The Scientist–Practitioner Model in Clinical Psychology: A Critique

Abstract

The scientist–practitioner philosophy of clinical practice is first placed in its historical context. The history of the scientist–practitioner stance is marked by an enduring struggle between advocates and opponents of a research-based profession of clinical psychology. The criticisms of the scientist–practitioner stance are discussed alongside counter-arguments in support of that position. Making the conceptual distinction between theoretical preference and philosophical position, three philosophies to underpin clinical psychology are outlined. It is concluded that the fundamental research to decide on the optimum stance for effective practice, and hence to inform training, is sadly lacking. © 1997 John Wiley & Sons, Ltd.

Full text

The Scientist±Practitioner Model in
Clinical Psychology: A Critique
Clive G. Long1, Clive R. Hollin2*
1St. Andrews Hospital, Northampton, UK
2University of Leicester, Leicester, UK
The scientist± practitioner philosophy of clinical practice is first placed
in its historical context. The history of the scientist± practitioner stance
is marked by an enduring struggle between advocates and opponents
of a research-based profession of clinical psychology. The criticisms
of the scientist± practitioner stance are discussed alongside counter-
arguments in support of that position. Making the conceptual distinc-
tion between theoretical preference and philosophical position, three
philosophies to underpin clinical psychology are outlined. It is
concluded that the fundamental research to decide on the optimum
stance for effective practice, and hence to inform training, is sadly
lacking. #1997 by John Wiley & Sons, Ltd.
Clin Psychol Psychother 4, 75± 83, 1997.
No. of Figures: 0. No. of Tables: 0. No. of Refs: 78.
INTRODUCTION
In the sense in which it is used here, the term
`scientist± practitioner' is used to refer to those
clinicians who personally conduct research and/or
draw on research to inform their practice. Such
clinicians possess therapeutic competence as well as
expertise in applied research, and may carry out
`clinical research' (Wilson, 1981) or `applicable
research' (Watts, 1984). The undertaking of research
is, of course, fundamental to the goal of an
empirically-based approach to the profession of
clinical psychology.
For four decades this research approach has
been the dominant model of professional practice
in clinical psychology. However, since it was first
proposed by the American Psychological Associa-
tion (APA) Committee on Training in Clinical
Psychology (1947), the Scientist± Practitioner model
of professional training has been the source of much
debate (Albee, 1970; Thelen and Ewing, 1970;
Peterson, 1976, 1985; Shakow, 1978; Barlow, 1981;
Frank, 1984; Garfield, 1986; Prokasky, 1968; Pilgrim
and Treacher, 1992). One focus of the debate is
whether clinical psychologists actually do combine
the roles of scientist and practitioner (Bornstein and
Wollersheim, 1978; Barlow, 1981) and, indeed,
whether there is any need to do so (Peterson,
1976; Shakow, 1978). The abandonment of the
scientist± practitioner model in parts of the USA
has led to the development of professional schools
of psychology that emphasize the training of
clinical psychologists for professional practice with
less emphasis on research training. In contrast a
number of British clinicians see the relatively new
extended 3-year training course for clinical psychol-
ogists as an opportunity to `realize more effectively
the scientist± practitioner philosophy by improving
the levels and breadth of research training skills and
practice' (Carr, 1990, p. 18). Before considering
the implications of the adoption of a scientist±
practitioner approach, it is helpful to place the issue
in historical context.
SCIENTIST±PRACTITIONER IN
HISTORICAL CONTEXT
As discussed by Raimy (1950), in the late 1940s the
scientist± practitioner model of training in clinical
* Correspondence to: Clive R. Hollin, Center for Applied
Psychology, University of Leicester, Leicester LE1 7RH, UK.
We would like to thank an anonymous reviewer for their
helpful and constructive comments.
CCC 1063±3995/97/020075 ±09$17.50
#1997 by John Wiley & Sons, Ltd.
Clinical Psychology and Psychotherapy, Vol. 4 (2), 75± 83 (1997)

psychology was given strong support by the
American Psychological Association. Such public
support represented an attempt to achieve two
aims: to heal the rift between academic and applied
psychologists; and, through advocating training in
research, to help clinical psychologists establish a
position independent of psychiatry in the mental
health field.
The often cited scientist± practitioner split
continued however, to be a source of discontent,
influenced by a climate in which some clinical
procedures were judged as unproven (e.g. Eysenck,
1965; Bergin and Strupp, 1972). In reality, several
other issues fuelled the debate: a view that natural-
istic research was scientifically unacceptable (Bergin
and Strupp, 1972; the position taken by some
clinicians that scientifically rigorous group compar-
ison research designs had difficulties in dealing with
the complexities and idiosyncrasies of the individual
(Chassan, 1967); and an opinion that process
research was incapable of evaluating the clinical
effects of treatment (Barlow and Hersen, 1984).
Indeed, prominent clinicians such as Matarazzo
stated that applied research had little influence on
their clinical practice: `Even after 15 years, few of my
research findings effect my practice. Psychological
science per se doesn't guide me one bit. I still read
avidly, but this is of little direct practical help. My
clinical experience is the only thing that has helped
me in my practice to date' (Bergin and Strupp, 1972,
p. 340). As this view prevailed among clinicians
who were well acquainted with research method-
ology, it is hardly surprising that clinicians without
such training were unaffected by the promise of
scientific evaluation of behaviour change.
Cohen's (1977, 1979) analysis of data from a series
of surveys, showed that 40% of mental health
professionals thought that there was no relevant
research to inform clinical practice, and that the
remainder believed that less than 20% of research
articles had any applicability to professional set-
tings. Regardless of where research stands, on a
continuum from poorly controlled single case
studies to multicentre randomized control trials,
tensions between scientists and practitioners con-
tinued to exist. This tension led to frustrations, as
seen in Wilson's (1981) pleas to `put aside the
invincible obstinacy of the fringe psychotherapists
who eschew a scientific approach to the develop-
ment and evaluation of treatment methods and
those negativists who merely bemoan the complex-
ity of our subject matter' (p. 218).
The growth of behaviour therapy provided the
setting conditions for an increase in the quantity of
clinical research (Agras and Berkowitz, 1980). As
this growth took place, so scientific quality grew
as flexible and innovative methodologies and
measures were introduced, ranging from group to
single case experimental designs (e.g. Barlow and
Hersen, 1984; Peck, 1985). In this climate, the
scientist± practitioner model, most elaborately docu-
mented by Barlow et al. (1984), has maintained its
impetus in recent years (e.g. Watts, 1984, 1992).
A series of articles in American Psychologist has
sought to reaffirm a commitment to the scientist±
practitioner model for training clinical psychologists
(Belar and Perry, 1992; O'Sullivan and Quevillon,
1992).
Nonetheless, criticisms and reservations about
the validity of a scientist±practitioner model in
clinical psychology have been expressed. These
reservations rightly continue to inform current
debate on the appropriate model for training and
practice in clinical psychology.
CURRENT DEBATE AND THE
SCIENTIST±PRACTITIONER
Clinician and Researcher: Conflicting Roles?
Frank (1984) suggested that there were essentially
two major criticisms of the scientist ±practitioner
model: (i) there is no need for a research training to
be a clinician; (ii) the interests and talents necessary
for research work and clinical work are incompat-
ible and impossible to combine.
There is no Need for Research Training
for the Clinician
This view has been supported by a number of
writers who argue that there is no need to combine
the role of researcher and clinician (Meehl, 1971;
Peterson, 1976; Frank, 1984; Pilgrim and Treacher,
1992). The position is held that the therapeutic task
of the clinical psychologist is sufficiently compli-
cated without including the role of researcher
(Pilgrim and Treacher, 1992). Yet further, if a
research role is given to all students, it might crush
intellectual curiosity especially in those clinically-
minded trainees who are forced to conduct it
(Barrom et al., 1988).
The Roles of Scientist and Practitioner
are Incompatible
Frank (1984) advances the position that person-
ality differences between scientist/researchers and
professional/practitioners cause major problems for
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
76 C. G. Long and C. R. Hollin

the scientist± practitioner model. The basis of this
position is that the two groups cannot coexist
within the same profession because of invisible
but powerful repelling forces between the two
extremes. Phillips (1989), however, reminds us that
all psychologists can be seen as sharing a common
frame of reference regardless of area of application.
Although research has found differences in value
systems aligned with the research/practitioner
distinction, this separation is not as distinct or
clear-cut as often supposed.
However, judged overall, there is no evidence to
support or refute these assertions. Indeed, the con-
trary view has been expressed that a clinical
training that neglects research skills or fails to
describe accurately an empirically-based practice
framework results in an education that emphasizes
`mystery over mastery' (Gambril, 1990, p. 343).
More forcefully, Medawar (1984) argues that quasi-
scientific psychologists `are getting away with a
concept of truthfulness that belongs essentially to
imaginative literature' (p. 58). He describes this
approach as `poeticism' which `stands for the belief
that imaginative insight and mysteriously privil-
eged sensibility can tell us all the answers that are
truly worthy of being sought or being known'
(p. 60). Such a preference for mystery results in the
neglect of research data concerning the effectiveness
of different styles and types of intervention. Indeed,
invoking mystery may be a major reason for the
`imposter syndrome': the feeling expressed by
some therapists that they are not really therapists
(DeAngelis, 1987).
In addition to the two points raised by Frank
(1984), there are practical questions concerning the
degree to which research is both applicable and
useful.
Research is Inapplicable and Poorly Supported
A number of clinicians hold negative attitudes
towards research, which is portrayed as irrelevant
to practice and ranking below more pressing service
commitments (Pilgrim and Treacher, 1992). Thus,
the criticism is made that the scientist±practitioner
model is not based on the reality of the settings in
which clinicians must operate. It may be that for
these reasons research often meets organizational
constraints and resistance (Salkovskis, 1984).
As O'Donohue and Szymanski (1994) note,
nominal and global judgements about research
such as `relevant/not relevant' are inappropriate.
The relevance of research is best seen as a
continuum from directly relevant to much less so,
so that practitioners must make real-world tests of
the applicability of an individual piece of research.
Further, as Abrahamson and Pearlman (1993) point
out, a research base is a necessity for psychology's
continued development as a profession, with an
identity relative to other mental health disciplines.
The poor UK tradition of conducting applicable
research is not a reason for the abandonment of
research. The last two decades have, for example,
seen a shift towards a greater emphasis on research
that is directly relevant to clinical practice (Barlow,
1981; Omer and Dar, 1992). The distinction between
applied and applicable research is important here
when considering the scientist± practitioner model
(Belbin, 1979). The essential characteristic of `applic-
able' research is a strong orientation towards a
practical problem with the goal of making convin-
cing recommendations for its solution. Applied
research, in contrast, is oriented more towards the
core area of psychology that is being applied or the
methodology employed, than to the solution of a
problem. As Watts (1984) notes, the tradition of
applicable research is weak and the research
undertaken by UK psychologists is applied but
often not applicable. However, the twinning of
research and practical objectives is central to what is
sometimes called action research. There are many
examples of research that have directly improved
clinical services. Paul and Lentz (1977) demon-
strated the advantages, in terms of efficacy and cost
effectiveness, of social learning methods in the
treatment of chronic institutionalized mental
hospital patients. In Britain, Lavender's (1985) work
on long-term psychiatric wards provided reliable
measures of the extent to which a ward met quality
standards of patient care and management, and led
to improvement of standards in the hospital.
Yet further, research can actually act as an agent
of change. The effect of research needs to be
measured not only in terms of its potential
advantage in answering important therapy
questions, but also in terms of its indirect effect
on clinical practice and the treatment milieu. James
et al. (1990) found that staff discussion of research
findings using the Ward Atmosphere Scale contrib-
uted to several changes in unit procedures and to
reduction in the real± ideal discrepancies for the
systems maintenance elements of the programme.
There is evidence that research and treatment may
enhance each other (Sacks et al., 1975; Braff et al.,
1979), and that research can even be therapy
(Rashkis, 1960; Leigh, 1975). Carroll et al.'s (1980)
research supports the notion that `patients have a
tendency to improve in a setting where structure
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
Scientist± Practitioner 77

has been imposed by a research design' (p. 379).
Specifically, it was found that the introduction of
the research protocol led to a more accurate
diagnosis, and more individualized and rigorously
monitored treatment regimes: alongside which the
methodology necessitated increased staff involve-
ment so that the research programme served as an
impetus to improve treatment. While acknow-
ledging the potential for a Hawthorn effect, Carroll
et al. were able to detect a cause ±effect relationship
that flowed directly from the implementation of
research.
A similar conclusion can be drawn from a study
by Ben-Arie et al. (1990) who found that their
research cohort had fewer readmissions to a
psychiatric hospital despite the presence of factors
(e.g. significantly greater number and frequency of
previous admissions in the research groups) that
would be expected to be associated with a contrary
result.
In the field of working with offenders the impact
of the meta-analytic studies (e.g. Lo
Èsel and Ko
Èferl,
1989; Andrews et al., 1990; Lipsey, 1992; Lo
Èsel, 1995)
has been dramatic. From a position of `nothing
works' the issue has quickly become one of `what
works', leading to a revival of a once moribund area
of clinical endeavour (Palmer, 1992; Hollin, 1993;
McGuire, 1995). Indeed, one of the findings of note
from the meta-analyses is that programmes with
high treatment integrity have the greatest impact on
recidivism (Hollin, 1995). One of the key correlates
of high integrity is an active research programme
evaluating the intervention, providing a perfect
example of the positive interplay between research
and effective practice.
Clinical Psychologists Rarely Produce
or Utilize Research
Although many clinicians pay lip service to the
importance of the scientist± practitioner model,
adherence to the model may be an ideal rather
than a reality (Prochaska and Norcross, 1983).
Studies show that clinicians have a low involve-
ment in research after graduating (Garfield and
Kurtz, 1976) with low productivity (Kelly et al.,
1978), and a modal frequency of research publica-
tions of zero (Barrom et al., 1988). Further,
psychologists consume and use research to only a
limited extent (Cohen, 1977, 1979). If this is true,
why train practitioners in research skills?
However, at least three studies have questioned
the validity of these findings in recent years, and
have supported the viability of the scientist±
practitioner model (Bornstein and Wollesheim,
1978; Barrom et al., 1988; Milne et al., 1990).
Bornstein and Wollersheim (1978), who surveyed
the scientist± practitioner activities of the APA's
Division of Clinical Psychology, found that
although both university and non-university beha-
viourists were more active than non-behaviourists
with regard to publications and formal papers, the
latter were more active producers of research than
was generally assumed.
Both Barrom et al. (1988) in the USA, and Milne
et al. (1990) in Britain, have noted that apparent
problems in research production, consumption and
utilization spring from too narrow a definition of
research. They argue that research needs to be
construed along a continuum from poorly con-
trolled single case studies, to multicentred random-
ized controlled trials. Barrom et al.'s (1988) random
sample survey of the scholarly activities of PhD and
PsyD trained psychologists found, like other
studies, that practising clinicians produced few
publications but that they were active in research in
other ways, and were consumers of research. In
particular the majority of clinicians either were
currently or had recently been involved in some
form of scholarly production, were active consu-
mers of scholarship, had a positive attitude towards
scholarship, and thought that research should
continue to be an important part of clinical
training. Analysis also suggested that setting
characteristicsÐsuch as the number of paid work
hours that could be devoted to research, and the
percentage of colleagues doing researchÐ
determined whether clinical psychologists actually
involved themselves in scholarly activity. They
concluded that given the low level of support for
such activities in most clinical settings, the scien-
tist± practitioner model was doing as well as could
be expected. Likewise, Milne et al.'s (1990) survey of
73 psychologists, found a high reported level of
producing (undertaking and publishing), consum-
ing (reading), and utilizing (applying to practice)
research. As O'Donohue and Szymanski (1994)
point out, the growing number of review publica-
tions and advances in technology such as CD ROM
has made research much more accessible.
However, from the standpoint of clinical psychol-
ogy, a study by Agnew et al. (1995) suggests that
there is no room for complacency in terms of
research output. Agnew et al. compared the research
output of clinical psychologists and psychiatrists
working in and around part of London. Across the
whole sample of clinical psychologists and psychi-
atrists working in both academic and practice
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
78 C. G. Long and C. R. Hollin

settings, the psychiatrists significantly outerformed
the psychologists on five out of six indices of
research performance (there was no difference on
the other). In academic settings the psychologists
more than matched the psychiatrists, but outside of
academic settings the difference was even more
marked, with the psychiatrists doing better on all
markers of research output.
Of course, as Agnew et al. note, this is a relatively
small scale study in a particular geographical
region, and the actual research indices say more
about volume than quality of research. Nonetheless,
Agnew et al. pose an interesting point when
they suggest that clinical psychologists may have
inflated opinions about their research skills. Is
it possible that so many clinical psychologists
have moved away from the scientist± practitioner
modelÐ in the study by Agnew et al. almost half the
non-academic clinical psychologists had not pub-
lished in a peer review journalÐthat practitioner-
based clinical research is at risk of becoming an
endangered species?
Another approach to criticizing the scientist±
practitioner model questions the basic assumptions
most closely associated with conducting research to
suggest that research is misleading and irrelevant.
The list of criticisms below that fall under this
general heading is mainly extracted from Pilgrim
and Treacher (1992).
Research is Misleading and Irrelevant
Researchers are (Not) Impartial
The stereotype of the impartial researcher was
examined by Broad and Wade (1985) in their book
Betrayers of the Trust: Fraud & Deceit in Science. Broad
and Wade attacked the view of research scientists as
impartial seekers of truth, citing examples of
malpractice and fraudulence. Pilgrim and Treacher
(1992), admitting that Broad and Wade do not
discuss psychological research in detail, with the
exception of `Burt's fabrication of data in his papers'
(p. 65), applied this position to the scientist±
practitioner model in clinical psychology. (The
allusion to Burt's fraudulence is made despite
continued debate on the accuracy of such claims.)
Thus, the view is advanced that research in clinical
research cannot be trusted as it is methodologically
unsound and largely irrelevant; while, in a handful
of studies, some researchers are unwilling or unable
to share their data and can make mistakes. Thus
claims that the scientist±practitioner model is
successful (e.g. Strickland, 1983) are seen as politic-
ally motivated, being `a crucial part of the profes-
sion's (i.e. clinical psychology) rhetoric in
establishing its superiority vis-a
Á±vis other profes-
sions such as psychiatry, social work and nursing'
(Pilgrim and Treacher, 1992, p. 82).
Research is Not Value Free
The socially myopic scientist is perhaps one of the
great myths about the research community. It is
abundantly clear from diverse material such as
biographies and interviews that many scientists
struggle with the moral and ethical issues associ-
ated with their work. Indeed, Prigogine and
Stengers' (1984) book Order Out of Chaos, makes
the point that all measurements, experiments, and
observations are only truths within given situations
and contexts. In their view, with which most
contemporary researchers in the human sciences
would surely agree, researchers always remain
`at the mercy of triviality and poor judgement'
(p. 42). It is precisely for that reason that profes-
sional bodies have codes of ethics to guide research.
Similarly, many centres of applied research are
concerned to have ethics committees to oversee
their research programmes.
Empiricism Equals Positivism
In a discussion of tactics used to destroy and
discredit clinical research evidence, Andrews (1989)
notes that invoking the demon of positivism is a
common ploy. To label a body of research as
positivistic is, it seems, to negate any value that the
research might have. At a more rational level,
there are problems simply in coming to an under-
standing of the term `positivism'. In a critique,
Halfpenny (1982) was able to identify no fewer than
12 positivisms. To equate empirical research with
positivism is, it follows, rather meaningless without
specifying what positivism is implicated. However,
the real issue is not one of philosophical exactness,
but a knee-jerk reaction by those bent on
knowledge destruction. As Halfpenny notes, anti-
positivists, `use the term loosely and indiscrim-
inately to describe all sorts of disfavoured forms of
inquiry' (p. 11).
Research Relies on Statistics
It is true that some research does use statistics, but
it is not clear why this is such a bad thing. Further,
many researchers use a range of approaches to
understanding their data, including qualitative as
well as quantitative methods. In recent years, for
example, the usefulness of single case study designs
and small ndesigns in bridging the scientist±
practitioner gap has been highlighted (e.g. Long
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
Scientist± Practitioner 79

and Hollin, 1995), along with sophisticated qualita-
tive research methods such as Grounded Theory
(Strauss and Corbin, 1990). This is not to say that
statistical tests have always been used correctly, as
shown by Dar et al. (1994), this is patently not
the case. However, as Dar et al. point out, this
simply means that researchers need to think care-
fully about their style and methods of statistical
analysis. Inappropriate use of statistical tests in
some research is not an argument for abandoning
statistically informed evaluation of therapeutic
effects.
Clinical Researchers are Behaviourists
The same argument applies here as for posi-
tivism: what type of behaviourism is implied?
While it may be the case that much clinical research
is carried out by those who espouse some form of a
behavioural stance, it is nonsense to imply that this
is true for all clinical research.
ALTERNATIVES TO THE SCIENTIST±
PRACTITIONER MODEL
Pilgrim and Treacher (1992) discuss three alterna-
tives to the scientist±practitioner model for clinical
psychology: the psychodynamic model, eclecticism,
and the concept of the reflective practitioner (Schon,
1983). However, with the possible exception of the
reflective practitioner, these are theoretical models
for the practice of clinical psychology, not philoso-
phies than underpin the discipline. For example,
one can embrace the principles of psychodynamic
theory to inform practice and be committed (or not)
to a scientist± practitioner stance. Once theory and
philosophy are conceptually disentangled, it is self-
evident that there are three philosophies: the
scientist model, the practitioner model, and the
scientist± practitioner.
The Scientist Model
The clinical psychologist whose practice followed a
pure scientific model would, presumably, adhere
strictly to a set of theoretical principles, and would
only use techniques of empirically proven effective-
ness in the given circumstances. Eventually the
accumulation of empirical knowledge would lead to
theoretical modification and change, in turn leading
to the development of new techniques to inform
clinical practice. Of course, it is difficult to see how
this pure model could ever be realized in practice.
Given current knowledge, clinical psychologists will
always be in a position where they will need to
innovate and work on accepted best practice rather
than on scientifically proven methods.
The Practitioner Model
In a practitioner model, the clinical psychologist
learns about practice from other practitioners,
typically their teachers. Barlow et al. (1984) make
the point that: `Those early practitioners who could
communicate their theories clearly, or who pre-
sented new and exciting principles, tended to attract
many followers. If those same people wrote
prolifically and/or gave numerous workshops, then
their procedures would be far more likely to be
adopted through sheer exposure than would
procedures of those who gave fewer workshops or
wrote, very little about what they were doing'
(p. 34).
Thus practice is more likely to be influenced by a
charismatic leader than by mundane research. It
seems likely that once absorbed the exciting new
principles would be applied by practitioners on a
piecemeal, trial and error basis. (There is also the
phenomenon of individual practitioners becoming
their own charismatic leaders, devising their own
principles and style of practice that `works for
them'.) It is sobering to realize, as Barlow et al.
(1984) point out, that cognitive therapy and biofeed-
back were developed, popularized, and widely
adopted before compelling research data on their
effectiveness had been gathered. While there is now
a respectable database of knowledge on those two
methods, there are doubtless many other therapies
in widespread use for which there is no research
evidence of effectiveness. However, if the words of
the charismatic leader and one's own experience
and intuition are the benchmarks, data matter little.
As O'Donohue and Szymanski (1994) note, clinical
intuition and personal observation and judgement
does not stand up terribly well, involving a variety
of errors that can lead to harm for clients and the
profession of clinical psychology.
The Scientist± Practitioner
It is evident that within clinical psychology empir-
ical research can embrace a range of types and
forms of enquiry, have a basis in several theoretical
approaches, and use a multitude of methods and
styles of analysis. The clinical psychologist as a
scientist± practitioner will seek to use the findings of
such research to augment and inform their practice.
The scientist± practitioner therefore strives for a
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
80 C. G. Long and C. R. Hollin

symbiosis between research and their own experi-
ence. The model of the reflective practitioner
actually informs the latter half of this process in
suggesting ways to structure and use constructively
one's own experience of practice.
CONCLUDING REMARKS
It is not difficult to mount an argument that
dual training in practice and research skill lends
a valuable quality to the profession of clinical
psychology. Edwards (1987) notes that since
problem solving is a central process for both
researchers and practitioners, it follows that the
roles of scientist and practitioner are compatible.
The ability to consider intelligently both clinical and
research issues is arguably mutually beneficial to
the researcher and the clinician in the performance
of their respective functions. Indeed, practice brings
to research the insight, judgment, and experience
that comes from daily interaction with difficult
practical issues and problems (Schopler, 1987). The
problem with the gap between professional know-
ledge and the demands of real world professional
practice, highlighted by Schon (1983) and Rein and
White (1981), is potentially capable of resolution
within the scientist± practitioner framework. Simon
(1972) argues that all professional practice is
centrally concerned with `design', i.e. the process
of `changing existing situations into preferred ones'
(p. 56).
In his presidential address to the British Psycho-
logical Society's Annual Conference, Watts (1992)
commented on the perennial danger of psychology
`falling apart' and fragmenting into an academic
discipline concerned with basic research that had
little connection with a profession with valuable
applied skills. In describing the obstacles that the
practitioner has in applying science to practice,
Phillips (1989) points out that while it is important
to strengthen the interface between science and
practice, the practical application of science is not
generally well understood. There is probably some
force to the argument that researchers do not
always consider an applied perspective. Rush
(1994), for example, notes the need to make user-
friendly to practitioners the vast amount of well-
established information on the treatment of depres-
sion. In a thought provoking article on the obstacles
to disseminating applied psychological science,
Beutler et al. (1993) report survey findings suggest-
ing that while clinicians believe research findings
are and have been important in modifying their
practice, they receive `research' information not
from research journals but from workshops,
popular books and practice-oriented journals.
The implication of this is that scientists need to
modulate their findings through vehicles of com-
munication valued by practitioners. The importance
of this is highlighted by Burnham's (1987) persuas-
ive argument that the popularization of science by
journalists in fragmented bits and pieces is one of
the major reasons for the widespread acceptance of
superstitious beliefs. Indeed, a major problem in
applying science to practice, is overcoming attitudes
that tend to justify and reinforce isolation of the
scientific community on the one hand, and the
practitioner community on the other. Evaluation is
long overdue of the variety of ways of integrating
the work of the scientist and the practitioner for
their mutual benefit (Heppner and Anderson, 1985;
Garmezy and Masten, 1986; Beutler et al., 1993).
Adoption of the scientist± practitioner model
means that development in clinical psychology is
guided by both practice and empirical research.
However, to make this statement does not mean
a blind acceptance of stereotypes of empirical
research. It is a matter of debate, even empirical
debate, whether a dual training in clinical and
research skills produces better or worse clinical
psychologists. Further, it is necessary to be clear on
the criteria by which to judge the success or failure
of a model. While there are several criteria that
might be used, it is surely the effectiveness in
practice of clinical psychologists working to differ-
ent models that should provide the benchmark of
success or failure. The fundamental research to
inform the choice of optimum model of practice for
clinical psychology is waiting to be carried out.
REFERENCES
Abrahamson, D. J. and Pearlman, L. A. (1993). The need
for scientist-practitioner employment settings. American
Psychologist,48, 59± 60.
Agnew, S., Carson, J. and Dankert, A. (1995). The
research productivity of clinical psychologists and
psychiatrists: A comparative study. Clinical Psychology
Forum,86, 2± 5.
Agras, W. S. and Berkowitz, C. (1980). Clinical research in
behavior therapy: Halfway there? Behavior Therapy,11,
472± 487.
Albee, G. W. (1970). The uncertain future of clinical
psychology. American Psychologist,25, 1071±1080.
American Psychological Association, Committee on
Trainee and Clinical Psychology (1947). Recommended
graduate training programme in clinical psychology.
American Psychologist,2, 539± 558.
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
Scientist± Practitioner 81

Andrews, D. A. (1989). Recidivism is predictable and can
be influenced: Using risk assessments to reduce
recidivism. Forum on Corrections Research,1, 11± 18.
Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J.,
Gendreau, P. and Cullen, F. T. (1990). Does correctional
treatment work? A clinically relevant and psycho-
logically informed meta-analysis. Criminology,8, 369±
404.
Barlow, D. H. (1981). On the relation of clinical research to
clinical practice: Current issues, new directions. Journal
of Consulting and Clinical Psychology,49, 147± 155.
Barlow, D. H. and Hersen, M. (1984). Single Case
Experimental Designs, 2nd edn. New York: Pergamon
Press.
Barlow, D. H., Hayes, S. C. and Nelson, R. O. (1984). The
Scientist± Practitioner: Research and Accountability in
Clinical and Educational Settings. New York: Pergamon.
Barrom, C. P., Shadish, W. R. and Montgomery, L. M.
(1988). PhD's, PsyD's and real world constraints
on scholarly activity: Another look at the Boulder
model. Professional Psychology: Research and Practice,19,
93± 101.
Belbin, E. (1979). Applicable psychology and some
national problems. Bulletin of the British Psychological
Society,32, 241± 244.
Belar, C. D. and Perry, N. W. (1992). National conference
on scientist± practitioner education and training
for professional practice of psychology. American Psy-
chologist,47, 71± 75.
Ben-Arie, O., Coch, A., Welman, M. and Teggin, A. F.
(1990). The effect of research on readmission to a
psychiatric hospital. British Journal of Psychiatry,156,
37± 39.
Bergin, A. and Strupp, H. (1972). Changing Frontiers
in the Science of Psychotherapy. Chicago IL: Aldine-
Atherton.
Beutler, L. E., Williams, R. E. and Wakefield, P. J. (1993).
Obstacles to disseminating applied psychological
science. Applied & Preventive Psychology,2, 53± 58.
Bornstein, P. H. and Wollersheim, J. P. (1978). Scientist±
practitioner activities among psychologists' behavioral
and non-behavioral orientations. Professional Psychology,
9, 659± 664.
Braff, D. L., Bachman, J. and Glick, I. D. (1979). The
therapeutic community as a research world. Myths and
facts. Archives of General Psychiatry,36, 355±360.
Broad, W. and Wade, N. (1985). Betrayers of the Truth:
Fraud and Deceit in Science. Oxford: Oxford University
Press.
Burnham, J. C. (1987). How Superstition Works and Science
Lost: Popularising Science and Health in the United States.
New Brunswick, NJ: Rutgers University Press.
Carr, A. (1990). Doctoral degrees in clinical psychology.
Clinical Psychology Forum,25, 35± 37.
Carroll, R. S., Miller, A., Ross, B. and Simpson, G. M.
(1980). Research as an impetus to improved treatment.
Archives of General Psychiatry,37, 377± 380.
Chassan, J. B. (1967). Research Design in Clinical: Psychology
and Psychiatry. New York: Appleton-Century-Crofts.
Cohen, L. H. (1977). Factors affecting the utilisation
of mental health evaluation and research findings.
Professional Psychology,8, 526±534.
Cohen, L. H. (1979). The research readership and
information source reliance of clinical psychologists.
Professional Psychology,10, 780±785.
Dar, R., Serlin, R. C. and Omer, H. (1994). Misuse of
statistical tests in three decades of psychotherapy
research. Journal of Consulting and Clinical Psychology,
62, 75±82.
DeAngelis, T. (1987). Therapists who feel as if they're not
therapists: The imposter syndrome. American Psycholo-
gical Society Monitor,18, 14.
Edwards, R. (1987). Implementing the scientist±
practitioner model: The school psychologist as data-
based problem solver. Professional School Psychologist,
2, 155± 161.
Eysenck, H. J. (1965). The effects of psychotherapy.
International Journal of Psychiatry,1, 97±178.
Frank, G. (1984). The Boulder model: History, rationale
and critique. Professional Psychology, Research and
Practice,15, 417± 435.
Gambril, E. (1990). Critical Thinking in Clinical Practice.
San Francisco CA: Jossey-Bass.
Garfield, S. L. (1986). Some comments on a revolution in
the training of professional psychologists. American
Psychologist,41, 1175± 1176.
Garfield, S. L. and Kurtz, R. (1976). Clinical psychologists
in the 1970's. American Psychologist,32, 1±9.
Garmezy, N. and Masten, A. S. (1986). Stress, com-
petence and resilience: Common frontiers for
therapist and psychopathologist. Behavior Therapy,17,
500± 521.
Halfpenny, P. (1982). Positivism and Sociology: Explaining
Social Life. London: Allen & Unwin.
Heppner, P. P. and Anderson, W. P. (1985). On the
perceived non-utility of research in counselling. Journal
of Counselling and Development,63, 539± 544.
Hollin, C. R. (1993). Advances in the psychological
treatment of delinquent behaviour. Criminal Behaviour
and Mental Health,3, 142± 157.
Hollin, C. R. (1995). The meaning and implications of
`programme integrity'. In: J. McGuire (Ed), What Works:
Reducing Reoffending. Chichester: Wiley.
James, L., Milne, D. L. and Firth, H. (1990). A systematic
comparison of feedback and staff discussion in chang-
ing the ward atmosphere. Journal of Advanced Nursing,
5, 329± 336.
Kelly, E. L., Goldberg, L. R., Fiske, D. W. and Kilkowski,
J. M. (1978). 25 years later: A follow-up study of the
graduate students in clinical psychology assessed at the
VA selection research project. American Psychologist,33,
746± 755.
Lavender, A. (1985). Quality of care and staff practices in
long stay settings. In F. N. Watts (Ed), New Develop-
ments in Clinical Psychology. Leicester: The British
Psychological Society.
Leigh, V. (1975). Attitudes and fantasy themes of patients
on a psychiatric research unit. Archives of General
Psychiatry,32, 598±601.
Lipsey, M. W. (1992). Juvenile delinquency treat-
ment: A meta-analytic inquiry into the variability
of effects. In T. D. Cook et al. (Eds), Meta-Analysis
for Explanation: A Casebook. New York: Russell Sage
Foundation.
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
82 C. G. Long and C. R. Hollin

Long, C. and Hollin, C. R. (1995). Single case design:
A critique of methodology and analysis of recent trends.
Clinical Psychology and Psychotherapy,2, 177± 191.
Lo
Èsel, F. (1995). Increasing consensus in the evaluation of
offender rehabilitation? Lessons from recent research
syntheses. Psychology, Crime, & Law,2, 19± 39.
Lo
Èsel, F. and Ko
Èferl, P. (1989). Evaluation research on
correctional treatment in West Germany: A meta-
analysis In: F. Lo
Èsel and J. Hiasan (Eds), Criminal
Behavior and the Justice System: Psychological Perspectives.
New York: Springer-Verlag.
McGuire, J. (Ed). (1995). What Works: Reducing Reoffending.
Chichester: Wiley.
Medawar, P. B. (1984). Plato's Republic. Oxford: Oxford
University Press.
Meehl, P. (1971). A scientific, non-researched doctorate for
clinical practitioners: Arguments pro and con. In: R. R.
Holt (Ed), New Horizon for Psychotherapy. New York:
International University Press.
Milne, D., Britton, P. and Wilkinson, I. (1990). The
scientist± practitioner in practice. Clinical Psychology
Forum,30, 22± 30.
O'Donohue, W. and Szymanski, J. (1994). How to win
friends and not influence clients: Popular but problem-
atic ideas that impair treatment decisions. The Behavior
Therapist,7, 29± 33.
Omer, H. and Dar, R. (1992). Changing trends in three
decades of psychotherapy research: The flight from
theory into pragmatics. Journal of Consulting and Clinical
Psychology,60, 88± 93.
O'Sullivan, J. J. and Quevillon, R. P. (1992). Forty years
later: Is the Boulder model still alive? American
Psychologist,47, 67±70.
Palmer, T. (1992). The Re-emergence of Correctional, Inter-
vention. Newbury Park, CA: Sage Publications.
Paul, G. L. and Lentz, R. J. (1977). Psychological Treatment
of Chronic Mental Patients. Cambridge, MA: Harvard
University Press.
Peck, D. F. (1985). Small Nexperimental designs in clinical
practice. In: F. N. Watts (Ed), New Developments in
Clinical Psychology. Leicester: The British Psychological
Society.
Peterson, D. R. (1976). Need for the doctor of psychology
degree in professional psychology. American Psycholo-
gist,31, 792± 798.
Peterson, D. R. (1985). 20 years of practitioner training in
psychology. American Psychologist,40, 441±451.
Phillips, B. N. (1989). Role of the practitioner in applying
science to practice. Professional Psychology: Research and
Practice,20, 3±8.
Pilgrim, D. and Treacher, A. (1992). Clinical Psychology
Observed. London: Tavistock/Routledge.
Prigogine, I. and Stengers, I. (1984). Order Out of Chaos.
New York: Bantam Books.
Prochaska, J. O. and Norcross, J. C. (1983). Contemporary
psychotherapists: A national survey of characteristics,
practices, orientations and attitudes. Psychotherapy:
Theory, Research & Practice,20, 161± 173.
Prokasky, W. F. (1986). Preparation of regulation of
professional psychologists. American Psychologist,41,
1176± 1177.
Raimy, V. C. (Ed) (1950). Training in Clinical Psychology,
New York: Prentice-Hall.
Rashkis, H. A. (1960). Cognitive restructuring: Why
research is therapy. Archives of General Psychiatry,2,
612± 621.
Rein, M. and White, S. (1981). Knowledge for practice.
Social Science Review,55, 1±41.
Rush, A. J. (1994). Research and clinical practice: Bridging
the gap. Current Opinion in Psychiatry,7, 3±4.
Sacks, M., Carpenter, W. T. and Richmond, M. B. (1975).
Psychotherapy in hospitalised research patients. Arch-
ives of General Psychiatry,32, 581± 585.
Salkovskis, P. (1984). Psychological research by NHS
clinical psychologists: An analysis and some
suggestions. Bulletin of the British Psychological Society,
37, 375±377.
Schon, D. (1983). The Reflective PractitionerÐHow Profes-
sionals Think in Action. New York: Basic Books.
Schopler, E. (1987). Specific and non-specific factors in the
effectiveness of the treatment system. American Psychol-
ogist,42, 376± 383.
Shakow, D. (1978). Clinical psychology seen some 50 years
later. American Psychologist,33, 148±158.
Simon, H. (1972). The Sciences of the Artificial. Cambridge,
MA: MIT Press.
Strauss, A. and Corbin, J. (1990). Basics of Qualitative
Research: Grounded Theory Procedures and Techniques.
Thousand Oaks, CA: Sage.
Strickland, B. (1983). Responding to current concerns
in clinical psychology. The Clinical Psychologist,37,
25± 27.
Thelen, M. H. and Ewing, D. R. (1970). Roles, functions
and training in clinical psychology: A survey of
academic clinicians. American Psychologist,25, 550±554.
Watts, F. N. (1984). Applicable psychological research in
the NHS. Bulletin of the British Psychological Society,37,
41± 42.
Watts, F. N. (1992). Is psychology falling apart? The
Psychologist: Bulletin of the British Psychological Society,5,
489± 494.
Wilson, G. T. (1981). Some thoughts about clinical
research. Behavioral Assessment,3, 217±225.
#1997 by John Wiley & Sons, Ltd. Clinical Psychology and Psychotherapy, Vol. 4, 75±83 (1997)
Scientist± Practitioner 83