The Montessori community are committed to carrying out research to demonstrate that Montessori works. It is only real data that will convince parents to ask for Montessori for their children and the government to provide Montessori as a choice for schools in the State sector.
The process of getting real data was launched in March 2011 with a workshop at the Maria Montessori Institute given by Dr Steve Hughes, chair of the AMI Global Research Committee, who is going to lead the UK community in this research effort.
This is a report from the workshop:
We all gathered at the Maria Montessori Institute in Hampstead on a Saturday morning, coming from all over Europe, some visiting their local Montessori haunt, others discovering the venerable building, and a few like myself coming back to their alma mater. There was quite a lot of socialising and hugs exchanged over tea and biscuits, in a very English way until we finally all got together for what we came for. Steve Hughes started talking and we all listened. For a less cumbersome reading, I'll put myself in his shoes and borrow his voice, which is in a sense easier and fitting since I was the only other male in the room. Oh, Montessori brothers, where were you?
The essence of the talk could be summarised in two parts: what is a good school and how can we prove to the world that our Montessori schools are good schools? As an introduction, it is important to realise that the Montessori community is faced with a typical ‘locker room’ syndrome: we're only talking with people from our team. We're convinced of the goodness of our actions and the righteousness of our path, and we confirm each other in it. Yet, self-assuredness and conviction only gets you so far, when you need to convince other people, not other Montessorians.
When trying to prove something, we can determine five levels of evidence, from the least powerful to the strongest.
The first level is opinion. Everybody has plenty of them, and when it comes to education they are usually very eager to share them with you. They can be about anything: ‘I believe that the Earth is flat’, or ‘I'm quite sure that children can read before the age of 7 without being taught’. Opinions can be right or wrong, but they can't really be discussed. Not because they have to be respected, but because there is nothing to discuss about it. Opinions simply are opinions.
The second level is anecdote. Again, combined with opinions, these first two levels make the massive bulk of every argument on a given internet forum or at any Christmas dinner. Anecdotes are opinions backed by a handful (sometimes just one) fact. And again, they can be right or wrong, and we can throw them at each other without much ability to prove anything with it except that we have a fascinating life and we can tell stories in an engaging way.
The third level is summary statistics. It is in essence a compilation of anecdotes, but it adds up to something much more forceful. When a teacher can say ‘94% of the children that came from my classroom have graduated from high-school whereas the average of the neighbourhood is 50%’, it has a lot more weight than the heart-warming story of your nephew. This is already something done by some schools, and figures like this which was quoted on the East Dallas Community School website definitely look shiny.
The fourth level is a tested hypothesis. There we enter the realm of experimental science: a researcher gives himself something to prove and goes forth to do so. The Angeline Lillard study on Montessori in the schools of Milwaukee was such a case. She formulated the hypothesis that Montessori children would fare better than children of equivalent socio-economic background, with parents as involved and caring. She gathered data in a way consistent with her hypothesis to not skew the results, and came to the conclusion that she was right.
The fifth level, a replicated hypothesis, would be another scientist, preferably someone doubtful of the method who would replicate the experience of Dr Lillard and come to the same conclusions - or slightly different ones. But that would be an experiment in a scientific and controlled fashion.
You start doing research at summary statistics, but you really enter science at the fourth level. Most Montessorians are not scientists, and formulated hypothesis might be difficult to prove for us, mere teachers. Yet, at the very minimum, we must shun opinions and anecdotes and turn to facts, numerous and organised to make our points. The main point being that we are good schools. But, what is a good school? What defines a school as good? What are the goals of schools?
From a neuropsychologist’s point of view, a good school is one that fosters brain development and respects its rules. From a more traditional perspective, a good school would be one that helps children perform well. In a massive study in the US, Bob Pianta, Dean of the Curry school of education at the University of Virginia tracked which schools were able to fight back social determinism and to reduce the gap between children which is there from the start and stays there unless acted upon. Such schools would be defined as good schools in Dr Pianta’s study. In both cases, traditional schooling seems to fail pretty miserably.
The traditional method of teaching [which can be tracked back about 2500 years at least] consists of a teacher that tells ‘stuff’ to students who listen to it and are subsequently tested to check how much of the aforementioned ‘stuff’ is still there, stuck to them and hasn't been washed away. From a neurological perspective, it makes no sense. This type of learning has a half life of one year, meaning that you lose half of it after one year, and then half of this remaining half next year and so on. The concept behind this method is that all students arrive at the beginning of the school year with the same skills and knowledge and that by the end of the school year we gave to each of them the same additional set of skills and knowledge with which they can move on to the next year. Any honest academic knows that, of course, it doesn't work. Children arrive with a wild array of capacities, and stay that way in the best of cases, or amplify the gap between their achievements in the worse.
To ensure cooperation and compliance, traditional schools use a mixture of carrots and sticks, namely grades [which can be both], punishments and rewards. This has the result of making schools an extremely crushing environment for many children. An interesting test of this is to type ‘school makes me...’ into Google and to let this modern oracle tell you what the most common endings to this sentence are (hint : it's not ‘happy’).
So a good school would not work any of these ways. A good school would be so different in its workings that it is worth labelling it a ‘school 2.0’, a whole new version of the old concept of school. Such a school would act in compliance with what we know of the brain. First and foremost, the brain is fed by sensory inputs, and the biggest provider of it is the hands. It is through touching that the children will learn first. A ‘school 2.0’ would provide exploration through movement, through the work of the hands, and a wide and diverse range of sensory inputs. It would focus on life goals, not just end of year test ones. We want ‘school 2.0’ to accompany children in the lifelong endeavour to become a well developed human being. What are the qualities we seek in others? In employees, colleagues, or even friends? We like them to be efficient, trustworthy, independent, self-motivated, caring and responsible...all these qualities can't be taught, but in a ‘school 2.0’ their growth would be fostered.
‘School 2.0’ would accept and embrace the diversity of the children that arrive in it and would not try to take them all to the same point at the end of the year, both a futile and fruitless task. A ‘school 2.0’ would focus on nurturing the pre-frontal cortex, the zone of the brain that distinguishes us from the rest of the living creatures. It's the part which is responsible for decision-making by providing cognitive flexibility, the capacity to control impulses and working memory. It requires the opportunity to make choices in the classroom and to have to control your behaviours without outside rewards or punishments.
So, after reading all this, all Montessori teachers are probably nodding a lot, and patting themselves on the back for being at the forefront of pedagogical research. But the problem is double: how to prove that we are doing the right thing and how to deal with the great number of people who still strongly believe that ‘school 1.0’ goals (good test scores) are what matters.
For the second part, it is mostly a question of hope and time that the ongoing teaching debate will finally catch up in Science. That day might dawn sooner than expected here in England with initiatives like the one at the Royal Society, ‘Neuroscience:
implications for education and lifelong learning’ which seems determined to take neuroscience into consideration when it comes to education reform. But in the meantime, it is important for us to prove that Montessori education is relevant on both fronts: that Montessori children fare well on standardised tests and that they achieve ‘school 2.0’ goals.
We don't have the capacity to conduct nationwide research projects able to convince the movers and shakers. We have neither the resources nor the skill. But that's okay because there is plenty that we can do and, by doing it well, we can create an amount of data that will help us reach the ‘Powers that Be’. In the meantime, our research will concentrate on the people we need to convince immediately: the parents.
They are the ones who have to make the difficult decision of entrusting the most precious thing in their life to a bunch of weird people who don't believe in gold stars, worship an old Italian woman and freak out when you put a mat on a painted line on the floor. The amazing things we pretend will happen to their child is only hearsay for them, and they are desperately in need of hard data and facts to help them make the difficult choice to go against the flow. The most obvious data we can gather is the answer to the questions they always ask: ‘Are our children actually learning anything since you don't teach them ?’ ‘What happens to them after they leave your school ?’
We can put together a programme evaluation along these two lines. It can take different aspects, none of which are exclusive, and ideally would be combined. The first way to evaluate if your children fare well on standardised tests, is, well, to have them do standardised tests. But it's not about bringing multiple choice questions to the children about which cube is on top of the pink tower [the answer is b, the smallest one}. We want to be able to prove how children in our classrooms compare to how children from traditional schools do, and for that there is a vast battery of tests available. These tests are easy to administer, assess a wide range of skills and developmental abilities and most importantly, have been tried on a great number of children, giving you an average of results, helping you know where your children are on the scale. Tests like the Bracken and the Vineland are invaluable in tracking down precise results, which we sometimes fail to see.
Steve showed an evaluation that was made at Cornerstone Montessori School in the US, that caters to socially disadvantaged children, using both these tests proving that Montessori children are not geniuses way above the average, but are all on the good side of normal. Though this may seem a very unimpressive result, with nothing to write home about, it's in fact crucially important in two regards: Children from disadvantaged neighbourhoods usually fare much lower than that, so this proves the impact of the school. By giving small but real results, it reinforces us: we stop selling pipe dreams of excellent children with dazzling intellects. Yes, I once had a 7 year old knowing all his multiplication facts all the way to 16². but I can hardly pretend all children do. A realistic, measured and consistent positive effect is a better argument than a few anecdotes of amazing children that sometimes make us sound like snake oil peddlers (even though our snake oil works).
The second part, ‘what happens to our children’ is a bit more complex to answer but not that much. We can start with the low hanging fruits, the information that is a phone call away. Keep track of your alumni and ask for their report sheets, their grades, and file all that. A quick compilation after a few years will probably end up making a compelling answer to that most annoying and recurring question. But since we want a more elaborate answer we need to collect more information and another way to do it is through questionnaires.
The first person we can send questionnaires too is parents and former parents. We can also send them to teachers. Teachers questionnaires are especially valuable since they have a greater sense of normality than parents, seeing our former students among tens of other children of the same age and background.
When preparing the questionnaires there is a few important things to keep in mind: Open ended questions are difficult to quantify even though they can be informative. The questions that can be answered through a Likert scale (range tidiness from 1 to 7 for example) should have language attached to it and not just numbers. It helps gets more standardised results and they can be used more easily: ‘80% of teachers think the children from our school were well-prepared for school’ is more convincing than ‘rated them 6 out of 7’. It shouldn't be too long, and you should make things as easy as possible - offer incentives, pre stamped envelopes, make it simple.
The last important way to collect data is through classroom observation. Montessori courses provide a lot of tools for observing classrooms. A standard behavioural observation procedure could be by tracking specific behaviours in children every thirty seconds and ticking them on an observation sheet. This would allow us to assess how much our children are actively working compared to passively listening to us. We could also use this way to measure the amount of undesirable behaviours that are actually happening. Again, being able to prove that this ratio of appropriate/inappropriate behaviour is better than traditional schools is more convincing than pretending that these behaviours don't exist in ours. Such classroom observation allows us to compare us with other schools. For example, the Bob Pianta study showed that on average the ratio of listening to acting in traditional classrooms was 10 to 1 (1 hour of activity for 10 hours of listening).
So, equipped with summary statistics of alumni results, with the results of these parent and teacher questionnaires and with classroom observation reports we will now hopefully be able to create a set of data, that can create a compelling argument in our favour.
And of course, all these are what we do at our own small local level. But if, after a first round, we were able to gather results, questionnaires and maybe were able to put together a Standard Montessori Assessment kit, something of a much larger scale could become feasible. This would give us something that could give us weight and momentum.
In this day and age, it is very politically incorrect to make claims of superiority of any kind, and since Montessorians don't encourage competition, we are severe offenders when it comes to not being loud enough and try not to antagonise traditional schools. But maybe this has to stop. Probably now is the time to be arrogant, to make our claims and to back them with facts.
Science aims for truth, and that's a pretty arrogant goal. But so was Maria Montessori's aim: to create a system of education that would discover and follow the universal rules of the child, that would be as undisputed as are the rules of gravity. So let's not be nice anymore, and let's be right: let's get real.