Article Categories
[ Show ] All [ Hide ]
Clean Language
Article Selections
[ Show ] All [ Hide ]
First published in Rapport, November 2008.
Based on a paper by Doyle, Walker, Nixon, Walsh and Michell presented at
the British Educational Research Association Conference, 2007.

Cleaning up the ‘F’ word in coaching

Nancy Doyle and Caitlin Walker

Coaching has become a central part of CPD and work based learning.  Within the organisational framework, coaching is seen as a fundamental moderator between individual skill and performance outputs – the mechanism by which we can learn and upskill.  Since we know that this kind of 1:1 attention, apprentice-style learning, is high effective (Kolb, 1984) it all sounds spot on.  So what’s wrong with coaching? Well, it all depends on the quality of the feedback ….

Feedback can be a ‘dirty’ word in some organisations.  Managers hate giving it – they skirt around negative issues, over generalise and fail to give sufficient details (Levy & Williams, 2004).  London (2003) reports that receivers of feedback also delete and distort it so as to avoid personal responsibility for things they didn’t like and take credit for those they did like (or vice versa, depending on personality)!  Early research on feedback found these accuracy problems insurmountable whilst at the same time concluding that objective feedback was critical to the coaching process having any effect on performance (Kluger & DeNisi 1996).  Molden (2007) explains the most useful feedback is free of judgement about its nature, intention, emotion or any other type of external change. It should be seen as ‘clean’ – like a scientist in a laboratory test describing precisely what you see and hear.

Nancy Doyle and Caitlin Walker, of Training Attention Ltd, have spent the past 5 years developing a model of feedback which is easy to send and easy to receive.  It can be hard to learn but once you have adopted Clean Feedback you need never be afraid of it again.  Rather, it becomes a nurturing part of the learning process.  We’ve been working with Sarah Nixon, Barbara Walsh and Beth Mitchell at Liverpool John Moores University to test the model with trainee teachers.  We presented the research so far at the British Educational Research Association conference in September 2007.  Sound to good to be true?  This is how it works…

A learner needs to know exactly how she did something wrong, so that she can avoid the behaviour in the future.  Equally, how exactly did she do that right, so she can repeat it?  Here are some examples of real feedback that was given to trainee trainers:

Feedback set 1
“You weren’t aware of the groups needs”
“You were friendly and welcoming”
“Your style is like a school ma’am”

This feedback is not ‘clean’ – it is full of interpretation and assumption.  It does not tell me what those trainers did or said and I have no way of replicating or avoiding the behaviour.  Compare it to the following:

Feedback set 2
“You talked with your back to the group while you wrote on the board which meant some people couldn’t hear you.  My interpretation of that was that you weren’t aware of the group’s difficulty in hearing you from the back”.
“You smiled and introduced yourself personally to each person as they came in the room.  That felt warm and welcoming to me”.
“You spoke very quickly and loudly at first, whilst standing up, which caught their attention straight away, like a school ma’am”.

So what could I learn from this?  
1. That talking with my back to the group is inconsiderate;
2. personal introductions and smiling is friendly and;
3. standing whilst talking loudly gets attention.  

Can I repeat those behaviours when I want those results? Yes.  Can I avoid them when I don’t want attention or do want to be considerate? Yes.

We call the nature of the second feedback set ‘sensory specific’ i.e. it refers to what I have seen or heard through my senses, rather than what I have made up or interpreted from it.  This is carefully separated from the feedback giver’s interpretation.  The first set of feedback is conceptual and symbolic.

Audia and Locke (2003) describe ‘cognitive elaboration’; the process by which receivers of conceptual and symbolic feedback discern which of their actions caused it.  If I receive feedback like the first set, I have to process it through all my own beliefs about myself, the feedback sender, his / her intentions, through my own capacity for self knowledge and change.  These loops of thought require an investment of time and are subject to inaccuracies as described above.  The second feedback set works it out for me – the sender has to do the cognitive elaboration in clarifying which actions led to the evaluation.  However, does that make the evaluation accurate and do I therefore have to accept it?

With Clean Feedback the answer is no.  Talking with your back to the group doesn’t matter to some, personal introductions are time wasting for another and standing talking loudly like a school ma’am can be overbearing and a turn off.  However, we do have to accept that for the sender of the feedback, the evaluations were true and demonstrate their personal preference.  Now I have a choice.  I can accept the feedback and change my behaviour or not.  This is the double edged sword of feedback in coaching.  Personal preference is rife but is it irrelevant?  

Again, the answer is no. Since all communication and social learning is littered with personal preference perhaps instead of avoiding it we should honour it.  It’s good to know the personal preferences of your boss and co-workers!  The trick is to expand your coaching process to ensure that each individual is getting feedback from at least 3 different people.  The sensory specific information will ensure that you are clear about your actions and from 3 people you can get a broader understanding of how different people interpret you behaviour.  If everyone tells me that my back being turned is inconsiderate, I know this is something I have to work on.  If only 1 person does then I can become aware of it, modify my behaviour in that persons’ company and look out for others who might feel the same.  The point, is that if anyone gives you feedback, you have to assume there will be other people in the world who feel the same.  The question is how many, and should it affect what you do?

We have observed that this kind of giving and receiving feedback fits best into a peer group coaching model, where everyone’s opinion is valued equally.  This fits with research that suggests that feedback from multiple sources is more effective (London, 2003).  We’ve also found that in order to break down the barriers to giving and receiving negative (or ‘developmental’) feedback we have to set up a protocol which requires each sender to give 1 positive, 1 negative and 1 development piece in each interaction.  By framing the negative as a required action decreed by the process people find it less personal (Wang, in press).

Figure 1: The Clean Feedback Model

Something that you said or did that worked well for me was …
I interpret this as meaning …

Something that you said or did that didn’t work so well for me was …
I interpret this as meaning…

Something I prefer you to say or do is …
I interpret this as meaning …

We’ve implemented this model with each other since 2002.  We’ve built on the wisdom of organisational trainers over the years who are clear that feedback needs to be observable and balanced (i.e. positive and negative).   Our contribution to the received wisdom about giving and receiving feedback, we think, is the way we loop the interpretations back to learning about your colleagues.  This makes the model useful as a systemic group learning tool – a dynamic process that builds teams.

We started training it to others when working in a failing school, as a remedy to the belief that coaching is a punitive measure and to help teachers become less isolated from each other.  Since then we’ve used it in business coaching, teams where communication was fraught with conflict, with groups of long term unemployed people and we’ve taught it to pupils as young as 11.  We use this feedback model as a mechanism to engender peer coaching and social learning.  

For more information on the research at LJMU, Clean Feedback for teachers and pupils or using Clean Feedback to create peer coaching networks please contact


Audia, P. G. and Locke, E. A. (2003).  Benefiting from negative feedback.  Human Resource Management Review, Vol. 13, pp 631 - 646

Kluger. A. N. and DeNisi, A. (1996).  The effects of feedback interventions on performance: A historical review, a meta-analysis and a preliminary feedback intervention theory.  Psychological Bulletin, Vol. 119:2 pp 254 - 284

Kolb D. A. (1984) Experiential Learning.  Prentice hall, Eaglewood Cliffs, NY

Levy, P. E. and Williams, J. R. (2004). The social context for performance appraisal: A review and framework for the future.  Journal of Management, Vol 30:6 pp 881 - 905

London, M. (2003).  Job Feedback: Giving, Seeking and Using Feedback for Performance Improvement.  Mahwah, NJ: Lawrence Erlbaum Associates
Molden, D (2007). Managing with the power of NLP. NJ: Prentice Hall

of original paper:

Background to topic

Feedback is an essential part of learning and improving performance. Methods of feedback delivery in addition to, characteristics of the source and recipient may influence improvements in learner performance. However problems exist in the feedback process which relate to possible misinterpretation and influence (in both source and recipient) and specificity of the meaning.

The clean feedback model (Walker 2006) which is designed to keep feedback separate from emotional reactions, acknowledges and actively utilises the bias of the source and the recipient, and gives clear guidelines as to how to make information specific and behaviour based. This is done using ‘clean’, sensory based content.  This means being clear about the difference between a) what the recipient actually did or said and b) what the source interprets from it.  This enables the source to give information about how the learner could alter their behaviour to achieve different results, rather than to a value judgement being made about what the person is like.

The present research aims to bring clean feedback to trainee PE teachers to enhance their skills at giving and receiving feedback with each other, their mentors and their pupils. This action research is currently being undertaken will be evaluated in June with results on the impact and successes available by July.  

Research question and or focus of enquiry

In the relation to the development of trainee teachers, the research sets out to find out whether the clean feedback model can lead to improved learning, performance and coaching?  How this model might impact on creating successful peer coaching networks and what is the impact of giving and receiving clean feedback on the learning experience of trainee teachers.

Research method and / or mapping of literature

A group of 12 PGCE PE trainee teachers and 3 staff went through two intensive days of developing their skills with the clean language model and the feedback model. Written examples of the trainees lesson evaluations and observations on themselves and their mentors were randomly selected for analysis. The participants were given feedback on their evaluations and observations on a two weekly basis to enable them to develop the clean language technique. Changes and developments were evidenced which helped to record progress. Individual and focus group meetings were conducted with participants, school mentors and groups of children to determine effects of the model in practice.

Analytical and/ or theoretical frame

The clean feedback model actively investigates and uncovers an individual’s way of constructing their world. Feedback models which seek to reduce or eliminate bias are a false endeavour, since the setting in which we perform are themselves biased.  Instead we use the understanding of bias to deliver feedback in a way that will be palatable to the received and will enable the source to learn as well.  Our theory is that feedback delivered in this way will lead to greater improvements in performance and a systemic approach to learning for the teachers and the pupils. 

Research findings and/or contribution to knowledge

The research builds on the established knowledge base about the difficulties and errors in delivering and receiving feedback by proposing a tool for communicating performance.  This will contribute to improvements in practice in training, teaching and coaching.  It will also offer a new direction for theoretical development on the nature and purpose of bias or social constructs of learning.

 »  Home  »  Applications  »  Business & Organisations  »  Cleaning up the F word
Article Options

Clean Language
Symbolic Modelling

launching soon...

James Lawley
Penny Tompkins

register your interest for the launch discount

more info
view all featured events