Abstract
Many reading difficulties may be prevented if detected and effectively addressed early (Greenwood & Carta, 2011). In New Zealand, recent policy documents specify instructional targets after one year of school (Ministry of Education, 2009). Therefore, assessment tools are required for this developmental period that are effective and efficient.
Some aspects of beginning reading, such as learning the alphabet, may be constrained, no longer growing once they are achieved due to ceiling effects (Paris, 2005). However, because reading skills are still developing, oral passage reading tasks may not be sensitive to individual differences and change in performance until approximately the middle of Year 2 due to floor effects. A measure is required that adequately indexes children’s developing reading proficiency during this intermediate period.
Research suggests that context-free word recognition speed contributes to syntactic processing in reading (Klauda & Guthrie, 2008) and differs between skilled and unskilled readers (Jenkins, Fuchs, Van Den Broek, Espin & Deno, 2003). Fuchs, Fuchs and Compton (2004) found Word Identification Fluency (WIF) to be a good indicator of overall reading proficiency among Year 2 aged children in the US. Due to linguistic (Robb, Maclagan, & Chen, 2004; Robb & Gillon, 2007) and curricular (Smith & Elley, 1997) differences between NZ and America, small-scale field trials are recommended before introducing measures to the New Zealand educational context (Croft, Stafford & Mapa, 2000).
This thesis aimed to create and evaluate a New Zealand adaptation of the WIF task (NZWIF). Following the approach taken by Fuchs et al. (2004), we developed New Zealand Word Identification Fluency (NZWIF) for the New Zealand (NZ) educational context, informed by research on the development of word usage by children in NZ and consideration of words NZ children encounter in beginning reading texts.
Participants were 73 children from 3 Dunedin schools who were Year 2 in 2009 or 2010. NZWIF demonstrated alternate format reliability (r = .96 - .98) and robust linkages across time (r = .90 - .94). Robust concurrent and predictive correlations were found between NZWIF and age-appropriate phonemic awareness, decoding and passage reading measures, which have already demonstrated validity in NZ (Schaughency & Suggate, 2008) and a passage reading task of highly decodable text which has not yet been validated for use in NZ. The measure also differentiated between children who were and were not meeting National Standards criteria. The NZWIF measure showed strong concurrent and predictive relations with book level and differed for students who were and were not of concern at the end of Year 2. Those who were not of concern at the end of Year 1 performed significantly better on the NZWIF measure the following year than those for whom concern was indicated. Evaluation of growth suggests greater rates of progress in the second half of the year, particularly for students considered “at risk”. Findings suggest NZWIF may be an efficient and effective indicator of beginning reading proficiency with developing readers and support future research examining NZWIF as tool for capturing developmental trajectories and informing educational decision-making.