Research in Developmental Disabilities

23 (2002) 332±341

Early intensive behavioral intervention

for children with autism: parental
therapeutic self-ef®cacy
Richard P. Hastings*, Matthew D. Symes
Center for Behavioural Research Analysis and Intervention in Developmental Disabilities,
Department of Psychology. University of Southampton, Southampton, UK
Received 21 February 2002; received in revised form 30 April 2002; accepted 13 May 2002

Abstract

Several authors have suggested that thequality of therapist performance accounts forsome
of the variability in outcomes observed in early intensive behavioral intervention for children
with autism. However, there is a distinct lack of theoretical and empirical work addressing
therapist performance in this context. In the present study, we explored predictors of one
variable, beliefs about one's ef®cacy in the therapeutic role, that may be related to therapist
performance. Eighty-®ve UK mothers who were acting as therapists for their child's program
completed a questionnaire survey. Results showed that program variables (e.g., number of
hours of therapy each week, time since program started) were unrelated to maternal
therapeutic self-ef®cacy. However, support received from the program, the severity of the
child's autism, and maternal stress were signi®cant predictors. Regression analysis also
showed that maternal stress mediated the impact of support from the program and autism
severity on maternal therapeutic self-ef®cacy. Limitations of the study, suggestions for future
research, and practical implications are brie¯y discussed. In particular, we advocate a role for
supervisors in analyzing and developing interventions for therapist behavior.
# 2002 Elsevier Science Ltd. All rights reserved.

Keywords: autism; self-ef®cacy; applied behavior analysis; early intervention

1. Introduction

Compelling empirical data focus on outcomes in clinic-directed applied

behavior analysis (ABA) and early intervention programs for children with
*
Corresponding author. Present address: School of Psychology, University of Wales, Bangor,
Gwynedd LL57 2AS, Wales, UK. Tel.: ‡44-1248-388214; fax: ‡44-1248-382599.

0891-4222/02/$ ± see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 8 9 1 - 4 2 2 2 ( 0 2 ) 0 0 1 3 7 - 3
 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341 333

autism (e.g., Lovaas, 1987; McEachin, Smith, & Lovaas, 1993). Within clinic-
directed programs, parents are often encouraged to be a full member of the
therapy team (Lovaas & Buch, 1997). However, increasingly parents are playing a
more central role in the organization and delivery of ABA interventions. This is
due, in part, to practical problems in obtaining services from suitably quali®ed
ABA practitioners and the ®nancial costs to families who are not in receipt of
assistance from government-funded agencies (Jacobson, 2000; Johnson & Hast-
ings, 2002). Data on programs where parents act as therapists, and on programs
that are parent-directed, are generally encouraging (e.g., Sheinkopf & Siegel,
1998; Smith, Buch, & Gamby, 2000). However, treatment gains have typically
been found to be less than those achieved in clinic-directed programs (Bibby,
Eikeseth, Martin, Mudford, & Reeves, 2001; Smith et al., 2000).
The sources of variance in outcomes of parent versus clinic-directed ABA
programs have received little systematic study. Mudford, Martin, Eikeseth, and
Bibby (2001) observed that parent-directed programs in the UK often failed to
meet standards of a model intervention (Lovaas, 1987). Speci®cally, children
were often older than 40 months at intake, the average number of therapy hours
per week was 32, no program was supervised on a weekly basis, and it was
dif®cult to establish the credentials of those supervising the programs. In a
subsequent analysis, these treatment variables were not signi®cantly related to
child outcomes (Bibby et al., 2001) apart from the child's age at intake. Younger
children had better outcomes, a pattern that has been found by other investigators
(e.g., Fenske, Zalenski, Krantz, & McClannahan, 1985; Harris & Handleman,
2000).
Smith et al. (2000) also addressed dimensions of the quality of parent-directed
programs. In a study of six young children with autism, observations of the
discrete trial implementation performance of the children's therapists were
compared with the performance of therapists working in a clinic-based program.
Although presentation of discriminative stimuli, discrete trials, and consequences
was correct on the vast majority of occasions in both groups, the clinic therapists
showed superior performance. Smith et al. (2000) suggest that such differences
may explain why early success on children's simple skills can be demonstrated in
parent-directed programs but this is rarely translated into best outcomes in the
longer term.
Although there has been some analysis of the impact on program outcome of
child variables such as IQ, age at intake, and language ability (Fenske et al., 1985;
Harris & Handleman, 2000; McEachin et al., 1993; Smith, Eikeseth, Klevstrand,
& Lovaas, 1997), therapist variables have not been related directly to intervention
ef®cacy. A working hypothesis would be that therapist behavior (e.g., adherence
to treatment protocols) is a salient variable in contributing to the success of ABA
programs for children with autism. This question is empirical and remains to be
tested. Assuming that therapist performance is in¯uential, it will also be important
to consider factors that account for variation in therapist behavior or performance.
At present, there is no theory of therapist performance in ABA programs for
children with autism. However, we might expect that child, program, and therapist
334 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341

variables will be salient (cf. Graziano & Katz, 1982). Each of these is considered
in turn. First, behavioral researchers have shown for some time that child variables
affect parent, teacher and staff behavior (e.g., Berberich, 1971; Emery, Binkoff,
Houts, & Carr, 1983; Patterson, 1982; Sherman & Cormier, 1974). In particular,
adults reliably behave in a manner that serves to avoid or escape severe behavior
problems that function as aversive stimuli (Carr, Taylor, & Robinson, 1991;
Taylor & Carr, 1992). In the context of ABA programs for children with autism,
this negative reinforcement process may play a role but one would also hope that
positive reinforcement of therapist behavior also occurs due to improvements in
children's adaptive behaviors.
Second, therapist performance is also likely to be affected by various program
variables. These factors might include the quality of training provided, frequency
and quality of supervision, support provided by therapy team members, and the
underlying ef®cacy of the behavioral technology. By way of example, a key factor
in determining performance of therapists applying various ABA techniques is
continued monitoring and feedback by a supervisor (e.g., McConnachie & Carr,
1997). A third set of factors that may in¯uence therapist performance relate to the
therapists themselves. Aspects of therapists' personality (e.g., extroverts may ®nd
it easier to apply social reinforcers), mental health/stress, and beliefs may all be
important. For example, staff working in mental retardation services have been
found to interact less positively with clients if they are experiencing increased
stress or burnout (Lawson & O'Brien, 1994; Rose, Jones, & Fletcher, 1998). The
integrity of ABA interventions may well be compromised under such conditions.
The present research addresses one therapist factor that may have an impact on
their performance in ABA programs for children with autism: therapists' beliefs
about their self-ef®cacy in the therapeutic role. This factor may have bene®ts at
the behavioral level. In particular, there are strong contingencies acting to punish
therapist adherence to behavioral technologies. These include temporal distance
between therapist behavior and positive changes in the child and social dis-
approval from members of the therapists' verbal community (Allen & Warzak,
2000). It is possible that therapists' belief in their ef®cacy in the therapeutic role
will help to mitigate against these contingencies (e.g., improving the chances that
they will adhere to the program long enough for their behavior to come under the
control of child improvement contingencies). In this manner, therapeutic self-
ef®cacy may interact with establishing operations set up by supervisors (cf. Allen
& Warzak, 2000) such as: ``You may experience some tantrums and other
dif®culties to start with, but these are signs that what you are doing is working.
They will soon settle down if you continue with the program.'' Although this is a
speculative argument at the present time, the research is signi®cant as one of only
a very small number of studies of therapist factors that may account for some of
the variance in therapist behavior and the outcomes of ABA programs for young
children with autism.
In the present study, we evaluated the relationship between the severity of the
child's symptoms of autism, social support provided by the program team, and
parents' perceptions of their ef®cacy as a therapist on ABA early intervention
 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341 335

programs. Speci®cally, we hypothesized that parent therapists would have lower

levels of self-ef®cacy when they were working with children with more severe
symptoms of autism and when they felt less supported by the program team. We
also explored a potential mechanism for this hypothesized effect: that child-
related stress might mediate the impact of symptom severity on therapeutic self-
ef®cacy. In previous research in the developmental disabilities ®eld, strong
relationships have been found between stress variables and therapeutic self-
ef®cacy (Hastings & Brown, 2002a, 2002b). The present analysis focused on a
sub-group of parents from a larger survey study of families where a young child
with autism was engaged in an ABA program (Hastings & Johnson, 2001). These
parents were acting as one of the therapists on their child's program team. Thus,
the present data are also signi®cant as they represent the ®rst systematic study of
the perceptions of parents in this therapeutic role in ABA programs for children
with autism.

2. Method

2.1. Participants

Participants were drawn from a larger survey of primary parental caregivers of

children with autism engaged in intensive ABA early intervention programs in the
UK (Hastings & Johnson, 2001). One hundred and forty-one parents participated.
In this paper, data are presented on a sub-sample where one or both parents were a
part of the therapy team for their young child with autism. Due to the small number
of fathers responding to the survey (11 from 141, nine of whom were a therapist for
their child's program), only mothers were included. Of the 130 mothers from the
full sample, 85 were working as a therapist on their child's program.
In all cases, the mothers were the biological parents of the child with autism.
The age of these mothers ranged from 26 to 47 years, with a mean of 36.78
(SD ˆ 4:50) years. Seventy-eight (91.8%) of the mothers were married, and
living with their spouse, four were living with a partner, and three were divorced,
separated, or single and not living with a partner. Almost every mother (96.5%)
reported that they were currently in a general state of ``good health.'' Further
demographic details of the mothers, their young child with autism, their family,
and the family's involvement in the ABA intervention program can be found in
Table 1. In every respect bar one, the present sample were representative of the
full sample reported by Hastings and Johnson (2001). The difference was that
more of the partners of the present sub-sample were also engaged as a therapist on
their child's program when compared with the full sample.

2.2. Materials

In addition to a questionnaire designed to elicit the demographic information

summarized above, four measures were included.
336 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341

Table 1
Further demographic characteristics of the sample

Characteristic Percentage of
sample/mean (SD)

Mother has bachelors, masters or doctoral degree 45.9

Mother in full or part-time employment (including ABA therapy) 96.5
Mother's partner in employment 95.3
Sex of child with autism
Male 88.2
Female 11.8
Age of child with autism (years) 4.96 (1.45)
Child with autism lives with mother 98.8
Mother's hours per week working on program 6.79 (3.62)
Mother's partner acting as therapist on child's program 32.9
Partner's hours per week working on programa 3.96 (1.99)
Length of time on program (months) 14.18 (11.63)
Age of child when program started (months) 45.52 (13.72)
a
Mean hours calculated for those acting as therapists on their child's program only, and not
across whole sample.

2.3. Autism Behavior Checklist

The Autism Behavior Checklist (ABC: Krug, Arick, & Almond, 1980) was used
to generate an overall index of the level of autism severity by symptoms. The
validity of the ABC has been questioned on several occasions (e.g., Sevin, Matson,
Coe, Fee, & Sevin, 1991; Wadden, Bryson, & Rodger, 1991). However, in the
present study, it was used simply as an index of autism symptomatology and not as a
diagnostic instrument. The checklist was presented to parents in the simple yes/no
format (Volkmar et al., 1988) and the total score only, calculated using the
weightings suggested by Krug et al. (1980), was used in the analyses reported below.

2.4. Maternal stress

Maternal stress was measured using the Parent and Family Problems sub-scale
of the Friedrich Short Form of the Questionnaire on Resources and Stress (QRS-
F: Friedrich, Greenberg, & Crnic, 1983). Because ®ve of the items in this sub-
scale of the QRS-F form a conceptually and psychometrically distinct depression
scale (Glidden & Floyd, 1997), they were removed from scoring the measure.
Thus, maternal stress was measured using a slightly shorter version of the Parent
and Family Problems scale.

2.5. Maternal therapeutic self-ef®cacy

Mothers rated perceptions of their ef®cacy as a therapist on the child's program

using a ®ve-item scale adapted from previous research on staff and parent self-
ef®cacy in dealing with severe problem behaviors in children with developmental
 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341 337

disabilities (Hastings & Brown, 2002a, 2002b). The items addressed dimensions
of mothers' beliefs about their ef®cacy as a therapist:
1. How con®dent are you in carrying out the tasks that make up your child's
Lovaas/ABA program?
2. How dif®cult do you personally ®nd it to carry out the tasks that make up
your child's Lovaas/ABA program?
3. To what extent do you feel that your contribution as a tutor/therapist on your
child's program makes a difference to his/her progress?
4. How satis®ed are you with the way in which you carry out the tasks that make
up your child's Lovaas/ABA program?
5. To what extent do you feel in control of your child's Lovaas/ABA program?
Each item was rated on a seven point scale anchored at each end with appropriate
descriptors (e.g., ``not at all con®dent'' to ``very con®dent'' for the ®rst item). A total
score was derived by summating the scores on all ®ve items. In previous research,
this self-ef®cacy scale has shown good reliability with Cronbach's alpha in excess
of .90 (Hastings & Brown, 2002a, 2002b). In the present sample of mothers, internal
consistency was also high (Cronbach's alpha ˆ :87).

2.6. Support from the ABA program team

The helpfulness of the ABA program in providing support to the mothers was
rated on a single ®ve-point item (ranging from ``not at all helpful'' to ``extremely
helpful'') from Dunst, Jenkins, & Trivette's (1984) Family Support Scale.

2.7. Procedure

The main survey sample (see Hastings & Johnson, 2001) were recruited
through the UK support organization for parents of children with autism engaged
in ABA programs. Four hundred and ®fty questionnaires were mailed out through
the support group to their parent members. Questionnaires were returned anon-
ymously to the research team. One hundred and forty-one members replied,
constituting a response rate of 31%. This response rate is at the low end of the
typical range for a postal survey of this sort. However, not all members of the
organization were parents of a child currently engaged in an early intervention
program. The support group's data at the time of the study suggested that
approximately 250 families were conducting intensive ABA early intervention
programs in the UK. If we assume that this ®gure is reasonably accurate, the true
response rate for the study is closer to 60%. There are no data available about non-
responding families as the survey was returned anonymously.

3. Results

Statistical analysis proceeded through two distinct stages. First, correlates of

maternal therapeutic self-ef®cacy were explored. Thus, all of the demographic
338 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341

Table 2
Regression analysis of maternal therapeutic self-ef®cacy

Step/predictor b p R2

Step 1 .22
Autism severity .294 .007
Support from ABA program .316 .004
Step 2 .31
Autism severity .171 .114
Support from ABA program .178 .106
Maternal stress .366 .003

variables and other measures described above were analyzed for their univariate
association with maternal therapeutic self-ef®cacy. Only three variables were
associated with maternal therapeutic self-ef®cacy at p < :05: (1) Mothers of
children with more severe symptoms of autism reported lower self-ef®cacy as a
therapist (Pearson r 85† ˆ :37, p < :001), (2) Mothers who experienced higher
levels of support from the ABA program reported higher self-ef®cacy
(r 76† ˆ :37, p < :001), and (3) Mothers who reported lower levels of stress
also reported higher self-ef®cacy (r 85† ˆ :53, p < :001).
For the main analysis, the procedure recommended by Baron and Kenny
(1986) for the exploration of mediated effects was followed. First, autism
symptom severity and support received from the ABA program were entered
as predictors in a regression analysis of therapeutic self-ef®cacy. Second,
maternal stress was added to the regression equation. A mediated effect would
be shown if autism severity initially made a signi®cant independent contribution
to the prediction of therapeutic self-ef®cacy but this effect reduced and became
non-signi®cant once maternal stress entered the equation and was shown as a
signi®cant predictor (Baron & Kenny, 1986). As no demographic or program
variables were related to therapeutic self-ef®cacy, these were not included in this
analysis. The results of the regression analysis are displayed in Table 2.
The regression analysis showed that the severity of the child's autism was a
negative predictor and support received from the ABA program was a positive
predictor of maternal therapeutic self-ef®cacy. However, once maternal stress
entered the regression equation the contribution of these variables became non-
signi®cant. Furthermore, maternal stress was a signi®cant negative predictor of
maternal therapeutic self-ef®cacy. Thus, there was evidence that maternal stress
mediated the impact of both autism severity and support from the ABA program
on maternal therapeutic self-ef®cacy.

4. Discussion

This study is one of a very small number addressing therapist variables that
may predict therapist performance in ABA programs for young children with
autism, and the ®rst to gather data from mothers acting as therapists on these
 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341 339

programs. Mothers' beliefs about their self-ef®cacy in the therapeutic role are
related to child, program, and maternal variables. Speci®cally, more severe autism
symptoms in the child, less supportive ABA program teams, and higher maternal
stress were predictive of lower levels of maternal therapeutic self-ef®cacy.
Furthermore, there was evidence that maternal stress mediated the impact of
the other variables on maternal therapeutic self-ef®cacy (i.e., their effect was
mainly via their impact on maternal stress).
It is important to remember that causal relationships have not been demon-
strated. Prospective studies of factors related to therapist performance are thus
required. There is also a need for the development of theoretical and empirical
work on therapist performance in ABA programs for young children with autism.
We have argued that perceived self-ef®cacy in the therapeutic role may well be a
salient variable (i.e., alongside objective assessments of therapist performance),
but its role is yet to be established empirically.
Another issue that requires research attention in this ®eld is the potential con¯ict
between mothers' and fathers' roles as parents and their role as a therapist for their
child. This is poorly understood, but a high proportion (67% in the present survey)
of ABA programs may involve one or both parents working in the therapist role.
Multiple roles with the child may enhance the effectiveness of a tutor, impede their
work, or have no measurable effect. Research is needed to answer this question so
that the most effective therapy teams are established for each child. There is also the
question of how multiple roles may affect the psychological well-being of the
parents acting as therapists. In the full sample, there was no evidence that parents
acting as therapists reported more stress than other parents (Hastings & Johnson,
2001) but more data are needed to address this issue.
Generalizability of the present results is a signi®cant question. A number of
potential sources of bias were operating in this study. In particular, the sample of
parents were a socially advantaged group, and there is a problem of source
variance (mothers reported on all of the main variables). Due to the response rate
obtained, it is unclear whether the results are representative of families engaged in
ABA interventions in the UK or elsewhere. A further point is that it is unknown
whether similar results would be obtained in research focusing on non-parental
therapists in ABA programs.
Although the practical utility of the data from the present study are yet to be
demonstrated, discussion of implications for practice emphasizes the signi®cance
of research efforts focusing on therapist behavior in ABA programs for young
children with autism. The present results suggest that therapists in less supportive
or cohesive ABA teams, working with children with more severe symptoms of
autism, and who are also experiencing higher levels of stress will have less of a
belief in their ef®cacy as a therapist. Attention to these variables, and factors
identi®ed in future research, may help in the selection and continued support of
therapy teams by supervisors of ABA programs. Given the potential mediational
role of stress, the primary variables for intervention with therapists might be team
support and techniques for reducing exposure to autism symptoms in the child, or
at least therapist perceptions of these symptoms.
340 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341

The main implication of the arguments put forward in this paper is that ABA
programs for young children with autism need to maintain a broad focus. Thus,
the supervisor's role should not simply be to review and advise on aspects of the
child's program but also to gather data and to develop support interventions for
the team working with the child. As Allen and Warzak (2000) state in the title of
their functional assessment of parental adherence to behavioral interventions:
``Effective treatment is not enough.''

References

Allen, K. D., & Warzak, W. J. (2000). The problem of parental non-adherence in clinical behavior
analysis: Effective treatment is not enough. Journal of Applied Behavior Analysis, 33, 373±391.
Baron, R. M., & Kenny, D. A. (1986). The moderator±mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173±1182.
Berberich, J. P. (1971). Do the child's responses shape the teaching behavior of adults? Journal of
Experimental Research in Personality, 5, 92±97.
Bibby, P., Eikeseth, S., Martin, N. T., Mudford, O. C., & Reeves, D. (2001). Progress and outcomes
for children with autism receiving parent-managed intensive interventions. Research in
Developmental Disabilities, 22, 425±447.
Carr, E. G., Taylor, J. C., & Robinson, S. (1991). The effects of severe behavior problems in children
on the teaching behavior of adults. Journal of Applied Behavior Analysis, 24, 523±535.
Dunst, C. J., Jenkins, V., & Trivette, C. M. (1984). The family support scale: Reliability and validity.
Journal of Individual, Family, and Community Wellness, 1, 45±52.
Emery, R. E., Binkoff, J. A., Houts, A. C., & Carr, E. G. (1983). Children as independent variables:
Some clinical implications of child effects. Behavior Therapy, 14, 398±412.
Fenske, E. C., Zalenski, S., Krantz, P. J., & McClannahan, L. E. (1985). Age at intervention and
treatment outcome for autistic children in a comprehensive intervention program. Analysis and
Intervention in Developmental Disabilities, 5, 49±58.
Friedrich, W. N., Greenberg, M. T., & Crnic, K. (1983). A short form of the Questionnaire on
Resources and Stress. American Journal of Mental Deficiency, 88, 41±48.
Glidden, L. M., & Floyd, F. J. (1997). Disaggregating parental depression and family stress in
assessing families of children with developmental disabilities: A multi-sample analysis. American
Journal on Mental Retardation, 102, 250±266.
Graziano A. M., & Katz, J. N. (1982). Training paraprofessionals. In A. S. Bellack, M. Hersen, & A.
E. Kazdin (Eds.), International handbook of behavior modification and therapy (pp. 207±229).
New York: Plenum Press.
Harris, S. L., & Handleman, J. S. (2000). Age and IQ at intake as predictors of placement for young
children with autism: A four-to six-year follow-up. Journal of Autism and Developmental
Disorders, 30, 137±142.
Hastings, R. P., & Brown, T. (2002a). Behavior problems of autistic children, parental self-efficacy
and mental health. American Journal on Mental Retardation, 107, 222±232.
Hastings, R. P., & Brown, T. (2002b). Behavioural knowledge, causal beliefs, and self-efficacy as
predictors of special educators' emotional reactions to challenging behaviors. Journal of
Intellectual Disability Research, 46, 144±150.
Hastings, R. P., & Johnson, E. (2001). Stress in UK families conducting intensive home-based
behavioral intervention for their young child with autism. Journal of Autism and Developmental
Disorders, 31, 327±336.
Jacobson, J. W. (2000). Early intensive behavioral intervention: Emergence of a consumer-driven
service model. Behaviour Analyst, 23, 149±171.
 R.P. Hastings, M.D. Symes / Research in Developmental Disabilities 23 (2002) 332±341 341

Johnson, E., & Hastings, R. P. (2002). Facilitating factors and barriers to the implementation of
intensive home-based behavioural intervention for young children with autism. Child: Care,
Health and Development, 28, 123±129.
Krug, D. A., Arick, J., & Almond, P. (1980). Behavior checklist for identifying severely handicapped
individuals with high levels of autistic behaviour. Journal of Child Psychology and Psychiatry,
21, 221±229.
Lawson, D. A., & O'Brien, R. M. (1994). Behavioral and self-report measures of burnout in
developmental disabilities. Journal of Organizational Behavior Management, 14, 37±54.
Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in
young autistic children. Journal of Consulting and Clinical Psychology, 55, 3±9.
Lovaas, O. I., & Buch, G. (1997). Intensive behavioral intervention with young children with autism.
In N. N. Singh (Ed.), Prevention and treatment of severe behavior problems: Models and methods
in developmental disabilities (pp. 61±86). Pacific Grove, CA: Brooks/Cole.
McConnachie, G., & Carr, E. G. (1997). The effects of child behavior problems on the maintenance
of intervention fidelity. Behavior Modification, 21, 123±158.
McEachin, J. J., Smith, T., & Lovaas, O. I. (1993). Long-term outcome for children with autism who
received early intensive behavioral treatment. American Journal on Mental Retardation, 97, 359±
372.
Mudford, O. C., Martin, N. T., Eikeseth, S., & Bibby, P. (2001). Parent-managed behavioral treatment
for preschool children with autism: Some characteristics of UK programs. Research in
Developmental Disabilities, 22, 173±182.
Patterson G. R. (1992) Coercive family process. Eugene: Castalia Press.
Rose, J., Jones, F., & Fletcher, C. B. (1998). Investigating the relationship between stress and worker
behaviour. Journal of Intellectual Disability Research, 42, 163±172.
Sevin, J. A., Matson, J. L., Coe, D. A., Fee, V. E., & Sevin, B. M. (1991). A comparison and
evaluation of three commonly used autism scales. Journal of Autism and Developmental
Disorders, 21, 417±432.
Sheinkopf, S. J., & Siegel, B. (1998). Home-based behavioral treatment of young children with
autism. Journal of Autism and Developmental Disorders, 28, 15±23.
Sherman, T. M., & Cormier, W. H. (1974). An investigation of the influence of student behavior on
teacher behavior. Journal of Applied Behavior Analysis, 7, 11±21.
Smith, T., Buch, G. A., & Gamby, T. E. (2000). Parent-directed, intensive early intervention for
children with pervasive developmental disorder. Research in Developmental Disabilities, 21,
297±309.
Smith, T., Eikeseth, S., Klevstrand, M., & Lovaas, O. I. (1997). Intensive behavioral treatment for
preschoolers with severe mental retardation and pervasive developmental disorder. American
Journal on Mental Retardation, 102, 238±249.
Taylor, J. C., & Carr, E. G. (1992). Severe problem behaviors related to social interaction 2: A
systems analysis. Behavior Modification, 16, 336±371.
Volkmar, F. R., Cicchetti, D. V., Dykens, E., Sparrow, S. S., Leckman, J. F., & Cohen, D. J. (1988).
An evaluation of the Autism Behavior Checklist. Journal of Autism and Developmental
Disorders, 18, 81±97.
Wadden, N. P. K., Bryson, S. E., & Rodger, R. S. (1991). A closer look at the Autism Behavior
Checklist: Discriminant validity and factor structure. Journal of Autism and Developmental
Disorders, 21, 529±541.