Post-primary reading fluency development: A latent change approach Marco van de Ven ⁎, Marinus Voeten, Esther G. Steenbeek-Planting, Ludo Verhoeven Behavioural Science Institute, Radboud University, The Netherlands abstract article info Article history: Received 23 June 2015 Received in revised form 22 December 2016 Accepted 20 February 2017 Available online xxxx In a longitudinal study, we investigated the post-primary reading fluency development of 1034 Dutch adoles- cents enrolled in four tracks of secondary education (high, intermediate, low, special education), using a multi- ple-group latent change approach. We assessed adolescents' word, pseudoword and text reading fluency, as well as their rapid naming abilities, at the beginning and after a half-year of education. The results showed that fluency development for real words, pseudowords and texts continues beyond the primary grades, and that the degree of improvement was similar across educational tracks. Word reading fluency and change in word reading fluency predicted text reading fluency development. No additional effect was found for pseudoword reading fluency. Finally, rapid naming ability was not only related to initial text reading fluency, but also to development of text reading fluency. Implications of these findings for assessment and intervention are discussed. © 2017 Elsevier Inc. All rights reserved. Keywords: Rapid naming Reading fluency Longitudinal Multiple-group latent change analysis Adolescents 1. Introduction Reading skill is an important determinant of success both within and beyond the school setting (O'Shaughnessy, Lane, Gresham, & Beebe- Frankenberger, 2002). Yet, many students in Dutch secondary education have insufficiently developed literacy skills (Hacquebord, Linthorst, Stellingwerf, & de Zeeuw, 2004). Dutch has a moderately transparent orthography (Seymour, Aro, & Erskine, 2003), and Dutch poor readers are mainly characterized by their slow reading speed rather than low ac- curacy (van der Leij & Van Daal, 1989). Hence, improving Dutch reading skills is mainly a matter of increasing reading fluency (Verhoeven & van Leeuwe, 2009). Oral reading fluency may serve as an indicator of basic reading competence (Fuchs, Fuchs, Hosp, & Jenkins, 2001), especially oral reading fluency of connected text. Fluency of reading words in iso- lation and fluency of reading connected text are related but different skills (Fuchs et al., 2001; Jenkins, Fuchs, van den Broek, Espin, & Deno, 2003; Kim, Wagner, & Foster, 2011). The present study addressed the development of Dutch oral reading fluency during the first half-year of secondary education (Grade 7). We operationalized reading fluency as the number of correct words per minute. The previous literature is mainly concerned with students in elementary education. For instance, Kim et al. (2011) studied oral and silent reading fluency of first-grade students in a longitudinal design, showing that at least for first-graders oral and silent reading fluency are distinct forms of reading fluency. The current study also involves a longitudinal design, in which we distinguished between four educational tracks in Dutch secondary schools. We included four measures of oral reading fluency, namely (1) Rapid Automatized Naming (henceforth RAN) of letters (this may be regarded as a basic measure of reading speed, as argued by Schatschneider, Carlson, Francis, Foorman, & Fletcher, 2002), (2) pseudoword reading fluency (a measure of basic decoding skill), (3) word reading fluency (primarily a measure of lexical access), and (4) text reading fluency. To disentangle the effects of word and pseudoword reading fluency within and across educational tracks, a multiple-group latent change approach was adopted. Like Kim et al. (2011) we used multiple indicators for reading fluency at the word level, but unfortu- nately for text reading fluency we had only one indicator available. Reading is a complex, multi-faceted skill that needs to be acquired (Byrne, 2008). Theoretical models of reading (e.g. Seidenberg & McClelland, 1989; Van Orden, Pennington, & Stone, 1990; Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001; Perfetti & Hart, 2001) suggest that both phonological decoding (i.e. mapping graphemes to phonemes) and semantics (mapping words to meanings) are required to become a competent reader. For instance, the Dual Route Cascaded model (Coltheart et al., 2001) claims that listeners use the direct as well as the indirect route during word reading. In the indirect (non-lexical) route, readers execute grapheme-to-phoneme routines to obtain words' phonological representations, while, in the direct (lexical) route, readers directly retrieve word forms from the mental lexicon. Word and pseudoword reading tasks can be combined to separate the non-lexical and lexical route during decoding. Pseudoword reading tasks, on the one hand, are expected to primarily tap into phonological decoding (i.e. pseudowords cannot be directly retrieved from the men- tal lexicon), whereas word reading (at least, in advanced learners) mainly taps into the lexical route. Another reason why we included word and pseudoword reading tasks in the present study was to Learning and Individual Differences 55 (2017) 1–12 ⁎ Corresponding author at: Radboud University Nijmegen, P.O. Box 9104, 6500 HD Nijmegen, The Netherlands. E-mail address: Marco.vandeVen@pwo.ru.nl (M. van de Ven). http://dx.doi.org/10.1016/j.lindif.2017.02.001 1041-6080/© 2017 Elsevier Inc. All rights reserved. Contents lists available at ScienceDirect Learning and Individual Differences journal homepage: www.elsevier.com/locate/lindif