Dimensions of Secondary School Effectiveness: Comparative Analyses Across Regions

Abstract

Recent UK government policy implementing new systems of evaluation and accountability have highlighted the use of performance data to inform judgements about secondary schools and stimulate school improvement. However, these developments have been informed by a relatively small number of research studies addressing the methodology of measuring school effectiveness, and often employing limited or incomplete datasets. This paper reports the ®ndings of an Economic & Social Research Council (ESRC) funded study that employs 6 extensive and detailed regional datasets (drawn from Lancashire, London, Jersey, Scotland, the Netherlands and England as a whole). The study aims to provide new evidence to assist school staff, policy-makers and academics in understanding the multi-faceted nature of school effectiveness and the need to evaluate school performance in detail. The objectives were to investigate 1) the optimal models for measuring secondary school effectiveness across a range of outcomes in the UK and abroad; 2) the extent of regional differences in the results; 3) the de®nition of the underlying dimension(s) of school effectiveness across different regional and policy contexts. The ®ndings show that at least 4 dimensions of secondary school effectiveness can be de®ned, speci®cally in terms of different outcomes, pupil groups, pupil cohorts and curriculum stages. In addition regional differences appear to exist in the size and impact of school effects, and these are mirrored by differences in regional context in terms of pupil selection. In conclusion it is argued that effectiveness at different levels of the education system (e.g., individual pupils; departments; whole school; region and nationally), as well as interactions between levels, needs to be continually monitored in order to map out the boundaries of school effectiveness and how these change over time. The ®ndings are discussed in relation to developing a value added framework for school evaluation in the UK.

Educational effectiveness in Chilean secondary education: comparing different ‘value added’ approaches to evaluate schools

“…The results obtained by this study regarding Research Question 1 are in line with previous research conducted in Chile and elsewhere, especially when pointing out that CVA models typically provide the best fit of the data (in terms of total variance explained) and that differences between schools remain, although to a lesser extent, after student background characteristics, school contextual factors and prior achievement have been taken into account (Thomas, 2001;Opdenakker et al, 2000Opdenakker et al, , 2002Opdenakker et al, , 2006Muijs and Reynolds, 2003;Webster andFisher, 2000, Cervini, 2009). …”

“…Previous research mostly conducted in Europe and USA has demonstrated that the structural levels in the education system below (classroom, department) and above (municipality, province, country, etc.) the school level, have played an important role in different contexts when exploring the relative importance of schools (Bosker and Scheerens, 1989;Goldstein, 1997;Opdenakker and Van Damme, 2000;Thomas, 2001;Luyten, 2003;Van den Noortgate, Opdenakker and Onghena, 2005;Cervini, 2009;Martínez, 2012). By not specifying levels both below and above the school in multilevel analyses when estimating school effects, model results can potentially be both miss-specified and misleading in terms of overestimating school effects.…”

The Limitations of using School League Tables to Inform School Choice

“…These studies mostly consider correlations for school effects over three, four and five years. At GCSE, Wilson and Piebalga (2008) report a correlation of 0.62 for value-added school effects two years apart whilst Gray et al, (1996) and Thomas (2001) report correlations of 0.56 and 0.55 respectively for four years apart. Similar magnitudes have been reported at A-level: Gray et al (2001) and Yang and Woodhouse (2001) both report correlations of around 0.55 for school effects four years apart.…”

Of Trends and Trajectories: Searching for patterns in school improvement

“…Moreover, it is likely to be a mistake to assume that the practices of the schools recognised as outstanding are ones that should be adopted by other schools. (p. 36) Our own research would tend to support the view that changes in schools' performance over time are a good deal more modest and variable than is sometimes supposed (Gray et al, 1996(Gray et al, , 2001Thomas, 2001). Only a small minority of institutions seems to improve consistently over time; the greater majority has ups and downs.…”