Semilingualism Applied to the Literacy Behaviors of Spanish-Speaking Emerging Bilinguals: Bi-illiteracy or Emerging Biliteracy?
by Kathy Escamilla - 2006
Language differences in the United States are largely viewed as problems that schools must remedy. This paradigm has created the pervasive belief that Spanish is a root cause of underachievement for Spanish-speaking English language learners (ELLs). This article examines teacher beliefs systems with regard to the above paradigm. It also examines informal writing assessments as tools to evaluate children’s bilingual writing development. Fourth- and fifth-grade students were given writing prompts in English and Spanish. Teachers at the school read and rated the samples. Findings indicated that children had strengths in writing, particularly in the area of complex ideas. In contrast, teachers’ discussions focused on perceived weaknesses. Findings raise questions about using writing-sample data and teacher judgment to identify ELLs for special education. Study findings raise some questions in terms of using writing-sample data and teacher judgment to identify ELL students for special education.
Ruiz (1988) and others have argued that language differences in the United States are largely viewed as problems that public schools must remedy. The language-as-a-problem paradigm has become ingrained in school policies and practices, particularly in districts and schools with large numbers of children who are second language learners of English, and most especially with regard to students whose first language is Spanish. Language-as-a-problem paradigms create the pervasive belief that Spanish is the root cause of underachievement of students who are native Spanish speakers.
An extreme view of the language-as-a-problem paradigm is the common belief among teachers and policy makers that some school-age children born to Spanish-speaking parents have no language (MacSwan, Rolstadt, & Glass, 2002). Evidence of this belief is provided by MacSwan et al., who documented that in 1996, the Los Angeles Unified School district reported having 6,800 children who were classified as non-nons, more commonly referred to as semilinguals. Semilinguals are children who are thought to be nonverbal in both English and their native language. MacSwan et al., and Valadez, MacSwan, and Martinez (2001) have raised questions about the validity of semilingualism as a socially constructed concept related to language proficiency. Sadly, the views that non-English languages are problems and that more than one language can cause semilingualism have become erroneous axioms upon which many school policies and programs are based.
This article will extend the concept of semilingualism beyond that of oral language and into the realm of literacy, and examine how concepts of semilingualism have been applied to students who are learning to read and write in Spanish and English in U.S. schools. For purposes of this article, I label the application of semilingualism to literacy as bi-illiteracy. Bi-illiteracy is a socially constructed concept that implies low levels of literacy in both English and Spanish. In this article, I argue that the language-as-a-problem belief among teachers and policy makers extends beyond beliefs about perceived oral language deficits of Spanish-speaking children into beliefs about literacy acquisition and achievement.
In their seminal work on bilingual special education, Baca and Cervantes (2004) demonstrated that the principal reason for overrepresentation of ELLs in special education is biased assessment practices. They primarily discuss this bias with regard to formal, norm-referenced testing. This article adds to that literature and demonstrates that even well-intentioned informal bilingual assessment may be biased.
REVIEW OF LITERATURE
Using the work of Bernhardt (1991, 2000, 2003), Grant and Wong (2003), and Garcia (2000), I contend that many perceived beliefs about bi-illiteracy are due to pervasive yet unexamined assumptions of teachers and other educators in schools with large numbers of second-language learners. These assumptions include an almost exclusive reliance on first-language reading and writing research in the development and implementation of literacy instruction and assessment programs for bilingual learners (Smith, Jimenez, & Martınez-Leon, 2003). First-language reading research is almost exclusively English-language based. This is further exacerbated by a reliance on the overwhelmingly English-speaking North American/British/ Australian literacy industry that drives teacher education and policy with regard to the development of literacy programs (Bernhardt, 2003). As some have coined this phenomenon, good teaching is good teaching, or one size fits all, implying that literacy practices found to be effective in suburban Pittsburgh with monolingual English students will be equally effective in inner-city Los Angeles with predominantly Spanish-speaking students with no need for modification (Grant & Wong, 2003). These assumptions also include the widespread belief that reading processes in a second language may not be significantly different from those in a first language (Fitzgerald, 1993, 1995a, 1995b) and that learning to read in English as a second language is the same for all language groups. This assumption stems from the homogenization in labeling all second-language learners as English language learners (ELLs) or limited English proficient (LEPs) and implies that learning to read in English as a second language is no different for a native speaker of Farsi than for a native speaker of Spanish. Dubin, Eskey, and Grabe (1986) suggested that different languages and their different orthographic systems may require different reading strategies . . . or at least suggest the likelihood that various universal processes interact differently for optimal processing in different languages (p. 29). Reliance on universal assumptions eliminates the need to study language-specific issues in the acquisition of literacy and biliteracy, and invites the assessment of second-language learners acquisition of English using monolingual English assessments and policies.
In a review of the research and pedagogy in writing in ones second language (L2), Hedgcock (2005) noted that overall, there is a dearth of L2 writing research in the literature. With regard to the understanding of L2 writing development, this literature review discusses three particular issues. The first is that the extant body of literature focuses on writing development in L2, not biliteracy development, with L2 writing development being assessed vis-a-vis the writing development of native speakers of the language. This is particularly true when English is the L2. Second, linguistic accuracy serves as the most influential, and therefore problematic, formal dimension known to influence raters perceptions of L2 writing. Finally, there is a gap in the attention in the research between outcomes and processes. L2 writing research has been focused on the outcomes of writing but rarely on the dimension of day-to day classroom writing instruction, which no doubt influences the outcomes.
Wiley (2005) added that there is also a dearth of research on biliteracy development, particularly when focused on language minority children in the United States. In his review of the research, Wiley identified that this dearth of research is partly attributable to definitional issues (e.g., how do we define literacy and biliteracy?), to an assessment system that only measures literacy attainment in one language (English), and to a policy environment that has never endorsed the kind of maintenance bilingual education programs that would ensure that minority students could attain biliteracy. In short, U.S. programs for Spanish-speaking children learning English have paid scant attention to conceptualizing how to develop bilingualism and biliteracy in school.
Added to the above is the literature on teacher beliefs and how these beliefs are translated into instructional practices. There is an extensive body of research documenting that both majority and minority teachers have taught in teacher education programs to believe that minority children come to school with deficits in background and experience that must be corrected by the school in order for these children to achieve. As hooks (1993) said, Most teachers have been taught in classrooms where styles of teaching reflected the notion of single norm of thought and experience. In the U.S. this has been just as true for non-white teachers as for white teachers (p. 91). Because of the prevalence of best practice paradigms, if students do not respond positively to instruction that is labeled as good for everyone, teachers frequently attribute the lack of effectiveness to problems within the student rather than to the possibility that the teacher method or approach was inappropriate given the population. If the norm is monolingual, then bilingual by definition is not normal. Researchers such as Halcon (2001) added that the manifestation of mainstream ideology in teacher practices and attitudes ultimately negatively affects the education of Latino children (p. 67). Other researchers, such as Bartolome and Balderrama, (2001), Nieto (1996), and Gay (1995), argued that although most teachers have good intentions for the children they teach and care deeply about them, they may have deep-seated racist tendencies and beliefs that they have never been asked to identify or examine. In short, teacher beliefs that language, culture, and social class is a problem, combined with a dearth of literature on the development of biliteracy, are enacted in teaching practices that often address the problem of language rather than the development of bilingualism and biliteracy.
The above assumptions provide the conceptual framework for discussing bi-illiteracy as a socially constructed category. Working with second-language students poses several challenges for teachers who are unfamiliar with how a students second language interacts with the first. Students learning to write in a second language make progress in their literacy development in ways that differ from literacy development of English monolingual students. The reason for this is that students come to school with a developed oral (and often written) first language, and certain aspects of the first language, including orthography, phonemic system, and rhetorical structure, do not transfer neatly (Luther, 1997; Montaño-Harmon, 1991; Odlin, 1989; Perez, 1998). As a result, linguistic rules from the first language that transfer to ELLs writing in English are often misinterpreted as a language problem rather than a natural progression of second-language literacy development (Cummins, 2001; Escamilla & Coady, 2001, 2003).
The purpose of this study was twofold. The first was to examine teachers perceptions of the writing behaviors of students who are learning to read and write in both Spanish and English in an elementary school in Colorado. Teacher perceptions play a significant role in labeling children as good writers or poor writers, and reading and writing achievement are significant factors in referrals of ELLs and others to special education.
Data for this study were collected in an elementary school in a large urban school district in Colorado. The elementary school is located in a Spanish-speaking neighborhood that is composed predominantly of Mexican immigrant and Mexican American families. The school is medium sized with a population of 475 children, of whom 89% are ethnically Latino, 47% are classified as ELLs whose first language is Spanish, and 97% are on free or reduced lunch. The Latino population of this school is overwhelmingly Mexican and Mexican immigrant. Although new immigrants arrive almost daily at the school, the majority (over 60%) of the fourth- and fifth graders at this school had been in the school since kindergarten. During the 2002-2003 school year, the school received a statewide rating of low (Denver Public Schools [DPS], 2004a). School ratings are based primarily on the outcomes of the Colorado Student Assessment Program (CSAP) tests. The CSAP is Colorados version of wide-scale high-stakes testing.
The elementary school implements an English language acquisition (ELA) program (Denver Public Schools, 2004b) for all students who are identified as ELLs. The ELA program is modeled after an early-exit transitional bilingual program. In that program, students who are identified as ELLs and speak Spanish as a native language are provided literacy instruction in Spanish for 4 years (kindergarten through grade 3). The program also mandates daily oral English as a second language (ESL) instruction. By grade 4, most of the students are transitioned into all-English-language instruction.
At the time of the study, very little was known about students academic achievement in Spanish, for reading or writing, because neither the school nor the district assessed student achievement in Spanish. Further, academic achievement in English was measured only in reading and language arts via the Iowa Test of Basic Skills (ITBS) test. Student achievement in writing in English was not formally assessed. ITBS data were not disaggregated by language. In 1997, the ITBS outcomes for fourth graders showed reading scores in the 20th percentile, and language arts outcomes at the 23rd percentile. For fifth graders, reading results were at the 18th percentile, and language arts results were at the 15th percentile. Teacher interviews with both bilingual and monolingual teachers in the study indicated that teachers overwhelmingly attributed these outcomes to the schools demographic situation, particularly to the fact that a majority of children in the school were native Spanish speakers and learning English as a second language. The bilingual teachers expressed confidence that if they had an achievement test in Spanish, their students would do well on it.
At the time of the study, the school had 18 full-time classroom teachers, 12 of whom were bilingual in English and Spanish, and 6 of whom were monolingual English. The bilingual teachers were split evenly between native Spanish-speaking bilingual teachers and native English-speaking bilingual teachers. Further, they were ethnically diverse, with 5 Latinas, 1 African American, and 6 Whites. It is important to note, however, that only one of the bilingual teachers had specialized training in the teaching of literacy in Spanish, because she had been an elementary school teacher in Mexico prior to immigrating to the United States. Of the 12 bilingual teachers, 9 had advanced degrees (masters degrees) and state-approved endorsements in bilingual education. All 12 of the teachers were considered by the district to be fully endorsed bilingual teachers because they had completed a district-approved 150-hour in-service training program. The focus of the training program was on compliance with a court order related to teaching second-language learners, and little training emphasis was given to pedagogical issues with regard to literacy and language teaching in ELA classrooms.
Of the six monolingual English teachers, three had bachelors degrees and three had masters degrees in elementary education. Five were White, female, and middle class, and 1 was Latino and grew up in a working-class family who spoke primarily English.
Table 1 summarizes characteristics of the participating teachers.
Research work in this school was structured through a partnership between the University of Colorado, Denver, and the local school district. The partnership model was a professional development school model, and as such, the university partnership had four specific components: (1) preservice teacher education, (2) professional development for teachers in partner schools, (3) curriculum and instruction innovation, and (4) research and inquiry. I spent 4 years as a leadership area professor in this partnership school (1994-1998). During 2 of these years, doctoral students worked with me on the research aspects of the partnership. As a part of the partnership program, teacher candidates, teachers, and professors were required to engage in research projects that could serve to improve instruction in the school.
The year that this study was conducted was the first year that Colorado began the implementation of high-stakes assessments in the form of the CSAP. Teachers at the school were concerned about how the Spanish-speaking students at the school would fare on the new assessment, and they were particularly concerned about the writing portion of this assessment in English and Spanish. For these reasons, the teachers, school principal, teacher candidates, and university professor decided that the research project for the year should focus on writing.
All teachers in the school, monolingual and bilingual, acknowledged that over the years, they had paid scant attention to the development of writing either in Spanish or English and that their literacy instruction was primarily focused on reading. We worked with them to develop a holistic writing rubric as a way to gauge writing development, in both Spanish and English, of students in the school. We understood that a students writing is greatly influenced by purpose and content. Thus, allowing students only a first-draft attempt at writing and evaluating that piece with a rubric might not capture what a student actually knows as a writer. Although we understood the limitations of using writing prompts and rubrics, we decided to use them to gather data on students writing for several reasons. First, the state had created a standards-based assessment system known as the CSAP, and the writing portion of the CSAP was going to be administered through the use of a writing prompt as the vehicle for assessing student writing proficiency. Thus, we felt that using the writing prompt would be a good way to give the students practice for the upcoming CSAP, to provide the teachers with information about how the students performed on this type of writing assessment, and to understand how a rubric would be used to assess their writing. Further, the CSAP writing assessment was going to be made available at third and fourth grade in Spanish and English. Finally, through the development of writing prompts, the collecting of writing samples, and the subsequent dialogues about student writing in both Spanish and English, we were able to gather data about teachers perceptions of student writing skills in both English and Spanish.
Because this particular research project was a schoolwide inquiry project, data in all classrooms were collected on the same days and in the same ways. Writing assessments were given by classroom teachers and were administered to the whole class on one day in early October 1997, and again in late March 1998. Teachers were given a protocol to administer the writing prompt that specified the following administration procedures: (1) Teachers were asked to write the title of the writing prompt on the chalkboard. They were asked not to write any other information on the board. (2) Teachers told the students that they had 30 minutes to write about the prompt and that they should use their own life experiences to help them write the story. (3) Teachers were told specifically not to do brainstorming or to discuss what the students might write about before they began writing. Although brainstorming ideas is an excellent technique for teaching writing, the distinction was made in this project between teaching writing and assessing writing. Students, therefore, did not have the benefit of other students ideas for this assessment. (4) Teachers told the students that they could do a rough draft, make an outline, and use a graphic organizer or other tool to first put their ideas on paper. (5) Teachers told students that they should then turn their rough draft into the final paper to be turned in. (6) Teachers were asked to remind students at the 15-minute mark so that they would have time to write a final draft. (7) At the end of 30 minutes, teachers collected the rough drafts (for students who chose to do them) and the final writing samples. (8) Fourth- and fifth-grade teachers who gathered writing samples in both Spanish and English were asked to collect writing samples in Spanish first, wait one week, and then gather the data in English.
We collected 409 writing samples in the fall and 353 in the spring from students in grades 1-5. We collected writing samples in Spanish only from students in grades 1-3 who were learning to read first in Spanish. We collected writing samples in English only from students in grades 1-3 who were in monolingual English classrooms. We collected writing samples in both English and Spanish for students in grades 4 and 5. Of the 364 writing samples collected, 110 were from fourth- and fifth-grade students who were making transitions from Spanish literacy instruction to English literacy instruction. Again, we did not advocate for the use of writing prompts by teachers, but we did recognize the increasing use of such prompts in state standardized tests. We also realized the value of using the writing samples as a tool for teachers to talk about their perceptions of students writing skills and development. We gave the first prompt in the fall (My best birthday ever). After some discussion with teachers about the prompt, we created a second prompt (If I could be someone else for a day) that was administered in the spring. All the students had 30 minutes in which to write.
It was decided to use a different prompt between the fall and spring for several reasons. Teachers at the school expressed a concern, often shared by researchers, that using the same forms of assessment over and over again creates test familiarity. If students are familiar with the testing itemsin this case, the prompttheir outcomes may be more reflective of test familiarity than of any improvement in their actual writing skills. Teachers wanted to keep the genre the same from fall to spring, but they wanted a different prompt. Each of the prompts requires a narrative style of writing, but the specific prompt was changed to accommodate the concern about test familiarity.
During the fall and spring, we structured four professional development opportunities (two in the fall and two in the spring) for teachers, teacher candidates, the principal, and the research team to meet to score and discuss the results of these writing samples and the possible instructional implications. Professional development opportunities included two half-day meetings (from noon to 3:30) and two meetings conducted after school (from 3:30 to 5:30). Meeting agendas included the following topics: (1) training teachers in how to use the writing rubric to assess students writing development; (2) identifying anchor papers at each grade level; (3) having teachers read and rate writing samples in order to achieve interrater reliability; (4) compiling results and comparing scores from the fall with those from the spring to assess student growth in writing in English and Spanish; and (5) examining student writing samples to assess skills, strategies, and knowledge that students were transferring across languages (see Coady & Escamilla, 2004; Escamilla & Coady, 2001). Writing assessments were scored holistically using the Escuela Brillante Spanish/English writing assessment (Adams County School District, 1996).
The Escuela Brillante writing assessment is based on a 7-point scale that looks at student writing development on a range from 1 (not competent) to 7 (superior). Each category has a narrative description of the writing behaviors that are characteristic of students at that level. For example, the level-5 descriptor is highly competent, and the descriptor says, The student can compose a series of ideas about a topic with basic skills at a level for the grade and with descriptive vocabulary and complicated sentence patterns. The paper does not contain characteristics of superior writing such as insight, fluency, creativity or vitality and richness of expression (Adams County School District, 1996, p. 24).
One of the professional development sessions was devoted to creating interrater reliability. Before the session, each of the 18 teachers was given 10 papers to score independently using the Escuela Brillante rubric. Papers given to teachers represented a variety of writing levels and abilities. All 12 teachers had the same 10 papers to score. Teachers brought scored papers to the professional development session, and scores from the independent reviews were compared one to another. When there were wide ranges in scores (e.g., one reviewer rated a paper as competent but another rated the same paper as superior), discussions were held about the descriptors on the rubric and how to get at a shared understanding of what the descriptors meant. Widespread discrepancies in scoring either in English or Spanish were uncommon, and in general, there was a high degree of interrater reliability. After the initial session establishing interrater reliability, samples were scored during the professional development sessions. Finally, the researchers identified anchor papers that were thought to be representative of each writing level. These anchor papers were used by teachers to score and interpret the samples. Student names were removed from all writing samples prior to scoring and replaced with numbers.
During these sessions, we also asked teachers to discuss their perceptions about the strengths and needs of their students writing, both in Spanish and English, and we interviewed teachers individually about what they thought could be done to improve and enhance the writing of students at the school.
Teacher candidates and researchers individually interviewed each of the 18 teachers. A total of 10 open-ended interview questions explored the teachers formal training in teaching literacy in Spanish and English, along with their beliefs about writing instruction in Spanish and English. Teachers were asked about how they go about teaching writing and, more specifically, how they help students make transitions from writing in Spanish to writing in English. They were specifically asked to identify issues in writing development when children transfer from Spanish to English writing and reading programs.
FINDINGS AND DISCUSSION
There were three major findings regarding student emerging biliteracy and teacher perceptions of student development. Each is discussed below.
FINDING 1: TEACHER KNOWLEDGE AND THEORIES ABOUT ELL BILITERACY
This section summarizes teacher knowledge and theories about ELL biliteracy as it relates to the development of writing. As will be demonstrated in the discussion next, teachers in the study expressed contradictory beliefs about bilingualism and biliteracy. All the bilingual teachers, for example, expressed a strong belief that teaching monolingual Spanish-speaking children to read in Spanish first was beneficial to the development of literacy in English. Time and again, they expressed that Spanish literacy provided the basis for transfer to English. The monolingual teachers also demonstrated a knowledge of the transfer theory, and for the most part, they agreed that using a childs native language was beneficial for the ELLs. In discussing this idea of transfer across languages, teachers emphasized reading and not writing, and never discussed or used the word biliteracy. They were very familiar with theories about cross-language transfer but had little knowledge about the development of biliteracy.
Bilingual teachers in the study expressed frustration about their lack of opportunity to develop their own skills and knowledge in teaching reading and writing in Spanish and in teaching emerging biliterate students. They only had knowledge about how teaching reading and writing in Spanish may differ from teaching reading and writing in English. Monolingual teachers expressed that they need to know more about how writing in English for ELLs differs from writing for monolingual English children. None of the teachers in the study had formal training in how literacy development in English for ELLs may be different than for monolingual English children.
When discussing the childrens writing, however, the majority of teachers in the study were dismayed at how poorly the children wrote in both Spanish and English. They believed that the childrens writing in Spanish was particularly problematic and expressed a belief that Spanish was easier to learn to write than English. The overall perception was that writing in Spanish is easier than writing in English. Further, because teachers perceived that students were poor writers in Spanish, they concluded that they had few skills in Spanish to transfer to English. Thus, they concluded that many of the students writing issues in English were due to few skills in Spanish to transfer to English.
Given the above, it would be easy to conclude that lack of appropriate teacher preparation led to persistent teacher belief in deficit explanations of bilingualism and emerging biliteracy. Data presented next do provide evidence that teachers in the study tended to discuss students writing development in deficit terms, and they did express concern that the students were not strong writers in either Spanish or English. However, during the course of the professional development sessions, other trends became clear and merit discussion in this article.
During interview and focus group sessions, teachers expressed frustration that the district provided a plethora of staff development opportunities in English literacy instruction, but not one of the teachers in the study had ever had an opportunity to attend a staff development session that was conducted entirely in Spanish on the topic of teaching of literacy in Spanish. Although 6 of the 12 bilingual teachers were native Spanish speakers, only one had formal education courses in Spanish (she immigrated to Colorado from Mexico). All the others had graduated from U.S. high schools and colleges and received all their teacher education courses, including literacy methods, in English. Further, they had all passed the district Spanish-language proficiency test required to be a designated bilingual teacher.
Studies by Guerrero (1997), Calderon and Diaz (1993), and Wink and Flores (1992) confirmed that the majority of bilingual teachers nationally reported that their university course work was rarely delivered in Spanish and that methods classes, including literacy methods, were taught in English. These same studies argue that bilingual teachers can pass Spanish proficiency tests but frequently report that they do not feel competent to deliver academic instruction in Spanish. Added to this, the monolingual English teachers had had very little opportunity to learn about ELL literacy development. Three reported having attending in-services over time, and 3 reported learning from their peers but not having formal training.
The fourth- and fifth-grade bilingual teachers in the study expressed the frustration that they had less training in second-language acquisition than all the other teachers in the school. In fact, the district ELA court order (DPS, 2004b) mandates that transition teachers (termed ELA-T) need only 60 hours of in-service training to be fully endorsed, whereas 150 hours are required for all other teachers. The fourth- and fifth-grade teachers in the study were all designated ELA-T teachers and expressed the frustration that they were ill prepared and unsure about how to teach literacy in Spanish; they were also uncertain about appropriate methods to use during transition from Spanish to English. Some researchers have argued that teachers at the transition stages need the most training to help students transfer skills and strategies across languages (Berman et al., 1992; De Avila, 2000; Gersten, 1996; Saunders, OBrien, Lennon, & McLean, 1998). Teachers at this school who were teaching transitional students had the least.
Data from professional development sessions and teacher interviews revealed that many teachers described themselves as people whose knowledge of English and Spanish was limited. They expressed concern that they did not know how to do the very things that they wanted their students to do. For example, many said that they did not know how or when to teach accent rules in Spanish literacy, nor did they feel comfortable having literary discussions in Spanish. They also did not know how to help develop students literacy in English as a second language. Again, without using the word, they described themselves as having a form of bi-illiteracy.
Further, all teachers reported that the literacy program, in general, focused on the teaching of reading, not writing, and that both reading and writing focused on what was considered to be a generic model of good practice. Teachers discussed their perceptions of how reading should be taught in English and Spanish. Most confessed that they put much less emphasis on writing than reading. Most notably, the bilingual teachers discussed beliefs about how reading and writing should be taught in Spanish and English, but none discussed literacy development through a biliteracy perspective. For the most part, teacher responses to interviews presented a view that Spanish skills could transfer to English and could provide the foundation for successful literacy acquisition in English, but they observed that student writing samples demonstrated negative transfer from Spanish to English.
FINDING 2: TEACHER ASSESSMENT OF ELL BILITERACY THROUGH WRITING SAMPLES
Findings from this study indicated that early teacher assessment in this study was characterized by a focus on student deficits (what the students did not do correctly) and the attribution that these deficits were a result of interference or negative transfer from Spanish to English. Over time, however, as teachers engaged in more in-depth conversations about these writing samples, teacher assessment evolved and increasingly focused on emerging biliteracy.
Monolingual English teachers independently read writing samples that were written in English, and bilingual teachers read some written in Spanish and some written in English. Again, they initially read the papers independently and then discussed their observations in a group. Bilingual teachers were asked to explain their observations and summarize the Spanish writing samples to their monolingual peers. After reading, scoring, and summarizing the results of the administration of each writing prompt, teachers were asked to discuss students writing development in both English and Spanish. To do this, teachers were asked to forget about the overall score and look at what could be learned from how the students wrote and what they chose to write about. This finding indicated three general trends. First, teachers focused on talking about students writing weaknesses rather than strengths. They emphasized student weaknesses by focusing their comments on conventions, mechanics, and organization rather than content or voice. Finally, there was overall consensus that lack of background knowledge (not knowing what to write about) was a significant issue in childrens writing in both languages. The following examples of student writing in Spanish serve to illustrate the basis upon which teachers came to these conclusions. All samples represent student writing to the spring prompt, If I could be someone else for a day.
Ana Barbara - Ganaria mucho dinero y me iria a repartir ese dinero a las personas pobres (Ana Barbara - I would make a lot of money and I would give that money to poor people).
Campos - alludar ajente pobre o mala para comprar una mansion para mi y otra para mi mama (Campos - to help poor people or sick people to buy a mansion for me and another one for my mother).
Gobernador de Denver colorado -yo seria muy rico y ayudaria alos pobres . . . ayudar ami papa y ami mama (Governor of Denver - I would be very rich and I would help poor people . . . to help my father and my mother).
Policia - para ensenar alos niños sobre la biolensia para que se los fueran aprendiendopoco apoco sobre eso . . . y para cuando sepresente algun caso de que el esposo anda tomado y que golpio a su mujer y que se escape y se yebo a sus hijos dela casa (Policeman - To teach children about violence so that they could be learning little by little about this . . . and so that when a case comes up where a husband is drunk and hits his wife so that she can escape and take the children).
Policia - para alludar a las personas y poner orden en la ciudad . . . arestaria a los rateros que roban y los pone en la carsel por muchos años de prision (Policeman - to help people and to help keep the city safe . . . I would arrest robbers and I would put them in jail for many years).
Maestra - ayudaria a los niños de calle les hace falta comida, ropa y sapatos. Aunque yo no tubiera dinero lo consigueria para ayudarlos (Teacher - to help the street children who do not have food, clothes, or shoes. Even though I might not have money, I would look for some way to help them).
Teachers expressed concern about student spelling errors, especially of words that they considered to be easy (e.g., povre, alludar, jente, yebo), their failure to separate words (alos, ami, dela), and especially their lack of knowledge of accent marks. They also expressed concern over lack of knowledge of conventions (e.g., punctuation, use of commas), and the use of long run-on sentences. There is ample evidence to support teachers concerns about student writing skills. Nonetheless, it is also important to note the general absence of dialogue about student strengths, particularly in the areas of expression of complex ideas and voice. Contrary to teacher beliefs that students lacked background knowledge, student writing indicated strong evidence that students in the study had deep knowledge of the social and economic realities of their communities and, further, had plenty of ideas for how to improve these conditions (see Coady & Escamilla, 2005, and Escamilla & Coady, 2001, for a more detailed analysis of the content of students writing in this study). Students in the study were able to express complex and sophisticated ideas, but teachers tended to discuss student writing in terms of their shortcomings and not their strengths. In short, teachers examination of Spanish writing samples led them to further conclude that these children were weak writers in Spanish.
In addition, early teacher assessment focused on the idea that writing issues in English could be attributed to negative transfer from Spanish. During interviews and professional development sessions, all teachers, both bilingual and monolingual, expressed the knowledge that skills and strategies learned in one language transfer to another. They understood that this was a rationale for the ELA program and for teaching children to read and write in Spanish. However, during the course of the project, it became apparent that the perception of a positive relationship between Spanish and English literacy held for reading but not for writing. As teachers examined the English writing samples of children in the study, the majority expressed concern that the problem with English writing was Spanish interference, particularly in the areas of syntax, spelling, and word endings. The writing samples below illustrate teachers beliefs. The first student wrote that she would like to be her teacher for a day; the second that she would like to be Lisa Simpson, the cartoon character; and the third student that she would like to be her mother.
Ms. S - I would be ms. S. She is good ticher. Sometime she get mad with us. I wish the class could pay attencion and behave so that ms. S. could be happy.
In the case of the first student, teachers observed that many of her spelling issues illustrated her use of Spanish sounds to write English words (e.g., ticher, wich, attencion). They were concerned with her use of Spanish syntax (e.g., mad with us) and her failure to hear the ending s in English words (e.g., sometime, get). They viewed her writing issues as caused by Spanish.
I would like to be Lisa Simpson. Lisa is my favorite cartoon character because is intelligent and so cool. She know everything in math and science. And she never get mad at nobody she is so cool. I would like to help kids to finich the school. And help people that need a job.
Similarly, in the case of the second student, teachers cited her omission of she (because is), her use of a double negative (never get mad at nobody), and her use of phrases like to finich the school, as signs of Spanish interfering with English.
May Mom Sara - o I wich tha I was my mom. Because shes so nice and perty olove my mom because shes nice to me and she lake my best f rind. Pore thing of my mom because she had es and fid us and whor for us and fold for es and my dad work for es and by stoffore us and by ore fode and care fore es, whim and my mom.
As with the other two students, teachers viewed this student as having writing problems that stem from her first language. For example, misspelled words such as wich (wish), whor (work), stof (stuff) and es (us) were attributed to using Spanish sounds to spell English words. Further, phrases such as pore thing of my mom were attributed to Spanish syntax, and the insertion of her dads role in caring for the family as a sign of Spanish digressive logic, causing her to get off topic. As with examination of Spanish writing samples, further examination of English writing samples illustrated that teachers believed that students were weak writers in English and Spanish, and that most of their writing issues were caused by negative transfer from Spanish. Again, the focus of these discussions was on childrens perceived weaknesses in writing, not on their strengths or the power of their ideas. Without using the word, teachers statements had added support to notions of bi-illiteracy. This finding was consistent for both monolingual and bilingual teachers in the study. In short, all teachers had perceptions that interference (negative transfer) was a major problem in the writing development of these children, and all teachers focused more on childrens writing weaknesses rather than their strengths.
Teachers are not alone in their perceptions that literacy in one language interferes with another. Black and Valenzuela (2004) reported that the Texas Education Agencys accountability system, the Reading Proficiency Test in English (RPTE), encourages schools to prepare students for passing the English Texas Assessment of Academic Skills (TAAS) and states that RPTE ratings of advanced are intended to indicate that with another year of effective instruction, the students can be expected to engage in standardized testing in English with minimal language interference.
An important finding in this study was that over time, teacher discourse moved beyond a focus on bi-illiteracy and toward discussions about emerging biliteracy. Evidence for this finding is presented in the following summary of the discussion from the third professional development session. One of the teachers in the group received her initial teacher education license from Mexico, and she regularly used teaching materials from Mexico in her literacy program. At this meeting, she raised several issues that had been troubling her since the beginning of the study. She asked simply how we knew that the two writing prompts in English and Spanish were equivalent simply because we asked children to do the same type of writing in Spanish and English. Initially other teachers in the group (and the researchers) were somewhat dismissive of her question, yet as she persisted, we had a collective aha.
We discovered that the writing prompts in English and Spanish were the same but were definitely not equal. Both writing prompts asked children to write a narrative story. Both required that the children write in the past tense. On the surface, these were seemingly equal tasks. However, when examining the two languages, it became apparent that the two tasks are hardly similar. In English, the formation of the past tense is relatively simple; that is, one simply adds ed to the end of a word. Of course, there are many exceptions, but the general rule is relatively simple. In Spanish, the task is not so simple because all past-tense expressions, whether preterit or imperfect, require the use of accent marks. Our writing rubric specified that words that required accent marks and did not have them were to be considered as misspelled words. As we examined the childrens writing, it became obvious that one of the major reasons that many children who wrote in Spanish were deemed marginal writers was that they did not know accent rules, particularly in using the various past tenses; yet, the task required them to know these rules. As we talked with teachers, we concluded that nascent understanding of accent rules in Spanish is most likely rather typical for fourth- and fifth-grade Spanish writers. Further, when we consulted the Mexican National Curriculum for teaching language arts (Secretarıa de Educacion Publica, 1995), we discovered that children in Mexico are not expected to master accent rules related to past tenses until sixth grade. Rather than being limited-Spanish writers, they were most likely developing at a typical pace.
In our conversation with teachers, it became apparent that the Spanish reading program at the school did not have specific points in the curriculum to teach children accent rules. Similarly, many teachers told us that they themselves were unsure of accent rules and how to teach them in Spanish. Teachers concluded that they were holding children responsible for mastery of content that they had never been taught. Further, teachers expressed a need to have specific staff development sessions on teaching literacy, especially writing, in Spanish. In short, teachers wanted to improve their skills and expertise in teaching literacy in Spanish. Teachers expressed a desire to further develop their own biliteracy. Monolingual teachers in the group expressed that these concepts were things that all teachers should know; even if they didnt know Spanish, they should be more suspicious of translated tasks.
Similarly, as we examined and reexamined childrens writing, it became apparent that when examining the writing of emerging biliterate children, we were too quick to blame Spanish. Upon closer examination, it was apparent that English writing of Spanish-speaking students was, in part, similar to writing issues that any 9- or 10-year-old might have. An example of a childs writing (Figure 1) and our analysis of it illustrate this point.
This childs writing shows evidence that he is doing some of the things that all fourth-graders do and some things that only Spanish speakers do. For example, native-English-speaking children frequently misspell words like because (becuase), earned (erned), and graffiti (grofitey). They write long run-on sentences, make mistakes such as me and my men instead of my men and I, and confuse homonyms such as there and their. That said, structures such as will built a school for the kids could go to most likely come from Spanish. We need to be able to separate what typical monolingual English fourth graders do in writing from what Spanish speakers do when they write in English. By separating out typical writing issues from ELL writing issues, we were able to look at childrens writing with broadened perspectives. In fact, we agreed that if we were to do this analysis again, we would create a chart asking teachers to separate into two different columns errors that are common to the grade level and errors that are transferred from Spanish to English. Recent research in this area (e.g., Howard & Coburn, 2005) has concluded that intermediate ELL students writing in English in U.S. dual-language programs is characterized by equal numbers of errors that come from Spanish and that would be considered typical of native English speakers writing at a grade level.
Finally, and perhaps most promising, as bilingual teachers discussed the content of childrens writing with each other and with their monolingual peers, they observed that children at this school had many things to write about and much to say about their world. Their background knowledge and content was their strength, and they frequently made cross-cultural connections that might go unnoticed. The writing sample in Figure 2 is an example of this content knowledge.
The content of this childs writing is very interesting and could easily have gone unnoticed had it not been for one of the teachers who pointed out that the child was making a comparison between Martin Luther King, Jr. and Emiliano Zapata, a famous Mexican civil rights leader. Perhaps Emiliano Zapatas most famous statement was, Es mejor morir de pie que vivir de rodillas (Its better to die on your feet than live on your knees). The students line, I would stay on my feet until freedom rings, is quite a remarkable connection between a civil rights leader from Mexico and one from the United States. Writing samples such as this one helped us understand that the children did not lack background knowledge and expressed quite sophisticated ideas. In short, during this project, all of us became more astute and enthusiastic observers of childrens writing. We made growth in changing perspectives from bi-illiteracy to emerging biliteracy.
FINDING 3: ELL BILITERACY OUTCOMES
After the fall collection of data, the mean score on the Escuela Brillante writing rubric was, on a 5-point scale, 2.6 in Spanish and 2.2 in English. In the spring, the mean score was 3.2 in Spanish and 3.0 in English. Over the year, the majority of students had shown growth in their writing in both Spanish and English. However, the numeric scores from both the fall and the spring indicated that the majority of students were not competent writers in either Spanish or English. The Escuela Brillante writing rubric describes writing scores in the 2-3 range as indicating beginning writing significantly deficient in several skill areas. Writers in the 3-4 range are considered marginally competent and deficient in one or more skill areas. Given these outcomes, it is not surprising that teachers and others at the school expressed the belief that students were poor writers in both English and Spanish, and the yearlong research project had only served to confirm with empirical data what they had suspected all along.
Interview data and data collected during discussions at the professional development sessions indicated that teachers had a range of perceptions as to why the students scored so low on the writing rubric. Reasons given for student performance included a strong belief on the part of bilingual teachers that the children had been rushed to English literacy without having had enough time to fully develop their literacy skills in Spanish. Other teachers argued that writing had never been a focus of the school literacy instruction, and there had been too much focus on reading and none on writing. All teachers, bilingual and monolingual, added that the background of the children (poverty, limited life experiences, and so on) impeded their writing development.
It is significant to note that at the projects inception, neither the teachers nor the researchers raised a concern about the use of a writing rubric that had been developed in English and then reconstructed into Spanish as being a possible problem in looking at writing development of developing bilinguals. In short, the school staff believed the numbers, and the researchers had legitimatized through the study that language was a problem and that biiliteracy was a fact. Black and Valenzuela (2004) argued that studies with this type of design employ the systematic use of valid, scientific instruments to produce legitimate knowledge to justify student deficits. Unfortunately, the design of the study and the focus on the numeric outcomes had served to affirm notions of bi-illiteracy among the staff at the school. The researchers, rubrics, and teacher beliefs and perceptions were part of the problem.
CONCLUSIONS AND IMPLICATIONS FOR SPECIAL EDUCATION FOR ELLS
Researchers in the field of bilingual special education (Artiles & Ortiz, 2002; Baca & Cervantes, 2004; Ortiz & Rivera, 1990) advocate that bilingual education that uses a childs first language in instruction, particularly initial literacy instruction, can positively impact student academic achievement and can possibly reduce the number of inappropriate referrals and placements of ELL students in special education. The use of L1 for initial literacy instruction is thought to be a particularly powerful intervention.
In this study, bilingual and monolingual teachers were asked to assess writing samples of children learning to write in both Spanish and English. They were asked to assess the quality of the childrens writing and discuss what they perceived to be the childrens strengths and needs. It is significant to note that both bilingual and monolingual teachers perceived that children were not good writers in either Spanish or English and that they had little to write about. It is also significant to note that teachers perceived that English spelling, syntax, and grammar problems were caused by Spanish.
Findings from this article suggest that both bilingual and monolingual teachers want to be effective advocates and teachers for their ELL students, but in many cases, their preparation as teachers has limited their views of emerging bilinguals.
I would submit that if the use of L1 literacy instruction will in fact increase academic achievement and reduce inappropriate referrals of ELLs to special education, then we must consider significant changes in our teacher education and professional development programs.
First, we must include in these programs training for all teachers that expands notions of cross-language transfer beyond the interference or negative transfer paradigm to more sophisticated notions of biliteracy development. As noted previously, there is a dearth of research in this area.
Next, we must help teachers and researchers understand that simply assessing a student in his or her first language without attention to cross-linguistic differences is unlikely to yield information that is more illustrative of a childs abilities even in contexts in which children are learning to read and write in two languages.
Of greatest concern here is the potential for teachers to conclude that children are underachieving in Spanish, that L1 instruction has not been effective and therefore should be eliminated from the curriculum, and/or that L1 instruction impedes learning literacy in English.
Bernhardt (2003) said that the most pressing issue for reading instruction is the preparation of teachers to ensure that they have the knowledge or skill to diagnose and assess childrens progress. Assessment does not mean high stakes testing but the everyday set of judgments that teachers make about the progress of individual children (p. 39). This study illustrated to us the need to help teachers and researchers make different judgments about the literacy development of emerging biliterate students. Findings from this study concur with those of Bernhardt in that we need to develop a theory of literacy for bilingual students not based on monolingual English theory. For example, we need to help teachers understand that many of the perceived writing problems of students, such as those in this study, were more likely typical of second-language learners than signs of low levels of development in both languages. We further have to help teachers see beyond conventions and mechanics and look at the content of childrens writing.
Teachers in this study, just as those in national studies (Guerrero, 1997), expressed a need and a desire to further develop their own biliteracy. It is doubtful that writing instruction in Spanish will improve until bilingual teachers are confident in their ability to teach literacy in Spanish and to help students make transitions to biliteracy.
Finally, it must be noted that some programs, such as the ELA program in this study, are riddled with terms and assumptions that label language as the problem (Ruiz, 1988). The label of the ELA program in this study places no value on the development of biliteracy or bilingualism. In fact, the expressed goal of the ELA program is acquisition of English (DPS, 2004b). Valenzuela (2004) concluded that it is difficult to encounter discussions anywhere in the United States of developing policies that focus on developing bilingualism or biculturalism as either a private or public asset, including their potentially significant contribution to academic achievement. It is in this increasingly hostile environment that we must work to transform teacher perceptions of the literacy behaviors of second-language children from bi-illiterate to emerging biliterate.
As a postscript to this study, it is encouraging to report the results on the Colorado Student Assessment Program (CSAP) for spring 2003 for the school in the study. At this school, third-grade results indicated that 76% of the students who took the CSAP in reading in Spanish were considered to be proficient or advanced, and in writing, 71% of the children were reported to be proficient or advancedencouraging signs that teachers are paying attention to writing and reading, and to laying the foundation for biliteracy.
Adams County School District No. 27J. (1996). Escuela Brillante Writing Assessment. Brighton, CO: Author.
Artiles, A., & Ortiz, A. (2002). English language learners with special education needs: Identification, assessment and instruction. McHenry, IL: Center for Applied Linguistics, Delta Systems Co.
Baca, L., & Cervantes, C. (2004). The bilingual special education interface. Upper Saddle River, NJ: Pearson.
Bartolome, L., & Balderrama, M. (2001). The need for educators with political and ideological clarity: Providing our children with the best. In M. Reyes & J. Halcon (Eds.), The best for our children: Critical perspective on literacy for Latino students (pp. 48-64). New York: Teachers College Press.
Berman, P., Chambers, J., Gandara, P., McLaughlin, B., Minicucci, C., Nelson, B., et al., (1992). Meeting the challenge of language diversity: An evaluation of programs for pupils with limited English proficiency (Executive Summary, Vol. I). Berkeley, CA: B.W. Associates.
Bernhardt, E. (1991). Reading development in a second language: Theoretical, empirical, and classroom perspectives. Norwood, NJ: Ablex.
Bernhardt, E. (2000). Second language reading as a case study of reading scholarship in the 20th century. In M. L. Kamil, P. B. Mosenthal, P. D. Pearson, & Z. R. Barr (Eds.), Handbook of reading research: Volume III (pp. 791-811). Mahwah, NJ: Erlbaum.
Bernhardt, E. (2003). Challenges to reading research from a multilingual world. Reading Research Quarterly, 38(1), 112-117.
Black, B., & Valenzuela, A. (2004). Educational accountability for English language learners in Texas: A retreat from equity. In L. Skrla & J. Scheurich (Eds.), Educational equity and accountability: Paradigms, policies and politics (pp. 215-234). Albany: State University of New York Press.
Calderon, M., & Diaz, E. (1993). Retooling teacher preparation programs to embrace Latino realities in schools. In Reshaping teacher education in the southwest: A response to the needs of Latino students and teachers (pp. 51-70). Claremont, CA: Tomas Rivera Center.
Coady, M., & Escamilla, K. (2005). Audible voices and visible tongues: Exploring social realities in Spanish speaking students writing. Language Arts, 82, 462-471.
Cummins, J. (2001). The role of primary language development in promoting educational success for language minority students. In California State Department of Education (Ed.), Schooling and language minority students: A theoretical framework (pp. 3-49). Los Angeles: Evaluation, Dissemination and Assessment Center, California State University.
De Avila, E. (2000). Teaching for transfer. Paper presented at the National Association for Bilingual Education (NABE) conference, San Antonio, TX.
Denver Public Schools. (2004a). Denver Public Schools: School ratings. Retrieved November 5, 2005, from http://www.dpsk12.org/general/assessment
Denver Public Schools (2004b). Denver Public Schools: English language acquisition program. Retrieved November 5, 2005, from http://www.ela.dpsk12.org
Dubin, F., Eskey, D., & Grabe, W. (1986). Teaching second language reading for academic purposes. Reading, MA: Addison-Wesley.
Escamilla, K., & Coady, M. (2001). Assessing the writing of Spanish-speaking students. In S. Hurley & J. Tinajero (Eds.), Literacy assessment of second language learners (pp. 43-63). Boston: Allyn & Bacon.
Escamilla, K., & Coady, M. (2003). Beyond skills and strategies: Assisting Spanish speaking students in
their transition to English literacy. Paper presented at the National Association for Bilingual Education, New Orleans, LA.
Fitzgerald, J. (1993). Literacy and students who are learning English as a second language. The Reading Teacher, 46, 638-647.
Fitzgerald, J. (1995a). English as a second language learners cognitive reading processes: A review of research in the United States. Review of Educational Research, 65, 145-190.
Fitzgerald, J. (1995b). English as a second language reading instruction in the United States: A research review. Journal of Reading Behavior, 27, 115-152.
Garcıa, G. E. (2000). Bilingual children’s reading. (In M. L. Kamil, P. B. Mosenthal, P. D. Pearson, & Z. R. Barr (Eds.), Handbook of reading research: Volume III (pp. 813-834). Mahwah, NJ: Erlbaum.
Gay, G. (1995). Mirror images on common issues: Parallels between multicultural education and critical pedagogy. In C. Sleeter & P. McLarend (Eds.), Multicultural education, critical pedagogy and the politics of difference (pp. 155-189). Albany: State University of New York Press.
Gersten, R. (1996). Literacy instruction for language minority students: The transition years. Elementary School Journal, 96, 226-244.
Grant, R., & Wong, S. (2003). Barriers to literacy for language minority learners: An argument for change in the literacy education profession. Journal of Adolescent and Adult Literacy, 46, 386-394.
Guerrero, M. (1997). Spanish academic language proficiency: The case of bilingual education teachers in the U.S. Bilingual Research Journal, 21, 1-18.
Halcon, J. (2001). Mainstream ideology and literacy instruction for Spanish speaking children. In M. Reyes & J. Halcon (Eds.), The best for our children: Critical perspectives on literacy for Latino students (pp. 29-47). New York: Teachers College Press.
Hedgcock, J. (2005). Taking stock of research and pedagogy in L2 writing. In E. Hinkel (Ed.), Handbook of research in second language teaching and learning (pp. 597-613). Mahwah, NJ: Erlbaum.
hooks, b. (1993). Transformative pedagogy and multiculturalism. In T Perry & J. Fraser (Eds.), Freedoms plow: Teaching in the multicultural classroom (pp. 91-97). New York: Routledge.
Howard, E., & Coburn, C. (2005). A developmental investigation of cross-linguistic spelling errors in Spanish/English bilingual student. Washington, DC: Center for Applied Linguistics.
Luther, F. (1997). First Nations preservice women teachers experiences and perceptions regarding technology. Unpublished doctoral dissertation, University of Saskatchewan, Canada.
MacSwan, J., Rolstad, K., & Glass, G. (2002). Do some school-age children have no language? Some problems of construct validity in the pre-LAS Español. Bilingual Research Journal, 26, 395-420.
Montano-Harmon, M. (1991). Discourse features of written Mexican Spanish: Current research in contrastive rhetoric and its implications. Hispania, 74, 417-425.
Nieto, S. (1996). Affirming diversity. New York: Longman.
Odlin, T. (1989). Language transfer. Cambridge, England: Cambridge University Press.
Ortiz, A., & Rivera, C. (1990). AIM for the BEST: Assessment and intervention model for bilingual exceptional students (Contract No. 300-87-0131). Washington, DC: Office of Bilingual Education and Minority Language Affairs.
Perez, B. (1998). Sociocultural contexts of language and literacy. Mahwah, NJ: Erlbaum.
Ruiz, R. (1988). Orientations in language planning. In S. McKay & S. Wong (Eds.),L diversity: Problem or resource? (pp. 3-25). Cambridge, MA: Newbury House.
Saunders, W., OBrien, G., Lennon, D., & McLean, J. (1998). Making the transition to English literacy successful: Effective strategies for studying literature with transition students. In R. M. Gersten & R. T. Jimenez (Eds.), Promoting learning for culturally and linguistically diverse students (pp. 99-134). Belmont, CA: Wadsworth.
Secretaria de Educacion Publica. (1995). Español, Grados 1-6. Mexico City: Secretaria de Educacion Publica.
Smith, P., Jimenez, R., & Martinez-Leon, N. (2003). Other countries literacies: What U.S. educators can learn from Mexican schools. The Reading Teacher, 56, 772-781.
Valadez, C., MacSwan, J., & Martinez, C. (2001). Toward a view of low achieving bilinguals: A study of linguistic competence in designated semilinguals. Bilingual Review, 25, 281-291.
Wiley, T. (2005). Second language literacy and biliteracy. In E. Hinkel (Ed.), Handbook of research in second language teaching and learning (pp. 529-544). Mahwah, NJ: Erlbaum.
Wink, J., & Flores, J. (1992). Induction programs for bilingual teachers: Addressing the needs of teachers in language minority education. The Journal of the Association of Mexican American Educators, pp. 73-80.