Theeboom Et Al - Does Coaching Work - Ameta-Analysis
Theeboom Et Al - Does Coaching Work - Ameta-Analysis
Theeboom Et Al - Does Coaching Work - Ameta-Analysis
To cite this article: Tim Theeboom, Bianca Beersma & Annelies E.M. van Vianen (2014) Does coaching work? A meta-analysis
on the effects of coaching on individual level outcomes in an organizational context, The Journal of Positive Psychology:
Dedicated to furthering research and promoting good practice, 9:1, 1-18, DOI: 10.1080/17439760.2013.837499
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
The Journal of Positive Psychology, 2013
Vol. 9, No. 1, 1–18, http://dx.doi.org/10.1080/17439760.2013.837499
Does coaching work? A meta-analysis on the effects of coaching on individual level outcomes
in an organizational context
Tim Theeboom*, Bianca Beersma and Annelies E.M. van Vianen
Department of Work and Organizational Psychology, University of Amsterdam, Weesperplein 4, 1018 XA, Amsterdam,
The Netherlands
(Received 12 February 2013; accepted 15 August 2013)
Whereas coaching is very popular as a management tool, research on coaching effectiveness is lagging behind.
Moreover, the studies on coaching that are currently available have focused on a large variety of processes and outcome
measures and generally lack a firm theoretical foundation. With the meta-analysis presented in this article, we aim to
shed light on the effectiveness of coaching within an organizational context. We address the question whether coaching
has an effect on five both theoretically and practically relevant individual-level outcome categories: performance/skills,
well-being, coping, work attitudes, and goal-directed self-regulation. The results show that coaching has significant posi-
tive effects on all outcomes with effect sizes ranging from g = 0.43 (coping) to g = 0.74 (goal-directed self-regulation).
Downloaded by [Juan Cortes] at 14:05 24 March 2014
These findings indicate that coaching is, overall, an effective intervention in organizations.
Keywords: coaching; coaching effectiveness; coaching interventions; coaching outcomes; meta-analysis
the effectiveness of coaching as a change methodology, more recent literatures have gradually moved towards a
it also has some serious limitations which we will generally accepted definition of coaching. Kilburg (1996)
discuss below. originally defined coaching of executives in organiza-
With the meta-analysis we present in this article, we tions as ‘a helping relationship between a managerial-
aim to provide a comprehensive quantitative review and client and a consultant that follows a formally defined
to answer the question: Does coaching1 work when coaching agreement’. Grant (2003) transformed this
provided in an organizational context by professionally definition into a more general one and defined coaching
trained coaches? Furthermore, by means of a systematic as a ‘result-oriented, systematic process in which the
review and integration of the types of coaching outcomes coach facilitates the enhancement of life experience and
that were included in prior studies, we aim to give an goal-attainment in the personal and/or professional life of
initial impetus to further theoretical development of normal, non-clinical clients’ (p. 254).
coaching research. That is, by organizing coaching This latter definition encompasses several important
outcomes that emerge from the literature into meaningful features, namely: it can be applied to a multitude of
categories, we aim to enable future studies to build on or coaching domains (e.g. personal coaching and organiza-
extend theory that explains the paths and processes that tion coaching) and coaches (executive and non-execu-
lead to these different categories of outcomes. tive), it emphasizes the self-directedness of the coaching
process, and it recognizes coaching as a systematic
process rather than just being empathic and ‘having good
Coaching conversations’ (Leonard-Cross, 2010). Furthermore,
Downloaded by [Juan Cortes] at 14:05 24 March 2014
Coaching has its roots in a multitude of disciplines, although the differences and similarities between coach-
including philosophy, sociology, anthropology, sports, ing and therapy are still a topic of debate (Bono et al.,
communication science and even natural sciences (Brock, 2009; Brunning, 2006; Hart, Blattner, & Leipsic, 2001),
2008). However, in terms of the number of articles Grant’s (2003) definition of coaching distinguishes
published in peer-reviewed journals, sub-disciplines of coaching from therapy by its focus on a non-clinical
psychology have shown to be the most fruitful areas of population.
research on coaching (Grant, Passmore, et al., 2010).
Initially, most of the research on coaching was conducted
within areas such as sports psychology (e.g. Gallwey, Coaching effectiveness: beyond return of investment
1974; Whitmore, 1992) and clinical psychology (e.g. measures
Berg & Szabo, 2005; De Shazer, 1988). More recently, The literature on coaching has grown exponentially in
research on coaching has particularly flourished in two the last 15 years. Whereas only 93 articles were
strongly related areas of psychology that emerged within published in the years between 1937 and 1999, the total
the last two decades: positive psychology (Seligman & number of articles and dissertations on coaching reached
Csikszentmihalyi, 2000) and coaching psychology 634 in 2011 (Grant, 2013) and the number has been stea-
(Passmore, 2010). dily growing ever since. However, the bulk of articles
Positive psychology focuses on studying three aspects still consists of descriptive papers and/or case studies as
that constitute the scientific notion of happiness: positive well as practitioner articles primarily aimed at emphasiz-
emotion, meaning, and engagement (Seligman, 2007). ing the benefits of certain coaching interventions (De
Coaching psychology focuses on studying behavior, Meuse et al., 2009).
cognition, and emotion within coaching practice to deepen Additionally, this predominantly practitioner-
understanding of coaching processes and to enhance generated research has resulted in an overemphasis on
coaching techniques (Passmore, 2010). While there are ROI measures. ROI as a measure of coaching effective-
some debates about how these areas of research are related ness is appealing because it provides some direct insight
to each other (some authors see coaching psychology as a into the tangible benefits of coaching interventions
sub-discipline or an applied form of positive psychology; (Fillery-Travis & Lane, 2006; Leonard-Cross, 2010).
e.g. Grant & Cavanagh, 2007), both areas share their However, the ROI measure has some serious limitations.
focus on performance enhancement, positive aspects of For instance, the factors included in the most frequently
human nature, and the strengths of individuals (Linley & used calculation of the ROI metric (benefits – costs/
Harrington, 2005). Therefore, positive psychology seems costs × 100) are highly idiosyncratic and tend to ignore
to offer a robust framework for researching coaching and context variables such as team input. Therefore, it is
as such may constitute ‘one of the solutions to the lack of often impossible to determine the degree to which the
a theoretical framework in the coaching field’ (Freire, financial benefits can be directly attributed to the coach-
2013, p. 428) ing intervention (Grant, Passmore, et al., 2010, p. 26).
While different psychological sub-disciplines initially Additionally, performance measures (direct benefits) are
developed their own specific conceptual framework, seldom available and a narrow focus on ROI and other
The Journal of Positive Psychology 3
performance-related measures neglects other – more several studies have shown that coaching can positively
indirect – ways in which organizations could potentially influence goal-attainment expectancy (Evers, Brouwers,
benefit from coaching, such as employee well-being and & Tomic, 2006; Moen & Skaalvik, 2009) and goal-
health. Therefore, the current meta-analysis investigates progression and commitment (Green, Oades, & Grant,
coaching effectiveness by looking at well-validated, more 2006). Because the relationship between goal-setting on
distal indicators of functioning in addition to perfor- the one hand and motivation and performance on the
mance measures: well-being, coping, work and career- other hand is well established (Locke & Latham, 2002),
related attitudes, and goal-directed self-regulation. improving employees’ self-regulation by enhancing their
The overall goal of coaching in a work context is to ability to set and strive for goals is yet another way in
optimize a person’s work-related functioning. First, which organizations could benefit from coaching.
individuals in organizations function better if they feel All in all, coaching could benefit organizations by
well, that is, if their basic needs are fulfilled (Deci & enhancing employees’ performance and skills, well-
Ryan, 1985, 2000) and if they do not struggle with being, coping, work attitudes, and goal-directed self-
health-related problems as caused by their job (Burton regulation. Because extant empirical studies (see Method
et al., 1999). With regard to the latter, work-related stress section below) lack a clear conceptual framework for
affects over 20% of workers in the European Union classifying coaching outcomes, we categorized available
(Brunn & Milczarek, 2007) which costs organizations outcomes into these outcome dimensions, which are both
about 20,000 million Euros per year. Indeed, research theoretically and practically relevant and also well
has evidenced negative relationships between ‘soft’ established within the psychological literature.
Downloaded by [Juan Cortes] at 14:05 24 March 2014
coaching, effects of coaching, and outcomes of coaching. Studies conducted in an educational context (e.g.
Initially, we searched for articles (i) that included undergraduate students) were included because the cli-
quantitative data on the effects of coaching; (ii) in which ents’ characteristics and needs are similar to those of
coaching was provided by professionally trained external clients in organizational settings, that is, they are simi-
coaches or trained peers; and (iii) in which the coaches lar with respect to their demographics and the chal-
belonged to a non-clinical population. Second, we lenges they face (working in teams, meeting deadlines
performed both backward and forward searches on the etc.). Finally, we excluded all studies in which not
studies we retrieved in phase 1. Third, we contacted sev- enough statistical information was available or could
eral scholars known to be active in the field of coaching not be made available after contacting the correspond-
psychology research in order to retrieve (yet) unpub- ing authors.
lished results. At the same time, we sent out a request This selection process resulted in a total of 18 studies
for published and unpublished studies via the mailing-list included in the final analysis. All studies that were
services of Academy of Management, the Society for included in the final analysis are indicated with an * in
Industrial and Organizational Psychology, and the our list of references. An overview of basic characteris-
European Association of Social Psychology. tics of these studies is displayed in Table 1.
In the fourth phase, we screened the 107 full articles
retrieved in the first three phases and excluded all cross-
sectional studies since these do not allow controlling for Outcome categorization
most threats to internal validity (Cook & Campbell, The categorization of outcomes was conducted in three
Downloaded by [Juan Cortes] at 14:05 24 March 2014
1979). Furthermore, we excluded all studies in which the steps. In the first step, the first and the second authors
authors did not perform quantitative analyses (e.g. case mutually assigned all study outcomes into one of the
studies). In the fifth and final phases, we used a priori following categories: performance/skills, well-being,
conceptual and methodological criteria to decide which coping, work attitudes, or goal-directed self-regulation
of these studies would be included in the final analysis. while the third author categorized the outcomes inde-
For example, we excluded all studies in which the pendently of the first two authors. In the second step,
described coaching process did not match our definition the interrater agreement was calculated. The interrater
of coaching or for which a description of the coaching agreement (Cohens’ κ) was 0.80, which is considered
process could not be obtained. to be large (see Landis & Koch, 1977). In the third
In addition to Grant’s (2003) definition of coaching, step, the authors discussed their discrepancies and
we took one more characteristic of coaching into account agreed on a final categorization. Several studies in the
when selecting studies, namely that the coaching was meta-analysis included multiple measures within the
provided by a (professionally trained) coach with no same outcome category (e.g. measures of stress and
formal authority over the coachee. The main reason for burnout both fall within the well-being category). An
this is that research on mentoring has shown that a average of these effect sizes was included in order to
mentor’s formal authority over the protégée can affect prevent violation of the independent sample assumption
the way in which the protégée behaves during coaching (Hunter & Schmidt, 2004).
sessions (Mullen, 1994; Tepper, 1995; Waldron, 1991). The performance/skills category includes both subjec-
Furthermore, the goals in managerial coaching (in which tive and objective outcome measures that either directly
a manager coaches an employee) are often strongly reflect performance (e.g. number of sales, supervisory
related to organizational performance (Cox, Bachkirova, rated job performance) or reflect the demonstration of
& Clutterbuck, 2009) which might influence the degree behaviors needed for an organization to be effective (e.g.
to which the coaching process is self-directed. Another transformational leadership behaviors). The well-being
reason for excluding studies on managerial coaching is category includes both subjective and objective outcome
that our aim for this research was to set a first step for measures that are a direct representation of peoples’
answering the question whether the high costs of hiring well-being, health, need fulfillment, and affective
external professionally trained coaches are justifiable for responses. Examples of these are measures of psychopa-
both individual clients and organizations. Thus, we thology (e.g. Depression Anxiety and Stress Scale;
excluded all studies in which the coach had a formal Lovibond & Lovibond, 1995) and burnout (e.g. Maslach
authority over the coachee. Burnout Inventory; Maslach & Jackson, 1986).
We also excluded all studies in which the influence The coping category includes outcome measures
of other interventions (e.g. leadership development pro- related to the ability to deal with present and future job
grams) could not be ruled out as a confounding factor. demands and stressors. Examples of these are measures
Because we aim to provide more insight into the use- of self-efficacy (e.g. General Self-Efficacy Scale;
fulness of coaching in organizations, we only included Schwarzer & Jerusalem, 1995) and mindfulness (e.g.
studies conducted in a work or educational context. Mindful Attention Awareness Scale; Brown & Ryan,
The Journal of Positive Psychology 5
Table 1. Study characteristics and outcome overview of studies included in the meta-analysis.
Nr.
Study n Intervention sessions Outcomes Design
Bozer and Sarros (2012) 96 Cognitive behavioral 11 Self-reports RCT
coaching
Career satisfaction
Job commitment
Job Performance
Self-Awareness
Supervisory ratings
Job performance
Self-awareness
Task performance
Egan and Song (2005) 103 Coaching on goal-setting Unknown Self-reports QEF
and goal achievement
Job satisfaction
Organizational commitment
Performance goal orientation
Performance rating
Supervisory ratings
Performance rating
Developmental planning
Developmental support
Openness to new behaviors
Positive affect
Self-efficacy
Anxiety
Depression
GAS
Insight
Self-reflection
Stress
Quality of life
Anxiety
Cognitive hardiness
Depression
GAS
Learning
Personal insight
Well-being
(Continued)
6 T. Theeboom et al.
Table 1. (Continued).
Nr.
Study n Intervention sessions Outcomes Design
Grant et al. (2009) 41 CB-SFC 10 Self-reports RCT
Anxiety
Depression
Gas
Resilience
Stress
Well-being
Work
Anxiety
Cognitive hardiness
Hope
Depression
Stress
Agency
Autonomy
Environmental mastery
Goal striving
Negative affect
Personal growth
Positive affect
Positive relations with others
Purpose in life
Satisfaction with life
Self-acceptance
Total hope
(Continued)
The Journal of Positive Psychology 7
Table 1. (Continued).
Nr.
Study n Intervention sessions Outcomes Design
Organizational commitment
Satisfaction with coworkers
Satisfaction with supervision
Satisfaction with work
Self-insight scale
Self-reflection scale
Stress
Turnover intentions
Quality of Life Inventory
Supervisory ratings
Effectiveness
Ratings by coach
Effectiveness
(Continued)
8 T. Theeboom et al.
Table 1. (Continued).
Nr.
Study n Intervention sessions Outcomes Design
Elaboration of supervisory
ratings
Performance
Rating by peers
Performance
Supervisory ratings
Performance
Rating by authors
Goal specificity
Autonomy
Environmental mastery
Goal commitment
Goal progression
Negative affect
Personal Growth
Positive affect
Positive relations with others
Purpose in life
Self-acceptance
Subjective well being
Notes: RCT = randomized control trial; QEF = quasi-experimental field study in which participants were non randomly allocated to experimental and
control groups; WSD = within-subjects design without control group, which includes both pre- and post-intervention measures; CB-SFC = cognitive
behavioral solution-focused coaching; and GAS = goal-attainment scaling.
a
Only the C-MT condition (see original) was included for the calculation of effect sizes.
2003). The work attitudes category includes outcome goal-attainment, and goal-evaluation. This category also
measures related to cognitive, affective, and behavioral includes goal-attainment scaling (GAS) measures, which
responses toward work and career, such as job are increasingly popular in coaching settings (see Spence
satisfaction (e.g. Job Description Index; Smith, Kendall, [2008] and Peterson [1993] for overviews of the use of
& Hulin, 1969), organizational commitment (e.g. Organi- GAS in coaching research).
zational Commitment Scale; Porter, Steers, Mowday &
Boulian, 1974), and career satisfaction (e.g. Career
Satisfaction Scale; Greenhaus, Parasuraman & Wormley, Calculating the effect sizes
1990). Finally, the goal-directed self-regulation category One of the most challenging steps in a meta-analysis is
includes all outcome measures related to goal-setting, combining the effect sizes of different studies in one
The Journal of Positive Psychology 9
analysis (McGaw & Glass, 1980). Since effect sizes After selecting the general approach for the meta-
based on means are easily interpretable and the studies analysis, the statistical model for the meta-analysis has
in our analysis employed a large variety of outcome to be designated. In terms of the model for the meta-
measures to assess the impact of coaching interventions, analysis, the (conservative) random-effect model was
the use of an effect size index based on standardized adopted as recommended by the National Research
means was the logical choice (Borenstein, Hedges, Council (1992). As opposed to the fixed effect model,
Higgins, & Rothstein 2009; Cohen, 1988). Since the the random-effect model allows that the true effect size
most popular index, Cohen’s d, tends to overestimate the varies from study to study based on both the variability
population effect size when small samples are included of the independent variable (e.g. intensity or duration of
in the analysis, we chose to use Hedges g which can still intervention) and differences in the samples of the
be interpreted as the mean difference expressed in research population such as age, educational background,
standard deviation units and applies a simple correction and type of job of the coaches (Borenstein et al., 2009;
to overcome this bias (Hedges, 1981). Hedges & Cooper, 1994).
Effect sizes can be defined in relation to pre-interven- Heterogeneity between studies was quantified by an
tion scores, post-intervention scores, or difference scores. assessment of both the classical Cochran Q statistic
In theory, it is possible to choose either definition (1954) and the I2 statistic as proposed by Higgins and
because effect sizes can be transformed into a common Thompson (2002), see also Higgins, Thompson, Deeks,
effect size index by using the correlation between pre- & Altman, 2003). While the Q statistic serves as a test
and post-intervention scores to estimate the sampling of significance for between-study heterogeneity, the value
Downloaded by [Juan Cortes] at 14:05 24 March 2014
bias (Morris & DeShon, 1997). Unfortunately, the corre- for I2 represents the proportion of between-study vari-
lations between pre- and post-intervention scores were ance in effect sizes that can be attributed to between-
seldom provided and the amount of studies that used a study heterogeneity rather than within-study variability
mixed design outnumbered the amount of studies that (Borenstein et al., 2009). When the value for I2 is large
used a within-subject design. Therefore, effect sizes (see Higgins & Thompson, 2002 for some guidelines on
based on the pooled (experimental and control groups) interpretation), one of the possible explanations for this
standard deviations of the post-intervention scores were is the existence of moderating variables. In what should
chosen as the referent effect size index. By doing so, the be considered exploratory analyses due to the relatively
estimation of parameters was minimized since only small amount of studies, we explored the influence of
the pre-intervention/post-intervention correlations for the two of these potential (methodological) moderating vari-
(minority of) studies that used a within-subject design ables in order to provide guidance for the future method-
needed to be estimated2. Finally, effect sizes based on ological approach of coaching research. First, following
post-intervention standard deviations are likely to be the example of Fusar-Poli et al. (2012), we conducted
slightly biased downward, and are thus the most conser- subgroup analyses for sets of studies that were different
vative choice (Carlson & Schmidt, 1999). in terms of study design (mixed designs vs. within-
subject designs). Second, we examined the influence of
the number of coaching sessions using meta-regression
Meta-analytic procedure and statistical analyses analysis. Finally, following recent recommendations by
The Hedges and Olkin (1985) approach to meta-analysis Sterne et al. (2011), we assessed the risk for publication
was used to calculate the effect sizes. Comparisons with bias (or small study bias) by visually inspecting funnel
other commonly applied methods such as by Hunter and plots and by applying the regression intercept of Egger,
Schmidt (1990) and Rosenthal (1991) suggest that differ- Smith, Schneider, and Minder (1997).
ences between the Hedges and Olkin methods and the
other methods are relatively small and only apply under
very specific circumstances. If anything, the Hedges and Software for the analysis and statistical corrections
Olkin method can be considered to be the most conser- The software that was used for the analysis was
vative approach because it does not allow for the statisti- comprehensive meta-analysis. CMA was developed by
cal corrections for artifactual sources of variance (e.g. Borenstein, Hedges, Higgins, and Rothstein (2005) and
measurement error and restriction of range) that tend to is based on the Hedges and Olkin’s (1985) approach to
result in an inflation of effect size estimates (Borenstein meta-analysis. CMA offers advantages over other soft-
et al., 2009). Additionally, the Hedges and Olkin ware in terms of its flexibility to handle multiple data
approach seems to provide the most conservative esti- entry formats (e.g. data from within-subject designs and
mate of the (lower limit of) confidence intervals (John- mixed designs) and its intuitive approach to sensitivity
son, Mullen, & Salas, 1995), which can be used for analysis and the detection of between-study heterogene-
determining the statistical significance of effect sizes. ity (Borenstein et al., 2009).
10 T. Theeboom et al.
CI (95%)
Study n Sessions Study design g Lower Upper p-value
Bozer and Sarros (2012) 96 11 RCT 0.36 −0.12 0.83 0.145
Cerni et al. (2010) 14 10 QEF 0.08 −0.15 0.30 0.501
Egan and Song (2005) 103 Unknown QEF 0.69 0.29 1.08 0.001
Finn (2007) 32 6 QEF 1.17 0.34 1.99 0.006
Grant (2003) 20 10 WSD 0.84 0.31 1.38 0.002
Grant (2008) 29 5 WSD 0.71 0.23 1.19 0.004
Grant et al. (2009) 41 10 RCT 0.63 0.00 1.25 0.049
Grant et al. (2010) 65 10 RCT 0.25 −0.34 0.83 0.406
Green, Grant, and Rynsaardt (2007) 44 10 RCT 0.61 0.02 1.20 0.042
Green et al. (2006) 50 9 RCT 0.70 0.13 1.28 0.016
Kochanowski et al. (2010) 84 10 QEF 0.25 −0.16 0.66 0.239
Luthans and Peterson (2004) 67 1 WSD 0.74 0.47 1.01 0.000
Moen and Skaalvik (2009) 19 4 QEF 0.82 −0.10 1.75 0.080
Peterson (1993) 100 50 WSD 2.33 1.93 2.72 0.000
Poepsel (2011) 28 4 RCT 0.92 0.16 1.70 0.018
Smither et al. (2003) 1243 3 QEF 0.13 0.00 0.27 0.049
Spence et al. (2008) 15a 4 RCT 0.10 −0.39 0.58 0.695
Spence and Grant (2007) 40 10 RCT 0.29 −0.33 0.91 0.355
Total 2090 – – 0.66 0.39 0.93 0.000
Notes: N = analyzed sample size; g = Hedges’ g; CI (95%) = 95% confidence interval for g; RCT = randomized control trial; QEF = quasi-experimen-
tal field-study in which participants were non randomly allocated to experimental and control groups; and WSD = within-subjects design without con-
trol group, which includes both pre- and post-intervention measures.
a
Only the C-MT condition (see original) was included for the calculation of effect sizes.
The Journal of Positive Psychology 11
CI (95%)
k N g Lower Upper p Q I2
Performance/skills 6 2007 0.60 0.04 1.16 0.036 112.24* 95.55
Well-being 10 564 0.46 0.28 0.62 0.000 7.72 0.00a
Coping 10 1703 0.43 0.25 0.61 0.000 6.36 43.01
Attitudes 7 507 0.54 0.34 0.73 0.000 8.64 30.51
Self-regulation 11 789 0.74 0.42 1.06 0.000 81.27* 53.38
Notes: k = number of studies included in the analysis; N = total sample size in k studies; g = Hedges’ g; CI = 95% random effects confidence intervals;
and Q = Cochran Q statistic. I2 = the proportion of total variation in the estimates of treatment effect that is due to heterogeneity between studies.
*
indicates that between-study heterogeneity significant at α = 0.000.
a
The I2 was truncated to zero because the Q statistic used for the computation of I2 was smaller than its degrees of freedom.
The weighted point estimates of the effect sizes seem sessions. The choice for the comparison of studies
to indicate that the effect sizes of studies that used a with less or equal to five sessions vs. more than five
within-subject design are larger than the effect sizes of sessions emerged from the available study data with
the studies that used an independent-group design. Sub- some studies reporting a maximum of five coaching
group analysis indicated that the overall effect size sessions whereas other studies reporting more than this
Downloaded by [Juan Cortes] at 14:05 24 March 2014
Table 4. Weighted effect sizes of coaching interventions on all outcome categories for different study designs.
CI (95%)
k N g Lower Upper p Q I2
Performance/skills
IDG 5 1372 0.18 0.04 0.32 0.013 5.51 27.43
Within-subjects 1 100 2.33 1.93 2.72 0.000 –b –b
Well-being
IDG 7 394 0.39 0.18 0.60 0.000 3.91 0.00a
Within-subjects 3 116 0.54 0.26 0.82 0.002 3.12 35.94
Coping
IDG 7 325 0.26 0.15 0.37 0.000 11.67 48.57
Within-subjects 3 116 0.47 0.20 0.74 0.001 2.16 7.45
Attitudes
IDG 6 169 0.48 0.29 0.67 0.000 6.23 19.77
Within-subjects 1 67 0.74 0.47 1.01 0.000 –b –b
Self-regulation
IDG 9 1565 0.32 0.21 0.43 0.000 37.37* 78.59
Within-subjects 2 116 1.33 0.85 1.81 0.035 –b 0.00a
Notes: k = number of studies included in the analysis; N = total sample size in k studies; g = Hedges’ g; CI = 95% random effects confidence intervals;
Q = Cochran Q statistic; and I2 = the proportion of total variation in the estimates of treatment effect that is due to heterogeneity between studies.
*
Indicates that between-study heterogeneity significant at α = 0.000.
a
The I2 was truncated to zero because the Q statistic used for the computation of I2 was smaller then it’s degrees of freedom.
b
The statistics could not be computed because not enough data was available.
12 T. Theeboom et al.
Table 5. Weighted effect sizes of coaching interventions for studies that differ in terms of the number of coaching sessions.
CI (95%)
k N g Lower Upper Q I2
Performance/skills
≤5 sessions 3 1329 0.26 0.02 0.53 0.00 58.57
>5 sessions 2 110 0.11b −0.09 0.32 0.00 0.00a
Well-being
≤5 sessions 4 195 0.47 0.15 0.79 5.83 38.36
>5 sessions 7 296 0.46 0.22 0.69 0.00 0.00a
Coping
≤5 sessions 5 1429 0.35 0.11 0.59 0.00 50.33
>5 sessions 5 182 0.54 0.30 0.79 0.00 0.00a
Attitudes
≤5 sessions 3 126 0.67 0.41 0.94 19.14 44.39
>5 sessions 4 150 0.35 0.08 0.61 0.75 0.00a
Self-regulation
≤5 sessions 6 1457 0.52 0.15 0.88 6.962 79.94
>5 sessions 5 215 1.02 0.67 1.36 0.00 32.18
Notes: k = number of studies included in the analysis; g = Hedges’ g; CI = 95% random effects confidence intervals; Q = Cochran Q statistic; and
I2 = the proportion of total variation in the estimates of treatment effect that is due to heterogeneity between studies.
Downloaded by [Juan Cortes] at 14:05 24 March 2014
a
The I2 was truncated to zero because the Q statistic used for the computation of I2 was smaller then it’s degrees of freedom.
b
One study reported an average of 50 coaching sessions and was therefore considered an outlier and excluded from the meta-regression analysis.
forms of therapy and coaching by its premise that there is interventions, individual characteristics of the coach and
no need for an extensive analysis and understanding of the coachee, and the relationship between the coach and
problems in order to create solutions (Berg & Szabo, the coachee. We provide some specific suggestions for
2005; Grant & O’Connor, 2010). Therefore, it is possible each of these areas below.
to jump directly to the ultimate aim of coaching, namely Research concerning the design of coaching interven-
the identification of solutions, potentially resulting in a tions may benefit from the literature on training and
smaller number of sessions needed to make progression mentoring which draws heavily on educational psychol-
(Kim, 2008). Future research could investigate whether ogy and theories on (adult) learning (e.g. theory on
solution-focused coaching is indeed more effective than transformative learning; Mezirow, 1991). Theories on
other coaching approaches and whether specific coaching adult learning and its underlying mechanisms can
effects also depend on significance and/or complexity of provide insights that are relevant for increasing the
coaches’ problems. ‘transfer of coaching’ (i.e. long-term effectiveness of
Although the results should be interpreted with coaching interventions). Furthermore, Spence and Oaedes
caution because of the exploratory nature of the analy- (2011) have suggested that Deci and Ryans’ (1985) self-
sis, the finding that coaching can be effective even determination theory (SDT) is a valuable theoretical
when the number of coaching sessions is relatively framework for future research on the design of coaching
small is encouraging for organizations and individuals interventions. Central constructs of SDT such as goal-
in need of coaching. However, while the difference in setting, intrinsic motivation and the human needs of
the number of sessions does not seem to impact the competence, relatedness, and autonomy, are crucial for
Downloaded by [Juan Cortes] at 14:05 24 March 2014
mean effect size, the examination of the heterogeneity facilitating durable change within coaches (Ryan, Lynch,
statistics does show that there is less variability in the Vansteenkiste, & Deci, 2011).
effect sizes for studies using a larger amount of ses- Research concerning the characteristics of coaches
sions. In other words, the robustness of the effects of may find a valuable starting point in the therapy litera-
coaching seems to increase with the number of coach- ture. For example, studies investigating the personal
ing sessions. This finding corroborates research on adult characteristics of effective therapists have shown that
learning which suggests that deeper levels of learning individual characteristics such as (perceived) empathy
(e.g. transformative learning; Mezirow, 1991) only are important predictors of therapy outcomes (Burns &
occur when there are sufficient opportunities for critical Nolen-Hoeksema, 1992; Elliott, Bohart, Watson, &
reflection and active experimentation. Greenberg, 2011). Recent work in the field of executive
coaching indeed suggests that non-specific factors such
as understanding, encouraging, and listening behaviors
Future research: a need for theoretical enrichment of the coach may be better predictors of coaching effec-
It is our hope that future research will not only con- tiveness than specific factors such as the coaching meth-
tinue to examine whether coaching is effective, but also odology (de Haan, Culpin, & Curd, 2011). In this light,
respond to the need for more theoretical development the influence of constructs related to coaches’ ability to
in coaching psychology. A strong theoretical framework perceive and manage the emotional states of coaches,
is expedient from both an empirical and a practical such as emotional intelligence (Salovey & Mayer, 1989),
perspectives (Grant, 2010; Spence & Oaedes, 2011). seems especially relevant to examine in future research.
For scholars working in the field of coaching psychol- Research concerning the characteristics of coaches
ogy, a strong theoretical foundation could purposefully may explore the concept of ‘coachability’ that originates
guide the construction of cumulative knowledge. For in the sports psychology literature. Coachability is a
practitioners, insight into how (rather than if) coaching multidimensional construct that reflects the combination
works can provide guidelines for the improvement of of personality traits (e.g. agreeableness, openness to
extant coaching interventions and the development of experience) and motivational components (e.g. achieve-
new interventions. ment motivation) needed to improve functioning and
One way in which theoretical enrichment of the performance (Giacobbi, 2000). Furthermore, therapy
coaching literature could be facilitated is by incorporat- research has shown that outcome expectations and self-
ing theoretical perspectives from several sub-disciplines efficacy of clients (coaches) play an important role in the
of psychology (Grant, 2010), particularly from research effectiveness of therapeutic interventions (e.g. Goldin
into related fields of developmental interactions such as et al., 2012). These constructs will be of similar impor-
therapy, mentoring, and training (D’Abate, Eddy, & tance in the context of coaching.
Tannenbaum, 2003). More specifically, the relative Finally, studies on coaching and therapy have shown
theoretical richness of these fields may serve as a source that the relationship (working alliance) between a coach
of inspiration for theoretical enrichment in four interre- and a coachee (therapist and client) has considerable
lated areas of coaching research: the design of coaching implications for the effectiveness of interventions (Baron
14 T. Theeboom et al.
& Morin, 2009; Del Re, Horvath, Flückiger, Symonds, subordinate and co-worker perceptions so as to assess
& Wampold, 2012). With this in mind, both the literature the indirect effects of coaching.
on similarity attraction (Byrne, 1971) and interpersonal A fourth limitation of this study is that the findings
trust (Mayer, Davis, & Schoorman, 1995) can be used as are based on a relatively small number of studies.
theoretical frameworks to examine how functional Although we did not find any evidence for publication
relationships between a coach and a coachee can be bias, Sterne et al. (2011) noted that analyses for publica-
established and sustained. tion bias could produce unreliable results when the num-
ber of studies is small and heterogeneity across studies is
substantial. For this reason, our findings should be inter-
Limitations preted with caution. At the same time, however, our
Five limitations of the current study should be findings consistently showed effectiveness of coaching
mentioned. First, the majority of the studies included in across a broad spectrum of outcome measures. Also, our
this meta-analysis relied on self-reports of outcome sensitivity analysis indicated that the removal of the
measures. According to Peterson (1993), there is a con- study by Peterson (1993) did not alter our conclusions.
siderable inconsistency between self-reports and other- Fifth, the 1243 participants in the study by Smither
reports (e.g. by the supervisor or coach) when evaluating et al. (2003) account for a large proportion of the total
change in the coachee: self-reports tend to overestimate number of participants in the studies we examined. How-
the effects of coaching interventions. Hence, self-report ever, since the effect sizes in this study were much smal-
measures of performance seem troublesome (Podsakoff ler than the average effect sizes over all studies, the
Downloaded by [Juan Cortes] at 14:05 24 March 2014
& Organ, 1986). Therefore, future studies on coaching inclusion of the study of Smither et al. has resulted in a
should rely less on self-reports and should include other conservative rather than optimistic estimation of the
sources for measuring coaching outcomes such as 360 effectiveness of coaching.
feedback (see Smither, London, Flautt, & Fargas, 2003 Finally, the general lack of empirical work on
for an example) as well as tangible results. coaching and its weak theoretical foundation has resulted
Another problem with self-reports is that it is difficult in a large variety of coaching interventions and
to establish actual change on the outcome measure outcomes. As a consequence, the number of comparable
(alpha change) rather than respondents’ redefinition of studies suitable for a meta-analytic synthesis was
the rating scale (beta change) and/or the concept that is relatively limited.
measured (gamma change). Both beta and gamma
changes are due to a shifting conceptualization of the Conclusion
outcome as a result of coaching (Peterson, 1993). It Despite its limitations, the current meta-analysis indicates
should be noted, however, that also beta and gamma that coaching can be effectively used as an intervention
changes can be conceived as relevant outcomes of in organizations. Furthermore, this study has pointed out
coaching. Transformative learning theory states that several methodological issues that need to be addressed
existing belief systems and frames of reference need to in future studies on coaching effectiveness. The biggest
be challenged before deep-level changes will occur overall limitation of the coaching literature is the lack of
(Mezirow, 1991). The ultimate aim of coaching is to rigorous examinations showing the causal mechanisms
facilitate deep-level changes and learning (de Haan et al., by which coaching interventions are effective. Thus, we
2011). Therefore, more insight into alpha, beta, and agree with Fillery-Travis and Lane (2006) that it is the
gamma changes and their underlying cognitive structures time to shift attention from the question ‘does it work?’
(Thompson & Hunt, 1996) is needed because this may to ‘how does it work?’. This second question can only
help researchers and practitioners to better design a be answered by building a firm theoretical framework
coaching intervention and measure its impact. that can be used to identify the underlying mechanisms
A second limitation is that most studies in our and processes.
meta-analysis did not measure coaching effectiveness
over time (at multiple time-points), making it difficult to
assess the long-term impact of coaching interventions. Notes
Third, the focus on individual-level benefits of coaching 1. While some scholars explicitly distinguish personal
in the studies included in our analysis neglected possible coaching from organizational coaching2 (e.g. Grant, 2010),
‘spillover’ effects that coaching could have on other we take the position that this distinction will be less clear in
people within an organization. For example, if the practice. For example, when coaching concerns stress issues,
it often taps into the domain of work–life balance and the
coachee is an executive and his or her coaching results intended changes and outcomes will thus affect both the pro-
in improved leadership skills (the functioning of) subor- fessional and the personal functioning of the coached. There-
dinates and co-workers will benefit as well. Future fore, we do not make this distinction in our endeavor to
research on the effectiveness of coaching could include answer the question whether coaching is effective.
The Journal of Positive Psychology 15
2. For the transformation of effect sizes based on change Carlson, K. D., & Schmidt, F. L. (1999). Impact of experimental
scores standard deviations (repeated measure designs) into design on effect size: Findings from the research literature
the referent effect size index (based on post intervention on training. Journal of Applied Psychology, 84, 851–862.
score standard deviations), the comprehensive meta- *Cerni, T., Curtis, G. J., & Colmar, S. H. (2010). Executive
analysis (CMA) software’s’ default option (r = 0.5) was coaching can enhance transformational leadership. Interna-
used. As recommended by Morris (2000) and Borenstein tional Coaching Psychology Review, 5, 81–85.
et al. (2009), additional analyses based on different values Cochran, W. G. (1954). Some methods for strengthening the
for r (i.e. r = 0.1 and r = 0.9) were conducted and these common χ2 tests. Biometrics, 10, 417–451.
demonstrated similar results. The interested reader can con- Cohen, J. (1988). Statistical power analysis for the behavioral
tact the first author for more information. sciences (2nd ed.). Hillsdale, NJ: Eribaum.
Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation:
Design and analysis issues for field studies. Boston, MA:
References Houghton Mifflin Company.
Couto, D., & Kauffman, C. (2009). What can coaches do for
Baron, L., & Morin, L. (2009). The coach–coachee relationship you? Harvard Business Review, 1–8. Retrieved from http://
in executive coaching: A field study. Human Resource hbr.org/2009/01/what-can-coaches-do-for-you/ar/1
Development Quarterly, 20, 85–106. Cox, E., Bachkirova, T., & Clutterbuck, D. (2009). The com-
Berg, I. K., & Szabo, P. (2005). Brief coaching for lasting solu- plete handbook of coaching. London: Sage.
tions. New York, NY: W. W. Norton & Company, Inc. D’Abate, C., Eddy, E., & Tannenbaum, S. T. (2003). What’s in
Bond, F. W., & Bunce, D. (2003). The role of acceptance a name? A literature-based approach to understanding men-
and job control in mental health, job satisfaction, and toring, coaching, and other constructs that describe devel-
work performance. Journal of Applied Psychology, 88, opmental interactions. Human Resource Development
1057–1067.
Downloaded by [Juan Cortes] at 14:05 24 March 2014
Review, 2, 360–384.
Bono, J. E., Purvanova, R. K., Towler, A. J., & Peterson, D. B. Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and
(2009). A survey of executive coaching practices. Person- self-determination in human behavior. New York, NY: Ple-
nel Psychology, 62, 361–404. num.
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, Deci, E. L., & Ryan, R. M. (2000). The ‘what’ and ‘why’ of
H. R. (2005). Comprehensive meta-analysis version 2 goal pursuits: Human needs and the self-determination of
[Computer software]. Englewood, NJ: Biostat. behavior. Psychological Inquiry, 11, 227–268.
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, De Haan, E., Culpin, V., & Curd, J. (2011). Executive coaching
H. R. (2006). Comprehensive meta-analysis (Version in practice: What determines helpfulness for clients of
2.2.027) [Computer software]. Organizational Research coaching? Personnel Review, 40, 24–44.
Methods, 11¸ 188–191. De Meuse, K. P., Dai, G., & Lee, R. J. (2009). Evaluating the
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, effectiveness of executive coaching: Beyond ROI? Coach-
H. R. (2009). Introduction to meta-analysis. Cornwall: ing: An International Journal of Theory, Research and
Wiley. Practice, 2, 117–134.
*Bozer, G., & Sarros, J. C. (2012). Examining the effectiveness Del Re, A. C., Horvath, A. O., Flückiger, C., Symonds, D., &
of executive coaching on coaches’ performance in the Wampold, B. E. (2012). Therapist effects in the therapeutic
Israeli context. International Journal of Evidence Based alliance-outcome relationship: A restricted-maximum likeli-
Coaching and Mentoring, 10, 14–32. hood meta-analysis. Clinical Psychology Review, 32, 642–
Brock, V. G. (2008). Grounded theory of the roots and emer- 649.
gence of coaching (Unpublished Doctoral dissertation). De Shazer, S. (1988). Clues: Investigating solutions in brief
International University of Professional Studies. Retrieved therapy. New York, NY: W.W. Norton & Company Inc.
from http://libraryofprofessionalcoaching.com/wp-content/ Duijts, S. F. A. P., Kant, I. P., van den Brandt, P. A. P., & Swa-
uploads/2011/10/dissertation.pdf en, G. M. H. P. (2008). Effectiveness of a preventive
Brown, K. W., & Ryan, R. M. (2003). The benefits of being coaching intervention for employees at risk for sickness
present: Mindfulness and its role in psychological well- absence due to psychosocial health complaints: Results of a
being. Journal of Personality and Social Psychology, 84, randomized controlled trial. Journal of Occupational &
822–842. Environmental Medicine, 50, 765–776.
Brunn, E., & Milczarek, M. (2007). Expert forecast on emerg- *Egan, T., & Song, Z. (2005). A longitudinal quasi-experiment
ing psychosocial risks related to occupational safety and on the impact of executive coaching. Paper presented at the
health (European Risk Observatory Report). Retrieved from 20th Annual Conference of the Society for Industrial and
http://osha.europa.eu/en/publications/reports/7807118/ Organizational Psychology, Los Angeles.
Brunning, H. (2006). Executive coaching: Systems-psychody- Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997).
namic perspective. London: Karnac. Bias in meta-analysis detected by a simple, graphical test.
Burns, D. D., & Nolen-Hoeksema, S. (1992). Therapeutic British Medical Journal, 315, 629–634.
empathy and recovery from depression in cognitive- Elliott, R., Bohart, A. C., Watson, J. C., & Greenberg, L. S.
behavioral therapy: A structural equation model. Journal (2011). Empathy. Psychotherapy, 48, 43–49.
of Consulting and Clinical Psychology, 60, 441–449. *Evers, W. J., Brouwers, A., & Tomic, W. (2006). A quasi-
Burton, W. N., Conti, D. J., Chen, C. Y., Schultz, A. B., & experimental study on management coaching effectiveness.
Edington, D. W. (1999). The role of health risk factors and Consulting Psychology Journal: Practice and Research, 58,
disease on worker productivity. Journal of Occupational 174–182.
and Environmental Medicine, 41, 863–877. Feldman, D. C., & Lankau, M. J. (2005). Executive coaching:
Byrne, D. E. (1971). The attraction paradigm. New York, NY: A review and agenda for future research. Journal of Man-
Academic Press. agement, 31, 829–848.
16 T. Theeboom et al.
Fillery-Travis, A., & Lane, D. (2006). Does coaching work or Grant, A. M., & O’Connor, S. A. (2010). The differential
are we asking the wrong question? International Coaching effects of solution-focused and problem-focused coaching
Psychology Review, 1, 23–35. questions: A pilot study with implications for practice.
*Finn, F. A. (2007). Leadership development through executive Industrial and Commercial Training, 42, 102–111.
coaching: The effects on leaders’ psychological states and Grant, A. M., Passmore, J., Cavanagh, M., & Parker, H.
transformational leadership behavior (Unpublished (2010). The state of play in coaching. International Review
Doctoral dissertation). Queensland University of Technol- of Industrial & Organizational Psychology, 25, 125–168.
ogy. Retrieved from http://eprints.qut.edu.au/17001/ *Green, L. S., Grant, A. M., & Rynsaardt, J. (2007). Evidence-
Freire, T. (2013). Positive psychology approaches. In J. Pass- based life coaching for senior high school students: Build-
more, D. Peterson, & T. Freire (Eds.), The Wiley-Blackwell ing hardiness and hope. International Coaching Psychology
handbook of the psychology of coaching and mentoring Review, 2, 24–32.
(pp. 426–442). West Sussex: Wiley-Blackwell. *Green, L. S., Oades, L. G., & Grant, A. M. (2006). Cogni-
Fusar-Poli, P., Bonoldi, I., Yung, A. R., Borgwardt, S., Kemp- tive-behavioural, solution-focused life coaching: Enhancing
ton, M. J., Valmaggia, L., & McGuire, P. (2012). Predicting goal striving, well-being and hope. Journal of Positive Psy-
psychosis: Meta-analysis of transition outcomes in individ- chology, 1, 142–149.
uals at high clinical risk. Archives of General Psychiatry, Greenhaus, J. H., Parasuraman, S., & Wormley, W. M. (1990).
69, 220. Effects of race on organizational experience, job perfor-
Gallwey, T. W. (1974). The inner game of tennis (1st ed.). mance evaluations, and career outcomes. Academy of Man-
New York, NY: Random House. agement Journal, 33, 64–86.
Giacobbi, P. R. (2000). The athletic coachability scale: Con- Hart, V., Blattner, J., & Leipsic, S. (2001). Coaching versus
struct conceptualization and psychometric analyses (Doc- therapy: A perspective. Consulting Psychology Journal:
toral dissertation). University of Tennessee, Knoxville. Practice & Research, 53, 229–237.
Downloaded by [Juan Cortes] at 14:05 24 March 2014
Retrieved from http://sunzi.lib.hku.hk/ER/detail/hkul/ Hedges, L. V. (1981). Distribution theory for Glass’s estimator
2688806 of effect size and related estimators. Journal of Educational
Goldin, P. R., Ziv, M., Jazaieri, H., Werner, K., Kraemer, H., Statistics, 6, 107–128.
Heimberg, R. G., & Gross, J. J. (2012). Cognitive reap- Hedges, L. V., & Cooper, H. (1994). The handbook of research
praisal self-efficacy mediates the effects of individual cog- synthesis. New York, NY: Sage.
nitive-behavioral therapy for social anxiety disorder. Hedges, L., & Olkin, I. (1985). Statistical methods for meta-
Journal of Consulting and Clinical Psychology, 80, 1034– analysis. San Diego, CA: Academic Press.
1043. Higgins, J. P. T., & Thompson, S. G. (2002). Quantifying het-
Grant, A. M. (2001). Towards a psychology of coaching: The erogeneity in a meta-analysis. Statistics in Medicine, 21,
impact of coaching on metacognition, mental health and 1539–1558.
goal attainment (Unpublished Doctoral dissertation). Mac- Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman,
quarie University, Sydney. Retrieved from http://www.eric. D. G. (2003). Measuring inconsistency in meta-analyses.
ed.gov/PDFS/ED478147.pdf British Medical Journal, 327, 557–560.
*Grant, A. M. (2003). The impact of life coaching on goal- Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analy-
attainment, metacognition and mental health. Social Behav- sis: Correcting error and bias in research findings. New-
ior and Personality, 31, 253–264. bury Park, CA: Sage.
Grant, A. M. (2008). Personal life coaching for coaches-in- Hunter, J. E., & Schmidt, F. L. (2004). Methods of meta-analy-
training enhances goal attainment, insight and learning. sis: Correcting error and bias in research findings (2nd
Coaching: An International Journal of Theory, Research ed.). Thousand Oaks, CA: Sage.
and Practice, 1, 54–70. International Coach Federation. (2012). Retrieved May, 16, 2012
*Grant, A. M. (2010). The development of coaching psychol- from, http://www.coachfederation.org/about-icf/overview/
ogy. International Coaching Psychology Review, 1, 12– Johnson, B. T., Mullen, B., & Salas, E. (1995). Comparison of
22. three major meta-analytic approaches. Journal of Applied
Grant, A. M. (2013). The efficacy of coaching. In J. Passmore, Psychology, 80, 94–106.
D. Peterson, & T. Freire (Eds.), Handbook of the psychol- Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K.
ogy of coaching and mentoring (pp. 15–39). West Sussex: (2001). The job satisfaction-job performance relationship:
Wiley-Blackwell. A qualitative and quantitative review. Psychological Bulle-
Grant, A. M., & Cavanagh, M. J. (2004). Toward a profession tin, 127, 376–407.
of coaching: Sixty-five years of progress and challenges for Kampa-Kokesch, S., & Anderson, M. Z. (2001). Executive
the future. International Journal of Evidence-based Coach- coaching: A comprehensive review of the literature. Con-
ing and Mentoring, 2, 1–16. sulting Psychology Journal: Practice and Research, 53,
Grant, A. M., & Cavanagh, M. J. (2007). Evidence-based 205–228.
coaching: Flourishing or languishing? Australian Psycholo- Kilburg, R. R. (1996). Toward a conceptual understanding and
gist, 42, 239–254. definition of executive Coaching. Consulting Psychology
*Grant, A. M., Curtayne, L., & Burton, G. (2009). Executive Journal: Practice and Research, 48, 134–144.
coaching enhances goal attainment, resilience and work- Kim, J. S. (2008). Examining the effectiveness of solution-
place well-being: A randomized controlled study. The Jour- focused brief therapy: A meta-analysis. Research on Social
nal of Positive Psychology, 4, 396–407. Work Practice, 18, 107–116.
*Grant, A. M., Green, L. S., & Rynsaardt, J. (2010). Develop- *Kochanowski, S., Seifert, C. F., & Yukl, G. (2010). Using
mental coaching for high school teachers: Executive coach- coaching to enhance the effects of behavioral feedback to
ing goes to school. Consulting Psychology Journal: managers. Journal of Leadership & Organizational Studies,
Practice and Research, 3, 151–168. 17, 363–369.
The Journal of Positive Psychology 17
Landis, J. R., & Koch, G. G. (1977). The measurement of Passmore, J., & Fillery-Travis, A. (2011). A critical review of
observer agreement for categorical data. Biometrics, 33, executive coaching research: A decade of progress and
159–174. what’s to come. Coaching: An International Journal of
Latham, G. P. (2007). Theory and research on coaching prac- Theory, Practice & Research, 4, 70–88.
tices. Australian Psychologist, 42, 268–270. *Peterson, D. B. (1993). Measuring change: A psychometric
Leonard-Cross, E. (2010). Developmental coaching: Business approach to evaluating individual training outcomes. Sym-
benefit – Fact or fad? An evaluative study to explore the posium conducted at the Eighth Annual Conference of the
impact of coaching in the workplace. International Coach- Society for Industrial and Organizational Psychologists,
ing Psychology Review, 5, 36–47. San Francisco, CA.
Linley, P. A., & Harrington, S. (2005). Positive psychology Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in orga-
and coaching psychology: Perspectives on integration. The nizational research: Problems and prospects. Journal of
Coaching Psychologist, 1, 13–14. Management, 12, 531–544.
Locke, E. A., & Latham, G. P. (2002). Building a practically *Poepsel, M. A. (2011). The impact of an online evidence-based
useful theory of goal setting and task motivation: A 35-year coaching program on goal striving, subjective well-being,
odyssey. American Psychologist, 57, 705–717. and level of hope (Unpublished Doctoral dissertation),
Lovibond, P. F., & Lovibond, S. H. (1995). The structure of Minneapolis, MN. Harold Abel School of Social and
negative emotional states: Comparison of the depression Behavioral Sciences. Retrieved from http://gradworks.
Anxiety stress scales (DASS) with the beck depression and umi.com/3456769.pdf
anxiety inventories. Behaviour Research and Therapy, 33, Porter, L. W., Steers, R. M., Mowday, R. T., & Boulian, P. V.
335–343. (1974). Organizational commitment, job satisfaction, and
*Luthans, F., & Peterson, S. J. (2004). 360-degree feedback turnover among psychiatric technicians. Journal of Applied
with systematic coaching: Empirical analysis suggests a Psychology, 59, 603–609.
Downloaded by [Juan Cortes] at 14:05 24 March 2014
winning combination. Human Resource Management, 42, Rosenthal, R. (1991). Meta-analytic procedures for social
243–256. research (Rev. ed.). Newbury Park, CA: Sage.
Maslach, C., & Jackson, S. E. (1986). MBI: Maslach Burnout Ryan, R. M., Lynch, M. F., Vansteenkiste, M., & Deci, E. L.
Inventory manual (research edition) (1st ed. 1981). Palo (2011). Motivation and autonomy in counseling, psycho-
Alto, CA: Consulting Psychologists Press. therapy, and behavior change: A look at theory and prac-
Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An tice. The Counseling Psychologist, 39, 193–260.
integrative model of organizational trust. Academy of Man- Salovey, P., & Mayer, J. D. (1989). Emotional intelligence.
agement Review, 20, 709–734. Imagination, Cognition and Personality, 9, 185–211.
McGaw, B., & Glass, G. (1980). Choice of the metric for effect Schwarzer, R., & Jerusalem, M. (1995). Generalized self-effi-
size in meta-analysis. American Educational Research cacy scale. In J. Weinman, S. Wright, & M. Johnston
Journal, 17, 325–337. (Eds.), Measures in health psychology: A user’s portfolio.
Mezirow, J. (1991). Transformative dimensions of adult learn- Causal and control beliefs (pp. 35–37). Windsor: NFER-
ing. San Francisco, CA: Jossey-Bass. NELSON.
*Moen, F., & Skaalvik, E. (2009). The effect from coaching on Seligman, M. E. (2007). Coaching and positive psychology.
performance psychology. International Journal of Evidence Australian Psychologist, 42, 266–267.
Based Coaching and Mentoring, 7, 31–49. Seligman, M. E., & Csikszentmihalyi, M. (2000). Positive
Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). psychology: An introduction. American Psychologist, 55(1),
Preferred reporting items for systematic reviews and meta- 5–14.
analyses: The PRISMA statement. Annals of Internal Medi- Smith, P. C., Kendall, L. M., & Hulin, C. L. (1969). The mea-
cine, 151, 264–269. surement of satisfaction in work and retirement. Chicago:
Morris, S. B. (2000). Distribution of the standardized mean Rand McNally.
change effect size for meta-analysis on repeated measures. *Smither, J. W., London, M., Flautt, R., Vargas, Y., & Kucine,
British Journal of Mathematical and Statistical Psychology, I. (2003). Can working with an executive coach improve
53(1), 17–29. multisource feedback ratings over time? A quasi-experi-
Morris, S. B., & DeShon, R. P. (1997). Correcting effect sizes mental field study. Personnel Psychology, 56, 23–44.
computed from factorial ANOVA for use in meta-analysis. Spence, G. (2008). New directions in evidence-based coaching:
Psychological Methods, 2, 192–199. Investigations into the impact of mindfulness training on
Morris, S. B., & DeShon, R. P. (2002). Combining effect size goal attainment and well-being (Unpublished Doctoral dis-
estimates in meta-analysis with repeated measures and sertation). University of Sydney, Sydney. Retrieved from
independent-groups designs. Psychological Methods, 7, http://ses.library.usyd.edu.au/bitstream/2123/2469/1/New%20
105–125. Directions%20in%20the%20Psychology%20of%20Coac
Mullen, E. J. (1994). Framing the mentoring relationship as an hing%20(Spence,%202006).pdf
information exchange. Human Resource Management Spence, G. B., Cavanagh, M. J., & Grant, A. M. (2008). The
Review, 4, 257–281. integration of mindfulness training and health coaching: An
National Research Council (1992). Combining information: exploratory study. Coaching: An International Journal of
Statistical issues and opportunities for research. Washing- Theory, Research and Practice, 1, 145–163.
ton, WA: National Academy Press. *Spence, G. B., & Grant, A. M. (2007). Professional and peer
Organ, D. W., & Ryan, K. (1995). A meta-analytic review of life coaching and the enhancement of goal striving and
attitudinal and dispositional predictors of organizational well-being: An exploratory study. Journal of Positive Psy-
citizenship behavior. Personnel Psychology, 48, 775–802. chology, 2, 185–194.
Passmore, J. (Ed.). (2010). Excellence in coaching: The indus- Spence, G. B., & Oaedes, L. G. (2011). Coaching with self-
try guide. London: Kogan Page. determination in mind: Using theory to advance evidence-
18 T. Theeboom et al.
based coaching practice. International Journal of Evidence cessing model to assess attitude structure. Academy of
Based Coaching and Mentoring, 9, 37–55. Management Review, 4, 655–690.
Sterne, J. A., Sutton, A. J., Ioannidis, J., Terrin, N., Jones, D. Waldron, V. R. (1991). Achieving communication goals in
R., Lau, J., & Higgins, J. (2011). Recommendations for superior-subordinate relationships: The multi-functionality
examining and interpreting funnel plot asymmetry in meta- of upward maintenance tactics. Communications Mono-
analyses of randomised controlled trials. British Medical graphs, 58, 289–306.
Journal, 343(7818), 302–307. Whitmore, J. (1992). Coaching for performance. San Diego,
Sue-Chan, C., & Latham, G. P. (2004). The relative effective- CA: Pfeiffer.
ness of external, peer, and self-coaches. Applied Psychol- Wright, T. A., & Cropanzano, R. (2000). Psychological
ogy, 53, 260–278. well-being and job satisfaction as predictors of job per-
Tepper, B. J. (1995). Upward maintenance tactics in supervi- formance. Journal of Occupational Health Psychology,
sory mentoring and nonmentoring relationships. Academy 5, 84–94.
of Management Journal, 38, 1191–1205. Wrzesniewski, A., & Dutton, J. (2001). Crafting a job: Employ-
Thompson, R. C., & Hunt, J. G. (1996). Inside the black box ees as active crafters of their work. Academy of Manage-
of alpha, beta, and gamma change: Using a cognitive-pro- ment Review, 26, 179–201.
Downloaded by [Juan Cortes] at 14:05 24 March 2014