Describing reading and numeracy assessments captured at a tertiary institute – Patterns of use for ethnicities, gender, as well as mother-tongue and non-mother-tongue speakers of English (2011-2014)

Abstract or Summary

In this report, we describe our attempt at making sense of literacy and numeracy assessment tool (LNAT) data collected at the institute from inception to December 2014. We describe the challenges we faced in coding and restructuring the LNAT data-extraction files to process all LN performance data. We excluded vocabulary and writing assessment scores. We embarked upon data processing that required line-by-line coding to create labels for processing all reading and numeracy assessment scores. The two coding categories involved the Score Categories variable with four categories of score, and the Pre-post Category variable consisting of three categories. Once approximately 15 219 numeracy and 15 854 reading records had been coded and cross-validated, we then used layered cross-tabulations and clustered bar charts to explore the value hidden in a somewhat messy LNAT data-extraction file layout. For both reading and numeracy, we report institutional performance data for cross-tabulations of the following variables: English as a first language, gender, ethnicity, pre/post categories, score categories, threshold exemptions. One of our aims is to illustrate that cross-tabulations and bar charts are useful tools in making sense of large institute-wide LNAT data sets. These findings should also be viewed against the wider background summarised in a Summary of Results (Greyling (2015e) and its four sub-reports (Greyling, 2015 b, c, and d).