January 10, 2007
Alternative Route Programs for Certification in Special Education
By Rosenberg, Michael S; Boyer, K Lynn; Sindelar, Paul T; Misra, Sunil K
The need for high-quality teachers-particularly in such high- demand areas as mathematics, science, and special education-has been a major impetus for the emergence and growth of alternative routes to certification (AR). Indeed, the U.S. Department of Education (2002, 2004) has proclaimed repeatedly that AR programs, as opposed to the traditional routes offered by colleges of education, are an effective means of streamlining the process of certification to move teachers into the classroom on a fast-track basis. In fact, recent federal legislation, including No Child Left Behind and IDEA, encourages these approaches to teacher preparation. These changing standards require alternative route candidates to pass certification or licensure exams to be highly qualified, but AR programs can alter, shorten, or waive entirely coursework in educational philosophy, pedagogy, and practice teaching.
In addition to the persistent shortage of qualified personnel and federal government influence, two other factors have contributed to the proliferation of AR programs. One is the acute need for multicultural personnel. Special education teachers are predominantly White (86%), whereas the student population requiring special education is 32% diverse (Tyler, Yzquierdo, Lopez-Reyna, & Flippin, 2004). On average, AR programs have been more successful than traditional programs in attracting African American and Hispanic participants into the field. The success of these programs is a result of the tendency for alternative route program participants to reflect the demographics of the communities in which the programs are located (Humphrey & Weschler, 2005). Not surprisingly, many urban school districts, routinely and desperately in need of large numbers of certified teachers, develop alternative routes in special education and other high-needs fields. They consider AR programs to be a more viable source of supply of special education teachers-both White and culturally and linguistically diverse (CLD)-than traditional programs. Not only is the number of candidates greater, but there are also greater numbers of opportunities to tailor program content and activities to address the challenges specific to urban and minority schools (e.g., Rosenberg & Rock, 1994).
The second additional factor is the strident sentiment from both inside and outside the educational establishment that traditional approaches to teacher preparation are self-serving, overregulatory, anachronistic, and ineffective. For the most part, those who advocate alternative methods of teacher preparation assert that little evidence links pedagogical training and teacher quality, and they claim that the traits most associated with teacher effectiveness are knowledge of content and verbal ability (Hess, Rotherham, & Walsh, 2004). Although this sentiment for reform has translated into a range of new methods for providing teacher preparation (Hassel & Sherburne, 2004; Rosenberg & Sindelar, 1998), there is little information indicating how these novel approaches would develop teachers for students with disabilities.
When policy makers and those involved in general-education teacher preparation contemplate the design and implementation of AR programs, special education is rarely on their radar screens. Indeed, most of what legislators often debate and what the education literature often reports (e.g., Goldhaber, 2004; Zeichner & Schulte, 2001) can best be characterized as the "secondary content model," that is, programs that focus on contentspecific pedagogy for individuals with subject-area expertise. Still, what researchers know about AR programs in mathematics, science, and other content areas may have limited application to special education teacher preparation, whereas universal design and pedagogy are more important than specific subject-area methods (Rosenberg & Sindelar, 2001). This factor is critical when considering the range of competencies (e.g., behavior management, social skills instruction, and content enhancements) required of educators who deliver high- quality, comprehensive educational programs to students with the range of special needs that students with exceptionalities exhibit (e.g., Council for Exceptional Children, 2003; Neel, Cessna, Borock, & Bechard, 2003; Rosenberg, Sindelar, & Hardman, 2004).
AR programs are proliferating at a rapid rate (Feistritzer, Haar, Hobar, & Scullion, 2005), and researchers know little abouf the range of program offerings in the extreme shortage area of special education. The stakes remain high because teacher certification, which many regard as a major component of teacher quality, is associated with public confidence in the schools. However, the literature does not include reports about most AR programs, and researchers know little about the nature and extent of these programs. In this study, we identified 235 AR programs in 35 states and the District of Columbia (we did not find evidence of special- education AR programs in 15 states); 101 programs in 25 states and the District of Columbia responded to the survey. As a result of the magnitude of the response, we know more about the scope of AR programs and about the design and delivery of these programs, as well as the characteristics of participants recruited. Our purpose, in partnership with the National Clearinghouse for Professions in Special Education (NCPSE), was to develop a database that lists and describes AR offerings in all states that have such programs. Specifically, we were interested in the following:
* How the prevalence of AR programs in individual states relates to the shortage of special education teachers within those states.
* The design and administration of such programs.
* The length and intensity of programs.
* Program characteristics, such as admission requirements, modes of instructional delivery, and completion rates.
* The experiences and demographic composition of AR program participants.
The methodology for the study involved developing a survey, generating a comprehensive listing of alternative programs, and administering the survey to AR program directors.
DEVELOPING THE SURVEY
Using resources from the available literature on AR programs (e.g., Hawley, 1992; Hillkirk, 2000; Rosenberg & Sindelar, 2001; Zeichner & Schulte, 2001) as well as our own experiences in designing and studying teacher programs, we identified four areas believed to differentiate among AR programs, including (a) program infrastructure, (b) program length and intensity, (c) specific program features, and (d) participant demographics. (For a copy of the full survey, contact the first author.) Table 1 presents descriptions of these areas, the scope of the questions asked, and the number of items developed for each area.
We conducted survey development and validation in several steps. First, we generated items for a telephone survey by using program infrastructure, program length and intensity, specific program features, and participant demographics as referents. We then compiled the items into a preliminary survey draft and circulated this draft among members of the research team, as well as associates of team members. These individuals rated each item for its face validity as well as its ability to assess the construct of the survey area. We used a 4-point scale to assess correspondence to the area and removed items that fell below a mean score of 3.25 or edited them for clarity. Team members employed a similar process to assess the relative importance of each item and generated additional items to include in the final survey. Team members also attempted to ensure that survey respondents would have clear definitions of terms and would be able to provide explanations and narrative responses that did not fit existing categories.
IDENTIFYING AR PROGRAMS IN SPECIAL EDUCATION
Uncovering existing special education AR programs throughout the nation required persistence and multiple search methods. Initially, we consulted publications (e.g., Feistritzer & Chester, 2001) that specialize in listing AR teacher preparation \programs. However, we found most AR programs by contacting the special education and certification/licensure offices within state departments of education. We completed this process by using both traditional mail and electronic mail. Another method used in the identification process was to search each state's university and college education department Web site for information indicating whether it provided AR special education programs. If such information existed, NCPSE contacted a member of that department by e-mail or by telephone to seek information regarding these programs. The final method used in seeking programs was follow-up research on programs revealed at conferences or by educators in the field. NCPSE staff placed follow- up telephone calls to each state department of education to ensure that we had included all known programs. During all contacts, we defined AR programs as preparation routes that provided access to a teacher credential and that circumvented conventional college and university preparation programs (Hawley, 1992). Before administering the survey, we only had limited information (e.g., location and primary program sponsor) about the identified programs, so we were unable to make meaningful comparisons between responding and nonresponding programs.
ADMINISTERING THE SURVEY
Trained telephone interviewers at the University of Florida's Bureau of Economic Business Research (BEBR) Telephone Survey Lab conducted the survey. Interviewers entered program directors' responses contemporaneously, a process facilitated by the questionnaire's use of multiple-choice and numeric formats. Entries were stored automatically in memory; and preprogrammed follow-up questions, when appropriate, were displayed. However, because of the unique nature of AR programs, many items included a text box that allowed for responses not provided in the survey, as well as statements that could qualify certain responses. BEBR collected survey data from March until September 2003. BEBR standard procedure is to use a computer program to call potential respondents repeatedly when there is no answer, the line is busy, or respondents are unavailable to talk (in which case, BEBR schedules a follow-up call); and BEBR maintains a careful log of contacts. The interviews took from 30 to 45 min to complete, and BEBR called respondents back if they needed additional time to think about the items or if they had other matters that conflicted with the interview. For each item, BEBR generated frequency distributions and two lists, one for other responses and one for text-box comments.
RESPONDENTS TO SURVEY: GENERAL INFORMATION
We identified a total of 235 AR programs in special education, of which 101 program directors responded to the survey. The 235 programs were located in 35 states and Washington, DC, with California (32.8%) and Texas (16.6%) accounting for the largest proportion of programs. The responses from 101 of the 235 programs represented a 42.9% response rate. The 101 programs that responded were in 25 states and Washington, DC. Of these programs, 51, or 50.5%, were in four states: 17 (16.8%) in Texas, 16 (15.8%) in California, 10 (9.9%) in Georgia, and 8 (7.9%) in Louisiana.
To determine the association between the prevalence of AR programs in individual states and the shortage of special educators within those states, we correlated the number of teachers who were not fully certified (U.S. Department of Education, 2005) with the number of AR programs. The resulting correlation, .64, was significant at the .01 level. Nationally, the overall percentage of special education teachers who are not fully certified has averaged 12.5%. In the 15 states that lacked identified AR programs, the average percentage of special education teachers who were not fully certified was 5.8%. In the four states with the highest number of AR programs, the average percentage of teachers not fully certified was 22.6%.
We also noted that many of the AR programs have been developed within the past few years. Of the 98 responding programs, 39.8% had been in operation for 3 years or less, and 51% had been in operation for 5 years or less. (Although we surveyed 101 programs, those interviewed did not always respond to each item. Throughout the results section, we indicate the number of participants who responded to the individual items.) Approximately 44% of the program directors indicated that their programs were operating at enrollment capacity, 17% reported being above capacity, and approximately 39% indicated that they were below capacity. Reasons for the shortfalls included program entry requirements, insufficient funding, and lack of awareness about the program.
AREA I: PROGRAM INFRASTRUCTURE
Regarding program infrastructure, we focused on funding, groups targeted by funding, groups involved in the design of the program, the agency that recommends that licenses be issued, and teaching standards (see Table 2).
Program Funding. The survey addressed several items related to program funding. First, the survey asked respondents to note all agencies responsible for funding their programs: 57.1% of respondents mentioned the state education agency (SEA) as the current funding source, 52.9% mentioned institutions of higher education (IHEs), 39.8% mentioned the local education agency (LEA), and 31.6% mentioned the federal government. Among the most common funding patterns were local education agency-state education agency (LEA-SEA) and IHE-SEA partnerships, accounting for 20.7% each. IHE- SEA-LEA funding partnerships accounted for 14.6% of the respondents. The percentage of respondents that reported that tuition was the current funding source was 10.2%. Approximately 32% of the respondents indicated that a federal grant assisted in the start-up of the program and that the program targeted specific groups for recruitment.
Groups Targeted by Funding. The 33 programs that used funding for specific groups were interested in recruiting a diverse group of participants: 45.5% of the respondents targeted both African Americans and Hispanics, 42.4% targeted Native Americans, and 39.4% targeted Asian/Pacific Islanders. The programs targeted retired persons more often than individuals with any other career status (57.6% of respondents), 48.5% of respondents targeted general educators and mid-career changers, 36.4% targeted recent bachelor's degree recipients, 30.3% targeted military service personnel, and 27.3% targeted paraeducators.
Program Design. In considering program design, we were interested in identifying stakeholders who played an active role in developing the AR program. The survey first asked respondents to note all agencies primarily responsible for the design of the program: 75.8% of the respondents reported that IHEs were the primary program designer, 71.7% reported that SEAs were responsible for program design, and approximately half the respondents (48.5%) indicated that LEAs were responsible for program design. Second, we were interested in determining the number of responding programs in which one agency was primarily responsible for the design of the program and in determining the extent and nature of design partnerships. Respondents mentioned the IHE-LEA-SEA collaboration most often, (approximately one third of the respondents), followed by IHE-SEA collaboration, with 22.6% of the respondents. Relatively few of the respondents engaged in program design alone; IHEs, SEAs, and LEAs, by themselves, accounted for 12.9%, 12.9%, and 3.2%, respectively, of the responses.
Agency Recommending Issuance of License. Although most programs shared design input, one lead agency made recommendations to the state that a license be issued. Of the 99 programs responding, 67% of the respondents identified IHEs as the lead agency, SEAs recommended that licenses be granted in 16.2% of the programs, and LEAs recommended that licenses be granted in 14.1% of the programs.
Teaching Standards. Most respondents, 87.8%, indicated that program designers used nationally recognized professional teaching standards to develop their program. Of the 86 respondents who used standards, 82.5% reported the use of CEC professional standards in designing the program, 36% used the Interstate New Teacher Assessment and Support Consortium (INTASC), and 34.8% used the National Board for Professional Teaching Standards (NBPTS). A considerable proportion of respondents (43%) employed other standards in program design. For the most part, these standards were individual state standards. Although many respondents reported the use of professional standards in designing their programs, approximately one half (52.3%) of the respondents reported that their programs had received CEC/NCATE approval.
AREA II: PROGRAM LENGTH AND INTENSITY
Table 3 summarizes data associated with program length and intensity. Central to assessing program length and intensity was determining whether the program required full-time teaching in schools and the length of the preparation that participants received before assuming full classroom teaching responsibilities. We were also interested in the actual length of the training program, so we assessed program duration by using a full professional teaching credential as the end point. Of the 101 respondents, 84.2% reported that their program requires full-time teaching in school, and 85.8% used university credits as an index of progress. Approximately one half of the credit-earning programs were part-time programs (less than 11 credits per semester), and the remainder were full-time programs.
In those programs that required full-time teaching, 14.5% of teachers received no training before beginning teaching responsibilities. Approximately one half of the respondents, 51.9%, reported that participants received 3 months or less of training before assuming teaching responsibilities, and approximately two thirds (67.5%) required 6 mo\nths or less. Of the 93 programs responding, approximately 13% required 10 to 12 months of participation, and approximately one third (31.2%) of the programs awarded full licensure in 18 months or less. The largest number of respondents (approximately 34.4%) reported that the program required 24 months of program participation.
Program Infrastructure Data Summary
To develop a factor that would reflect critical aspects associated with fast-track preparation for special education teachers, we combined two measures: the length of preparation, in months, that program participants receive before assuming classroom (teaching) responsibilities while enrolled in the program; and the length of time that participants are in the program, in months, before they receive the full professional teaching license for which the program prepares them. We characterized 3 months or less for entry into the classroom as being a brief amount of preparation time before assuming classroom responsibilities and considered 4 months or more as an extended amount of time. Regarding length of time in a program before receiving a full professional license, we characterized 18 months or less as a brief amount of time and viewed programs that required more than 18 months as extended. Data on these two variables combined were available for 80 of the programs: 30% of the programs (n = 24) were extended in both entry to the classroom and in program length; 18.7% (n = 15) were extended in classroom entry but brief in program length; 40% of the programs (n = 32) were brief in classroom entry and extended in program length; and 11.3% (n = 9) of the programs were fast track, that is, brief for both classroom entry and program length.
Program Length and Intensity: Data Summary
AREA III: PROGRAM CHARACTERISTICS
In considering program characteristics, we asked program directors about the requirements for admission and program selectivity, types of degrees and licenses granted by the program, modes of instructional delivery and school-based mentorship, financial support, and program completion rates. Table 4 summarizes these data.
Admission and Selectivity. We asked respondents to identify all requirements for entry into the AR program. Most programs required participants to have a minimum GPA (86.1%) and a bachelor's degree (87.1%). Two thirds to three fourths of the programs required an interview (68.3%), a standardized test (70.3%), and letters of recommendation (74.3%). Respondents reported that 42.3% of the programs accepted 81% to 100% of their applicants, and an additional one third (33.3%) accepted 61% to 80% of applicants. However, when asked about the selectivity of their programs, 30.7% of the respondents characterized their programs as being very selective, 62.4% viewed their programs as selective, and only 6.3% indicated that their programs were not selective. Thus, although most respondents (93.7%) view their programs as selective, more than 42.3% accept between 81% and 100% of all applicants.
Completion. Respondents estimated the number and percentage of program completers during a typical year. Most programs tend to have a sizable number of yearly program completers, with 40.6% of respondents indicating that 31 or more participants finish their program. A rather small percentage (13.2%) of respondents indicated that 10 or fewer persons complete their programs each year. Approximately two thirds reported that between 90% and 100% of their participants completed the program in the previous year; only 12% of the programs had a completion rate of less than 75%.
Degrees, Licenses, and Age and Grade Emphases. Although 86% of respondents reported that participants need a bachelor's degree to enter their program, only 32.7% of the 101 responders indicated that participants receive a degree after completing the program. Many programs (47.5%) do not offer a degree, and approximately 20% of the programs provide an option for earning a degree. Of those that offer a degree, 61% offer only a graduate degree, 13.8% offer an undergraduate degree, and 25% offer both graduate and undergraduate degrees.
The programs offered a broad range of license types. When asked to note all the options that applied to their programs, 90.1% of respondents reported that their program offers an initial license, 42.6% reported that their program offers an add-on license to a previously earned license, and approximately 25% noted that their programs provide an opportunity to earn a dual or blended license. Another 25% of the respondents indicated that they provided information specific to the language of their state regulations (e.g., Class A, probationary, or credential rather than license).
Our variable data regarding age and grade levels that the AR programs address are consistent with the expansive age requirements typical of most state's certification requirements (see Geiger, Crutchfield, & Mainzer, 2003). Of the 96 respondents, 50% indicated that their programs address early childhood through secondary special education and 41.6% of responding programs reported that their licenses cover elementary/secondary special education. Only 8.3% indicated that their programs limited their emphasis to a particular age or grade group.
Modes of Instructional Delivery and Mentorship. We asked respondents to identify all modes of instructional delivery that their program employed. Most programs included supervised fieldwork (94%) and university or college-based coursework (89%); 68% of the programs employed distance learning; and 58% used districtbased staff development. Other methods, identified by 9% of the programs, included summer institutes and independent studies.
As the primary location of the instruction, IHEs accounted for the largest proportion, 43.9%, among the respondents. Approximately 30% reported that the program used a combination of IHE and LEA locations, and 13.3% reported that the LEA alone housed instructional efforts. Of the 13% who reported another location, most were regional resource or service centers for several states. Many programs (90.1%) reported that their AR programs include a mentorship component. We asked respondents to identify all types of mentorship and supervision that their programs provided. Most respondents, 91.1%, indicated that their program provided mentoring by the LEA, and 57.4% indicated that a university provided mentoring. Most of the 21.8% who reported another type of mentoring cited supervision options emanating from regional resource centers.
Program Characteristics: Data Summary
Accommodations for Individuals With Disabilities and Financial Support. Virtually all the respondents (99%) reported that their AR programs provide accommodations for those with disabilities. These accommodations include extra time for examinations and assignments, as well as the provision of adaptive technologies.
Approximately 90% of the respondents indicated that their programs provide one or more kinds of financial assistance for program participants, and approximately one half of the respondents (53.9%) indicated that the program furnishes tuition assistance. Approximately 40% provide salary support, and 31.5% provide some type of stipend. With many programs requiring full-time teaching in schools, it is likely that cooperating school districts are paying salaries. Other types of support that the respondents reported include eligibility for financial aid in the forms of loans, variable yet undefined state grants, and performance bonuses on program completion.
Participant Demographics: Data Summary
AREA IV: PARTICIPANT DEMOGRAPHICS
In addressing the area of participant demographics, we were interested in the demographic composition of the responding programs, as well as the experience of program participants before program entry. Table 5 summarizes the data.
Gender and Age. Forty percent of the respondents indicated that their program enrolled a class that consisted of 20% men; approximately three fourths of the respondents indicated that their programs included 30% or fewer males. The average percentage of females (75.5%) in AR prograrns, outnumbered by three to one the average percentage of males, (24.5%). Aggregating the age data indicated that more than one half (52.4%) of the participants in the responding programs were more than 31 years old; 26.4% of the participants ranged from 25 to 30; and in 21.2% of the programs, the average age ranged from 20 to 24.
Race. In the 33 programs that had funding to recruit specific groups of individuals, 45% of respondents indicated that they were seeking African American or Hispanic participants, 42% indicated they were seeking Native Americans, and 39% indicated that they were seeking Asians and Pacific Islanders. The largest percentage of students actually participating in the programs were White (70.5%); African American students represented 16.2% of participants, and Hispanic students represented 10.2%. Approximately one half of the programs (50.6%) reported that 80% to 100% of their participants were White; 42.5% of the programs reported having between 0% and 5% African Americans (14.5% reported having none), and 60% of the programs reported having between 0% and 5% Hispanic students (31.3% reported having none). Approximately 25% of the programs reported enrollments that included more than 20% African American participants, and 15% reported that their enrollments included more than 15% Hispanic participants. Involvement of other racial groups was quite limited.
Participant Experience Before Program Entry. We asked respondents about students' career experiences before they entered AR programs. Specifically, we were interested in the proportion of program participants who were retired, retired from military service, recent recipients of bachelor's degrees, career changers, general educators, and paraprofessionals. The average percentag\e of retirees in programs was small (2.7%); approximately 62% of the respondents indicated that their programs included no retirees. At the upper end, only 7.4% of the programs reported having a percentage of retirees ranging from 13% to 15%. The average percentage of participants who were retired from military service was also quite small (3.6%); approximately one half of the respondents indicated that no participants had retired from military service; 82% reported 5% or less.
The average proportion of AR program participants who had been recent recipients of bachelor's degrees (that is, within the previous 12 months) was quite sizable (28.7%). Most respondents reported that 1% to 10% of their program's participants were recent recipients of bachelor's degrees, yet 17.3% of the respondents indicated that 70% to 100% of their program's participants were recent college graduates. The average proportion of AR participants identified as midcareer changers was 45.9%. Approximately 26.6% of the respondents indicated that their programs consisted of between 71% and 100% midcareer changers; 10.1% of the respondents indicated that the program had no career changers.
The average proportion of AR program participants identified as general educators was a considerable 22.8%. Although a large proportion of respondents, 40%, reported that their program had no general educators, approximately 15% of the respondents indicated that their programs included between 61% and 100% general educators. Finally, the average proportion of AR special education students identified as paraeducators was 11.4%. Many respondents (45.1%) reported that their programs had no paraeducators, and 39% of the respondents indicated that their programs included between 1% and 20% paraeducators. No respondents reported that their programs were designed exclusively for paraeducators.
The results of this study confirm that a large number of stakeholders are designing and implementing a variety of alternative special education teacher training programs. Previous efforts (Rosenberg & Sindelar, 2001, 2005) to define the essential elements of AR programs were reviews of a very limited literature. Those efforts led to the conclusion that AR programs can be effective when they meet certain programmatic conditions.
In analyzing the data from the current study, several key themes emerged. First, the shortage of certified special educators appears to be the impetus for the proliferation of AR programs. second, although a defining characteristic of such programs is access to a teaching credential that circumvents conventional university preparation, IHEs are very much involved in the AR enterprise. Third, although most programs require fulltime teaching, the length of preparation and support vary greatly. Finally, the characteristics of the participants recruited into programs raise a number of issues that question whether the proliferation of AR programs in special education, as currently designed, is actually increasing the supply of certified and (in the aftermath of the reauthorized IDEA of 2004) highly qualified special education teachers.
AR PROGRAMS AND THE SHORTAGE OF SPECIAL EDUCATORS
Three major factors have contributed to the rapid growth of AR programs: the persistent shortage of certified personnel, the acute need for teachers from CLD backgrounds, and sentiment from political action groups (and the U.S. Department of Education) that traditional approaches to teacher education are flawed and require streamlining to ensure that rigid programs do not discourage those who wish to become teachers (Hardman & Muldur, 2003; Rosenberg & Sindelar, 2001; Rosenberg et al., 2004). In spite of the intense and often vitriolic rhetoric surrounding the shortcomings of conventional teacher preparation (e.g., Hess et al., 2004; Thomas B. Fordham Foundation, 1999), such criticism is rarely evident in discussions of special education teacher development. There is little argument that students who need special education require teachers with a wide range of specialized skills. In fact, many arguments surrounding what constitutes a highly qualified special education teacher involve the need for professionals who have content area expertise and the ability to provide accommodations for learning and behavioral support.
Our data provide additional evidence that the driving force behind the proliferation of special education AR programs is the acute shortage of certified and highly qualified special education teachers. Specifically, in our comparison of states with high numbers of AR programs and those that lack any programs, sharp differences occurred in the percentages of uncertified special education teachers. Although we cannot assert causality in this relationship, a link clearly exists between the severity of the shortage of certified special educators and the number of AR programs in individual states. One implication of these data is that the design of AR programs in states with severe shortages must ensure that candidates persist in their new careers. Effective AR programs are extended, rigorous, and programmatic; fast-track programs with limited support have high attrition (Darling-Hammond, 1999; Rosenberg & Sindelar, 2001). Because all AR programs are not created equally, states with large numbers of such programs should actively monitor the nature and extent of the programs' training activities and should ensure that only programs that result in high- quality teachers who persist over time receive approval and funding.
Furthermore, our findings regarding the diversity of participants in special education AR programs are consistent with findings from research on general and special education (e.g., Humphrey & Weschler, 2005; Tyler et al., 2004). Specifically, on average, the proportion of African American and Hispanic participants in AR programs exceeds their representation in traditional programs. In our data, this difference was attributable to a relatively small subset of AR programs that attracted highly diverse participant pools. These special education programs were often in highly diverse communities (as Humphrey and Weschler concluded about the general education programs that they studied). Thus, neither research about general education nor research about special education has identified inherent differences between alternative and traditional preparation-beyond the demographics of the populations from which they draw participants-that might account for the relative success of alternative preparation in attracting diverse participant pools.
THE INVOLVEMENT IN AR PROGRAMS
In recent years, a considerable amount of criticism has focused on how IHEs develop teachers (and how long it takes), and corresponding calls have requested that states break the monopoly of universities and consider empowering approved multiple providers for teacher preparation (e.g., Hassel & Sherburne, 2004). Rosenberg and Sindelar (1998) expressed concern that speed, convenience, and practicality are the primary methods used to entice students to change careers in this new marketplace. An even greater concern was the insidious effect that private-sector, for-profit institutions, such as the University of Phoenix, might have on mainstream providers of teacher preparation. Many feared that bottom-line economic factors (i.e., recruiting students and generating credit- hour enrollments) would influence the design, implementation, and viability of preparation programs.
Instead of being forced out of the teacher-preparation marketplace, traditional IHEs have responded to LEA concerns about shortages of special education teachers and have been instrumental in designing and delivering either some or all of the critical components of AR programs. In fact, mainstream IHEs were primarily responsible for the design of 76% of the programs that we surveyed; 89% of the programs used university coursework, with 73% of the programs employing IHEs as the venues for instruction. In 64.5% of the programs, IHEs were in partnerships with either or both LEAs and SEAs. Because most of the programs surveyed required classroom teaching, very few IHE programs were independent of substantive SEA or LEA involvement.
Why are IHEs so active in the AR marketplace? Rotherham and Mead (2004) suggest that IHEs embrace AR programs because such programs provide universities with the opportunity to diversify their considerable financial teacher education portfolio. With more than 200,000 undergraduate and graduate completers each year, teacher preparation programs nationally bring in more than $6.1 billion in tuition annually. Because of their low average faculty costs and an increasing reliance on adjunct instructors, teacher preparation programs are often seen as cash cows for their universities. As demand for traditional teacher preparation shrinks-among those who are teaching students with special education needs and are uncredentialed, 24% are bypassing traditional university and college programs (Connelly, 2003)-AR programs represent an effective means for mainstream universities to expand their offerings into the growing marketplace of nontraditional adult learners with little additional capital expenditure.
Nonetheless, many questions remain regarding the transformation of university programs for preparing special education teachers. For example, how will an institution's offering of multiple options for preparing special education teachers fare in the real world of practically minded students? Will the proliferation of AR programs result in the eventual extinction of comprehensive and pedagogically advanced traditional programs in the marketplace? Will the proliferation of AR programs reduce the shortage of highly qualified special education teachers? The authors' experiences have indicated that teacher candidates often take the path of least resistance to obtain a \teaching credential. Credentialism rather than learning is a dominant theme in our changing academic marketplace, and many who are new to the field and are allowed to merely meet reduced standards may do just that (Rosenberg & Sindelar, 1998; Zemsky, 1993). IHEs must face the new fiscal realities of the academic marketplace, that is, students want rapid and convenient access to needed requirements and the competition from nontraditional vendors of teacher preparation is considerable. Therefore, IHEs, the onetime mainstream provider of teacher education, may be willing to alter their offerings to maintain their place in the process. With multiple user-friendly paths to certification and almost immediate employment available in certain public schools, we must be careful that all approved programs include rigorous teacher education activities.
LENGTH OF PREPARATION AND SUPPORT
Because the shortage of certified-and now highly qualified- special education teachers is at a chronic crisis level, AR programs naturally seek to place qualified personnel in classrooms as quickly as possible. For the most part, respondents to our survey are doing just that-84% of the 101 programs require full-time teaching in schools. However, considerable variation exists in the amount of preparation that candidates receive before assuming classroom responsibilities and the amount of support that they receive throughout their program. In approximately one half, or 51.9%, of the programs that require full-time teaching, candidates receive 3 months or less of training before entering the classroom as the teacher of record, and 14.5% of the programs require no training at all. Although we know few specifics about the intensity of training during these periods, a major share of preparation necessary for success in the classroom clearly occurs while teacher candidates are on the job. Consequently, we assessed the amount of time that candidates were required to participate in the AR programs before receiving certification. Most programs require an ample amount of time, with approximately two thirds of the programs, 68.8%, requiring 18 months or more of participation in program activities.
Placing a novice teacher with limited training in a classroom of challenging and needy students with disabilities is a high-risk endeavor. The knowledge and skill base recommended for entry-level teachers of students with disabilities is considerable and includes competence in assessing learning differences, instructional planning, implementing research-based instructional strategies, and professional and ethical practice, as well as collaboration with families and fellow service providers (CEC, 2003). However, complementing the time in the classroom with substantive standards- based training and effective support minimizes the level of risk involved with an inexperienced teacher. To assess the extent that programs were providing training and support for on-the-job AR programs, we developed a factor that combined length of preparation before assuming teaching responsibilities and length of time participating in program activities. Most programs reflected a desire to provide adequate support for the AR candidates. Of the 80 programs that responded to the two items, 30% were extended for both entry to the classroom and for required program participation-more than 3 months of time before assuming teaching responsibilities and completion of a program that lasted more than 18 months; and 40% of the programs were fast in entry to the classroom (3 months or less) but had extended program length. Of those with limited program length (that is, less than 18 months), 18.7% were extended in entry to classroom teaching and 11.7% were fast.
We believe that this finding is a generally positive one. Regardless of the amount of time before participants assume full teaching responsibilities, most AR programs have a duration of more than 18 months. Since almost all the programs provide mentoring and supervision (more than 90% reported providing mentorship and supervised fieldwork) and a broad range of instructional activities, many novice special educators in AR programs are being educated and supported. However, we remain concerned that a significant proportion of programs, 11.7%, or one in every eight programs, include a fast track to the classroom and have limited program length. Such programs may arguably be appropriate for certified elementary educators or for developing secondary-level content-area teachers who are proficient in their areas of content but need basic pedagogical skills. However, in this era of educational accountability, newly prepared special education teachers are responsible for increasing the performance of their students, a challenge that requires enhanced instructional skills; expertise in diagnosing and assessing learning problems, instructional strategies for inefficient learners, and behavior management; and myriad skills in collaboration, consultation, and advocacy. Fortunately, most AR programs are making the effort to deliver supportive programs that include enough time to promote successful induction to the profession. Moreover, by providing on-the-job training as part of this support, AR programs are reducing the opportunity costs associated with professional development and induction, making the programs both instructionally supportive and financially attractive.
PARTICIPANTS IN AR PROGRAMS
In addition to circumventing traditional university methods of teacher preparation and facilitating the entry of members of underrepresented CLD groups into teaching, AR programs provide a means to enter teaching for individuals who lack traditional undergraduate teacher preparation. It is particularly noteworthy that the average proportion of students who were general educators was a sizable 22.8%, indicating that many credentialed educators are availing themselves of additional activities to attain certification. This phenomenon represents a new trend.
Boe, Cook, Bobbin, and Terhanian (1998) noted that the proportion of special education teachers who transferred to general education (8.8%) was significantly greater than the proportion of general education teachers who transferred to special education (1%). However, only a small net loss of special education teachers occurred because the number of general educators is nine times the number of special education teachers. Some general educators enrolled in the responding programs may have been unable to obtain a position in their chosen area and sought entry to teaching by accessing an AR program in special education. Whatever their reasons, these general educators represent a considerable proportion of participants entering the field. Moreover, with their background in education, these candidates are unlikely to require the same intensity of training and support needed by those who have little or no relevant background or experience. However, these candidates may be using the special education position as a way to enter a particular school system, making them a risky investment when general education positions become available.
By far the largest group of participants, approximately 46%, consisted of midcareer changers. AR programs desire such individuals because they represent a new and large pool of potential special education teachers. Reviewing the relative advantages and disadvantages of recruiting midcareer changers in AR programs, Dai, Denslow, Dewey, Sindelar, and Rosenberg (2006) noted that since most midcareer changers hold bachelors degrees, programs can often streamline the training process, particularly when trainees bring to training related experiences and backgrounds that can facilitate rapid acquisition of concepts related to education. Unfortunately, some midcareer changers are also high-risk prospects, particularly those who take substantial pay cuts to enter the field and those whose initial careers are least like teaching. Indeed, some individuals switch careers because of altruistic reasons; however, those who experience deep pay cuts are likely to return to their more lucrative careers when opportunities arise, and those who had careers far different from teaching may discover that they are not comfortable working with children and adolescents or being constrained by the routines of a school environment. Because large numbers of career changers enter AR programs and because both good and bad reasons exist for changing careers, we believe that programs must assess the motivation and dispositions of midcareer changers carefully and systematically.
The large percentage of individuals who were recent recipients of bachelor's degrees is also noteworthy. Across programs, approximately 28.7% of the participants had earned their degree less than 12 months before entering an AR program. Why do so many young adults enter special education AR programs? Perhaps the reason is an inability to enter the job market in their chosen field or a desire to pursue time-limited public service (e.g., Teach for America) through a temporary teaching assignment. Another possibility is that many of these individuals completed general education programs and were able to secure a teaching position only as a special educator. In fact, additional analysis of our data indicates that approximately 25% of those with recent bachelors degrees are also general educators. We believe that further analyses of these recent graduates are important; those with recent degrees constitute a diverse group of teaching candidates who, because of their previous experience and long-term goals, may require individualized levels and intensity of training. As mentioned previously, those who are general educators will likely require less direct instruction in certain aspects of pedagogy and child development. Regarding those well-meaning young people who are pursuing special education teaching as atemporary public service activity before enrolling in a professional school that is not related to education, policy makers will need to determine the relative benefits of allocating extensive and scarce training resources to inexperienced personnel who have little intention of staying in the field.
Finally, the rapid growth of special education AR programs has been justified on the grounds that schools and colleges of education, the traditional producers of special educators, are unable to keep up with the demands of the market. The objective of expanding the number of AR programs is to increase the existing pool of potential special educators and ultimately reduce the shortage of highly qualified teachers. Clearly, the large numbers of midcareer changers-particularly those who would be unable to change to a different profession unless they received a salary or faced low opportunity costs for the training needed to enter a different profession-represent a new supply of potential special education teachers. However, one could argue that other participants in AR programs, particularly those with recent bachelor's degrees and some of those who are general educators, would gravitate to existing programs if AR access were not easily available. We fear that new (and costly) programs recruiting such individuals may cannibalize existing programs and result in no real increase in the needed supply. Individuals may be merely migrating from comprehensive programs with extensive requirements to programs that offer the lure of an immediate salary yet a more limited opportunity to acquire, master, and reflect on evidenced-based practices. The impact of AR programs on the actual supply of certified special education teachers and on the design of traditional teacher preparation programs remains to be seen. Future efforts will need to determine whether widening program offerings is actually increasing supply or whether policy initiatives have only served to shepherd a segment of the special education teacher market around many of the professional standards and requirements that we hold as essential for success with students with disabilities.
CONCLUSIONS AND RECOMMENDATIONS
On the basis of our analyses of the broad programmatic features, we conclude the following about AR programs in special education.
* Large numbers of AR programs exist, particularly in states that experience significant shortages of special educators. Instead of being eased out of the AR market, many IHEs have streamlined their offerings and are-usually in partnerships with SEAs and LEAs- providing certification programs that feature the opportunity to earn a salary during on-the-job training. We are uncertain, however, of the long-term effects that streamlined programs will have on the quality of special education teachers entering the classroom.
* Although most programs feature rapid entry to classroom teaching, many programs provide extended opportunities for training through coursework, distance education, and on-site support through university supervision and mentoring. Unfortunately, a significant number of programs-11% of our respondents-do not provide an adequate amount of support for candidates who are placed rapidly in the classroom. AR programs are heterogeneous in length, support, and program intensity; and educators and policy makers must not view programs that deliver a research-based curriculum in a coherent programmatic fashion in the same light as those that make little or no effort to do so.
* Most AR programs in special education adhere to the U.S. Department of Education's (2004) suggested practice of recruiting widely and selecting carefully. Special education AR program respondents employ a wide range of strategies to recruit students and consider themselves selective in their choice of candidates. Most programs require a bachelor's degree for program entry, and relatively few offer an additional degree as part of the program. Students in AR programs tend to be older than students in traditional programs; but in both groups, females outnumber males.
* The backgrounds of participants who are recruited into AR programs and who ultimately persist in their new careers are a rich and important source of data for researchers and policy makers. By being aware of these data, we can begin to form generalizations regarding the backgrounds that may be best suited for success in an AR program. For example, we found many candidates with recent degrees who were also general educators, indicating a significant movement from general education to special education. Such individuals appear, at first, to be excellent candidates for AR programs-their previous experiences serve as a solid foundation for teaching and may eliminate the need for some basic aspects of a teacher preparation program. However, such candidates are also risky: They may have been unable to secure a general education position, they may use special education as a method of entering teaching, and they may plan to leave their current position if one in their chosen area becomes available. Programs for these individuals may not be cost-effective, because the commitment to special education may be short-lived and may not result in a sustained increase in the supply of new teachers.
* The largest group of AR participants is midcareer changers. Such personnel are desirable because they represent a new pool of potential special educators. However, midcareer changers are also risky candidates; those who migrated from positions with high salaries or professions that were not similar to teaching are likely to return to their previous careers if lucrative opportunities arise. Consequently, AR programs must institute selection criteria that carefully assess the motivations and dispositions of those who turn to education after a career in another segment of the economy.
* Many special education AR programs appear similar to traditional teacher preparation programs in a number of ways. A large percentage of AR programs have considerable IHE involvement, employ nationally recognized teaching standards, require coursework and supervised fieldwork, provide mentors, and claim to be selective in admission. Nonetheless, several critical factors appear to differentiate AR programs from traditional preparation options. These factors include the length, intensity, and sequence of training activities; the amount of preparation time before entering the classroom; and the characteristics and life experiences of those who choose to enter programs.
We recognize the limited generality of our findings and urge readers to be circumspect in their interpretations of our results. Our experience in this study and elsewhere suggests that the life of many AR programs may be fleeting. Two of the authors have undertaken AR programs of our own. Although our programs were OSEP-funded, state departments of education and districts also fund similar programs. These programs have come and gone, making them more like projects than like established programs. In defining our universe, however, we did not make this differentiation; as a result, our universe is inconstant and difficult to sample reliably. Other evidence of the ephemeral nature of our universe came from colleagues in the Telephone Survey Lab, who reported a large number of nonworking numbers (n = 55). We took this finding to mean that many of the programs that we had identified were no longer in existence 1 year later. Several colleagues who were familiar with programs in their home states have substantiated this speculation by recognizing programs in our database that were no longer operating. In short, because some AR programs are short-lived, the universe of programs is inconstant, and no sample can represent it reliably. By the time researchers take and survey a sample, the universe from which it is drawn will have changed. Although not necessarily representative, the responses from 101 programs form a substantial and illustrative source of information.
Second, we chose not to differentiate between individuals who explicitly refused to participate in the survey-true nonrespondents- and others who we could not reach at the numbers that we called, who had nonworking lines, or who did not answer calls in spite of frequent callbacks. In this sense, we believe that our estimate of the completion rate is low. In fact, of the 134 individuals who did not complete the survey, only 31 explicitly refused to participate. Others either could not be contacted despite repeated attempts, or their programs were no longer operating. Thus, the upper bound of percentage completion is 101/132, or 76.5%, and the true value lies somewhere between 76.5% and the minimum, 43%.
Finally, we call readers' attention to the potential bias in the self-report format of our survey items. Of course, biased responding is most likely when questions have a recognizable valence, as do our questions regarding standards and availability of mentorship. Who would disregard standards in developing a program? Who would not see value added by including a mentoring component? We urge readers to exercise caution in interpreting these findings and perhaps others.
With these limitations in mind, we recall that in a previous review of the literature on AR programs in special education, we (Rosenberg & Sindelar, 2001, 2005) characterized the limited database as an iceberg with a small visible portion of published research and conference presentations and most of the AR enterprise below the surface. We now know more of the nature and extent of AR programs and can begin to use these data to judge the long-term efficacy of various types of programs. After identifying program characteristics, we can assess efficacy through measures of attrition, cost, and teacher performance. Clearly, the marketplace has responded innovatively to the critical need for special educat\ors. The pressing need to develop solutions for the shortage has challenged all of us to revise our conventional thinking about now to best prepare special education teachers rapidly. It is essential, however, that we evaluate our efforts to ensure that solutions proposed in the new marketplace meet professional standards reliably and do not cut deeply into the essential elements of quality teacher preparation.
Uncovering existing special education AR programs throughout the nation required persistence and multiple search methods.
AR programs represent an effective means for mainstream universities to expand their offenngs into the growing marketplace of nontraditional adult learners with little additional capital expenditure.
Boe, E. E., Cook, L. H., Bobbin, S. A., & Terhanian, G. (1998). The shortage of fully certified teachers in special and general education. Teacher Education and Special Education, 21(1), 1-21.
Connelly, V. J. (2003, November). An analysis of alternatively prepared teachers using the school and staffing survey. Paper presented at the meeting of the Teacher Education Division, Biloxi, MS.
Council for Exceptional Children. (2003). What every special educator must know: The standards for the preparation and licensure of special educators (5th ed.). Reston, VA: Author.
Dai, C. Denslow, D., Dewey, J., Sindelar, P. T., & Rosenberg, M. S. (2006). Economic analysis and the design of alternative route teacher education programs. Manuscript submitted for publication.
Darling-Hammond, L. (1999). Solving the dilemmas of teacher supply, demand, and standards. Quality Teaching, 9, 3-4.
Feistritzer, C. E., & Chester, D. T. (2001). Alternative teacher certification. Washington, DC: National Center for Education Information.
Feistritzer, C. E., Haar, C. K., Hobar, J. H., & Scullion, A. B. (2005). Alternative teacher certification. Washington DC: National Center for Education Information.
Geiger, W. L., Crutchfield, M. D., & Mainzer, R. (2003). The status of licensure of special education teachers in the 21st century (COPSSE Document No. RS-7). Gainesville, FL: University of Florida, Center on Personnel Studies in Special Education.
Goldhaber, D. D. (2004). Why do we license teachers? In F. M. Hess, A. J. Rotherham, & K. Walsh (Eds.), A qualified teacher in every classroom? Appraising old answers and new ideas (pp. 81-100). Cambridge, MA: Harvard Education Press.
Hardman, M. L, & Mulder, M. A. (2003). Federal education reform: Critical issues in public education and the impact on students with disabilities. Eagle Summit on Personnel Preparation in Behavior and Emotional Disorders. Dallas: University of North Texas.
Hassel, B. C., & Sherburne, M. E. (2004). Model 2: Cultivating success through multiple providers: A new state strategy for improving the quality of teacher preparation. In F. M. Hess, A. J. Rotherham, & K. Walsh (Eds.), A qualified teacher in every classroom? Appraising old answers and new ideas (pp. 201-222). Cambridge, MA: Harvard Education Press.
Hawley, W. D. (1992). The theory and practice of alternative certification! Implications for the improvement of teaching. In W. D. Hawley (Ed.), The alternative certification of teachers (pp. 3- 34). Washington, DC: ERIC Clearinghouse on Teacher Education.
Hess, F. M., Rotherham, A. J., & Walsh, K. (Eds.). (2004). A qualified teacher in every classroom? Appraising old answers and new ideas. Cambridge, MA: Harvard Education Press.
Hillkirk, R. K. (2000). Effective models for alternative programs in teacher education. In D. J. McIntyre & D. M. Byrd (Eds.), Research on effective models for teacher education (Association of Teacher Educators Yearbook No. 8; pp. 195-202). Thousand Oaks, CA: Corwin.
Humphrey, D. C., & Weschler, M. E. (2005). Insights into alternative certification: Initial findings from a national study. Teachers College Record. Retrieved April 4, 2006, from http:// www.tcrecord.org/content.asp?contentid= 12145
Neel, R. S., Cessna, K. K., Borock, J., & Bechard, S. (2003). Quality program indicators for children with emotional and behavioral disorders. Beyond Behavior, 12(3), 3-9.
Rosenberg, M. S., & Rock, E. E. (1994). Alternative certification in special education: Efficacy of a collaborative, field-based teacher preparation program. Teacher Education and Special Education, 17, 141-153.
Rosenberg, M. S., & Sindelar, P. T. (1998). The impact of market driven higher education practices on professional teacher development. Teacher Education and Special Education, 21, 227-235.
Rosenberg, M. S., & Sindelar, P. T. (2001). The proliferation of alternative routes to certification in special education: A critical review of the literature. Arlington, VA: Council for Exceptional Children, National Clearinghouse for Professions in Special Education.
Rosenberg, M. S., & Sindelar, P. T. (2005). The proliferation of alternative routes to certification in special education: A critical review of the literature. Journal of Special Education, 39, 117- 127.
Rosenberg, M. S., Sindelar, P. T., & Hardman, M. L. (2004). Preparing highly qualified teachers for students with emotional or behavioral disorders: The impact of NCLB and IDEA. Behavioral Disorders, 29, 266-278.
Rotherham, A. J., & Mead, S. (2004). Back to the future: The history and politics of state teacher licensure and certification. In F. M. Hess, A. J. Rotherham, & K. Walsh (Eds.); A qualified teacher in every classroom? Appraising old answers and new ideas (pp. 11-47). Cambridge, MA: Harvard Education Press.
Thomas B. Fordham Foundation. (1999). The teachers we need and how to get more of them: A manifesto. Washington, DC: Author.
Tyler, N. C., Yzquierdo, Z., Lopez-Reyna, N., & Flippin, S. S. (2004). Cultural linguistic diversity and the special education workforce: A critical overview. The Journal of Special Education, 38, 22-38.
U.S. Department of Education. (2002). Meeting the highly qualified teacher challenge: The Secretary's annual report on teacher quality. Washington, DC: Office of Policy, Planning, and Innovation.
U.S. Department of Education. (2004). Alternative routes to certification. Washington, DC: Author.