Response to Donna Kerró Understanding the Sources of Teaching Competence: Choices, Skills, and the Limits of Training
by Richard J. Murnane - 1983
This response to an article by Donna Kerr (Teachers College Record, Spring 1983) stresses the importance of teacher salaries as a means of attracting talented college graduates to teaching. Teacher effectiveness is greatly influenced by classroom experience. Improving preservice teacher education may have a limited effect on teacher improvement. (Source: ERIC)
Conversations with Sharon L. Kagan and Edward Pauly were helpful in preparing these comments.
Attempting to make sense of a complex, politically sensitive policy issue is a challenging and risky endeavor. The author inevitably becomes exposed to a variety of criticisms, many of which are motivated by the critics self-interest. Since I believe that it is important to stimulate intelligent debate about the question of teacher competence, I would like to be supportive of Professor Kerrs efforts. Indeed, I believe that her article does contribute to the policy discussion by bringing together a variety of evidence, and by enumerating many of the dimensions of the problem and the actors involved in it.
However, I am disturbed by many of Kerrs policy recommendations because they seem to me to be inconsistent with the evidence-both the evidence she presents concerning the reasons for the apparent decline in teaching competence and also existing evidence concerning the determinants of teaching effectiveness, the factors affecting the career decisions of teachers, and the power of university-based scientific approaches for improving the performance of the nations schools.
Let me try to explain my concerns by focusing on the sources of teaching competence, a term used in the Kerr paper with varied connotations about the sources of competence.
One can usefully think of three sources of teaching competence: intellectual ability, on-the-job experience, and formal preservice teacher education. The three sources are clearly not independent, nor is it easy to determine their relative importance. However, these sources do suggest different strategies for improving the quality of teaching in American schools. Consequently, it is important to consider the nature of the evidence regarding the importance of each of these sources of teaching effectiveness.
The most compelling evidence concerns the importance of intellectual skills, as measured by a teachers score on a test of verbal ability or by the quality of the undergraduate college the teacher attended. Several studies have reported significant relationships between these variables and teaching effectiveness, as measured by student test-score gains.1 (Kerr mentions some of this evidence.)
There is also quite strong evidence that experience on the job is an important source of teaching effectiveness. Several studies have found that teachers with two or more years of experience are more effective than beginning teachers.2
The evidence is weakest in regard to the role of formal preservice teacher training. To my knowledge, no study has found undergraduate major or the choice of a particular teacher training institution or participation in a particular preservice training program to be significantly related to teaching effectiveness. In fact, one of the most consistent findings of research on the correlates of teaching effectiveness is that teachers with masters degrees are not more effective on average than teachers without masters degrees.
Clearly one should not interpret these results as indicating that intellectual ability should be the sole criterion used in recruiting teachers or that formal teacher training cannot make a difference. In fact, the lack of evidence supporting formal preservice training as a source of competence may be to some extent a result of limitations in the available data. For example, all data bases suitable for examining the correlates of teaching effectiveness as measured by student achievement gains pertain to a single school district. Since there is less variation in training among teachers within a district than among teachers in the country at large, these data bases do not permit the most powerful possible tests of the efficacy of alternative teacher training programs.
However, the evidence that does exist suggests two implications for thinking about how to improve teaching competence in American schools. First, it is important to look very carefully at what determines the career choices of college graduates, and how more intellectually able teachers can be recruited by school districts. Second, one must be skeptical about a strategy for improving teaching that relies on changes in formal preservice training. It is these aspects of the Kerr recommendationsthe lack of attention to recruiting more able applicants, and the emphasis on changes in preservice teacher trainingthat I find troubling, and not justified by what we know about the sources of teaching competence. Let me explain in more detail why I believe that a different emphasis is more appropriate.
Do we know how to influence the career choices of intellectually able college graduates who might be attracted to the teaching profession? We know that higher salaries are an important part of the answer. There is compelling evidence that districts with higher salary scales attract more intellectually able teachers. 3 There is also evidence that the career-mobility decisions of teachersboth decisions to change school districts and decisions to leave the professionare sensitive to salary differentials.4 Given the compelling nature of the evidence on the importance of salaries in determining career decisions, and given Kerrs mention of some of this evidence in her enumeration of the sources of the teacher competency problem, why are improved financial incentives for classroom teachers omitted from her list of recommendations? Two reasons seem plausible.
One may be the realization that there is little political support for salary increases for public employees, including teachers. This realization naturally leads the analyst to think about what can be done without significant increases in expenditures. However, given the importance of salaries in determining the job choices of college graduates, better salaries may be a necessary condition for upgrading the quality of the teaching staffs of U.S. schools.
A second reason for the lack of attention to salary increases may be the belief that increased compensation will not lead to improvements in teacher quality in many districts because the personnel offices will not hire the most able teachers. While Kerrs paper does not contain an explicit indictment of personnel office practices, such an indictment seems to follow from evidence she citesthat the graduates who were hired had lower scores on the National Teachers Examination (NTE) than did graduates who applied for teaching jobs but were not hired. It does not necessarily follow from this evidence that personnel offices do not hire the best applicants for jobs. Given that hiring is done by 15,000 independent school districts in the United States, each of which is choosing among its own pool of applicants, the aggregate evidence is consistent with a situation in which every local district offers jobs to those with the highest NTE scores in its pool of applicants. Thus, before we indict personnel offices, we need to know which teachers apply for jobs in particular school districts, which applicants are offered jobs, and which applicants accept the jobs. Clearly, this evidence is difficult to collect, but without it, it is not possible to determine reliably the extent to which the choices of personnel offices contribute to the teacher competency problem.
This caveat aside, it probably is true that some personnel offices, as a result of a lack of resources, political pressures, or poor judgment, do not hire the best applicants. In these districts higher salaries might not lead to improvement in the quality of teaching staffs. However, if the choices of personnel offices are the problem, they should be the focus of recommendations for change. Said differently, higher salaries are probably a necessary condition for improvement in teaching quality because they are a crucial determinant of the choices of potential teachers. Higher salaries may not be a sufficient condition because the composition of teaching staffs depends not only on teacher choices, but also on the choices of 15,000personnel offices, the actions of which are not well understood.
In describing the dimensions of the teacher competence problem, Kerr places considerable emphasis on the high attrition rate of teachers from the profession, and she presents limited evidence that the attrition rate is highest for the most able teachers. She argues that a large part of the competence problem is the structure of the teaching professiona structure that leaves teachers with twenty years of experience with the same tasks as those of new teachers. Kerr is probably right that for many teachers, but not all, what was challenging and rewarding for a few years becomes boring eventually. My disagreement with Kerr concerns what to do about this phenomenon.
Kerr suggests a differentiated staffing structure with the position of head teacher being created for teachers who earn doctorates in teaching. I have two reservations about this proposal. First, it is not clear that the job structure in the teaching profession influences the career decisions of teachers. Asking teachers why they left the profession provides questionable data; reliable evidence comes only from examining whether teachers who face different opportunities, both within the teaching profession and in other careers, make different choices about how long to stay in teaching. To my knowledge, there is no evidence that could be interpreted as indicating that the option of earning a doctorate and becoming a head teacher would improve the retention of teachers, or, if it did, would attract the most talented teachers. My second reservation concerns the requirement that head teachers should possess an earned doctorate. Would more formal training improve the skills of potential head teachers? I know of no evidence to support this position.
I am of the opinion that the way to attract more intellectually able college graduates into teaching is not to create a hierarchical structure within the teaching profession, but rather to make it easier for bright, well-educated graduates to become teachers for a few years and then to move on to alternative occupations, feeling that they had achieved something and had done important service, and without feeling the need to denigrate themselves or the teaching profession because they eventually wanted to do something else. Implicit in this suggestion is the belief that for many talented college graduates, spending four or five years in the classroomlong enough to learn to be a good teacher and to be of real service, but not a lifetimewould be an attractive part of a varied, rich professional life. In what other occupation is a new entrant given so much responsibility and so much independence of action as in teaching?
There are some things that could be done to attract more talented graduates into the classroom for a few years. One interesting possibility is Kerrs recommendation that alternative licensing requirements be developed for graduates who have not gone through formal teacher training. To my mind, however, even more important than specific policies are changes in rhetoric and preconceptions. I wonder how many talented college graduates are discouraged from teaching by rhetoric that suggests that anything less than a lifetime in the classroom implies both a failure in the structure of the teaching profession and a personal failure for the college graduate who leaves the classroom after a few years of successful, rewarding teaching. Kerrs uses of the words amateur and professional lend support for the conventionaland, I believe, unfoundedattitude that a person cannot be a successful teacher if he or she plans to teach for only a few years.
Let me turn to my second theme, the limits of teacher training. Kerr argues convincingly that formal preservice teacher training has not been successful in creating a corps of talented, effective teachers. While my intuition is that there is much more variation in the attributes and quality of teacher training in the United States than Kerr suggests, I believe that her basic description is correct. However, to point out that teacher training as it currently exists is not effective does not necessarily imply that changes in teacher training would make it an important source of teacher competence.
Kerrs recommendations for improving teacher trainingreliance on major universities, emphasis on developing a strong scientific baseare reminiscent of the recommendations made to improve educational research and development in the 1960s. When the educational research centers were established by the federal government in 1965, they were placed in major universities. Real efforts were made to develop a strong scientific base from which powerful educational technologies could be developed.5 While there is considerable disagreement about the value of the contributions of educational research and development (R and D), it is now recognized that the emphasis on the development of new technologies to dramatically improve student learning is inappropriate. If R and D makes a difference in education, it is by expanding the menu of alternative techniques and curricula from which teachers can draw and by helping teachers to adapt teaching strategies and curricula to meet the changing needs of their students.
Similarly, if training is to enhance teaching competence, it is likely to be in subtle ways that improve teachers abilities to diagnose, to search, to adaptin other words, to learn for themselves. Such training may be important, particularly if it is more closely linked to on-the-job experiences than most formal teacher training currently is. However, this kind of training, closely linked to practice, is not what Kerr appears to have in mind. Her training plan (for head teachers) emphasizes a program of clinical studies that would utilize the extant and growing stock of theoretical perspectives and empirical studies that illuminate human learning. I believe that such an approach will not produce dramatic improvements in teaching competence for the same reason that R and D has not produced dramatic successes: Teaching is simply not a process that consists of application of codified techniques and principles that can be developed in the laboratory or learned in the university classroom.
This does not mean that teaching is not a skilled profession. It clearly is. But the critical skills seem to be acquiredor not acquiredthrough experience rather than through formal preservice training. In other words, my reading of the evidence is that there is only limited potential for formal preservice training to enhance the quality of teaching in American schools. Consequently, I view skeptically recommendations for improving teaching that emphasize restructuring of the formal preservice education of teachers.
In conclusion, I agree with Professor Kerr that the low level of competence of the teachers in some American schools is a matter of real concern. I also agree that America will have better teachers only if it learns to value and reward effective teaching more highly. My disagreement with Professor Kerr concerns how to achieve this enhanced status for the teaching profession. I do not believe that restructuring teacher education can be the focal point for effective change. This strategy does not pay attention to the results of the many past efforts to restructure teacher training, or to the reasons many school districts in the United States have staffs of very talented, effective teachers at the same time that other districts are not able to attract competent teachers.