Monday, 18 January 2016

An investigation of the evidence John Hattie presents in Visible Learning


At the 2005 ACER conference (p5) Hattie said,

'We must contest the evidence – as that is the basis of a common understanding of progression.' 

Then in Visible Learning [VL] he quotes Karl Popper (p4)

'Those amongst us unwilling to expose their ideas to the hazard of refutation do not take part in the scientific game.'

The Maths curriculum for all Victorian schools (the state in which Hattie lives) details the following criteria for ALL students to achieve by the end of Year 10:

'Evaluate statistical reports in the media and other places by linking claims to displays, statistics and representative data.' Mathematics Statistics and Probability Levels 7-10A.

So we place a high priority on our students being able to evaluate statistical claims. 

Tom Bennett, the founder of researchEd,  wrote an influential paper 'The School Research Lead', where he states (p9&10),


'There exists a good deal of poor, misleading or simply deceptive research in the ecosystem of school debate.'

'Where research contradicts the prevailing experiential wisdom of the practitioner, that needs to be accounted for, to the detriment of neither but for the ultimate benefit of the student or educator.'

In his excellent analysis 'School Leadership and the cult of the guru: the neo-Taylorism of Hattie', Professor Scott Eacott says (p11),

'The uncritical acceptance of his work as the definitive word on what works in schooling, particularly by large professional associations such as ACEL, is highly problematic.'

Prof Adrian Simpson's detailed analysis of the calculation of effect sizes, The misdirection of public policy: comparing and combining standardised effect sizes states (p451), 

"The numerical summaries used to develop the toolkit (or the alternative ‘barometer of influences’: Hattie 2009) are not a measure of educational impact because larger numbers produced from this process are not indicative of larger educational impact. Instead, areas which rank highly in Marzano (1998), Hattie (2009) and Higgins et al. (2013) are those in which researchers can design more sensitive experiments.

As such, using these ranked meta-meta-analyses to drive educational policy is misguided.
"


Prof Dylan Wiliam writes in 'Getting educational research right', 

'Those ... who focus on ensuring that practice is based on ‘what works’, will find that no educational initiative can be implemented in the same way in every school. Adjustments need to be made, but they need to be made by people who understand the research so that the initiatives do not suffer what Stanford education professor Ed Haertel called “lethal mutations”. Teachers, leaders and policymakers all need to be critical consumers of research.'


The Aim of this Blog:


is to be a critical consumer of research and contest the evidence that Hattie presents in his 2009 book Visible Learning [VL] by using independent peer reviews and by analysing the studies that Hattie used.

The blog is broken up into different pages (menu on the right) designed so you can easily go to what interests you most

Firstly a critique of Hattie's methodology - Effect Size, Student Achievement, CLE and other errors, A Year's Progress and Validity/Reliability. 

Then an analysis of particular influences. I would recommend starting with what was his highest ranked influence 'Self Report Grades' and then look at the controversial 'Class Size'.

In his interview with Hanne Knudsen (2017) John Hattie: I’m a statistician, I’m not a theoretician Hattie states,

'What I find fascinating is that since I first published this back in the 1990s, no one has come up with a better explanation for the data...

I am updating the meta-analysis all the time; I am up to 1400 now. I do that because I want to be the first to discover the error, the mistake' (p7).


I find these comments hard to reconcile since, as you will see, there are many scholars who have identified significant problems in Hattie's work and have called into question his entire model.


I also recommend teachers look at the section - A Years Progress? It analyses what I think is Hattie's most dangerous idea that an effect size of 0.4 = 1 year's student progress.

If you want to contribute please let me know. Many of the controversial influences only have 1-3 meta-analyses to read. I can provide you copies of most of the research used.


Summary:


The peer reviews have documented significant issues with Hattie's work from major calculation errors, misrepresentation to questionable inference and interpretation.

In my view, the most serious of these errors is Hattie's use of studies that do not measure what he claims they do. This occurs in 2 ways:


Firstly, many studies do not measure achievement but something else, e.g., IQ, hyperactivity, behavior, and engagement. See Student Achievement for more details.

Secondly, most studies do not compare groups of students that control for the particular influence that Hattie claims. There is a litany of examples (see menu on the right), e.g., in Hattie's highest ranked influence 'self-report grades', he used a meta-analysis which reports the highest effect size of 3.1.

The author's, Kuncel et al (2005) state their aim (p64),

'Since it is often difficult to get results transcripts of student PREVIOUS GPA’s from High School or College, the aim of this study is to see whether self-reported grades can be used as a substitute. This obviously has time-saving administration advantages.'

These students are reporting their High School GPA afterward to College admissions, which is a measure of honesty, not a prediction of their future grade as Hattie stated!

I contacted Professor Kuncel to make sure I interpreted his study correctly, he replied that the high effect size and conclusion of the study was that: 

'Generally, people exaggerate their accomplishments.'

In the second meta-analysis that Hattie used for 'self-report grades' the study was about peer- review NOT Self!

In Hattie's three published defenses (20102015 & 2017), he never addressed the specific examples of misrepresentation but only generally defends the meta-analysis methodology and the use of his benchmark effect size of 0.4.

Prof's Snook, Clark, Harker, Anne-Marie O’Neill and John O’Neill respond to Hattie's 2010 defense in 'Critic and Conscience of Society: A Reply to John Hattie' (p97),

'In our view, John Hattie’s article has not satisfactorily addressed the concerns we raised about the use of meta-analyses to guide educational policy and practice.'

Prof Arne Kare Topphol responds to Hattie's defense,

'Hattie has now given a response to the criticism I made. What he writes in his comment makes me even more worried, rather than reassured.'

Darcy Moore posts,

'Hattie’s [2017] reply to Eacott’s paper does not even remotely grapple with the issues raised ..'

Prof Eacott also responded to Hattie's defense,

'Disappointed that SLAM declined my offer to write a response to Hattie's reply to my paper. Dialogue & debate is not encouraged/supported ...'

Prof Dylan Wiliam casts significant doubt on Hattie's entire model by arguing that the age of the students and the time over which each study runs is an important component contributing to the effect size. 

Supporting Prof Wiliam's contention is the massive data collected to construct the United States Department of Education effect size benchmarks. These show a huge variation in effect sizes from younger to older students. 

This demonstrates that age is a HUGE confounding variable or moderator since, in order to compare effect sizes, studies need to control for the age of the students and the time over which the study ran. Otherwise, differences in effect size can be due to the age of the students measured!

Given Hattie's conclusion in his 2015 defense (p8),

'The main message remains, be cautious, interpret in light of the evidence, search for moderators, take care in developing stories, welcome critique, ...'

I'm extremely surprised Hattie has not addressed the massive implication of this evidence to his work, all he says in his summary VL 2012 (p14),

'the effects for each year were greater in younger and lower in older grades ... we may need to expect more from younger grades (d > 0.60) than for older grades (d > 0.30).'

Hattie finally agrees (2015 defense, p3) with Prof Wiliam:

'Yes, the time over which any intervention is conducted can matter (we find that calculations over less than 10-12 weeks can be unstable, the time is too short to engender change, and you end up doing too much assessment relative to teaching). These are critical moderators of the overall effect-sizes and any use of hinge =0.4 should, of course, take these into account.'

Yet Hattie DOES NOT take this into account, there has been no attempt to detail and report the time over which the studies ran nor the age group of the students in the question nor adjust his previous rankings or conclusions.


Professor Dylan Wiliam summarises, 

'... the effect sizes proposed by Hattie are, at least in the context of schooling, just plain wrong. Anyone who thinks they can generate an effect size on student learning in secondary schools above 0.5 is talking nonsense.' 

The U.S Education Dept benchmark effect sizes support Wiliam's contention.


We need to move from evidence to quality of evidence:


There must now be at least some hesitation in accepting Hattie's work as the definitive statement on Teaching.

Beng Huat See, in her paper, ‘Evaluating the evidence in evidence-based policy and practice: Examples from systematic reviews of literature', suggests the direction where educational research must now go,

This paper evaluates the quality of evidence behind some well-known education programmes using examples from previous reviews of over 5,000 studies on a range of topics. It shows that much of the evidence is weak, and fundamental flaws in research are not uncommon. This is a serious problem if teaching practices and important policy decisions are made based on such flawed evidence.

Lives may be damaged and opportunities missed.

The aim of this paper is to show how widespread this problem is and to suggest ways by which the quality of education research may be improved. For example, funders of research and research bodies need to insist on quality research and fund only those that meet the minimum quality criteria.


The debate must now shift from Evidence to Quality of Evidence.

The US Dept of Education has done this and has developed clearly defined quality criteria in their What Works Clearing House.

Hattie's Aim:


Hattie uses the REDUCTIONIST approach by attempting to break down the complexity of teaching into simple discrete categories or influences.

Although, Nick Rose has alerted me to another form of reductionism defined by Daniel Dennett - 'Greedy Reductionism' which occurs when,

'in their eagerness for a bargain, in their zeal to explain too much too fast, scientists and philosophers ... underestimate the complexities, trying to skip whole layers or levels of theory in their rush to fasten everything securely and neatly to the foundation.'

I think this latter definition better describes Hattie's methodology.

Additionally, Hattie only uses univariate analysis but complex systems require multivariate analysis. As Prof Jordan Petersen states,

No social scientist worth their salt uses univariate analysis.



Hattie states: 'The model I will present ... may well be speculative, but it aims to provide high levels of explanation for the many influences on student achievement as well as offer a platform to compare these influences in a meaningful way... I must emphasise that these ideas are clearly speculative' (p4).

Hattie uses the Effect Size (d) statistic to interpret, compare and rank educational influences.

The effect size is supposed to measure the change in student achievement; a controversial topic in and of itself (there are many totally different concepts of what achievement is - see here). In addition, surprisingly, Hattie includes many studies that did not measure achievement at all, but rather something else e.g., IQ, hyperactivity, behavior, and engagement.

Also, Hattie claims the studies used were of robust experimental design (p8). However, a number of peer reviews have shown that he used studies with the much poorer design of simple correlation, which he then converts into an effect size (often incorrectly! see -Wecker et al (2016, p27)). Hattie then ranks these effect sizes from largest to smallest.

The disparate measures of student achievement lead to the classic problem of comparing apples to oranges and has caused many scholars to question the validity and reliability of Hattie's effect sizes and rankings, e.g., Higgins and Simpson (2011, p199):


'We argue the process by which this number has been derived has rendered it practically meaningless.'


Blatchford et al (2016, p96) state that Hattie's comparing of effect sizes, 'is not really a fair test'.

'The reconstruction of Hattie's approach in detail using examples thus shows that the methodological standards to be applied are violated at all levels of the analysis. As some of the examples given here show, Hattie's values are sometimes many times too high or low. In order to be able to estimate the impact of these deficiencies on the analysis results, the full analyzes would have to be carried out correctly, but for which, as already stated, often necessary information is missing. However, the amount and scope of these shortcomings alone give cause for justified doubts about the resilience of Hattie's results' Wecker et al (2016, p31).

'the methodological claims arising from Hattie's approach, and the overall appropriateness of this approach suggest a fairly clear conclusion: a large proportion of the findings are subject to reasonable doubt' Wecker et al (2016, p35).

'When taking the necessary in-depth look at Visible Learning with the eye of an expert, we find not a mighty castle but a fragile house of cards that quickly falls apart.Prof Pierre-Jérôme Bergeron.

'To believe Hattie is to have a blind spot in one’s critical thinking when assessing scientific rigour. To promote his work is to unfortunately fall into the promotion of pseudoscience. Finally, to persist in defending Hattie after becoming aware of the serious critique of his methodology constitutes willful blindness.' Prof Pierre-Jérôme Bergeron.

Dr Neil Hooley, in his review of Hattie - talks about the complexity of classrooms and the difficulty of controlling variables, 'Under these circumstances, the measure of effect size is highly dubious' (p44).

'The studies have different effect sizes for different contexts and different levels of schooling, thus averaging these into one metric is meaningless.' Dr Mandy Lupton.

The Common Language Effect Size (CLE) is a probability statistic which usually interprets the effect size. However, three peer reviews showed Hattie calculated all CLE's incorrectly (he calculated probabilities that were negative and greater than 1!). As a result, he now claims the CLE statistic is not important and he focuses on an interpretation that an effect size d = 0.4 is the hinge point, claiming this is equivalent to a year’s progress. Although, there are significant problems with this interpretation.

Although Hattie seems to have another highly doubtful interpretation of probability in a recent interview with Hanne Knudsen (2017) John Hattie: I’m a statistician, I’m not a theoretician Hattie states,

'The research studies in VL offer probability statements – there are higher probabilities of success when implementing the influences nearer the top than bottom of the chart' (p7).

Hattie's Interpretation of the Meta-analyses:


'No methodology of science is considered respectable in the research community when experiment design is unable to control a rampant blooming of subjectivity' Myburgh (2016, p10).

Meta-analysis, as a methodology, has been widely criticised, for not representing the original studies faithfully.

Yet, Hattie takes this interpretation problem to another level as his methodology is META - meta-analysis or MEGA- analysis (Snook et al, 2009).

'the methodology used (by Hattie), neglects the original theory that drives the primary studies it seeks to reviewMyburgh (2016, p4).

Wecker et al (2016, p35) are also critical of this META - meta-analysis methodology,

'... Hattie's work makes clear, a single meta-analysis cannot conclusively answer the question of the effectiveness of an influencing factor anyway. Therefore, meta-analyses should be updated when a significant number of additional primary studies have been added, but not in a second-stage meta-analysis, as in Hattie's work, but as a first-stage meta-analysis based on all existing primary studies ... '

But, 'Hattie assures the research community that he has arrived at sound conclusions based on his confidence that his mega-analysis of meta-analyses consists of quality studies, that the effect sizes faithfully represent a review of the original data and that he adequately explores moderators' Myburgh (2016, p19).


Yet Hattie uses a wide range of meta-analyses which use TOTALLY different experimental designs, on different groups of people (university students, doctors, nurses, tradesmen, and sometimes high school students!), with vastly different measures of student achievement or often no measure of achievement at all!

As Professor Peter Blatchford points out about Hattie's VL,

'it is odd that so much weight is attached to studies that don't directly address the topic on which the conclusions are made' (p13).

Wecker et al (2016, p28) Hattie mistakenly include studies that do not measure academic performance.

A typical example of this is Hattie's influence 'reducing disruptive behavior' where he uses 3 meta-analyses (see the menu on right for details) to get an average low effect size of 0.34. Hattie has often used the polemic 'THE DISASTERS and GOING BACKWARD' to describe influences with low effect sizes. Yet most teachers would say that reducing disruptive behavior is one of their major aims.

A detailed look at the 3 meta-analyses is revealing - one compares the achievement of students with ADHD with a normative group, the other compares the behavior of students with 'emotional/behavioral' disturbance (EBD) with a normative group. The third does focus on reducing disruptive behavior but measures behavior NOT achievement! In addition, one meta-analysis gets a large negative result, i.e., student achievement decreased!

HOW CAN REDUCING DISRUPTIVE BEHAVIOUR DECREASE STUDENT ACHIEVEMENT?

A thorough look at this study clearly shows Hattie's misinterpretation, the study compares emotionally disturbed kids with 'normal' kids! The NEGATIVE result comes from an effect size calculation using (emotional/behavioral disturbance achievement) - (Normative achievement). Yet in other behavioral studies, the effect size is calculated the opposite way giving a positive result!

Hattie should have adjusted for this inconsistency. If this were done the average effect size would rocket up in Hattie's rankings to #6. If studies were removed then ranking would rise further to #3.

This would then be consistent with teachers' experience.

A key tenet of the scientific method is reliability, this simple analysis demonstrates how unreliable Hattie's rankings are.

Professor Adrian Simpson insightfully identifies this problem (p455),

'Using unequal comparisons or using unspecified ones makes it impossible to compare or combine effect sizes meaningfully.'

With Evidence Like This Who Needs Your Opinion:


In spite of these significant errors, Hattie uses trite slogans like 'know thy impact' or 'statements without evidence are just opinions'. This belittles teacher experience and opinion and raises his so-called evidence and rankings above them.

Nick Rose and Susanna Eriksson-Lee in their excellent paper 'Putting evidence to work', quote a more provocative slogan from Kevan Collins, Chief Executive of the Education Endowment Foundation (EEF),

'if you're not using evidence to inform your decisions, you must be using prejudice' (p5).


In his interview with Hanne Knudsen (2017) John Hattie: I’m a statistician, I’m not a theoretician, Hattie seems to have retreated from this polemic,

'Evidence can also be related to experience – and the extensive experience of many teachers is legitimate evidence – to be contested, to be examined, and to be evaluated – in terms of the
best impact on the learning lives of students. When there are differences between the evidence from the research and from experience, then there is a need for examination, for reflection, for seeking more avenues of evidence – and I want this to be via the effects on the students
' (p7).


The Problem of Breaking Down the Complexity of Teaching into Simple Categories, Influences or 'Buckets':


'The partitioning of teaching into smallest measurable units, a piecemeal articulation of how to improve student learning, is not too removed from the work of Taylor over 100 years ago. Despite its voluminous and fast expanding literatures, educational administration remains rooted to the same problems of last centuryEacott (2017, p10).

By Professor Robert Sapolsky in his course 'Introduction to Human Behavioral Biology' (see 47:20 - 48:30).



Peer Reviews:


Professor John O'Neill has reviewed these influences: micro-teaching, professional development, providing formative evaluation, comprehensive interventions for learning disabled students, feedback, spaced vs. massed practice, problem-solving teaching, metacognition strategies, teaching strategies, co-operative vs. individualistic learning, study skills and mastery learning.

Dr Kristen Dicerbo has analysed self-report grades.

Dr Mandy Lupton has analysed Problem-Based and Inquiry-Based Learning.

Professors Higgins and Simpson have published Hattie's calculation errors.

Professor Arne Kare Topphol also published Hattie's calculation errors (in Norwegian) summary here.

Professor Ivan Snook et al, give a general critique of VL focusing on lack of quality studies, the problems of Hattie's ranks and generalisations. They use class size and homework as examples.

Professor Ewald Terhardt published a general critique of Hattie's methodology and issues of Hattie's conflict of interest.

Hattie's retort to Snook and Terhardt, which is basically a defense of meta-analysis as a methodology.

Snook, Clark, Harker, Anne-Marie O’Neill and John O’Neill, a reply to Hattie's retort.

Dr Myburgh analysed Hattie's retort to Snook, et al and Terhardt above. Myburgh focuses on the critique of meta-analysis as a methodology and not the specific critiques of Hattie's misrepresentations.

Professor Bergeron (2017) published Hattie's calculation errors plus other issues about correlation studies and misinterpretation.

'When taking the necessary in-depth look at Visible Learning with the eye of an expert, we find not a mighty castle but a fragile house of cards that quickly falls apart' (p1).


Plant (2014) critiques Hattie methodology showing major flaws in calculations and interpretations.

'an attempt was made to argue that the hitherto most comprehensive research findings synthesis in education, John Hattie's meta-analysis Visible Learning (2009), although in many places a commitment to the importance of differentiating considerations of the findings of meta-analyzes surrenders that his own methodological approach, however, this seems more like lip service' (p96).

Wecker et al (2016) published major issues of misrepresentation, major calculation & interpretation errors.

'the methodological claims arising from Hattie's approach, and the overall appropriateness of this approach suggest a fairly clear conclusion: a large proportion of the findings are subject to reasonable doubt' (p35).

Professor Timothy Shanahan in his article, Why You Need to Be Careful About Visible Learning (2017) shows that Hattie often counts the same studies twice resulting in unreliable effect sizes. Also, he shows that Hattie gives the same weighting to meta-analyses even though there are vast differences in their quality, which also leads to unreliable results.

Professor Scott Eacott's (2017) critique of the 'cult of Hattie'; how and why it came to be and its dangers.

Kelvin Smythe e.g., 'John Hattie: your research is now a con.'

Whilst not directly about Hattie's evidence for feedback, David Didau gives an excellent overview of the evidence for feedback here. Also, Gary Davies has an excellent blog - Is Education Research Scientific?

Why has Hattie become so popular?


'Hattie’s work has provided school leaders with data that appeal to their administrative pursuitsEacott (2017, p3).

'Garbage in, Gospel out' Dr Gary Smith (2014)

What has often been missed is that Hattie prefaced his book with significant doubt 'I must emphasise these are clearly speculative' (p4)Yet, his rankings have taken on 'gospel' status due to: the major promotion by politicians, administrators and principals (it's in their interest, e.g. class size), very little contesting by teachers (they don't have the time, or who is going to challenge the principal?) and limited access to scholarly critiques - see Gary Davies excellent blog on this.

'Materialists and madmen never have doubts' G. K. Chesterton

Interestingly, his reservation has changed to an authority and certainty that is at odds with the caution that ALL of the authors of his studies recommend, e.g., class size and ability group. Caution due to lack of quality studies, inability to control variables, major differences in how achievement is measured and the many confounding variables. Also, there is significant critique by scholars who identify the many errors that Hattie makes; from major calculation errors and excessive inference to misrepresenting studies, e.g., Higgins and Simpson (2011)Wecker et al (2016).


The Rise of the Policy Entrepreneur:


Science begins with skepticism, however, in the hierarchical leadership structures of Educational Institutions skeptical teachers are not valued, although ironically, the skeptical skills of questioning and analysis are valued in students.  This paves the way for the many 'snake oil' remedies and the rise of policy entrepreneurs who 'shape and benefit from school reform discourses'.

Professor John O'Neill in analysing Hattie's influence on New Zealand Education Policy describes the process well:

'public policy discourse becomes problematic when the terms used are ambiguous, unclear or vague" (p1). The "discourse seeks to portray the public sector as ‘ineffective, unresponsive, sloppy, risk-averse and innovation-resistant’ yet at the same time it promotes celebration of public sector 'heroes' of reform and new kinds of public sector 'excellence'. Relatedly, Mintrom (2000) has written persuasively in the American context, of the way in which ‘policy entrepreneurs’ position themselves politically to champion, shape and benefit from school reform discourses' (p2).

Hattie's recent public presentation in the TV documentary 'Revolution School' confirms Professor O'Neill analysis. Dan Haesler reports Hattie's remedy cost the school around $60,000.

Professor Ewald Terhardt (2011, p434), 'A part of the criticism on Hattie condemns his close links to the New Zealand Government and is suspicious of his own economic interests in the spread of his assessment and training programme (asTTle)'.


Ambiguous, Unclear or Vague?


There are many examples of ambiguity in the detailed analysis of each influence on the right menu. Although, the first striking one is in Hattie's preface to VL:

"It is not a book about classroom life, and does not speak to the nuances and details of what happens within classrooms."


However, many influences such as class size, teacher subject knowledge, teacher training, ability grouping, student control, mentoring, teacher immediacy, problem-based learning, exercise, welfare, and homework are considered to be about classroom life but Hattie has given them a low ranking. 


In Hattie's 2012 update of VL he does an about face and says,


'I could have written a book about school leaders, about society influences, about policies – and all are worthwhile – but my most immediate attention is more related to teachers and students: the daily life of teachers in preparing, starting, conducting, and evaluating lessons, and the daily life of students involved in learning' (preface).

Hattie also promotes Bereiter’s model of learning, 

'Knowledge building includes thinking of alternatives, thinking of criticisms, proposing experimental tests, deriving one object from another, proposing a problem, proposing a solution, and criticising the solution … ' (VL p27).

'There needs to be a major shift, therefore, from an over-reliance on surface information (the first world) and a misplaced assumption that the goal of education is deep understanding or development of thinking skills (the second world), towards a balance of surface and deep learning leading to students more successfully constructing defensible theories of knowing and reality (the third world)' (p28).

Prof Proulx, Critical essay on the work of John Hattie for teaching mathematics: Entrance from the Mathematics Education, explains the contradiction,

'... ironically, Hattie self-criticizes implicitly if we rely on his book beginning affirmations, then that it affirms the importance of the three types learning in education... '

'So with this comment, Hattie discredits his own work on which it bases itself to decide on what represents the good ways to teach. Indeed, since the studies he has synthesized to draw his conclusions are not going in the sense of what he himself says represent a good teaching, how can he rely on it to draw conclusions about the teaching itself?'


Also, in his presentations, he describes many of these low ranked influences as DISASTERS! 

This seems to DEFY widespread teacher experience.



Is Hattie’s Evidence Stronger than Other Researchers or Widespread Teacher Experience?


A summary of the major issues scholars have found with Hattie's work (details on the page links on the right):


  • Hattie misrepresents studies e.g. peer evaluation in 'self-report' and studies on emotionally disturbed students are included in 'reducing disruptive behavior'.
  • Hattie often reports the opposite conclusion to that of the actual authors of the studies he reports on, e.g. 'class-size', 'teacher training', 'diet' and 'reducing disruptive behavior'.
  • Hattie jumbled together and averaged the effect sizes of different measurements of student achievement, teacher tests, IQ, standardised tests and physical tests like rallying a tennis ball against the wall.
  • Hattie jumbled together and averaged effect sizes for studies that do not use achievement but something else, e.g. hyperactivity in the Diet study, i.e., he uses these as proxies for achievement, which he advised us NOT to do in his 2005 ACER presentation.
  • The studies are mostly about non-school or abnormal populations, e.g., doctors, nurses, university students, tradesmen, pre-school children, and 'emotionally/behaviorally' disturbed students.
  • The US Education Dept benchmark effect sizes per year level, indicate another layer of complexity in interpreting effect sizes - studies need to control for age of students as well as the time over which the study runs. Hattie does not do this.
  • Related to the US benchmarks is Hattie's use of d = 0.40 as the hinge point of judgments about what is a 'good' or 'bad' influence. The U.S. benchmarks show this is misleading.
  • Most of the studies Hattie uses are not high quality randomised controlled studies but the much, much poorer quality correlation studies.
  • Most scholars are cautious/doubtful in attributing causation to separate influences in the precise surgical way in which Hattie infers. This is because of the unknown effect of outside influences or confounds.
  • Hattie makes a number of major calculation errors, e.g., negative probabilities.


Misrepresentation:


The reducing disruptive behavior studies are are an example of Hattie's misrepresentation. Another key example is class size. Hattie interprets the meta-analysis differently to the actual authors of the study. I was surprised to find this a common issue in VL.


Class Size:


For example, Glass and Smith (1979), 1 of the 3 studies that Hattie uses for class size, summarise their data in a graph and table:



The trend and the difference between good and poor quality research are clearly displayed. The authors conclude,

'A clear and strong relationship between class size and achievement has emerged... There is little doubt, that other things being equal, more is learned in smaller classes' (p15).



Hattie uses an average (which is another issue discussed) from the above table of d = 0.09 (although it seems the average is closer to d = 0.25). Hattie concludes class size has minimal impact on student learning. In fact, he goes further than this, in his 2005 ACER presentation (using this research) he calls class size a DISASTER! Other times he interprets d < 0.40 as "going backward"!


The Quality of the Research:

'Extraordinary claims require extraordinary evidence.' Carl Sagan

Generally, Hattie dismisses the need for quality and makes the astonishing caveat, that there is, 

'... no reason to throw out studies automatically because of lower quality' (p11).

Hattie's constant proclamation (VL 2012 summary, p3),

'it is the interpretations that are critical, rather than data itself'  

is worrying, as it is opposite to the Scientific Method paradigm as Professor Ivan Snook et al (2009, p2) explain:

'Hattie says that he is not concerned with the quality of the research ..., of course, quality is everything. Any meta-analysis that does not exclude poor or inadequate studies is misleading, and potentially damaging if it leads to ill-advised policy developments. He also needs to be sure that restricting his data base to meta-analyses did not lead to the omission of significant studies of the variables he is interested in.'

Professor John O'Neill writes a significant letter to the NZ Education Minister regarding the poor quality of Hattie's research, in particular, the overuse of studies about University, graduate or pre-school students and the danger of making classroom policy decision without consulting other forms of evidence, e.g., case and naturalistic studies. 

'The method of the synthesis and, consequently, the rank ordering are highly problematic' (p7).

The U.S. Department of Education has set up the National Center for Education Research whose focus is to investigate the quality of educational research. Their results are published in the What Works Clearing House. They also publish a Teacher Practice Guide which differs markedly from Hattie's results - see Other Researchers.

Importantly they focus on the QUALITY of the research and reserve their highest ratings for research that use randomised division of students into a control and an experimental group. Where students are non-randomly divided into a control and experimental group for what they term a quasi-experiment, a moderate rating is used. However, the two groups must have some sort of equivalence measure before the intervention. A low rating is used for other research design methods - e.g., correlation studies.

Given most of the research that Hattie uses is correlation based, he has skillfully managed to sidestep the quality debate within school circles (but not within the academic community - see References).


Self-Report Grades - the Highest Ranked Influence??


Hattie concludes the ‘best’ influence is self-reported grades with d=1.44. Which Hattie interprets as advancing student achievement by 3+ years!

This is an AMAZING claim if true: that merely predicting your grade, somehow magically improves your achievement to that extent. I hope my beloved “under-achieving” Australian football team – The St Kilda Saints are listening – “boys you can make the finals next year just by predicting you will - you don't need to do all that hard training!"


Whilst it may be simpler and easier to see teaching as a set of discreet influences, the evidence shows that these influences interact in ways in which no-one, as yet, can quantify. It is the combining of influences in a complex way that defines the 'art' of teaching. 

A Teacher's Lament:


Gabbie Stroud resigned from her teaching position and wrote:
'Teaching – good teaching - is both a science and an art. Yet in Australia today [it]… is considered something purely technical and methodical that can be rationalised and weighed.

But quality teaching isn't borne of tiered 'professional standards'. It cannot be reduced to a formula or discrete parts. It cannot be compartmentalised into boxes and 'checked off'. Good teaching comes from professionals who are valued. It comes from teachers who know their students, who build relationships, who meet learners at their point of need and who recognise that there's nothing standard about the journey of learning. We cannot forget the art of teaching – without it, schools become factories, students become products and teachers: nothing more than machinery.'



John Oliver gives a funny overview of the problems with Scientific Studies:


Another overview the issues with published studies-