‘No excuses’ dealt a devastating blow?

Embed from Getty Images

On this blog, I have repeatedly expressed unease at the terms ‘no-excuses’ and ‘zero-tolerance’ when applied to schools. I have never visited a school that describes itself in this way, but I doubt whether the phrase is completely accurate. I am also uneasy about much of the criticism these schools attract because it seems to be a complaint against any kind of strong behaviour policy at all. And I am in favour of strong behaviour policies.

I was therefore interested to read a new critique, based on research funded by the Australian government. The article cautions us against adopting what the author, Dr. Linda Graham, sees as a British model of ‘no excuses’ schools. Graham draws on statements made by Jonathan Porter, a UK ‘no excuses’ advocate. Porter’s statements suggest that such policies enable all teachers to teach, not just the particularly charismatic or experienced ones.

I share Porter’s reasoning when I advocate strong behaviour policies. Students should respect their school as an institution and therefore they should respect any teacher who represents that institution, unless and until that teacher does something to forfeit this respect. There should be no need for a long and tortuous process whereby every individual teacher has to work to earn this respect. Not only is this wasteful of teaching time, it is inequitable. Teachers can be the victims of racism, sexism and homophobia, and experience suggests that tall, middle-aged teachers who have been at a school for a number of years and who hold positions of authority, find it easier to earn respect than small teachers, new teachers, young teachers, teachers with strong accents, casual relief teachers and so on.

Graham challenges this position, arguing that the factor which impacts most upon classroom behaviour is the quality of teaching. If students are forced into compliance, they cannot communicate that the quality of teaching is poor and so ‘no excuses’ offers cover for ineffective or lazy teachers.

I am dubious about this contention because I don’t accept some of the paper’s premises and I don’t agree that it is supported by the research evidence presented.

Firstly, I am deeply sceptical of the idea that poor behaviour is a form of communication. If that were the case, why would humans spend so much time and energy trying to hide their poor behaviour? Poor behaviour derives from personal and situational factors that are complex, and it often seems motivated by the desire to please friends or to avoid challenging or boring tasks or tasks that threaten the ego. Sometimes, it seems to be about asserting a position in the peer hierarchy. None of these motivations are necessarily about sending a signal to a teacher or other authority figure.

Secondly, the author seems to be worried that ‘no excuses’ is the project of a shadowy group of figures who I don’t recognise:

“The most recent strain of the crisis rhetoric appears to be driven largely by teachers identifying as ‘neo traditional’, particularly in England where these teachers are well-connected by social media. The neo traditional teacher favours teacher-centred instruction, deplores inclusion and differentiation, and promotes strict whole-school ‘no excuses’ discipline policies modelled on an extreme interpretation of behaviourism.”

I know a lot of people who favour teacher-centred instruction and it is true that the advocates of ‘no excuses’ tend to fall into this camp. However, I don’t think they identify as ‘neo traditional’ because this oxymoronic term is mostly used pejoratively. Neither do they ‘deplore’ inclusion or differentiation; a strangely emotive term to use in an academic paper, particularly when you realise that this paper was published in the International Journal of Inclusive Education and you consider the likely views of this journal’s readers.

I am aware of some well reasoned critique of common approaches to inclusion and differentiation – I’ve written some myself – but this is far more nuanced than Graham allows. Does Graham wish to force us into a binary; that we must either accept inclusion and differentiation entirely on her terms or we must ‘deplore’ them?

From there, the paper becomes quite strange. In an attempt to show that ineffective teachers are the major cause of student misbehaviour, we are presented with a series of anecdotes about two teachers who have been observed as part of a research project. Mr Smith is a really bad teacher who manhandles the students and is constantly berating them, often unfairly. Miss Jones, on the other hand, is really good. She sets clear rules and routines, such as getting the children to sit with their hands on their heads, and reinforces these with rewards. She is also really nice and says, ‘bless you,’ when a child sneezes. (I’m just imagining what the reaction would be on Twitter if a ‘no excuses’ school ‘forced’ children to put their hands on their heads, but that’s another matter…)

These observations left me feeling uncomfortable. I am pretty sure that if I were one of these teachers and I read this paper then I would recognise myself from the details given. The way these passages are written, and the way the teachers are described, reminded me that Graham is an academic who has never been a teacher herself. Mr Smith is not described at all sympathetically.

Moreover, I find it odd that Graham builds her case based on anecdotes about just two teachers and I doubt whether such evidence can prove anything much. As it stands, it actually offers some support for the kind of behaviourist approach espoused by ‘no excuses’ advocates. The obvious point to raise is that these two teachers are not working in a ‘no excuses’ school. Perhaps Mr Smith would be a more effective teacher in an environment where rules, routines and consequences were determined at a whole-school level – rather than one where these decisions were his to make – and where he received some training in how to apply these procedures. We don’t know, but he clearly needs help and support. I’m not sure that he needs researchers earnestly documenting his every failure.

Perhaps aware that her own evidence offers support for rules, routines and consequences, and therefore behaviourism, Graham attempts to use a lesson observation rating scale – CLASS – to show that the teachers differ in other aspects of their abilities and not just classroom management. I am unconvinced by this, given the validity and reliability issues faced by classroom observation instruments of this kind.

And so, there we are.

It’s hard to know what to make of it all. Graham clearly wants to put the boot in to ‘no excuses’. The trouble is that rather than brandishing a boot, she appears to be waving an old sock with the words, “there might be something in ‘no excuses’ after all,” stitched into it.


30 thoughts on “‘No excuses’ dealt a devastating blow?

  1. Tom Burkard says:

    In 1999 I inspected three of East Anglia’s top comprehensives for the Telegraph Good Schools Guide, and the one which impressed me most strongly was Framingham Earl, which had an unapologetic ‘no excuses’ policy. In every class I observed, behaviour simply wasn’t an issue: everyone was singing from the same song sheet, and class discussions were lively. Lunch didn’t interrept the even temper of the day–there was no sense of taking the weight off the pressure cooker, like it was at the school where I was teaching part-time.

    For the final period I was left alone in the library with all of the Year 11s, who lost no time in telling me what a shock it had been coming from their ‘anything goes’ primary schools and finding that you could get sent to time out just for dropping a pencil on the floor. Yet however swift and sure the sanctions were, they were never harsh or disproportionate, and there was complete agreement that classes were far more interesting when they weren’t disrupted by attention-seekers. One pupil volunteered that teachers were always willing to help you in their free periods ‘because they weren’t stressed out’. And indeed the staff turnover was exceptionally low.

    Later, when I was flogging our Wave 3 literacy intervention, I visited a lot of very good primary schools where teachers had no problem whatever with authority. Now I’m working with three Inspiration Trust primary schools that serve some of the most disadvantaged pupils in Great Yarmouth, and they’re an absolute delight; pupils and staff are cheerful, relaxed.and above all purposeful. Then of course there’s Michaela, which I visited in its first year.

    The problem with educators like Dr Graham is that they cannot conceive that pupils are like adults: they’re much happier when the bosses know what they’re doing, and everyone is working cooperatively and productively. However, too many educators and teachers have never left school, and their perspective hasn’t really moved on from that of the truculent teenager. In an ideal world, no one should teach until they’ve had at least two years’ experience working with grown-ups.

    • I’ve heard your final argument before. I went straight into teaching from university and I sometimes wonder whether I would have a different perspective if I had worked in industry for a few years first.

      • I sort of did it after-the-fact; spent three years in teaching straight after Uni, and then five years out of it before coming back. The experience was helpful, I feel.

        WRT the rest of your piece: you’re being very kind to Dr. Graham.

  2. Greg this is an important debate and I’d like to offer some (hastily prepared) reflections of mine to the mix. My comments are not so much about the merits, or otherwise of “no excuses” behaviour policies, but of how research is done in education, and by whom.

    You say that you are “…..deeply sceptical of the idea that poor behaviour is a form of communication”. This is a statement I often make when I am running PD for teachers stemming from my twenty years of research in the youth justice sector. Our research (and that of colleagues in the USA, UK and NZ) shows that (conservatively) around 50% of young people adjudicated in the youth justice system have language skills that place them in a clinical range on formal, standardised measures. Simply put, this means that these young people are not understanding an awful lot of what is being said to and around them. So “simple” things (on the face of it) like complying with a teacher instruction (particularly if this comes in the form of a slightly syntactically complex sentence) will be challenging, if not beyond the reach of these students. You might want to retort that young people in the justice system are a small, extreme proportion of the school-age body. We would certainly want them to be a small proportion, but they are not insignificant (in any sense) because of this. Further, we have recently found (reported in a paper that is under review) similarly high rates of language disorder (and low literacy levels) in a sample of secondary students attending so-called “flexible” or alternative education settings. These students cannot cope in the mainstream because (even if their emotional and behavioural issues were resolved) they do not have the language and literacy skills to process and respond to everyday curriculum demands. So when we say “behaviour is a form of communication, we don’t mean that students are consciously “choosing” behaviour to communicate with adults – rather this shorthand expression might be better interpreted as “behaviour can be an *indicator* of factors that are not visible to the casual observer” (not nearly as catchy, I know, as “behaviour is a form of communication”. Of course not all students who display challenging behaviour have language disorders, but the comorbidity between these phenomena is much higher than people might think.

    I’d also like to take issue with your criticism of the use of the CLASS (Pianta et al.) tool. Yes, there are “issues” with classroom observation tools, in the same way that there are issues with the tools we use to measure people’s blood pressure, or blood sugar levels, or anxiety states, satisfaction with their partners, or income…or just about anything. The fact that there are limits to tools does not mean we should not use them – it means we should use them judiciously and with a clear understanding of their limitations, while all the while contributing to the body of knowledge that can improve them. Accordingly, I don’t think it’s reasonable to refer to the case examples provided as “anecdotes” when they are clearly data – collected and analysed in accordance with the CLASS peer-reviewed protocols. If we’re not going to use tools such as CLASS to get a systematic understanding of what is happening in classrooms, what exactly should we be using??

    It is always possible that research participants might be indirectly identifiable, either to themselves or others, and that is one of the many ethical challenges we juggle as researchers. In this case though, I would say that we have to weight the greater good of this discussion taking place, against the sensitivities of individuals who may experience discomfort. This applies too when health researchers conduct studies such as Graham’s in hospital emergency departments. Sometimes researchers observe and report on sub-optimal professional practice, and that creates discomfort, sometimes even disbelief on the part of stakeholders, but we all need to move beyond those reactions so we can use the data to improve what is happening on the ground, as well as making upstream changes to pre-service training.

    I am particularly uncomfortable with the notion that because Prof Graham has never been a classroom teacher, her results can somehow be dismissed in one sweep. I am also not a classroom teacher, but was, for example, lead author on the study “Oral language supports early literacy: A pilot cluster randomized trial in disadvantaged schools” (see here – http://www.tandfonline.com/doi/abs/10.3109/17549507.2013.845691). I’m also a lead Chief Investigator on the Classroom Promotion of Oral Language randomised controlled trial (see http://bmjopen.bmj.com/content/7/11/e016574) We can’t cherry-pick when it’s OK to accept non-teachers as researchers in schools, and when it’s not. Medical science would be in a parlous state if doctors rejected research conducted by pharmacologists, biochemists, physiologists etc – many of whom never step inside a hospital, let alone interact with patients, but never-the-less, their contribution to the everyday knowledge, reasoning, decision-making, and skills of medical practitioners is essential.

    Any one piece of research contributes only one piece of a much larger, complex puzzle, and every study has limitations. That’s how we should be interpreting this, and every other study that comes across our desks.

    • I don’t disagree with much of what you say. Your qualification of the claim that behaviour is a form of communication makes some sense, but also, of course, demonstrates that this claim is not an accurate one.

      My criticisms of CLASS are based on my reading of the MET project research. I don’t think it’s a simple as saying all measures are flawed. In the way that CLASS is typically used, I simply don’t believe it is capable of measuring the quality of teaching, which is pretty fundamental. If you follow my link, you will find my detailed reasons.

      The fact that Graham has never been a teacher clearly does not invalidate her research. I was commenting more on my personal response to the way this paper has been framed and how this may make teachers feel, many of whom read this blog.

      Questions about terminology aside, do you think that this data about just two teachers who don’t work in a ‘no excuses’ school in any way supports the contention of the paper? That is the main issue.

      • OK, I need to stop doing this in between other tasks at my desk. Try this response with (hopefully all) typos removed:

        Thanks for responding Greg. My point is that behaviour is something to be observed and understood, and all is not always as it first seems. I have had many teachers tell me that when they consider students’ behaviour through the lens of compromised language skills (especially poor comprehension skills), they make different accommodations in their own behaviour (eg how tasks are presented) which can avert some behavioural difficulties, by better setting students up for success in the first place. Of course all teachers would want to do this, but perhaps that doesn’t always happen as we would like it to.

        I will have a closer look at your link re CLASS when I get a moment, but my question remains: what tools would you have us (in the broadest sense of the word) use instead of CLASS in order to describe, understand, and report on what is happening on the ground in classrooms?

        I think that Graham does present a critical analysis of the construct she is interested in, and uses data to do so. The teachers in question don’t work in no-excuses schools, but they do work in schools that have very high rates of disciplinary suspensions – which was the sampling frame for the study. Like all research, it is open to criticism, and extrapolation from these settings to no-excuses settings may be a valid one. Do we have similar observational data on what goes on in such settings? (this is not a rhetorical question BTW – I am genuinely interested in the answer).

      • “What tools would you have us (in the broadest sense of the word) use instead of CLASS in order to describe, understand, and report on what is happening on the ground in classrooms?”

        I am not sure. Why would I want to do this? If it is simply to describe classrooms then why do I need an instrument like CLASS? CLASS is not just a tool for describing classrooms, it is intended for making inferences about the quality of the teaching taking place (and scoring it), as in the Graham article. This is something I think is largely invalid (unless, as in the MET project, you video a number of different lessons by the same teacher, show the videos to multiple trained observers and not inform the teachers of the criteria being used and, even then, notoriously unreliable teacher value-added scores give a better measure of quality). If I want to be more open-minded in my data collection i.e. not make assumptions about which teacher behaviours are superior at the outset, then I would probably want to develop a similar protocol to those used in the process-product research of the 60s and 70s.

        However, I go back to the question of why I would want to do this. Graham seems to use this data to make the rather trivial point that student behaviour is affected by the teacher. The evidence she provides largely relates to classroom management (although this is quickly elided with teacher quality in general). I think everyone in education would agree that teachers’ classroom management skills affect behaviour in the lesson. That’s why, in my new book for graduate teachers, I have written nearly 8000 words on the topic. The point of ‘no excuses’ or the kinds of strong behaviour management systems that I favour is that each teacher doesn’t have to start from scratch, negotiating their own rules and setting their own norms with every class. This is the default position in many schools in Australia, I suspect, and I don’t think it is healthy for the reasons I have stated; it’s harder for new teachers or those from disadvantaged groups and it leads to inconsistency. I believe a strong behaviour system can mitigate this lottery, support teachers and ultimately lead to happier and more productive schools.

        But I might be wrong and so this is an area that research could address. We could examine ‘no excuses’ schools, for instance, and compare them to other schools using correlational and observational data. It may even be possible to use the random assignment ‘lottery’ system used by some U.S. charters to address this question. Does my hypothesis stand? Simply pointing out that teachers affect behaviour, and using quite extreme examples, doesn’t really address this at all. It might be difficult research to conduct but that doesn’t mean we can substitute it with research that doesn’t address these issues and then make claims as if it did.

  3. doctorgaz says:

    You’re being a big too measured here Greg. Let’s call a spade a spade and call out this “research” for what it is: nonsense.

  4. From what I have read about schools tagged as “no excuses”…..(a misleading term for having a strong behaviour policy as you say) both in the UK and in the US (in the form of particular charter school chains), student voice seems to indicate happy students, low rates of bullying, teachers are less stressed and academic results are high. It could be a worthwhile research project to collect academic and student voice data from a representative sample of such schools to confirm this or prove Graham’s contentions. Unfortunately, I can’t read her paper as it is behind a US$172 paywall, shame.

    • TBF, Graham did have 50 versions made available for free yesterday.
      I think you are correct in your thoughts. It does seem that a number of schools do extremely well with ‘no excuses’ discipline. I think a research project into how these schools get it so right would be more valuable than constantly telling teachers that it is their fault. It’s seems that things can be done to minimise problems without blaming teachers’ choices.

  5. Mitch says:

    I always find no-excuses arguments interesting as in general I try to not be a punitive teacher.
    I must say I didn’t read the paper as it was behind a wall but the abstract did say to me, ‘victim blaming’, which I had hoped we had moved past.
    Once again I think we are talking different things. Many of us may think of ‘no-excuses’ being that misbehaviour is noticed and acted upon whereas others may look at it as mandatory detention where students are directed onto the school-to-prison pipeline for reasons out of their control.
    I believe like you Greg that in reality many of these schools allow for mitigating factors, try and not punish if not necessary and aim to be reasonable. We may hear the occasional ‘outrageous’ story, like students being put in detention because their parents didn’t give them lunch which everyone gets in arms about but we never hear the outrageous stories in the other direction such as students receiving just a disapproving face when they have hit a student or thrown a pencil into the fan.

    • Tom Burkard says:

      This is a common but totally misinformed interpretation of the way a ‘no excuses’ policy works (at least in any school I’ve visited). It’s not a matter of allowing for mitigating factors or using draconian punishments, but rather of nipping misbehaviour before it excalates into a confrontation. Perhaps one of the more notable British examples was Tommy MacKay’s West Dunbartonshire project, which took place circa 1990–2005. He worked with all 34 primary schools in WD, which is second only to Glasgow in terms of urban squalour and deprivation in Scotland. It began by convincing teachers that their pupils could be held to the same standards of behaviour as middle-class children, and after three years this had succeeded to the extent that Dr MacKay was able to introduce a literacy programme based upon synthetic phonics. Within another 10 years, all pupils were going up to secondary school with a reading age of at least 9 1/2. I visited WD and talked to a lot of these teachers, and their confidence and pride in their achievement was palpable.

      An important feature of the WD initiative is that parents were treated with respect, and once they found that their children weren’t doomed to lead the same crappy lives in high-rise slums, it transformed the whole sense of community. Another being the fact that Dr MacKay worked tirelessly in these schools, and this wasn’t just another initiative dropped on teachers from on high.

  6. Greg, just coming back to your comments/query about CLASS in your earlier response to me – interestingly I think you and Professor Graham came to some broad consensus on this in your interaction here a couple of years ago: https://gregashman.wordpress.com/2015/04/03/can-teaching-be-given-a-score/

    Maybe I’m naive, but my answer to the “why would we want to do this?” question is simply to understand a complex ecology better – one that is “high stakes” for both teachers and students. As a social scientist, I grapple in every research project with the validity and reliability of my measures and where possible use multiple forms of measurement for the same construct, so that triangulation is occurring. The bottom line for me is that we all need to titrate research findings carefully and objectively.

    I think rather too much weight has been assigned by some in the Twitterspere today to the fact that Prof Graham is not/has not been a classroom teacher. Until I did research in education, I had never encountered such a sentiment and it still pulls me up in my tracks. I’ve never heard anything like it in youth justice and certainly not in health, where multidisciplinary perspectives are the norm. Maybe I need to get my thoughts together at some point and blog about it 🙂

    • Here is an analogy that might help explain my argument: Imagine that someone was opposed to systematic synthetic phonics instruction and wrote a paper on this. Imagine that this paper began by arguing against ‘barking at print’, by stating that phonics alone is not reading and by warning about the shadowy figures who promote phonics. Imagine that the author then went on to quote evidence from two whole language classes that showed that the reading proficiency of the students was highly dependent upon which teacher they had. Imagine that this evidence was then used to argue that phonics instruction was irrelevant because the major factor affecting reading ability was the quality of the teaching. I doubt whether you would be convinced by such an argument, in the same way that I remain unconvinced by this one.

      • Yes, the analogy does explain your argument, I’m just not sure that it completely holds up. As I noted earlier, the sampling frame for the study was schools with high rates of disciplinary exclusions and the data described in the paper are presented as (in part) a refutation to the “no excuses” argument. Some may see that as an over-reach, and that’s a position I can understand to some extent. If nothing else, though, it has prompted a debate that involves teachers and researchers and that has to be a good thing.

        Now, this particular researcher is not on holidays, unfortunately, so I’ll have to bow out of the discussion at this point, but it has been good to engage, as always, Greg.

  7. Jo says:

    Many academics in education frequently compare educational research to medical research (as Pamela Snow has done here). In the field of medical science, it would be very rare for anecdotal evidence, such as that which has been presented by Dr Graham, to be used to support ways of treating wide spread problems experienced by patients or their doctors. It would be even rarer, for any research that was based on less than several hundred case observations, to be published in a peer reviewed medical journal and used as evidence to support the practise of the majority of medical doctors. Why then, do we insist on publishing and accepting papers within the faculty of education that do not reflect the rigour of other areas of academia? I appreciate that including participants such as children and young adults in research is problematic for obvious reasons, but surely we can support our theories with more than just two anecdotal references? Furthermore, I was always led to believe that the protection of participants, young or old, was paramount.

    • Hi Jo, thanks for your comment and for continuing the conversation. I have to disagree with you on a couple of counts though. First of all, Professor Graham’s paper was based on observational data, collected and analysed in accordance with the scoring protocol of a peer-viewed classroom observation tool; it was not based on “anecdotal evidence”. Secondly, I can assure you that there are many, many studies published in medical and health sciences that have small samples. In the sub-field of medical sociology, for example, very small samples are studied in depth. Sample size is not the only measure of study quality.

      You are right that each study needs to be based on its own markers of methodological rigour, but as a field, we don’t (or at least, should not) change tack based on one paper – we need to look at the strength of the evidence overall. The latter is something I think health and medicine tend to do better than education, partly because they have a longer tradition of doing so. Maybe this also reflects the fact that when medical practitioners “get it wrong” sometimes people die – a great incentive to be cautious in applying new findings too precipitously.

      • SSP advocate in a sea of WL proponents says:

        Hi Pamela, Was not referring to the wider category of allied health (that of SP, OTs or nurses) only medical science ie. that which is pertinent to a medically trained doctor or physician.

  8. Hi Jo the same applies in medicine not all studies are large samples – they are probably the exception in fact. That’s why medicine relies on the overall pattern of findings from the literature (e.g., using meta-analyses), rather than advocating change of practice based on one or a small number of studies. The notion of “levels of evidence” is very important in health, as summarised here: https://www.rch.org.au/uploadedFiles/Main/Content/rchcpg/hospital_clinical_guideline_index/Hierarchy%20of%20Evidence%20holter%20monitor.pdf

    It would be nice to think that all medical evidence comes from large representative samples, but that’s not the case.

    • SSP advocate in a sea of WL proponents says:

      Hi Pamela, Thank you for your response. Just as not all studies in medicine involve large samples, not all studies are worthy of being part of a meta analysis either. As a teacher, I often hear people in education saying “The research shows this and that…” What research??? In terms of comparing education to medicine, if a cardiologist were to recommend a course of medication for cardiovascular disease to GPs, based on a meta-analysis involving studies with two or three participants, they would not be taken seriously (nor would the govt have approved the medication in the first place). I strongly believe however, and I’m sure you would agree, that professionals in education can have just as big an impact on the long term outcomes of the population, as professionals in medicine. Therefore, as teachers, I feel, we would be naive to be basing decisions that may have long term consequences, on meta-analyses that involve studies based on a mere two or three participants.

      • I think we are generally in agreement Jo, though I would say that sample size alone is not usually the deal-breaker determining whether a study is included in a meta-analysis; several methodological factors are considered (e.g. characteristics of the sample, choice of measurement tools), and sample size can be “weighted” in meta-analysis. Another difference in medicine is that panels of experts synthesise the current evidence into clinical practice guidelines and/or care pathways and these are then rolled out for use (and continuous feedback from users to result in improvements). Again, I don’t see this level of agility in using evidence in education, but hopefully we are moving in that direction.

  9. Pingback: A tale of two unions – Filling the pail

  10. Pingback: An interview with Barry Smith – Filling the pail

  11. Pingback: The first step in tackling a problem is to acknowledge that it exists – Filling the pail

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.