Abstract
Two hundred fifty-four intermediate-grade teachers (88% female) in urban public schools in Chile indicated how frequently they made writing instructional adaptations for their weakest writers, which included students with learning and other disabilities. A majority of teachers reported making the following adaptations for their weakest writers at least once a week: individual tutoring, assistance from a peer when writing, and extra instruction in grammar/spelling, planning/revising, handwriting, sentence construction, and text structure. On a monthly basis or more often, a majority of teachers reported they provided weaker writers with choice about writing assignments, the opportunity to complete an alternative writing assignment, and extra writing instruction via computer technology. How often an adaptation was applied did not differ by grade. The perceived adequacy of teacher undergraduate preparation to teach writing, their efficacy to teach writing, and the proportion of students with disability in their classes each made a unique and statistically significant contribution to predicting how frequently teachers applied adaptations for their weakest writers. The Chilean teachers in this study reportedly made adaptations for weaker writers in their class more often than teachers in studies conducted in other countries, including the United States.
Writing is a powerful and versatile tool. We use writing to communicate, inform, persuade, record information, express feelings, chronicle experiences, explore the meaning of events, create imaginary worlds, entertain others, and heal psychological wounds (Graham, 2019). In school, students write about materials read and presented in class (Bangert-Drowns et al., 2004). At work, workers use writing every day to perform their jobs (Light, 2001). At home, writing is used to initiate and maintain personal connections (Freedman et al., 2016). Consequently, writing is a key, competence students must master (UNESCO, 2015).
Despite its importance, there is considerable concern worldwide about students’ writing competence (Graham & Rijlaarsdam, 2016). Many youngsters, including students with learning disabilities (LD), do not acquire adequate proficiency as writers by the time they finish their compulsory education (e.g., Michaelowa, 2001; UNESCO, 2015). This is true in both affluent and less affluent countries. For instance, on the last national assessment of writing in the United States, two out of every three students did not write well enough to meet grade-level demands (National Center for Educational Statistics, 2012). In Chile, where this study took place, results from the last national assessment (i.e., Simce Escritura) revealed that almost one half of Chilean students experienced problems writing at an intermediate or advanced level in terms of development of ideas, text organization, and coherence (Agencia de Calidad de la Educación, 2017).
Instructional Adaptations
For students who do not master writing sufficiently, including students with LD (Graham et al., 2017), there is a price to pay. Poor writing can limit personal, academic, and occupational attainments. Although many factors influence writing development (Bazerman, 2009), schools are tasked with teaching writing. This requires that schools provide strong writing instruction for all students as well as instructional adaptations for students who need such modifications (Common Core State Standards et al., 2010). The basic idea underlying this proposition is that the success of weaker writers, including those with LD, is aided by making instruction responsive to students’ needs. As Corno and Snow (1986) indicated 30 years ago, improved outcomes in schools depend on designing instruction to meet student individual differences.
We think that instructional adaptations are especially critical for students who experience difficulty learning to write. They need additional assistance mastering different kinds of writing as well as essential writing skills and processes (McKeown et al., 2019). This is particularly important in the intermediate grades (Grades 4–6), as students are expected to increasingly use writing in more complex ways, including as a learning tool (Bangert-Drowns et al, 2004). Adapting writing instruction for weaker writers during these grades is important for three additional reasons. One, it is easier to address problems relatively early before they become more severe later (Lyon, 1996). Two, opportunities to enhance students’ writing via instructional adaptations in secondary schools are more limited, as middle and high school teachers spend little time teaching writing (Graham, 2019). Three, instructional adaptations may result in enough writing growth so that some children will not need special education services.
The study reported here examined intermediate-grade teachers’ adaptations for the weakest writers in their class, including students with LD and other disabilities. It further examined whether the frequency of teachers’ adaptations was related to beliefs about preparation to teach writing and efficacy for teaching writing. In contrast to prior research on instructional adaptations in writing, which took place mostly in the United States (see Graham, 2019), with one study involving teachers in Portugal and Brazil (Veiga Simão et al., 2016), there are no previous studies that have investigated this issue with a sample of Chilean teachers (Bañales et al., 2018). Therefore, the current investigation took place in Chile, providing a more international flavor to research in this area.
Chile provides an interesting context for studying instructional adaptations in writing. It is one of South America’s most prosperous countries, but income inequality is high (World Bank Group, 2015). To address these equity challenges and to improve education more broadly (Santiago et al., 2017), Chile introduced policies designed to improve teacher preparation and provide effective education to vulnerable students. Curriculum changes now place a strong focus on writing via learning objectives established in the Language and Communication Program of the Basic Education Curriculum in Chile from first to sixth grades (Ministerio de Educación, 2018).
The reform efforts undertaken across Chile have been quite extensive. This includes professionalization of the teaching profession, including increased funding for education, establishment of standards for the teaching profession (i.e., The Good Teaching Framework), creation of an extensive teacher evaluation performance system (with accompanying award systems for stronger teachers), institution of initial teacher education for all teachers provided by universities and professional institutes, development of guidelines for preparing teachers, installation of various forms of professional development, and the promotion of inclusive education in school across Chile (Santiago et al., 2017). At each level, these reform efforts promote the importance of adapting instruction for vulnerable students and students with special needs (Ministerio de Educación, 2015, 2016b, 2017b)
The current study focused on the types of adaptations made by intermediate-grade teachers in urban schools. In Chile, these are schools with more than 800 students, and 88% of elementary-grade language arts teachers in Chile work in such schools (Ministerio de Educación, 2017a). In addition, almost 90% of the population in Chile live in urban areas (Santiago et al., 2017).
Research Questions and Predictions
The current study addressed the following research questions:
We asked teachers to indicate how frequently they instituted specific instructional writing adaptations for their weakest writers (RQ1). We concentrated these adaptations on two aspects of instruction that improve intermediate-grades students’ writing: facilitating students’ writing and teaching important writing skills and strategies (see Graham et al., 2012; Koster et al., 2015).
For adaptations to facilitate writing, we focused on weaker writers choosing their own writing topic, completing an alternative writing assignment, and pairing them with a peer to complete writing tasks. There is evidence that pairing weaker writers with peers enhances writing performance (MacArthur et al., 1991). Although the impact of choice and alternative assignments has not been investigated empirically, we included these two adaptations as they are likely to increase weaker writers’ engagement to write (Patall, 2013).
For teaching adaptations, we asked teachers about providing extra instruction through individual mentoring as well as computer technology. We further asked them how often they provided extra grammar/spelling, planning/revising, sentence, and text structure instruction. Students who find writing challenging experience difficulty with these aspects of writing (Graham et al., 2017). There is also evidence that each of the teaching adaptations assessed can improve weaker students’ writing (Gillespie & Graham, 2014; Rogers & Graham, 2008).
In prior studies examining writing adaptations, U.S. elementary-grade teachers reported they commonly applied a variety of adaptations with their weakest writers (e.g., Graham et al., 2003, 2008a), with more than 50% of teachers making a majority of the individual adaptations assessed on a weekly (Graham et al., 2016) or monthly basis (Gilbert & Graham, 2010; Troia & Graham, 2017). A majority of teachers in Portugal and Brazil made virtually all their adaptations once a month. We anticipated that teachers in Chile would also make a variety of adaptations frequently.
For RQ2, we expected that use of specific adaptations by teachers would vary by grade. For example, teachers in a higher grade (Grade 6) may be less likely to teach handwriting than teachers in a lower grade (Grade 4), as handwriting is typically taught in lower grades (Graham et al., 2008b). In contrast, teachers in a higher grade (Grade 6) may be more likely to allow weaker writers to choose their own writing topics or modify them than teachers in a lower grade (Grade 4), as they may view their students as more capable of making such choices.
For RQ3, we anticipated that teacher beliefs would predict teachers’ instructional adaptations in writing. This prediction was based on the writer(s)-within-community (WWC) model (Graham, 2018). Although the WWC model situates writing within specific communities (e.g., classrooms) and argues that writing and the teaching of writing are influenced by features of the community as well as macro factors involving history, culture, institutions, society, and politics, it also contends that writing and the teaching of writing are shaped by the agency and actions of the members of a community. Pertinent to the present study, the WWC model proposes that what teachers do when teaching writing is influenced by their beliefs, including their beliefs about their own competence. Accordingly, such beliefs can foster or hinder writing instruction, as they can affect teaching behaviors, including how often teachers make instructional adaptations for their weakest writers. Thus, we expected teachers’ beliefs about their preparation to teach writing and efficacy to do so would collectively and uniquely make statistically significant contributions to predicting how often teachers made adaptations. These beliefs should serve as a guide for action (Fives & Buehl, 2012), with more adaptions made by teachers who are more positive about their preparation and capabilities.
Although teachers’ beliefs about preparation to teach writing and efficacy for doing so commonly predict how frequently teachers apply various writing practices in general (Graham, 2019), we know very little about the value of these two beliefs in predicting how often Grades 4 to 6 teachers apply adaptations with their weakest writers. Although we predicted that teachers who believe they are better prepared and more efficacious about teaching writing would make more adaptations, a study by Troia and Graham (2017) with Grades 3 to 8 teachers did not find that this was always the case. Teacher preparation predicted the frequency of two types of instructional adaptations (instructional and evaluative supports), but not a third type of adaptation (technological aids). Teacher efficacy only predicted the frequency of the third type of adaptation, but not the other two types of adaptations. In other studies, with elementary-grade students (Gilbert & Graham, 2010; Graham et al., 2003, 2016), teacher efficacy did not predict adaptations for weaker writers.
To bring greater precision to our analysis involving the predictive value of teacher beliefs, we first controlled for variance due to teaching experience, class size, percent of students with special needs, and percent of students who are weaker writers. It was important to control for these variables as each could hypothetically accounted for unique variance in the frequency of teachers’ adaptations. More experienced teachers may make more adaptations than less experienced ones, as they have had more time to learn how to address the needs of their weakest writers. A large number of students in a class may make it more difficult for teachers to make adaptations, as it is harder to provide individual attention to large numbers of students. Finally, teachers may make more adaptations when their class has larger numbers of weaker writers and students with disabilities, as the need for differentiated instruction is increased.
Method
Participants
The principals of the 2,473 municipal and private subsidized public urban schools serving students in Grades 4 to 6 in Chile, in the 15 regions of the country, were contacted. Three hundred nine of these principals agreed to allow their teachers to participate in this study, and the teachers in these schools were sent our survey by mail. Two hundred fifty-four of the Grades 4 to 6 teachers in these schools completed the survey. Based on a population of 3,131 teachers in the 2,473 urban schools in Chile, the sampling error for the most critical type of question teachers answered (items with seven choices asking about the adaptations teachers made for weaker writers) was 0.06 of a point assuming a statistical confidence interval of 95% (Dillman et al., 2009). Information about participants is provided in the Results section.
Survey Instrument
Participating teachers were asked to complete 43 survey items. These items asked teachers to provide information about themselves (nine items), their class (five items), instruction (six items), beliefs about writing (13 items), and adaptations for weaker writers (10 items). The items were drawn from a previous U.S. survey (Gilbert & Graham, 2010), and were converted to Chilean. Chilean teachers provided feedback on wording to ensure items were understandable.
Teacher information included gender, age, years teaching language arts, and education (i.e., undergraduate elementary education, undergraduate elementary education with a second major, undergraduate secondary education with a content major, graduate coursework in education/language arts, or a master’s degree in education/language arts). Items about class asked about class size, number of students with special education needs, and percent of students who were good, average, and weaker writers. For instruction, teachers were asked to indicate what percent of their instructional time in writing was spent teaching the whole class, small groups, and individual students. Teachers were also queried about the amount of time they spent each week teaching writing, reading, and speaking. They responded to these three questions using a six-item scale: no time devoted (score of 1), less than 1 hr (score of 2), 1 to 2 hrs (score of 3), 3 to 4 hrs (score of 4), 5 to 6 hrs (score of 5), more than 6 hrs (score of 6).
Teachers were asked to rate the adequacy of their preparation to teach writing based on their college experiences as an undergraduate (preservice), after college through the experiences provided by their schools/districts (in-service), and postgraduate studies (master’s, specialization courses). They were asked to respond to each of these items using a Likert-type scale with items not prepared (score of 0), minimally prepared (score of 1), adequately prepared (score of 2), and extensively prepared (score of 3).
Teachers’ beliefs about efficacy to teach writing was assessed using 10 items taken mostly from Gilbert and Graham (2010). These items assessed teachers’ beliefs about their overall effectiveness to teach writing, effectiveness to teach specific aspects of writing (e.g., writing concept or skill), and effectiveness to address specific issues (e.g., students’ difficulties with a writing assignment). Teachers responded to each item with a 6-point Likert-type scale ranging from strongly disagree (score = 1) to strongly agree (score = 6). A higher score indicated stronger efficacy (the scale is presented in the appendix).
Ten items assessed teachers’ adaptations for the weakest writers in their class, including students with LD and other disabilities (Gilbert & Graham, 2010). These items assessed adaptations for teaching writing (e.g., individual mentoring, extra instruction in sentence construction) and facilitating the writing process (e.g., writing with peers, providing choice). Teachers indicated how often they applied these adaptations with the weakest writers in their classroom using a 7-point Likert-type scale: never (1), once a year (2), 2 to 3 times a semester (3), once a month (4), once a week (5), 2 to 3 times a week (6), daily (7). Each semester in Chile lasts for approximately 5 months. This scale differs.
Procedures
A letter describing the study was sent via email to all principals in elementary municipal and private subsidized public schools in urban areas in all regions of Chile. The names and contact information of these principals were obtained from the Official Directory of Establishments in Chile (Ministerio de Educación, 2016a). In this letter, we presented the purpose of the study and that participating teachers would be part of a raffle distributing 10 digital gift cards valued at approximately US$37.00 each. Once we obtained teacher emails from a principal, the informed consent letter describing the study, its risks, and the raffle were electronically sent to teachers as was the survey. One week after the first survey was emailed, a follow-up email was sent 1 and 2 weeks later, and again 4 weeks later. A final email was sent out at 8 weeks.
When teachers emailed us their survey, teacher names were changed to an identification number and then deleted. This ensured anonymity of responses from teachers, as there was no link between teacher responses and email addresses. It also provided a mechanism to ensure follow-up emails did not go to teachers who had already completed the survey. Teachers not interested in participating in the study had the option to opt out at any time.
Results
Before presenting the findings for our three research questions, we first present descriptive information on the teachers who completed the survey, followed by the outcomes from analyses examining the factor structure of the 10 adaptation and the 10 efficacy items. We then describe how frequently teachers made each adaptation for their weakest writers, determining which adaptations were made most frequently (RQ1) and whether adaptations varied by grade (RQ2). Next, we present the results of a regression analysis examining whether teacher beliefs about preparation and efficacy to teach writing predict frequency of adaptations (RQ3).
Participating Teachers
Descriptive information about the participating teachers is provided in Table 1. Most of the 254 teachers were female (88%). Slightly more than one fourth of the teachers taught sixth (26%) and fifth grades (26%), whereas the remaining teachers taught fourth grade (48%). Most of the teachers (90%) majored in elementary education as undergraduates, with slightly more than half of these elementary education majors completing a second major (51%). A relatively small minority of teachers (10%) were secondary education undergraduate majors, whereas three teachers had not majored in education at all as undergraduates (1%). Slightly more than one fourth of the teachers (28%) had taken graduate coursework in the area of language arts, but only 10% of the 254 teachers had completed a master’s degree in this area. Teachers were mostly positive about their preparation to teach writing, with 60%, 65%, and 49% indicating their undergraduate, in-service, and graduate preparation were either adequate or extensive.
Teacher Characteristics.
As a group, the teachers taught language arts for an average of 6 years (SD = 7.26 years; see Table 2). Their fourth- to sixth-grade classrooms averaged 33 students (SD = 20.02), and about one fifth of their students (M = 7.78, SD = 20.02) had a disability. They indicated that almost one fourth of their students were good writers (24%), approximately one half of them were average writers (45%), and slightly more than one fourth of them were weaker writers (27%).
Means and Standard Deviation of Control Variables and Predictors.
Note. Scale for teacher efficacy: strongly disagree with efficacy statement (score = 1) to strongly agree with efficacy statement (score = 6); scale for preparation: not prepared (score of 0), minimally prepared (score of 1), adequately prepared (score of 2), and extensively prepared (score of 3); scale for teacher adaptations: never (1), once a year (2), 2 to 3 times at semester (3), once a month (4), once a week (5), 2 to 3 times a week (6), and daily (7).
Teachers indicated they commonly devoted 1 to 2 hr (38% of teachers) or 3 to 4 hr (33% of teachers) teaching writing each week (see Table 3). They devoted similar amounts of time to teaching speaking, with the majority of teachers spending 1 to 2 hr (43% of teachers) or 3 to 4 hr (22% of teachers) teaching it. Time devoted to reading instruction was greater, as most teachers devoted 3 to 4 hr (39% of teachers), 5 to 6 hr (25% of teachers), or more than 6 hr (20%) to teaching it each week. The majority of writing instruction involved teaching the whole class (51%), followed by small group instruction (28%) and individualized instruction (25%).
Frequency and Means of Time Devoted to Reading, Writing, and Oral Instruction.
Note. Teachers responded to a scale with 6 points: not devoted time (1), less than 1 hr (2), 1 to 2 hrs (3), 3 to 4 hrs (4), 5 to 6 hrs (5), more than 6 hrs (6).
Factor Analyses
To determine the factor structure of the adaptation and self-efficacy items, we conducted two factor analyses. The first factor analysis of the 10-item Adaptation Scale using data from this study yielded a single factor with an eigenvalue of 6.11 accounting for 61% of the variance. All 10 items loaded on this single factor at 0.51 or higher. Coefficient alpha for this scale was .94. The score for the adaptation factor was the average of the 10 items.
A second factor analysis of the 10-item Self-Efficacy Scale using the data from this study produced a single factor with an eigenvalue of 4.14. This factor accounted for 68% of the variance, and all 10 items loaded on this single factor at 0.47 or higher. Coefficient alpha was .95. The score for self-efficacy factor was the average of the 10 items.
How Frequently Do Teachers Apply Adaptations for Weaker Writers (RQ1)?
Table 4 presents information on how frequently the participating teachers made adaptations for their weakest writers. More than one half of the teachers indicated they provided additional instruction in grammar and spelling (58% of teachers), planning and revising (56% of teachers), handwriting (59% of teachers), sentence construction (62% of teachers), and text structure (54% of teachers) to these students at least once a week. More than two thirds of the teachers (69%) reported they provided individual mentoring to their weakest writers at least once a week, whereas a slight majority of teachers (55%) paired weaker writers with peers to write at least once a week. Just more than one half of teachers (51%) provided weaker writers with a choice on what to write about at least once a week, and about three out of every four teachers (72%) allowed these students to complete an alternative writing assignment at least once a month. Slightly more than one half of teachers (52%) provided weaker writers with extra instruction using technology at least once a month, but close to one third (36%) made this adaptation once a year or less often.
How Frequently Teachers Applied Specific Adaptations with Struggling Writers.
Note. Teachers responded to a scale with 7 points: never (1), once a year (2), 2 to 3 times at semester (3), once a month (4), once a week (5), 2 to 3 times a week (6), and daily (7).
We conducted a one-way analysis of variance (ANOVA) with a repeated measure to determine whether some adaptations were applied more often than others. This analysis was statistically significant, F(1, 211) = 1,932.92, p < .001. Follow-up analyses revealed teachers were more likely to engage in individual mentoring with students than all other adaptations (all ps < .007), except for extra sentence instruction. Furthermore, teachers were more likely to provide extra sentence instruction than all other adaptations (all ps < .007), except for individual mentoring as well as extra grammar/spelling and handwriting instruction. Teachers were more likely to provide choice, pair weaker writers with peers, and provide extra grammar/spelling, handwriting, and text structure instruction than apply alternative assignments or provide extra instruction using computer technology (all ps < .03). Finally, teachers were more likely to provide extra planning/revising instruction and assign alternative writing assignments than provide extra writing instruction with computer technology (both ps < .001).
Do Teacher Adaptations for Weaker Writers Vary by Grade (RQ2)?
For each of the 10 adaptations assessed in this study, a separate one-way ANOVA was conducted to determine whether teachers in the three grades (4–6) differed in how often they applied an adaptation. To control for Type 1 errors, we made a Bonferroni adjustment (α of .05/10 analyses), setting alpha at .005. Using this criterion, grade was not statistically related to frequency of use for any of the adaptations (all ps > .030).
Do Teachers’ Beliefs Predict Teachers’ Adaptations for Weaker Writers (RQ3)?
Analysis
We conducted a hierarchical regression analyses to determine whether teachers’ beliefs about their preparation and efficacy for teaching writing collectively and separately predicted adaptations for weaker writers after first accounting for variance due to the following control variables: experience teaching language arts, class size, percent of students with special needs, and percent of students who were struggling writers (see Table 5). The control variables were first entered as a block. Then, the teacher belief measures (preparation and teacher efficacy) were entered as a second block. This allowed us to determine whether preparation and efficacy together made a statistically significant contribution to predicting adaptations after accounting for variance due to the control variables. It also allowed to determine whether measures of teacher beliefs (e.g., teacher efficacy) made a statistical contribution to predicting adaptations after accounting for variance due to the control variables and the other belief measures (the three preparation items).
Multiple Regression Analysis for Adaptations for Struggling Writers.
Note. Coefficients are standardized.
The scores for adaptations and teacher efficacy were the average of the 10 items measuring each of these constructs, respectively. Three separate scores assessed teacher preparation. This included scores for adequacy of undergraduate preparation, graduate preparation, and in-service preparation. We did not calculate an average score for these items as less than 20% of teachers had completed any preparation beyond the undergraduate level.
The regression analysis involved 205 of the 254 teachers who returned the survey. It was necessary to drop 43 teachers from this analysis because they did not provide information on questions concerning preparation, number of special education students in their class, or number of years teaching language arts. Descriptive statistics for the variables included in the regression analysis were adaptations (M = 4.31, SD = 1.39), teacher efficacy (M = 4.75, SD = 1.07), undergraduate preparation (M = 3.14, SD = 0.76), graduate preparation (M = 3.69, SD = 1.08), in-service preparation (M = 3.36, SD = 0.98), years teaching language arts (M = 5.86, SD = 7.08), class size (M = 35.45, SD = 20.02), percent of students with disabilities (M = 26.03%, SD = 18.62%), and percent of weaker writers (M = 26.75%, SD = 17.80%).
Correlations among variables
Although the three items measuring teacher preparation were statistically related (correlations ranged from .34–.37), only the undergraduate preparation score was statistically related to teacher efficacy (.15). In addition, teacher efficacy was statistically related to the adaptation score (.21), but the scores for the three preparation items were not.
For the control variables, percent of students with special needs (.36) and class size (.18) were statistically related to the adaptations score, but years teaching language arts and percent of weaker writers were not. The only statistically significant relationship among control variables involved class size and percent of students with special needs (.44).
Regression analysis
After first controlling for variance, R2 = .13, F(3, 201) = 9.95, p < .001, related to the four control variables (years teaching language arts, class size, percent of poor writers, percent of students with disabilities), the four teacher belief items (teacher efficacy as well as undergraduate, in-service, and graduate preparation) collectively accounted for a statistically significant amount of variance in frequency of adaptations, R2 = .06, F change(5, 196) = 2.94, p = .014. Unique statistically significant contributions across all control and predictor variables involved percent of students with disabilities (t = 4.78, p < .001), efficacy (t = 2.54, p = .012), and beliefs about undergraduate preparation (t = 2.29, p = .023).
Discussion
Even though it is assumed that school success depends on adapting teaching to students’ needs (Corno & Snow, 1986), there are less empirical data on how often elementary-grade teachers make adaptations for children who find school challenging, especially for those who experience difficulty learning to write. There are just a few studies examining the adaptations elementary teachers make for their weakest writers, and they are mostly limited to teachers in the United States (Gilbert & Graham, 2010; Graham et al., 2003, 2016; Troia & Graham, 2017), with some additional study in Europe and Brazil (Veiga Simão et al., 2016). The current study extends this previous work by surveying how often Grades 4 to 6 teachers in urban schools in Chile make instructional adaptations for their weakest writers.
Teachers in Chile Make Writing Instructional Adaptations Frequently
When Chilean teachers were asked to indicate how frequently they made 10 adaptations for their weakest writers, most teachers indicated they made these adaptations weekly or more often. Each of the adaptations for teaching (extra handwriting, grammar/spelling, planning/revising, sentence construction, and text structure instruction as well individual mentoring) was provided by a majority of teachers at least weekly. The only exception involved extra writing instruction using computer technology, which was applied by a majority of teachers at least monthly. Adaptations for facilitating writing were also applied at least weekly (weaker writers working with a peer) or monthly (providing choice or alternative writing assignments). Not surprisingly, some adaptations were applied more frequently than others (e.g., individual tutoring and extra sentence instruction were applied most frequently), but we did not find that teachers in different grades made specific adaptations more or less often. Additional research is needed to replicate these findings and explore why teachers favor some adaptations over others.
Like elementary-grade teachers in the United States (Gilbert & Graham, 2010; Graham et al., 2003, 2016; Troia & Graham, 2017) and Portugal and Brazil (Veiga Simão et al., 2016), teachers in Chile applied a variety of different instructional adaptations for writing. The Chilean teachers appeared to apply these adaptations more frequently than their U.S., Portuguese, or Brazilian counterparts though. When we examined equivalent adaptations in U.S. investigations, we found that a majority of these adaptations (60%) were applied at least several times a month (Gilbert & Graham, 2010; Troia & Graham, 2017) or monthly (Graham et al., 2016). Chilean teachers in this study applied 70% of these adaptations at least weekly. Although it was difficult to compare our findings with those from Portugal and Brazil (Veiga Simão et al., 2016), as there was virtually no overlap in the adaptations studied, a majority of Portuguese and Brazilian teachers made their adaptations once a month or less often.
It is not clear why Chilean teachers made more adaptations than U.S., Brazilian, or Portuguese teachers. One possible explanation is that Chile is undergoing reforms designed to enhance schooling generally and writing instruction specifically. However, two of the studies conducted in the United States (Graham et al., 2016; Troia & Graham, 2017) were enacted during the Common Core State Standards era (Common Core State Standards et al., 2010). This reform movement, like the one in Chile, proposed standards aimed at improving the teaching of writing, including the importance of instructional adaptations. It is possible, that teachers and schools in the United States did not place as much importance and value on these reform efforts as teachers in Chile did, leading to differences in how often they applied instructional adaptations for weaker writers in the classroom. It is also possible that differences in the specifics of the reform efforts in these two countries, and how the resulting standards were to be achieved, interacted with political, social, cultural, and historical variables (Graham, 2018), to affect the frequency of teachers’ adaptations for their weakest writers. This does not, however, explain differences between Chilean teachers and those from Portugal and Brazil, as writing reform was not emphasized in these countries when Veiga Simão et al. (2016) conducted their investigation.
Another possible reason for why teachers in Chile adapted writing instruction for their weakest writers so frequently involves preparation to teach writing. The Chilean teachers in this study were much more positive about their preparation than teachers in most studies conducted in the United States (e.g., Gilbert & Graham, 2010) and moderately more positive than Portuguese and Brazilian teachers (Veiga Simão et al., 2016). Chilean teachers’ positive beliefs about their preparation may reflect the establishment of a stronger skill set to teach writing to their weakest writers.
Differences between Chilean and U.S. teachers in the frequency of adaptations, however, may reflect more mundane methodological differences in investigations. For example, our study involved Chilean teachers working in urban areas, whereas the U.S. studies sampled teachers from all strata of society (e.g., Troia & Graham, 2017), although Portuguese and Brazilian teachers in Veiga Simão et al. (2016) were from metropolitan schools. Furthermore, the available studies did not ask teachers the exact same questions about instructional adaptations for writing, nor did they use the exact same scale for responding to items. For example, Veiga Simão et al. applied a 5-point Likert-type scale (never, several times a year, monthly, weekly, and daily), and even studies that applied a 7-point Likert-type scale differed to some degree in terms of the anchor points described. We do not think this accounts for why Chilean teachers indicated they made more adaptations than teachers in previous studies, as differences centered mostly on weekly adaptations versus bimonthly or monthly adaptations.
Finally, Chilean teachers’ responses to our survey may have been positively skewed by the selection process or teachers’ desire to respond in socially desirable ways. It is possible that principals did not share our survey with teachers unless they thought they were strong writing teachers. It is also possible that the teachers who completed the survey wanted to present themselves in a positive light, and they indicated they made more adaptations than they actually did. This last explanation does not adequately explain why teachers in Chile were more likely to adapt writing instruction for weaker writers, as teachers in studies conducted in other countries would also be likely to do the same.
Although we cannot establish that reform efforts in Chile are responsible for the relatively positive outcomes in the present study, this investigation does provide an initial baseline for future studies in this area. In any event, research is needed that compares the adaptations made by teachers not only in Chile and the United States but also in other countries. It is also important in future investigations to query Chilean teachers about how reform efforts in Chile influenced how often and the types of adaptations they made. As we noted earlier, adaptations and inclusion are emphasized at each level of the reform movement, including standards for the teaching profession, teacher evaluation performance system, initial teacher education, guidelines for preparing teachers, and ongoing professional development (Santiago et al., 2017). Teachers should be asked to describe how each of these reform efforts influenced whether and how they adapted writing instruction for their students.
It is also important that future research examines a broader range of adaptations in writing, including asking teachers more specifically about the writing adaptations they make for children with LD and other disabilities. Although we asked Chilean teachers to indicate how often they made adaptations for their weakest writers, we did not ask them specifically about adaptations for students with LD and other disabilities. No study to date has asked about adaptations for students with LD, although Troia and Graham (2017) asked about students with disabilities in general. It is possible that the types of instructional writing adaptations teachers make for students with different types of disabilities vary.
More research is needed to explore how adaptations for weaker writers are enacted in real time, how teachers sequence and combine instructional adaptations to meet instructional writing goals, and how such adaptations influence students’ writing and growth as writers. Such studies will require the use of multiple methodologies, including qualitative, observational, correlational, and instructional research.
Teachers’ Beliefs Predict Their Use of Adaptations for Their Weakest Writers
Research conducted with elementary-grade teachers demonstrates that what they believe predicts their observed (e.g., Rietdijk et al., 2018) and reported writing practices (De Smedt et al., 2016). Two beliefs that consistently predict teachers’ general writing practices are beliefs about preparation to teach writing and efficacy to do so (Graham, 2019). Even so, the value of these two beliefs in predicting teachers’ adaptations for their weakest writers has been inconsistent. For example, Troia and Graham (2017) reported that preparation and efficacy predicted some types of adaptations and not others.
In the present study, we found that teachers’ beliefs about the adequacy of undergraduate teacher preparation as well efficacy to teach writing made statistically unique contributions to predicting how often teachers made adaptations for their weakest writers. Graduate-level preparation and in-service preparation did not make such a contribution. Additional research is needed to replicate these findings and extend them to other contexts.
Future research also needs to take a more nuanced look at how beliefs about preparation and efficacy predict teachers’ adaptations for weaker writers, examining more than just a global judgment about each belief. This includes asking teachers to judge multiple aspects of their preparation, including whether their preparation provided them with the knowledge and skills to (a) understand how writing functions and develops, (b) teach writing to typical students as well as to specific populations of students (such as those with LD), and (c) conduct formative assessment that allows them to change their writing instruction to meet their students’ needs as well as make adaptations for specific students. Likewise, teacher efficacy needs to address teachers’ confidence in carrying out these same types of functions in their classroom. Such research would more precisely pinpoint what aspects of beliefs about preparation and efficacy predict writing practices more broadly, and adaptations more specifically.
Finally, we found that the proportion of students in a class with disabilities uniquely and statistically predicted how often teachers made instructional adaptations for weaker writers. This was not the case for proportion of students in a class who were weaker writers. These findings suggest that teachers did consider disability status when making adaptations. This provides additional support for the earlier recommendation that future research should examine how teachers make instructional writing adaptations for different types of students in their classroom.
Implications for Practice
Teachers in the present study reported they frequently used a variety of adaptations to meet the needs of their weakest writers. Furthermore, teachers’ beliefs about their preparation and efficacy to teach writing predicted how frequently they made adaptations for these students. Although this study is descriptive, self-reported, and correlational, it provides a positive demonstration that teachers do adapt their writing instruction for their weakest writers. Such demonstrations provide a script for what other teachers can and hopefully will do. It is important that preservice teachers and those already teaching are aware that adaptations for writing are common and expected, even in classes with the highest proportion of students with disabilities
The findings that beliefs about preparation to teach writing and efficacy to do so positively predict reported writing adaptations imply that it may be fruitful to attend to these two ingredients in efforts to increase teacher use of instructional adaptations with their weakest writers. Both preparation and efficacy are malleable factors, and teachers are not overly positive about either when evaluating them. A majority of teachers in studies worldwide typically describe their preservice and in-service preparation as less than adequate, and they are only slightly positive about their competence to teach writing (Graham, 2019). It may be particularly useful for teacher preparation programs and the schools in which teachers work to devote the energy and resources needed to prepare teachers who have the know-how and confidence to teach writing effectively as well as make effective adaptations for their weakest writers, including students with LD.
Footnotes
Appendix
Self-Efficacy Scale to Teach writing.
| 1. When a student’s writing performance improves, it is usually because I found more effective ways of teaching that student. |
| 2. If a student does not remember what I taught in a previous class, I know what to do to improve their learning. |
| 3. If a student can quickly master a new concept or skill related to writing, it is because I knew how to teach him/her. |
| 4. If I work hard, I can help my students solve the most difficult writing problems. |
| 5. When a student improves his writing performance, it is due, in part, to my extra efforts. |
| 6. If a student is unable to do a writing task, I can accurately judge whether the task has the appropriate level of difficulty. |
| 7. When a student experiences difficulty with a writing task, I know how to adjust it to his or her level. |
| 8. When a student’s writing performance improves, it is usually because I found more effective teaching approaches. |
| 9. I effectively manage the class while teaching writing. |
| 10. I consider myself effective in teaching writing. |
Note. Teachers responded to a scale with 6 points: completely disagree (1), moderately disagree (2), slightly disagree (3), slightly agree (4), moderately agree (5), completely agree (6).
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
This research was funded by Consejo Nacional de Ciencia y Tecnología (CONICYT; Fondecyt Regular Number 1171898).
