Coaches have long sought toimprove their coachingconversations and have

frequently questioned theeffectiveness of their work and theimpact it has on helping clients tomeet their objectives.

However, despite being curiousabout the effectiveness or outcome oftheir coaching practices, there havebeen few serious attempts to explorethe propensity of their practice in areliable and validated way.

We estimate there are probablyfewer than 20 robust quantitativeoutcome studies in the executivecoaching literature, none of whichsatisfy the gold standard of thedouble-blind randomised control trialoften used in medicine orpsychotherapy. The main reasonsbehind this are the prohibiting costsand the formal requirements of arigorous outcome study. The coachingindustry is relatively small andfragmented, and coaches prefer,rightly, to prioritise their coachingcommitments, and often don't findthe time or the right researchenvironment to objectively study theirprofession's effectiveness. This givesus a very limited understanding ofoverall coaching outcomes, one basedlargely on assumption rather thantrue scientific evidence.

From what little we do know{De Haan & Duckworth, 2013), weexpect the very reliable results inpsychotherapy outcome research tobe more or less replicated in executivecoaching. We think overall objectiveeffectiveness has now beendemonstrated, even if the effects areby no means as high as inpsychotherapy. We also believe thatthose factors identified as the 'activeingredients' of psychotherapy, such asquality of coaching relationship,positive expectations, personalitiesof therapist and patient, and so on,will be active ingredients of coachingas well.

l o Coaching at Woric

Outcome report:conversationsare key to resuitsErik de Haan and Nadine Page report on thelargest quantitative coaching outcome studyto date which they beheve breaks new groundand will help to improve the effectiveness ofcoaching relationships

]uly/August I Vol 8 Issue 4 © 2013 | www.coaching-at-work.com

Although recent studies haveillustrated the rise of coachingoutcome research, they have notnecessarily conveyed a coherent orreliable message about theeffectiveness of coachingconversations. For example, studiesfavouring a field study method withouta control group have found very largeeffects based on clients' self-reports{tg,McCovern etal, 2OOl;Levenson,2009), and large effects when objectivemeasures are used (eg, Bow/es et ai,2007; Perkins, 2009). In contrast, thosestudies with a control group andobjective measures (eg, Sm/tiier et ai,2003; Evers et al, 2006) found smalleffects. It appears that if the client is the

The greatest Coaching Outcome Research Ever:our key findings• The coaching relationship is the best predictor of outcome

• Clients, coaches and sponsors agree on what they see as coaching outcome

• Ciient seif-efpcacy (the client's seif-motivation) is also an importantactive ingredient

• There are only small correlations between personality in terms of MBTI andthe other variables

sole source of the data, the outcometends to be very positive. However,when such same-source bias iscontrolled for, the effect is muchsmaller, although still positive.

Likewise, studies exploring thevarious aspects of the coach or theclient that might have a positive effecton outcome are also on the rise. Thesehave included both coach and chentcharacteristics, including coachpersona (eg, warmth, status, health)and ideology (eg, allegiance), chentunderstanding or hope about therelationship and the strength of thecoaching relationship {DeHaan &Duckworth, 2013). However, due to thedesign and method constraints of thecurrent research approaches, andthe variability in levels ofmeasurement, there is a degree ofambiguity about the activeingredients in executive coaching.

Our goalIn this study we explore the importanceand relative impact of some of thepotentially active ingredients commonto all coaching approaches, and fromthe perspective of all stakeholders onthe coaching journey (coach, client andorganisational sponsor). We believeour findings break new ground incoaching outcomes research and willhelp improve the effectiveness ofcoaching relationships.

Over the past 18 months we havebeen reaching out to coaches, clients,and sponsors internationally using an'open source' approach. We believethat the best way to develop a good

understanding of effective coachingis to engage with those who arepresently involved in the process.We invited experienced coaches withan interest in doing sohd research tojoin forces and gather high-volumedata collectively. This enabled us tocollect as many measurements of real-life coaching assignments as possible.It was our hope to obtain the largestsample of coaching relationships inthe coaching literature. We firmlybelieve we have achieved this goal.

Our sampleIn the preliminary findings reportedhere, which represent approximately75 per cent of the final sample, wehave reliable data for more than 1,100coaches, 1,800 coaching clients and 82organisational sponsors (linemanagers or directors) from morethan 34 different countries. Thisdataset is already substantially largerthan the largest we have identified inthe literature (in Smither et al, 2003).

We know that participation levelshave increased significantly since ourlast download of the data, but due tothe sheer volume of responses we havebeen unable to process and verify thereliability of all of our data. We hopeto be able to share results from thefinal dataset by the end of the year, inthe form of a peer-reviewed articletogether with many of the coacheswho have joined forces with us for thisresearch. But for now, we would like tooffer a taster of what we think aresome very interesting, novel andsignificant findings.

]uly/August I Vol 8 Issue 4 © 2013 | www.coaching-at-work.com Coaching at Work

Our approachBuilding on our previous research{DeHaanetal,20n;2013)tiiisresearch has taken a big step forward inunderstanding coaching outcomes.Engaging all stakeholders in thecoaching journey (coach, client, andsponsors) is a method we believeunique to our research project. It hasenabled us to provide new insights intocoaching relationships from threeperspectives. In the online surveys thatwe distributed to coach, client andsponsor, we measured coachingeffectiveness as our main outcomevariable. We also assessed the strengthof the coaching relationship in threeareas - bond, task and goal; personallevels of self-efficacy and personalitypreferences (as characterised withMBTI), as the predictors in our model.

Our predictionsFigure J displays the relationshipsamong the variables we measured.

COACHINgOUTCOMES

COACHINQRELATIONSHIPS

COACHSELF-EFFICACY

CLIENTSELFEFFICACY

PERSONALITYDIFFERENCES

\

COACHPERSONA

CLIENTRSONALITY

Figure 1 Relationship among measured variables

The coaching relationshiphas the most powerful connectionwith coaching outcomes ^ß

We make four predictions aboutcoaching outcomes: 1 ) the strength ofthe coaching relationship as reportedby both coach and client will predictcoaching outcomes, 2) client andcoach personahty, and client-coachpersonality dissimilarity will predictcoaching outcomes, 3) client andcoach self-efficacy will predictcoaching outcome, and 4) thestrength of the coaching relationshipwill mediate the effect of personalityand self-efficacy as predictors ofcoaching outcomes.

Our findingsCoaching outcomes from three lensesIn line with our expectations, we foundpositive relationships among the

variables overall. The coach, client andsponsor perceptions of the coachingoutcome for clients were positivelyrelated {Pearson's r ranged from.20-.33**), suggesting stakeholders inthe coaching process have broadagreement on the effectiveness of thecoaching contracts. We perceive this asa positive, unique result.

Seif-efficacyWe also found that client self-efficacyhad a direct relationship withcoaching relationship and coachingoutcome as perceived by the client,but not the coach (Pearson's r = .26**;r = .29**). And in reverse, coachself-efficacy related to coachingrelationship and coaching outcomes

as perceived by the coach, but not theclient (Pearson's r=í5**;r=.22**).

There was no cross-over betweenclient and coach. It seems that anindividual's personal self-efficacy levelscan determine their own coachingrelationship and outcome, but notthose of others. This finding supportsprevious research showing a person'sself-efficacy expectations have a directbearing on their personal and careerdevelopment {Anderson &Betz, 2001).

PersonaiityWe found some effects of personality(dis)similarity on coaching outcomesand coaching relationship. Personalitydissimilarity as measured by thesensing-intuiting (S/N) dimension ofthe MBTI, was related to bettercoaching outcomes. The N/Scombination for client-coach was themost effective and an S/S match was theleast effective. A personality match onthe judging-perceiving (J/P) dimensionwas more important for the coaching

12 Coaching at Work July/August I Vol 8 Issue 4 © 2013 | www.coaching-at-work.com

relationsbip. A P/P match was the mosteffective and a P/J mismatch for client-coach was the least effective for thegoal aspects of the coachingrelationship. These results partiallysupport previous research {Scoular&Linley, 2006) and indicate that differentperspectives can be more effective forcoaching outcomes, but profilesimilarity is important for quality ofthe coaching relationship.

Coaching relationshipAbove and beyond the precedingresults, the clearest messageemanating from this research is thattbe coaching relationship has themost powerful connection withcoaching outcomes. We measured thisrelationship in three ways: for coachand client scores separately, and thenfor the correspondence across coachand client data. We also consideredthe task, goal and bond aspects of theworking alliance to see which was themost 'active ingredient' of thecoaching relationship.

There was a similar pattern of resultsfor all pairings {Pearson's r range from.20-.60**). We found that the task andgoal dimensions of the coachingrelationship had stronger connectionswith coaching outcomes compared tothe bond aspect. However, all threedimensions of the client-coachrelationship are important.

Watch this spaceWe have also been exploring severalother 'common factors' that mightrelate with coaching outcome andcoaching relationship. These include:client and coach gender, type of coachand length of coaching relationship.

We have found some interestingconnections here, but it is too early toreport on these. We are erring on theside of caution and want to estabhshthe reliability of these results beforesharing them. We are finding it veryeasy to pick up significant findingsbecause of the large dataset and do

not want to misconstrue our findings.What we have reported here is

just a snippet of the data that wehave available and the relationshipsthat we could explore. We are excitedabout what we have found so farand the future possibihties of thiswork. We wanted to share ourfindings with coaches as soon as wehad them ourselves, to honour theirimmense contributions to theresearch project and also perhaps toaffect their current coaching-conversations positively. To delaywould simply be a disservice tothe profession. We will, however,continue with our endeavours tomake this the 'greatest ever' coachingoutcome project.

Over the next couple of months wewill be replicating and conductingfurther analyses on the final dataset,and then sharing these findingsinternationally at the end of the year,in a peer-reviewed paper. We hope ourcurrent findings have given you someinsight into your own coaching-conversations, and that both you andyour clients will benefit as a result.

Erik de Haan is lead researcher andNadine Page lead statistician, atAshridge Centre for Coaching,Ashridge Business School

Note** denotes a significance level ofp<0.01 for a false positive

References• s L Anderson N E & Betz, 'Sources of social seif-efficacy expectations: theirmeasurement and relation to career development', ir\ Journal of VocationalBehavior, 58, PP98-117,2001

• S V Bowles, C ] L Cunningham, q M De La Rosa & ] ] Picaño, 'Coaching leaders inmiddle and executive management: goals, performance, buy-in', in Leadership andOrganization Development Journal, 28(5), PP388-408,2007• E De Haan, V Culpin & ] Curd, 'Executive coaching in practice: what determineshelpfulness for clients of coaching?', in Personnel Review, 40(1), PP24-44,2011• E De Haan, A Duckworth, D Birch & C Jones, 'Executive coaching outcomeresearch: the predictive value of common factors such as relationship, personalitymatch and self-efficacy', in Consulting Psychology Journal: Practice and Research,65(1), PP40-57,2013

• E De Haan & A Duckworth, 'Signaling a new trend in coaching outcome research',in International Coaching Psychology Review, 8(1), pp6-20,2013• W ] q Evers, A Brouwers & W Tomic, 'A quasi-experimental study on managementcoaching effectiveness', in Consulting Psychology Journal: Practice and Research,58, PP174-182,2006

• A Levenson, 'Measuring and maximizing the business impact of executivecoaching', in Consulting Psychology Journal: Practice and Research, 61,PP103-121,2009

• ] Mcqovern, M Lindemann, M Vergara, S Murphy, L Barker & R Warrenfeltz,'Maximizing the impact of executive coaching: Behavioral change, organizationaloutcomes, and return on investment', in The Manchester Review, 6, ppi-9,2001

• R D Perkins, 'How executive coaching can change leader behaviour and improvemeeting effectiveness: an exploratory study', in Consulting Psychology Journal:Practice and Research, 61(4), PP298-318,2009• A Scoular & P A Linley, 'Coaching, goal-setting and personality type: Whatmatters?', in The Coaching Psychologist, 2, pp9-ii, 2006• ] W Smither, M London, R Flautt, Y Vargas & I Kucine, 'Can working withan executive coach improve multisource feedback ratings over time?A quasi-experimental field study', in Personnel Psychology, 56, PP23-44,2003

July/August I Vol 8 Issue 4 © 2013 | www.coaching-at-work.com Coaching at Work 13

Copyright of Coaching at Work is the property of Coaching at Work Limited and its contentmay not be copied or emailed to multiple sites or posted to a listserv without the copyrightholder's express written permission. However, users may print, download, or email articles forindividual use.