Faulty reasoning, an extensive analysis of another analysis

I hope that by sharing this, really show that we are capable of doing a lot as teachers. We are able to take a look at the big picture and ask our school leaders and colleagues if we are on the right path to learning. I did not get any response on this analysis in my previous school. But, I am hoping that you who are out there, might be able to use what I have written below to see where you can do what I did, but hopefully much better. And I sincerely hope you will be heard.

What can I say about this? I presented the analysis below as a response to another analysis from a previous school I worked in. I felt that the latter-mentioned original analysis did not go deep into its main focus, namely answering fairly why students fail – it settled by simply saying the reason the school did not do so well was because the teachers were bad teachers. The reason? The authors based it all on a single person performing interviews with dissatisfied students and parents. I am all for that, it was just such a claim alone + some shaky hearsay and anecdotal evidence about poor teacher behavior which somehow would define everyone, was the only base of evidence for this grand conclusion.

What was even more of a travesty were the conclusions and the action plan mapped out based on that. Simply put: conclusion: bad teachers that do not have special adaptations to student needs and the action plan was conveniently as a magic pill, we would all focus on special adaptations. This resulted in a major overload in documentation on things that we were in fact already doing, with no clear direction as to what it was for, other than to show we were not those bad teachers. I made the attempt to question those practices and map out how much we were all doing the same thing with no systematic evaluations.

This, plus my analysis of the analysis, were met with silence from the school leadership. I received no feedback on it.

The analysis which is below has been stripped of names that could possibly identify the school or any individuals.

Analysis of an analysis on why students fail their classes

The big question which I shall attempt to shine the light on and analyze is the following: “Why have so many students failed their classes, the previous school year?”

A factor which has been put to the forefront as almost the single, most important one is in a previous analysis by the school leadership at [insert random school name] is the teachers’ lack of ability in their professional capacity, especially with regards to specifically adapt the teaching and that teachers, among other things do not feel any responsibility for failed students. This conclusion rests mainly on interviews done with students who did not do well during the previous school year. However, it is not clear from the analysis what has been said nor what questions have been asked to get to these conclusions. The evidence is poorly formulated. Furthermore, it is not clear if the interviews were semi-structured (which seems most likely to be part of what implicitly needed to be interpreted), structured or entirely open. Additionally, there is no account of how many students who were interviewed, how many who said this or that, what questions were asked and so on.

Why is this problematic? Well, because less structure for the interviews means subsequently less reliability. Secondly, it has been proven that if an interview is done in front of a panel instead of a single person, this would grant the interview more reliability. There could possibly have been two people present as a kind of panel during the interviews but at a closer study of those, there seems to have been only one person conducting the interviews. It is not entirely clear how this is all laid out.

The reliability for the source material is in itself therefore of a substandard, since there has been no complement of other studies nor sources whatsoever. The actions and action plans which have been proposed lack evidence and timetable. What further makes the poor standard of the analysis, is the fact that the results are hard to separate from what the students are saying and what is being interpreted by the authors of the analysis. Are they conclusions of the results? What is the foundation for these conclusions? Many questions are left unanswered. Take for example what is said right below in the results of the analysis (not the conclusions nor discussion) section:

“There is a culture on the school which is that one (meaning the teachers) is not responsible for the failed students. It has been brought to light that one almost boasts about the number of failed students one has. “The collegial shame” is lacking. We who have done this analysis finds that also surprisingly few teachers ask each other for help, despite that they obviously have a hard time reaching through to the students. This is a culture which we will have to work for getting in place.”

How can it be so solemnly clear that there is a culture on this school that “one is not responsible for the failed students”? Due to the faulty presentation of the source material and the methodology, it makes it hard to see in any way on what grounds this perception rests. Is it all completely anecdotal? Or is it because the students said so, every single one of them and/or were there many students who said this, as in more than half so at least it could be said that a majority of the students asked deems this the case? The questions remain unanswered.

Do the students even know what is good for them?

The top basis of data on which the analysis rests are the interviews with students and parents. Asking them is of course desirable and worthwhile. However, the blaming of the teachers and the conclusions that were being drawn from these conversations does not hold up for much scrutiny. Especially, as learners usually do not know what is good for them. In particular and in tandem to this, there needs to be a holistic picture to be able to draw such conclusions. There needs to be data drawn from several different sources of input which are continually collected with validity.
This could probably look something like figure 1 here below.
figure 1

This is just a small part of something which is a lot more complex really; it is about collecting data on the efficiency of an education. The following four categories of data should be used within education to get a more nuanced and extensive view:
1.) Input data, which is divided into two categories of character, namely teacher and student. The student character consists of the following descriptions and subsets of data such as demographics, former academic performance, student documentation which follows the students from transferring between different schools and native language. The teacher character consists of teacher competence, academic qualifications and professional experience.
2.) Data process, which takes place in the actual teaching, learning and processes of assessment – both inside and outside of the physical classroom’s premisses – such as lesson plans, method of assessment and the management of the classroom.
3.) Context data is about the school culture, the curriculum for the different subjects (including desirable outcomes of learning) and other extra curricular educational programs, the school’s HR, infrastructure and the financial planning which consists of: the school staff, buildings, software/hardware and expenses.
4.) Outcome data consists of the students’ successes in classroom-based formative assessments, homework, standardized tests and national tests. This also includes the students’ wellbeing, social and emotional development such as safety, support, respect for diversity and special needs. It is also about data regarding studies and work after graduation.

“[L]earners often misregulate their learning, exerting control in a misguided or counterproductive fashion and not achieving the desired result. This is due to (a) not having the necessary standards upon which to judge their learning state, (b) not having the necessary knowledge to monitor their own state in comparison with the standards, and/or (c) not being able to initiate the proper processes to change their current state when their behavior falls short of the standards (Baumeister & Heatherton, 1996; Taminiau et al., 2013). The second problem is that learners often choose what they prefer, but what they prefer is not always what is best for them.”

In summary as previously stated: Students/course participants/novice learners are poor at regulating their own learning. They often choose what they prefer and it is not always best for them. All this is founded in lack of knowledge and ability to be able to monitor one’s own behavior and change it.
Another myth which is being peddled around within education is unfortunately the one about learning styles. The evidence for this is weak and has been strongly opposed for many good reasons. This is summarized as: “The individually preferred way of learning is often a bad predictor of the way people learn most effectively; what people prefer is often not what is best for them.” It would be better to instead trust oneself with cognitive ability with regards to working memory + new knowledge and cognitive load instead of what way is best preferred to be learning in.

The level of the organisation and the teacher

For a long time there has been a sort of truism within the world of education: no school is better than its teachers. All of that sounds fine and vaguely enough expressed to make it seem as if actually were true. Nevertheless, this expression is problematic due to a number of different factors: 1.) It lacks evidence (2007 McKinsey) 2.) A teacher is not an all-powerful ruler who will affect her/his subjects through all means in perfect detail and omnipotence. 3.) The only way to verify this claim would be to either try several “exceptional teachers” on several different schools with different kinds of backgrounds and see if, no matter the circumstances, they will still excel and perform admirably. Especially then at schools that are struggling with poor results. The other way would be if education and mankind’s learning as well as performance could possibly be sorted in to clear, measurable and moldable factors. These factors would then in turn have to be able to be affected by the teacher.

So what then should we do together, as a school?

If we were to use a checklist to look at the reasoning behind different actions and action plans, then Daniel Willingham’s specifically designed one for teachers is superb in this manner.

1. What problem is being solved and what is the thought that this would be able to improve?

Answer: “Many students fail their classes. We need an education of what is actually part of the teacher’s job description. There is a lack of knowledge about what special adaptations are and that they are a part of the job description. We need the teachers to be educated in grading and assessment. There is on many levels a lacking of a holistic look on knowledge, instead the knowledge requirements are simply ticked off. The teachers must trust their own assessment competence and do more of continuous assessments of the students’ knowledge, trust in it and not demand that everything needs to be put in examinations on paper. We must get a culture where one asks for help and actively seeks other pedagogues to enhance one’s own practice.” That was the answer in the analysis to this question. Now, to the urgent question at hand, what are the evidence to support all of this and can clearly show that it will lead to betterment? Could it not actually also be about a devaluing of the worth of the grades?

2. How will we know of this will work? How will we measure this success and when will we know if we are successful? How can we be certain of that there is a meaningful way to measure whether this will lead to something better than what has been done before? Since we gladly celebrate the new and change then there must of course be a better answer than “this feels right”.

Answer: No answer to be found, so that has forced me to extrapolate the implications of the analysis into the following: “The Swedish School Ministry says so, ‘kind of’”. Well, what we have done before is also something which the Swedish School Ministry has told us to do. How can we measure any of this? And if we are going to measure, is there not a huge possibility that this will lead to an increased workload? Who will be responsible for this? What other assignments shall be taken away in order for us to be able to do this?

3. When will it be obvious that we have done the necessary improvements? Is there any backup plan? What kind of evidence do we have that it will work? When can we say that we have failed? It feels reasonable that we may know how this progress should be evaluated beforehand.

Answer: No answer.

4. What will happen if the goals are achieved, or not achieved? If these initiatives do not go as planned, when do we admit to them no longer doing so? Will we adjust what we are doing? Is there any kind of backup plan?

Answer: No answer.

5. What evidence are there which support that backing up this plan is a good idea? Where does the research come from? How certain can we be about the findings behind this? Is it classroom based, from laboratories or is it simply pure theorising?

Answer: No answer, apart from the implied “The Swedish School Ministry says so, kind of.”

6. Are the expertise and the experience of teachers being acknowledged in the decisions?

If teachers are asked to test something which sounds weird then it is not enough to simply dismiss it by saying “All research supports it.” Where is all of this research? Show the basis for it. Teachers need to be provided with convincing arguments to ignore the evidence based on their own experiences.
Answer: No answer and no recognition of any teacher competency at all, in fact the very opposite, apart from the implied “The Swedish School Ministry says so, kind of.”

The teachers’ fault?

The thing is that if you put a certain teacher into a really good context with the right circumstances, then he/she will most likely have a better chance to achieve amazing results. If it is about the entirely wrong context and circumstances, then the teacher might still perform really well but this is the highway to burnout. There needs to be the right conditions and structures, but above all, a trust to build up a professionalism.
Most teachers can be good teachers with the right conditions and most teachers can also be bad teachers with the wrong conditions.

Focus on assignments vs. focus on knowledge?

Now is that even a thing? To even get a notion of where the student is in her/his own development towards the required knowledge and level of abilities will require points of assessing that ability. You can call these assignments, tests, examinations or whatever you like, but these points of assessing the quality of student work are still necessary. If a student does not do anything and if we do not expect them to adapt and work under pressure, then we are doing them a huge disservice.

Special/extra adaptations?

Of course, as teachers we always try to get to know our students well enough so we can adapt our method of teaching and assessing. But, according to The Swedish School Ministry, we also have the obligation to meet the needs of the students who are not challenged enough by our teaching. Extra adaptations shall take place within the framework of regular teaching and should therefore, one can assume, be possible to execute under these conditions as well? The thought was not that there would have to be additional time spent on documenting special adaptations. Besides, extra adaptations take place every day in every classroom.
But, there is an enormous trap lurking within this line of thinking: To only presume that if we execute these special adaptations, then the student should be able to keep up. The problem with working from such a premise is that it will eventually lead to that the groups as a whole will risk lagging behind. Human beings have more things in common than we would perhaps at times like to think. This is also true for learning. So, we start we the class as a group. I have seating plans which place the students at random in the classroom with walloping effects on peace and quiet with one class. I do not necessarily do this every time, but I have realized that especially in the afternoons, there tends to be some turmoil with certain classes. Another class which I have in the morning, I can be less strict with, more so because they are at large less troublesome.
These are but a few examples of what we all can recognize ourselves in: Obviously, we start with the premise of how learning happens and that in general, we are all more alike than different. We then adapt our teaching, our way of interacting with the students based in part on the groups of students we face in order to accommodate everything and everyone in the best possible way, as optimal to learning as possible, to the best of our abilities. The different factors which this is all dependent on also include, but are not excluded to: time of the day, the students’ mood for the day, our mood, our curriculum and many other factors.
What is then the thought behind all of this? No statistical evidence is presented for what we can do and cannot do. So, why not start then with some kind of evidence-informed reasoning from the simple case that if you do have a population of students to look into then it will be hard or truly measure anything by any sense of validity: “This is because the data is based on children’s results, and children are complicated and individual, and the school population in any given school is statistically too small to make meaningful generalisations.. “ (Didau 2015)
It goes on a bit more saying that:

“Maybe a bit more data might be useful. Except it wouldn’t because, for most schools, the data varies with the cohort, not the quality of teaching. [my choice of bold-letter text] […] There is no pattern. Unless a school consistently records 100 per cent, there never is a pattern for any school, in any historical data. This is because the data is based on children’s results, and children are complicated and individual, and the school population in any given school is statistically too small to make meaningful generalisations. The lack of a trend may be why the data isn’t presented over a number of years, because anyone looking at the data would realise that it is random. As any financial advertisement will tell you in the small print, past performance is no guarantee of future success. Long-term trends in something as complex as educational outcomes are – unless you mess with the data by, you know, making the tests easier, selecting by ability or dis-applying certain children from assessment, or simply not reporting stuff – always random.” (Didau 2015)

So what is the question we are asking? The number of failed students? The school has had these students and let them go further into their programmes, even when they failed almost all previous classes. In a sense, this is what the school had to do, as by school law we cannot deny the students the right to their education (even though it is a ludicrous statement if there ever was one) The teachers have on numerous occasions brought the attention to the school leadership and tried to accommodate the students as best as we could to make some kind of change. Alas, it has been all in vain, as the stated reason has been: we cannot deny them the education they have a right to by making them for example dropping behind one class and catch up, since how could they possibly move on from the first year being filled with mostly failed classes and then all of a sudden “just get it” the next year?

How would one then go about turning around a school, according to research? First model, building a strong organisation with strong leadership.

There is no one-way for all action plan, so there will be a few propositions based on research and sound reasoning from the world of research in the different takes below. Firstly, let’s start with the most common-sensical drastic change where there are usually four steps that are mentioned in the world of educational research when the topic at hand is how to turn around a school from failure to success:
1.) Signal the need for dramatic changes with a strong leadership.
2.) Maintain a consistent focus on improving instructions. This shall all be upheld by the school leadership within the school in order to support the teachers as well as gathering relevant data from a number of sources to make it clearer, transparent and to solidify the impression of reliability of the proposed actions.
3.) Make the changes visible early on in the process. These improvements shall be clearly formulated and verifiable. It should be very clear that we are on our way to a tangible goal.
4.) Create a dedicated and engaged staff. It has to be pointed out, that while this is one of the more evidence-strong studies which have been made, they still get very low scores in terms of reliability and validity because there being too many factors at play all at once when dealing with such a complex thing as turning around a whole school.
So this is one take, so far so good. It has one of the strongest basis of evidence in the world of research on education, but still a very poor basis. Are there perhaps other ways to look at this? If you wonder then, where is the accountability of the teachers? How high in hierarchy of influence would we place the teacher’s role in all of this? We are dealing with the success of a school after all?

The second model, working with the culture and curriculum

First step: The school culture plays a huge role in school. Is it cool to study or is it cool not to study? The goal for most children is not to be a successful adult, but a successful child in that particular social setting this child finds himself/herself in. Depending on what kind of school culture there is in the school for the children, will of course heavily influence what is seen as cool or not cool. How can this then be achieved? There are a number of different steps for this, which are recommended in the world of research. What unifies all of them is that all of this has to be initialized and governed from an organizational level. Unfortunately, in the school, there has been no such leadership during this problematic period when so many students failed their classes.
The second most important thing is not the teachers, but what the curriculum is all about and how it is being followed (which of course the teachers partake in, but it should be done by the support of the organizational level).
But, it is also said that dedicating time for general competencies such as 21st century skills (widely criticized among others professor in educational psychology, Paul A. Kirschner) should not at all be prioritized. What should be prioritized are the subjects in school and not some kind of umbrella of general competencies or skills, which by the way would still be coming along nicely if the curriculum and the subject-domain knowledge of the students develop! If schools focus on the curriculum and the subject-domain knowledge correctly, then also the “weaker” teachers will be better and the “best” teachers will be even more amazing!
In third place comes assessment. This also requires an awareness on an organisational level which makes it possible for the teachers to continuously develop their assessment. Equally important, there has to be a consensus on what assessment is supposed to do and what it is not supposed to do. This of course encroaches on to the territory of the teachers’ competence, but as the other two aforementioned pillars, the teachers are not responsible and cannot possible be responsible for this. If the school does all it can in order for it to find a meaningful assessment of achieved successes of the students, then the teaching will inevitably become more efficient. We cannot measure progress in a way which is either reliable or entirely valid, but we can say much about achieved successes.
When then all of these three factors are in place: the culture of the school, the right focus in the curriculum and correct approach towards assessment. Then, and only then, can we talk about the quality of the teaching in the school.

Why this discrepancy? Answer: Cognitive biases and lack of knowledge

There have possibly been a number of honest mistakes, partly dependant on cognitive biases that we all are subject to on a daily basis, but which nonetheless are not as forgiving when it comes to putting the blame on teachers and what direction the school should take based on this. Some examples which can be found in the analysis are the following: Availability bias. We have a tendency to put too much weight and focus into the information which is available, so much so that it blinds us for other factors. In the case of the analysis there seems to have been the case of not incorporating other sources other than of course the aforementioned dissatisfied students and occasional anecdotal evidence.

Now obviously, as I previously stated, this ground of evidence is in itself not bad, on the contrary, but it cannot be the sole basis for such grandeur conclusions. Besides the fact that it is naturally a part of something within the Swedish education system ever since 2015 – when the School inspection (like OFSTED) and the School Ministry started to create and act on directives regarding special adaptations as well as specific support. What was measured was on the big whole the number of students who passed courses and not so much how that happened. It also misses looking at the long-term consequences pertaining the student’s maturing process and independency, among other things. The terms specific support, extra adaptations, leadership and stimuli and so on are not so easy to keep track of. The Swedish specialized blog in this particular area, specialpedagogen.blog, summarizes effectively what the core of the thinking behind extra adaptations and aforementioned terms was:

Try not to focus on whether what is done is an extra adaptation or not, the most important thing is that the student gets the chance to develop her/his knowledge and that we follow this in a formative process. What we call it is less interesting than whether it has an effect on the student’s learning and the school’s development of guidance and stimuli, that is to say, successful teaching.

This is however not a part of the analysis. It is not clear what is meant by adaptations and extra adaptations. Neither is there anything which in any tangible way prompts the conclusion below.

There has been a view, which is that if the teacher delivers teaching and the student does not learn, then it is the student’s problem. Very rarely is the reasoning about the group and the teacher’s teaching. “How can I change my teaching so that the students will learn?” “When I now see that several students do not learn, I need to change my teaching in accordance with the assessments I do of the students’ learning.” (From the original analysis, my own translation)

Many questions arose when I read this the first time: How do the people behind the analysis know that this is the case? Through interviews with the students? Would the teacher perhaps be reasoning as such in front of the students about said questions? This was unknown to my colleagues and I. This is not of course expressed in any curriculum or job description. How is it clear that this reasoning which is being prompted did not happen between colleagues? Whose view is it that we are addressing? It is expressed as if it could be everyone. As can be clearly demonstrated, there are extraordinary claims which are being made, even though they are being done so with inadequate indices which are supposedly what are the basis to this.
This leads us to the next and perhaps most dangerous faulty reasoning in its consequences: Result bias, which is about that if we assess a decision dependant on what led to it and not the situation and the circumstances when we analyze a problem. There was not any kind of recognition or admonition of the objectionable and tumultuous organisational circumstances around the teachers during the particular period in the analysis. In other words, it was only seen as such that the teachers failed with x amount of students and whatever decisions or circumstances that were behind this was wrong. In this case the teachers were blamed for not having adapted the teaching enough.

But, if the situation was such that the decisions the teachers made were such that the teachers tried to do the very best possible with regards to the circumstances? Teachers worked in accordance with the competence and directives that were given, even though this was lacking from the leadership. The turmoil was understandable and explainable due to some chaos within the school leadership at the time and organisation changes, but none of that was mentioned in the analysis. It was not mentioned that the support structure around the teachers were all but made of plywood, it was essentially just set pieces.

Why then is this particular bias so dangerous? Well, because this can and will deter teachers from trying something new. If you make the wrong decisions even though you possibly could have known or made the right decision, then you will always at the end be blamed for this, you and you alone. This chokes effectively all propensity to try something new – or for that matter dare to talk about your own vocational worries with your school leadership. It is instead a manner of play which should take place in teaching. Where emotional wellbeing is the Alpha and the Omega. Motivation and comfort are poor proxies for learning, by the way. And may God have mercy on your soul if you ever say that you have made no adaptations. Then it is entirely your fault.

Next, confirmation bias, which is about how we only search for the information which confirms our own presuppositions. Other than simply being satisfied with a cold conclusion from the students’ testimony being painted in broad strokes over the landscape painting, there are such statements which had no trace of critical thinking or analysis whatsoever which the authors reached as a conclusion.
Last, but not least, we have the clarity bias, which is to say, that we focus on what is most prominent with people or situations. An example from the world of education is in a school in England that presented results for its staff. This result implied that girls were doing better on average than boys in the school. The interpretation of these results from the school leadership led of course to that this difference mattered in itself and there had to be something to do about this immediately. A teacher decided to scrutinize the results, as there were no proper statistical foundation for how these results were shown. The teacher ran through a lot of data and factors to see how well they correlated with the students’ results.
These variables were

• Sex
• Free school meals
• Previous schools
• Results from Key Stage 2 English/maths/science results
• Results from Key Stage 3 English/maths/science results
• Reading age
• Student attendance
• Teacher attendance
• The need for special support
• English as a foreign language (EAL)

All of these had a correlation with how well students did in school. The most prominent factors were student attendance, teacher attendance and the results from Key Stage 2 English. Some similarities with regards to the analysis can be seen: 1.) A factor is put forth independant on the other factors. Sex as the main factor in the analogy and the teacher in the analysis. 2.) No “secure” statistical grounds of evidence, or evidence that support the conclusion. One source, one interpreter, uncertain structure and so on.

What then affects a school’s results, according to meta research?

So, what affects a school’s results? Hattie’s summary is updated on a yearly basis. In it, the teacher is one of the most prominent factors of success in education. But, and this is a huge but, it is not the only one. And, this is a big and, such meta studies and summaries without the proper analysis risk being misunderstood and read as there only being the teacher who is the sole most important factor. For an untrained eye which has no scientific inquiry as a main interest, nor the tools to address this critically, it is an easy conclusion to reach and stay there in the line of reasoning.

What is important in any case is the student-teacher relationship. That is all true, furthermore, what has to be taken into account is that it is hard to get a positive relationship with a certain number of students during a limited time. A teacher needs time with his/her students. A teacher needs a solid foundation to stand on. A school needs routines. A school needs leadership. A leadership which supports and alleviates the teacher to perform his/her mission. This has not been found during the time that the analysis focuses on. Furthermore, some students have changed teachers on numerous occasions during their time in the school. How will the teachers who are left be able to supplement this on their own? How can we ask for help if there is no time or possibility of this?
It is also important to regard the following with Hattie’s meta studies: 1.) They are taken as a method directly from research within for example medicine. This area of study has variables which are easier to control and isolate when the studies are done. In the case of education it is impossible to isolate variables and do any kind of research which is reliable in the one or the other area. Of course, the case of RCT will help in case there are correlations. But, the amount of studies replicated that are published is astonishingly low (around 0.13 % or so). It will in the best case scenario provide a very unclear piece of the puzzle of the complex whole.

2.) As just mentioned, a study has shown that less than one percent of all research which is being published is replication studies. A majority of these (68 %) will successfully replicate the same results as the original study, but if there is another research team than the original one who is doing the replication studies, then only 54 % are being replicated successfully.

3.) As if that was not enough, the effect size which Hattie uses is problematic. There are of course gains to be found by having meta analyses of huge studies. However, this should not just be read uncritically and without the understanding of the limitations of Hattie’s methods. An overhaul which has been made regarding his methods, by a professor in statistics and mathematics as well as a professor in education, yields the following, very unflattering description of Hattie’s work:

“In summary, it is clear that John Hattie and his team have neither the knowledge nor the competencies required to conduct valid statistical analyses. No one should replicate this methodology because we must never accept pseudoscience. This is most unfortunate, since it is possible to do real science with data from hundreds of meta-analyses.”

There is no unit known as the unit of education. It is not clear if we read Hattie’s research without these very important caveats: critical thinking, understanding of research methodology and the problem of effect sizes. This could lead to us easily being swayed to believe two things: it is only about using formative feedback to students and class sizes do not matter at all.
Without any objective kind of measurement – or at least measurements from different data points – it is very hard to in any serious way put forth an exact analysis. It is therefore crucial to be nuanced and analytical in one’s proceedings. The research from pedagogy, psychology and medicine will help to triangulate – together with the research and knowledge about human kind and learning – there will be a possible pathway through the muddy marshes to get to probable explanations. To simply proclaim one factor as the ruling one is in itself dangerous.

What should we do then?

What can then possibly affect students’ academic results? There is a plethora of answers which naturally places how well the teachers work with the students and what makes good teaching. There is often something which is overlooked in this reasoning, a very key detail which is common sensical: The attendance of the students have a direct correlation with grades. In the previous analysis there were in no way any presentation of data regarding the attendance of the students who had been failing classes. Attendance certainly affects grades. Among other things one can read from a report on schools in Chicago:

“High schools in Chicago have shown substantial success at getting more students on-track in the ninth-grade year by monitoring students’ grades and attendance closely and reaching out right away when students fall behind.”

In substantive respects, it is this which is suggested in the original analysis, which is that we should map out students who are in need of support. This is all good thinking but it is not entirely the teachers’ fault. There was no such possibility earlier and there were no such initiatives from the leadership. Furthermore, a lot of the contact in the role as mentors were being put on the teachers, who would get in touch about issues, which were then if possible, being pursued whether the student needed something more. A lot of time had to be spent to call home, send emails and actively seeking contact with the students, which could all take up a lot of time of one’s workday. That was still not enough. There was no recipient to our concerns, nor any leadership which would put through and follow-up on initiatives regarding the students. The calls could help at times, but other times not at all. There was no plan as to any consequences on student behavior, or at least it was not implemented.
Thus, the same thing which is said above is shown in the following excerpt from an analysis with regards to the importance of attendance affecting the academic results of students:

It is easy to lose focus on supporting students’ academic behaviors, with all of the changing demands around curriculum, pedagogy, and assessments […] figuring out why students have low grades and poor attendance, and addressing those issues, is essential to establish an engaging classroom environment with challenging instruction. Otherwise, the bestplanned lesson falls flat. Getting students to be more engaged in their classes—coming every day they are not sick, turning in all assignments, and putting in their best effort—is what matters the most for their later outcomes.

So, no matter how good the implementation is in classes, it will not matter if the students are not there. Naturally, this is not the one and only considerable factor behind success in school, but it is a large, contributing factor which helps to paint a more nuanced picture as to the number of students failing classes. Some studies have shown that it is in fact possible to already at an early stage, in September, to spot which students are at risk for the rest of the year, in the case of the correlation between attendance and academic results.
The next factor which makes a difference: special adaptation and special support. There was a structural problem and not a problem which should be pinned to the teachers, where there were no coherent nor clear routines as to what to do with students in need of adaptations. Teachers and mentors warned many times of students that were and were not in school. There was a conference every term for every class. In those conferences, we talked about our worries, put forth where there were concerns and tried to implement different actions from there. However, no one seemed to be on the receiving end, as it came to be only a teacher question to do something about this.
This leads us to what really should be done to make sure that there are fewer cases of having to do special adaptations and extra adaptations and so on. Namely, provide the school and teachers with structure and routines. This structure and these routines will make the teachers and students feel safe. If there are no clear structures or routines, then there is no solid ground to stand on.

How do we move on?

The proposed changes from the at large deficit analysis rest on unsteady grounds. The teacher colleague has essentially been introduced to extra adaptations and special adaptations as a sort of magic bullet which will solve everything. There is however, no plan from the analysis as to in how these steps will be implemented. A simple analytical tool from the professor in cognitive psychology, Daniel Willingham, helps us to see if this is something constructive. Before we get to the checklist, Willing believes that if we teachers are going to change our practice, then we will need to be convinced and not ordered. So how do we move on? What steps should we take? The following is suggested:

1. What is the problem that is going to be solved and what is it supposed to improve?
2. How can we know if this will be successful?
3. When is this improvement expected to be visible?
4. What will happen if the goals are not achieved?
5. What evidence are there which support that we should spend time and resources for this improvement?
6. Are the teachers’ expertise and experience being acknowledged? Are the teachers part of the process?