This is the fourth post in a series, which together make up a literature review submitted for my MEd in Educational Research at the University of Cambridge. You can read also read part one, part two, part three, part five and part six. I hope that this is of interest to anyone considering undertaking an MEd, or for those who wish to dig a little into the academic literature around educational effectiveness Research and School Improvement. I would be most grateful to anyone who can provide any critique to what is written. References can be found at the end of part six.
- History of Educational Effectiveness Research
It is difficult to precisely define EER, for two principal reasons. Firstly, EER describes not just one field of research, but rather a ‘conglomerate of research in different areas’ (Creemers & Kyriakides, 2012, p.3). Secondly, the discipline has has undergone fairly major shifts over the last three or four decades, largely in response to criticisms from the research community. The more prominent and transfigurative criticisms will be explored in §5, but a good overview is provided by Townsend (2001).
It is also worth noting that I am referring to the field as educational effectiveness research throughout this review, but this name is a fairly recent development. For a long time the discipline was known as School Effectiveness Research, before the new label EER was adopted to reflect the need for a broader base, partly as a result of a rapidly changing educational landscape (Muijs, 2006). During this review I shall refer to the overarching discipline as EER, but may speak of School Effectiveness Research if the context compels such terminology. This relabelling is a good example of what could be termed a ‘retrospective reactive, not purposive’ (Reynolds et al., 2014, p.201) shift in the content of the field. It would not be entirely unfair to suggest that much of the discipline’s content has arisen as reaction to its critics, although a more recent commitment to a theoretical underpinning to the field is reversing this (Creemers and Kyriakides, 2012).
As a result of the reactive nature of EER, the field is best understood by tracing its history. It is customary to begin with the aforementioned studies by Coleman and Jencks, since it is usually agreed that EER was born as a reaction to the conclusions of these papers, which raised serious questions about whether schools made any difference to pupil outcomes in comparison to background factors such as socio-economic factors and ability. As we will see, the particular circumstances of the field’s origin continues to influence its identity. In particular, the discipline’s initial purpose to demonstrate that schools did make a difference – contra Coleman and Jencks – has resulted in a preoccupation with a quantitative study of effectiveness, focusing at the school level, which may contribute to the limited uptake of findings by practitioners.
Reynolds et al (2011) suggest that the history of EER has five distinctive phases, and this is a useful framework from which to understand the field. Greater focus will be given to the more recent developments in EER, which will allow us to arrive at the dynamic model of educational effectiveness and appreciate how it addresses key concerns from the literature. This will then guide future directions, allowing me to isolate the links between EER and school improvement, in particular practice at the classroom level.
Phase One – Demonstrating Variation in School Effects
The notion that schools were impotent to alter predetermined student outcomes was set in the 1960s and the view remained entrenched until the 1980s. Alongside the aforementioned works of Coleman and Jencks, the Plowden Report was published in England in 1967 adding weight to the idea that school influences paled in significance compared to home factors. In response to this educational researchers undertook empirical work to demonstrate that schools did vary in their effects. The case was set out most emphatically by Rutter et al. (1979) in their book book length report of a nine year longitudinal study in secondary schools in England. Many of the findings still echo in what we now know to be effective in securing good educational outcomes – the idea that children ‘live up (or down) to what was expected of them’ was considered a ‘provocative finding’ in an early review of the book (Bardon, 1980, p. 557).
Phase Two – Methodological Development, Including Multi-Level Methodologies
Once it had been established that schools do have an effect on student outcomes, the field was compelled to demonstrate more scientifically the effects. This meant that methodological advances were required, and in particular the adopting a multi-levelled understanding of schooling. Such an understanding took into account that educational effectiveness was complex, with factors at the national, school, classroom and student level all interacting.
Phase Three – Investigating Why Schools Had Different Effects
By the mid 1990s, the educational research community had successfully rebuked the idea that schools did not have a meaningful effect on student outcomes. The second phase of EER had begun to explore the relationships between the different levels of factors involved in outcome and had also examined the outcomes (and their stability) with increased nuance (Goldstein, 1995). What followed was attempts to outline why schools differed in their effects (as opposed to the earlier efforts of establishing that they differed in their effects). Thus, the contextual factors at the school level were further probed and studies moved away from an ‘input/output’ understanding towards an ‘input/process/output’ understanding (Teddlie & Reynolds, 2000).
Phase Four – Internationalisation and Links to School Improvement
The quantitative methodology approach up until this point was necessitated by the field’s initial goals, but by the mid 1990s there was greater acceptance of the qualitative methods preferred by research into school improvement. As such, the ‘rich, thick descriptions’ which could help ‘explain the relationships’ (Reynolds et al., 2014, p. 201) between effective factors and student outcomes began to be addressed, although Kyriakides and & Creemers concede that EER has continued to focus too strongly on school characteristics ‘without looking at the processes that are need to change the situation’ (2012, pp. 7-8).
During this period, however, the was successful in collaborating internationally and grew rapidly as a result. The establishment in 1988 of the International Congress of School Effectiveness and Improvement helped to spearhead this development and unify disciplines under the umbrellas of EER and school improvement. The internationalisation has allowed joint projects, helping to identify global factors associated with school improvement.
Phase Five – Development in Statistical Analysis and the Emergence of the Dynamic approach
The final phase of EER’s journey brings us up to date with the field as it stands, and is characterised by the development of increasingly sophisticated statistical analysis along with a reconceptualisation of EEF as a dynamic set of relationships. This includes not only the interaction between student and teacher, but between the different levels of influence on student outcomes: national level, school level, classroom level and student level. Each of these is related to broader outcomes that take into account what Creemers and Kyriakides term the ‘new goals of education’ including affective outcomes such as affective and psychomotor outcomes.
This dynamic model will be explored in greater depth in § 6, but for now we will turn to the more compelling criticisms that the EER faces, which will result in implications for research questions emerging in the final section (§7).