Still waiting for Eureka: the problem with schools as a science

Image source:
Image source:

This is the first post with a particular theme in the series “Why is everyone an expert of education?“. The series is written collaboratively by Ira Socol, Dr Greg Thompson and Tomaz Lasic.

Science is an awesome, admirable pursuit of measurable, testable, replicable truth. It is a pursuit of certainty that something has, does and will work as we imagine it. We want to be certain that a medicine we give to our child has been tested and declared safe, a bridge we drive across stable enough to carry us. Data and methods we use give us control to predict, improve, avoid. That’s all fine when dealing with inanimate objects. Yet ask an engineer and they will tell you that even concrete and steel could behave unpredictably, hence the overdesign and consideration of safety margins. Enter humans.

Data and scientific methods give us the delirious joy of thinking that things are repeatable and replicable with human beings too. Because something works/does not work in one context, we assume that that must be the case for all contexts. This applies to curriculum, to standards, to teaching strategies, to discipline and management strategies, to educational psychology, to the very organisation of schooling.

Scientifically observing and measuring actors in schools fools us into believing we are capable of absolute knowledge, absolute categories, and once we got those right – absolute solutions. Fools us? Just look at performance pay for teachers debate, curriculum wars, ad absurdum arguments over accuracy of assessment, validity and usefulness of standardised, statewide literacy and numeracy testing and more. In all these we want to measure so we can be ‘certain’, to establish norms and through them (as the term suggests) normalise people and put them in brackets. Once they are in boxes, brackets and percentages, we can apply the right dose of whatever it is we have come up with as the solution.

Now please – categorisation is a perfectly human thing and helps us live our daily lives, no doubt. But when categorisation becomes unexamined, unquestioned purpose of what we do with human beings in education, then we have a few problems. Or put bluntly: It is rubbish, and it bewilders and alienates students (and with it many teachers, parents and others).

When seeing education as a science, we tend to break it down to its elements, particularly those we are familiar with. Instead of having a more holistic view of an actor in the process of schooling, we prefer a cleaner, reductionist, building-block view. We build walls around and guard our areas of expertise. There we propose solutions to fix things we are certain of. And when we want to justify, we look for figures. In this need, the efficient simplicity of “68% on the test” becomes very seductive. We are insecure in trusting the ‘soft’, slippery immeasurables like intuition, freedom or hope etc. so we invent ever more complex, ‘hard’ systems of quantifying people to cling to and value: “Is this a 72% or 74% answer?” Sounds like “is this Class B or Class C concrete”? But hey, “it has worked for me at school so why not for these kids too, they should be motivated by it.”

So, apart from all having a lot to say and guard as ‘experts’ – where does this lead us?

The problem with education as a science is amplified by the centralised nature of most school systems, where control is exerted from a distance. If you remove relationships from schooling (let’s face it, the majority of system administrators/policy makers spend very little time dealing with relationships in schools), what is left is the idea that schools are largely the same, and therefore, the same tests should elicit the same responses. If they don’t there must be something wrong – and the finger is normally pointed at teachers!

In Australia, USA at least, and we imagine many other countries, the people making decisions on education have probably not done more than tour a school in the last twenty years, yet they still are able to provide ‘expert’ commentary. A few problems rise immediately from this.

First is the modernist view of scientific solution – a belief in the inevitability of human progress through science. These decision-makers enter the debate with a belief that each adaptation in the educational system has been “progressive,” and thus that the education system which they succeeded in was “near perfect” – the result of centuries of continuous improvement. A bit of “tinkering”  (Tyack & Cuban, Tyack & Hansot) will thus result in “perfection.”  Their authority flows from certainty and the promise of being led to such perfection. The second problem is the idea that success is quantifiable and that the idea of it applies to all of us equally. The third problem is the leaders presuming that their framing of experiences is both correct and applicable universally.

This all adds up to create an understanding of education as a science where results are replicable and reliable. We call this phenomena “white coat replicability”. Governments, the media, and bureaucracies deal in white coat replicability when they talk about effective schools and teachers, about ranking schools based largely on benchmark testing.We are hearing a lot in the media at the moment about Joel Klein and New York public schools. Consider this article which talks about the ways that data collected can be misused and misrepresented.

Today, schools are dominated by externally developed testing, reporting on student achievement that uses mandatory standards and systems, and continual reform. What is this doing to students? Do we think that schools cater more or less for student difference and uniqueness now than they used to? After years of education as a science, the system is at least as discriminatory as ever. We think that we collect and control so much information now, it is slowly paralysing us from dealing in intangibles such as freedom and hope.

The interaction between students and teachers is becoming less relational and more normalised, where outputs are measured and teachers and students are measured and measure each other. The problem with this is that the measurement fixes students as good/bad, successful/unsuccessful, likeable/unlikeable and students increasingly live/perform these measurements. Ever wonder why there are groups of students who are Rebels in every school regardless of context? Schools need rebels to show students who not to be.

This fixing of students into neat packages are things that schools do all the time. After all you can measure deviance/compliance can’t you? Students find these packages very hard to change. We are indebted to the work of Stephen Ball from the University of London and the notion of performativity.

To put it simply, students and teachers are trapped in their performances as a result of the ways they are measured and organised within education as a science. The trick of performativity seems to be that individuals come to see themselves in terms of the data collected. They perform as they feel they should be expected to/come to expect themselves. For example, a “dumb” student may hardly try to do well in a test because he does not see himself smart – and the grades simply confirm it. The data assigns the well-known and well-rehearsed roles. These roles are so reified by repetition that even altering the performance to accommodate (a change in) their self-image becomes unacceptable, much like when one actor discomforts audiences if he/she replaces another in a television series. When students act out of their accepted role (for example, “good” girls getting drunk), they experience negative feedback, often from students as well as teachers. Ontologically, schools teach students to limit their being – limit who they are and can be. Change anyone?

You can’t collect all this data and employ these evaluators in schools without people performing their results (positionality). Want to know why many students feel alienated from schools? One of the key factors is that they are continually judged as being less or even non-successful. The same is true for teachers! Who hasn’t felt despondent when the school they are working in has performed at unacceptable levels in their subject through something like NAPLAN (Australia’s nationwide literacy and numeracy testing)? One of our colleagues had this experience and the school spent the next 6 months ‘diagnosing’ the problem, counselling the subject teachers, exhorting the students to do more, aim higher etc. The end result? Teachers were forced to run after school study classes, they began to ‘teach for the test’ and the school results stayed the same. The drop in teacher morale was curiously never measured, or seen as significant. We don’t think that the educational experiences of the students would have improved as a result either – teaching for the test negates most of what is wonderful in the relationships between teachers and students and the richness of student learning.

So what is being learnt when education is run and organised as a science? That your performances measure you – that you are quantifiable. You learn how to see yourself as a particular type of student/teacher, and grow to see this as ‘normal’ – the business of schools becomes normative measuring and pedagogy becomes sterile, limited, controlling and superficial. But gee it looks clean!


  1. Matthew Ferrinda

    I will admit I am slightly exhausted after a full day at uni, whilst juggling a social life and earning a living, but hey just the joys of being a student teacher!

    On topic, from a quick read of this great post I’ll add my two cents as a future teacher.
    Having recently being part of the education system as a student I now can express feelings from both points of view. I do agree that there are social constructions of stereotypes in schools that students and teachers alike adhere to and conform. I have seen this as a student, as a teacher, and even through my family and friends. Discussion today during a tutorial at uni when talking about assessment and what to teach next (marking a literacy work sample) the suggestion came up to send ‘struggling’ students for remedial help, this was quickly shut down by the class and tutor. This ‘labeling’ of students creates stereotypes and a mentally that leads them to believe they cannot achieve, as future teachers and past students we understand the damage this can cause!

    Drawing nearer to the end of my degree (3rd year second semester) there is a ‘drilling’ of what assessment is. Safe to say we are learning not to teach to the test and to integrate assessment into our everyday curriculum. Hopefully I’ll be able to master this term ‘assessment’ in my class and use my students work samples to better their learning and my teaching practice! Time will tell!

    The general consensus from my peers is we want to know where the system is going from here, and what do these ‘experts’ expect us to do next?
    The message is being drilled into us though to not fall into past practices of treating education as you put it a ‘science’ which in many places it still happening. WE hope a change will come… but when and how…we do not know

    P.S. Apologies for the rant, just a few things that came to mind after a long day at uni thinking all about assessment and teaching!

  2. Deven Black

    Another problem with scientific approaches to education is that the science practiced has minimal validity, The “scientific” examinations of education fail the most common tests of validity:

    1. There is no double-blind testing – at best the examiners know who they are testing, what they are expecting to see, at worst they know what they want to see, and either situation can affect what they actually see and distort conclusions drawn from marginal results. In valid scientific testing the examiners do not know who they are testing because of the use of…

    2. Random assignment of test subjects — in valid scientific tests assignment to groups is completely random for the purpose of having mostly identical subsets within each group, but schools don’t work that way. In schools, classes are engineered as much as possible to create certain conditions in the class and in the school. This raises the problem of….

    3. Peer effects — the performances of any individual student, and often groups of students, are effected by the students in the rest of the class and the class culture as evident in student actions, not teacher beliefs. If you put a student in a class with “better” peers he will perform better; if you put him with “worse” peers he will perform worse.

    This all leads to the single biggest problem of education science, no placebo. In biomedical testing one of the randomly chosen groups gets the treatment and the other randomly chosen groups gets a placebo, a sugar pill or some other innocuous thing. Neither the subjects of the test or the examiner know which group is which until after the all the data is gathered. Only then is which group got the treatment and which got the placebo revealed. If groups are not randomly chosen using a placebo is meaningless because you’re not comparing apples to apples. Using a proper placebo protocol may be impossible in an education setting, but it is the only way to give solid empirical evidence instead of prettied-up anecdotal reporting.

    The final problem of education science is generalization. Generalization is the most difficult aspect of any scientific inquiry, especially so for those dealing with policy. Generalization usually requires reproduction – separate double-blind trials conducted by other people in other places that get the same results as the original. This, too. may be completely impossible to do, but it is the only way a conclusion drawn from a “scientific” approach can be proven to have validity.

    The main problem with “data-driven” education practice and policy is that the data used fails every validity test applied to scientific inquiries. The net effect is that all education measurement is anecdotal, and anyone’s anecdotes are generally as worthless as mine.

  3. Victoria

    Firstly, great starting image! It made me laugh quite a lot 😛

    You bring up some great points, but as someone who has been through ‘normal’ schooling, open-learning schooling AND home schooling, I’ve actually preferred the more ‘constricted’ style of education right up until I started my TEE.

    Open-learning schooling, although good in concept (it allows you to learn at your own pace and of things that take your interest) I found that because there weren’t certain boundaries and goals to achieve, that I (and many of my peers) weren’t learning at a steady pace and to the best of our abilities. To me, the boundaries in curriculum based schooling allows a framework for students to see what they could possibly achieve. In senior school, however, I found this framework more constricting, mainly because I was able to think more constructively and criticise content presented to me and my peers. I am now able to critique the framework that once I once followed and appreciated.

    I don’t think everyone fits into one particular way of learning, and THAT is the downfall of the education system we know today. If we didn’t put so much emphasis on teaching the curriculum in ONE particular way, and MAKING the students and teachers conform to this, then I feel there would be a higher level of education all over the world. What we, as students and teachers, need to do is not so much individualise EVERY student’s education, or even continue just having “good/bad, successful/unsuccessful, likeable/unlikeable” groups, but provide a wider selection of schools and/or programs that will allow the individual to learn to the best of their abilities. Yes, you would still have to have grades and goals, but they would serve more as a milestone and indicator for self progression. This would result in less rebellion within the system, because there would be something for everyone and therefore less limitations.

    I don’t think it’s so much the schooling system that teaches us to file different people under certain ‘types’, but the worth society places upon being ‘educated’ in a certain way. However, by allowing for an alteration in the education system, this social worth would contain more equality, and possibly allow for entire nations to progress together.

  4. Joseph Thibault

    You’ve outlined the dangers of scientific education well, my thought is that we do this (normalize our experiences) because it makes things easier to code later. Certainly saying that “this result is attained by this inputs, so more of this inputs!” is a leap of faith because, when it comes to children, adults and human beings in general, it’s impossible to measure/account for all of the variables that come into play. Learning is subjective, but it’s much easier to plan for, execute and deliver if we view it as an objective practice. Which is why textbooks are largely still the norm (canned lesson plans worked for a test group, they must have at least a similar effect elsewhere, or so the thinking goes).

    The argument is that this thinking is detrimental, and if it is not the way we should go, then what/where/how?

  5. Rob Abbey

    Dear Tomaz While I love the intent and content of this discussion I am forced to question the assumptions behind any easy equating of education being a science.
    1. You argue(roughly) that science is a good since it examines things in great detail and gives much good knowledge
    2. Things are best if examined,
    Therefore education should be a science.
    The conclusion begs the question. There is a logical fallacy in this (much over)simplification of the valuable discussion above. Why is it a good that education should be scientific? This is the hidden assumption in the topic. It is difficult to genuinely examine that hidden assumption of “Education should be a science” without first asking is scientific knowledge the only way to get to truth? Is there substance in the notions of education being an art rather than scientific? Are there limits to scientific strains of knowledge? What is scientism? Where and when does education become ideology (just as science proponents often fall into ideological traps). Accepting the task of examining “School as a Science” I am left with a thousand prior questions here. What if thinkers like Popper or Polyani are right about knowledge is personal? and science is ultimately unprovable. You often correctly link knowledge with power and domination and notions of education as a science too often neglect this dimension. Where is the role of morals in science let alone education? Is truth(whatever that is) too important to be left to the managerial class? And so it goes. And it was ever thus. Schools will improve “real soon now” I wonder if, and hope, that there will be alternate modes of learning that leave schools out of the picture?

  6. Tomaz Lasic


    Hello old friend. Oh, I think we bat for the same side here! We (Greg, Ira & myself) posit that it is indeed problematic to treat education as a science, pretty much for the reasons you state here. You mention Polanyi, Popper (throw in the good ol’ Kuhn for good measure)…they are but some of the first, if not most prominent, thinkers who have questioned the ever-shifting ‘rightness’ of scientific approach to ‘managing’ people, losing the all important nuance of (personal) context. I think it is some of fearless edu-leaders (enter Gillards and Kleins of this world), who commit the logical fallacy you speak of.

    Hope of the alternative? In my lifetime I hope. School as we know it is broken but until the same sources of power that have sustained it over the centuries benefit from it (and others convinced it is the only way while accepting it…enter Gramsci) it will be a bloody hard task to move beyond the nice words and superficialities.

    Oh, and thanks for the encouragement for my PhD. You’ll be one of those people I’ll be blaming in the doldrums of writing 😛

    Cheers Rob and thanks for chipping in to the dialogue.

  7. Pingback: Tweets that mention Still waiting for Eureka: the problem with schools as a science « Human --

Post a comment

You may use the following HTML:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>