The National Literacy and Numeracy Strategy (DES, 2011) calls for greater use to be made of assessment data at both school and system levels. At school level, it is proposed that the results of standardised tests should be used to monitor aspects of student learning, to inform school self-evaluation processes, and to track the achievement of groups of students at risk of underperformance. At system level, the Strategy calls for aggregated assessment information to be used to form an overall national picture of performance in literacy and numeracy (see also OECD, 2012).
Internationally, the use of sophisticated statistical techniques such as value-added modelling to compare performance across different schools is common place. To date, comparatively little attention has been given to using quantitative approaches and large scale assessment data for assessing school effectiveness in the Republic of Ireland.
The current project makes use of existing datasets in order to consider whether or not schools can be said to perform significantly different to expectations, after controlling for student background characteristics. Datasets explored as part of these analyses include the 2009 National Assessments, the Programme for International Student Assessment, Junior Certificate examination results and data gathered as part of the DEIS evaluation.
Preliminary findings suggest that existing datasets are of limited use for distinguishing between schools on the basis of performance as the majority of large scale educational assessments currently conducted in Ireland are cross-sectional in nature and therefore do not include measures of achievement over time. The limitations of cross-sectional studies for these types of analyses have previously been identified in the literature (see e.g., OECD, 2008). A further limitation of many existing datasets is that data are available for samples of students only. Although population data are available in state examination databases, these provide very limited data on student background.
