Abstract
In this study, we investigate the distribution of qualified special educators across elementary neighborhood schools and exclusionary public and private special education schools. Using the 2011–2012 Schools and Staffing Survey, we provide a descriptive analysis of measurable teacher qualifications (e.g., years of experience, preparation type, degree, formal preparation, certification) across the three settings of interest. Findings indicate that each type of school relies heavily on special educators who lack qualifications in special education. Furthermore, special educators in neighborhood schools were statistically significantly more qualified in terms of experience, degrees, and special education certification than special educators in public and private exclusionary special education schools. These findings have important implications for policy makers, school leaders, and researchers.
Keywords
Among education policy researchers, there is broad consensus that teachers matter for student outcomes and that teachers’ experience and qualifications contribute to their effectiveness (Chetty, Friedman, & Rockoff, 2012; Clotfelter, Glennie, Ladd, & Vigdor, 2007; Henry, Bastian, & Fortner, 2011; Ladd & Sorenson, 2014; Rivkin, Hanushek, & Kain, 2005; Xu, Ozek, & Hansen, 2013). Although debates continue on how best to prepare teachers (Darling-Hammond, 2010), increased general and special education teacher preparation does appear to be associated with stronger student academic achievement (e.g., Boyd, Grossman, Lankford, Loeb, & Wyckoff, 2009; Feng & Sass, 2013; Ronfeldt, Schwartz, & Jacob, 2014). Moreover, the academic achievement of poor, minority, and non-English-speaking children appears to rely more heavily on the quality of their classroom teachers than their peers’ achievement (Clotfelter, Ladd, Vigdor, & Wheeler, 2006; Iatorola & Stiefel, 2003; Lankford, Loeb, & Wyckoff, 2002; Sass, Hannaway, Xu, Figlio, & Feng, 2010). Thus, the equitable distribution of qualified teachers to the students who need them most continues to warrant attention in research and policy.
Within special education, few studies have examined how qualifications (e.g., preparation, experience) contribute to instructional quality and effectiveness. In one exception, Feng and Sass (2013) demonstrated that students with disabilities were most likely to experience positive gains in reading when their special educators were well qualified in special education, as defined by certification status, coursework, and undergraduate and graduate degrees. Their study, in combination with smaller studies demonstrating the effects of preparation on instructional quality (e.g., Nougaret, Scruggs, & Mastropieri, 2005), provides justification for ensuring all students with disabilities have access to well-qualified special educators. Yet estimates consistently indicate that many schools continue to rely heavily on special educators who lack basic qualifications, such as full certification or a degree in special education, teaching experience, or coursework in teaching methods (Fall & Billingsley, 2011; Mason-Williams, 2015; Mason-Williams & Gagnon, 2017). Apparently, although a sufficient knowledge base exists as to how to intervene to raise the achievement of students with disabilities, we lack qualified individuals capable of applying those interventions in classrooms.
In the following sections, we summarize research indicating that teachers with important qualifications may be inequitably distributed among schools. Next, we discuss the ongoing utilization of exclusionary schools (i.e., public and private schools serving primarily students with disabilities in noninclusive, or exclusionary, settings) and concerns with teacher sorting in these schools. Subsequently, we present our investigation of special education teacher sorting across elementary neighborhood and exclusionary (public and private) special education schools.
Teacher Sorting in Special Education
Students who live in poverty and students who are Black or Latino/a are disproportionately served by less-qualified general educators (e.g., Clotfelter et al., 2006; Kalogrides & Loeb, 2013; Kalogrides, Loeb, & Beteille, 2013). Emerging research indicates that similar patterns of “teacher sorting” occur in special education as well, contradicting the U.S. Department of Education’s (USDOE) Annual Report to Congress, which suggests 94% of special educators meet highly qualified teacher (HQT) standards (USDOE, Office of Special Education and Rehabilitative Services, Office of Special Education Programs, 2016). For instance, Fall and Billingsley (2011) found that special educators in high-poverty districts were significantly less likely to be fully certified than special educators in low-poverty districts. Similarly, using the 2003–2004 Schools and Staffing Survey (SASS:04), Mason-Williams (2015) found that students with disabilities attending high-poverty schools were more likely to receive instruction from special educators who were not certified in special education, were new to teaching, and had completed an alternative certification (AC) program. Such differences raise important questions regarding the extent to which all students with disabilities receive an equitable education and the impact this has on student outcomes.
Research in both special and general education on teacher sorting has focused almost exclusively on regular school settings, ignoring the extent to which teachers may unevenly distribute among special education and alternative school settings. To fill this gap in the research literature, Mason-Williams and Gagnon (2017) demonstrated that secondary alternative and special education schools disproportionately employed less-qualified teachers than neighborhood schools. Using data from the 2007–2008 SASS (SASS:08) dataset, the researchers found that, compared to neighborhood schools, teachers in special education and alternative secondary schools were significantly less likely to have preparation in academic content areas, while content area teachers in secondary exclusionary schools more often lacked special education preparation. In addition, they noted a limited number of content area teachers in any school setting (i.e., neighborhood, exclusionary) held a degree and/or certification in their content area. Results of their study demonstrated that teacher sorting affects the extent to which students in exclusionary special education schools have access to qualified content and special education teachers. However, to date, no comparable studies have been conducted to examine the distribution of well-qualified teachers across neighborhood and public or private special education schools at the elementary level.
Exclusionary Special Education Schools
Access to a continuum of placement options has been a hallmark of federal special education policy since the passing of the Education of All Handicapped Children Act in 1975. Despite efforts to more fully include students with disabilities, those who exhibit the most substantial learning and behavioral needs (e.g., students requiring more intensive therapeutic and instructional supports) continue to attend public and private special education schools (referred to collectively as “exclusionary schools”; Kurth, Morningstar, & Kozleski, 2015). Nationally, 5.3% of school-age students with disabilities attend exclusionary schools, including 29.0% of students with deaf–blindness, 24.3% of students with multiple disabilities, 17.5% of students with emotional disturbance, 12.8% of students with hearing impairments, 11.1% of students with visual impairments, 9.2% of students with autism spectrum disorders, 7.6% of students with intellectual disability, and 8.3% of students with traumatic brain injuries (USDOE, Office of Special Education and Rehabilitative Services, Office of Special Education Programs, 2016]. Theoretically, exclusionary schools are designed to provide highly specialized, intensive, and individualized supports as part of the continuum of least restrictive environments (Bullock & Gable, 2006; Hoge, Liaupsin, Umbreit, & Ferro, 2013). Districts and families may choose to send students to exclusionary schools when neighborhood schools’ resources are deemed insufficient to meet students’ substantial needs, often assuming these schools are staffed by qualified special educators who will better support students’ substantial learning, behavioral, mental/physical health, and/or security needs (Bullock & Gable, 2006; Hoge et al., 2013).
To fulfill these expectations, teachers in exclusionary schools must be prepared to address student behavioral difficulties and academic skill gaps. Simultaneously, teachers are challenged with ensuring students’ access to the general education curriculum, which is both a student right and a necessity for student success following their eventual return to a neighborhood school. Presumably, accomplishing these difficult tasks would necessitate that special educators in exclusionary settings meet state certification and licensure standards. However, to date, little research has been done to investigate the qualifications of teachers in private special education schools and how their qualifications compare with their colleagues in public schools. At the same time, teachers in private schools, generally, do not need to meet federal or state certification requirements (Goldhaber, Destler, & Player, 2010), which may explain differences in the distribution of certified teachers in public schools versus private schools. For instance, Goldhaber and colleagues found that private school teachers were less likely to be fully certified, had lower average experience, and were less likely to hold an advanced degree (e.g., master’s degree, PhD) compared to their colleagues in public schools. If similar discrepancies occur between elementary neighborhood schools and exclusionary placements, as at the secondary level (Mason-Williams & Gagnon, 2017), students with the most substantial learning and behavioral needs could inadvertently be deprived of access to qualified teachers, thereby limiting their ability to meet Individualized Education Program (IEP) goals and experience a free appropriate public education (Individuals With Disabilities Education Act [IDEA], 2004).
Though evidence exists of teacher sorting among secondary neighborhood and exclusionary school settings, no research has addressed the issue at the elementary level. If inequitable teacher sorting affects schools at the elementary level, as it does at the secondary level among exclusionary schools, it would have important implications for policies and practices focused on ensuring all students with disabilities have access to teachers capable of meeting their needs. In this study, we investigate the distribution of qualified special educators within and across neighborhood schools and exclusionary public and private special education schools. Two research questions guided our analysis:
Method
To investigate the research questions, we conducted a descriptive analysis comparing the qualifications of elementary special educators across neighborhood and exclusionary schools (including both public and private special education options), using data from SASS:12. As in prior studies investigating the distribution of teachers across schools (e.g., Clotfelter et al., 2006; Mason-Williams, 2015; Mason-Williams & Gagnon, 2017), we examined measurable teacher qualifications (e.g., years of experience, preparation type, degree, formal preparation, certification). Appropriate approval was obtained from the Institutional Review Board at the university where the analysis was conducted.
Data Source
Since 1987, the USDOE has administered the SASS every 2 to 4 years to a sample of teachers, administrators, and district personnel. SASS provides a snapshot of resources and programs available in public and private schools and districts nationwide. For this analysis, we obtained restricted-access data from the 2011–2012 administration, the most current version of the SASS available. The complex sampling structure of SASS allows researchers to use it to make inferences about the preparation and qualifications of all U.S. teachers. Other scholars in special education have used SASS datasets to investigate issues related to teacher qualifications, retention, and attrition (e.g., Boe, 2006; Boe, Shin, & Cook, 2007; Mason-Williams, 2015). Variables on teacher qualifications and school setting came from the public and private school versions of the Teacher Questionnaire (TQ) and the School Questionnaire (SQ). Downloadable copies of the questionnaires are available on the SASS website (http://nces.ed.gov/surveys/sass/index.asp).
Sample
The National Center for Education Statistics (NCES) surveyed 48,829 public schoolteachers and 6,686 private schoolteachers for the SASS:12 (Goldring, Taie, Rizzo, Colby, & Fraser, 2013). According to documentation provided by NCES, national-level estimates can be produced for multiple target populations, including teachers and schools. The stratified probability design, which began at the school level for both the public and private versions of the survey, ensures sufficient numbers of schools were sampled across all school sectors (public) and affiliations (private). NCES made appropriate sampling adjustments to account for small school enrollment, using number of full-time equivalent (FTE) teachers reported as a measure of school size (Goldring et al., 2013). Detailed information regarding sampling frames, sample selection, and nonresponse bias are provided within the SASS documentation accompanying the restricted dataset from NCES (Goldring et al., 2013).
For this analysis, we merged data from the public and private school versions of the TQ to include those individuals who identified their main teaching assignment as special education in any of grades kindergarten through sixth. The current analysis included teachers who filled positions as regular full- or part-time teachers, as well as itinerant teachers, long-term substitute teachers, administrators, and professional staff members. These positions represent a small portion of the overall sample but excluding these individuals from the analysis may not accurately reflect the qualifications of the variety of individuals providing special education services to students with disabilities. Using variables from the public and private versions of the SQ, the sample was further limited to only teachers from “regular” public elementary schools (i.e., neighborhood schools), public special education schools, and private special education schools, eliminating teachers in public and private schools identified as special program emphasis, alternative, early childhood, or career/vocational/technical schools.
The complex sampling design of SASS requires the application of appropriate weights to produce better estimates of the population. In this analysis, the sampling weights were renormed for the reduced sample size, following procedures described by Thomas and Heck (2001). As this analysis relied on descriptive analysis, other methods for estimating variance (e.g., AM or WesVar software) were not necessary. The final sample yielded 1,455 elementary-level special educators across all settings of interest.
Measurement
Teaching experience
An NCES-created, dichotomous variable was used to identify teachers from the SASS:12 who were in their first 3 years of teaching. This variable is similar to those used in previous studies of teacher sorting (Clotfelter et al., 2006; Mason-Williams, 2015; Mason-Williams & Gagnon, 2017) and is consistent with research showing that rapid gains in teacher effectiveness occur during a teacher’s first 3 years (e.g., Henry et al., 2011). Teacher experience was thus defined dichotomously as (a) 3 or fewer years of experience or (b) 4 or more years of experience.
Alternative Certification
One question on the SASS:12 TQ asked respondents whether they began teaching through an AC program, which NCES defined as “a program that was designed to expedite the transition of non-teachers to a teaching career, for example, a state, district, or university or alternative certification program” (NCES, 2011, p. 27). This variable was used to identify teachers who were alternatively certified.
Degree status
As in prior studies of teacher sorting (Clotfelter et al., 2006; Mason-Williams, 2015; Mason-Williams & Gagnon, 2017), a dichotomous variable was created to indicate whether a teacher had (a) a bachelor’s degree or less or (b) a master’s degree or more.
Formal preparation
Three variables measured teachers’ formal preparation, replicating and extending a variable used by Boe and colleagues (2007). First, the number of undergraduate and graduate courses a teacher completed in teaching methods was defined as extensive (more than nine courses), some (three to nine courses), or minimal (two or fewer courses) to create the Coursework variable. The second variable (i.e., Practice Teaching) is the length of time a respondent indicated completing a practicum or internship experience, also defined as extensive (12 weeks or more), some (5–11 weeks), or minimal (4 weeks or fewer). Last, similar to the variable used by Boe and colleagues (2007), we calculated an Amount of Teaching variable by combining the Coursework variable with the length of Practice Teaching variable. For this variable, extensive teacher preparation included a combination of at least 5 weeks of practice teaching and five or more courses focused on teaching methods (with more coursework required for individuals with lesser practice teaching and vice versa). Some teacher preparation included any amount less than 12 weeks of practice teaching and at least some courses focused on teaching methods. No teacher preparation indicated teachers who had no time for practice teaching and only one or two courses focused on teaching methods.
Degrees in elementary grades, special education, or both
Questions on the TQ allow a respondent to provide information about major fields of study for the variety of degrees a teacher may attain at various levels, including associates, bachelor’s, master’s, and doctoral degrees. For this investigation, several variables were combined to identify whether a teacher held a degree in elementary education, special education, or both elementary and special education.
Certification in elementary or in special education
Respondents indicated the array of certifications they possessed on the TQ. Although the SASS provides a variety of areas of specialization within special education for respondents to choose from (e.g., autism, special education—general, learning disabilities), state-by-state variability in licensure categories makes it problematic to form useful comparisons (Geiger, Crutchfield, & Mainzer, 2003; Sindelar, Myers, & Fisher, 2016). Moreover, the use of both cross-categorical licensure and disability-specific options within some states, in combination with distinctions among grade levels (e.g., cross-categorical secondary), makes a more generic special education certification code more appropriate (Goldrick, Sindelar, Zabala, & Hirsch, 2014). Therefore, we created a dummy-coded variable to indicate whether teachers held full certification (i.e., regular, probationary) in elementary education and/or in special education. Teachers who held provisional or temporary certifications were not considered certified, as they had not yet completed all coursework or other requirements necessary for full certification.
Classification of school settings
The SASS sampling design allows national estimates to be produced according to school characteristics, including public sectors (i.e., special education schools) and private affiliations (i.e., nonsectarian special education schools). Using questions on the public and private school versions of the SQ, schools were classified as regular-public, special education-public, or special education-private based on information provided by a school representative (i.e., principal, administrative staff member). Both the public SQ and the private SQ allowed respondents to indicate whether they were a “regular” or a “special education school,” defined by NCES as a school that “primarily serves students with disabilities” (NCES, 2011, p. 6). Other options provided for respondents not included in this analysis were alternative schools, special program emphasis schools (i.e., magnet or charter schools), and vocational/technical schools. In the first category, “regular” or “neighborhood” schools included 1,353 elementary special educators, or 93.0% of the full sample. The public special education school sample included 63 elementary special educators, or 4.3% of the full sample. The private special education school sample included 40 elementary special educators, or 2.7% of the full sample.
Analytical Approach
To answer the research questions, descriptive statistics and two-way contingency tables were constructed to describe the qualifications of elementary special educators in neighborhood versus exclusionary schools, considering public and private special education schools separately. The appropriate follow-up tests of statistical significance (chi-square) were analyzed, as well as the Cramer’s V statistic, a measure of the strength of the association between the variables (Blaikie, 2003). A Cramer’s V of .5 or above is considered a strong association, while .1 to .3 is considered a moderate association, and 0 to .1 is considered a weak or negligible association (Green & Salkind, 2005).
In addition, we conducted post hoc analyses to inspect the relative and absolute contribution of each cell (i.e., the school type) to the overall variance. To do this, we analyzed the adjusted z-score residual, allowing determination of which cell(s) contributed most to the more omnibus chi-square test (Beasley & Schumacker, 1995). In addition, due to the large number of comparisons, we used a more conservative p value to avoid Type I errors using the Bonferroni adjustment (Sidak, 1967).
Results
Elementary Special Educators’ Qualifications
To answer Research Question 1, we inspected the qualifications of elementary special educators collectively (see Table 1). Across all schools, 86% of elementary special educators had more than 3 years of teaching experience. Nearly two thirds held an advanced degree (63%) and the majority identified completing traditional preparation (85%), rather than alternative preparation (15%). In respect to their preparation, three out of four sampled special educators identified completing at least three courses in teaching, while 90% completed at least 5 weeks of practice teaching. Approximately two thirds had a degree in special education, with 26% holding both elementary and special education degrees. Nearly half (46%) held certification in elementary education, and 85% held certification in special education. Approximately, 60% of elementary special educators held both a degree and certification in special education.
Qualifications of Elementary Special Educators in SASS:12 in Neighborhood, Special Education-Public, and Special Education-Private School Settings (Weighted; N = 1,455).
Note. SASS = Schools and Staffing Survey; Special Ed = special education; Alternative Cert = alternative certification; ns = not significant; Amt of prep = amount of teacher preparation; Elem Ed = elementary education; Elem + Spec Ed = elementary and special education degrees; Deg + Cert-Spec Ed = degree and certification in special education.
p < .05. **p < .01. ***p < .001.
Qualifications in Neighborhood Versus Exclusionary Schools
In Research Question 2, we investigated the relationship between the qualifications of sampled elementary special educators with their school setting. In neighborhood schools, elementary special educators appeared statistically significantly more experienced (85%) than their colleagues in public (76%) and private (65%) special education schools—Pearson χ2(2, N = 1,455) = 19.66, p < .001. The Cramer’s V statistic (.12) indicated that a moderate relationship existed between teaching experience and working in a neighborhood or an exclusionary school setting. The post hoc test (see Table 2) identified statistically significant adjusted chi-squares for both neighborhood—Pearson χ2(1) = 17.14, p < .001—and special education-private schools—Pearson χ2(2) = 14.29, p < .001. Apparently, neighborhood schools had statistically significantly more experienced special educators than expected (87%), while special education-private schools had statistically significantly fewer than expected (65%).
Post Hoc Comparisons of Statistically Significant Qualifications of Special Educators in SASS:12 by Neighborhood, Special Education-Public, and Special Education-Private School Settings (Weighted; N = 1,455).
Note. SASS = Schools and Staffing Survey; Special Ed = special education; Elem Ed = elementary education; Cert-Spec Ed = certification in special education.
p < .05, with Sidak (1967) adjustment, p < .008.
Statistically significant differences were also noted in the relationship between school type and holding an elementary education degree or degrees in both elementary and special education. In neighborhood schools, approximately 37% of special educators held a degree in elementary education, significantly more than the 21% in public special education schools and 28% in private special education schools—Pearson χ2(2, N = 1,455) = 7.75, p < .05. However, Cramer’s V statistic indicated a weak association (.073) and the post hoc analysis identified no statistically significant difference among settings. At the same time, one fourth of special educators in neighborhood schools held both elementary and special education degrees, compared to 14% to 15% in exclusionary schools—Pearson χ2(2, N = 1,455) = 7.29, p < .05. The Cramer’s V statistic indicated this was also a weak association (.071). The post hoc analysis identified that neighborhood schools accounted primarily for the differences noted—Pearson χ2(1) = 7.29, p < .008. In neighborhood schools, statistically significantly more special educators held both a degree in elementary and in special education than the expected count.
There was also a statistically significant relationship between school type and special education certification—Pearson χ2(2, N = 1,455) = 10.77, p < .05. Of the special educators sampled in SASS:12, 85% of those in neighborhood schools held certification in special education, compared with nearly 80% in public special education schools and 66% of those in private special education schools. The Cramer’s V statistic indicated a weak association (.086). The post hoc tests yielded similar results: Neighborhood schools had statistically significantly more special educators with special education certification than expected—85%, Pearson χ2(1) = 7.84, p < .008, where special education-private schools had statistically significantly fewer than expected—67%, Pearson χ2(1) = 9.36, p < .008.
Discussion
In this investigation, we examined the distribution of qualified special educators within and across neighborhood schools and exclusionary public and private special education schools. Consistent with prior research, we found the overall supply of qualified elementary special educators to be insufficient and that teacher sorting contributes to significant inequities in the distribution of special educators across neighborhood and exclusionary schools (Fall & Billingsley, 2011; Mason-Williams, 2015; Mason-Williams & Gagnon, 2017).
Insufficient Overall Supply of Qualified Special Educators
Collectively, the majority of elementary special educators sampled within the SASS:12 dataset were experienced, completed traditional preparation programs that included some/extensive coursework and practice teaching, and held a degree and certification in special education. However, approximately one fourth of elementary special educators completed minimal coursework in education-related topics, one third lacked a degree in special education, and three fourths did not have dual preparation in special and elementary education. In addition, although only 15% of special educators reported completing an AC program, approximately three out of four special educators reported engaging in none or some coursework. This may imply that a substantial proportion of traditional preparation programs are not providing experiences that meet Boe and colleagues’ (2007) criteria for extensive preparation. In light of the extensive responsibilities all special educators have with implementing intensive interventions within a multitiered support system (such as Response to Intervention and Positive Behavioral Interventions and Supports), these finding are problematic.
At the same time, although our findings are consistent with prior studies that indicate an alarming shortfall of well-qualified special educators in the workforce (e.g., Mason-Williams, 2015; Mason-Williams & Gagnon, 2017), it contradicts the USDOE’s most recent Annual Report to Congress. This discrepancy is potentially due to USDOE’s methods for counting FTE positions and vagueness in the HQT definition (Steinbrecher, McKeown, & Walther-Thomas, 2013). For instance, a state may consider a special educator who is engaged in teacher preparation as a HQT, even if they have not yet completed courses or practica (e.g., Arizona Department of Education, n.d.). Moreover, USDOE does not count short-term substitute teachers or positions left unfilled. Thus, the Report to Congress may mask the true status of the qualifications of the special education workforce.
Sorting Across Neighborhood and Exclusionary Elementary Schools
The existence of a continuum of placements has traditionally been justified using the concept of vertical equity, the idea that specialized services are sometimes needed to provide an equitable education to students with more substantial needs (Berne & Stiefel, 1984; McLaughlin, 2010). Yet we found that fewer special educators in exclusionary schools were experienced, fewer held a degree in elementary education or in both elementary and special education, and fewer held special education certification than their colleagues in neighborhood schools. These differences are troubling, given that special educators in exclusionary schools often must simultaneously address student behavior problems and academic skill gaps, while also ensuring that students have access to the general education curriculum.
If selecting an exclusionary school is based on the assumption that teachers in the school hold substantial expertise for working with students with particular learning, behavioral, mental/physical health, and/or security needs, then our findings may actually underestimate the severity of the problem. For instance, the use of a generic category for special education degrees and certification may underestimate the proportion of special educators not qualified in their main teaching assignment as we did not distinguish among different special education licensure areas (e.g., mild/moderate, severe, category specific), due to the wide variety of licensure areas across states. Therefore, some special educators who appear to hold both qualifications and a position in special education may in fact be teaching outside their area of expertise. This potentially could leave even more classrooms for students with the most substantial learning and behavioral needs to be staffed by individuals insufficiently trained for their position.
Implications for Research
As scholars have noted (e.g., Brownell, Sindelar, Kiely, & Danielson, 2010; Sindelar, Brownell, & Billingsley, 2010), providing well-qualified special educators to all students with disabilities will require a simultaneous solution that addresses both the shortage of special educators willing to continue teaching (i.e., the quantity problem) and the qualifications of special educators currently teaching (i.e., the quality problem). Research is needed to identify and to test promising recruitment strategies. Some scholars have begun to evaluate the effects of financial incentives (e.g., loan forgiveness, tuition remission) on inducing prospective teachers and teachers from surplus areas to work in special education (Clotfelter et al., 2007; Feng & Sass, 2015; Steele, Murnane, & Willet, 2010). More scholarship is needed to continue evaluating components and effects of similar programs. Complementing this research should be a focus on a deeper understanding of preservice teachers’ job preferences, how those preferences contribute to their decisions regarding where to teach, and what factors contribute to those preferences. Future qualitative studies and surveys should examine how matriculating preservice special educators make decisions about which positions to apply for and accept.
Research on the role attrition plays and strategies to ameliorate it is also needed, specifically for special educators in exclusionary settings. There are numerous setting complications (e.g., inadequate space, resources, training) and student characteristics (e.g., high intensity learning and behavior needs, mental health issues) that complicate special educators’ work in exclusionary settings that could cause significant stress (Gagnon & Barber, 2015). Researchers should examine how attrition rates in exclusionary schools compare to neighborhood schools and to what extent attrition contributes to inequitable sorting. If attrition does contribute, understanding causes of differential attrition rates could inform strategies for retention in exclusionary schools.
Implications for Policy and Practice
Our findings have multiple policy implications for state and local leaders, as well as practical implications for administrators. We recommend several courses of action that states and districts could take, in isolation or together, to increase the likelihood that students with the most substantial needs are served by well-qualified special educators, regardless of school setting.
Systemically identify and monitor inequitable access to qualified special educators across all settings
Every Student Succeeds Act (ESSA; 2015) requires states to develop Equity Plans detailing racial/ethnic, linguistic, and socioeconomic disparities in students’ access to effective, in-field, and qualified teachers. However, ESSA does not currently require states to monitor equity gaps across exclusionary and neighborhood schools. Our findings indicate that it may be important for states to pass legislation requiring that this data be collected and that it be included within Equity Plans. District leaders should also consider monitoring disparities in students’ access to well-qualified special educators across neighborhood schools and any exclusionary schools to which they send students, including private settings.
Maintain high standards for licensure in special education
One strategy we advise strongly against is lowering criteria for entry into the special education workforce. In the past, this has been a popular strategy for filling shortages (Rosenberg, Boyer, Sindelar, & Misra, 2007), and there are indications that many states continue to consider this option (Felton, 2016). It is unlikely that teachers will have the sophisticated knowledge and skills to effectively serve students with disabilities in exclusionary schools by programs that have reduced entry criteria, such as fast-track alternative routes (Nougaret et al., 2005; Rosenberg et al., 2007; Sindelar, Daunic, & Rennells, 2004). Reducing entry criteria is also likely to be financially costly in the long term, as the short-term gains of hiring someone with reduced entry criteria may be offset by costs of higher attrition that is more common among less-qualified personnel (Rosenberg & Sindelar, 2005).
Recruit and retain qualified special educators to exclusionary schools
Rather than a single solution, attracting and retaining skilled special educators to serve in exclusionary schools will likely require coordinated, systemic efforts. One effort with growing research support is the use of financial incentives, such as higher salaries or stipends, tuition reimbursement, and generous scholarships for qualified personnel willing to work in exclusionary schools (Clotfelter et al., 2007; Feng & Sass, 2015; Steele et al., 2010). Improvements in working conditions could also help to retain teachers in exclusionary schools (Albrecht, Johns, Mountstevens, & Olorunda, 2009). This may include providing additional planning time, fostering collegial collaboration among teachers within exclusionary schools and with teachers in neighborhood schools, cultivating healthy school cultures, and developing mentorship programs. Furthermore, states and districts should consider collecting data on attrition rates from these schools, as well as special educators’ perceptions of their working conditions. When attrition rates are understood, states can provide leaders with support for developing plans to improve teacher supports and working conditions.
Develop knowledge and skill of less-qualified special educators in exclusionary schools
Policy makers can enhance the knowledge and skills of personnel currently serving as special educators in neighborhood and exclusionary schools who do not hold qualifications. Similar to recruitment strategies aimed at paraprofessionals, states can offer scholarships for these teachers to pursue teacher certification, and districts should consider providing release time and/or bonuses to pursue certification in special education. In addition, because exclusionary schools are disproportionately staffed by less-skilled special educators, states should consider targeting these settings for ongoing and comprehensive professional development (Gagnon, Houchins, & Murphy, 2012).
Limitations
Several limitations are important to note. Although the SASS dataset has several unique advantages for investigations of the teacher workforce (i.e., national generalizability, robust sample size, replication across time), it lacks student outcome data, leaving researchers to rely on observable characteristics rather than direct measures of effectiveness. A similar limitation is that our measures of preparation and teaching experience are relatively blunt; knowing teachers have coursework in methods does not tell us whether the courses addressed important strategies for teaching students with disabilities. Similarly, knowing a teacher has experience in teaching does not tell us whether that experience included serving students with disabilities or how that experience might have supported their learning about effective instruction for students with disabilities.
In attempting to create a snapshot of all teachers in all schools in the United States, the developers of the SASS questionnaires need to balance extensiveness versus depth. In this investigation, we defined “in-field teaching” as an individual with a main teaching assignment in special education who held a degree and certification in special education. Limitations in the SASS data prevent more precise investigation of the potential mismatch of, for instance, a special educator trained to work primarily with students with high incidence disabilities, such as learning disabilities, who works in a classroom with students with intellectual disability. More precise measures of preparation and experience could be useful in future analyses and provide a more precise account of the extent to which special educators across settings have the necessary skills.
Conclusion
Students in public and private elementary special education schools are removed from their neighborhood schools, so they can receive skilled, specialized educational services. Although they are in these exclusionary schools, they often rely on special educators to provide the totality of their core instruction, as well as research-based academic and behavioral interventions that support their progress in general education curricula and toward their IEP goals. Yet we found special educators in these settings have fewer qualifications for fulfilling these responsibilities than special educators in neighborhood schools. A concerted effort is necessary to address both the quality and the quantity of special educators serving students with disabilities in both neighborhood and exclusionary schools. Such an effort may be difficult, but there is a serious need to address the fact that the least qualified educators are currently teaching students with the greatest academic and behavioral needs.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.