Evaluation of a Large-Scale Bottom-Up-Developed Reading Intervention: Compensatory Effects for Nonnative Elementary School Students?

• In Event: 25.072 - Systematic Reading Interventions in Schools and Families: Effectiveness for Disadvantaged Groups

Abstract

Given the importance attributed to promoting reading competence of all students, the objective of the study was to examine the impact of a nation-wide bottom-up reading intervention in Germany. Students with non-native language background have been identified to be in strong need of promotion of reading skills. Therefore, the study compares effects for native and non-native speakers.
Promoting reading comprehension is a major goal of education (Connor et al., 2014). Studies repeatedly pointed to a considerable number of students who systematically lag behind their peers in reading skills already in the elementary school grades (NCES, 2007). In many countries non-native speakers of the language at school are overrepresented in this group. On the one hand an increasing gap between good and struggling readers can be assumed (Matthew effect, compare Stanovich, 1986), but empirical evidence pointed to rather week effects (Pfost et al, 2014). In the context of interventions compensatory trajectories of reading development of non-native speakers seem to be possible. The study examined main and differential effects of a nation-wide intervention program (Tier I approach) focusing on reading fluency and reading competence.
1,032 students were involved in the first and second measurement point of a quasi-experimental study at the beginning and the end of fourth grade (Intervention group [IG]: 656; Mage-t1= 9.72 [SDage= 0.50], 43.6 % girls; 40.8 % non-native; Control group [CG]: 376; Mage-t1= 9.66 [SDage= 0.48], 42.6 % girls; 32.2 % non-native). IG and CG did not differ systematically by percentage of female and social background, but in age and percentage of non-native speakers. The intervention was based on bottom-up (school) developed approaches to promote reading competence. IG-schools worked together in networks. CG-schools were not involved in the large-scale intervention program and reported not to follow a structured reading concept. Reading comprehension and fluency were measured with established measures. Ten plausible values were used for analyses with good reliabilities (α > .79). To answer the first research question the effect of the intervention on reading competence at the end of grade 4 was tested with structural equation model (SEM) controlling for reading competence at the beginning of the school year, gender, language background, social status, SEN status, and age. With respect to research question 2 an interaction between being non-native and the intervention was examined.
SEMs were evaluated with respect to RMSEA and CFI and showed appropriate fit. No main effect of the intervention was found. Being non-native had a significant negative effect on reading comprehension at the beginning of grade 4. Contrary to or expectations no interaction effect was identified between non-native students and the intervention with respect to reading fluency. However reading fluency was interacted negatively with the intervention, even when previous reading achievement was controlled for. Non-native students profited less from the intervention compared to native speakers.
Findings suggested that a bottom-up reading intervention, which was not tailored to the specific needs of non-native students, did not lead to compensatory development in elementary grade 4. We rather found support for the Matthew effect in reading.

Authors

• Franziska Schwabe, Technical University of Dortmund

• Theresa Schlitter, Technical University Dortmund

• Jennifer Igler, IFS - TU Dortmund

• Annika Ohle, TU Dortmund

• Annika Teerling, Leibniz Institute for Science and Mathematics Education (IPN)

• Olaf Koeller, Leibniz Institute for Science and Math Education

• Nele McElvany, TU Dortmund University