Randomised controlled trials (RCTs) – a rarity in coaching – can help build a compelling argument for executive coaching. Erik de Haan, Sally Bonneywell and Sue Gammons report on what’s thought to be the largest ever RCT of a corporate coaching programme
Executive coaching has increased substantially in organisations in the past 20 years and is now firmly established as a development intervention for leaders in large organisations (Ridler, 2016).
While increasingly rigorous standards and qualifications have been developed and a wide variety of research into coaching has been undertaken, as shown in recent meta-analyses (Theeboom et al, 2014; Jones et al, 2015), very few corporate coaching programmes have been formally evaluated (McDermott et al, 2007).
Large-scale randomised controlled trials (RCTs) in executive and workplace coaching are needed to determine to what extent any seen benefits can be attributed to the coaching itself, and how independent factors might influence the outcome. This helps to determine which individuals might benefit most from coaching, or what preparation is optimal for clients.
In today’s budget-constrained world, organisations often need to make some tough decisions about when and how coaching is made available to executives. The results of RCTs can play an important role in making such delicate choices about investment in coaching and the matching of clients and coaches.
In this article, we report on an RCT conducted within a large global healthcare corporation, involving 180 clients, 66 coaches and 140 managers of the clients. We contend that this is the largest RCT of a corporate coaching programme to date.
The study involved two consecutive groups of female leaders on a leadership development programme (Accelerating Difference) designed to increase the proportion of women in leadership roles. This 14-month programme consisted of a maximum of 12 individual one-hour coaching sessions, together with six of group coaching.
Within the design of the programme, participants (clients) were randomly assigned to one of two cohorts, the first of which started in April and the second in September, creating a six-month time-lagged control group. The coaches were qualified, part-time internal coaches, most of whom were managers with a ‘day job’ in the organisation.
Data was collected through online questionnaires sent to all clients, their managers and the coaches by email. The first questionnaires were completed before the coaching for cohort 1 (the experimental group) started, the second questionnaires went out six months later before cohort 2 (the control group) started their coaching and the third measurement was six months later, once coaching for cohort 1 had ended.
Coaching effectiveness
There was strong evidence that executive coaching is an effective intervention, not only in the eyes of the clients (see Figure 1a), but also in the eyes of their line managers (see Figure 1b). Coaching effectiveness was assessed using four items on a seven-point response scale (‘Strongly Disagree’ to ‘Strongly Agree’): “The outcome of my coaching objectives so far: 1) I have been successful in creating reflective space for me; 2) I have been successful in creating new insight for me; 3) Through (preparation for) coaching I have successfully engaged in new action or behaviour, and 4) I would consider this coaching journey successful.”
Use of a randomised control group verified that the effectiveness can be attributed to the coaching itself and confirmed meta-analysis studies such as Jones et al (2015). The difference between the experimental and control groups was larger than anticipated, which may be due to the addition of group coaching and managerial support as part of the leadership development programme.
Organisational buyers can feel reassured that coaching is generally a very effective intervention in the eyes of clients and their managers and may warrant investment for leadership development. While this study was undertaken in a healthcare company, we believe these findings are applicable in many industries, because the clients were globally mobile, senior and mostly general managers, not healthcare experts.
Factors Influencing Coaching Effectiveness
- Working alliance
It’s now well recognised that coaching effectiveness goes hand in hand with a strong relationship between coach and client. The study confirmed this in finding a strong positive correlation between the working alliance and coaching effectiveness as rated by coach and client.
The client’s experience of the working alliance also predicted coaching effectiveness at later times, as reported previously by Gessnitzer & Kauffeld (2015). This strongly suggests that matching is best done by the clients themselves, through a chemistry meeting or trial session, and not by HR or other intermediaries, as indicated in previous research by Boyce et al (2010), Page & De Haan (2014), and Bozer et al (2015).
- Client preparedness
Clients who rated themselves more highly in terms of their self-efficacy, psychological wellbeing, resilience and social support at the start of the coaching were more likely to have higher coaching effectiveness scores as rated by the coach and client. These measures may be regarded as indicators of the client’s mental preparedness for coaching and this extends earlier work by Stewart et al (2008) and De Haan et al (2013; 2016), which focused on self-efficacy only.
Self-efficacy is the belief in one’s competence to cope with a broad range of stressful or challenging demands and has been found to be a valid construct in a broad range of ethnic and cultural settings globally (Luszczynska et al, 2005).
Coaching also, in turn, improved their psychological wellbeing and sense of support. This confirms previous work by Grant et al (2009; 2010) who found that coaching enhanced workplace wellbeing and resilience.
- Personality
Coaching showed the first signs of supporting the development of the personality factors Adjustment (confidence, self-esteem and composure under pressure), Ambition (initiative, competitiveness and desire for leadership roles) and Interpersonal sensitivity (tact, perceptiveness and ability to maintain relationships) on the Hogan Personality Inventory, which describes ‘bright-side’ personality – qualities that describe how we relate to others when we are at our best. These findings were relatively small and inconsistent, so require further research, but they confirmed earlier work by Stewart et al (2008).
- Coaching behaviours
There was very little evidence that specific coaching behaviours had any impact on coach- or client-rated coaching effectiveness. These behaviours were measured using the recently validated Ashridge Coaching Behaviours Questionnaire (De Haan & Nilsson, 2017), based on John Heron’s model of coaching interventions: Prescribing, Confronting, Informing, Releasing, Exploring and Supporting. This finding was expected and in line with many meta-analyses in the field of psychotherapy (summarised in De Haan, 2008), where factors such as the strength of the working alliance have been found to have a bigger impact on the effectiveness of the therapy.
Personality derailment patterns
There has long been an expectation that coaching can be particularly helpful with understanding and in changing unhelpful, maladapted and hidden personality characteristics of executives (e.g. Kets de Vries, 2006). Because of its personalised nature, the high confidentiality and possibility for deep understanding and challenge, coaching seems to work at relational and personal depth, touching on personality derailment patterns of executives (Nelson & Hogan, 2009).
Coaching is expected to make a difference in important and personal aspects of an executive’s leadership and it has been suggested coaching might be uniquely targeted at reducing “personality derailment tendencies” of leaders, which can be costly to large organisations (Kaiser et al, 2015).
This study is the first to demonstrate the impact of coaching on leadership derailment patterns, by measuring these personality characteristics before and after the intervention. Coaching effectiveness as rated by the client correlated with changes in two personality measures on the Hogan Personality Inventory, namely Prudence (self-discipline, responsibility and conscientiousness) and Excitable (moody, easily annoyed, hard to please and emotionally volatile). Prudence rises significantly while Excitable lowers significantly in the experimental group.
Coaching seems to have a demonstrable calming or containing effect, which is exactly what it sets out to do in being a conversational, reflective practice. This result confirms a large-scale meta-analysis study into the effect of psychotherapy on similar personality traits, such as emotional stability and extraversion (Roberts et al, 2017).
More research is needed to determine if coaching has the capacity to change those bright or dark sides of personality, or if coaching only affects the client’s mood state, helping them ‘feel’ better or calmer. It is plausible that coaching helps executives gain new insights and sensitises them to their own habitual responses, resulting in them becoming more self-aware, adaptive or self-regulated without fundamentally changing their deeper ‘personalities’. However, this would still positively affect their reputation in the workplace and alter how their personality was perceived by others.
Key lessons
- Organisations can feel reassured that coaching is an effective intervention that warrants investment for leadership development.
- Coaches can explore their client’s preparedness for coaching – their motivation, support and mental wellbeing – and advise clients in the initial contracting conversations to also consider those aspects.
- Coaches should keep enquiring into the quality of the working alliance, particularly from the perspective of their clients, since it is such an important predictor of effectiveness.
- More large-scale RCTs are needed to further explore coaching effectiveness, including the influence of ‘preparedness’ factors such as self-efficacy, wellbeing and social support, and to unpick the differential effects of group and individual coaching conflated in this study.
The authors
- Erik de Haan is director of the Ashridge Centre for Coaching, Ashridge/Hult International Business School School for Business and Economics, and Professor of Organisation Development & Coaching at the School of Business and Economics, VU Amsterdam
- Sally Bonneywell is founder and CEO, Bonneywell Development Ltd
- Sue Gammons is coaching director, GSK
References
- L A Boyce, R J Jackson and L J Neal, ‘Building successful leadership coaching relationships: Examining impact of matching criteria in a leadership coaching program’, in Journal of Management Development, 29, 914-931, 2010
- G Bozer, J Baek-Kyoo and J C Santora, ‘Executive coaching: does coach-coachee matching based on similarity really matter?’, in Consulting Psychology Journal: Theory and Practice, 67(3), 218-233, 2015
- E de Haan, Relational Coaching: Journeys towards mastering one-to-one learning. Chichester: John Wiley & Sons, 2008
- E de Haan and V Nilsson, ‘Evaluating coaching behavior in managers, consultants and coaches: A model, questionnaire, and initial findings’, in Consulting Psychology Journal: Practice and Research, 69(4), 315-333, 2017
- E de Haan, A Duckworth, D Birch and C Jones, ‘Executive coaching outcome research: The contribution of common factors such as relationship, personality match, and self-efficacy’, in Consulting Psychology Journal: Practice and Research, 65(1), 40-57, 2013
- E de Haan, A Grant, Y Burger and P-O Eriksson, ‘A large-scale study of executive coaching outcome: the relative contributions of working relationship, personality match, and self-efficacy’, in Consulting Psychology Journal: Practice and Research, 68(3), 189-207, 2016
- S Gessnitzer and S Kauffeld, ‘The working alliance in coaching: Why behavior is the key to success’, in Journal of Applied Behavioral Science, 51(2), 177–197, 2015
- A M Grant, L Curtayne and G Burton, ‘Executive coaching enhances goal attainment, resilience and workplace well-being: a randomized controlled study’, in The Journal of Positive Psychology, 4(5), 396-407, 2009
- A M Grant, L S Green and J Rynsaardt, ‘Developmental coaching for high school teachers: Executive coaching goes to school’, in Consulting Psychology Journal: Practice and Research, 62(3), 151-168, 2010
- R Jones, S Woods and Y Guillaume, ‘The effectiveness of workplace coaching: A meta-analysis of learning and performance outcomes from coaching’, in Journal of Occupational and Organizational Psychology, 89(2), 249-277, 2015
- R B Kaiser, J M LeBreton and J Hogan, ‘The dark side of personality and extreme leader behavior’, in Applied Psychology, 64(1), 55-92, 2015
- M F R Kets de Vries, The Leader on the Couch: A Clinical Approach to Changing People and Organizations. Chichester: John Wiley & Sons, 2006
- A Luszczynska, U Scholz, U and R Schwarzer, ‘The General Self-Efficacy Scale: Multicultural Validation Studies’, in Journal of Psychology, 139(5), 439-457, 2005
- M McDermott, A Levenson and S Newton, ‘What coaching can and cannot do for your organization’, in Human Resource Planning, 30(2), 30-37, 2007
- E Nelson and R Hogan, ‘Coaching on the dark side’, in International Coaching Psychology Review, 4, 7-19, 2009
- N Page and E De Haan, ‘Does coaching work? …and if so, how?’, in The Psychologist, 27(8), 582-586, 2014
- Ridler & Co. The 6th Ridler Report: Strategic trends, 2016
- B W Roberts, J Luo, D A Briley, P I Chow, R Su and P L Hill, ‘A systematic review of personality trait change through intervention’, in Psychological Bulletin, 143(2), 117-141, 2017
- L J Stewart, S Palmer, H Wilkin and M Kerrin, ‘The influence of character:Does personality impact coaching success?’, in International Journal of Evidence Based Coaching and Mentoring, 6(1), 32-42, 2008
- T Theeboom, B Beersma and A E M van Vianen, ‘Does coaching work? A meta-analysis on the effects of coaching on individual level outcomes in an organizational context’, in The Journal of Positive Psychology, 9(1), 1-18, 2014