THE CHALLENGING ROLE OF A READING COACH, A CAUTIONARY TALE

Abstract

The purpose of this case study is to describe the challenges one coach faced during the initial implementation of a coaching initiative involving 33 teachers in an urban, high-poverty elementary school. Reading coaches are increasingly expected to play a key role in the professional development efforts to improve reading instruction in order to improve reading achievement for struggling readers. Data sources included initial reading scores for kindergarten and first-graders, pretest and posttest scores of teachers’ knowledge, a teacher survey, focus group interviews, project documents, and field notes. Data were analyzed using a mixed methods approach. Findings revealed several challenges that have important implications for research and practice: that teachers encountered new information about teaching early reading that conflicted with their current knowledge, this new information conflicted with their core reading program, teachers had differing perceptions of the role of the reading coach that affected their feelings about the project, and reform efforts are time-intensive.

Children who enter school with limited literacy experience are more sensitive to the quality of beginning reading instruction than children who enter school with relatively richer experiences and who have stronger vocabulary and alphabetic awareness (Al Otaiba, Connor, Lane, Kosanovich, Schatschneider, Dyrlund, Miller, & Wright, 2007; Connor, Morrison, & Katch, 2004; Connor, Morrison, & Petrella, 2004; Foorman, Francis, Fletcher, Schatschneider, & Mehta, 1998; Juel & Minden-Cupp, 2000; Scanlon & Vellutino, 1996; Spira, Bracken, & Fischel, 2005). Darling-Hammond (2000) reported that the largest variance in student reading achievement on the National Assessment of Educational Progress was explained by variables related to teacher professional development and training even after controlling for student poverty and language backgrounds. Because the link between teacher quality and student achievement is so clear, it is very disturbing that children in high-poverty schools are likely to have the least qualified, least knowledgeable teachers (Pearson, 2001). It is critical, therefore, to provide teachers with broad, deep, and ongoing professional development that focuses on attributes of effective practice (Roller, 2001).

There is some indication that part of effective professional development includes the use of reading coaches (Joyce & Showers, 1995). The International Reading Association (IRA; 2004) defines reading coaches as professional educators who “provide ongoing, consistent support for the implementation and instructional components” of reading. Reading coaches collaborate with teachers to achieve specific professional development goals; help teachers develop clinical knowledge about teaching and learning; and organize professional development to improve its coherence, duration, and consistency over time (Dole, 2004). As a matter of fact, under the Reading First Initiative of the No Child Left Behind Act (NCLB; 2001), the federal government has recommended that districts and schools employ reading coaches as a critically important part of their professional development plan.

As a consequence of Reading First, the number of reading coaches has grown exponentially over the last few years. Yet, because reading coaches are still relatively new, little is known about the process of coaching or its effectiveness. For example, the International Reading Association (IRA; 2003) expressed concern about the rapidly growing number of reading coaches given the lack of information about the quality of new coaches’ training and skills.

The purpose of this study is to describe the challenges one coach faced during the initial implementation of a school reform project in an urban, high-poverty elementary school serving a culturally and linguistically diverse student population. This is the population similar to the Reading First population of children, and this coach is an example of what the federal government and the IRA would consider an exemplary reading coach with the knowledge and skills to lead effective professional development. Nevertheless, a number of challenges existed at the school level to prevent the reading coach from effectively carrying out the level and quality of the professional development she had planned.

We first describe the conceptual framework that guided our case study of the coaching process in this project. Second, we describe the background and context for the case study. Finally, we discuss study findings and identify the challenges involved in the coaching process.

Conceptual Framework

The conceptual framework for the coaching model used in this study is grounded in two bodies of literature, one in the behavioral consultation literature (Fuchs & Fuchs, 1989; Grimes, 2002; Telzrow, McNamara, & Hollinger, 2000) and the other in the reading and professional development literature (Bean, Cassidy, Grumet, Shelton, & Wallis, 2002; Bean, Swan, & Knaub, 2003; Dole, 2004; IRA, 2004; Jaeger, 1996; Joyce & Showers, 1995). The behavioral consultation model is inductive and dynamic, beginning with identifying a problem, for example, poor reading performance, and clarifying the intensity of this problem in reliable, measurable terms. The next step in the model involves analyzing the problem to plan professional development training in evidence-based interventions that can improve reading behavior, implementing the interventions as intended, then using progress monitoring data to evaluate the effectiveness of the interventions. If students respond, the interventions were effective; if not, then the next step of the model is to modify the interventions to make them more successful for students. The process is iterative rather than linear and continues until progress is made. This coaching model is similar to a responsive problem-solving model for the school, teachers, and students (Denton, Hasbrouck, & Sekaquaptewa, 2003). A critically important aspect of coaching, according to this model, is to use student assessment data to develop professional development activities that show teachers how to use data to organize their classroom instruction and to provide flexible, small group, individualized interventions for struggling readers (Foorman & Torgesen, 2001; Pressley et al., 2001).

The second body of literature used to conceptualize this study can be found in the fields of reading and professional development. Reading coaches are an extension of the reading specialist, whose traditional purpose in schools has been to tutor struggling readers to improve their reading achievement. The reading specialist also “provides school and district-based leadership by designing, monitoring and assessing reading achievement progress, providing professional development and improving reading achievement” (IRA, 2004).

Recently, reading coaches are being asked to take on the leadership role and to forgo the teacher role. They are being asked to work only with teachers, assisting them to develop high-quality instruction to improve reading achievement for all students (Dole, 2004). This role is different from the reading specialist role where specialists did work directly with students. This aspect of the reading coach’s role became important in this study.

The role of coaching can also be informed and understood through the professional development literature that has roots in the early work of Joyce and Showers (Showers, Joyce & Bennett, 1987). They reported on the value of coaching as a part of professional development. In their professional development model, feedback to teachers and in-class coaching, when combined with theory, demonstration, and practice, accounted for significant increases in teacher knowledge and skills. These increases were most readily seen in teachers’ daily work practices in their classrooms (Joyce & Showers, 1995). Thus, another part of the reading coach’s role is to provide teachers with in-class coaching and feedback as they instruct their students.

In this study, the role of the reading coach was conceptualized as a combination of the behavior-consultation model, the reading specialist/reading coach model, and the coach’s role in the professional development model developed by Joyce and Showers (1995). The reading coach in this study was an instructional leader who acted as a consultant to teachers. She began by setting the stage for change, building rapport with teachers, and then gradually, after developing trust, observing teachers, assessing their students, and modeling lessons in grade-level meetings.

Teacher Knowledge

As an instructional coach, the reading coach is in a position of significantly increasing teachers’ knowledge about reading. Increase in teacher knowledge should be an important dimension of any professional development plan. In this study we examined growth in teacher knowledge of reading over the year of implementation of the school reform in reading. To date, few studies have examined teacher growth in knowledge of reading, including language structure. In one study, Al Otaiba (2005) reported a pre- to posttreatment improvement by preservice teachers from 52% to nearly 100% mastery on the Mather, Bos, & Babur (2001) measure. However, this study involved preservice rather than inservice teachers, who took a semester-long three-credit undergraduate reading methods course coupled with a 20-hr clinical experience that allowed them to apply what they learned while tutoring a student.

A second investigation of teacher knowledge involved a year-long intensive professional development (McCutchen et al., 2002). This study reported that kindergarten and first-grade teachers significantly improved their knowledge of phonology and orthography on a measure developed by Moats (1994). Further, during classroom observations, the investigators observed that teachers began to use their knowledge more often, which in turn led to significant improvement on phonological awareness for their kindergarten students and on oral reading fluency for first-grade students.

The Current Study

The perspective guiding this case study of one coach’s experience in school reform is that, just like students, teachers enter school with prior knowledge, background, and experiences (Darling-Hammond & Youngs, 2002; Dieker, Voltz, & Epanchin, 2002). This means that reform initiatives are not starting with a blank slate. They must include a variety of methods of ongoing professional development and must be flexible enough to accommodate teachers’ differences while working toward a similar outcome (McCutchen et al., 2002; McCutchen & Berninger, 1999; Simmons, King, Kuykendall, Cornachione, & Kameenui, 2000; Taylor, Pearson, Clark, & Walpole, 1999). The Learning First Alliance (2000) recommended that

improvement in teaching is a lifelong enterprise that requires mentoring, observation, follow-up evaluation, and problem solving with peers. Improved teaching is most likely to occur within a supportive, collaborative context that allows sufficient time for understanding of new ideas and approaches. (http://www.learningfirst.org)

This study addresses the research gap on the implementation of coaching in the reading school reform effort. In this study the coach had been an expert classroom teacher; she has also had a private clinical practice that deepened her knowledge of reading assessment and instruction; she had experience coaching and collaborating with teachers; she was skilled at presentation and was competent at observing teachers and providing them with feedback. Further, she was a national reviewer for Reading First state grants. Our focus in this study is primarily on the challenges experienced by the reading coach, as reported by quantitative and qualitative data.

Method

School Setting and Participants

The school was a K–4 Title I school located in an urban setting in the southeastern United States. The principal had 17 years of administrative experience and a recent doctorate in educational leadership. The 33 K–4 teachers were slightly more racially diverse than national patterns: 14% were African American, 3% were Hispanic, 3% were Kurdish, and 77% were White. The teachers had taught between 1 and 33 years, with a mean of 14.8 years. Only 1 was male. There were 7 kindergarten, 6 first-grade, 7 second-grade, 7 third-grade, 4 fourth-grade, and 2 English as a Second Language (ESL) teachers.

Nearly 70% of the approximately 700 students received free or reduced lunch. Over 35% of the students were English Language Learners (ELLs) with 10 different primary languages. The students were predominantly African American (38%) but also White (29%), Latino\a (28%), Asian\Pacific Islander (3%), and American Indian (0.3%). Twenty-five percent of their parents did not have a high school diploma or equivalent, 60% had a high school diploma or a general education diploma (GED), and 15% had college degrees.

Project Context and Goals

The school principal applied for a private grant to support a full-time coach to lead a professional development program to train her teachers to apply scientifically based reading research (SBRR) (National Reading Panel, 2000; Snow, Burns, & Griffin, 1998). The grant proposal requested funds for a three-year pilot program that she hoped would “demonstrate a successful strategy for alleviating much of the reading crisis” faced by her school and by other similar schools within the district.

The project goals described by the principal in the proposal were deceptively simple: (a) increase teachers’ confidence and ability to teach reading through knowledge of SBRR and (b) improve student reading achievement. The principal’s initial vision, expressed in the grant proposal, was for the coach to accomplish these two goals within three years through the following activities:

1. Plan and develop a comprehensive, research-based, school-wide reading program that included strategies for both prevention and remediation. This reading program would include strategies and materials currently in use, involve evaluating those materials to identify strengths and weaknesses, and identify and incorporate additional materials and strategies to address the weaknesses.

2. Provide ongoing professional development in SBRR for all teachers following recommendations from the Learning First Alliance (2000). This aligns with the first activity because teacher knowledge of SBRR would help them to evaluate the materials and strategies and to identify new or additional ones that met the criteria.

3. Train teachers to collect and use data to differentiate instruction. In order to identify student needs and plan appropriately for instruction, teachers would need to know the fundamentals of assessment and how to use assessment results for instructional planning.

Coaching Procedures

Timeline

Table 1 shows a timeline describing the coaching process across the school year that was developed after reviewing the coach’s e-mails and field notes. What follows is a description of these activities clustered in phases.

TABLE 1.

Timeline for Project

Date	Coaching theme	Description of coaching activities
Summer		Reading Coach Project Grant written by principal.
September–October	Problem identification	Project is funded.
		Coach is hired by the principal.
		Coach begins to observe and listen to teachers’ and principal’s ideas about the needs of the school.
		Coach conducts grade-level focus groups and individual meetings with teachers.
		Coach observes teachers’ literacy instruction to identify teachers’ strengths and weaknesses.
End of October	Problem identification and initial solution planning.	Coach becomes aware of the gap between what and how teachers taught and where the research said they needed to be.
		Coach recognizes that the core reading program was at odds with the intent of the project.
		Coach and principal begin to plan priorities for professional development: (a) describe the research and provide a compelling rationale; (b) train teachers in code-focused instruction; (c) identify teacher and child outcome measures; (d) train teachers to conduct progress monitoring (DIBELS); (e) because of the need to prevent reading difficulties, focus and allocate most resources to K–1.
		Coach was trained as a DIBELS trainer.
November–December	Continued problem identification and begin intervention.	Teacher Knowledge Assessment (Mather, Bos, & Babur, 2001) administered to all teachers.
		Coach provides two 1-day workshops to all teachers to describe the research and the scope of the plan for the project.
January	Continue implementing intervention.	Coach provides another 1-day workshop about assessments and the role of data in differentiating small group instruction.
		Coach trains a schoolwide team to administer DIBELS.
		Team tests kindergarten and first grade with DIBELS.
		Team tests second through fourth grade with Words Their Way.
February–March	Clarify the problem for the teachers.	K-1 teachers receive their child data.
	Continue intervention. Begin to evaluate the success of the intervention by observing teaching practices.	K-1 teachers receive a research-based phonological awareness text (Adams, Foorman, Lundberg, & Beeler, 1997).
		Coach models selected lessons at grade-level meetings.
		Coach begins teaching all teachers to assess their own students.
		Coach observes in K-1 and helps teachers use DIBELS data to identify at-risk students in order to tailor instruction.
		Coach recommends teachers use flexible small heterogeneous grouping strategies.
		Coach asks teachers to share what was most helpful strategy of the week.
		Coach gives teachers in second to fourth grade the text Words Their Way (Bear, Invernizzi, Templeton, & Johnston, 2000) and the data about their students’ spelling errors.
April May	More formal evaluation. Begin planning coaching for the following year.	Teacher Knowledge Assessment readministered to all teachers.
		Teachers are surveyed and participate in focus groups.
		Coaching project ends. Evaluation report is prepared and submitted.
		Principal is transferred and says farewell at the last faculty meeting.
		Coach’s contract is not renewed.

Phase 1: Identifying the needs and building a community (problem identification and planning solutions). The reading coach acknowledged her need to build knowledge about the school environment and emphasized the need to develop a sense of trust with teachers in order to build a community of learners. Therefore, her field notes reflected that she dedicated much of September and October to “observing in each classroom … talking with teachers, finding out what worked for them, and determining their concerns about teaching reading to struggling readers.” By the end of October, the coach reported that the teachers appeared interested in learning more but appeared to be overwhelmed by the requirements of the new district-mandated core reading program. Unfortunately, this program had a whole-language focus that contrasted sharply to project training. Another concern was that teachers’ editions for this program did not offer a structured scope and sequence that would support teachers in providing explicit and systematic instruction in phonological awareness and phonics (Al Otaiba, Kosanovich-Grek, Torgesen, Wahl, & Hassler, 2005).

Once the coach had identified some initial problems, she turned to researching potential solutions. First, she asked the principal for permission to order supplemental intervention materials that would support more explicit code-focused instruction. Specifically, she decided to order Phonemic Awareness in Young Children: A Classroom Curriculum (Adams, Foorman, Lundberg, & Beeler, 1997) and Words Their Way (Bear, Invernizzi, Templeton, & Johnston, 2000). Further, she reviewed the literature and contacted researchers engaged in professional development efforts. She read the Simmons et al. (2000) article titled “Implementation of a School-wide Reading Improvement Model: No One Ever Told Us It Would Be This Hard,” which described how data, specifically the Dynamic Indicators of Basic Early Literacy Skills (DIBELS) (Good & Kaminski, 1996, 2002) were used to screen, progress monitor, and individualize instruction. She read more about DIBELS, then she and the principal contacted the evaluators for help. We developed a plan for the coach herself to be trained as a trainer of DIBELS and for a university team to help her train a school team (i.e., the principal, vice principal, and Title I assistants) to help her screen the first wave of students. Training the schoolwide team was the first step, as it was more feasible than trying to train all of the teachers. Teacher training would occur in phases. Further, the evaluators suggested that the coach use the just-published Teacher Knowledge Assessment (Mather et al., 2001) as a pre- and posttest of teachers’ initial level of understanding and growth in understanding about explicit code-focused instruction.

Phase 2: Introducing plans for professional development and seeking baseline data. The first schoolwide meeting to kick off the project was in November. Teachers were given the Teacher Knowledge Assessment (Mather et al., 2001) pretest. Then the coach presented an overview of SBRR to introduce a rationale for explicit instruction and described the scope of the yearlong plan for the project during a teacher planning day in November. She explained to teachers that by the end of the first year, all teachers would learn about the research and would be trained to collect data on their children’s reading skills. She presented data about the Matthew effect (Stanovich, 1986), and then using the medical analogy of “triage,” explained that the project would focus on prevention in the first year.

In November, the coach received her DIBELS training and trained her school team. Then after the winter break, in January, the team tested all kindergarten and first-grade students on grade-appropriate DIBELS subtests. Students’ beginning reading achievement was low. Table 2 shows the percentage of students meeting performance criteria (benchmarks) as established or on grade level. It was concerning that only about a third of kindergartners met this criteria for Letter Naming Fluency (LNF), the subtest that measures how many letters a child names in a minute, as this measure is highly predictive of future reading ability (Adams, 1990). Only 6% met criteria for Phoneme Segmentation Fluency (PSF), which measures students’ ability to say the individual phonemes in three- and four-phoneme words. Only 1% of kindergarteners met criteria on the Nonsense Word Fluency (NWF), which measures students’ ability to read letter-sounds and to blend letters into words.

TABLE 2.

Percentage of Students Meeting Winter DIBELS Performance Criteria (i.e., Benchmarks)

DIBELS Performance Criteria	Kindergarten	First grade
Letter Naming Fluency (LNF)a
Established	33	—
Emerging	32	—
Deficient	35	—
Phoneme Segmentation Fluency (PSF)b
Established	6	46
Emerging	24	42
Deficient	69	12
Nonsense Word Fluency (NWF)
Established	1	32
Emerging	15	54
Deficient	84	13
Oral Reading Fluency (ORF)c
Established	—	29
Emerging	—	50
Deficient	—	22

Note. “Established” means the percentage of students who have met the criterion for mastery; considered on grade level. “Emerging” means the percentage of students who have not met the mastery criterion but are not predicted to be at significant risk for future reading failure. “Deficient” means the percentage of students predicted to not read fluently by the end of third grade.

^aLNF is not administered in first grade.

^bPSF is generally not administered after spring of kindergarten. Comparisons here are made with Spring K benchmarks.

^cORF is not administered until the middle of first grade.

Furthermore, fewer than half of first-graders had reached the end of kindergarten grade level criteria on the PSF measure. Fewer than a third of students were successful enough to reach first-grade criteria on either NWF or Oral Reading Fluency (ORF). ORF measures the number of words a student can read correctly in one minute in connected text.

Phase 3: Whole-school professional development and specific grade level focus. The second whole-school workshop was held in January. The coach provided an update of the reading initiative with a focus on the upcoming events for the spring. All kindergarten and first-grade teachers would be trained to assess their own students using the DIBELS (Good & Kaminski, 2002). Second- through fourth-grade teachers would learn to administer the Developmental Spelling Analysis from Words Their Way (Bear et al., 2000). The coach described that the purpose of training teachers in each of these assessments was to enable them to provide differentiated instruction across each classroom at all grade levels during Year 2 of the project.

Additional grade-level specific workshops were held late in the spring. At this time, kindergarten teachers learned to administer the LNF, Initial Sound Fluency, (ISF), PSF, and NWF DIBELS measures. First-grade teachers learned LNF, PSF, NWF, and ORF. Second-, third-, and fourth-grade teachers were taught to administer the Developmental Spelling Analysis (Bear et al., 2000).

Because of the strategic decision to work more intensely on early intervention, the coach met weekly with kindergarten and first-grade teachers but had bimonthly meetings with the second-through fourth-grade teachers. Grade-level meetings occurred during teachers’ regular planning time. Initially, at all grade levels, the coach discussed the assessment materials and overall class results. Intervention materials were brought into team meetings in order to model their use.

The coach focused predominantly on the kindergarten and first-grade levels. After the January data were collected, she explained the scores and benchmarks to teachers and provided them with a general rationale for using differentiated instruction. Teachers were given grade-appropriate supplementary intervention materials (Adams et al., 1997) that the coach modeled and discussed in their weekly team meetings. Teachers were then encouraged to select activities to improve blending and segmentation skills. Subsequently, as teachers began to use the materials within their classrooms, the coach helped teachers use their DIBELS scores as a way to group children for differentiated instruction.

A parallel, albeit later and less intensive, process occurred for the upper grades using the Developmental Spelling Analysis (Bear et al., 2000) results to identify students needing more intensive help. The coach met with the teachers and discussed specific strategies that might be helpful to these students. She showed them how to develop literacy centers using activities taken from Words Their Way (Bear et al., 2000), which addressed levels of need identified through students’ spelling errors. Teachers participated in a “make and take,” that is, they made several copies or sets of activities so that each teacher had a copy of the activity for his or her classroom. However, the coach spent very little time modeling in these upper elementary classrooms.

Phase 4: Project termination. The project was abruptly terminated shortly before the last day of school. The project evaluation had just recommended that the project be continued and suggested that a new core reading program that was consistent with SBRR should replace the implicit district-mandated program. When the principal asked the district for permission to purchase a core reading program, which would be research-based using Title I monies or monies from the coaching grant, she was turned down. At the final faculty meeting of the year, the principal informed the faculty she had been immediately transferred to a middle school that was even lower performing. Approval to continue the project was denied. Thus, just as the coach had established a foundation of early understanding of SBRR, with prevention and early identification as the core, and primed the teachers for the next steps—that is, asking for more information about differentiated instruction based on assessment results—the grant ended.

Evaluation Methods

We used a mixed methods research approach that would allow us to combine qualitative and quantitative research techniques (Onwuegbuzie & Teddlie, 2003; Sechrest & Sidana, 1995; Shadish, Cook, & Campbell, 2002). We collected quantitative data that describe the reading levels of kindergarten and first-grade students at the school and the knowledge of teachers. The qualitative aspect of our work involved our acting as participants-as-observers during the project (Gold, 1958; LeCompte & Preissle, 1993). For example, we consulted with both the reading coach and principal and conducted a focus group with teachers while at the same time acting as “outside” evaluators. This involvement allowed us to describe how the coach planned, developed, and implemented the professional development; to interpret “coaching” from the perspective of the coach and teachers; and to describe the project as it was situated within the context of one school. Consequently, as we interpret our findings from a situated perspective, we are mindful of Putnam and Borko’s (2000) statement, “We are inevitably part of the contexts in which we seek to understand teachers’ knowing and learning” (p. 13).

Data Sources and Collection Procedures

Data for this case study are drawn from three sources we used during the 2001–2002 school year. The pretest of teachers’ knowledge about literacy was administered prior to professional development in November and the posttest was given in late April. The coach administered both pre- and posttest surveys to all participating teachers. The teacher survey and focus group interviews were conducted by Stephanie Al Otaiba in May. The last source was project documents such as the grant proposal and the reading coach’s field and presentation notes.

Teacher knowledge

The Teacher Knowledge Assessment: Structure of Language developed by Mather et al. (2001), is a 22-item multiple choice measure that assesses teachers’ knowledge of the structure of the English language at the level of the word (e.g., If “tife” were a word, the letter “i” would probably sound like the “i” in: (a) if, (b) beautiful, (c) find, (d) ceiling, (e) sing) and at the level of the individual phoneme (e.g., How many speech sounds are there in the word “grass”? (a) two, (b) three, (c) four, (d) five). In addition, questions also address methods of reading instruction (e.g., Mark the statement that is false: (a) Phonological awareness is a precursor to phonics; (b) Phonological awareness is an oral language activity; (c) Phonological awareness is a method of reading instruction that begins with individual letters and sounds; (d) Many children acquire phonological awareness from language activities and reading). The authors reported that the test-retest reliability of the measure was .83.

Teacher survey

The teacher survey was created for the project and consisted of three sections. Section 1 contained background information such as the number of years teaching and the number of courses in reading, phonics, or reading assessment taken either at universities or through inservice training. Section 2 contained opinion questions about the effectiveness of the coaching project such as, “What is the most powerful thing you have learned that has affected your teaching of reading this year?” Section 3 addressed suggestions for improving professional development. In May, each teacher completed the survey independently, then participated in grade-level (kindergarten, first grade, etc.) focus groups held on school grounds and led solely by Stephanie Al Otaiba (i.e., the coach was not present). Each of the four grade-level groups met for between 1 and $1\frac{1}{2}\text{hr}$. The purpose of conducting focus groups was to discuss responses on the surveys, thus providing a member check. Therefore the interviewer had read all surveys prior to the meetings, brought the teachers a copy of their surveys, and identified common themes across teachers. Teachers were also asked if they had additional ideas or suggestions about the coaching process. The interviewer kept field notes about how teachers interacted, their body language, and whether their communications were consistent with their responses on the surveys.

Project documents

The third source of data was approximately 1,200 pages of project documents collected by the coach and principal. These documents included (a) the grant application to the funding agency; (b) samples of the curriculum being used; (c) notes from the grade-level and leadership meetings; (d) e-mail exchanges among the coach, evaluators, and leadership; (e) PowerPoint presentations from all workshops; and (f) the coach’s field notes. Because of the large volume of these data, in the year after the project ended, the coach reviewed her field notes and journals. She prepared a file formatted as a date log of activities. Then she met with the other authors to narrow the dimensionality of the data so it could be organized thematically and related to the quantitative data.

Data Analysis

Teacher data

The teacher survey and field notes related to focus group responses were coded by the first author for themes within each topic on the survey: prior coursework in reading and assessment, preferred mode of professional development training, strategies on which additional training was desired, attitude toward the coaching project, and challenges to implementing what they had learned from coaching using a constant-comparative method (LeCompte & Preissle, 1993). The themes were compared and differences resolved through discussion with the second author. These themes were then presented to the coach for validating and legitimating. In this way, both triangulation and member checking were used to validate the themes (Patton, 1990). In addition, the first author used peer debriefing to discuss whether similar issues were found in the initial year of Florida’s Reading First. The teacher literacy knowledge survey was analyzed using a paired samples t test to determine if there was a significant difference between the pretest and the posttest. Because of the small number of teachers, t tests by grade level could not be conducted. In addition, teacher performance on the measure was examined to analyze errors.

Project documents

Using initial categories taken from the teacher surveys, similar themes and disconfirming data were identified using a data reduction process similar to that used with the teacher survey data. Data summaries and drafts were also presented to the principal and coach as a means of member checking. The trustworthiness of data collection and analysis was supported by prolonged engagement at the school site and triangulation of sources (e.g., multiple sources and multiple respondents) and analysts (LeCompte & Preissle, 1993; Patton, 1990). We did not attempt to examine causal relationships but rather sought to identify descriptive themes.

Findings

The first sections of the study findings are organized following sections in the survey, including teachers’ knowledge, teachers’ descriptions of what they would like to learn more about, the most important thing they had learned, and challenges faced during implementation. The final section relates the roles and challenges that were identified by the coach.

What Knowledge Did Teachers Develop About Reading Instruction and Assessment?

“It’s all new to me.” The first section of the survey explicitly asked teachers questions about their background knowledge about phonemic awareness, phonics, or reading assessments. Only 4 of the 33 teachers who completed the survey recalled taking any specific university coursework on phonics, only 2 recalled any coursework covering phonological awareness, and only 5 had taken a course that covered reading assessment. Moreover, they reported that most professional development they had received prior to the project had focused on whole-language approaches rather than on SBRR. During their focus group interviews, the teachers all reiterated that this was “new material” for them. Similarly, the coach wrote in year-end field notes, “What I did do is raise their awareness of phonemic awareness. However, they had never taught it before.”

Teachers’ initial performance on Mather et al.’s (2001)Teacher Knowledge Assessment confirmed their limited knowledge about the structure of language. As shown in Table 3, in November before any training, teachers’ scores averaged 13.52 out of 22 (61%) items correct and by April, the mean score was slightly higher: 14.67 (67%) items correct (Mather et al., 2001). Although the paired samples t test revealed this was significant growth from pre- to posttest, this should be considered descriptive information and no causal claims inferred because there was no control group. Examining the types of items that teachers continued to miss in April, it appeared they continued to be confused about (a) the difference between phonics and phonological awareness; (b) the number of phonemes or sounds in words; and (c) phonics concepts such as consonant blends, consonant digraphs, and diphthongs. Teachers were given the results of their knowledge test and during the focus interviews expressed eagerness to learn more in the coming year.

TABLE 3.

Performance by Grade Level on the Teacher Knowledge of Language Structure Survey

		Pre (January)	Post (April)	Gain
Grade Level	n	M (SD)	M (SD)	M (SD)
Kindergarten	7	11.4 (2.7)	13.6 (3.2)	2.4 (2.6)
1st Grade	6	14.8 (2.6)	16.7 (3.3)	1.8 (2.3)
2nd Grade	7	12.9 (3.0)	14.0 (3.8)	1.1 (2.0)
3rd Grade	7	14.0 (3.4)	13.3 (3.1)	−0.7 (2.3)
4th Grade	4	14.8 (2.1)	15.0 (1.8)	0.3 (2.2)
ESL	2	16.0 (—)	19.0 (1.4)	3.0 (1.4)
Total	33	13.5 (3.0)	14.8 (3.3)	1.2 (2.4)

Note. Total possible is 22. Raw scores are reported.

“What do I want to learn more about and how?” The next section of the survey asked teachers for feedback regarding additional strategies they would like to learn next year and their preferred mode of professional development. With regard to strategies, teachers’ answers fell into one of five categories listed here in order of most to least popular: (a) improving literacy centers, (b) practicing DIBELS, (c) learning how to adapt instruction after gathering data, (d) phonics instruction, and (e) prioritizing and integrating what they had learned this year.

Their answers varied somewhat by grade level and their preferences were validated during the grade-level focus groups. Findings indicated that many first-, second-, and third-grade teachers were frustrated at having to “give up” planning time for professional development. This group preferred half or whole day training workshops. A first-grade teacher expressed her frustration by saying, “How can I plan how to improve my instruction when we lose our planning time? Why can’t the school give us additional time during the school day or get us a sub?” By contrast, kindergarten and fourth-grade teachers preferred to use planning time to get feedback from the reading coach. During the focus group, several kindergarten teachers nodded when the lead teacher said, “It really helps when we all hear the same thing from the coach and we can process it together.” Nearly all the teachers requested additional mentoring and modeling of the instructional strategies in their own classrooms, as illustrated by one fourth-grade teacher’s comment, “I just want to see her show me how to create centers for small group interventions without losing control of my classroom!”

“The most important thing I learned.” The survey asked teachers to identify the most important thing they had learned by participating in the project. Teachers’ answers fell into four categories: learning assessment strategies, learning new ways to teach reading, receiving helpful textbooks, and learning the importance of phonological awareness and phonics. For example, one kindergarten teacher said the most important thing she learned was the awareness of the development of reading skills. She added, “I am more aware of children’s reading development, the scope of learning to read, and the possible problems”. Several teachers did not identify anything but stated, “Ask me next year when I have had a chance to implement …“

What were challenges faced in implementing what I learned? Teachers named four primary challenges to implementing what they had learned: lack of time, knowledge, materials, and support. Upper-grade teachers also complained that they had not yet had enough in-class support to know how to use Words Their Way materials. When reviewing the survey feedback orally during the focus groups, a few third- and fourth-grade teachers shared their frustrations specifically with regard to the referral process initiated by the coach. Their comments revealed that they had expected the referral process to result in the coach’s performing the role of a reading specialist, that she would provide one-toone tutoring with struggling readers. For example, one teacher claimed the coach was always “too busy, doing I don’t know what since she doesn’t have a classroom of students.”

The majority of teachers reported feeling “overwhelmed by needing to learn a new district-adopted core reading program as well as the professional development under the project,” in the words of one first-grade teacher. During the focus groups, teachers reported that they liked the core reading program, which was neither explicit nor systematic. They seemed unaware of the mismatch between this program and what they were learning about scientifically based instruction through the reading coach. However, with news of the project termination, teachers reported to the coach that they felt abandoned and frustrated because they had understood the potential of SBRR and were eager to learn more and keep growing. The coach was not able to give the teachers any explanation for the sudden termination of the grant or the transfer of their principal, who had been their academic leader for 17 years.

Coach-Identified Roles and Challenges

The coach saw herself as having several roles: leader/collaborative consultant, resource, collaborator, diagnostician, and student advocate. As leader/consultant, she arranged staff development (led workshops and weekly grade-level meetings), arranged and selected intervention materials that were scientifically based, and coordinated all of the new reading initiative assessments. As a resource, the coach advised teachers regarding the needs of individual students. The principal also sought advice from the coach regarding assessment, planning, materials, curriculum, professional development, and other areas. As a collaborator, the coach worked daily with the teachers and administrators in her school but also with allied professionals such as speech-language professionals, the school psychologist, AmeriCorp tutors, educational assistants, ELL teachers, and the guidance counselor. Another role was that of diagnostician. The coach provided individual testing to a few students. Teachers were asked to complete a referral form and students were given a battery of criterion-referenced tests to determine current reading strengths and weaknesses with summary and programmatic recommendations provided to their teachers. Finally, the coach saw herself as a student advocate. She helped improve self-esteem and sought to motivate students through successes in reading during and after tutoring sessions. She advocated for students with teachers by helping teachers understand why students were struggling.

The reading coach identified four primary challenges to her role: the district-mandated basal reading program, the limited resources, the number of long-term “seasoned” teachers who were resistant to change, and the premature end of the project. The district-mandated basal reading program did not match what scientifically based reading research has found to help students learn to read. It was an implicit program that did not include explicit, systematic instruction in all five critical components of good reading instruction. Specifically, there was little emphasis on phonemic awareness or linking phonemic awareness to phonemic decoding. There was no instruction on fluency beyond reading with expression. Vocabulary instruction focused more on sight words than on word meanings. Comprehension instruction offered many strategies, but the core reading program did not show teachers how to explicitly teach them.

Another challenge the coach noted was the limited resources and subsequent focus on kindergarten and first grade, which had the unintended consequence of alienating many second- through fourth-grade teachers who had been teaching for many years. Fueling this alienation was the fact that most of the veteran teachers were unfamiliar with SBRR, they were aware of the mismatch between their core reading program and SBRR, and they complained that the coaching project was “one more thing” on their busy plates. Some of these veteran teachers were resistant to the notion of changing from whole group to small group differentiated instruction. Many complained that the coach did not provide enough direct service for their students.

The premature end of the project was challenging to the coach for several obvious reasons. First, she had established a “buy-in” from the teachers about the importance of SBRR despite the contradictory district-mandated training and core reading program. Not only were the teachers frustrated and confused but the coach herself experienced disappointment. In her notes, she appeared to go through a mourning process—experiencing several stages of disappointment, sadness, and a profound sense of loss. Although the 1st year of “setting the stage” was viewed as positive by the coach, she had anticipated the 2nd year as the first real “meat” of the 3-year project, teaching the teachers how to teach more effectively. Not being able to get to that aspect of the grant was especially disappointing for the coach. Furthermore, she had no explanation for her teachers about the sudden termination of the grant or the transfer of their principal, who had been their academic leader for 17 years.

Discussion

The initial three-year school reform plan was ambitious. The goals were to (a) plan and develop a comprehensive, research-based, schoolwide reading program; (b) provide ongoing professional development in SBRR for all teachers following recommendations from the Learning First Alliance (2000); and (c) train teachers to collect and use data to differentiate instruction. Nevertheless, the project ended at the end of the first year. Although funding was still available, the district suspended the project and transferred the principal. The principal was highly respected within the system and was, unfortunately for our project, transferred to a failing middle school to help make necessary changes. A possible explanation was that the new district administration was not yet ready to embrace the reform model presented and instead preferred to continue the old system. Thus, it may have been easier for the new administration to discontinue the project, which had been approved by the previous administration, prematurely than to reconsider (i.e., change) their model and implement SBRR.

Lessons Learned

Our involvement with the coaching project taught us several lessons that have important implications for others conducting research and working with reading coaches to begin reform-type professional development (Cochran-Smith & Lytle, 1999). First, it takes hard work for a coach to change teachers’ views about instruction. Second, we learned that it is particularly challenging for coaches to help teachers use systematic and explicit approaches in teaching phonological awareness and phonics when this approach is not aligned with the core reading program or part of their preservice learning. Third, teachers at various grade levels have differing views of the role of the reading coach, with many not understanding what the role is and who it is for. Finally, reform takes time. These lessons are explained in greater detail in the sections that follow.

One challenge is teacher learning of new information that conflicts with existing knowledge. Teachers’ comments on the end-of-year survey indicated they felt more aware of the importance of phonological awareness and phonics, which the coach presented in her workshops. It is important to note that the professional development in the core reading program did not focus on phonological awareness or phonics but rather on comprehension. On average teachers improved from 61% to 67% correct on the Teacher Knowledge Assessment (Mather et al., 2001). However, most teachers still struggled with important concepts needed to explicitly teach code-focused instruction, including (a) the difference between phonics and phonological awareness; (b) the number of phonemes or sounds in words; and (c) phonics concepts such as consonant blends, consonant digraphs, and diphthongs. Additionally, in-class modeling of code-focused instruction was not intended as the major focus of the initial year of the project. It is therefore not surprising that teachers’ gains on knowledge were not as strong as in Al Otaiba (2005) or McCutchen et al. (2002). However, it is difficult to directly compare findings because both studies offered much more intensive training and in the Al Otaiba study, participants were undergraduates.

A significant challenge to the coaching process in this study was that teachers held what many would call conflicting views and ideas about how to teach early reading. Teachers’ existing knowledge about using whole language to teach early reading was inconsistent with the SBRR presented to them by the reading coach. Thus, not only was the reading coach teaching something new to teachers, she was teaching something that conflicted with their existing knowledge about early reading instruction. So, a particular challenge to this reading coach was helping teachers deal with information she presented to them that conflicted with their existing ideas and experiences.

Need for alignment between the core reading program and SBRR. A related challenge for this coach was the lack of alignment between SBRR and the district-mandated, meaning-based core reading program. The process of integrating ideas and practices into ongoing instruction is neither simple nor straightforward. Joyce and Showers (1995) have reported that many teachers need to use a strategy as many as 20 times before it is fully integrated into practice. Research on teacher learning has suggested that expert teachers’ knowledge is event-structured (Putnam & Borko, 2000); so we would expect that teachers’ new knowledge about teaching the components of phonological awareness and phonics would be developed in the workshops but would be applied through instructional routines within the classroom. Instead, teachers in this study were forced to think of these two components as “add-ons” to be completed on top of the plethora of activities associated with the core reading program. Mesmer and Karchmer (2003), researchers who themselves have been coaches, noted that their Reading Excellence Act (REA)-funded project competed with many other initiatives. “As a result, teachers faced competition for their already precious time. The tremendous pull on them from so many sources may have diluted the REA initiative” (p. 644).

Differing perceptions of role of coach. Our findings are consistent with prior investigations (Quatroche et al., 2001) that reported that some teachers expected a direct service model (reading specialist) rather than a coaching model. Our study adds to the literature, however, by showing that responses varied by grade level and as a function of the amount of resources the coach dedicated to different grade levels. The majority of teachers in the lower grades appreciated the support and training from the coach. In contrast, teachers in the upper grades were frustrated because they wanted the coach to provide their low-achieving students with regular one-to-one tutoring (i.e., more of a specialist role). Anecdotally, the teachers wondered what the coach did that she did not have time to serve their kids, “if she had all the SBRR answers.”

In all likelihood the coach, who had a doctorate in school psychology and was a national reviewer for Reading First, may be more knowledgeable about SBRR than most coaches. Perhaps because of her expertise, even though she had been a classroom teacher, teachers may not have viewed her as “one of them.” Coaches who are former teachers and come up through the ranks within a school may be more accepted (Dole, 2004). Nevertheless, implications of our findings are that it is important for school leadership to clarify the role of coaches to teachers. Clarification of the coach’s role would help reduce teacher dissatisfaction about not having needs met that are not part of the coach’s role. In addition, clarification of roles may serve to increase the use of the coach as a resource and model, due to a common understanding of her knowledge, expertise, and responsibilities. As Margolis, Denny, and Hollander (1994) suggested, coaches may help teachers develop individualized instructional plans. However, because reading coaches do not provide individualized instruction for struggling readers, schools will need to provide some way to continue to provide direct service (tutoring) support for struggling readers (Allington & Baker, 1999).

School reform takes time. Clearly the teachers in this case study personify the adage “Awareness is the first step.” It took virtually the whole year for teachers to become “coachable”— to learn what it is that they did not know with regard to phonological awareness and phonics and to recognize the importance of these components of SBRR for their struggling students in particular. Perhaps if teachers had the opportunity to use a core reading program or intervention materials that were consistent with SBRR for a year, they would have improved further. During the second half of the year, the coach provided materials and began to model their use in group or grade-level trainings, but the teachers had not yet developed a schema for SBRR. Neither their prior teacher preparation coursework nor the district-mandated professional development to date had focused on phonological awareness, phonics, or assessment. Although the coach was well-meaning in wanting to develop grade-level, collaborative communities of learners, teachers in all likelihood needed a more intensive learning experience than they received.

Limitations and Directions for Future Research

There are several important limitations of this study. First, the case study occurred in only one school and for a limited time period. Findings are therefore preliminary. Given that there is so little research about the effects of coaching, which is a requirement of NCLB (2001), future research should evaluate whether change in teacher behaviors and student outcomes are causally related to coaching using a stronger research design with a comparison group, classroom observations of teaching behavior, and more frequent progress monitoring of students’ reading. Second, there are particular disadvantages of a participant-as-observer methodology. For example, we were not privy to all points of view in this study. We have data from the teachers’ and coach’s points of view but not the principal and other professionals in the school and district. Data relevant to their viewpoints would help interpret the challenges of coaching in this particular context. In addition, because the participant was an active and overt observer, there is the potential for reactivity, that is, the participants reacting to their being observed (Drew, Hardman, & Hosp, 2008). Third, the coach was more highly trained than typical coaches and there are no uniform standards for coaches in terms of either reading or leadership coursework (IRA, 2004). It is unreasonable to expect that all coaches would have the same degree of knowledge and experience with SBRR and with assessment.

Although the case presented here is only one school in a specific social and political context, the difficulties encountered are not unique. Others are apt to encounter at least some similar obstacles and challenges. Thus, the study contributes several important findings related to the coaching process that inform the currently limited research base on reading coaches. First, before initiating coaching, it is vital to identify potential discrepancies or gaps between teachers’ current knowledge and what teachers are to be taught through professional development. Hopefully, as teacher preparation programs more consistently train teachers to implement SBRR, this gap will narrow. A variety of sources of information may be used in this identification process ranging from student data, teacher knowledge measures, and interviews and surveys about teacher concerns. In this study, the coach identified that it was even more difficult for her to bridge this gap due to the lack of alignment with the core reading program. Thus, a second salient issue is generating buy-in, not only from teachers but also from district administrators, regarding the need for an explicit core reading program. Third, it is critical that the role of the coach and what coaching entails is explained in clear terms that are repeated frequently to teachers, preferably by school leadership. Although the coach in this study was brought in from outside the school faculty, this issue may also impact coaches who were former teachers or reading specialists within the school. We will never know what would have happened had the coaching project not been terminated, but our findings reinforce that coaching is not a short-term fix. Future research is needed to establish what reasonable timelines are for generating sustainable change. Such knowledge has never been so critical as now, with thousands of schools or districts in the process of implementing SBRR and struggling to meet the requirements of NCLB.

Biographies

Stephanie Al Otaiba, Ph.D., is an Associate Professor of Special Education at Florida State University and is also a Faculty Associate of the Florida Center of Reading Research (FCRR). Al Otaiba has published over 35 articles and chapters related to her research interests: early literacy interventions, response to intervention, and teacher training.

John L. Hosp, Ph.D., is an Assistant Professor of Special Education at Florida State University and Research Faculty at the Florida Center for Reading Research (FCRR). His research and teaching interests revolve around response to intervention (RtI), particularly the disproportionate representation of minority students in special education programs and aligning assessment and instruction/intervention.

Susan Smartt, Ph.D., was educated at Peabody College of Vanderbilt University and Tennessee State University. She has worked as a teacher, reading specialist, principal, school psychologist, consultant, researcher and teacher trainer. She currently is a Senior Research Associate at Vanderbilt University and privately provides consultation, technical assistance, and professional development to local education agencies and state departments of education nationwide. Her research focuses on involving parents in early literacy intervention, connecting research to classroom practice, and preparing preservice teachers.

Janice A. Dole, Ph.D., is Director of the Utah Center for Reading and Literacy and Professor in the Department of Teaching and Learning at the University of Utah. Her research interests include school reform in reading, professional development and comprehension instruction. She has published widely in research and educational journals. For the last few years, she has been co-evaluator of Utah’s Reading First project. She is also working on a national study examining four comprehension interventions on fifth-grade students comprehension.