Attitude items and low ability students: the need for a cautious approach to interpretation

The results of the Progress in International Reading Literacy Study (PIRLS 2001) were published in 2003. In addition to data about the reading achievements of 10-year-olds in 35 countries, PIRLS 2001 also collected questionnaire information from children, their teachers, headteachers and parents. The results showed not just how well students can perform in various reading tasks, but also the relationship between reading abilities and other characteristics, including the characteristics of their homes and schools, the students’ attitudes to reading, reading enjoyment, selfconfidence, engagement and perceptions of their reading abilities. Surprisingly enough, children in England were reported as having poor attitudes to reading, compared to children in many other countries, despite high achievement on the reading tests. This paper raises some concerns about the interpretation of results from the study, suggesting a more complex picture than that presented in the summary index published in 2003. Our secondary analysis of the attitude and achievement data from PIRLS shows that pupils with low ability levels have misunderstood the questions which tested their attitudes towards reading. The authors point out that caution is needed when making cross-country comparisons to avoid naïve approaches to interpretation. The results suggest that the precise construction of items concerning attitudes is crit- ical, especially when the instrument is being administered to young children and trial- ling and review of attitude instruments might be needed before their large-scale use. This study can give rise to further work to assess whether the outcomes found can be extended and justify the need for a review of the attitude instruments currently in use.