Wednesday, 3 October 2012

Measuring and Rewarding Teacher Effectiveness

Recently, a teacher-friend sent me a link to a resource for measuring teacher effectiveness. There are many such resources out there and most Western education jurisdictions now have some sort of initiative that is designed to measure how 'good' individual teachers are. Most of these initiatives are still housed in the performance processes of schools and systems and have not been directly linked to teacher pay scales or other reward mechanisms, or at least not as tightly linked as many 'reformers' would like...

CAN you 'measure' the effectiveness of individual teachers?

In short, I think you can measure the relative quality of individual teachers, or at least get a very good approximation. I certainly acknowledge there are alternative viewpoints out there regarding this!
Measuring this sort of broad, 'intangible' concept - a latent variable, to use a statistical term - is statistically achievable, but it requires a complex and detailed measurement approach. The statistical process of measuring the broad, difficult to quantify concept [eg: the quality of a teacher], is to measure the observable traits of the given concept. In other words, you measure what can be easily quantified and directly relates to the broader concept. A vital aspect of this approach, is to take into account as many of the observable traits that make up the broader concept as possible.

HOW can you accurately measure a teacher's effectiveness?

One thing I liked about the resource linked to earlier, is that they acknowledge the importance of multiple methods being used to gather information about a teacher's performance. Specifically, they cite 'student test scores', 'classroom observation' and 'surveys'. I agree that each of these are valid sources of information about a teacher's performance. 
However, the key to ensuring that a measurement tool / instrument / process is as valid and accurate as possible, is to emphasise [or 'weight'] the items that give the best information about the broader concept being measured.
In the case of measuring teacher quality, both student results and surveys are external indicators of a teacher's performance. Both of these sources of information can be impacted upon by other factors, outside of the teacher's control. 
Classroom observation, IMHO, is the most significant source of information in any determination of teacher quality - how can you accurately judge someone's capacity to perform, if you haven't watched and analysed them performing!?!
So we need to watch teachers teaching and analyse the wide variety of observable traits that make up a teacher's practice. The more specific and deconstructed from larger concepts that these observable traits are, the better. For example, one domain within the broader concept of 'Teacher Quality' would likely be something like 'Professional Knowledge'... This domain should be broken down into a series of items that are more easily quantified. eg:

  1. to what extent do the teacher's plans reflect understanding of the content?
  2. to what extent do the teacher's plans reflect understanding of the students?
  3. to what extent do the teacher's plans reflect understanding of the available resources?
  4. how appropriate are the teacher's instructional outcomes?
  5. how effectively do assessment methodologies match the goals of the learning sequence?
By rating / scoring the broad range of specific traits that are part of a teacher's role, we can 'build' an accurate picture of the relative quality of an individual teacher.

What about performance pay..?

My own views on this are evolving... 

I don't think the current system of paying people based upon how old [oops... 'experienced'... ] they are is fair or motivating.
Like anyone who has worked in a school, I've seen several examples of gun early-career teachers working for significantly less money than old, outdated, uninspired, ineffective teachers - purely because they haven't 'been there' for as long! I don't think this situation is fair and I think it can negatively impact upon staff harmony, as the underpaid guns get frustrated with the overpaid ineffectives... 

However... the science and research regarding motivation quashes the notion of financial incentives being effective in raising performance. Indeed, Daniel Pink, in his acclaimed 2009 book, Drive: The Surprising Truth About What Motivates Us, reveals that using financial incentives to motivate people to perform better invariably leads to worse performance!
Pink does not suggest that we work for nothing [phew... ]... but his research found that, if we "pay people enough to take the issue of money off the table" [page 33], then any further financial incentives are ineffective in terms of improving performance and, indeed, can often have the opposite effect, by way of reducing performance over time. Instead, Pink identified three aspects of intrinsic motivation as the best, long-term motivators in the workplace:

  1. autonomy
  2. mastery
  3. purpose

Aside from the ineffectiveness of financial incentives for improving performance, I fear the competition element that this would inevitably ignite between teachers. Teaching needs to be a co-operative, collaborative profession - initiatives that are designed to have some people 'succeed' [earn the financial rewards] and others 'fail' [miss out on the financial rewards] are doomed to see teachers taking more individualistic approaches to improvement and development.

So how should we use information about individual teacher quality?

I won't go into professional improvement processes here, but in terms of how we remunerate teachers, perhaps a fairer and more effective way than present is to pay all teachers a common salary - "take the issue of money off the table"... 
If we are going to reward extra effort, high performance, etc, perhaps we focus on intrinsic motivations and, if we do use extrinsic motivations, use them in a 'now - that' fashion, rather than 'if - then'... eg:


  • provide greater autonomy to high-performing teachers - "Hire good people, then leave them alone." [William McKnight, 3M President, 1930-1940]
  • provide greater opportunities for high-performing teachers to pursue mastery. eg: provide time / resources / etc. for the pursuit of further knowledge and skills, or allow greater time to work on areas of passion and interest.
  • give greater purpose to the role of high-performing teachers - create opportunities for them to lead and influence more broadly.


  • 'now that you've done [x, y, z... ], here is [insert reward]... '
  • don't make these external rewards the reason for performing well at the task [being a good teacher!].

Phew, that was a long post - what was it again that I promised myself for finishing this..?


  1. Hey Peter - I think your thoughts on merit pay are bang on. Merit pay is that idea that never works but never dies (Ravitch). Teachers need to be paid well but more importantly, they need to be included in the purpose and the chance for mastery. Autonomy is a powerful tool but the key is PROFESSIONAL autonomy. Effective teachers use their professional autonomy to learn, reflect, take risks and always hone their practice.

    I am not sold that we need a standardized evaluation tool. Education is something that has so many ways that work that it is difficult to pinpoint measurable qualities. As you said, teachers and admin in classrooms participating in professional learning dialogue can be so powerful.

    1. Hi Chris,

      Agree - the real 'drive' comes from having the autonomy to apply one's own creativity and pursue professional growth in a personalised fashion.
      Take your point about measurement of teachers - it is a difficult and complex task. Precisely why we need to steer away from any initiative that attempts to simplify [eg: over-reliance on student test results... ] this 'intangible' quality.


  2. One point I forgot to mention is that Pink's work is based upon the work of Ed Deci and Richard Ryan... my favourite quote from them - "we cannot motivate others... but we can create the conditions for others to motivate themselves". What conditions are we creating for our teachers?

    1. Agree fully - motivation is the essence of leadership. If staff are motivated, there is no need for carrots or sticks...