Differences in Beginning Special Education Teachers: The Influence of Personal Attributes, Preparation, and School Environment on Classroom Reading Practices
Bishop, Anne G., Brownell, Mary T., Klingner, Janette K., Leko, Melinda M., Galman, Sally A. C., Learning Disability Quarterly
Abstract. Little research exists to help us understand why some beginning special education teachers of reading engage in more effective classroom practices than others. Factors that may influence these differences include personal attributes, preparation, and school environment. This mixed-methods study examined beginning special education teachers (N = 25) who taught reading to elementary students. Teachers were identified as most accomplished, moderately accomplished, and least accomplished, as defined by an overall classroom practice score. Interviews, observational field notes, and survey data on preparation and work environment revealed that the most accomplished beginners were consistently reflective, resourceful, and relentless and used these attributes to improve instruction, whereas others varied in this regard. Furthermore, while adequately prepared in special education, beginners reported inadequate preparation in reading. The interplay of personal attributes, preparation, and school environment seems to be a powerful determinant of a teacher's level of accomplishment.
Beginning teachers are often portrayed as struggling: scrambling to deal with the complexity of teaching and left to their own devices to maneuver a complex school environment. Specifically, beginning teachers often have trouble (a) with student discipline, (b) engaging
their students in the curriculum, (c) meeting their students' individual needs, (d) figuring out what to teach and how, and (e) navigating the school environment (Reynolds, 1995). In addition, many appear focused on themselves and their own survival rather than on making their instruction more responsive to students. In many cases, their practice can seem mechanical and unresponsive to students (Kagan, 1992).
Not all beginning teachers experience the same struggles in their practice, however, and some appear better equipped than others to deal with the demands of teaching. In special education, three studies have shown marked differences between the instructional practices of beginning special education teachers, with some teachers demonstrating remarkable sophistication in their ability to engage and effectively teach students with disabilities (Brownell et al., 2009; Nowacek, McKinney, & Hallahan, 1990; Seo, Brownell, Bishop, & Dingle, 2008).
At a time when teacher quality is receiving much national attention (e.g., the No Child Left Behind Act, U.S. Congress, 2001; the Individuals with Disabilities Education Improvement Act, U.S. Department of Education, 2004), we need to understand why some beginning special education teachers engage in more effective classroom practices than others. Such understanding of special education teachers of reading is especially important given the preponderance of students with reading disabilities served and the numbers of special education teachers who are likely to enter the classroom through alternative routes.
Most of the research examining beginning special education teachers has examined challenges these teachers face and workplace supports they need to encourage them to stay (Billingsley, 2004; Billingsley, Griffin, Smith, Kamman, & Israel, 2009). Less is understood about the factors that may contribute to differences in their practices. In general education, there is a healthier literature base explaining how differences in individual teacher qualities, initial preparation, and school context explain variations in beginning teacher practice. We acknowledge that some of these findings may apply to beginning special education teachers, but we can only surmise that the complexity of special education makes the struggle to enact their practice more exaggerated. Such complexities include applying their broad-based teacher training to address a multitude of student needs and the multiple roles they are likely to assume in schools (Blanton, Sindelar, & Correa, 2006).
To better understand why some beginning special education teachers were more adept in their practice than others, we selected teachers from a larger study who were deemed as more or less effective in teaching reading to students with learning disabilities. We built on this larger study to further our field's understanding of beginning special education teachers, their practice, and influences on their practice. In establishing a rationale for the study, we analyzed literature in general and special education that demonstrates the relative importance of three major clusters of variables in influencing beginning teacher practice: personal attributes, teacher preparation, and school environment.
Factors That Influence Classroom Practices
Personal attributes. Personal attributes and unique experiences of beginning special education teachers are likely to play a powerful role in shaping their classroom practices as well as their understanding of teaching and learning. Personal attributes include teacher beliefs and individual personal traits. It is well established that one's prior experience as a student in elementary and secondary schools has a strong influence on classroom practice (Richardson, 1996). Students of teaching bring powerful, personal history-based lay theories about classroom practice to their teacher education experiences (Holt-Reynolds, 1992). Such theories or beliefs about teaching, learning, and students, in turn, play strong mediating roles in what novices learn about teaching and how they apply it in the classroom (Richardson, 1996). In special education, however, entering beliefs may not play the same role in classroom practice, as prospective teachers have spent little time in a special education setting (Pugach, 2005).
One individual personal trait that is prominent in the teacher education literature is a propensity for reflection. Reflective teachers consistently analyze their teaching practices and search for ways to better help their students (Ross, Bondy, & Kyle, 1993). Reflective teachers are introspective, open-minded, responsible, and have the "ability to view situations from multiple perspectives" (Ross et al., 1993, p. 48; Zeichner & Liston, 1996).
Although reflecting on their practice should enable teachers to improve their practice, barriers such as time constraints, fear of being judged, weak knowledge of the content and how students might learn it, and the difficulty of analyzing one's own actions can hinder teachers' abilities to reflect (Pickett, 1996; Turner-Bissett, 1999). Reflection continues to be an area of interest for researchers, as the ability to reflect is a common trait among accomplished teachers (Loughran, 2002).
Individual teacher traits also play an important role in how teachers respond to the challenges they confront when entering the special education classroom. General education teachers, working in inclusive classrooms, are more likely to engage in cognitively demanding instructional interactions and provide more active instruction when they perceive themselves as being able to assist students with disabilities instructionally compared to their less efficacious peers (Stanovich & Jordan, 1998).
The role that personal attributes play in the learning and practices of special education teachers, however, has not been subject to much study to date (Renzaglia, Hutchins, & Lee, 1997). Lessen and Frankiewicz (1992) reported that successful special education teachers displayed humor, enthusiasm, fairness, empathy, flexibility, and self-control. In addition, they maintained good relations with individuals and groups of students. Teachers' ability to forge positive relationships with students with disabilities may be due to their level of self-efficacy with regard to their competence as teachers. Thus, in a large survey study of special education teachers, Carlson, Lee, and Schroll (2004) identified self-efficacy as a key dimension of teacher quality.
Beginning special education teachers' beliefs about instruction, their beliefs about their role in helping students with disabilities, their ability to reflect on their instruction, and their confidence in their ability to work with students with disabilities, are likely to influence their ability to learn from what they are doing. If beginning special education teachers are able to identify the instructional improvements they must make and are confident that they will be able to make them, their instructional practices are likely to be stronger.
Teacher preparation. Early teacher training represents the integration of beliefs, knowledge, and practices of beginners (Wilson, Floden, & Ferrini-Mundy, 2001). What teachers learn prior to entering the classroom makes a difference in their success. The quality and intensity of as well as the components that are incorporated into the learning directly influence a beginner's ability to teach. While it is often difficult to separate preparation experiences from resulting knowledge and beliefs, we think it is important to examine preparation as a separate entity, acknowledging that preparation experiences are key in acquiring the specialized knowledge necessary for effective classroom practices in reading.
The general education literature includes a series of studies creating linkages between teacher education, teacher practice, and student achievement specific to reading. Researchers have demonstrated that exemplary teacher education programs have an advantage over generic elementary programs in preparing beginning teachers to provide reading instruction. Graduates of exemplary teacher education reading programs reported being more responsive to the students' needs and demonstrated more effective classroom practice using a validated observation system (International Reading Association, 2003; Maloch et al., 2003). This work supports the importance of high-quality preparation to promote beginning teacher quality.
Thus far, four studies have investigated the contribution of special education preparation to beginning teaching. For example, Sindelar, Daunic, and Rennells (2004) examined the practices of beginning special education teachers who participated in three preparation routes, using a validated observation system, Praxis III (Dwyer, 1998). These researchers found that teachers graduating from campus-based and alternative programs run through a district-university collaborative were more competent than those graduating from a district-only alternative program.
Nougaret, Scruggs, and Mastropieri (2004) also found that preparation plays a role in beginning special education practice, as teachers in their study who received minimal preparation were less sophisticated in their practice than teachers graduating from a special education preparation program. Feng and Sass (2009) provided preliminary data that demonstrate connections between preparation and student achievement. Specifically, these researchers found that special education teachers, with approximately 30 hours of preparation, made stronger contributions to the reading achievement gains of students with disabilities than teachers without such preparation.
Finally, Leko (2008) demonstrated that not all special education preservice teachers benefited equally from their preparation in special education and reading. Using qualitative interviews and observations as well as ratings on a validated observation tool, Leko found that the opportunities preservice teachers had to apply their knowledge of teaching reading to students with disabilities, as well as their individual beliefs about teaching reading and their competence for teaching, seemed to work together to influence the quality of their instruction.
These studies demonstrate connections between teacher preparation and beginning special education practice. Further, the Leko study (2009) shows that variations in preparation experiences and individual qualities of beginning special education teachers are likely to shape teachers' practices.
School environment. External factors also come into play as beginners engage in classroom practice, and these factors may be pronounced for special education teachers given the complex challenges they encounter in schools. Environments marked by collaborative colleagues, mentorship, access to materials, and supportive administrators enable special as well as general education teachers to develop their skills. Across several studies, beginners described how collegial support was important in learning to teach. Support allowed teachers to take risks, provided new ideas for instruction, enabled teachers to secure feedback on instruction, and helped them to feel more self-efficacious about teaching and their influence in the school (Appleton & Kindt, 1999; Chester & Beaudin, 1996).
Johnson, Kardos, Kauffman, Liu, & Donaldson (2004) found evidence of strong instructional leadership in schools that inducted beginning teachers successfully. Principals played a key role in establishing supportive academic cultures. Effective principals hired teachers who understood the school mission and provided ongoing feedback aimed at validating and improving beginning teachers' practice. This helped teachers develop more certainty about their teaching and improved their self-efficacy (Chester & Beaudin, 1996). Additionally, effective principals played a strong role in crafting high expectations for behavior and learning and supported beginners in fulfilling those expectations. In schools with effective leadership, beginners believed they were not left to "sink or swim."
Access to quality curriculum also seems to influence a beginning teacher's ability to provide instruction. Research on beginners demonstrates that they often rely on textbooks to inform their instruction, even when they believe doing so constrains their creativity. Well-specified curricula that help teachers approach instruction and make decisions assist them in feeling secure about their teaching (Kauffman, Johnson, Kardos, Liu, & Peske, 2002).
In a qualitative interview with 50 first- and second-year teachers, Kauffman et al. (2002) found that teachers who had access to a well-articulated curriculum were likely to feel confident when approaching instruction. In contrast, when teachers did not have access to structured curriculum, they felt lost. The pressure to create curriculum occurred while they were learning to maintain discipline, engage students in classroom activities, navigate a new school context, communicate with parents, deal with state standards, and evaluate student work. In such situations, beginning teachers felt exhausted, scrambling to stay one step ahead of the students.
Although the literature on beginning teachers provides a useful foundation, few studies have been conducted in special education on the confluence of factors affecting beginning teacher practice. In sum, we have insufficient research to help us explain why some beginning special education teachers of reading engage in more effective classroom practices than others.
The purpose of this study was to examine accomplished, moderately accomplished, and less accomplished beginning special educators, defined by an overall classroom practice score using a valid and reliable observation instrument. Specifically, we sought to ascertain if there were commonalities and differences among teachers with varying instructional abilities in reading by focusing on personal attributes, preparation, and school environment.
We employed a mixed-methods strategy of inquiry. We deemed this to be the most appropriate approach for helping us understand the confluence of personal attributes, preparation, and school environment of participating beginning teachers. Such a method allows for more indepth study of factors than is possible using other approaches. This strategy of inquiry enabled us to examine teachers in context (LeCompte & Schensul, 1999).
Participants included 25 teachers with one to three years of experience who were assigned to teach reading to students with high-incidence disabilities in grades 3-5 (all names are coded by state name and an assigned number). The teachers taught in seven school districts in Colorado and Florida, with 18 of the 25 teachers working in Florida. The percentage of students in schools receiving free or reduced-cost lunch ranged from 2.7% to 87%; the median was 71%. A majority of teachers (n = 20) were in resource settings. All but three used core reading programs such as Reading Mastery Plus and Harcourt Brace Trophies. Use of the reader's workshop model combined with guided reading was reported by the three teachers who implemented a less prescribed curriculum delivery model. Fifteen of the 25 teachers had completed a traditional degree in special education, 4 had majored in general education and taken a special education certification test to become qualified, and 6 had participated in alternative programs that added special education to an existing noneducation bachelor's degree.
Teachers were observed using the Reading Instruction in Special Education Observation Instrument (RISE), an instrument developed to capture effective special education reading practices and used in an earlier study to measure beginning special education quality (Brownell et al., 2009). The RISE is comprised of 22 items that address the following subscales: Instructional Practices, General Instructional Environment, Decoding, Com-prehension, and Classroom Management. These subscales were designed to address components of effective reading instruction as well as instructional and classroom management practices known to define effective instruction in special education (Gersten, Fuchs, Williams, & Baker, 2001; National Reading Panel, 2000).
Items in each subscale are based on a 4-point Likert scale, and observers may score using mid-point ratings such as 2.5. A score of 1 represents "Low Quality" and a 4 represents "High Quality." Teachers were observed teaching reading on three to four occasions by one or two researchers. Upon completion of the observation, a score for each item was assigned. An overall classroom practice (OCP) score was calculated to represent overall effectiveness. The OCP score was the average score across all items on the RISE.
The validity and reliability evidence for the RISE was established in a larger study (Brownell et al., 2009). Its internal structure validity was established by examining corrected item-total correlation coefficients for the entire instrument and for each of the subscales. The coefficient alpha for the total instrument was 0.96. Corrected item-total correlation coefficients ranged from .5 to .9, providing evidence that the internal structure of the instrument was consistent with its intended structure.
Criterion validity was established by employing hierarchical linear modeling analyses to determine the proportion of variance contributed to student reading achievement gains by average overall practice as well as subscale scores on the RISE. OCP, general instructional environment, and classroom management scores contributed a significant proportion of variance to oral reading fluency gains (.37, .48, and .59, respectively). The connection between the OCP performance score on the RISE and student achievement gains provides further evidence of its validity in quantifying quality reading instruction in special education.
Reliability of the RISE was established by calculating alpha coefficients for the entire scale (.96) and the subscales (.88 to .94). We established interrater reliability between and within sites. Further, we established anchor persons in all sites, who trained the group to 80% reliability before going into the field. We also maintained within-site observer agreement, with interrater observer agreement calculated on 10% of the observations; 70% of the observations were at 81% or above.
Teachers were categorized based on results from the RISE: most accomplished, moderately accomplished, and least accomplished. We identified the most accomplished beginners as the seven teachers who had an overall classroom practice score of 3.5 or 4 (an average of all RISE items). Twelve teachers fell into the moderately accomplished range, with overall classroom practice scores that ranged from 2.0 to 3.0. Six teachers fell at the other extreme of the teacher practice continuum, with overall classroom practice scores at 1.5 or below. These teachers were considered least accomplished. These cut scores represent the ranges that cluster above and below the score of 2.5, the score that reflected average performance for each item.
Teacher interviews. To understand the differences among the teachers, we developed and administered a teacher interview (see Appendix). Questions were structured around (a) preparation and learning opportunities to gain knowledge, (b) teachers' perceptions of their ability to enact their new craft in the classroom, and (c) school context.
Preparation questions tapped into how well prepared teachers felt in special education and, in particular, reading. One line of questioning read as follows: "Did you have any reading instructors who were particularly effective in helping you see how to put what you were learning into practice? Can you give me examples of what they did?" Teachers were also asked to describe their successes and challenges in the first couple of years of teaching. Teachers responded to questions such as: "What aspects of reading instruction are most challenging to you? What was your biggest success in teaching this year?" Questions relating to school environment focused on the level of support they felt they received. For example, teachers responded to the following: "How does your teaching environment help or hinder you from putting your ideas for teaching reading into practice? What type of curricula are you using in your classroom and what types of supports have you received?" This one- to two-hour interview was conducted at the school site with the research team recording responses on a tape recorder or computer.
Secondary data sources. Two data sources were used to support findings from the interviews. First, the research team developed and administered The Special Education Beginning Teacher Survey: Influences on Practice. This two-part survey was used to collect teacher preparation data relating to (a) the quality of learning opportunities to teach reading instruction and (b) characteristics of the school environment that support teacher learning. Subscales include (a) the content of both special education and reading specific coursework ([alpha] = .92), (b) opportunities for active learning ([alpha] = .90), (c) coherence of learning opportunities for teaching reading ([alpha] = .88), (d) relevance of learning opportunities ([alpha] = .90), (e) opportunities to learn from the curriculum ([alpha] = .83), (f) schoolwide efforts to improve reading ([alpha] = .87), (g) collective teacher practices in reading instruction ([alpha] = .81), (h) school climate ([alpha] = .91), and (i) reading beliefs ([alpha] = .66). All items were scored on a 5-point Likert scale.
Trained observers took extensive field notes during all classroom observations to further inform the findings. These notes provided detailed descriptions of teachers' classroom reading and management practices; evidence drawn from these notes supported the scoring of the observation instrument. All observers were former reading and special education teachers and had considerable experience conducting classroom observations. Additionally, observers were trained to be reliable on the RISE (as described earlier).
Quantitative data were used to sort beginning teachers into groups based on an overall classroom practice rating derived from the RISE. Teacher interviews played a primary role in the qualitative data analysis, and secondary data sources such as the teacher survey and observation field notes were also part of the analysis. By incorporating multiple data sources and a mixed methodology, we were able to establish triangulation and deepen our understanding of beginning teachers.
We followed both deductive and inductive processes in a multistep fashion. Our approach was deductive in that we established a priori which general factors we wanted to examine; it was inductive in the sense that within each of these broad categories, we themed the data primarily from the "bottom up" using a grounded theory or constant-comparative approach (Glaser & Strauss, 1967; Strauss & Corbin, 1990). However, at times we followed our intuitions, checking to see if the data confirmed our understanding, in more of a deductive manner. In their description of qualitative research in special education, Brantlinger, Jimenez, Klingner, Pugach, and Richardson (2005) noted that, "It seems the more experienced the researchers, the more their studies would anticipate findings and be designed to document rather than discover phenomena" (p. 197). Thus, they point out that researchers often moved back and forth between the data and emerging theories and their own intuitions or expert beliefs.
Interviews served as our primary data source. Our first step was to read through all interviews to get a broad sense of what the data showed. Then we chunked all responses, putting brackets around text that applied to one idea. Next, we indicated whether each chunk applied to one of the three broad factors that we believed would influence teacher practice based on the literature (i.e., personal attributes, preparation, context), or "other." We then analyzed the interviews of the teachers who received the lowest overall classroom practice score to determine if certain characteristics or constructs would emerge. This was done in a "bottom-up" fashion, whereby we gave a code or label to each chunk of data (i.e., portion of interview response) and then looked for ways to cluster chunks with the same or similar codes. As explained by Glaser and Strauss (1967), we compared each chunk "with the previous incidents in the same and different groups coded in the same category ... [This process] soon starts to generate theoretical properties of the category" (p. 106).
The next step was to examine the code clusters for patterns or themes, constructing "a set of relational statements that can be used to explain, in a general sense, what is going on" (Strauss & Corbin, 1990, p. 145). We also looked for negative cases and worked to refine our themes. In addition, we made note of quotes that exemplified emerging themes.
The interviews with moderately and most accomplished teachers were analyzed in a similar fashion, coding our data, clustering the data, and looking for themes, in a recursive manner. Once this was done within the teacher types (i.e., least, moderately, and most accomplished), we checked themes across types. For example, reflectiveness, resourcefulness, and relentlessness emerged as personal attributes that could be attributed to the most accomplished teachers. We then looked at the low and moderately accomplished teachers for examples of these qualities. We triangulated interview data with Likert-scale ratings from The Influences on Practice Survey and the observed practices from field notes on the RISE.
For example, we compared the interview, survey, and observation data from a moderately accomplished teacher who told an interviewer she was inadequately prepared to teach word study. Both the interview and the score on the Preparation Survey indicated her concern about teaching phonics. Similarly, her word study score on a lesson teaching blends supported her report of struggling to teach word study. By analyzing data from teachers along a continuum, the findings provided a glimpse into the role personal attributes, preparation, and context can play in supporting or hindering a beginner's ability to thrive during the early years.
Trustworthiness and Credibility
Several steps were taken to account for our biases and to ensure the trustworthiness and credibility of our findings. From a qualitative perspective, the researcher is a critical instrument in collecting and analyzing data (Patton, 2002). Thus, knowledge and experience play a key role in data interpretation and should be acknowledged. As researchers, teacher educators, and former classroom teachers, we have more than 80 years of combined experience in special education, general education, and administration, with specific expertise in reading instruction for students with disabilities. Thus, we brought well-informed views to our research of what good general instructional practices look like and pedagogical content knowledge that is reading specific, as well as knowledge of learners, educational contexts, and curricula.
In addition, we have also accumulated several years of experience conducting qualitative research and are well trained in how to account for biases when carrying out a study, analyzing data, and interpreting findings. Thus, we were aware of our preconceptions and openly discussed differing opinions. Specifically, we discussed our perspectives about the use of the observation instrument, data collection procedures, and the data analysis process. For example, in one meeting where all of us were in attendance, we watched a videotaped lesson in which the teacher was using a scripted approach to provide reading instruction. We realized that those of us who favored this method tended to rate the teacher more favorably than those of us who did not; however, we reached group consensus and developed a common mindset that reflected a comprehensive direct instruction approach to teaching reading. In a rigorous, systematic fashion, we followed this interactive process with every item on the RISE, making sure we were in agreement about the factors addressed by each item.
After establishing interrater reliability within the group, we conducted at least 15% of our observations in teams, independently completing the RISE and comparing our ratings and our impressions. Similarly, we followed a multistep process to data analysis that involved reaching consensus about every code and theme.
To account for and monitor our biases as well as our impressions, we engaged in reflection while conducting observations and kept track of our thoughts and reactions to what we were seeing by recording observers' comments, or memos, while writing field notes. For example, in one observation, the researcher noted, "[It seems to me she is emphasizing and saying the incorrect form much more often than the correct form.] (KH, 5-13-04)." Memos such as these reminded us to check our impressions against the data to determine if they held true. As another way to ensure the validity of our findings, we looked for counterexamples during the data-analysis process whenever findings were emerging.
Finally, to ensure trustworthiness and credibility, we also engaged an external auditor. This step