’s d (Adjusted to Hedge’s g)
-0.36
Cohen’s d (Adjusted to Hedge’s g)
0.58
Cohen’s d (Adjusted to Hedge’s g)
0.87
Cohen’s d (Adjusted to Hedge’s g)
-0.16
Cohen’s d (Adjusted to Hedge’s g)
-0.22
Outcome 3a. CBCL
(Hyperactivity) CB SC
pre-to-post-test
Cohen’s d (Adjusted to Hedge’s g)
0.17
Outcome 3b. CBCL
(Hyperactivity) CB SC
pre-to-follow-up
Cohen’s d (Adjusted to Hedge’s g)
0.18
Outcome 3c. CBCL
(Hyperactivity) CB SC
vs. PT Control post-test
Cohen’s d (Adjusted to Hedge’s g)
0.00
Outcome 3d. CBCL
(Hyperactivity) CB SC
vs. PT control follow-up
Cohen’s d (Adjusted to Hedge’s g)
0.10
Outcome 1c. CBCL
(Total Problems) CB
SC vs. PT Control posttest
Outcome 1d. CBCL
(Total Problems) CB
SC vs. PT Control
follow-up
Outcome 2a. CBCL
(Externalizing) CB SC
pre-to-post-test
Outcome 2b. CBCL
(Externalizing) CB SC
pre-to-follow-up
Outcome 2c. CBCL
(Externalizing) CB SC
vs. PT Control post-test
Outcome 2d. CBCL
(Externalizing) CB SC
vs. PT Control followup
58
Doctorate in Educational and Child Psychology
Cara Southworth
Full Reference: Miranda, A., & Jesús Presentación, M. (2000). Efficacy of cognitivebehavioral therapy in the treatment of children with ADHD, with and without
aggressiveness. Psychology in the Schools, 37(2), 169 – 182.
Weight of Evidence A
A.1 Quality of Measures
Check
Reliability and Validity
Multi-method – Different rating scales (parent: DSM-III-R; EPC)
Multi-source
Outcomes 1: DSM-III-R adapted questionnaire
Reliability Validity –
Outcome 2: The Scale of Behavioural Problems (EPC)
Reliability = This scale has an alpha coefficient of .88 and a test-retest correlation of .79 in

n/a
various subscales.
Validity = It also has adequate level of construct and criterial validity.
Outcome 3:
Reliability =
Validity =
Overall Rating for measurement
(3) Strong evidence: case for validity, reliability above 0.85 (in majority or primary
outcomes), multiple methods and multiple sources
(2) Promising evidence: reliability above 0.70, multiple methods or multiple sources
(1) Weak evidence: reliability above 0.50, multiple methods or multiple sources –

Judgment: multiple methods, reliability not calculated for this study population.
(0) None of the above
A.2 Quality of Control Measure
Check
Control Measure –
Type of comparison group
Typical intervention
Attention placebo
Intervention element placebo (intervention minus active ingredient linked to change)
Alternative intervention
Pharmacotherapy
No intervention
Wait list/delayed intervention
Minimal contact
Unable to identify type of comparison
Overall judgement of type of comparison group
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)
Unable to identify comparison group
Counterbalancing of change agent (e.g. teacher or therapist)
By change agent
Statistical
Other
59
Doctorate in Educational and Child Psychology
Cara Southworth
Not reported/none
Group equivalence established
Random assignment
Posthoc matched set
Statistical matching
Post hoc test for group equivalence
Equivalent mortality
Low attrition (less than 20% for post)

Low attrition (less than 30% for follow up)

Intent to intervene analysis carried out (include findings):
Overall Rating for Control Measure
(3) Strong evidence: at least one active comparison group, group equivalence must be
established (preferably random assignment), counterbalancing, low attrition at post and
follow up (if applicable)
(2) Promising evidence: at least one ‘no intervention’ comparison group. Plus at least two
of: group equivalence must be established (preferably random assignment), or
counterbalancing or low attrition at post and follow up (if applicable)
(1) Weak evidence: at least one comparison group. Plus at least one of: group equivalence
must be established (preferably random assignment), or counterbalancing or low attrition
at post and follow up (if applicable)
(0) no efforts made to control

A.3 Quality of Intervention
Group Experimental Design
Primary outcomes in appropriate unit of analysis

Primary outcomes familywise/experimenter wise error rate controlled when applicable

Primary outcomes have sufficiently large N
At least 75% of primary outcomes significant

50-74% of primary outcomes significant
25-49% of primary outcomes significant
Less than 25% of primary outcomes significant
Overall Rating for Intervention
(3) Strong evidence: Significant outcomes for at least 75%. Plus appropriateness of
analysis, familywise/experimenter wise error controlled and sufficiently large N.
(2) Promising evidence: Significant outcomes for 50-74%. Plus appropriateness of

analysis, familywise/experimenter wise error controlled and sufficiently large N. –
Judgement – appropriate significance and familywise error rate controlled.. but not
sufficiently large N.
(1) Weak evidence: Significant outcomes for 24-49%. Plus appropriateness of analysis,
familywise/experimenter wise error controlled and sufficiently large N.
(0) No/Limited evidence: Less than 25% significant outcomes and none of the above
Overall Rating for WoE A (0+1+2)/3 = 1
(3) High
(2) Medium
(1) Low

60
Doctorate in Educational and Child Psychology
Cara Southworth
Weight of Evidence B
B.1 Design
Check
Single Case Design
Check Group Experimental Design
Type of Design:
Within –Series: Simple phase change
Non-randomised design
Within – Series: Complex phase
Non-randomised block design (between
change
participants)
Between – Series: Comparing two
Non-randomised block design (within
interventions
participants)
Between – Series: Comparing
Non-randomised hierarchical design
interventions with no interventions
Combined – Series: Multiple baseline
Optional coding for quasi-experimental

across participants
design
A pretest – posttest (+ follow-up) design – no
Combined – Series: Multiple baseline
control group.
across behaviours
Combined – Series: Multiple baseline
across settings
Combined – Series: Multiple probe
Mixed single participant design (tick
two of the above)
Mixed single participant and group
design (tick one of the above and one
in the right hand column)
Other (specify):
If randomisation used:
Completely randomised
Randomised block design (between participants e.g. matched classrooms)
Randomised block design (within participants)
Randomised hierarchical design (nested treatments)
Units of randomisation:
Individual
Classroom
School
Other (specify):
Overall confidence of judgement on how participants were assigned:
Very low (little basis)
Low (guess)
Moderate (weak inference)
High (strong inference)
Very high (explicitly stated)
N/A
Unknown/unable to code
Overall Rating for WoE B
(3) High
(2) Medium
(1) Low

61
Doctorate in Educational and Child Psychology
Cara Southworth
Weight of Evidence C
C.1 Sample
N
Diagnosed by the school psychologists as having chronic problems of inattention,
impulsivity, and over activity. Teachers interviewed to corroborate. Then had to meet this
criteria: score 15 + on Conners rating Scale, score 16 or higher on DSM-III-R, 1+ persistence
of symptoms, IQ score 80+ of WISC, absence of psychosis or any gross neurological, sensory
or motor impairment. Iowa Scale for Teachers screened for aggressiveness.
32
Sample size at start of study
Current exposure to alternative intervention – only reported significance data for whole
non-aggressive group & therefore some receiving AM as well as CB SC. Unable to
isolate only CB SC – so sample also receiving AM.
Prior exposure to alternative intervention
No historical exposure to alternative intervention
Single Case Design Participant
N
Group Experimental Design
additional information:
P1
8
ADHD + CB SC
P2
8
ADHD + CB SC & AM
P3
(8
ADHD aggression + CB SC)
(8
ADHD aggression + CB SC & AM)
Attrition (drop out)
Overall Rating for WoE C
(3) High
(2) Medium
(1) Low


Weight of Evidence D
Overall Rating for WoE D (1+1+1) /3 = 1
(3) High
(2) Medium
(1) Low

Effect Sizes
Relevant
Outcomes
Type of effect size
Effect size
Comment
Outcome 1a. DSM-IIIR (Hyperactivity) preto-post-test
Unable to calculate
effect size
Outcome 1b. DSM-IIIR (Hyperactivity) preto-follow-up
Unable to calculate
effect size
Outcome 2a. EPC
(Antisocial Behaviour)
pre-to-post-test
Unable to calculate
effect size
Outcome 2b. EPC
(Antisocial Behaviour)
pre-to-follow-up
Unable to calculate
effect size
Outcome 3a. EPC
(Internalisation) pretest-to-post-test
Unable to calculate
effect size
62
Doctorate in Educational and Child Psychology
Relevant
Outcomes
Type of effect size
Effect size
Cara Southworth
Comment
Outcome 3b. EPC
(Internalisation) poreto-follow-up
Unable to calculate
effect size
Outcome 4a. EPC
(School Problems) preto-post-test
Unable to calculate
effect size
Outcome 4b. EPC
(School Problems) preto-follow-up
Unable to calculate
effect size
Outcome 5a. EPC
(Psychopath disorders)
pre-to-post-test
Unable to calculate
effect size
Outcome 5b. EPC
(Psychopath disorders)
pre-to-follow-up
Unable to calculate
effect size
63
Doctorate in Educational and Child Psychology
Cara Southworth
Appendices References
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.).
Hillsdale, NJ: Erlbaum.
Gough, D. (2007). Weight of evidence: A framework for the appraisal of the quality
and relevance of evidence. In J. Furlong & A. Oancea (Eds.) Research Papers
in Education, 22(2), 213-228.
Kratochwill, T. R., & Steele Shernoff, E. (2004). Evidence-Based Practice: Promoting
Evidence-Based Interventions in School Psychology. School Psychology
Review, 33, 34-48.
NICE (2009). Attention deficit hyperactivity disorder: Diagnosis and management of
ADHD in children young people and adults. London, National Institute for
Health and Clinical Excellence.
64