This may be an unpopular opinion. Brace yourself.
There is so much research into education that it has become, at minimum, problematic; at most, toxic. There are five C’s to explore why educational research is writing itself into irrelevance. Call it the “C’s of Evidence” if you’d like.
Clocking-In
Teachers are already time poor. I can’t remember the last time I got the chance to sit down and read anything work-related, more than scanning through to read the lesson outline or writing resource to implement in my class that week. Much less any educational white-paper that someone came up with with new evidence that I should now be implementing in my classroom.
And yet that is what teacher’s are expected to do.
Evidence-based Practice became the catchphrase of the 2020’s in education circles as teachers sought to justify themselves and their classroom decisions.
But how many actually look at the actual evidence?
And how much of the evidence is actually classroom tested?
Time to read white-papers is one thing. But it’s a whole other thing to implement the findings into your own practice, and apply it across multiple curriculum areas, whilst remaining within school policy and expectation. The time that takes is greater than you might initially think, especially when considering overviews, planning, guides and resource creation. Doing this while continuing to teach a class on a daily basis can often be measured in years rather than months.
This can be sped up of course if the research has already been synthesized for the classroom, a resource developed, and the overview written in a way that classroom teachers can pick it up and run with it. However, in many cases, what this leads to are teachers following the research second hand – trusting the resource in their implementation of the research, rather than ever reading the research and what it is based off themselves.
Conditions
Research is like a Science experiment. You have a series of variables that you are observing and trying to test, with constants to be able to judge the variables against, and some control conditions that allow for baseline results to be compared against. Research in education is conducted with a series of conditions, with the likes of control groups, limited sample sizes, and narrow focuses. There are many, many issues with the way in which education research is conducted. It’s not so much an issue just with education research, but any research in which humans are the test subject.
Multi-factor Generalisations
There are simply so many factors that can contribute to the make up of a person, what they do, why they do it, and what we can learn from that. A child who scores 12 out of 100 in Maths can have a long list of reasons that can contribute to such a score. Did they wake up on the wrong side of the bed? Are they hungry? How long is their attention span? How much learning support at home have they had in the past two years? Are they struggling with a learning disability? Have they experienced significant trauma in their lives. Is this their first, second, third, or fourth school? Does their family hold a positive view of education? Has there been health issues that impact on their attendance at school? These are all variables, for which there are zero controls. None of these remain constant, and each impacts on a person to differing degrees.
Of course, all of these factors are impossible to take into account for the research to be conducted. You can’t possibly control all of these, and so they are largely ignored.
Don’t work with animals or children
Children are notoriously inconsistent. They are still learning the world around them. Whilst this makes for some naivety and therefore a degree of honesty that adults tend to be able to mask or fake, children also are much more in-tune with the imaginary and make-believe that there are serious questions over the validity of any results involving them.
Small Samples
Along with the myriad of variables that are trying to be accounted for, these then get multiplied by the number of students in a class that a teacher has to work with. Research is inherently dealing with smaller sizes. Rather than try and conduct in-depth research using every student in the class, they might take a sample of 5 students at random. Rather than interview every class, they might select one or two at the school. Rather than go to every school in the area, they might select one or two. The list of ways that the research is limited in order to make it manageable.
But when was the last time a teacher had five students where factors could just be ignored? Never. Each day, there are 20-30 students to account for in the class, any number of factors impacting on them.
So how applicable is any research conducted in the name of education?
Contradictions
Recently the education community has jumped on the Science of Learning bandwagon (one in which I believe the wheels are falling off quicker than they said ‘click click’ to the horse). This brought with it the Science of Reading, which looked at ways in which we actually learn to read and therefore to direct effective ways in which to teach Reading.
Marie Clay, one of New Zealand’s most honoured education researchers, was a world power in educational research. She single-handedly lead the drive for programmes like Reading Recovery in every school to support struggling readers in the 1980’s, and was a proponent of ‘whole-language’ literacy.
However, with the influx of the Science of Reading, as well as, researchers from various universities (particularly in the United States), her methods are now criticised as being ineffective, showing no improvement and even purported to have negative effects later on.
Yet, while looking through a pile of seriously old education books, I stumbled upon her book in which she introduces Reading Recovery. In it, she outlines her research and what she did in order to reach her conclusions (which was that early, specific and individualised intervention supported struggling readers to get back up to speed with their peers). The depth and detail in which she goes into, the considerations made and the testing of her intervention was evident in the brief skim through I had of her initial chapters.
What it showed me was real proof that educational research is somewhat insignificant. Because here was someone who has been lambasted and rubbished in recent years for implementing a programme in many countries all in effort to support and raise literacy levels of students who find reading difficult. She conducted her own research in order to find out what it was that made the most difference. She was meticulous and her findings were evidence-based practice.
Now I’m not saying research can’t be disproved, and I’m not saying that additional research can’t find shortcomings, or even find better, more effective strategies. Of course they can. But at the same time, who is to say that if the likes of Marie Clay, whose research is just as legitimate, can be tossed out and the proponents vilified, then in my mind; so can absolutely any other piece of research.
The vile statements of utter contempt recent converts of the Science of Reading have for the very mention of Clay’s name is something to behold. ‘She’s a disgrace. Caused terrible disservice to an entire generation of children.’ and other similar comments.
So we have one piece of research saying one thing. Another piece of research saying the opposite. Undoubtedly both with a different range of variables and an indeterminably amount of factors, and neither of which had the combination of students, needs, families, experiences as those in the classroom in front of you as the teacher today.
Shortly after, this came across my social media feed, and I simply had to screenshot it.
This statement was made, quoting the following aritlce about research carried out by Becky Taylor in the United Kingdom.
https://www.ucl.ac.uk/news/2026/jun/comment-should-we-stream-school-students-based-ability
It found streaming Year 7–8 students was beneficial for higher-ability students. These students made three months’ more progress in schools that streamed students for maths that those that did not. Streaming also did not damage the results or self-belief of students from disadvantaged backgrounds.
In fact, this study found students who were not streamed made only one month less progress overall than students who were, after two years. Students with disadvantage and students with low prior achievement made similar amounts of progress regardless of how they were grouped.
These findings were surprising because they seem to contradict previous claims streaming can harm some students without benefiting others. This new study can and has been interpreted as showing streaming benefits some students while creating no harm to others.
However, the findings do reinforce previous research suggesting streaming is inequitable and widens achievement gaps. The new study found high-achieving students made much more progress than low-achieving students when streaming was used.
When presented with contradicting research, how are teachers meant to know which piece of research is correct? Which one do they listen to? The only way to really know would be for them to read both and take into consideration the sample sizes, the measures used and the methods applied for data retrieval and how the data was analysed and the evidence collated.
And we’ve already found out that no teacher has the time to do that. By the time any teacher has been able to do that, new research would have been found and contradicted everything you’ve just read.
Count them All
With the coining of the phrase ‘Evidence-based practice’, or ‘Evidence-based approach’, education now relies on there needing to be research in order to do anything in classrooms. It would appear to me that this is clearly a supply-and-demand issue. Teachers are wanting research conducted so that they can say they are following the evidence, and so research is conducted en mass.
There has been a noticeable increase in educational research in the past decade.
Is this true? Or does it just seem that way because everyone is talking about it?
Well, I did a quick Google Search (with AI search disabled for the record!), and was thrown to this paper where researchers are now researching whether there has been an increase in research. Go figure!
https://link.springer.com/article/10.1007/s41979-026-00184-5
I, of course, do not have time to read it in any depth.
Instead, I simply jumped to the results tab which shows exactly what I suspected;
The sheer number of research papers being published has exploded since 2020.
Not only do we not have time to read all the research, there is now so much of it to read through, that it has become an impossible task for any classroom teacher to keep up with the sheer amount of it.
To be clear, what these numbers above show;
- In 2009 (when I began teaching) there were about 150 journals of education research being published. That’s on average one journal published every two days throughout the year. I read a grand total of 0 of these at the time.
- A decade later, in 2019, the total number of journals published had pretty much doubled, to over one for every weekday of the year. The number of these I had read to date had also doubled. Double nothing is still nothing.
- Come 2020, and suddenly we have 750 journals published. That’s 750 journals in a year. That’s 3 journals for every workday that year.
- It remains around 750 published journals for the next 5 years. That’s a mere 3,750 journals published. To be clear, that’s the same number of journals that were published in the 20 years from 1997 to 2016. I did the maths.
Put simply, researchers have written themselves into oblivion. What I mean by that is that there is simply so much research that we simply cannot keep up with it and we are simply numb to all of it now. The sheer magnitude in the number of research papers in education is creating a saturation of the market, sending them into obsolescence and irrelevance.
Teachers are routinely told ‘The Research says this…’, ‘The Research says that…’, and the vast majority blindly follow along, accepting of whatever approach we’re being told about. This is mostly because we don’t have the time, and simply because we can’t be bothered looking up the research, analysing the methods, and processing their conclusions. Let alone searching through the 750 journals published this year, which could be disproved within the next year at the rate that we’re going. It’s easier to nod our heads and just believe what is being told to us. We rely on our own experience of our classrooms, of teaching, our own understanding of how student’s learn, and the plausibility of what is being said as being true. If it sounds reasonable, there are many who will just give it a go until we’re told otherwise.
All this research, where findings can vary, and just about any hypothesis can be backed, is essentially rendering research as completely unimportant. Evidence-based practice isn’t a status to be touting if research can back and support anything and everything. With so much research out there, not only is it obsolete because very few have the time to read it, but with so many articles out there, evidence can probably be found for anything and everything. The saturation of the education research market is like an empty glass that is being filled, yet once the cup is full, the tap was left on and the water is now overflowing. The value of the water is depleted and we are no longer thirsty.
Cash is King
Why do we need all this research? One reason is the ‘evidence-based practice’ rhetoric. And because of that, in order to push through any new resource, or any new approach to teaching, providers need to have evidence for which to claim their product as being ‘evidence-based’. What this leads to of course, is every large educational conglomerate as well as governments providing funds for research to be carried out.
Education is big money. In 2025, the total income from education was a staggering $1,106,000,000.00. Yes. You read that right. $1.1 trillion. Of course, there is expenditure to take into account, but it leaves the country with $95 billion dollars in profit.
And that’s just New Zealand.
– Figure.nz: https://figure.nz/chart/kIpNw1H5r3cVIesv-O0u2toDAfNpAyifH
This then brings in a whole other reason to be skeptical of research that is conducted. Which master is the research serving? The wider educational community, or the company that is funding them to find what needs to be found in order to push their products through?
Concluding Evidence
One of the key statements I will hold onto was from one such researcher; Steve Graham. I was lucky enough to hear a lecture about Writing Instruction in which he outlined some of his findings from his unironically named ‘Evidence-based recommendations for teaching writing‘ paper.
You can view it here: https://www.tandfonline.com/doi/abs/10.1080/03004279.2024.2357893

Graham essentially said that upon reflection, any evidence, including his own, needed to be seen for what it is; evidence from a research environment.
He said (in not so many words) that just because it happened in a research environment, regardless of sample size, “…does not mean it is going to replicate in every situation and circumstance.”
In this one statement Graham made it clear that just because they found this in their research conditions, it does not mean it will translate into every classroom all of the time. Some classrooms may find benefit. Others may not. Some students will benefit. Some won’t.
It is probably the single most self-reflective statement I have heard from any education researcher – a real, truthful, and down-to-earth statement about the reality every piece of research, including his own. It provided an unusual confidence in which one can stand on to challenge any research. It provided confidence to take on board research, but not make it the be-all and end-all that some try to make it.
Moreover, I am of the opinion that each and every day, teachers around the world are conducting their own educational research. Observing, questioning, recording, planning, implementing strategies and lots of trial and error. It may not be published anywhere, and it may not carry with it the backing of a distinguished university, but that does not mean it is not important, nor less valuable. The evidence for my evidence-based practice? The 15 years I’ve spent in the classroom.
That’s now my evidence based approach.