Tuesday, June 26, 2007

Teacher Quality Stats: Experience Matters

It has been a bit longer to post on this subject than I had intended, but personal and professional matters have gotten in the way of what is becoming my hobby on finding new ways to measure teachers' effectiveness. If you are looking for more posts on this subject, please see the Topic Teacher Quality Stats on the left.

In this post I would like to talk about something fairly common sensical--teacher experience. I hope to add some additional ideas into the mix about how we as consumers of education look at matters of teacher experience and how those of us looking to improve the quality of education look at experience as a factor in teacher effectiveness. In this post, I will talk about time in service, i.e., the number of years of experience a teacher has; time in position, that is the number of years a teacher as worked at a given school, and time in class, which I define as the number of years working in a grade level for elementary teachers and/or subject matter for secondary teachers.

Time In Service.
When looking at teacher experience, we generally look at one number--the number of years a teacher has been doing their job. After all, this is a relatively easy number to figure out--simply count the number of years since that teacher started teaching and then subtract out any gaps for say a sabbatical or maternity/paternity leave. Voila, there you have an experience statistic--number of years on the job. Given that everyone improves in their job over time as they become more confident and more experienced, measuring the length of service of a teacher is a reasonable, and respectable, measure.

This is not an irrelevant statistic, just an incomplete one. Studies have shown that most teacher rapidly gain effectiveness in teaching in the late second to the fifth year of experience. Thus, one would reasonably predict a teacher in their fifth year to be more effective than a teacher in their second year of teaching. One could also assume that a teacher with 20 years of experience would be more effective than a teacher with 10 years of experience. However, that has not proven to be the case, as teacher effectiveness tends to level off after the fifth year of experience.

But can we break down teacher experience a little deeper to find some methods, manner and or procedures that can lengthen the learning and effectiveness curve of teachers beyond five years. From a procedural standpoint, one way to increase teacher effectiveness over a longer period of time might be to alter than school operations to involve a more collegial atmosphere. Susan J. Rosenholtz studied the effects of teacher quality and the school's organization, arguing many times that an atmosphere that encourages interaction and learning from each other produces teachers more likely to continue to develop their effectiveness.

So one measure of teacher experience would be to also look at the relative collegiality of the school in which the teacher works. While it may be hard to fully quantify something like collegiality you could develop an arbitrary scale of 1 to 5, where 1 is the least collegial--say barely any talking in the teacher's lounge and 5 is the most collegial, using team teaching, fully developed mentorship programs, regular peer appraisals outside of performance appraisals, lesson plan review and critique, regular departmental and interdepartmental conferences on teaching techniques, etc. In short, how much support and input from fellow teachers of all experience levels do the teachers in a school get. If a collegial atmosphere is fostered by the school administrator, it is possible that the teachers themselves will produce more experience beyond the simple count of years in the classroom.

Time in service is, in reality, a limited measure. After five years, studies seem to point to little statistical significance in effectiveness between a five year teacher and a ten year teacher. If the goal is to extend the time in service for teachers, then we must look to other factors that can be used to judge teacher quality. The degree of collegiality and how it impacts a teachers effectiveness and quality after the first five years hints at something that many people who have studied teachers and teaching have guessed as an impact. External stimuli and challenges often create additional experiences which lead to greater quality in teachers.

Given that teaching is, by and large, a solitary experience for the teacher, finding ways to stimulate the teacher would seem to be a plus for school systems. Increasing the collegiality of a school certainly seems to help and should be implemented as best as possible. But collegiality may only go so far and other measures must be examined.

Time in Position.
If collegiality is a factor in teacher experience and quality, what about a negative factor as well, complacency. When a person has worked in the same place of for a number of years, in the same job, it is easy to get complacent; to take certain things for granted. While the influx of new children each year may lessen the impact or extend the period before complacency takes hold, it is a natural progression. However, unlike collegiality, complacency has something of a built-in indicator, the number of years in a particular school, or time in position.

If the number of years of teaching can be referred to as time in service, time in position would be simply the number of years a teacher has been at the school where she works. So far, in my research (which admittedly is not exhaustive), I have found little on this question of time in position on teacher quality. Hypothetically, the time in position would add to teacher effectiveness in the first few years in much the same way the first five years of time in service works, quality and effectiveness would increase significantly over the first five year in position as the teacher learns and adapts to the school, the administration, her peers, and the neighborhood and its students. As more and more understanding of the people and their neighborhood increases so too does that teacher's effectiveness at that school. But as the five year mark passes, the teacher's effectiveness may begin to slow in its growth.

One implication of this hypothesis would be that some slow teacher mobility may actually be a good thing both for the teacher and the school system. In general, society and parents have looked at the longer time a teacher has at the school, the better or more effective their teaching is. However, if the hypothesis of time in position being similar to time in service holds true, the after five or six years, a teacher may actually stop growing in terms of effectiveness at one school. If time in position approached five, six or seven years, it may behoove both the teacher and the school to transfer to another school for a period of five or six years. Since the population of the school's students turns over every five or six years, there can be some continuity for students and teachers, but with a transfer every five or six year, the teacher is "refreshed" by the challenges of a new environment and colleagues in much the same way as a new teacher is challenged during the first few years in service.

The primary difference, however between a new teacher and a transferred teacher is that a transferred teacher comes to the school with skills and knowledge of an experienced teacher. They may be able to make an immediate impact in the new environment, an impact that a new teacher may not be able to make.

Time in Class.
If time in service and time in position argue for regular, if long term, changes in teacher assignments, time in class would arguably work in the opposite direction. Time in class is the time in which a given teacher has been teaching at a particular grade level for elementary students or a particular subject area for secondary teachers. Again, this is a simply calculation of time.

Time in class is predicated upon a teacher subject area knowledge, more than his or her skill as a teacher. Thus, an elementary school teacher that works with second grade students would be expected to understand the physical, emotional and educational development of seven and eight year olds (those children most likely to be in her class). The more time a teacher spends with such students, the more experience she would gain as a result of the regular and repeated interactions, even thought the class roster might change from year to year. Similarly a secondary teacher who teaches say, American History, would be expected to develop a deeper and broader understanding of the subject, as well as how to teach that subject as she gains more experience in the class.

If time in service and time in position can be graphed in a curvilinear fashion, that is a steep increase in effectiveness in the early years followed by a steady decrease in the rate of growth after year five or six, time in grade would, ideally, have a relatively constant growth pattern, with perhaps some spikes based upon a master's degree or other professional development instruction.

However, in order to achieve a time in class chart of steady growth, teachers would have to have a professional development program based upon subject matter expertise rather than on the current fad in education or pedagogy. Thus, elementary teachers would have to have a professional development program based upon the psychology and physiology of their students at a given age, as well the latest changes in curricula and pedagogy. Secondary teachers would need professional development strongly related to their subject matter. Ideally, subject matter seminars presenting the latest information in their fields would form the basis of the a secondary professional development.

In addition to primary subject matter professional development, schools should look to expand teacher skills and knowledge into other subject matters. According to the NCES many teachers are not teaching in their field. Given the push by NCLB and the general public for highly qualified teachers, why not embrace diversity of assignment and allow teachers to explore other fields in which they may have an interest and aptitude. Would the teacher have a degree in the field? Perhaps, if schools and school systems would equate a second bachelor's degree on par with or even more valuable than a master's degree.

Imagine, a science teacher who teaches chemistry and/or physics must have a solid grounding in mathematics. What if that teacher were to obtain a second bachelor's degree in mathematics? Given that obtaining a second bachelor's degree would take roughly the same time as a master's degree without all that attendant "general education" requirements most colleges have, the economics are not all that different for schools or the teachers. Now that teacher is qualified to teach both science and math. The resulting diversity in subject matter and the ability of the teacher to meet multiple needs for the school keeps that teacher engaged, and effective. Just as adding to a subject matter knowledge is important for secondary teachers, having different knowledge allows for a deeper understanding of both knowledge fields.

Time in class may be different for most teachers than their time in service. Indeed, elementary school teachers may start teaching first graders, but a few years later move to third or fifth graders. Each time, a new time in class clock begins. Similarly, secondary education teachers may have five classes of history and one class of government, each with a different time in class "clock." Time in class is a bit more precise than simply time in service and allows for other comparison and measurement. Indeed as the NCES study points out, many of the measurements of teacher assignments and "out-of-field" teaching are either over or under inclusive. Some provide a bit more precision, such as percentage of course and percentage of students measurements. (see discussion on pages 11-12) When combining the later two, in particular the percentage of students measurements, with a time in class clock, you can get an accurate measure of how much time a student spends with a "out-of-field" teacher and perhaps measure the impact upon their learning.

Conclusion.
Time in a classroom and experience as a teacher are undoubtedly important measurements of teacher quality. As we have learned in multiple manners, the quality of education a child receives is greatly dependent upon the quality of that child's teachers. While it would be great to have nothing but experienced teachers with impeccable academic credentials, such a Stepford school system is impossible. But if schools could look beyond simple measures of seniority and experience found in a time in service measure, much could be learned about teacher effectiveness.

In large part, much of this post is somewhat hypothetical and based upon conjecture. Right now the only measure of experience regularly used is time in service and as we have learned, time in service as a predictor of teacher quality tends to diminish after year five in the classroom. While school conditions such as a collegial atmosphere or other outside inputs may enhance more senior teacher effectiveness, it may not be enough to truly make an impact.

But delving deeper, if we studied effectiveness of teachers based on their time in position as well, we may see a similar pattern. A steep increase in the first five years followed by a reduction in rate of growth after a teacher has spent time in a particular school. But since a teacher with five or six years of time in service may but transfers to a new school and thus resets her time in position clock has the advantage of time in service.

Time in class, however, is different because as a teacher spends more time in a given grade level or subject matter their actual effective would, one would hope, increase or at the very least grow on a geometric progression. While subject matter knowledge may be based in large part upon the prior education of the teacher, the key to successfully using time in class measurement will be found in the effective use of professional development. A poor professional development program for teachers, particularly secondary school teachers, will likely negate any effectiveness benefit gained from a teacher with significant time in class.

Regardless, more research on the impact of time in position and time in class needs to be done. Simply measuring time in service loses its relevance after too short a period and runs counter to the effort to identify and retain truly effective teachers. If we can find out where and how teachers lose the the effectiveness increases of their early careers, school systems can fashion policies, such as transfer policies and professional development programs that can keep the effectiveness curve as steep as possible much to the benefit of our students.

No comments: