Overview

The Promoting Science among English Language Learners (P-SELL) Scale-Up study investigated the effectiveness of a stand-alone, year-long, fifth grade science curricular and professional development intervention implemented large-scale over 3 years. P-SELL is aimed at improving the science achievement of all students with a focus on ELLs in the context of high-stakes assessment and accountability policy in science.

P-SELL is organized around three key features: state science standards, science inquiry, and language development for all students including ELLs. The P-SELL implementation involves three components. First, teachers receive science curriculum materials, including consumable student books, a teacher’s guide, science supplies, and online resources and supplements. Second, teachers engage in professional development workshops during the summer and throughout the school year. Third, teachers are provided support at school sites by a P-SELL District Coordinator in each district.

To investigate the effectiveness of P-SELL, a randomized controlled trial was conducted. The study involved 33 treatment schools implementing P-SELL and 33 control schools across three demographically diverse school districts in the state of Florida. All fifth grade science teachers and students in these schools participated in the study over 3 years, with a total of approximately 270 teachers and 7000 students participating each year.

Main Findings

On a researcher-developed assessment, the impact of P-SELL was both statistically significant and practically important across Years 1 through 3. On the state science assessment, the impact of P-SELL was both statistically significant and practically important in Years 1 and 3. The fact that a significant impact of practical importance was found not only on a researcher-developed assessment but also on the state science assessment (on 2 out of 3 years) is noteworthy. The only other large-scale randomized controlled trial of an elementary science intervention that found a positive impact on students’ science achievement was Maerten-Rivera et al. (2016), who implemented an earlier version of P-SELL in one large urban school district. Other large-scale randomized controlled trials of elementary science interventions found no impact on students’ science achievement.

Even though overall P-SELL had a positive impact on student achievement, over the three-year period, it was more consistently effective for non-ELLs, followed by former ELLs, recently reclassified ELLs, and lastly ELLs. On the researcher-developed assessment, P-SELL had a positive impact for non-ELLs and former ELLs in all 3 years of the study, for recently-reclassified ELLs in Years 1 and 3, and for ELLs only in Year 1. On the state science assessment, P-SELL had a positive impact for non-ELLs in all 3 years and for former ELLs in Year, 1 but it did not have an impact for recently-reclassified ELLs and ELLs.

The fact that the impact of P-SELL for former ELLs was similar to the impact for non-ELLs on the researcher-developed assessment was encouraging. This finding suggests that students who were once ELLs and were reclassified as English proficient before third grade (i.e., had exited ELL programs more than 2 years ago) benefitted from P-SELL similarly to those students who were never ELLs. On the other hand, the results for ELLs were less encouraging. The positive impact on the researcher-developed test for ELLs in Year 1 was not replicated in Years 2 and 3. This suggests that more work needs to be done to identify approaches to effectively support ELLs in the science classroom, especially when an intervention is implemented large-scale with all students. One promising venue are the instructional shifts promoted by the Next Generation Science Standards, as science and engineering practices are language-intensive (e.g., argue from evidence, construct explanations) and thus provide language learning opportunities to all students (Lee, Quinn, & Valdés, 2013).

P-SELL had positive impacts on teachers’ science knowledge in Year 1 and teachers’ instructional practices in all 3 years. In Years 1 and 3, compared to control group teachers, P-SELL teachers reported using more teaching for understanding practices, teaching for inquiry practices, language development strategies, and home language use strategies. In Year 2, P-SELL teachers reported using more teaching for inquiry practices.

P-SELL had a positive impact on teachers’ perceptions of various school resources (material and social) over the 3 years of implementation. In terms of material resources, compared to control group teachers, P-SELL teachers found time to be less of a barrier to their science teaching in all 3 years, and materials to be less of a barrier to their science teaching in Years 1 and 2. In terms of social resources, P-SELL teachers perceived parents and family as less of a barrier to their science teaching in Year 1 and reported more collaboration with other science teachers for science practices in Year 3.

In summary, the P-SELL study is one of very few rigorous experimental evaluations of an elementary science intervention implemented large-scale to find overall positive impacts on student achievement, teacher instructional practices, and teachers’ perceptions of school resources.

 

 

Last Modified: 10/13/2017
Modified by: Okhee Lee