This is a follow on from the previous post, as it surprisingly raised a number of questions. I have updated the post with some extra discussion on the research design to include the following:
The “teacher” was a research assistant who “read aloud from a book… which is a common instructional activity in kindergarten”. Two pages were read to the children. The assessment of learning consisted of “asking the children to select the correct answer from four pictorial response options” in a workbook. Now this is all a far cry from a standard classroom, where a teacher knows the children, and uses the room as part of the educational environment. The researchers made the point that “distractibility…decreases markedly with age” and hence would not be likely to be an issue with older learners.
In the study, “distraction” was measured by the “direction of the pupils’ gaze”; anyone not looking at the teacher was marked as “distracted”. “Learning” was measured by a written test after the lesson which consisted of the pupils being read to. The correlation between ‘time off task’ (the measure of which the study admits is contentious) and learning score is minimal. As is the case with studies in this genre, considerable attention has been given to measurement and the calculation of statistics – down to two decimal places in some cases. In that sense, the results are quite statistically reliable, that is, we are all likely to get the same results. However it has weak explanatory validity since there is likely to be disagreement over exactly what it claims to be measuring – distraction and learning.
In a comment to the first post from Mike Ollerton, he raises the following questions:
“Why might learners become distracted beyond the fact there was some display work in their classroom? Might some of the reasons be due to:
A) the teacher does too much talking and telling so learners have to do too much listening?
B) the tasks a teacher offers learners lack interest and stimulation?
C) the teacher asks too many rhetorical/closed questions?
D) the teacher is too quick to accept answers from students which the teacher wanted to hear in the first place?
E) learners are not given sufficient time nor are encouraged to discuss the questions their teacher asks?”
Now these are all reasonable questions. Given the content of the “lesson” however, it is difficult to see many opportunities for pupils to engage; the teacher did all the talking, the task was merely an assessm
ent, there were no discussion opportunities.
So in any evaluation of research, it is critical to consider alternative hypotheses or explanations which fit the data but which don’t require the initial assumptions. In some ways these are termed “counterfactuals“, exploring possible implications if the principle asumptions were incorrect. In this research this is the assumption that the visual environment is influential in children’s attention and this influences their learning.
So, let me consider an alternative, counterfactual, explanation to the one offered that does not require the level of display to be the independent variable.
The activity (an unknown adult reading two pages from a book) was uninspiring to the children and they lost focus and some looked around the room for stimulus, because it was all a novelty to them being the first they has seen it (distraction). All sorts of interesting things were on display. As their attention shifted, they stopped listening. When the room was stripped out, they had nothing interesting to look at so were forced to listen to more of the story. As a result they focussed on and remembered more (learning).
What we do not know, because it wasn’t measured, was what they did attend to, and maybe learn, when the room was decorated. This was presumably not considered because the researchers were looking for distraction, rather than stimulus. Both of the variables (distraction and learning) were specifically designed to be easily measurable, locating the work within a positivist tradition. In some places, as mentioned in the previous post, measurement is reported to two decimal places. This is quite typical of such (quantitative) work, where considerable attention is given to the calculation of measures and statistics, at the expense of interpretation. Consequently, the research has high construct reliabilty (we would all get the same values), but is low on explanatory validity (there is space to disagree on what is exactly being calculated – in this case distraction and learning)
This was a small scale clinical study after all and I strongly suspect, designed as a small part of a doctoral study. If that being so, why was it published? Academics and researchers need to write, communicate and publish. In the USA it forms a large part of one’s promotion. In the UK it forms a central part of one’s contract of employment. Your publications list becomes who you are! For those studying for a doctorate, publishing is part of the expectation and initiation into academia. There are advantages and disadvantages in this, not least in how one’s published work becomes interpreted by the public. But also because sensationalist headlines in the popular press are rarely the result of systematic literature reviews.
One response to the Fisher et al. article comes from Kana Imuta and Damian Scarf, of the Universities of Queensland and Otago. After a similar critique of the research, they present a re-analysis of Fisher’s et al.’s data. (I am very grateful to Anne Watson for bringing this to my attention). They focussed on the idea of “novelty” which I mentioned above. By looking at the data for individual lessons they found, unreported by Fisher et al. who had aggregated the data from each lesson, that “distraction” reduced with each lesson as the novelty wore off even over three lessons. They finish with this:
Beyond a momentary distraction, classroom decorations may actually be beneficial in the longrun. For example, allowing children to decorate the classroom with their own work may improve their self-esteem and beliefs about the value of their work .Also, when classroom decorations are related to what the children are learning, the decorations may act as “reminder” cues, improving children’s long-term memory of educational information. Taken together, when considering the valuable question of how to optimize the classroom visual environment, we should consider how the relevance as well as the novelty of the visual displays interact to influence children’s learning outcomes.
This is a far cry from where we started and the recommendation in the Times Educational Supplement of all places for teachers to “ditch their display“.
An Alternative Perspective on the Visual Environment
Let me offer an alternative research perspective to those who might still argue that classroom display detracts from learning. This time from three colleagues at the School of Education at the University of Nottingham, Pat Thomson and Chris Hall and Lisa Russell. Lisa now works at the University of Huddersfield, and her research is worth exploring. I refer to their research article titrled “If these walls could speak: reading displays of primary children’s work” which you can download here:
In this three year ethnographic study, Pat, Chris and Lisa studied one Primary school. The summary of their work (the abstract) reads as follows. As you read this contrast it with Mark Enser’s call to “ditch your display“.
The first thing a visitor notices when entering Hollytree primary school is the art-work displayed on every wall. This paper, based on a three-year ethnographic study of the school, mobilizes field notes and interview and photographic data to probe the meanings of this visual ‘display’. We argue that the walls (re)produce and promote normative meanings of ‘good work’, the ‘good student’, the ‘good teacher’ and the ‘good school’, which serve both internal and external purposes. They are also a means of promoting an inclusive culture which, while true of arts activities, may not always be the case in mainstream classes. In addition, the school walls support aspects of the school timetable of collective work, and also constitute resources for children to construct narratives about their collective and individual histories in the school.
What I feel is important to understand here is there are two quite different research narratives – a quantitative clinical-positivist study by two psychologists and a statistician (Fisher, Godwin & Seltman, 2014) and a qualitative interpretist-ethnography study by three educational rearchers and ex-teachers (Thomson, Hall & Russell, 2007).
As Pat, Chris and Lisa say in their paper, primary schools celebrate “children’s work on walls, hanging from ceilings and adorning school prospectuses and websites”. Furthermore, creating a vibrant environment is actively promoted in official Goverment policy; the support materials for the English primary strategy, Excellence and Enjoyment state:
The physical environment has a significant influence on learning. It gives children clear messages about how we value them and how we value their learning. It can be supportive of independent learning. (Department for Education and Skills (DfES), 2004, p. 56).
Notably, in 2004 when this was written, the UK had a Labour Government. Whether the current Conservative Government values children and their independent learning is a moot point.
Now, the school is described by Pat, Chris and Lisa in the following terms:
Half the children live in local authority housing on the two council estates close to the school. A higher than average percentage of children qualifies for free school meals (24%), and the school has never excluded a pupil. The school has a reputation as successful, and inspection reports which deem it ‘outstanding’. […] every available area of wall is covered in displays. The hall, foyer and shared areas are dominated by children’s portraits in various media, sculpted figures, pages from an alphabetized poetry book, and mounted pages of autobiographies and biographies. Some of the work, such as the frieze in the hall, is clearly the work of a professional artist. Classrooms typically display mounted collages of children’s curriculum-based work, much of it with teachers’ comments attesting to its worth.
There was no escaping the human face and body when walking through or working in Hollytree primary. The school was person-centred and the walls literally said it is people who are important here, people of all shapes, sizes and colours, all have a place in our school. This was an individualised notion of ‘recognition’, but one which was highly inclusive of diversity. Hollytree was a school where even the walls suggested that everyone could ‘become somebody’ worthy of notice.
There are a number of things to take away from this. Firstly one needs to be careful about how one interprets and uses research articles.
I will finish here, for now with a final quote from the conclusions of Pat, Chris and Lisa’s article, because it returns us to where I started on classroom display, and because it offers a sense of hope not despair:
We were surprised to hear the children’s stories which were clearly important to their sense of belonging in the school. We suggest that further analyses of the displays in different schools in different locations might offer more such surprises, and, thus, help our understandings of the ways that schools ‘work’ in and as everyday practice.
Educational research is a very specialised business drawing on skills and dispositions, and technical knowledge accumulated over a number of years. It does the teaching profession no favours to jump on bandwagons, use a limited set of sources as a totem to push an agenda, whether that be cognitive load theory, direct instruction, variation etc. without understanding the philosophical, political and methodological conflicts that underpin social and psychological and cognitive research.
Thanks you for reading this far! More to come. I am on a roll.
Anna Fisher, Karrie Godwin and Howard Seltman (2014a). Visual environment, attention allocation, and learning in young children: when too much of a good thing may be bad, Psychological Science 25, 1362–1370.doi:10.1177/0956797614533801
Kana Imuta and Damian Scarf (2014) When too much of a novel thing may be what’s “bad”: commentary on Fisher, Godwin, and Seltman (2014) Frontiers of Psychology, 23 December 2014, doi.org/10.3389/fpsyg.2014.01444
Pat Thomson, Christine Hall, and Lisa Russell (2007) ‘If these
walls could speak: reading displays of primary children’s work’, Ethnography and Education, 2:3, 381 – 400, DOI: 10.1080/17457820701547450