The heart and soul of these high-quality alternative programs is the on-the-job supervision and support candidates receive as they face the daunting challenges of being a new teacher in what is often a very difficult classroom setting. In the programs studied, support is structured at three levels: (1) program-provided supervisors; (2) site-based mentors; and (3) peer cohort support. All six programs had some variation of these three, which interweave to form a new-teacher safety net. Rather than strand candidates to sink or swim, support structures ensure that candidates will fulfill their promise or, as McKibbin puts it, that "they will obtain the skills to succeed and the commitment to stay."

On-Site Supervision

All of these programs include direct and indirect support. Direct support comes in the form of classroom observations, done by the program (or university) supervisor, the mentor (who is often an experienced teacher at the candidate's site), or a school administrator, such as the principal, who has partnered with the program to provide such support.

In Georgia, mentors observe frequently, give candidates feedback, and act as role models by coaching and demonstrating lessons. They also arrange for candidates to visit and observe in other classrooms.

In Hillsborough, the coaching cycle is key to the program. Because candidates can enter at different times during the year, the program is organized into a series of observation-and-coaching "loops" within three cycles or phases, as depicted in figure 7. Within weeks 1-2, for example, the support provider—the candidate's program administrator—conducts a preobservation conference to schedule observation times and introduce the candidate to the Florida Performance Measurement System instrument. Observations will be based on Florida's Accomplished Practices for Educators, and the administrator will use this instrument in evaluating the candidate on those practices. The candidate will self-assesses on the same competencies. After the initial observation has taken place, the administrator and the candidate, together with a trained peer teacher, write an action plan to determine methods and time lines for addressing competencies that have not been successfully demonstrated. This plan guides subsequent observations and conferences and is updated at the end of each cycle.

Chico supervisors are also course instructors, ensuring that there is no disconnect between course work and classroom practice. As one Chico supervisor explains, "I know what's being taught in reading courses, and if I go out and see that it's not happening, I say, 'You just finished the course—where is it?'" On-site support is planned but also highly individualized—tailored according to Chico candidates' individualized plans and expressed needs. And the support team—supervisors, mentors, and school administrators—zeroes in on potential crises. "Need someone there next Tuesday?" queries another Chico supervisor. "We'll make that happen. We do visits on top of visits."

In all of the programs, support is carefully coordinated. In Georgia, supervisors facilitate regular reporting and communication. In Hillsborough, principals take that role. In Chico, it's the university supervisors, each regionally assigned and working with 10 to 15 mentors and roughly the same number of candidates. As they follow their candidates and link with mentors throughout the four semesters, Chico supervisors also communicate and develop rapport with school principals and other district or county education administrators.

All program leaders agree that the success of a support structure rests, fundamentally, on an environment of trust. Interns must continually give honest answers to the support providers' core question: "How are you doing?" Since candidates are simultaneously dealing with course work, teaching, supervision, and mentoring, everyone knows they are having a struggle. "In a traditional program, people expect your competencies to be there," says a graduate of Chico's program. "Here, supervisors know you will be floundering. I invited my supervisor to 'Come see my worst part of the day.' I was at four schools. She came to each one. She saw the diverse environments and knew my challenges. She understood. Then later, I said, 'Come again and see how much better I am doing.' There is no intimidation."

That sense of trust and bolstered confidence was echoed by a Georgia candidate: "From the beginning of the program, I felt I was set up to succeed."

On-Site Mentoring

While supervisors keep classroom practice grounded in course work, on-site mentors—"treasured advice givers," as one candidate called them—are critical to day-to-day survival. The programs pay strong attention to the selection and training of mentors, pay mentors a stipend, and are very clear about what is expected of them. (As an example, figure 8 is New York's mentor position description.) In Texas, mentors are selected by principals who receive guidance from Region XIII on what qualities to look for in a mentor. Mentors attend 15 hours of professional development provided by Region XIII. The mentor and each candidate must complete six observations during the school year—Region XIII suggests three times with the candidate observing the mentor and three with the mentor observing the candidate. In addition, the two also hold a minimum of four discussion meetings.

Georgia mentors—who are themselves classroom teachers—receive training on coaching and communication. They spend a minimum of 100 hours working with each candidate the first year and 50 hours the second year. One mentor responsibility is to support the candidate throughout all phases of the program by providing feedback based on the Danielson framework (see figure 6 on page 17).

Mentors in Hillsborough are former administrators. Not only do these individuals bring a wealth of expertise, but they have a vested interest in the district and can speak to principals and veteran teachers with the authority needed to make the candidates' lives easier. For example, candidates might be tempted to take on extra or peripheral responsibilities as good school citizens. Mentors would counsel principals to restrict such duties, to make candidates' experiences less taxing.

Peer Support

Besides this very direct support, successful alternative programs offer a more distanced yet crucial kind of support, in the form of seminars. Such seminars create a bridge—between theory and practice and also between the program's course work and its system of support. These sessions offer candidates the opportunity to share frustrations and engage in problem solving, not only with program faculty but with fellow candidates, whose insights come from being in the same boat. These kinds of discussions allow candidates to travel an arc: They take theory learned in course work, try it out with students, return to the group to analyze what succeeded or failed, get advice, and then go back and try again—each time growing in terms of teaching, reflection, and self-analysis.

Georgia offers an example of how such peer support operates. The RESA program makes available a series of professional, problem-based seminars. The seminars are facilitated by teachers with successful classroom experience, positive experience teaching adults, and expertise in particular specialty areas. Candidates are required to attend six seminars in the first year and four in the second year. If the support team determines that a candidate needs help with, say, behavior management, it recommends a classroom management seminar. The support comes in a form that is easy to digest, as well as relevant.