What Do Middle Grades Preservice Teachers Believe about Writing and Writing Instruction?

AbstractAfter third grade, students’ motivation and enjoyment of writing begins to wane, and this trend continues through most of their education. Middle grade students especially need high-quality writing instruction; however, many teachers report feeling inadequately prepared to teach writing. To combat these issues, teacher preparation programs should understand how their preservice teachers feel about writing and teaching writing. The present study surveyed 150 middle grade preservice teachers to determine their self-efficacy beliefs about writing and writing instruction. Results indicate that preservice teachers valued writing, but did not feel confident with many specific aspects of writing instruction.

Writing instruction in the middle grades also presents unique challenges for teachers. According to a 2016 What Works Clearinghouse Practice Guide, writing instruction in middle and secondary contexts should include: (a) explicitly teaching writing strategies using modeling, practicing, and reflecting; (b) integrating reading and writing; and (c) using assessments to inform instruction and provide feedback to students (Graham et al., 2016). While these recommendations provide global guidelines for enhancing writing instruction in middle grades classrooms, many teachers report concerns that limit their ability to implement such practices. Specifically, when providing reasons as to why writing is not taught extensively, teachers often report a lack of preparation from their teacher training program, little time in the school day, under-developed personal skills for writing, and piecemeal curricula (Cutler & Graham, 2008;Graham, Harris, & Chambers, 2017).
The benefits to teaching writing effectively can be stark for students and teachers. For example, recent research indicated that writing more frequently in school, as little as 30 minutes per day, can lead to a 12 percentile-point increase in writing quality (Graham & Harris, 2016). Yet, according to the National Center for Educational Statistics (NCES), only about 25% of students reported writing at least 30 minutes during the school day (NCES, 2012). Furthermore, several national surveys of elementary and secondary teachers reported that most teachers spend less than 30 minutes per day engaging students in writing (Cutler & Graham, 2008;Drew, Olinghouse, Faggella-Luby, & Welsh, 2017;Gilbert & Graham, 2010).
In addition to the academic benefits of increasing writing instruction, middle grades' students represent a unique group who are highly malleable in their beliefs, attitudes, and academic achievement (Schaefer, Malu, & Yoon, 2016). Importantly, students are heavily influenced by their teachers, and several studies reported that teachers are one of the most important factors influencing student achievement (Hodges, 2015;Myer et al., 2016). For example, teachers who had more positive beliefs about writing and higher efficacy for teaching writing positively influenced their students' beliefs about writing (see Hodges, 2015). These teachers also integrated writing throughout the school day and provided more opportunities for students to practice writing. Therefore, teacher beliefs, attitudes, and instruction in specific content areas, such as writing, will impact students. As one of the primary sources of training and pedagogical knowledge comes from preservice teacher preparation programs, the present study examines middle-level preservice teachers' beliefs about writing and teaching writing. Specifically, the study analyzes quantitative and qualitative data from the Preservice Teacher Self-Efficacy for Writing Inventory (PTSWI) to determine beliefs about writing, teaching writing, and teaching writing elements.

Research and Theoretical Foundations
Writing beliefs are not a new topic in education, and researchers agree that focusing on the middle grades is important (Howell, Faulkner, Cook, Miller, & Thompson, 2016;Schaefer et al., 2016). The following sections outline recent policy changes and prior research on writing instruction in middle grades, as well as how self-efficacy beliefs about writing influence teachers' instruction in writing.

Policy and Research on Writing Instruction in the Middle Grades
The focus on writing in the U.S. has shifted in recent years, providing both hope and areas of concern for the future (Schaefer et al., 2016). While the 2000 National Reading Panel (NRP) report did not include writing as one of the five main components of literacy education, the newer Common Core State Standards (National Governors Association, 2010a) emphasize both learning to write and writing-to-learn as important constructs in literacy development (Graham et al., 2012). By contrast, the College Board and 2016 edition of the SAT (Scholastic Aptitude Test) place differing priority on writing. According to a recent news report, the new SAT includes a 50-minute essay instead of the traditional 25minute essay; however, this essay is optional. Moreover, the optional writing section will no longer receive its own score, but the score will be added to the reading portion to comprise only half of a student's overall score (O'Shaugnessy, 2014). While focusing standards on writing is movement in a positive direction, the other changes reflect steps backward for writing education and cultural values for writing.
In addition to shifts in policy, educational research has provided few evidence-based instructional practices for writing, and even fewer that are targeted for middle-level students. In a recent review of instructional practices for elementary-aged children, Jones (2015) found that two widely used writing instructional methods represented the majority of practices: writing workshop and interactive writing. Writing workshop includes collaborative time to write and review peer writing as well as mini lessons on specific skills with which students are struggling. In contrast, interactive writing is based on teacher scaffolding and modeling through mentor texts and guided writing practice (Jones, 2015). While these practices are promising, they provide little variety for teachers and may be difficult to fit into all curricula.
Because of insufficient professional preparation, relatively vague standards, and inconsistent assessments as described above, writing is presented unequally within and across schools (Martin & Dismuke, 2018;Villalón, Mateos, & Cuevas, 2015). Researchers have suggested that if a teacher did not have excellent instruction in writing throughout school, then that teacher will not have the skills to provide quality writing instruction (Graham & Perin, 2007), thus perpetuating the problem for future generations. Additionally, if a teacher did have good modeling of writing throughout school and the teacher only relies on his or her perspective as a student, the teacher may not realize why the strategy worked from the vantage point of the instructor (Graham & Perin, 2007). This may result in the strategies working less effectively for this teacher, or not helping students in the same manner. In other words, many of these practices are based on anecdotes or testimonials, not on evidence-based research, which can lead to misuses or misunderstandings in their effectiveness (Graham & Harris, 2014).
Beyond individual classroom variability, overall research with in-service teachers shows that writing is not emphasized as a priority (Cutler & Graham, 2008). Commonly, teachers state that they value writing but feel unprepared to teach the skill to their students (Cutler & Graham, 2008). For example, in a study of 294 randomly selected primary-grade teachers from across the United States, only 28% stated that their "preparation to teach writing was either very good or outstanding" (Cutler & Graham, 2008, p. 911). An additional 42% of the participants described their preparation to teach writing as adequate, and 28% reported that their preparation in writing was poor or inadequate. When the teachers evaluated their own writing practices, they only moderately agreed that they (a) enjoyed the task of writing, (b) were effective writing teachers, and (c) managed writing time in the classroom effectively.
The findings of Cutler and Graham (2008) led other researchers to question how teacher self-efficacy for writing may impact students' writing achievement. As a result of these low levels of self-efficacy, teachers spend little time instructing writing or providing opportunities for students to engage in writing tasks (Applebee & Langer, 2009;Cutler & Graham, 2008;Graham, Harris, & Santangelo, 2015). When a specific skill does not receive ample attention, the perpetuated idea is that the skill is not important. These factors often lead to teachers avoiding writing in the classroom, creating the disconnect between the workforce demands of writing and amount of writing instruction in K-12 classrooms (Yancey, 2004).

Self-Efficacy Beliefs about Writing and Teaching Writing
Transitioning from instructional practices to teacher cognition and affect, we consider teachers' self-efficacy regarding writing. At the core, self-efficacy is defined as a person's belief that she or he can accomplish a task successfully even if the task is or becomes difficult (Bandura, 1986). Furthermore, Bandura (2001) and Pajares (2003) explained that self-efficacy beliefs originate from four sources: (a) interpreting the results of previous performance; (b) models and observations of more knowledgeable others; (c) social persuasions and interactions; and (d) emotional states. In considering self-efficacy within specific contexts, such as writing, self-efficacy for writing is a person's belief that she or he can accomplish a writing task successfully, even if the task is challenging. Of the limited research on self-efficacy beliefs in teaching writing, the majority has focused on inservice teachers (Cutler & Graham, 2008), not preservice teachers. A recent study by Wang, Hall, and Rahimi (2015) found that teachers with high selfefficacy in writing spend more time engaging students in writing tasks each week than teachers with low self-efficacy. High self-efficacy teachers also engage more in teaching the writing process, grammar, and usage skills (Graham, Harris, Fink, & MacArthur, 2001). Considering the theoretical framework for self-efficacy from Bandura (2001) and Pajares (2003) with the findings from Cutler and Graham (2008), teachers with high self-efficacy beliefs for writing are more likely to offer writing activities, model writing practices, and create environments in which writing is supported for their K-12 students and is likely to impact the writing achievement of those students. To develop teacher's self-efficacy for teaching writing, a change should begin at the teacher preparation level. Changes in self-efficacy take dedicated time and practice (Bandura, 1997), and such an intervention is more easily implemented within the structure of teacher preparation programs. By the time teachers reach the classroom, they are inundated with tasks and gain limited professional development for teaching writing. Reaching teachers who are still developing their beliefs about writing and writing instruction has the potential to proactively prepare teachers to more successfully integrate writing into their future classrooms rather than to reactively try to change entrenched behaviors.
Assessing preservice teachers' beliefs about writing involves studying two separate belief systems-selfefficacy for writing and self-efficacy for writing instruction. First, preservice teachers must have high self-efficacy for writing (as a personal writer). In other words, teachers must view themselves as writers to be an effective writing teacher (Morgan, 2010;Zimmerman, Morgan, & Kidder-Brown, 2014). The Peter Effect (Applegate & Applegate, 2004) states this as teachers cannot teach what they themselves have not learned. If a teacher has not learned to be an effective writer and cannot perform the task, the teacher will most likely avoid writing in a K-12 classroom and not emphasize writing with students (Tshcannen-Moran & Woolfolk Hoy, 2001). Preservice teachers need to develop writing skills prior to entering the teaching profession, and this development should occur in their preparation programs.
Second, preservice teachers must develop selfefficacy for writing instruction. Self-efficacy for writing instruction is the preservice teachers' belief in their ability to teach writing effectively. Grossman, Hammerness, and McDonald (2009) stated that understanding how to write and being able to implement writing in a classroom can make an effective writing teacher. In the present study, we argue that both constructs-self-efficacy for writing and self-efficacy for writing instruction-make effective writing teachers, and this work helps consider the degree that these constructs are unique or overlapping. Therefore, in teacher preparation programs, preservice teachers need to be exposed to methods for instructing students on writing, developing writing skills in others, and modeling writing practices.

Gaps in the Research
Based on the inconsistent policy changes that influence the classroom context, time for writing, and writing curricula and the links between self-efficacy and writing instruction, it is clear that how teachers perceive writing and writing instruction greatly influence their middle-level students. However, researchers have tended to focus on in-service teachers rather than consider how preservice teachers view writing and writing instruction. This study begins to fill this gap by analyzing the beliefs about writing and writing instruction of preservice teachers who are pursuing initial teaching licensure in the middle grades. Specifically, we were guided by the following three research questions: 1. What are preservice middle-level teachers' selfefficacy beliefs about writing? 2. What are preservice middle-level teachers' selfefficacy beliefs about writing instruction? 3. What are preservice middle-level teachers' selfefficacy beliefs about teaching specific writing elements?

Methods
The present study surveyed preservice teachers enrolled in education programs in the Southwest United States. The survey described three components of writing beliefs: (a) self-efficacy for teaching writing elements; (b) self-efficacy for writing; and (c) self-efficacy for writing instruction.

Participants and Sampling
Preservice teachers (n = 150) were sampled from the Southwestern part of the United States. All preservice teachers in the present study sought initial licensure through a traditional teacher preparation program and focused on licensure for grades four through eight. While middle-level education is commonly defined as grades five through nine (Association for Middle Level Education, 2014), the teacher preparation programs in the present study defined middle level as grades four through eight. The sample was fairly homogenous in terms of race, with 84.7% of the sample identifying as White, and gender, with 88.7% identifying as female (see Table 1). These numbers are representative of the total population of teachers located in the state.
Additionally, most participants identified as junior classification (58.7%) and another 26.7% identified as sophomore classification, indicating that most participants were at least mid-way through their licensure program. Finally, almost half (48.7%) of participants sought English language arts certification and another 30.0% sought mathematics certification. Science and social studies certification were represented by 8.7% and 11.3% of participants, respectively.

Instrument
The participants were surveyed using the Preservice Teacher Self-Efficacy for Writing Inventory (PTSWI)  (Hodges, 2015). This instrument measures three factors: self-efficacy for teaching writing elements, self-efficacy for writing, and self-efficacy for writing instruction. Self-efficacy for teaching writing elements includes 11 items specifying qualities of writing instruction and asks preservice teachers to indicate the degree to which they feel prepared to teach those elements (see Table 2). One example item asks preservice teachers, "I feel prepared to teach voice (i.e., presence of the author in the text, tone)." The preservice teacher then can indicate "to a great extent," "somewhat," or "not at all," resulting in a score of 3, 2, or 1, respectively. Self-efficacy for writing asks preservice teachers to rank on a 5-point Likert scale their beliefs about their own writing abilities (see Table 3). For example, "I am confident in writing for a variety of audiences," shows how preservice teachers internalize their own abilities to write effectively. Finally, self-efficacy for writing instruction asks preservice teacher to rank on a 5-point Likert scale their beliefs about utilizing writing in the classroom and teaching middle grades' students to write (see Table 4). One item, "Providing consistent assessment of writing is important to developing writing confidence in students," asks preservice teachers to consider how assessment improves students' writing achievement.

Data Collection
During the final week of courses, in early December, the preservice teachers were recruited in person and provided with a link to access the survey online. The first page of the online survey provided participants with the option of completing informed consent and digitally signing to indicate their voluntary participation in the study. The survey took less than 20 minutes for most participants to complete.

Data Analysis
Data were coded both quantitatively and qualitatively. First, the mean and standard deviation for each item on the survey were calculated. This provided information about which items provided preservice teachers with the most confidence and the most concern. Next, the frequencies for each Likert response for each item were calculated. Again, this showed how preservice teachers clustered and how often preservice teachers chose the  (Lincoln & Guba, 1985).

Results
In the following sections, we first present the parametric results for middle grades preservice teachers' self-efficacy for teaching writing elements, self-efficacy for writing, and self-efficacy for writing instruction. Then, we present the qualitative results from the preservice teachers' responses to the openended question.

Self-Efficacy for Teaching Writing Elements
Self-efficacy for teaching writing elements includes preservice teachers' perceptions of their ability to teach specific components of writing. Across preservice teachers surveyed, great variation existed in which elements of writing they felt most confident in teaching and which elements of writing caused them the most anxiety in teaching. Preservice teachers reported the highest mean score for organization of ideas (M = 2.71) and the lowest mean score for spelling (M = 2.18). The results suggest preservice teachers felt confident that they could instruct students how to develop a wellorganized piece of writing, but they felt less prepared to teach students how to use appropriate spelling. Preservice teachers also felt less confident in their abilities to teach voice (M = 2.19). The other items on the survey clustered around means of 2.33-2.65, and looking at the individual responses indicates that most preservice teachers felt prepared to teach these components to a great extent or somewhat.

Self-Efficacy for Writing
Next, we determined preservice teachers' selfefficacy for writing or their perception of their abilities to complete writing tasks, products, and processes. Interestingly, the highest mean for any item was for Writing helps me accomplish daily tasks (i.e., complete to-do lists, journaling, note-taking), which indicates that preservice teachers acknowledged the importance of writing. The lowest item mean was for The majority of time I spend writing for enjoyment, showing that while preservice teachers saw the value in writing and understand its importance, they did not enjoy the task. Generally, preservice teachers reported more positive means (over 2.50) for most items, meaning they had strong self-efficacy for their own writing abilities.

Self-Efficacy for Writing Instruction
Self-efficacy for writing instruction focuses on preservice teachers' perception of their abilities to instruct writing in middle-grade classrooms. The highest scored item was for Writing is an important skill to teach to students. As in the self-efficacy for writing section, preservice teachers acknowledged how important writing is and that it should be a component of their instruction. Additionally, most preservice teachers agreed strongly or agreed that writing is important to their content area instruction. The lowest-rated item concerned using state standards focused on writing, which preservice teachers indicated they feel uncomfortable using.
Qualitative Coding: Types of Writing Preservice teachers were not provided with a definition of writing when asked about their writing productivity in a typical week. From the qualitative results, a variety of interpretations for what constituted writing were presented. Interestingly, the participants reported varying degrees of writing frequency in a given week. About 30.7% of participants reported that they wrote daily (see Table 1). The most frequent response was "In a typical week, I write 3-5 times per week," which represented 34.0% of participants. Another 28.0% of respondents noted that they only wrote one to two times per week, and 7.3% indicated they wrote less than one time per week. These results indicate that preservice middle grades' teachers were writing and using writing, but did not illuminate how writing was used.
The PTSWI also included one open-ended question that asked preservice teachers to detail the types of writing they did over the past week. For the 150 participants, 1 participant indicated "none," 145 reported an academic purpose for writing, and 37 included personal purposes for writing. The total number of responses exceeded the total number of participants (n= 150) because some participants reported that they wrote for both academic and personal purposes.
Interestingly, only eight respondents provided examples of writing specifically for educational purposes, such as writing lesson plans or preparing materials for their inschool practicum. Three of these responses specifically addressed using writing to prepare materials for mathematics teaching, which revealed a link between literacy and mathematics instruction. The majority of responses focusing on academic writing indicated they used writing to show knowledge from courses or practicum in the form of essays, reflections, or discussions.
In contrast, the most common responses for personal writing included journaling, often spiritual reflections, text messages, or social media posts. Of the students who indicated they wrote for personal reasons, 40.5% reported using writing for religious reflection or study. These results show that preservice teachers viewed writing in many ways. However, the results also reveal that preservice teachers may not have identified or acknowledged writing for teaching purposes. The total number of preservice teachers reporting that they used writing for teaching purposes seems low, which may indicate that preservice teachers did not consider lesson plans or preparing materials for class to be writing activities.

Quantitative and Qualitative Results: Converging or Diverging
The quantitative and qualitative results converge to indicate that preservice teachers valued writing and were willing to use writing to complete academic tasks, such as writing notes, but were less likely to engage in writing during their personal time. However, because most of the respondents who indicated that they wrote for personal reasons engaged in writing for spiritual purposes, a conclusion may still be made that the participants valued writing.

Discussion
After analyzing the qualitative and quantitative results, two themes emerged from the data: (a) many preservice teachers showed neutrality in teaching writing elements, and (b) preservice teachers' responses indicate a potential struggle in writing for various purposes and audiences. We explore these two themes further in the following sections.

Neutrality of Teaching Writing Elements and Implications for Middle-Level Education
The present study revealed interesting trends related to how preservice teachers felt about teaching specific elements of writing. Preservice teachers revealed the highest means for teaching organization of ideas and the revising and editing process. Both of these elements are abstract, with few set rules for completion. Editing typically follows the conventional rules of the English language; however, revising does not follow specific rules and organization is often viewed as highly abstract. In contrast, the lowest mean scores belonged to spelling and voice. Spelling follows conventions of English along with phonics rules. Low scores for teaching spelling may be attributed to preservice teachers' lack of preparation to teach morphological awareness as well (Washburn, Joshi, & Binks-Cantrell, 2011). On the other hand, voice is another abstract skill. Often voice is not taught in K-12, yet teachers expect that students can mysteriously and successfully use this skill (Elbow, 2007).
Overall, preservice teachers felt well-prepared to teach specific writing elements such as organization, word choice, and sentence structures regardless of course type. The concept of teaching apprenticeship (Zimmerman et al., 2014, which suggests that preservice teachers have been exposed to writing instruction for many years throughout their own K-12 schooling, could be affecting the high means for writing elements. More specifically, preservice teachers felt most prepared to evaluate the overall quality of student writing and organization of ideas. Many rubrics focus on the overall piece of writing, such as the 6 + 1 Traits Rubric that preservice teachers likely experienced during their own schooling. Additionally, standardized tests focus on organization, giving preservice teachers preparation in this element of writing. Preservice teachers feel least prepared to teach voice (i.e., presence of the author in the text, tone). Even though voice is a component on many rubrics, it is typically not a focus of direct instruction as are other elements of writing, such as grammar and clarity. Elbow (2007) described how the concept of voice in the teaching of writing has shifted since the 1960s, often jockeying between being prominent and losing momentum. At the same time, Witte (2007) emphasized how writing using voice could enhance students' online writing products. Both researchers indicated that voice may have been absent from most writing curricula through much of the 1990s and 2000s, which may explain why preservice teachers who attended school during these years felt less confident teaching this element of writing.
Interestingly, preservice teachers also rated their preparation to teach spelling lower than other elements, though many early childhood and elementary classrooms still give routine spelling tests. Joshi, Treiman, Carreker, and Moats (2008) explained that spelling rules are often overlooked and rote memorization is emphasized with teaching spelling. This might explain why preservice teachers did not feel adequately prepared to teach spelling; they likely knew many spellings by memorization but could not explain the underlying rules and principles.
The most interesting finding about teaching writing elements was that for more than half of the items preservice teachers selected "somewhat," which represented the neutral response for a category (see Table 2). A neutral response is difficult to interpret. One possibility is that preservice teachers did not have a strong opinion about their efficacy to teach writing elements. Alternatively, it is possible they had not reflected on their beliefs or did not want to reveal those beliefs. Moreover, they may have been unaware of their efficacy to teach these skills if they have had little practice in teaching writing during their teacher preparation program. This finding indicates that preservice teachers may require more exposure, practice, and guidance about how to teach writing and increase their knowledge of writing pedagogy.

Struggles with Recommendations for Writing: Various Purposes and Audiences
In the present study, preservice teachers indicated they highly valued writing. They mentioned using writing to complete daily tasks and that they were comfortable self-monitoring during the writing process; however, these same preservice teachers indicated that, overall, they did not have positive feelings toward writing and typically did not write for enjoyment. There appears then to be a disconnect between value and enjoyment. How will these preservice teachers approach writing instruction-as a task they value but do not enjoy? What changes can teacher educators make to enhance preservice teachers' positive feelings toward writing?
One reason the preservice teachers did not enjoy writing may be that they did not have high selfefficacy for the task. Specifically, preservice teachers indicated that they felt neutral about their ability to teach many elements related to writing, which may connect to their lack of confidence in successfully using these elements in their own writing.
Additionally, preservice teachers consistently reported that they did not feel confident writing for various purposes or audiences. Again, this finding may reveal a gap in their preparation to write and teach writing.
Preservice teachers communicated that they did not feel confident writing for various purposes and audiences, which may be a direct result of policy changes that came about during their K-12 experiences. The Common Core State Standards have a renewed focus on writing for various genres and audiences; however, these standards came about in 2010. Traditionally aged preservice teachers (i.e., those who entered college directly after high school) at the time of this study would have been in high school when Common Core was passed, so the writing instruction they experienced as students was likely not impacted by this change. These preservice teachers were taught during a time when they likely did not have to complete as much writing; therefore, they may not feel confident in their abilities to write or teach writing.
According to a 2012 practice guide for writing instruction, Graham and other prominent researchers suggested that teachers integrate writing across the disciplines and for various purposes and audiences. How will preservice teachers do this when they do not feel confident approaching these writing tasks themselves? Future research can work to answer these questions, specifically addressing how middle-level preservice teachers will approach writing instruction.

Limitations
While the present study considered confounding variables and issues in both data collection and analysis, several limitations are present. The data collected by the PTSWI are self-report data, which inherently subject itself to skepticism. Preservice teachers could be rating their efficacy more highly than they actually feel; this Hawthorne Effect could be the result of knowing they are part of a research study (Patten, 2009;Thompson, 2006). They may also be rating certain items higher simply because they suspect the items are related. In reality, preservice teachers might feel that writing is important but not feel prepared to teach certain aspects of writing. These factors were minimized by having a third-party researcher administer the surveys who was associated with neither the preservice teachers' course nor their grade.
Limitations to the testing and instrumentation included the questions asked on the survey, omitting some critical information about their past experiences. For example, preservice teachers' selfefficacy for writing might have been the result of factors outside of the classroom such as writing courses in other college courses (e.g., composition, English, or even science-related writing courses). In an effort for the survey to be completed during a small portion of class time, an exhaustive number of questions about the writing history of preservice teachers could not be asked.

Future Directions
Teacher education programs can use the PTSWI to measure the preservice teachers' self-efficacy beliefs for writing and writing instruction and then use the results to gauge the effectiveness of their program. These results can inform the education program in making changes to improve the quality of teacher preparation for writing instruction. Teacher education programs can also use the results of this study and their own results to advocate for including writing in teacher education.
The results of this study are important for policymakers. Currently, education policy is moving toward a focus on value-added scores, which to date only consider mathematics and reading achievement (Amrein-Beardsley et al., 2013;Scherrer, 2012). However, reading and writing are two sides to literacy that are equally important to K-12 students' success. Studies have shown that increasing writing achievement shows subsequent increases in other subject area achievement, such as mathematics (Kenney, Shofner, & Norris, 2014;National Commission on Writing, 2004). Writing is not often a focus of teacher education programs (Myers et al., 2016), which could result in increased gaps in K-12 students' writing achievement. Teacher education programs can use the PTSWI to identify weaknesses in their own preservice teachers' writing preparation, specifically in how they prepare middle-level educators. The programs can see in which areas preservice teachers feel efficacious and in which areas they need further instruction. Teacher educators can design writing instruction, research, and practice focused on what their individual students identify as areas of weakness.
The greatest influence for the field of writing comes from current and future policy changes. Currently, writing research is under-represented in No Child Left Behind (NCLB), value-added legislation, and the Every Student Succeeds Act (ESSA). However, these results show that preservice teachers have mixed views on writing. When studying in-service teachers, research has shown that these perspectives influence the amount of writing and types of writing instruction K-12 students receive. These findings indicate that policy-makers need to readmit writing to the national conversation when considering education.

Conclusion
The middle grade is a time when students face great changes emotionally, academically, cognitively, and behaviorally (Howell et al., 2016). For students to succeed academically during this challenging time, they need a safe, comfortable, and rigorous instructional context. Middle-level schools also present unique challenges in instruction. Students leave elementary school with foundational skills, and middlelevel schools should build on those skills while preparing students for high school. If the instruction students receive in the middle grades is poor, they are less likely to be successful in later years (Graham & Perin, 2007). Therefore, middle grades' teachers should feel confident in their ability to teach specific content to students (Campbell & Filimon, 2018;Howell et al., 2016). If middle grades' teachers do not feel confident in their abilities to teach writing, they will provide less efficacious instruction to middle grades' students, thereby perpetuating negative attitudes toward writing and not improving students' skills.
While policy-makers may be shifting the emphasis on writing, teachers are not passively waiting for something to happen-they are organized, proactive, and self-regulating individuals with specific goals in mind (Amrein-Beardsley et al., 2013). This social cognitive view of teachers is founded on the notion that they have some control over their environment (in this case, the classroom) and make decisions regarding the instructional content (Bandura, Barbaranelli, Caprara, & Pastorelli, 2001). Furthermore, just as a child's beliefs about his or her ability to master difficult tasks will affect his or her academic engagement and interest (Bandura, Barbaranelli, Caprara, & Pastorelli, 1996), so too will a teacher's belief that he or she can persevere through difficult writing tasks affect the teacher's motivation to integrate writing instruction in the classroom.
However, highlighting the key role of the teacher is not to say that policy is unimportant. Recent policy changes depict a roller-coaster of beliefs about writing. Some policy changes have advanced writing instruction, such as the development of the Common Core State Standards with a renewed focus on the importance of writing as a tool for learning, while the emphasis on value-added scores continues to ignore writing instruction (Amrein-Beardsley et al., 2013). If writing instruction is not emphasized in teacher preparation programs, preservice teachers will likely not value writing and it will not be a focus of their pedagogy once they have responsibility for their own classrooms. Though most research conducted on beliefs about writing and self-efficacy for writing is related to in-service teachers, the beliefs about writing and self-efficacy for writing of preservice teachers cannot be ignored. The findings from this study can be used to inform teacher education programs about the necessity for writing instruction courses while demonstrating the connection between self-efficacy beliefs and practice.