Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/338118465
What do clinical supervision research reviews tell us? Surveying the last 25
years
Article in Counselling and Psychotherapy Research · December 2019
DOI: 10.1002/capr.12287
CITATIONS
0READS
260
1 author:
Clifton Edward Watkins, Jr.
University of North Texas
130 PUBLICATIONS 1,690 CITATIONS
SEE PROFILE
All content following this page was uploaded by Clifton Edward Watkins, Jr. on 23 December 2019.
The user has requested enhancement of the downloaded file.
Couns Psychother Res. 2019;00:1–19. wileyonlinelibrary.com/journal/capr | 1© 2019 British Association for Counselling and Psychotherapy
1 | INTRODUC TION
Supervision research has been around for over 60 years now, and its importance for the field's advancement has been increasingly rec-ognised (Inman et al., 2014). As supervision studies have continued to accumulate and supervision has acquired greater empirical mass, supervision research reviews have become ever more common fare. Reviews of supervision studies have long been valued for providing a critical perspective on the available research, identifying (a) what can be learned to guide practice, (b) what obstacles emerge as lim-itations across investigations and (c) what areas are in need of at-tention going forward (cf. Mulrow, 1994; Munn, Peters, et al., 2018). Such reviews serve as barometers of progress, ideally reflecting in-creasing methodological sophistication and data base solidification. But is that the case for clinical supervision? Do supervision research reviews reflect such ‘progress’? Those questions are subsequently examined.
2 | RESE ARCH IN CLINIC AL SUPERVISION: A SHORT STATUS REPORT
Five broad areas of supervision research study are as follows: (a) su-pervision effects on client outcomes; (b) supervision effects on the supervisor–supervisee interaction; (c) supervision's direct effects on supervisee competence; (d) factors that mediate and moderate super-visor impact on supervisee competence; and (e) supervisor and super-visee characteristics (Bernard & Goodyear, 2014). Most research has focused on mediating and moderating factors (e.g. relationship varia-bles, developmental considerations; Bernard & Goodyear, 2014), what Ladany and Malouf (2010) refer to as ‘inside supervision’ matters. Data support the supervisor–supervisee relationship, perhaps the most sub-stantial ‘inside’ matter, as being integral to fostering supervisee change (Carifio & Hess, 1987; Ellis, 2010; Park, Ha, Lee, Lee, & Lee, 2019). Data further suggest that supervision positively impacts supervisees, result-ing in such gains as enhanced self-awareness, enhanced self-efficacy
Received: 13 September 2019 | Revised: 2 December 2019 | Accepted: 3 December 2019
DOI: 10.1002/capr.12287
R E V I E W A R T I C L E
What do clinical supervision research reviews tell us? Surveying the last 25 years
C. Edward Watkins Jr.
Department of Psychology, University of North Texas, Denton, Texas
CorrespondenceC. Edward Watkins, Jr., Department of Psychology, University of North Texas, 76203, Denton, TX, USA.Email: [email protected]
AbstractWhat do clinical supervision research reviews across the last 25 years tell us? That question is subsequently examined. Based on database and literature searches, 20 reviews appearing from 1995 through 2019 were identified for survey examination; consistencies, inconsistencies and other defining features were determined across reviews; and the survey findings and their implications are considered. Primary find-ings are as follows: (a) ‘proof’ for supervision appears to be more ‘proof by asso-ciation’ than otherwise, being primarily a product of ex post facto, cross-sectional, correlational study; (b) evidence supporting supervision impact of any type is weak at best, especially so for worker and client outcomes; (c) supervision models gener-ally lack empirical foundation; (d) evidence-based supervision appears to be more a hope and dream than supervision-based reality at present; and (e) the primary meth-odological problems that plagued supervision research in the 1990s are still all too frequent in modern research. Some questions to entertain about supervision going forward, and some remedies for improving its research, are proposed.
K E Y W O R D S
clinical, counselling, psychotherapy, research, review, supervision
2 | WATKINS
and enhanced skill acquisition (Goodyear & Guzzardo, 2000; Inman et al., 2014; Wheeler & Richards, 2007). But research supporting super-vision's impact on skill transfer remains limited, as does research ad-dressing supervision's impact on client outcomes.
The current state of supervision research can be compared to psy-chotherapy research in the 1950s or 1960s: measurement and effec-tiveness issues loom large as major concerns (Milne et al., 2012). It remains the case that ‘…although our knowledge and understanding of supervision has bourgeoned…, that which we do not understand or understand well continues to be vast’ (Inman et al., 2014, p. 86). Hampering that understanding, supervision research has often been, and continues to be, criticised for the following reasons: small sample sizes, over-reliance on self-report measures, limited number of valid supervision measures, ex post facto designs, limited attention to client outcomes and lack of longitudinal data (Ellis, D’Iuso, & Ladany, 2008; Ellis & Ladany, 1997; Hill & Knox, 2013; Russell, Crimmings, & Lent, 1984). As Milne et al. (2012) have stated, ‘… we are currently about “half-way there”, working on the “search for scientific rigour”...’ (p. 144).
3 | SURVE YING SUPERVISION RESE ARCH RE VIE WS: R ATIONALE , FOCUS AND APPROACH
3.1 | Rationale
In continuing that search, I wondered this: Might a survey of super-vision research reviews add to our current understanding about su-pervision's impact, how previously identified supervision research needs are being addressed and how supervision study has changed over time? Furthermore, if ‘…the evidence of supervisor impact has been well established’ (Bernard & Goodyear, 2014, p. 301), how might that evidence be on display across reviews?
Survey is defined here as a ‘detailed critical inspection and com-prehensive considering’ (see WordWeb; https ://wordw eb.info/). To my knowledge, no survey of supervision research reviews has been conducted. It could be informative to assemble review findings, so as to potentially reinforce and bolster current perspective, identify new, promising findings that might not have been properly recognised, or indicate areas that continue to show lack of progress and sorely need attention. Moreover, because a spike in supervision research reviews has occurred across these last five years alone (e.g. Alfonsson, Parling, Spännargård, Andersson, & Lundgren, 2018; Barrett, Gonsalvez, & Shires, 2019; Forshaw, Sabin-Farrell, & Schröder, 2019; Kühne, Maas, Wiesenthal, & Weck, 2019; Newman, Simon, & Swerdlik, 2019), the time for such a survey would seem especially apropos.
3.2 | Focus
The last 25 years, 1995 through 2019, was selected as the review period. That period of time was chosen for three reasons: (a) it covers the most recent generation of supervision scholarship;
(b) research is reported to have increased in both quantity and quality during that time period (Inman et al., 2014); and (c) that particular period captures a time of significant transformation in supervision's history, where the shift to a ‘competency-based’ and ‘evidence-based’ perspective has been increasingly embraced and become defining (e.g. Falender & Shafranske, 2017; O'Donoghue, Ju, & Tsui, 2018).
Focus was given to what might broadly be thought of as social (as opposed to medical) services, the specific provision of coun-selling, psychotherapy or emotional support being an important part of professional role functioning. Social work, counselling, psy-chology, psychiatric nursing and psychiatry were all included for study. However, reviews were excluded that combined both social and medical services (Cutcliffe, Sloan, & Bashaw, 2018; Dawson, Phillips, & Leggat, 2013; Ducat & Kumar, 2015; Farnan et al., 2012), focused exclusively on medical services (Snowdon, Leggat, & Taylor, 2017), relied primarily on a noncounselling/psychother-apy database (e.g. learning disabilities; Milne & James, 2000) or were primarily scoping (e.g. Goodyear, Bunch, & Claiborn, 2006; Sewell, 2018) or summary (MacDonald & Ellis, 2012; Weerasekera, 2013) in nature. Focus was given to systematic research reviews, defined here as ‘a type of research synthesis…to identify and re-trieve international evidence that is relevant to a particular ques-tion or questions and to appraise and synthesize the results of this search to inform practice, policy and in some cases, further re-search…’ (Munn, Stern, Aromataris, Lockwood, & Jordan, 2018, p. 144). Such reviews are conducted for a host of reasons: to confirm current practices, identify new practices, identify conflicting re-sults, identify areas for future research, uncover international ev-idence and produce statements to guide decision-making (Higgins & Thomas, 2019; Munn, Peters, et al., 2018). Journal articles, which met that definition and comported with those reasons, were included. Both in-print (already in bound journal form) and online (early view) review articles were allowed. Book chapters, with but two exceptions, were excluded. The decision was made to include the two chapters by Ellis and colleagues (Ellis et al., 2008; Ellis & Ladany, 1997) because they are (a) well-done, highly rigorous sys-tematic research reviews of supervision study methodology and (b) unique in the thoroughgoing attention and scrutiny given to methodological matters.
3.3 | Selection/Analysis approach
Four steps were taken to identify articles for examination: (a) da-tabase searches (e.g. Google Scholar, PsycINFO), using such search words as ‘supervision’, ‘clinical’, ‘research’ and ‘review’, were con-ducted for the 1995–2019 period; (b) reference sections of iden-tified articles were examined to identify reviews that might have been missed; (c) supervision journals or journals that publish some supervision material were examined for recent reviews; and (d) recent supervision texts (e.g. Bernard & Goodyear, 2019; Scaife, 2019) were also examined to further find possible missed work.
| 3WATKINS
Potential articles/chapters were identified and examined to deter-mine their being a systematic research review, a set of reviews was then identified for further examination, and each review was then read and studied so as to map its main features. Mapped features across the set of reviews included review focus, inclusion criteria, appraisal procedure, findings and limitations/issues. Other impor-tant features (e.g. attention to/omission of multicultural factors) were also examined and noted. Next, all reviews were examined for consistencies, inconsistencies and defining features, and all such facets were identified and recorded, the hope being that the resulting supervision information might confirm current practices, identify new practices, identify conflicting results, identify areas for future research and uncover international evidence (cf. Munn, Peters, et al., 2018).
Examination of reviews was approached in two ways. First, a broad view was taken, with consistencies, inconsistencies and de-fining features being determined across the entire body of identi-fied reviews. Second, the body of reviews was divided into two groupings of 12 years each: (a) the 1996–2007 years; and (b) the 2008–2019 years. No supervision reviews were identified as having appeared in 1995, so that year was excluded in order to make for two groupings equivalent in number of years covered. Those two periods were examined for any defining differences.
4 | RESULTS
A total of 20 clinical supervision research reviews were identified for study. Table 1 provides the mapped features of each review.
4.1 | What do these reviews tell us? The broad view
4.1.1 | Focus of reviews
The reviews were varied in focus and can be roughly grouped as follows: general reviews, where ‘the whole’ of supervision was ex-amined (Bogo & McKnight, 2006; Borders, 2006; Buus & Gonge, 2009; Newman et al., 2019; O’Donoghue & Tsui, 2015; Tsui, 1997); reviews primarily focused on supervision effects on supervisees and/or clients (Alfonsson et al., 2018; Carpenter et al., 2013; Freitas, 2002; Kühne et al., 2019; Reiser & Milne, 2014; Watkins, 2011; Wheeler & Richards, 2007); single reviews about evidence-based practice (Barrett et al., 2019), supervision models (Simpson-Southward, Waller, & Hardy, 2017), the supervisee's experience of receiving supervision (Wilson, Davies, & Weatherhead, 2016) and the supervisor's experience of providing supervision (Forshaw et al., 2019); and reviews focused on methodology and rigour (Ellis et al., 2008; Ellis & Ladany, 1997; Ellis et al., 1996).
Quantitative, qualitative, single-case or otherVirtually all reviews focused exclusively on quantitative stud-ies. However, two reviews were exclusively qualitative in nature
(Forshaw et al., 2019; Wilson et al., 2016), and one was a content analysis (Simpson-Southward et al., 2017).
The reviewers reviewingThe reviews cut across several mental health disciplines, being provided by representatives from social work, counsellor educa-tion, nursing and psychology. Countries from which the reviews originated were Australia, Canada, China, Denmark, Germany, New Zealand, Sweden, the UK and the United States.
Journals in which reviews appeared (excluding two book chapters)Reviews appeared in a host of varied journals: BMC Psychology, the British Journal of Social Work, Children and Youth Services Review, Clinical Psychologist, Clinical Psychology and Psychotherapy, Clinical Supervisor, Cognitive Behaviour Therapy, Counselling and Psychotherapy Research, International Journal of Mental Health Nursing, Journal of Social Service Research, Mental Health Review Journal, Psychology in the Schools, Psychotherapy and Training and Education in Professional Psychology.
4.1.2 | Inclusion criteria
Two inclusion criteria were consistently applied across most reviews: that the selected articles have been peer-reviewed and be in English. Where not explicitly stated, those criteria still seemed to be in play. Other criteria typically were reflective of any given review's particu-lar focus (e.g. that a measure of outcome be included where outcome was the primary concern).
4.1.3 | Appraisal procedure/criteria
Thirteen of 20 reviews involved the use of some type of objective ap-praisal procedure, with some type of rating of study quality typically being made. Different rating tools mentioned included the Critical Appraisal Skills Programme, Weight of Evidence approach and re-searcher-created evaluation checklists. PRISMA reporting guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) were applied in at least five reviews, and in three cases, a research protocol was produced and registered (with PROSPERO; International Prospective Register of Systematic Reviews) before the review's initiation. With PRISMA (Moher, Liberati, Tetzlaff, Altman, & the PRISMA Group, 2009) and PROSPERO (Page, Shamseer, & Trico, 2018) both being products of the last approximate decade, their use across supervision reviews is understandably a most recent phenomenon.
4.1.4 | Features of reviewed publications
The number of reviewed articles across reviews ranged from a low of four (Barrett et al., 2019) to a high of 131 (Ellis et al., 1996), with about
4 | WATKINS
TAB
LE 1
Su
perv
isio
n re
view
s ac
ross
the
last
25
year
s
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Alfo
nsso
n et
al
. (20
18)
Effe
cts
of s
uper
vi-
sion
on
supe
rvi-
sees
and
pat
ient
s
a. T
hat s
uper
visi
on fo
cuse
d on
ong
oing
cas
esb.
Tha
t tre
atm
ent c
ondu
cted
w
ithin
bro
ad c
ogni
tive-
beha
viou
ral f
ram
ewor
kc.
Tha
t exp
erim
enta
l, qu
asi-
expe
rimen
tal o
r sin
gle-
case
de
sign
use
dd.
Tha
t qua
ntifi
able
out
com
e m
easu
re o
f sup
ervi
see
and/
or p
atie
nt e
ffec
ts u
sed
e. I
n En
glis
h la
ngua
ge
Jada
d sc
orin
g sy
stem
us
ed a
nd C
ON
SORT
gu
idel
ines
follo
wed
Revi
ew p
roce
ss fo
l-lo
wed
gui
delin
es in
Co
chra
ne H
andb
ook
of
inte
rven
tions
and
by
PRIS
MA
Revi
ew p
roto
col r
eg-
iste
red
pros
pect
ivel
y w
ith P
ROSP
ERO
and
pu
blis
hed
5 ar
ticle
s re
view
ed (o
ut o
f 4,
103
initi
al h
its)
Year
s co
vere
d:
2006
–201
64
stud
ies
used
ran-
dom
ised
con
trol
led
desi
gn, a
nd o
ne u
sed
nonr
ando
mis
ed c
on-
trol
led
desi
gn2
stud
ies
cond
ucte
d in
th
e U
nite
d St
ates
, 1 in
A
ustr
alia
, 1 in
Ger
man
y an
d 1
in th
e U
K; U
K st
udy
invo
lved
sup
er-
vise
es fr
om R
ussi
a an
d U
krai
ne
a. M
ost f
requ
ent m
etho
dolo
gica
l sho
rtco
m-
ings
: fai
lure
to re
port
pow
er a
naly
ses,
in
adeq
uate
repo
rtin
g of
rand
omis
atio
n pr
oced
ures
and
inad
equa
te re
port
ing
of
post
hoc
ana
lyse
sb.
Sup
ervi
sion
may
impa
ct s
uper
vise
e co
mpe
tenc
ec.
Any
sup
ervi
sion
impa
ct o
n pa
tient
s is
w
eak
at b
est
d. N
o su
ppor
t pro
vide
d fo
r any
par
ticul
ar
supe
rvis
ion
mod
ele.
Nee
d to
exp
lore
sup
ervi
sion
pro
cess
, a
negl
ecte
d va
riabl
e
a. N
umbe
r of s
uper
viso
rs a
nd
patie
nts
invo
lved
unr
epor
ted
in
two
stud
ies
b. D
urat
ion
of s
uper
visi
on u
n-sp
ecifi
ed in
one
stu
dyc.
Fre
quen
cy o
f sup
ervi
sion
un
spec
ified
in o
ne s
tudy
d. S
uper
visi
on in
terv
entio
n in
on
e st
udy
cons
iste
d of
a s
ingl
e 30
-min
sup
ervi
sion
ses
sion
per
m
onth
for t
hree
mon
ths
in o
ne
stud
y
Barr
ett e
t al.
(201
9)To
revi
ew e
vide
nce
for s
cien
tist-
prac
titio
ner
and
evid
ence
-ba
sed
prac
tice
(EBP
) in
clin
ical
su
perv
isio
n
a. T
hat s
uper
visi
on b
e pr
ovid
ed b
y ac
cred
ited
supe
rvis
ors
b. E
vide
nce-
base
d pr
actic
e an
d/or
sci
entis
t-pr
actit
ione
r co
mpe
tenc
e as
an
outc
ome
c. P
ublic
atio
n in
pee
r-re
-vi
ewed
jour
nal
App
rais
al to
ol fo
r Cro
ss-
Sect
iona
l Stu
dies
Revi
ew p
roce
ss
follo
wed
PRI
SMA
gu
idel
ines
Revi
ew p
roto
col
deve
lope
d in
acc
ord-
ance
with
Coc
hran
e H
andb
ook
of s
yste
m-
atic
revi
ews,
regi
ster
ed
pros
pect
ivel
y w
ith
PRO
SPER
O a
nd
publ
ishe
d
4 ar
ticle
s re
view
ed (o
ut o
f 1,
287
initi
al h
its)
Year
s co
vere
d:
2015
–201
72
stud
ies
cond
ucte
d in
A
ustr
alia
and
2 o
ther
s fr
om th
e U
nite
d St
ates
a. V
irtua
l abs
ence
of E
BP/s
uper
visi
on
rese
arch
b. S
uper
viso
rs e
ncou
rage
d to
use
evi
denc
e-ba
sed
supe
rvis
ory
activ
ities
(e.g
. vid
eo
feed
back
, mod
ellin
g)c.
Sup
ervi
sors
enc
oura
ged
to u
se c
ompe
-te
ncy
eval
uatio
n ra
ting
form
s in
mon
itor-
ing
supe
rvis
ee p
rogr
ess
Nar
row
incl
usio
n cr
iteria
Focu
s on
clin
ical
psy
chol
ogy
trai
nees
onl
y
Bogo
&
McK
nigh
t (2
006)
To e
xam
ine
soci
al
wor
k su
perv
isio
n re
sear
ch
a. T
hat p
ublic
atio
n be
pe
er-r
evie
wed
b. A
ppea
r dur
ing
desi
gnat
ed
ten-
year
revi
ew p
erio
d
Non
e sp
ecifi
ed13
art
icle
s re
view
edYe
ars
cove
red:
19
94–2
004
All
stud
ies
from
the
Uni
ted
Stat
es
a. D
eart
h of
em
piric
al s
tudi
es o
n so
cial
w
ork
supe
rvis
ion
acro
ss th
e pa
st d
ecad
eb.
Mos
t stu
dies
use
d sm
all,
conv
enie
nce
sam
ples
and
wer
e cr
oss-
sect
iona
l, si
ngle
pr
ojec
tsc.
Rac
e/et
hnic
ity s
tudi
ed in
onl
y tw
o pr
ojec
tsd.
No
evid
ence
that
sup
ervi
sion
aff
ects
w
orke
r or c
lient
out
com
ese.
Sup
ervi
sion
mod
els
rem
ain
unte
sted
f. Su
perv
isio
n's
evid
ence
bas
e em
bryo
nic
All
stud
ies
from
the
Uni
ted
Stat
esC
riter
ia fo
r art
icle
sel
ectio
n un
spec
ified
No
obje
ctiv
e ap
prai
sal p
roce
dure
us
edPr
oces
s of
arr
ivin
g at
con
clus
ions
un
spec
ified
(Con
tinue
s)
| 5WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Bord
ers
(200
6)To
revi
ew c
once
p-tu
al a
nd e
mpi
rical
su
perv
isio
n lit
-er
atur
e fr
om th
e fie
lds
of c
ouns
el-
ling
and
coun
sel-
lor e
duca
tion
a. T
hat p
ublic
atio
n ap
pear
in
Am
eric
an C
ouns
elin
g A
ssoc
iatio
n jo
urna
l or
one
of th
ree
inte
rnat
iona
l co
unse
lling
jour
nals
(In
tern
atio
nal J
ourn
al
for t
he A
dvan
cem
ent o
f Co
unse
lling
, Brit
ish Jo
urna
l of
Gui
danc
e an
d Co
unse
lling
an
d Ca
nadi
an Jo
urna
l of
Coun
selli
ng)
b. A
ppea
r dur
ing
desi
gnat
ed
five-
year
revi
ew p
erio
d
Non
e sp
ecifi
ed84
art
icle
s re
view
edYe
ars
cove
red:
late
199
9 to
ear
ly 2
005
a. T
he s
uper
viso
ry re
latio
nshi
p em
erge
d as
hi
ghly
impo
rtan
t, th
at b
eing
esp
ecia
lly s
o fo
r mul
ticul
tura
l sup
ervi
sion
b. M
ultic
ultu
ral s
uper
visi
on re
ceiv
ed
incr
ease
d at
tent
ion,
feed
back
was
gi
ven
min
imal
rese
arch
att
entio
n, u
se
of te
chno
logy
beg
an to
be
inve
stig
ated
, an
d sc
hool
cou
nsel
ling
rece
ived
the
mos
t at
tent
ion
acro
ss c
ouns
ellin
g sp
ecia
lties
c. F
ew s
tudi
es e
mpl
oyed
an
expe
rimen
tal
desi
gn, a
nd q
ualit
ativ
e st
udie
s w
ere
com
para
tivel
y m
ore
freq
uent
, with
eve
n a
few
mix
ed-m
etho
ds s
tudi
es a
lso
bein
g co
nduc
ted
d. W
hite
fem
ales
wer
e pr
epon
dera
nt a
s re
sear
ch s
tudy
par
ticip
ants
e. S
tudi
es o
ften
invo
lved
sm
all s
ampl
es,
wer
e ba
sed
in b
ut a
sin
gle
prog
ram
me
and
wer
e ex
pos
t fac
to in
des
ign
Aut
hor s
ole
eval
uato
rN
o ob
ject
ive
appr
aisa
l pro
cedu
re
used
Buus
&
Gon
ge
(200
9)
To s
umm
aris
e an
d cr
itica
lly e
valu
-at
e al
l em
piric
al
stud
ies
of c
linic
al
supe
rvis
ion
in p
sych
iatr
ic
nurs
ing
a. T
hat p
ublic
atio
n be
em
piri-
cal s
tudy
of c
linic
al s
uper
vi-
sion
in p
sych
iatr
ic n
ursi
ngb.
In
Engl
ish
lang
uage
Art
icle
s sy
stem
atic
ally
ex
amin
ed u
sing
spe
-ci
alis
ed c
heck
lists
from
CO
NSO
RT, S
TRO
BE
and
CORE
Q
34 a
rtic
les
revi
ewed
Year
s co
vere
d:
1990
–200
7D
iver
se re
pres
enta
tion
of
geog
raph
ical
con
trib
u-tio
n; S
wed
en p
rimar
y co
ntrib
utor
; oth
er
cont
ribut
ors
incl
uded
A
ustr
alia
, Fin
land
, the
U
K an
d th
e U
nite
d St
ates
a. C
linic
al s
uper
visi
on in
psy
chia
tric
nur
sing
co
mm
only
vie
wed
as
a go
od p
ract
ice,
but
em
piric
al e
vide
nce
supp
ortin
g th
is v
iew
is
lim
ited
b. S
tudi
es w
ere
ofte
n sm
all-s
cale
, did
not
co
ntro
l for
con
foun
ding
fact
ors
and
had
relia
bilit
y/va
lidity
issu
esc.
Una
ble
to c
onfid
ently
say
sup
ervi
sion
in
psyc
hiat
ric n
ursi
ng s
ettin
gs h
as d
esire
d ou
tcom
es o
n nu
rses
or t
heir
patie
nts
d. I
dent
ified
rese
arch
obs
tacl
es in
clud
e th
e fo
llow
ing:
gen
eral
lack
of c
onse
nsus
am
ong
rese
arch
ers
abou
t wha
t res
earc
h in
stru
men
ts to
use
; and
gen
eral
lack
of
con
sens
us a
bout
whi
ch d
efin
ition
s an
d m
odel
s sh
ould
gui
de s
uper
visi
on
rese
arch
. Fur
ther
mor
e, s
uch
varia
bles
as
shi
ftin
g w
orkl
oads
, str
essf
ul e
vent
s,
inte
rper
sona
l and
inte
rpro
fess
iona
l co
nflic
ts, a
nd p
atie
nts’
dege
nera
ting
or
vola
tile
cond
ition
s re
nder
rand
omis
atio
n of
par
ticip
ants
and
con
trol
ling
for c
on-
foun
ding
fact
ors
part
icul
arly
pro
blem
atic
Con
cept
ualis
atio
ns o
f nur
sing
pr
actic
es s
how
som
e va
riatio
n ar
ound
the
wor
ld, m
akin
g da
ta
com
paris
ons
diff
icul
t
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
6 | WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Car
pent
er,
Web
b, &
Bo
stoc
k (2
013)
To d
eter
min
e w
hat
is k
now
n ab
out
supe
rvis
ion
ef-
fect
iven
ess;
and
ap
prai
se e
vide
nce
base
for s
uper
vi-
sion
in c
hild
w
elfa
re s
ervi
ces
a. T
hat p
ublic
atio
n be
pe
er-r
evie
wed
b. T
hat s
tudy
be
quan
titat
ive
or q
ualit
ativ
e an
d in
clud
e so
cial
wor
kers
or o
ther
ch
ild w
elfa
re p
ract
ition
ers
c. I
n En
glis
h la
ngua
ge
Wei
ght o
f Evi
denc
e ap
-pr
oach
em
ploy
ed (r
at-
ings
pro
vide
d fo
r eac
h st
udy
on tr
ustw
orth
i-ne
ss, a
ppro
pria
tene
ss
of s
tudy
des
ign
and
topi
c re
leva
nce)
Revi
ew p
roce
ss fo
l-lo
wed
PRI
SMA
gui
de-
lines
and
met
hods
de
velo
ped
by S
ocia
l C
are
Inst
itute
for
Exce
llenc
e
22 a
rtic
les
revi
ewed
(out
of
1,5
90 in
itial
hits
)Ye
ars
cove
red:
20
00–2
012
All
stud
ies
from
the
Uni
ted
Stat
es
a. P
rimar
y st
udy
prob
lem
s in
clud
ed th
e fo
llow
ing:
no
rand
omis
ed c
ontr
olle
d tr
ials
or q
uasi
-exp
erim
enta
l inv
estig
a-tio
ns; s
mal
l sam
ple
size
s; a
nd in
adeq
uate
de
scrip
tions
b. T
he v
ast m
ajor
ity o
f stu
dies
wer
e cr
oss-
sect
iona
l and
cor
rela
tiona
l, w
ith o
nly
one
stud
y be
ing
inte
rven
tion
(pre
/pos
t) in
de
sign
c. S
uper
visi
on fo
und
to b
e po
sitiv
ely
asso
-ci
ated
with
job
satis
fact
ion,
job
rete
ntio
n an
d ab
ility
to m
anag
e w
orkl
oad
d. S
uper
visi
on w
orks
bes
t whe
n at
tend
ing
to th
e su
perv
isor
–sup
ervi
see
rela
tion-
ship
, pro
vidi
ng s
ocia
l and
em
otio
nal
supp
ort a
nd p
rovi
ding
task
ass
ista
nce
e. I
mpa
ct o
f sup
ervi
sion
mod
els
on o
ut-
com
e no
t stu
died
f. Ve
ry li
mite
d, w
eak
evid
ence
for s
uper
vi-
sion
impa
ct o
n w
orke
r out
com
esg.
Im
pact
of s
uper
visi
on o
n co
nsum
er
outc
omes
an
unst
udie
d is
sue
All
stud
ies
from
the
Uni
ted
Stat
esIn
clus
ion
crite
ria p
ossi
bly
too
broa
dW
eak
rese
arch
des
igns
per
haps
ad
mitt
edRe
stric
ted
time
fram
e
Ellis
et a
l. (2
008)
To p
rovi
de u
pdat
e of
ear
lier r
evie
w
(Elli
s &
Lad
any,
19
97),
givi
ng
spec
ific
focu
s to
sup
ervi
sion
m
easu
res
and
eval
uatin
g 37
va
lidity
thre
ats
a. T
hat a
rtic
le b
e ab
out c
lini-
cal s
uper
visi
onb.
Tha
t art
icle
be
data
-bas
edc.
Tha
t art
icle
focu
s on
sup
er-
visi
on m
easu
rem
ent
Thre
e ra
ters
trai
ned
for
five
hour
s in
ratin
g 37
th
reat
s90
% ra
ter a
gree
men
t re
ache
d be
fore
beg
in-
ning
stu
dy
6 ar
ticle
s re
view
edYe
ars
cove
red:
19
95–2
007
All
stud
ies
from
the
Uni
ted
Stat
es
a. N
ew s
uper
visi
on m
easu
res
cont
inue
to
be d
evel
oped
; som
e ev
iden
ce th
at m
ore
strin
gent
app
roac
h is
bei
ng ta
ken
to
mea
sure
dev
elop
men
tb.
Mos
t res
earc
h st
ill n
ot s
cien
tific
ally
rig-
orou
s, w
ith n
umer
ous
thre
ats
to v
alid
ity
bein
g on
dis
play
c. M
ore
soun
d co
nstr
uctio
n of
sup
ervi
sion
-sp
ecifi
c m
easu
res
nece
ssar
y fo
r fie
ld to
ad
vanc
e
The
sam
e as
Elli
s, L
adan
y, K
reng
el,
and
Schu
lt (1
996)
/ El
lis a
nd
Lada
ny (1
997)
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
| 7WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Ellis
and
La
dany
(1
997)
Repl
icat
ion
and
exte
nsio
n of
Elli
s et
al.
(199
6), c
on-
duct
ing
a m
ore
circ
umsc
ribed
m
etho
dolo
gica
l re
view
spe
cific
ally
ta
rget
ing
supe
r-vi
sees
and
clie
nts
and
eval
uatin
g 37
va
lidity
thre
ats
The
sam
e as
Elli
s et
al.
(199
6)Th
e sa
me
as E
llis
et a
l. (1
996)
96 p
ublic
atio
ns re
view
ed
(one
em
piric
ally
bas
ed
book
and
95
rese
arch
ar
ticle
s)Ye
ars
cove
red:
198
1–19
95O
f ast
eris
ked
stud
ies
in
chap
ter R
efer
ence
s,
indi
catin
g th
eir i
nclu
sion
in
revi
ew, m
ost f
rom
the
Uni
ted
Stat
es
a. Q
uant
itativ
e fin
ding
s cl
osel
y pa
ralle
l ea
rlier
revi
ew (s
ee E
llis
et a
l., 1
996,
ent
ry
belo
w)
b. Q
ualit
y of
sup
ervi
sion
rese
arch
acr
oss
15-y
ear p
erio
d ju
dged
sub
stan
dard
c. A
bout
75%
of s
tudi
es e
x po
st fa
cto
in
natu
re (n
o ra
ndom
ass
ignm
ent,
no in
de-
pend
ent v
aria
ble
man
ipul
atio
n)d.
Sup
ervi
sion
mod
els
rem
ain
min
imal
ly
stud
ied
e. S
uper
visi
on re
plic
atio
n st
udie
s ar
e sc
arce
f. C
ontin
ued
lack
of s
ound
sup
ervi
sion
m
easu
res
g. C
lient
s ge
nera
lly n
ot in
clud
ed in
sup
ervi
-si
on s
tudy
; qua
lity
of s
uper
visi
on/c
lient
ou
tcom
e re
sear
ch p
oor
The
sam
e as
Elli
s et
al.
(199
6)Po
ssib
le th
reat
s to
hyp
othe
sis
valid
ityRe
view
sub
ject
to T
ype
II er
ror
Som
e re
view
-bas
ed c
oncl
usio
ns
poss
ibly
spu
rious
Ellis
et a
l. (1
996)
Crit
ique
sup
ervi
-si
on re
sear
ch
with
rega
rd to
sc
ient
ific
rigou
r an
d m
etho
dolo
gi-
cal q
ualit
y; e
valu
-at
e 49
thre
ats
to
valid
ity a
cros
s st
udie
s; th
reat
s in
clud
ed s
tatis
ti-ca
l con
clus
ion
valid
ity, i
nter
nal
valid
ity, c
onst
ruct
va
lidity
, ext
erna
l va
lidity
, hyp
oth-
esis
val
idity
and
m
etho
dolo
gica
l th
reat
s
a. T
hat a
rtic
le b
e co
nsis
tent
w
ith p
rovi
ded
defin
ition
of
clin
ical
sup
ervi
sion
b. T
hat a
rtic
le b
e da
ta-b
ased
c. T
hat a
rtic
le in
volv
e su
perv
isio
n of
indi
vidu
al
coun
selli
ng/t
hera
py
Four
rate
rs tr
aine
d fo
r 10
hr in
ratin
g 49
th
reat
s90
% ra
ter a
gree
men
t re
ache
d be
fore
beg
in-
ning
the
stud
y
131
publ
icat
ions
(one
em
-pi
rical
ly b
ased
boo
k an
d 13
0 re
sear
ch a
rtic
les)
Year
s co
vere
d: 1
981–
1993
Of a
ster
iske
d st
udie
s in
ar
ticle
Ref
eren
ces,
indi
-ca
ting
thei
r inc
lusi
on in
re
view
, virt
ually
all
from
th
e U
nite
d St
ates
a. A
bout
75%
of s
tudi
es e
x po
st fa
cto
in
natu
reb.
On
aver
age,
inve
stig
ator
s co
nduc
ted
thei
r res
earc
h w
ith a
100
% p
roba
bilit
y of
co
mm
ittin
g on
e or
mor
e Ty
pe II
err
ors
c. A
t lea
st 8
0% o
f the
stu
dies
wer
e ju
dged
to
hav
e in
flate
d Ty
pe I
or T
ype
II er
ror
rate
s or
unr
elia
ble
mea
sure
sd.
Sel
ectio
n bi
as a
nd a
mbi
guity
of c
ausa
l di
rect
ion,
resp
ectiv
ely,
wer
e th
reat
s to
77
% a
nd 6
9% o
f the
stu
dies
e. M
ono-
met
hod
bias
, con
stru
ct c
onfo
und-
ing
and
inad
equa
te p
reop
erat
iona
l ex
plic
atio
n w
ere
thre
e co
nsis
tent
thre
ats
to c
onst
ruct
val
idity
f. In
cons
eque
ntia
l hyp
othe
ses,
am
bigu
ous
hypo
thes
es a
nd d
iffus
e st
atis
tical
hy-
poth
eses
freq
uent
ly e
mer
ged
as th
reat
s to
hyp
othe
sis
valid
ityg.
Mos
t stu
dies
had
inad
equa
te s
ampl
e si
zes
(78%
) and
use
d on
ly s
elf-
repo
rt
data
(66%
)h.
Mos
t sup
ervi
sion
stu
dies
wer
e si
mul
ta-
neou
sly
likel
y to
find
spu
rious
sig
nific
ant
resu
lts a
nd u
nlik
ely
to d
etec
t tru
e ef
fect
s
Crit
eria
def
initi
ons
coul
d ha
ve
prec
lude
d de
tect
ing
desi
gn v
ari-
atio
ns in
sup
ervi
sion
rese
arch
49 th
reat
s no
t all
incl
usiv
eN
o di
rect
com
paris
on g
roup
Sele
ctio
n bi
as p
ossi
ble
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
8 | WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Fors
haw
et a
l. (2
019)
To e
xam
ine
supe
r-vi
sors
’ exp
erie
nce
of p
rovi
ding
su-
perv
isio
n: ‘W
hat
are
the
supe
rvi-
sor's
exp
erie
nces
of
pro
vidi
ng
clin
ical
sup
ervi
-si
on to
qua
lifie
d th
erap
ists
?’ (p
. 53)
To p
erfo
rm
qual
itativ
e m
eta-
ethn
ogra
phic
syn
-th
esis
of r
evie
wed
st
udie
s
a. S
uper
viso
r's e
xper
ienc
e of
do
ing
supe
rvis
ion
bein
g th
e fo
cus
of th
e st
udy
b. Q
ualit
ativ
e m
etho
dolo
gy
empl
oyed
c. S
uper
viso
rs b
eing
eith
er
psyc
holo
gist
s, p
sych
othe
ra-
pist
s or
cou
nsel
lors
d. I
n En
glis
h la
ngua
ge
Crit
ical
App
rais
al S
kills
Pr
ogra
mm
eRe
view
pro
cess
fo
llow
ed P
RISM
A
guid
elin
es
6 ar
ticle
s re
view
ed (o
ut
of in
itial
set
of p
ossi
ble
236)
Year
s co
vere
d:
1999
–201
64
stud
ies
cond
ucte
d in
th
e U
K, 1
in A
ustr
alia
an
d 1
in th
e U
nite
d St
ates
a. P
rimar
y fin
ding
s re
volv
ed a
roun
d fo
ur
them
es: e
xper
ienc
ing
diff
icul
ties
in
supe
rvis
ion,
resp
onsi
bilit
y, s
imila
ritie
s to
trea
tmen
t and
feel
ing
capa
ble
as a
su
perv
isor
b. S
uper
visi
on, b
ecau
se o
f its
mul
tifac
-et
ed n
atur
e, in
here
nt p
ower
imba
lanc
e an
d co
mpe
ting
dem
ands
, can
be
high
ly
chal
leng
ing
c. D
oing
sup
ervi
sion
can
hav
e si
gnifi
cant
pe
rson
al a
nd p
rofe
ssio
nal i
mpa
cts,
pos
i-tiv
e an
d ne
gativ
e, o
n th
e su
perv
isor
d. D
oing
sup
ervi
sion
oft
en le
ads
to p
er-
ceiv
ed g
row
th a
s a
supe
rvis
ore.
Sup
ervi
sor t
rain
ing
can
rend
er s
uper
vi-
sor u
npre
pare
dnes
s an
d be
wild
erm
ent
avoi
dabl
e
Thre
e st
udie
s fa
iled
to s
peci
fy
part
icip
ant g
ende
r and
race
/et
hnic
ity, t
wo
faile
d to
indi
cate
su
perv
isor
s’ nu
mbe
r of y
ears
pr
ovid
ing
supe
rvis
ion
Two
stud
ies
unpu
blis
hed
Art
icle
qua
lity
ratin
gs v
aria
ble
Frei
tas
(200
2)To
revi
ew in
de
tail
rese
arch
on
sup
ervi
sion
ou
tcom
e, re
-ex
amin
ing
seve
ral
stud
ies
from
Elli
s an
d La
dany
(199
7)
and
also
add
ing
in th
ree
othe
r st
udie
s (a
ppea
ring
from
199
3 to
19
97) t
hat t
hey
did
not c
over
That
sup
ervi
sion
out
com
e be
ad
dres
sed
Non
e sp
ecifi
ed10
art
icle
s re
view
edYe
ars
cove
red:
19
81–2
001
8 ar
ticle
s fr
om th
e U
nite
d St
ates
, 1 fr
om C
anad
a an
d 1
from
Sw
eden
a. B
eing
info
rmed
by
and
draw
ing
from
re
view
of E
llis
and
Lada
ny (1
997)
, Frie
tas’
findi
ngs
high
ly c
onsi
sten
t with
thei
rsb.
Tw
o m
ost c
omm
on p
robl
ems
to e
mer
ge
acro
ss s
tudi
es w
ere
as fo
llow
s: fa
ilure
to
cont
rol f
or T
ype
I and
Typ
e II
erro
r; an
d fa
ilure
to p
rovi
de p
sych
omet
ric d
ata
on
mea
sure
s em
ploy
ed
Aut
hor s
ole
eval
uato
rN
o ob
ject
ive
appr
aisa
l pro
cedu
re
used
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
| 9WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Kühn
e et
al.
(201
9)To
revi
ew c
urre
nt
stat
us o
f sup
ervi
-si
on in
terv
entio
ns
and
met
hodo
logi
-ca
l qua
lity
of e
m-
piric
al li
tera
ture
a. S
uper
visi
on o
f psy
chot
her-
apy
the
focu
s of
eac
h st
udy
b. A
dult
patie
nts
only
c. P
ublic
atio
n pe
er-r
evie
wed
d. O
utco
me
mea
sure
incl
uded
Revi
ewer
-dev
elop
ed
com
preh
ensi
ve to
ol fo
r ev
alua
ting
met
hodo
-lo
gica
l qua
lity;
ratin
gs
prov
ided
of h
ow w
ell
key
met
hodo
logi
cal
issu
es a
ddre
ssed
(e.g
. co
ntro
l of c
onfo
unds
, so
urce
s of
bia
s)PR
ISM
A g
uide
lines
fo
llow
ed a
nd re
view
pr
otoc
ol re
gist
ered
an
d pu
blis
hed
with
PR
OSP
ERO
19 a
rtic
les
revi
ewed
(out
of
8,7
86 in
itial
hits
)Ye
ars
cove
red:
20
01–2
017
5 st
udie
s us
ed ra
n-do
mis
ed c
ontr
olle
d de
sign
, and
1 u
sed
clus
ter r
ando
mis
ed c
on-
trol
led
desi
gn; f
ollo
w-u
p da
ta c
olle
cted
in o
nly
3 st
udie
sM
ajor
ity o
f rev
iew
ed
stud
ies
from
the
Uni
ted
Stat
es; o
ther
revi
ewed
st
udie
s fr
om A
ustr
alia
, G
erm
any,
Hon
g Ko
ng,
Swed
en a
nd th
e U
K
a. S
uper
visi
on g
ener
ally
see
n as
hel
pful
by
supe
rvis
ees,
as
bene
fittin
g th
eir t
hera
-pe
utic
com
pete
nce
b. A
var
iety
of s
uper
visi
on in
terv
entio
ns
wer
e em
ploy
ed, w
ith c
ase
disc
ussi
on a
nd
prov
idin
g fe
edba
ck b
eing
mos
t evi
dent
c. A
sup
ervi
sion
man
ual o
r sup
ervi
sor
trai
ning
not
rout
inel
y a
part
of r
esea
rch
stud
ies
d. M
ost s
tudi
es w
ere
unco
ntro
lled
or u
sed
smal
l sam
ples
, with
abo
ut h
alf r
elyi
ng
excl
usiv
ely
on s
elf-
repo
rt q
uest
ionn
aire
se.
Virt
ually
all
stud
ies
wer
e at
hig
h ris
k w
ith
rega
rd to
thre
e m
etho
dolo
gica
l qua
lity
issu
es: p
artic
ipan
t sel
ectio
n, c
ontr
ol o
f co
nfou
nds
and
othe
r sou
rces
of b
ias
f. Su
perv
isio
n re
sear
ch la
gs w
ell b
ehin
d ps
ycho
ther
apy
rese
arch
g. T
hese
que
stio
ns s
till r
emai
n w
ithou
t de-
finiti
ve a
nsw
ers:
Wha
t are
sup
ervi
sion
's ac
tive
ingr
edie
nts?
Doe
s su
perv
isio
n ef
fect
clie
nt c
hang
e?
Eval
uatio
n to
ol
revi
ewer
-dev
elop
edSt
ricte
r ope
ratio
nalis
atio
n of
in
clus
ion
crite
ria m
ay h
ave
resu
lted
in fe
wer
revi
ewed
stu
d-ie
s be
ing
sele
cted
, acc
ordi
ngly
lim
iting
dra
wab
le c
oncl
usio
ns
New
man
et
al. (
2019
)To
revi
ew c
urre
nt
evid
ence
abo
ut
supe
rvis
ion
in
scho
ol p
sych
olog
y
Stud
ies
requ
ired
to fo
cus
conc
eptu
ally
or e
mpi
rical
ly
on s
choo
l psy
chol
ogic
al
supe
rvis
ion
proc
esse
s an
d pr
actic
es
Syst
emat
ic m
appi
ng; r
e-vi
ew d
escr
iptiv
e as
op-
pose
d to
inte
rpre
tive
37 a
rtic
les
revi
ewed
(out
of
initi
al s
ampl
e of
70
artic
les)
, with
21
bein
g em
piric
al a
nd 1
6 be
ing
conc
eptu
al; o
f 21
empi
ri-ca
l pub
licat
ions
, 13
wer
e su
rvey
s, fo
ur w
ere
qual
itativ
e, tw
o w
ere
mix
ed-m
etho
ds, a
nd tw
o w
ere
quan
titat
ive
Year
s co
vere
d:
2000
–201
7M
ost s
tudi
es fo
cusi
ng
on s
choo
l psy
chol
-og
y su
perv
isio
n in
the
Uni
ted
Stat
es; o
ther
co
ntrib
utio
ns c
omin
g fr
om A
ustr
alia
, Hon
g Ko
ng, N
ew Z
eala
nd a
nd
the
UK
a. S
choo
l psy
chol
ogy
supe
rvis
ion
liter
atur
e qu
ite li
mite
d, it
s re
sear
ch la
ggin
g w
ell
behi
nd s
uper
visi
on re
sear
ch in
rela
ted
field
sb.
Top
ics
rece
ivin
g m
inim
al a
tten
tion
incl
ude
the
follo
win
g: s
uper
visi
on
proc
esse
s; o
utco
me;
mod
els,
met
hods
an
d fo
rmat
s; te
chno
logy
; sup
ervi
sor
trai
ning
; cer
tain
are
as o
f div
ersi
ty; a
nd
gate
keep
ing
c. A
met
hodo
logi
cally
plu
ralis
tic a
ppro
ach
to d
oing
sch
ool p
sych
olog
y su
perv
isio
n re
sear
ch re
com
men
ded
Empi
rical
rigo
ur o
f rev
iew
ed s
tud-
ies
not a
sses
sed
Met
hods
and
met
hodo
logy
de-
scrip
tions
take
n at
face
val
ue
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
10 | WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
O’D
onog
hue
& T
sui
(201
5)
To p
rovi
de c
om-
preh
ensi
ve re
view
of
rese
arch
on
prac
tisin
g so
cial
w
orke
rs
a. T
hat a
rtic
les
be
peer
-rev
iew
edb.
Tha
t art
icle
s in
clud
e fir
st-
hand
em
piric
al in
form
atio
nc.
Tha
t art
icle
s be
abo
ut
prac
tisin
g so
cial
wor
kers
, no
t stu
dent
s
Non
e sp
ecifi
ed86
art
icle
s re
view
edYe
ars
cove
red:
19
70–2
010
Abo
ut 6
0% o
f stu
dies
fr
om th
e U
nite
d St
ates
; ot
her n
otab
le c
on-
trib
utor
s w
ere
Aus
tral
ia,
Hon
g Ko
ng, I
srae
l and
N
ew Z
eala
nd
a. S
ocia
l wor
k su
perv
isio
n re
sear
ch
incr
ease
d in
num
ber a
nd g
eogr
aphi
cal
spre
adb.
Res
earc
h de
sign
s ha
ve in
crea
sing
ly d
iver
-si
fied
and
show
incr
easi
ng s
ophi
stic
atio
n of
dat
a an
alys
esc.
Sup
ervi
sor e
mot
iona
l sup
port
with
in
trus
ting
rela
tions
hip
miti
gate
s w
ork
stre
ss im
pact
and
pos
itive
ly re
late
d to
jo
b sa
tisfa
ctio
nd.
Cul
tura
l diff
eren
ces
exam
ined
in a
few
st
udie
se.
Clie
nts
min
imal
ly in
clud
ed in
stu
dies
, in
fluen
ce o
f sup
ervi
sion
on
clie
nt o
ut-
com
es la
rgel
y un
exam
ined
f. Em
piric
ally
sup
port
ed s
uper
visi
on p
rac-
tice
mod
el la
ckin
g
No
appr
aisa
l pro
cedu
re u
sed
Proc
ess
of a
rriv
ing
at c
oncl
usio
ns
unsp
ecifi
ed
Reis
er &
M
ilne
(201
4)A
pply
ing
fidel
ity
fram
ewor
k to
su
perv
isio
n ou
t-co
me
stud
ies
a. A
rtic
les
from
pee
r-re
view
ed
jour
nals
ove
r the
last
30
yea
rsb.
Dire
ct m
easu
re o
f clie
nt
outc
ome
incl
uded
c. S
uper
visi
on p
rovi
ded
had
to b
e of
psy
chot
hera
py o
r co
unse
lling
ser
vice
sd.
Sup
ervi
sion
pro
vide
d co
nsis
tent
with
Miln
e's
(200
7) e
mpi
rical
def
initi
on
of c
linic
al s
uper
visi
on
Fide
lity
Fram
ewor
k C
heck
list
12 a
rtic
les
revi
ewed
(out
of
initi
al s
earc
h sa
mpl
e of
48)
Year
s cov
ered
: 198
1–20
107
stud
ies
from
the
Uni
ted
Stat
es; 2
from
Aus
tral
ia,
2 fr
om th
e U
K an
d 1
from
Can
ada
a. A
reas
of i
nfid
elity
wer
e id
entif
ied
acro
ss
mos
t rev
iew
ed s
tudi
es, w
ith th
e m
ost
trou
blin
g is
sues
bei
ng re
ceip
t, tr
aini
ng
and
deliv
ery
b. F
idel
ity fr
amew
ork
cons
ider
ed u
sefu
l fo
r ass
essi
ng a
dher
ence
in s
uper
visi
on
stud
ies
c. A
rgum
ent m
ade
for a
lso
givi
ng ‘a
cid
test
’ w
eigh
t to
othe
r out
com
es b
eyon
d cl
ient
ou
tcom
es a
lone
Poss
ible
rate
r bia
sEx
clus
iona
ry c
riter
ia m
ay h
ave
been
ove
rly re
stric
tive
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
| 11WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Sim
pson
-So
uthw
ard
et a
l. (2
017)
Exam
inat
ion
of
cons
iste
ncie
s ac
ross
sup
ervi
-si
on m
odel
s
a. T
hat t
he s
uper
visi
on m
odel
de
scrib
es w
hat h
appe
ns in
su
perv
isio
nb.
Tha
t mod
el b
e ab
out o
ne-
to-o
ne s
uper
visi
onc.
In
Engl
ish
lang
uage
Con
tent
ana
lysi
s co
nduc
ted
Art
icle
s ev
alua
ted
with
fiv
e qu
estio
ns in
min
d:a.
Was
sup
ervi
see
lear
ning
and
/or
dev
elop
men
t ad
dres
sed?
b. W
ere
man
age-
rial a
nd/o
r eth
ical
re
spon
sibi
litie
s co
nsid
ered
?c.
Wer
e em
otio
nal
effe
cts
of th
e w
ork
cons
ider
ed?
d. W
as th
e m
odel
ba
sed
on e
mpi
rical
ev
iden
ce?
e. D
oes
the
mod
el ta
ke
into
acc
ount
sup
ervi
-so
r, su
perv
isee
and
cl
ient
?
52 a
rtic
les
revi
ewed
Year
s co
vere
d:
1964
–201
5
a. M
ost s
uper
visi
on m
odel
s, w
hile
focu
sing
on
sup
ervi
see
lear
ning
and
/or d
evel
-op
men
t, la
ck c
onsi
sten
cy a
nd a
re n
ot
empi
rical
ly b
ased
b. S
uper
visi
on m
odel
s ar
e la
rgel
y as
sum
p-tio
n-ba
sed
in n
atur
ec.
The
clie
nt re
mai
ns th
e ne
glec
ted
fact
or
in s
uper
visi
on m
odel
s
The
auth
ors
conc
lude
that
‘The
re
is li
ttle
evi
denc
e fo
r the
ef
fect
iven
ess
of s
uper
visi
on’ (
p.
1,22
8). W
hile
rese
arch
abo
ut th
e pa
rtic
ular
mod
els
in q
uest
ion
was
con
side
red,
sup
ervi
sion
ou
tcom
e re
sear
ch w
as n
ot
spec
ifica
lly e
valu
ated
as
part
of
conc
eptu
al re
view
. Per
haps
the
mos
t pre
cise
sta
tem
ent i
s th
is:
‘The
re is
litt
le e
vide
nce
for t
he
effe
ctiv
enes
s of
sup
ervi
sion
m
odel
s’.Va
rious
asp
ects
of t
he c
onte
nt
anal
ysis
can
be
ques
tione
d, s
ome
exam
ples
bei
ng a
s fo
llow
s:A
lthou
gh a
t lea
st tw
o m
odel
s w
ere
labe
lled
and
desc
ribed
as
dev
elop
men
tal,
they
wer
e si
mul
tane
ousl
y id
entif
ied
as n
ot
addr
essi
ng s
uper
vise
e de
velo
p-m
ent (
see
Hes
s an
d W
ard/
Hou
se
entr
ies)
und
er th
e ap
prai
sal
ques
tion,
‘was
sup
ervi
see
lear
ning
and
/or d
evel
opm
ent
addr
esse
d?’;
Som
e su
perv
isio
n m
odel
s ar
e id
entif
ied
as n
ot a
ddre
ssin
g a
part
icul
ar s
uper
visi
on d
imen
-si
on th
at th
ey d
o in
deed
see
m
to a
ddre
ss (s
ee e
ntrie
s fo
r the
Fr
awle
y-O
’Dea
/Sar
nat r
elat
iona
l m
odel
; Haw
kins
/Sho
het s
even
-ey
ed m
odel
; Mue
ller/
Kell
copi
ng
with
con
flict
mod
el; a
nd S
caife
m
odel
)O
ne m
odel
, a m
odel
abo
ut s
uper
-vi
sor d
evel
opm
ent,
is fa
ulte
d fo
r no
t foc
usin
g on
the
supe
rvis
ee
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
12 | WATKINS
Aut
hors
Focu
s of r
evie
wIn
clus
ion
crite
riaA
ppra
isal
pro
cedu
re/
crite
riaFe
atur
es o
f rev
iew
ed
publ
icat
ions
Find
ings
/con
clus
ions
Stud
y lim
itatio
ns/i
ssue
s
Tsui
(199
7)To
pro
vide
com
-pr
ehen
sive
revi
ew
of re
sear
ch o
n pr
actis
ing
soci
al
wor
kers
a. T
hat a
rtic
les
be
peer
-rev
iew
edb.
Tha
t art
icle
s in
clud
e fir
st-
hand
em
piric
al in
form
atio
nc.
Tha
t art
icle
s be
abo
ut
prac
tisin
g so
cial
wor
kers
, no
t stu
dent
s
Non
e sp
ecifi
ed30
art
icle
s re
view
edYe
ars
cove
red:
19
70–1
995
Men
tion
mad
e of
sup
ervi
-si
on re
sear
ch c
omin
g fr
om A
ustr
alia
, Can
ada,
H
ong
Kong
, Isr
ael a
nd
the
Uni
ted
Stat
es
a. S
ocia
l wor
k su
perv
isio
n re
sear
ch
embr
yoni
cb.
Mos
t stu
dies
wer
e on
e-sh
ot, c
ross
-se
ctio
nal s
urve
ys, r
elie
d on
sel
f-re
port
qu
estio
nnai
res,
and
had
pro
blem
s w
ith
inte
rnal
, ext
erna
l, an
d/or
con
stru
ct
valid
ityc.
Lac
k of
pro
gram
mat
ic in
vest
igat
ion
d. N
o st
udy
addr
esse
d cu
lture
e. C
lient
out
com
es m
inim
ally
add
ress
ed
Aut
hor s
ole
eval
uato
rN
o ob
ject
ive
appr
aisa
l pro
cedu
re
used
Proc
ess
of a
rriv
ing
at c
oncl
usio
ns
unsp
ecifi
ed
Wat
kins
(2
011)
To e
xam
ine
the
effe
cts
of s
uper
-vi
sion
on
clie
nt
outc
omes
a. U
sed
iden
tifie
d ou
tcom
e st
udie
s fr
om fo
ur p
revi
ous
revi
ews
b. I
dent
ified
add
ition
al a
rtic
les
for i
nclu
sion
app
earin
g si
nce
four
revi
ews,
usi
ng
‘supe
rvis
ion’
and
‘out
com
e’
as s
earc
h w
ords
Non
e sp
ecifi
ed18
art
icle
s re
view
edYe
ars c
over
ed: 1
981–
2010
11 s
tudi
es fr
om th
e U
nite
d St
ates
; oth
er
cont
ribut
ions
com
-in
g fr
om th
e U
K (3
), A
ustr
alia
(2),
Can
ada
(1)
and
Swed
en (1
)
a. M
isid
entif
icat
ion,
dat
a du
plic
atio
n an
d in
adeq
uate
mea
sure
men
t ide
ntifi
ed a
s pr
oble
ms
with
set
of s
tudi
esb.
Not
abl
e to
con
fiden
tly s
ay th
at s
uper
vi-
sion
pos
itive
ly a
ffec
ts c
lient
out
com
ec.
Thr
ee s
tudi
es id
entif
ied
as p
oint
ing
the
way
forw
ard
for f
utur
e su
perv
isio
n ou
tcom
e in
vest
igat
ions
Aut
hor s
ole
eval
uato
rN
o ob
ject
ive
appr
aisa
l pro
cedu
re
used
Whe
eler
&
Rich
ards
(2
007)
Revi
ew c
ondu
cted
so
as
to a
nsw
er
this
que
stio
n:
Wha
t im
pact
doe
s su
perv
isio
n ha
ve
on c
ouns
ello
rs/
ther
apis
ts, t
heir
prac
tice
and
thei
r cl
ient
s?
a. S
tudi
es e
ither
pub
lishe
d or
un
publ
ishe
d, q
uant
itativ
e or
qu
alita
tive
(but
no
unpu
b-lis
hed
stud
ies
incl
uded
in
final
sam
ple)
b. V
alid
, rel
iabl
e m
easu
re o
r rig
orou
s qu
alita
tive
anal
ysis
us
ed to
sho
w s
uper
visi
on's
impa
ct o
n su
perv
isee
c. S
uper
vise
es h
ad to
be
prac
tisin
g co
unse
llors
or
ther
apis
tsd.
In
Engl
ish
lang
uage
Revi
ew c
ondu
cted
us
ing
EPPI
-Rev
iew
er
soft
war
eIn
clus
ion/
excl
usio
n cr
iteria
pro
gram
med
in
to s
oftw
are
Dat
a ex
trac
tion
proc
e-du
re a
lso
prog
ram
med
in
to s
oftw
are,
so
that
va
rious
mat
ters
of
met
hodo
logy
and
de-
sign
cou
ld b
e re
cord
edQ
ualit
y ra
tings
mad
e as
to
eac
h st
udy'
s m
eth-
odol
ogic
al s
ound
ness
an
d ab
out h
ow w
ell
each
stu
dy fi
t with
the
incl
usio
n cr
iteria
18 a
rtic
les
revi
ewed
(out
of
8,2
95 in
itial
hits
)Ye
ars
cove
red:
19
81–2
005
14 s
tudi
es fr
om th
e U
nite
d St
ates
, 2 s
tudi
es
from
Sw
eden
and
2
stud
ies
from
the
UK
a. S
uper
visi
on s
how
n to
hav
e so
me
impa
ct
on s
uper
vise
e se
lf-aw
aren
ess,
ski
ll de
vel-
opm
ent a
nd s
elf-
effic
acy
b. M
inim
al re
sear
ch a
tten
tion
give
n to
the
clie
nt: ‘
…no
stu
dies
…of
fer s
ubst
antia
l ev
iden
ce to
sup
port
impr
ovem
ent i
n cl
i-en
t out
com
es’ (
p. 6
3)
Base
d on
qua
lity
ratin
gs, o
nly
two
of 1
8 st
udie
s ra
ted
‘ver
y go
od’
Spec
ifics
abo
ut e
mpi
rical
rigo
ur
of re
view
ed s
tudi
es (o
r lac
k th
ereo
f) no
t pro
vide
dSt
udie
s id
entif
ied
as b
eing
sup
-po
rtiv
e of
any
giv
en a
rea
(e.g
. su
perv
isio
n an
d se
lf-aw
aren
ess)
qu
ite li
mite
dSe
lf-re
port
mea
sure
s pr
edom
inan
t14
stu
dies
wer
e fr
om th
e U
nite
d St
ates
TAB
LE 1
(C
ontin
ued)
(Con
tinue
s)
| 13WATKINS
half of the reviews involving anywhere from 10 to 40 articles. The total span of years covered ranged from 1964 (Simpson-Southward et al., 2017) up to 2017 (Barrett et al., 2019), with most reviews account-ing for a span of 10–30 years. The vast majority of reviewed studies came from the United States. Other primary contributors were the UK and Australia, followed (in alphabetical order) by Canada, Finland, Germany, Hong Kong, Israel, New Zealand, Norway and Sweden.
4.1.5 | Findings/conclusions
Findings/conclusions can be placed together into four rough group-ings: supervision impact, supervision models, methodological mat-ters and other (miscellaneous concerns or conclusions).
Supervision impactSupervision, found to be positively associated with job satisfaction, job retention and ability to manage workload (Carpenter et al., 2013; O’Donoghue & Tsui, 2015), appears to be seen as helpful by supervi-sees (Kühne et al., 2019) and may even benefit their therapeutic compe-tence (e.g. enhanced self-awareness, enhanced sense of self-efficacy; Alfonsson et al., 2018; Kühne et al., 2019; Wheeler & Richards, 2007). But supervision's favourable impact on worker outcomes is weak at best, yet to be proven (Bogo & McKnight, 2006; Carpenter et al., 2013). Furthermore, the client has been, and continues to be, summarily ne-glected in supervision research: supervision's impact on client outcome has yet to be proven (Alfonsson et al., 2018; Bogo & McKnight, 2006; Buus & Gonge, 2009; Carpenter et al., 2013; Ellis & Ladany, 1997; Ellis et al., 1996; O’Donoghue & Tsui, 2015; Simpson-Southward et al., 2017; Tsui, 1997; Watkins, 2011; Wheeler & Richards, 2007).
Study data about giving and receiving supervision also accentuated this: supervision, because of its multifaceted nature, inherent power imbalance and competing demands, can be highly challenging and have significant personal and professional impacts, positive and negative, on both supervisor and supervisee, and both parties are advised to remain mindful of those realities (Forshaw et al., 2019; Wilson et al., 2016).
Supervision modelsSupervision models largely went untested, with no clear evi-dence supporting any given model (Alfonsson et al., 2018; Bogo & McKnight, 2006; Buus & Gonge, 2009; Carpenter et al., 2013; Ellis & Ladany, 1997; Ellis et al., 1996; Simpson-Southward et al., 2017). Supervision lacks evidence-based practice research and an evidence-based practice model (Barrett et al., 2019; O’Donoghue & Tsui, 2015; Simpson-Southward et al., 2017).
Methodological mattersMethodological issues and study shortcomings were consistently iden-tified and can be placed into three (not necessarily mutually exclusive) broad groupings: (a) failure to control for compromising variables; (b) type of study conducted; and (c) recurring issues about critical study components. Frequently identified compromising variables included unaddressed threats to validity, sources of bias and confounds, as well A
utho
rsFo
cus o
f rev
iew
Incl
usio
n cr
iteria
App
rais
al p
roce
dure
/cr
iteria
Feat
ures
of r
evie
wed
pu
blic
atio
nsFi
ndin
gs/c
oncl
usio
nsSt
udy
limita
tions
/iss
ues
Wils
on e
t al.
(201
6)To
exa
min
e tr
aine
e th
erap
ists
’ exp
eri-
ence
s of
rece
ivin
g su
perv
isio
nTo
per
form
qua
lita-
tive
met
a-sy
nthe
-si
s of
revi
ewed
st
udie
s
a. T
hat q
ualit
ativ
e de
sign
be
empl
oyed
, whe
re in
ter-
view
s or
focu
s gr
oups
wer
e us
edb.
Tha
t foc
us b
e on
trai
n-ee
s’ pr
evio
us s
uper
visi
on
expe
rienc
esc.
In
Engl
ish
lang
uage
Crit
ical
App
rais
al S
kills
Pr
ogra
mm
e15
art
icle
s re
view
ed (o
ut
of 1
,019
initi
al h
its)
Year
s co
vere
d:
1996
–201
29
stud
ies
from
the
Uni
ted
Stat
es, 3
from
the
UK
, 1
from
Aus
tral
ia, 1
from
C
anad
a an
d 1
from
N
orw
ay
a. F
our c
ruci
al th
emes
iden
tifie
d: s
uper
vi-
sion
as
a le
arni
ng o
ppor
tuni
ty; t
he
supe
rvis
ory
rela
tions
hip;
pow
er in
sup
er-
visi
on; a
nd s
uper
visi
on im
pact
. Pos
itive
an
d ne
gativ
e as
pect
s of
eac
h th
eme
cons
ider
edb.
Sup
ervi
sion
can
con
trib
ute
to s
uper
vise
e pe
rson
al a
nd p
rofe
ssio
nal d
evel
opm
ent,
but a
lso
has
the
pote
ntia
l to
caus
e su
per-
vise
e st
ress
, anx
iety
and
sel
f-do
ubt
c. S
uper
viso
rs s
houl
d re
mai
n m
indf
ul o
f th
e po
wer
diff
eren
tial a
nd it
s im
pact
on
supe
rvis
ion
Maj
ority
of p
artic
ipan
ts fe
mal
e
Not
e: C
ON
SORT
, Con
solid
ated
Sta
ndar
ds o
f Rep
ortin
g Tr
ials
; EPP
I, Ev
iden
ce fo
r Pol
icy
and
Prac
tice
Info
rmat
ion
and
Co-
ordi
natin
g C
entr
e; P
RISM
A, P
refe
rred
Rep
ortin
g Ite
ms
for S
yste
mat
ic R
evie
ws
and
Met
a-A
naly
ses;
PRO
SPER
O, I
nter
natio
nal P
rosp
ectiv
e Re
gist
er o
f Sys
tem
atic
Rev
iew
s; S
TRO
BE, S
tren
gthe
ning
the
Repo
rtin
g of
Obs
erva
tiona
l Stu
dies
in E
pide
mio
logy
.EP
PI-R
evie
wer
sof
twar
e is
a W
eb-b
ased
sof
twar
e pr
ogra
m fo
r man
agin
g an
d an
alys
ing
data
in li
tera
ture
revi
ews.
TAB
LE 1
(C
ontin
ued)
14 | WATKINS
as inadequate reporting and description of study specifics (Alfonsson et al., 2018; Buus & Gonge, 2009; Carpenter et al., 2013; Ellis et al., 2008; Ellis & Ladany, 1997; Ellis et al., 1996; Freitas, 2002; Kühne et al., 2019; Tsui, 1997). The vast majority of studies were described as being ex post facto, cross-sectional and correlational (i.e. no randomi-sation, no independent variable manipulation), with very few being experimental investigations (Bogo & McKnight, 2006; Borders, 2006; Carpenter et al., 2013; Ellis & Ladany, 1997; Ellis et al., 1996; Tsui, 1997). Recurring problematic issues included small sample sizes, an over-reliance on self-report measures and retrospective accounts, and a lack of sound supervision measures from which to draw (Alfonsson et al., 2018; Bogo & McKnight, 2006; Borders, 2006; Buus & Gonge, 2009; Ellis et al., 2008; Ellis & Ladany, 1997; Ellis et al., 1996; Kühne et al., 2019; O’Donoghue & Tsui, 2015; Tsui, 1997).
4.1.6 | Other study limitations/issues
Other study limitations/issues included inclusion criteria being overly restrictive or perhaps not restrictive enough (e.g. Barrett et al., 2019; Carpenter et al., 2013), some studies selected for review having incomplete information (Alfonsson et al., 2018) and being of variable quality (Wheeler & Richards, 2007), and about one-third of the reviews involving no objective appraisal procedure (e.g. Bogo & McKnight, 2006; Watkins, 2011). Additional issues also mentioned or identified were as follows: supervision process is a neglected variable and could benefit from study (Alfonsson et al., 2018); and neither a supervision manual nor supervisor training is routinely in-cluded in supervision research (Kühne et al., 2019).
4.1.7 | What do these reviews tell us? The two 12-year periods compared
Perhaps the clearest comparative results to emerge would be the fol-lowing: (a) the primary problematic features of and limitations that at-tended clinical supervision research from 1996 to 2007 (e.g. ex post facto designs, small samples) have loudly lingered into the 2008–2019 period and remain as current problematic features and limitations; (b) the use of some type of study appraisal procedure was standard fare for most reviews appearing from 2008 to 2019, but the reverse was the case for reviews appearing during the earlier time period; and (c) systematic supervision reviews appearing in the last several years have been more apt to be conducted in accordance with established sys-tematic review guidelines (e.g. PRISMA and PRISMA-P [PRISMA for Protocols; Moher, Shamseer, et al., 2015; Shamseer et al., 2015]).
5 | DISCUSSION
These survey findings can be viewed in two ways, as the ‘glass being half empty’ and the ‘glass being half full’. It seems important to hold both views in mind.
5.1 | The ‘Glass Half Full’
These reviews appear to reflect (a) a growing and lively research interest in supervision that increasingly stretches around the globe, that interest being most clearly on display across these last 15 years (cf. Pelling, Abbott, & Lack, 2017); (b) some diversification in study de-signs and sophistication of statistical analyses, particularly the wel-come embrace of qualitative studies (Borders, 2006; O’Donoghue & Tsui, 2015); and (c) continuing efforts being made across mental health disciplines to closely scrutinise supervision research so as to advance the field. Those observations are also reflected in other re-cent assessments about supervision and its research and seemingly are givens going forward (Bernard & Goodyear, 2019; Inman et al., 2014; Sewell, 2018). Supervision researchers are clearly ‘working on the “search for scientific rigour”...’ (Milne et al., 2012, p. 144), and all indications are that they will continue to vigorously do so.
5.2 | The ‘Glass Half Empty’
Four issues are most concerning: supervision impact, supervision models, methodological matters and evidence-based supervision.
5.2.1 | Supervision impact
Scholarly opinion—based upon the supposed weight of empirical evi-dence—is that supervision contributes to supervisee competence de-velopment and skill transfer, but any impact on client outcome has yet to be proven (e.g. Bernard & Goodyear, 2019). But what is most sur-prising from this survey of reviews is the limited evidence that affirms any type of supervision impact at all: although attention to clients and client outcome was identified frequently and repeatedly as being sorely needed (e.g. Alfonsson et al., 2018; Kühne et al., 2019), any evidence that supervision benefits supervisees was highly tentative and quite thin at best. Supervision has indeed been found to be as-sociated with some positive outcomes (e.g. increased sense of super-visee self-efficacy; Carpenter et al., 2013; O’Donoghue & Tsui, 2015; Wheeler & Richards, 2007), but beyond saying that, these reviews offer little else that can be definitively concluded about supervision's impact. If ‘…the evidence of supervisor impact has been well estab-lished’ (Bernard & Goodyear, 2014, p. 301), where is that evidence?
The question ‘Does supervision work?’ still seems to be a wide open question, not answered affirmatively and conclusively on any front across these reviews. Could it be that claims about supervision's impact and effectiveness have been overstated, maybe even over-sold (Alfonsson et al., 2018; Carpenter et al., 2013; White, 2017)?
5.2.2 | Supervision models
Supervision models generally went unexamined across these re-views, any such study appearing to be more often a product of the
| 15WATKINS
past than present. The idea of evidence-supported supervision models appears to be far more hope and dream than reality at pre-sent. Even when the one model-specific review about cognitive-behavioural supervision (Alfonsson et al., 2018) is closely examined, one could question whether it is truly a model-specific review. For example, one of the five reviewed ‘so designated’ CBT supervision studies—the often cited research investigation by Bambling, King, Raue, Schweitzer, and Lambert (2006)—mentions CBT minimally, instead the focus being on alliance and problem-solving conditions, and involved the use of an unsound, nonspecific supervision manual (Milne, 2016). Three of the other four studies, while occurring within the context of supervised CBT, were examinations about the impact of different modes of supervision delivery (e.g. bug-in-the-eye vs. video feedback, cotherapy treatment/supervision vs. single-pro-vider treatment/supervision). The fifth study's supervision condition consisted of but one 30-min Skype contact per month for a total of three months. Again, that these purportedly ‘carefully selected stud-ies’ are actual tests of CBT supervision can be questioned.
Pushing this line of thinking even further, how much do supervi-sion models really impact daily supervision practice anyway? Could it be that ‘most supervisors eventually develop their own unique inte-grationist perspectives..., [that that outcome] probably is inevitable’ (Bernard & Goodyear, 2004, p. 100)? If that is the case, what specific purpose does the multitude of models serve (Simpson-Southward et al., 2017)? Perhaps such models provide supervisors with the bits and pieces from which they construct those unique integrationist per-spectives, with some of those larger bits and pieces (e.g. supervisory alliance; Park et al., 2019) ideally enjoying some level of empirical support. But beyond possible piecemeal support, empirical backing for the vast majority of models (with the discrimination and devel-opmental models excepted; Bernard & Goodyear, 2019; Rønnestad, Orlinsky, Schröder, Skovholt, & Willutzki, 2019) appears lacking. Based on this survey of reviews, it does not seem beyond reasonable to wonder, ‘Is research on supervision models dead?’
5.2.3 | Methodological matters
The primary methodological problems that plagued supervision re-search in the 1990s are still the problems that plague research today: ‘The conclusions from previous systematic reviews of clinical super-vision, that there is a dire need for more empirical evaluations and that many studies in this area lack adequate methodological rigor, unfortunately still hold’ (Alfonsson et al., 2018, p. 219). Supervision's long lingering methodological issues—predominance of ex post facto, cross-sectional, correlational designs, small sample sizes, over-reliance on self-report measures, lack of psychometrically sound supervision measures, lack of experimental designs and longitudinal study—remain in force, temper any resulting research findings, com-promise the field's advancement and leave us with this reality: we still know relatively little about the effects of supervision on supervisees, supervisors or clients (Hill & Knox, 2013). Numerous supervision scholars agree that substantive research benefits will only accrue
when supervision studies do a better job of the following: employ-ing a pluralistic research approach that, in conjunction with ex post facto, cross-sectional study, also includes experimental investiga-tion and randomisation as well as data-driven case studies (Borders, in Goodyear et al., 2016; Pelling et al., 2017); conducting multisite studies so as to increase sample size (Hill & Knox, 2013; Russell et al., 1984); developing psychometrically sound supervision measures and subjecting existing substandard measures to proper validation procedures (Ellis et al., 2008; Ellis & Ladany, 1997); employing a mul-timethod, multitrait measurement approach, where self-report is but one type of data gathered (Ladany & Malouf, 2010; Lambert & Ogles, 1997); attending seriously to particular methodological issues (e.g. inadequate statistical power, threats to validity) during study devel-opment and execution (Ellis, 1991; Wampold & Holloway, 1997; cf. Schutt, 2012); examining supervision longitudinally (Holloway, 1992; Hill & Knox, 2013; cf. Potts, 2018); and studying supervision as a full-fledged triad, including the client and client outcomes as important parts of the supervisory picture (Ladany & Inman, 2012; Lambert & Ogles, 1997). Although those remedies have been identified for decades now, their implementation still lags far behind, negatively affects supervision research product and awaits realisation.
5.2.4 | Evidence-based supervision
Evidence-based supervision—built on a foundation of research evidence, expert consensus and practitioner expertise—can be understood practically as involving two fundamental features: (a) supervisor utilisation of research-supported supervisory methods and techniques during supervision; and (b) supervisor encourage-ment of supervisees to use empirically validated psychological interventions during therapy (Barrett et al., 2019; Milne, 2018). Evidence-based supervision (along with that which is competency-based) has indeed emerged as supervision's Zeitgeist, is gener-ally touted as being a widely beneficial scientific stance to adopt (Bernard & Goodyear, 2019) and is considered crucial for good su-pervision practice and supervision's advancement (Reiser & Milne, 2012; O'Donoghue et al., 2018). But one contrary reality remains: any evidence-based supervision studies seem to be a most scant commodity (Barrett et al., 2019), highly desirable but largely uni-dentifiable in the literature. If supervision is an evidence-based or evidence-informed affair, or is at least moving increasingly in that evidence-based direction, where is the actual evidence that points to that being so (cf. Barrett et al., 2019; O’Donoghue & Tsui, 2015)?
5.3 | An emerged trend? An emerging trend? And a caution
5.3.1 | An emerged trend?
That effort has been increasingly made to bring some type of more objective appraisal procedure to the systematic review process
16 | WATKINS
seems good. Whereas 13 of these 20 reviews involved the use of an appraisal procedure or criteria, almost all of those 13 reviews appeared during the last approximate decade alone. This would seem to reflect increasingly ongoing efforts to render the review process ever tighter and more rigorous. Those efforts seem des-tined and designed to bring benefit to any resulting supervision product.
5.3.2 | An emerging trend?
Efforts have also been increasingly made across these last 15 years to put more ‘systematic’ in the systematic review process and its product. Examples of that include the (2012) establishment of the journal, Systematic Reviews (Moher, Stewart, & Shekelle, 2015), re-lease of the second edition of the Cochrane Handbook of Systematic Reviews of Interventions (Higgins & Thomas, 2019), introduction and widening embrace of the PRISMA (systematic review) guide-lines (Moher et al., 2009), and the widening embrace of proto-cols that document the systematic review plan and procedures (e.g. PRISMA-P; Moher, Shamseer, et al., 2015; Shamseer et al., 2015) and the subsequent registration of those protocols (e.g. with PROSPERO, the international register launched in 2011; Page et al., 2018). All indications are that these systematic review changes are highly valued and here to stay because they (a) provide a pur-pose-driven organising structure for reviews to follow, (b) increase the rigour of the review process, (c) make prespecification explicit and (d) promote consistency, accountability, research integrity and transparency (Moher et al., 2009; Moher, Shamseer, et al., 2015). That three of the most recent supervision reviews surveyed here (Alfonsson et al., 2018; Barrett et al., 2019; Kühne et al., 2019) followed these systematic review and registration guidelines also seems good, reflecting efforts by supervision researchers to maintain the highest research standards, be highly rigorous in the process and produce the promised product for study and scrutiny. Future supervision systematic reviews would do well to follow the lead of these three efforts (Alfonsson et al., 2018; Barrett et al., 2019; Kühne et al., 2019).
5.3.3 | A caution
But just as there is much value to be found in their review ap-proach, two of these reviews (Alfonsson et al., 2018; Barrett et al., 2019) also raise this question for consideration: At what point does restriction of allowed studies become so restrictive that find-ing meaningful results becomes a virtual impossibility? Four stud-ies were reviewed in Barrett et al. (2019) and five in Alfonsson et al. (2018). It would seem that whereas more studies reviewed do not necessarily lead to quality findings, studies that are both more variable in content and fewer in number will be increasingly apt to lead to highly limited, less robust review findings. For example, Alfonsson et al. (2018), in commenting on their selected studies,
stated: ‘The five studies varied greatly regarding background, scope and quality, making it difficult to compare the studies and summarize the findings…’ (p. 219). The combination of study num-ber and variability would seem a most important consideration to bear in mind when thinking about supervision study selection for a systematic review.
5.4 | Survey limitations/Cautions
Three particular limitations or cautions attend this survey of re-views and merit mention. First, being a single-author effort, this survey does not benefit from having another set or multiple sets of eyes and perspectives brought to bear on what was done. Although I have tried to be careful in my thinking about this sur-vey and be meticulous in its conduct, readers may reasonably take issue with some aspects of what was done here and how it was done (e.g. my decision about what reviews to include and exclude). Second, review findings are a product of the reviewed studies; these 20 reviews, while providing generally good coverage of the supervision literature, did not cover the entire expanse of super-vision investigations. It is possible that some of my conclusions, while consistent with what is in these 20 reviews, are not in sync with the broader body of supervision literature. Third, I have taken a highly critical view here; some might even say too much so. If nothing else, perhaps this highly critical perspective might serve as a stimulus for counterpoint discussion.
6 | CONCLUSION
Practising supervisors and supervisees tend to believe in, and have con-viction about, the benefits, power and potential of supervision (Rast, Herman, Rousmaniere, Whipple, & Swift, 2017). But belief and convic-tion do not necessarily translate into empirical reality. It could be rea-sonably argued that supervision is still too much a product of ‘proof by association’, that supervision's methodological morass long lingers (e.g. compare Hill & Knox [2013] with Holloway & Hosford [1983]), and that if supervision is to ever substantially and fruitfully advance, research has to move beyond a preponderance of ‘proof by association’, more studiously address methodological shortcomings in research planning and execution and seriously abide by long-available, well-informed, ad-mittedly challenging yet doable recommendations (e.g. Ellis, 1991; Hill & Knox, 2013; Russell et al., 1984). If supervision is to ever be evidence-based, then there is sore need for more, better and broader evidence.
ORCIDC. Edward Watkins https://orcid.org/0000-0001-9137-5526
R E FE R E N C E SAlfonsson, S., Parling, T., Spännargård, Å., Andersson, G., & Lundgren,
T. (2018). The effects of clinical supervision on supervisees and pa-tients in cognitive behavioral therapy: A systematic review. Cognitive
| 17WATKINS
Behaviour Therapy, 47, 206–228. https ://doi.org/10.1080/16506 073.2017.1369559
Bambling, M., King, R., Raue, P., Schweitzer, R., & Lambert, W. (2006). Clinical supervision: Its influence on client-rated working alliance and client symptom reduction in the brief treatment of major depression. Psychotherapy Research, 16, 317–331. https ://doi.org/10.1080/10503 30050 0268524
Barrett, J., Gonsalvez, C. J., & Shires, A. (2019). Evidence-based prac-tice within supervision during psychology practitioner training: A systematic review. Clinical Psychologist, https ://doi.org/10.1111/cp.12196 (early view, available online)
Bernard, J. M., & Goodyear, R. K. (2004). Fundamentals of clinical supervi-sion, 3rd ed. Upper Saddle River, NJ: Pearson.
Bernard, J. M., & Goodyear, R. K. (2014). Fundamentals of clinical supervi-sion, 5th ed. Upper Saddle River, NJ: Pearson.
Bernard, J. M., & Goodyear, R. K. (2019). Fundamentals of clinical supervi-sion, 6th ed. Upper Saddle River, NJ: Pearson.
Bogo, M., & McKnight, K. (2006). Clinical supervision in social work: A re-view of the research literature. The Clinical Supervisor, 24(1–2), 49–67. https ://doi.org/10.1300/J001v 24n01_04
Borders, L. D. (2006). Snapshot of clinical supervision in counseling and counselor education: A five-year review. The Clinical Supervisor, 24(1–2), 69–113. https ://doi.org/10.1300/J001v 24n01_05
Buus, N., & Gonge, H. (2009). Empirical studies of clinical supervision in psychiatric nursing: A systematic literature review and method-ological critique. International Journal of Mental Health Nursing, 18(4), 250–264. https ://doi.org/10.1111/j.1447-0349.2009.00612.x
Carifio, M. S., & Hess, A. K. (1987). Who is the ideal supervisor? Professional Psychology: Research and Practice, 18, 244–250. https ://doi.org/10.1037/0735-7028.18.3.244
Carpenter, J., Webb, C. M., & Bostock, L. (2013). The surprisingly weak evidence base for supervision: Findings from a systematic review of research in child welfare practice (2000–2012). Children and Youth Services Review, 35(11), 1843–1853. https ://doi.org/10.1016/j.child youth.2013.08.014
Cutcliffe, J. R., Sloan, G., & Bashaw, M. (2018). A systematic review of clinical supervision evaluation studies in nursing. International Journal of Mental Health Nursing, 27, 1344–1363. https ://doi.org/10.1111/inm.12443
Dawson, M., Phillips, B., & Leggat, S. (2013). Clinical supervision for allied health professionals: A systematic review. Journal of Allied Health, 42(2), 65–73.
Ducat, W. H., & Kumar, S. (2015). A systematic review of professional su-pervision experiences and effects for allied health practitioners work-ing in non-metropolitan health care settings. Journal of Multidisciplinary Healthcare, 8, 397–407. https ://doi.org/10.2147/JMDH.S84557
Ellis, M. V. (1991). Research in clinical supervision: Revitalizing a scientific agenda. Counselor Education and Supervision, 30(3), 238–251. https ://doi.org/10.1002/j.1556-6978.1991.tb012 04.x
Ellis, M. V. (2010). Bridging the science and practice of clinical su-pervision: Some discoveries, some misconceptions. The Clinical Supervisor, 29(1), 95–116. https ://doi.org/10.1080/07325 22100 3741910
Ellis, M. V., D’Iuso, N., & Ladany, N. (2008). State of the art in assessment, measurement, and evaluation of clinical supervision. In A. K. Hess, K. D. Hess, & T. A. Hess (Eds.), Psychotherapy supervision: Theory, re-search, and practice, 2nd ed. (pp. 473–499). Hoboken, NJ: Wiley.
Ellis, M. V., & Ladany, N. (1997). Inferences concerning supervisees and clients in clinical supervision: An integrative review. In C. E. Watkins (Ed.), Handbook of psychotherapy supervision (pp. 447–507). New York, NY: Wiley.
Ellis, M. V., Ladany, N., Krengel, M., & Schult, D. (1996). Clinical su-pervision research from 1981 to 1993: A methodological cri-tique. Journal of Counseling Psychology, 43(1), 35–50. https ://doi.org/10.1037/0022-0167.43.1.35
Falender, C. A., & Shafranske, E. P. (2017). Supervision essentials for the practice of competency based supervision. Washington, DC: American Psychological Association.
Farnan, J. M., Petty, L. A., Georgitis, E., Martin, S., Chiu, E., Prochaska, M., & Arora, V. M. (2012). A systematic review: The effect of clinical supervision on patient and residency education outcomes. Academic Medicine, 87, 428–442. https ://doi.org/10.1097/ACM.0b013 e3182 4822cc
Forshaw, G., Sabin-Farrell, R., & Schröder, T. (2019). Supervisors’ expe-rience of delivering individual clinical supervision to qualified thera-pists: A meta-ethnographic synthesis. Mental Health Review Journal, 24(1), 51–68. https ://doi.org/10.1108/MHRJ-09-2018-0028
Freitas, G. J. (2002). The impact of psychotherapy supervision on client out-come: A critical examination of 2 decades of research. Psychotherapy, 39, 354–367. https ://doi.org/10.1037/0033-3204.39.4.354
Goodyear, R. K., Borders, L. D. A., Chang, C. Y., Guiffrida, D. A., Hutman, H., Kemer, G., … White, E. (2016). Prioritizing questions and methods for an international and interdisciplinary supervi-sion research agenda: Suggestions by eight scholars. The Clinical Supervisor, 35, 117–154. https ://doi.org/10.1080/07325 223. 2016.1153991
Goodyear, R. K., Bunch, K., & Claiborn, C. D. (2006). Current supervision scholarship in psychology: A five year review. The Clinical Supervisor, 24(1-2), 137–147. https ://doi.org/10.1300/J001v 24n01_07
Goodyear, R. K., & Guzzardo, C. R. (2000). Psychotherapy supervision and training. In S. D. Brown, & R. W. Lent (Eds.), Handbook of counsel-ing psychology, 3rd ed. (pp. 83–108). New York, NY: Wiley.
Higgins, J. P., & Thomas, J. (Eds.) (2019). Cochrane handbook for systematic reviews of interventions. West Sussex, UK: Wiley.
Hill, C. E., & Knox, S. (2013). Training and supervision in psychotherapy. In M. J. Lambert (Ed.), Bergin and Garfield’s handbook of psychotherapy and behavior change, 6th ed. (pp. 775–811). Hoboken, NJ: Wiley.
Holloway, E. L. (1992). Supervision: A way of teaching and learning. In S. D. Brown, & R. W. Lent (Eds.), Handbook of counseling psychology, 2nd ed. (pp. 177–214). New York, NY: Wiley.
Holloway, E. L., & Hosford, R. E. (1983). Towards developing a prescrip-tive technology of counselor supervision. The Counseling Psychologist, 11(1), 73–77. https ://doi.org/10.1177/00110 00083 111012
Inman, A. G., Hutman, H., Pendse, A., Devdas, L., Luu, L., & Ellis, M. V. (2014). Current trends concerning supervisors, supervisees, and cli-ents in clinical supervision. In C. E. Watkins, & D. Milne (Eds.), Wiley international handbook of clinical supervision (pp. 61–102). Oxford, UK: Wiley.
Kühne, F., Maas, J., Wiesenthal, S., & Weck, F. (2019). Empirical research in clinical supervision: A systematic review and suggestions for fu-ture studies. BMC Psychology, 7(1), 1–11. https ://doi.org/10.1186/s40359-019-0327-7
Ladany, N., & Inman, A. G. (2012). Training and supervision. In E. M. Altmaier, & J. Hansen (Eds.), Oxford handbook of counseling psychology (pp. 179–207). Oxford, NY: Oxford University Press.
Ladany, N., & Malouf, M. A. (2010). Understanding and conducting su-pervision research. In N. Ladany, & L. J. Bradley (Eds.), Counselor su-pervision (pp. 353–388). New York, NY: Routledge.
Lambert, M. J., & Ogles, B. M. (1997). The effectiveness of psychother-apy supervision. In C. E. Watkins (Ed.), Handbook of psychotherapy supervision (pp. 421–446). New York, NY: Wiley.
MacDonald, J., & Ellis, P. M. (2012). Supervision in psychiatry: Terra incognita? Current Opinion in Psychiatry, 25, 322–326. https ://doi.org/10.1097/YCO.0b013 e3283 54ecc
Milne, D. (2007). An empirical definition of clinical supervision. British Journal of Clinical Psychology, 46(4), 437–447. https ://doi.org/10.1348/01446 6507X 197415
Milne, D. (2016). Guiding CBT supervision: How well do manuals and guidelines fulfil their promise? The Cognitive Behaviour Therapist, 9, https ://doi.org/10.1017/S1754 470X1 5000720
18 | WATKINS
Milne, D. L. (2018). Evidence-based CBT supervision: Principles and prac-tice, 2nd ed. Chichester, UK: Wiley.
Milne, D. L., & James, I. (2000). A systematic review of effective cog-nitive-behavioural supervision. British Journal of Clinical Psychology, 39(2), 111–127. https ://doi.org/10.1348/01446 65001 63149
Milne, D. L., Leck, C., James, I., Wilson, M., Procter, R., Ramm, L., & Weetman, J. (2012). High fidelity in clinical supervision research. In I. Fleming, & L. Steen (Eds.), Supervision and clinical psychology: Theory, practice and perspectives, 2nd ed. (pp. 142–158). London, UK: Routledge.
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G., and the PRISMA Group (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine, 151(4), 264–269. https ://doi.org/10.7326/0003-4819-151-4-20090 8180-00135 (PRISMA guidelines currently being updated; www.prisma-state ment.org/)
Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., … Stewart, L. A. (2015). Preferred reporting items for system-atic review and meta-analysis protocols (PRISMA-P) 2015 state-ment. Systematic Reviews, 4(1), 1. https ://doi.org/10.1136/bmj.g7647
Moher, D., Stewart, L., & Shekelle, P. (2015). All in the family: Systematic reviews, rapid reviews, scoping reviews, realist reviews, and more. Systematic Reviews, 4(1), 1. https ://doi.org/10.1186/s13643-015-0163-7
Mulrow, C. D. (1994). Systematic reviews: Rationale for system-atic reviews. BMJ, 309(6954), 597–599. https ://doi.org/10.1136/bmj.309.6954.597
Munn, Z., Peters, M. D., Stern, C., Tufanaru, C., McArthur, A., & Aromataris, E. (2018). Systematic review or scoping review? Guidance for authors when choosing between a systematic or scoping review approach. BMC Medical Research Methodology, 18(1), 143. https ://doi.org/10.1186/s12874-018-0611-x
Munn, Z., Stern, C., Aromataris, E., Lockwood, C., & Jordan, Z. (2018). What kind of systematic review should I conduct? A proposed typol-ogy and guidance for systematic reviewers in the medical and health sciences. BMC Medical Research Methodology, 18(1), 5. https ://doi.org/10.1186/s12874-017-0468-4
Newman, D. S., Simon, D. J., & Swerdlik, M. E. (2019). What we know and do not know about supervision in school psychology: A system-atic mapping and review of the literature between 2000 and 2017. Psychology in the Schools, 56, 306–334. https ://doi.org/10.1002/pits.22182
O'Donoghue, K., Ju, P. W. Y., & Tsui, M. (2018). Constructing an evi-dence-informed social work supervision model. European Journal of Social Work, 21, 348–358. https ://doi.org/10.1080/13691 457.2017.1341387
O’Donoghue, K., & Tsui, M. (2015). Social work supervision research (1970–2010): The way we were and the way ahead. British Journal of Social Work, 45, 616–633. https ://doi.org/10.1093/bjsw/bct115
Page, M. J., Shamseer, L., & Tricco, A. C. (2018). Registration of system-atic reviews in PROSPERO: 30,000 records and counting. Systematic Reviews, 7(1), 32. https ://doi.org/10.1186/s13643-018-0699-4
Park, E. H., Ha, G., Lee, S., Lee, Y. Y., & Lee, S. M. (2019). Relationship be-tween the supervisory working alliance and outcomes: A meta-anal-ysis. Journal of Counseling and Development, 97(4), 437–446. https ://doi.org/10.1002/jcad.12292
Pelling, N., Abbott, D., & Lack, C. W. (2017). The state of supervision scholarship in the 21st century. In N. Pelling, & P. Armsrtong (Eds.), The practice of counselling and clinical supervision, 2nd ed. (pp. 109–116). Samford Valley, Qld: Australian Academic Press.
Pott, T. (2018). How highly effective psychotherapy supervisors supervise: A longitudinal study of supervisees’ experiences (unpublished doctoral dissertation). Edmonton, CA: University of Alberta.
Rast, K. A., Herman, D. J., Rousmaniere, T. G., Whipple, J. L., & Swift, J. K. (2017). Perceived impact on client outcomes: The perspectives of practicing supervisors and supervisees. SAGE Open, 7(1), https ://doi.org/10.1177/21582 44017 698729
Reiser, R. P., & Milne, D. (2012). Supervising cognitive-behavioral psychotherapy: Pressing needs, impressing possibilities. Journal of Contemporary Psychotherapy, 42(3), 161–171. https ://doi.org/10.1007/s10879-011-9200-6
Reiser, R. P., & Milne, D. L. (2014). A systematic review and reformulation of outcome evaluation in clinical supervision: Applying the fidelity framework. Training and Education in Professional Psychology, 8(3), 149–157. https ://doi.org/10.1037/tep00 00031
Rønnestad, M. H., Orlinsky, D. E., Schröder, T. A., Skovholt, T. M., & Willutzki, U. (2019). The professional development of counsellors and psychotherapists: Implications of empirical studies for supervi-sion, training and practice. Counselling and Psychotherapy Research, 19, 214–230. https ://doi.org/10.1002/capr.12198
Russell, R. K., Crimmings, A. M., & Lent, R. W. (1984). Counselor training and supervision: Theory and research. In S. D. Brown, & R. W. Lent (Eds.), Handbook of counseling psychology (pp. 625–681). New York, NY: Wiley.
Scaife, J. (2019). Supervision in clinical practice: A practitioner's guide, 3rd ed. London, UK: Routledge.
Schutt, M. A. (2012). Replication and extension of Ellis, Ladany, Krengel, and Shult (1996); Clinical Supervision and Research from 1981 to 1993: A Methodological Critique (unpublished doctoral dissertation). Bethlehem, PA: Lehigh University.
Sewell, K. M. (2018). Social work supervision of staff: A primer and scop-ing review (2013–2017). Clinical Social Work Journal, 46, 252–265. doi.10.1007/s10615-018-0679-0.
Shamseer, L., Moher, D., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., … Stewart, L. A. (2015). Preferred reporting items for systematic re-view and meta-analysis protocols (PRISMA-P) 2015: Elaboration and explanation. BMJ, 349, g7647. https ://doi.org/10.1136/bmj.g7647
Simpson-Southward, C., Waller, G., & Hardy, G. E. (2017). How do we know what makes for “best practice” in clinical supervision for psy-chological therapists? A content analysis of supervisory models and approaches. Clinical Psychology & Psychotherapy, 24, 1228–1245. https ://doi.org/10.1002/cpp.2084
Snowdon, D. A., Leggat, S. G., & Taylor, N. F. (2017). Does clinical su-pervision of healthcare professionals improve effectiveness of care and patient experience? A systematic review. BMC Health Services Research, 17(1), 786. https ://doi.org/10.1186/s12913-017-2739-5
Tsui, M. S. (1997). Empirical research on social work supervision: The state of the art (1970–1995). Journal of Social Service Research, 23(2), 39–54. https ://doi.org/10.1300/J079v 23n02_03
Wampold, B. E., & Holloway, E. L. (1997). Methodology, design, and eval-uation in psychotherapy supervision research. In C. E. Watkins (Ed.), Handbook of psychotherapy supervision (pp. 11–27). New York, NY: Wiley.
Watkins, C. E. Jr (2011). Does psychotherapy supervision contribute to patient outcomes? Considering thirty years of research. The Clinical Supervisor, 30(2), 235–256. https ://doi.org/10.1080/07325 223.2011.619417
Weerasekera, P. (2013). The state of psychotherapy supervision: Recommendations for future training. International Review of Psychiatry, 25(3), 255–264. https ://doi.org/10.3109/09540 261.2013.769431
Wheeler, S., & Richards, K. (2007). The impact of clinical supervision on counselors and therapists, their practice and their clients: A system-atic review of the literature. Counselling & Psychotherapy Research, 7, 54–65. https ://doi.org/10.1080/14733 14060 1185274
White, E. (2017). Claims to the benefits of clinical supervision: A critique of the policy development process and outcomes in New South
| 19WATKINS
Wales, Australia. International Journal of Mental Health Nursing, 26(1), 65–76. https ://doi.org/10.1111/inm.12292
Wilson, H. M. N., Davies, J. S., & Weatherhead, S. (2016). Trainee ther-apists’ experiences of supervision during training: A meta-synthe-sis. Clinical Psychology & Psychotherapy, 23(4), 340–351. https ://doi.org/10.1002/cpp.1957
AUTHOR BIOG R APHY
C. Edward Watkins, Jr., Ph.D., is Professor of Psychology at the University of North Texas. His primary professional interests focus on psychotherapy supervision and integrative and psy-choanalytic theories, practice, and research. He is editor of the Handbook of Psychotherapy Supervision (1997), co-editor (with Derek Milne) of the Wiley International Handbook of Clinical Supervision (2014), and volume coordinator (with Loredana-Ileana Vîşcu) of Metode si Tehnici de Interventie in Consilierea si Psihoterapia Copilului si Adolescentului (2019). He currently serves as Associate Editor (the Editor being Dr. Vîşcu) of the International Journal of Supervision in Psychotherapy. He is a Fellow of Divisions 29 (Psychotherapy) and 17 (Counseling Psychology) of the American Psychological Association.
How to cite this article: Watkins CE. What do clinical supervision research reviews tell us? Surveying the last 25 years. Couns Psychother Res. 2019;00:1–19. https ://doi.org/10.1002/capr.12287
View publication statsView publication stats