Early identification or broken promises? A comparison of literacy assessment performance in 112 Swedish third graders

Abstract

The national standardised literacy assessment (NSLA) for Swedish Language was introduced in 2009 as a grade-three compulsory assessment and includes the assessment of reading ability. It was introduced as a measure of relatively early identification of reading difficulties among nine-years old students. The primary objective of this study was to examine whether the NSLA is able to identify students with word decoding problems from a sample of third graders (N = 112; n = 57 girls; n = 55 boys) attending six schools in a Swedish municipality. Eleven students (10%) performed below the cut-off value for word decoding ability in students of this age group. Only three of these students were identified as not achieving the goals posed by the NSLA. In contrast, eight students with low word decoding ability managed to meet the NSLA requirements. Gender differences were observed, since all of the students who passed the NSLA, notwithstanding having low performance in WD, were boys. Although the NSLA was specifically introduced at this level to identify weakness in reading at an early stage, the results of this study indicate that approximately three quarters of students with poor word decoding ability may remain unidentified by the NSLA. These findings call into question the validity of the NSLA in recognising pupils in need of additional support in reading.

Keywords:

• national standardised literacy assessment
• word decoding
• early identification
• gender differences
• literacy tests

View correction statement:

Corrigendum