Supporting struggling students
Hingstman, Mariëtte
DOI:
10.33612/diss.169170090
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2021
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Hingstman, M. (2021). Supporting struggling students: Prevention and early intervention with Success for All. University of Groningen. https://doi.org/10.33612/diss.169170090
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
6
Introduction
In this chapter, we describe the overarching findings with regard to the support of struggling students in the early grades of primary education, based on the four studies of this dissertation. Previous research, predominantly conducted in the United States, has demonstrated that part of the struggles could be prevented or remediated by providing effective support at an early stage (Fletcher et al., 2018; Morgan et al., 2008; Reschly, 2010), but research on the effectiveness of interventions used in primary education in the Netherlands is scarce. Therefore, the main aim of this dissertation was to investigate whether and how the school reform program Success for All (SfA) could contribute to improved outcomes of struggling students in the Netherlands. The primary area of focus of this dissertation was reading development, because reading skills are crucial for success in school and life (O’Connor & Vadasy, 2011; Snow et al., 1998). In addition, we paid attention to behavioral issues that might hinder students’ reading development, as SfA aims to improve both reading achievement and student behavior.
In this dissertation, we made use of the Response to Intervention (RTI) framework to describe how SfA supports students at each tier. The Tier 1 program is targeted at all students (Grades K-5) and aims to prevent learning difficulties. In SfA, the Tier 1 program consists of daily 90-minute reading lessons in which explicit instruction and cooperative learning are fundamental. Student engagement is further stimulated by the use of appealing fiction and non-fiction books,
predictable routines, and several effective classroom management strategies, such as asking randomly chosen students to respond. Within the 90-minute lessons, different domains of reading/language instruction (e.g. fluency, comprehension, vocabulary, oral language) are integrated. Student progress is monitored through frequent assessments, and students are regrouped across grades during SfA lessons on the basis of their reading level. For students who are struggling, early intervention is provided in addition to the whole-class reading lessons. SfA
6
provides (computer-assisted) tutoring in pairs or small groups at Tier 2 and one-to-one tutoring at Tier 3. During tutoring sessions, students receive additional instruction and practice their reading skills, usually under the guidance of a paraprofessional. In most cases, tutoring is provided to first graders and aimed at improving phonemic awareness, letter knowledge, word reading, and fluency skills, though tutoring could also be provided to students in higher grades.
Although the core of SfA is engaging reading instruction that responds to students’ needs, SfA is more than just a reading curriculum. At first, by the frequent use of cooperative learning, the program aims to improve social-emotional skills alongside academic achievement (Veldman, Hingstman, Doolaard, Snijders, & Bosker, 2020). Furthermore, SfA contains components that focus on issues that might affect learning, such as parental involvement and attendance, and SfA places emphasis on continuous school improvement. Schools receive training and coaching from someone who is affiliated with the SfA foundation, and every school appoints a SfA facilitator who supports teachers in their implementation of SfA. Teachers also receive manuals in which the SfA lessons are described in detail.
SfA has its origins in the United States, where the program has been implemented for over 30 years and has repeatedly demonstrated significant positive effects on literacy outcomes (Borman & Hewes, 2002; Borman et al., 2007; Quint et al., 2015; Slavin et al., 2011), and the program is also being implemented in schools in England for over 20 years (Education Endowment Foundation, 2017a). Recently, SfA has expanded to the Netherlands. The Dutch version of SfA was developed through a Research & Development (R&D) project that started in 2015. This project was initiated because the reading performances and motivation of Dutch students have declined, and inequality between students from different socioeconomic backgrounds has increased in recent years (Inspectie van het Onderwijs, 2016, Mullis et al., 2017; OECD, 2019a). Although the developers of the Dutch version of SfA strived to adhere to the original program as much as possible, there were some differences between the two versions that are important
to bear in mind. At Tier 1, the SfA lessons were initially (school year 2015-2016) only provided to students in Grade 1, as the materials for all grades could not be developed at once. In every subsequent school year, a grade was added. For this reason, students in the Netherlands were not regrouped across grades during SfA lessons at the time of study. Regarding tutoring at Tier 2, all sessions in Dutch schools were provided face-to-face, because a computer-assisted tutoring program was not developed yet. Furthermore, the SfA components that focus on issues that might affect learning were only present to a small extent in the Dutch version of SfA, and the SfA facilitator was only appointed for four hours a week, instead of full-time. For a detailed description of the Dutch version of SfA, we refer to Mullender-Wijnsma et al. (2020).
In this dissertation, a variety of quantitative and qualitative research methods was used to address the topic of effective support of struggling students with SfA in a comprehensive way. After summarizing and integrating the main findings from the four studies and outlining the limitations of the research project, we describe implications for future research and practice.
Summary of main findings
In Table 6.1, we present a new version of Table 1.1 (presented in Chapter 1), highlighting the main outcomes of the four studies.
6
Table 6.1
Overview and highlights of the studies included in this dissertation
Chapt er Title Resear ch question(s) Highlights 2
Preventing special education assignment for students with learning or behavioral difficulties: a review of interventions What is known about the effectiveness of interventions that potentially reduce the number of students assigned to special education?
•
Reading R
ecovery and PBIS
plus
were found to significantly
reduce the number of special education assignments
•
Fast T
rack and Success for All demonstrated significant effects
on some, but not all measures of special education assignments
•
Start Making A R
eader T
oday, F
amilies and Schools T
ogether
and School-Based Intervention T
eams seem to have potential to
reduce special education assignments
•
Structure of Intellect and Individualizing S
tudent Instruction
did not seem to be effective in reducing special education assignments
•
Effective interventions for students with reading difficulties had a strong emphasis on tutoring
3
Supporting young struggling readers at Success for All schools in the United S
tates and the
Netherlands: comparative case studies
1. What are the similarities regarding the support of struggling readers between two U.S. schools and two Dutch schools that implemented Success for All? 2. What are the differences regarding the support of struggling readers between two U.S. schools and two Dutch schools that implemented Success for All, and which factors might explain these differences?
•
Interviewees in both countries agreed on the strengths of SfA for struggling readers: the integrated approach to reading/ language, the books, the cooperative learning activities, and tutoring
•
At all four schools, we observed a tension between the scriptedness of SfA and the need for flexibility in the support of struggling readers
•
The schools in the U.S. had a higher amount of specialized teachers and paraprofessionals, which made it easier to implement SfA
•
Relentlessness and inclusiveness seemed to be more embedded in the U.S. culture, which are important values for the support of struggling readers
•
Chapt er Title Resear ch question(s) Highlights 4
The effects of Success for All in the Netherlands on the reading achievement of first-grade students at risk of reading problems 1. What are the effects of Success for All on the reading achievement of students at risk of reading problems? 2. Is there a relationship between the intensity of the Success for All tutoring intervention and students’ reading achievement?
•
A positive significant effect of SfA on reading comprehension of students at risk of reading problems was found for the first cohort
•
The effects of SfA on the other outcomes were mostly small and positive, though not statistically significant
•
W
e observed a significant negative association of tutoring
intensity with word and text reading skills in the second cohort
•
Differences were observed in the implementation of both the Tier 1 classroom intervention and the Tier 2/3 tutoring intervention
5
Behavioral engagement of students with attention difficulties in Success for All lessons To what extent does Success for All lead to an increase in behavioral engagement of students with attention difficulties?
•
No significant main effect of SfA on behavioral engagement of students with attention difficulties was found in both substudies
•
6
In Chapter 2, the outcomes of a systematic literature review were presented. In the past decades, the number of students assigned to special education has increased, in spite of efforts for more inclusion. The aim of the study was to describe what is known about the effectiveness of interventions that potentially reduce the number of students assigned to special education. Similar to the other studies in this dissertation, we focused on interventions that were carried out in elementary schools for general education. Interventions were targeted at students with mild learning and/or behavioral difficulties. 12 studies that were published between 1999 and 2015 were included, in which the effects of nine different interventions were evaluated by using experimental or quasi-experimental research designs. Although studies eligible for inclusion could have been conducted in any OECD country, all included studies were conducted in the U.S. This is important to note, as most students with special educational needs in the U.S. still attend schools for general education, while similar students in the Netherlands are more often educated in segregated settings. Of the nine interventions, that were mainly multi-tiered in nature, four aimed to improve both academic achievement and behavior, three targeted only academic
achievement, and two targeted only behavior. The interventions differed in terms of effectiveness. The intensive Tier 3 tutoring intervention Reading Recovery and the multi-tiered behavioral intervention PBISplus were found to significantly reduce the number of special education referrals and/or placements, though both interventions were only evaluated in one study. Fast Track, a multi-tiered intervention targeted at students at high risk for long-term antisocial behavior, and Success for All, being a multi-tiered school reform program, were evaluated in multiple studies and at different points in time, and demonstrated significant effects on some, but not all measures. The Tier 3 tutoring intervention Start Making A Reader Today (SMART), the Tier 2 intervention for students with emotional and/or behavioral problems Families and Schools Together (FAST), and the Tier 3 consultative process School-Based Intervention Team (SBIT) seemed to have
potential to reduce special education assignments, though the reported effects were not statistically significant. The multi-tiered interventions for differentiated instruction Structure of Intellect (SOI) and Individualizing Student Instruction (ISI) did not seem to have an impact on special education assignments. With regard to intervention characteristics, we noticed that effective programs for students with reading difficulties had a strong emphasis on tutoring. Other patterns regarding intervention characteristics (e.g. a focus on academic achievement, behavior, or both) and effectiveness of the intervention were not observed. In sum, the review indicated that preventing special education assignment is possible for part of the students with mild learning and/or behavioral problems, if they receive effective support at an early stage.
In Chapter 3, we described the outcomes of a multiple case study, in which we investigated similarities and differences regarding the support of struggling readers between two U.S. schools and two Dutch schools that implemented SfA. The aim of this study was to obtain more insight into the topic of expanding an American program for struggling readers to a new context. In both countries, first- and second-grade teachers and other staff members were interviewed about the support of struggling readers in their school in general and about their experiences with implementing SfA. Additionally, the participating teachers were asked to select two struggling readers from their homeroom classes. To increase comparability, we asked the teachers to select students who were performing approximately six months below age-appropriate level on reading. During the interviews, we discussed how these students were supported, which provided us with 14 examples of the daily practices at these schools. Data was collected in school year 2017-2018 (the Netherlands) and school year 2018-2019 (U.S.). We concluded that American and Dutch interviewees agreed on a number of strengths of SfA for teaching struggling readers: the integrated approach to reading/language, the books, the cooperative learning activities, and the possibility to provide tutoring. Furthermore, in both countries we observed a tension between the scriptedness of the SfA
6
program and the need for flexibility in the support of struggling readers. To deal with this issue, schools made some adaptations to the SfA lessons and provided additional instruction and/or programs alongside SfA. With regard to differences between the two U.S. and the two Dutch schools, we observed that the number of specialized teachers and paraprofessionals was far higher in the U.S. schools. This can be explained by multiple factors: schools in the U.S. receive a (slightly) higher budget per student, include more students with special educational needs, and serve more students from disadvantaged backgrounds for whom they can receive additional funding. We also noticed that the educational cultures in the Netherlands and the U.S. seemed to differ in terms of relentlessness and inclusiveness, which might have been a hindering factor in the implementation of SfA in the Netherlands. In the context of teaching struggling readers, relentlessness – a core value of SfA – means not giving up until success is achieved (Slavin et al., 2009). We observed that Dutch teachers seemed to be less optimistic than U.S. teachers about the opportunities of students who fall behind. We supposed this could be related to the less inclusive system in the Netherlands, where relatively many students with special educational needs are educated in special schools (European Agency for Special Needs and Inclusive Education, 2020; NESSE, 2012). On average, Dutch teachers who work at schools for general education have negative to neutral attitudes towards teaching students with special educational needs, for example because they experience a high workload, or because they consider the possibilities for support at their school to be insufficient(Pijl, 2010; Smeets et al., 2019). Our case studies demonstrated that SfA can be of great help in setting up an effective multi-tiered support system in schools in the U.S. as well as in the Netherlands, though some adaptations to the SfA program (e.g. increasing the time spent on phonics) might be needed to ensure a good fit with the local context, the needs of the school and/or the needs of individual students.
Chapter 4 focused on the effects of SfA on the reading achievement of students at risk of reading problems in the Netherlands. Because SfA appeared
to be especially effective for initially low-achieving students in the U.S. (e.g. Quint et al., 2015), we wanted to investigate whether similar effects could be observed in the Dutch context. Students whose scores fell into category IV or V (the lowest 40%) at one or more literacy pretests, measured at the end of kindergarten, were included in the quasi-experimental study. The achievement of students from six SfA schools was compared with the achievement of students from four control schools. In total, 299 students from two different cohorts (2016-2017 and 2017-2018) were involved. The first cohort sample consisted of 148 students at risk of reading problems (89 SfA, 59 control) from 19 classes (9 SfA, 10 control). The second cohort sample consisted of 151 students at risk of reading problems (75 SfA, 76 control) from 19 classes (9 SfA, 10 control). We evaluated the effects of SfA on five reading subskills: phonological awareness, letter knowledge, reading comprehension, word reading skills and text reading skills, which were assessed at the end of Grade 1. Multilevel models were estimated for each outcome measure. We found a positive significant effect of SfA on reading comprehension in the first cohort. The effects of SfA on the other outcomes were mostly small and positive, though not statistically significant. These findings should be considered preliminary, as the study concerned the first years of implementation of SfA in the Netherlands. Within the group of students at risk in the SfA condition, we also examined whether there was a relationship between the intensity of the SfA tutoring intervention (which was provided to more than half of the students at risk) and reading achievement. In the second cohort, we observed a significant negative association of tutoring intensity with word and text reading skills. We want to emphasize that this finding does not mean that the SfA tutoring intervention is thus not effective. Interviews with tutors, school records and observations of tutoring sessions demonstrated some differences between schools and between tutors in the implementation of the tutoring program, for example with regard to setting and evaluating goals and engaging students. Furthermore, we suppose that it might have been difficult to detect a positive effect of tutoring intensity, as the most intensive types of
6
intervention are often provided to the most struggling students. All in all, we consider SfA to be a promising intervention for Dutch students at risk of reading problems, though more research is needed to better understand of the relationship between tutoring intensity and reading achievement.
In Chapter 5, we investigated whether SfA led to an increase in behavioral engagement of students with attention difficulties in the Netherlands, as high levels of engagement are associated with better academic outcomes. SfA includes several strategies that are expected to promote engagement, such as reading and discussing a variety of texts, frequent use of cooperative activities, and predictable routines. Two complementary substudies were conducted to answer the research question. The first substudy was carried out in school year 2016-2017. In this substudy, the behaviors of 40 first-grade students from 10 SfA classes were compared with the behaviors of 32 first-grade students from 8 control classes, who were all observed during reading lessons (between-subjects design). Similar to the study in Chapter 4, six SfA schools and four control schools were involved. The second substudy was carried out in school year 2017-2018 and only concerned students at the six SfA schools. In this substudy, 38 first- and second-grade students from 10 classes were compared with themselves by observing them during a SfA reading lesson and a non-SfA mathematics lesson (within-subjects design). We used a time-sampling observation instrument to measure on-task behavior of students with relative attention difficulties. Per classroom, the four students with the lowest scores on the subscale Task-performing of the Social Competence Observation List (SCOL) questionnaire were selected for participation in the study. Multilevel binomial logistic regression models were used to
investigate whether students’ behavioral engagement differed between SfA lessons and non-SfA lessons. As our data consisted of repeated measures within a lesson, we were able to model changes in on-task behavior during the lesson. In the first substudy, students in the SfA condition were on-task for 74% of time, versus 68.1% in the control condition, though no significant main effect of SfA or interaction
effect between SfA and measurement moment was found. In the second substudy, we again did not find a significant main effect of SfA. Students were on-task for 67.1% of time during SfA (reading), and for 67.2% of time during mathematics. Nevertheless, we did observe that patterns of behavioral engagement during SfA lessons were significantly different from patterns in mathematics lessons. While students seemed to peak in their on-task behavior in the middle of a mathematics lesson, during SfA they showed more on-task behavior at the start and at the end of the lesson. Altogether, we could not conclude that SfA led to an increase in behavioral engagement of students with attention difficulties. However, our study suggested that some lesson components or instructional strategies might lead to higher levels of behavioral engagement.
Integrative findings
In this dissertation, we aimed to investigate whether SfA could improve the outcomes of struggling students, and in particular, of struggling students in the Netherlands. In advance, we hypothesized that SfA would be beneficial for struggling students because of the combination of evidence-based strategies, such as explicit instruction and cooperative learning, at Tier 1 (the classroom program), and tutoring at Tier 2 (small-group) and Tier 3 (one-to-one). During interviews with Dutch and American teachers, it became apparent that the integrated approach to reading/language, the books, the cooperative learning activities, and tutoring were valued as effective features of SfA for struggling students. However, teachers also mentioned that they sometimes found it difficult to balance between the scriptedness of SfA and the need for flexibility in the support of struggling students. With regard to effects of SfA in the Netherlands, we found that SfA led to a significant improvement in reading comprehension skills of first grade students in one cohort. This is a very encouraging finding, given that understanding the meaning of text is crucial for learning in a broad sense (Snow et al., 1998). We did not find significant effects of SfA on the other academic and behavioral outcomes
6
we measured, contrary to our hypotheses. It is important to keep in mind that all studies were conducted during the first years of implementation of SfA in the Netherlands.
Limitations
As outlined earlier, this dissertation was embedded in an R&D project. Although simultaneously developing and studying a program has several advantages (such as being able to quickly respond to problems), we also want to highlight some limitations of this approach. At first, the Dutch version of SfA was less comprehensive than the American version at the time of study, because not all materials could be translated at once. The step-wise development led to additions and changes to the program each year. Secondly, implementation issues have likely impacted the outcomes of the effect studies. Teachers need time and guidance to get used to a new program, and numerous factors within a school might hinder implementation of the program as designed (Desimone, 2002; Durlak & DuPre, 2008). In the overall evaluation of the SfA program (Mullender-Wijnsma et al., 2020), it was concluded that there was room for improvement in the implementation of the reading lessons, which was probably related to the insufficient amount of coaching teachers received. In our observational study (Chapter 5), we noticed that the SfA lessons were often shorter than the prescribed 90 minutes, and that the time spent in pair and group settings during SfA lessons was relatively low, despite the emphasis of SfA on cooperative learning. Regarding tutoring, we observed that schools experienced organizational challenges, and we noticed that the quality of the sessions varied between tutors and schools (Chapter 3 and 4). Because of these issues regarding development and implementation, we emphasize that the conclusions in this dissertation only concern the effectiveness of the Dutch SfA program as implemented in school years 2016-2017 and 2017-2018, and should not be considered final outcomes. We suppose that larger impacts of SfA on the outcomes of struggling students could have been expected
with a fully developed program that is implemented as designed (Durlak & DuPre, 2008). However, maximum fidelity might not be possible and not desirable, in particular when teaching struggling students, for whom some adaptations to a standard program might actually enhance learning outcomes (Harn et al., 2013). In Chapter 3, we described which adaptations to the SfA program were made at two U.S. schools and two Dutch schools that implemented SfA. Although in most cases deliberate choices were made in order to improve student learning, some schools deviated from the program because of other (practical) reasons, and used additional interventions that were not necessarily evidence-based. We observed that solid partnerships between SfA coaches and school staff can help to achieve a good balance between fidelity and flexibility.
A limitation that concerns the effect studies (Chapter 4 and Chapter 5) is that we had little information about the daily practices at the control schools. Hence, it is difficult to determine how much the classroom instruction and
additional interventions at control schools differed from those in the SfA condition. The contrast between conditions was probably less prominent than in previous studies conducted in the U.S. context. However, the interviews in Chapter 3 indicated that Dutch teachers did saw clear differences between SfA and other programs that are commonly used in Dutch schools, for instance regarding the amount of cooperative activities.
Furthermore, three studies have in common that the samples were rather small. For the qualitative study in Chapter 3, this was a conscious choice, because we wanted to provide the reader with a detailed picture of how struggling readers were supported at these SfA schools, not pretending to draw generalizable conclusions. For Chapter 2 and 5, the small sample size is more clearly a
limitation. We found out that there were little high-quality studies on the effects of interventions on special education assignment, resulting in only 12 included studies in the review presented in Chapter 2. In Chapter 5, we focused on the subgroup of students with attention difficulties, which resulted in a small sample in
6
which significant effects are harder to detect.
During our research project, determining who is a struggling student and therefore eligible for participation was something we struggled with ourselves. There are many students, especially in the early years of primary education, who experience difficulties without being identified with a learning or behavioral disorder. According to Dutch teachers, about a quarter of students in their classes have special educational needs (Van der Veen et al., 2010). From the RTI perspective, these students should receive additional intervention at an early stage, and the response to this additional intervention will reveal whether the student’s problem is temporary or persistent. Although one of the aims of RTI is to decrease the number of students with identified disorders (Kirby, 2017; Slavin, 2018), who often experience negative consequences of their label such as lowered teacher expectations (e.g. Pit-ten Cate & Glock, 2018), some form of identification/labeling is inevitable in order to determine who is at risk and should receive additional intervention. It is common to select at risk students early on the basis of their test scores, which we did in Chapter 4 as well. According to Frans (2019), this is debatable, because young students’ scores are often unstable, and because test norms are often relative and thus subject to inflation (i.e., part of students will always fall within the ‘red zone’). Furthermore, test scores only tell part of the story. The same holds for scores on student behavior questionnaires (Konold & Pianta, 2007), which we used to select students with attention difficulties in Chapter 5. Ideally, teachers and researchers should make use of different instruments and observations to determine who is at risk. Thereafter, the at risk status of a student should not lead to lowered expectations, as the difficulty in learning may be due to insufficient instruction, instead of due to a deficit of the student (Kirby, 2017).
Directions for future research
In this dissertation, we have used various methods to gain more insight into the topic of support of struggling students, though several questions have remained
unanswered. Firstly, we believe future studies could pay more attention to the interplay between learning and behavior. Quite often, problems in learning and behavior co-occur (Reschly, 2010; Van der Veen et al., 2010). While many large-scale experimental studies have evaluated the effects of SfA on reading achievement (e.g. Borman & Hewes, 2002; Borman et al., 2007; Quint et al., 2015; Slavin et al., 2011), to our knowledge the relationship between SfA and student behavior is only explored in a few smaller studies (Datnow & Castellano, 2000; Hopkins et al., 1999). Because of the strong relationship between behavior and academic achievement (Fredricks et al., 2004), we recommend to include behavioral measures in evaluation studies more often. Furthermore, the outcomes of our study on the relationship between SfA and behavioral engagement (Chapter 5) suggested that there might be specific lesson components or instructional strategies that lead to more engagement, which requires further research.
Secondly, we think students’ voices should be more often heard in evaluation studies, to obtain more insight in their needs and perceptions
(Lundqvist, 2014). Also in this dissertation, effects on students were only measured indirectly: through tests, observations, and teacher interviews. Looking back, inviting struggling students to share their experiences with SfA would have been a valuable addition. In particular for the target group of this dissertation (struggling students), it would have been interesting to investigate whether they actually enjoy learning and experience feelings of success, which SfA aims to achieve for all students. However, we had to keep in mind that during the R&D project a lot of data was already being collected by multiple researchers. For future studies, we recommend to pay more attention to the student perspective throughout the research project.
Thirdly, this dissertation demonstrated that the local context impacts the feasibility of implementing an intervention like SfA. This is important to note, because our knowledge about evidence-based education predominantly stems
6
from research conducted in schools in the U.S. Although many evidence-based practices are transferable to the Dutch context, there are notable differences between the U.S. and the Dutch educational systems, for instance regarding the amount of support staff members at schools for general education. Therefore, it is important to continue the development of a Dutch version of SfA that fits well within the local context, and to keep evaluating the effects of the program on students’ development. The implementation issues described in this dissertation and by Mullender-Wijnsma et al. (2020) showed that there is still room for improvement.
In a broader sense, we think it is problematic that commonly used programs in the Netherlands for reading instruction in the early grades are not evaluated through (quasi-)experimental studies. As the reading performances of Dutch students are worrisome (Inspectie van het Onderwijs, 2018; Mullis et al., 2017), it is important that researchers help schools to determine which interventions are (not) effective in teaching children how to read, for instance by developing a database. An American example of such a database with
educational interventions is Evidence for ESSA (www.evidenceforessa.org). A similar Dutch database with youth care interventions is developed by het Nederlands Jeugdinstituut (www.nji.nl/nl/Databank/Databank-Effectieve-Jeugdinterventies).
Implications for practice
We would like to end this chapter with some implications for educational practice. Although we observed fewer effects of SfA in the Netherlands than expected, we believe our findings can still contribute to more effective support of struggling students in the Netherlands.
From our studies as well as from previous studies (e.g. Klingner et al., 2006; Quint et al., 2015), we know that implementing such a comprehensive
program as SfA is challenging for schools. The scripted nature of the program can be considered both a strength and a weakness. On the one hand, the detailed manuals set a clear standard, while on the other hand, some students may benefit more from a tailored approach. In line with Durlak and Dupre (2008) and Harn et al. (2013), we conclude that some flexibility in the implementation of SfA can improve the effectiveness of the program for struggling students, under the condition that the core values of SfA are broadly supported within the school. We will attempt to present these core values, which are based on decades of research, in such a way that they are also applicable to schools not working with SfA.
Improving reading performance begins with spending sufficient time on reading. This may sound obvious, but the durations of the reading lessons we observed (Chapter 5) indicate that schools often do not meet the guideline of 8 to 10 hours a week for Tier 1 instruction (Gijsel et al., 2011). Furthermore, it matters what happens within these reading lessons. Fortunately, there is consensus in the literature about the foundations of literacy instruction. “The most essential
thing, when it comes to learning to read, is to read a lot” (Brokamp, Houtveen &
Van de Grift, 2019, p. 1). For beginning readers, explicit instruction is important to become familiar with the alphabetic principle. To practice fluency and reading comprehension, students should be exposed to a variety of (fiction and non-fiction) texts throughout their school career. Moreover, teachers should help students to build up background knowledge and vocabulary, which is particularly important for students from disadvantaged backgrounds (Education Endowment Foundation, 2017b; Gersten et al., 2020; National Early Literacy Panel, 2008; National Reading Panel, 2000; Snow et al., 1998). Other studies have shown that engagement during reading lessons can be promoted through the use of effective classroom strategies, such as predictable routines and randomization techniques, and cooperative learning activities, such as Think-Pair-Share (Greenwood et al., 2002; Dietrichson et al., 2017). By providing effective instruction at Tier 1, the number of struggling readers will probably be decreased (Reschly, 2014).
6
1 instruction, it is important to keep in mind that most of them first and foremost need more time and instruction to become fluent readers (Snow et al., 1998). Therefore, tutoring (Tier 2/3) should always be provided in addition to Tier 1 instruction. Another recommendation from studies about RTI models is to assess students frequently and to intervene at an early stage, in order to prevent reading problems from becoming worse (Hagans & Good, 2013; Reschly, 2010). In our review study, we noticed that effective (tutoring) interventions can even prevent students from being assigned to special education. The effectiveness of tutoring interventions can be increased by setting and evaluating goals, aligning activities with the Tier 1 program, teaching skills explicitly, providing feedback, and ensuring that students experience feelings of success (Gersten et al., 2020; Gijsel et al., 2011; Neitzel et al., 2020). In our studies, we observed that tutoring was valued as an effective tool to support struggling readers at schools in the Netherlands (Chapter 3 and 4), but sustainable implementation was often hindered by organizational challenges. Although in general teachers are found to be the most effective providers of tutoring (Slavin et al., 2011), most schools need to look for less expensive solutions. Fortunately, recent studies have yielded promising results of tutoring provided by volunteers who are supported by a professional at school and/or a computer-assisted tutoring program (Neitzel et al., 2020; Zijlstra, 2015).
Lastly, we hope this dissertation encourages teachers and others involved with student development to reflect on how struggling students are viewed and treated. A deficit perspective presumes that a learning difficulty is mainly something internal and stable, while from a support perspective (reflected in the RTI approach and in the SfA program), a learning difficulty is presumed to be responsive to the instructional environment. Although the latter perspective becomes more apparent in Dutch education (Schölvinck & Jansen, 2014; Ledoux & Waslander, 2020), educational reform is a long-haul process, as the studies in this dissertation have shown. We believe the SfA program can inspire Dutch educators to establish a support system in which evidence-based prevention and flexible
