Yesterday I had parent teacher conferences, and afterwards I met with some of the other teachers at my school for a little after party. While there I asked the following question:
If you could choose between having a 10% increase in your salary, or capping your classes at 20 students, which would you choose?
The consensus seemed to be that the cap on class size would be preferable because it would improve quality of life. Then someone mentioned that in one of their previous jobs they got paid $10/week for every student in the class over 25. I asked whether teachers in that system had any control over how many students they had in their classes and the answer was that they didn't have much say in the matter.
The reason why this interests me is that two of the more common proposals for improving the quality of teaching in our schools are to improve teacher compensation and reduce class size. The rationale behind improving teacher compensation is that more highly qualified people might be motivated to teach if the job paid more. I thought that the main reason for reducing class size was that it would lead to improved classroom management and more individualized instruction. I didn't think too much about the idea that smaller classes improve the the quality of life for teachers.
The point was also made that some teaching positions are more labor intensive than others. In particular, grading papers takes a lot of time for English teachers. Several people mentioned that some districts give English teachers a reduced class load but require them to assign more essays. Most people didn't seem to object too much to that, but there was a little discomfort at the idea of paying some teachers more than others within a school.
I wonder what would happen if teachers were paid according to the number of students in their classes, and had to compete for students. It may create an incentive to make the class easy, but this could be mitigated by removing the responsibility for grading from the teachers. Make grades based on exams that are standardized within the school...
As I start to think of all the changes that would have to be made it occurs to me that the reforms would be very extreme if taken all at once. But the idea of giving teachers some say in the trade off between class size and compensation is very appealing to me.
Saturday, February 26, 2011
The Death and Life of the great American School System
I recently read Diane Ravitch's book on education reform. In it she tells the story of how she became disillusioned with testing, accountability, and school choice. Her main points are that charter schools aren't any better than public schools on average, and that testing and accountability lead to a narrow curriculum and teaching to the test. Her preferred approach to improving the education system is curriculum reform.
It is hard for me to give the book a simple thumbs up or thumbs down. It engaged me with a number of interesting stories about different attempts at reform and the mixed results. It encouraged me to learn more about curriculum reform and to question my beliefs about choice and accountability. But I am left disappointed. She states at the outset that she isn't going to offer a "silver bullet," and she certainly stays true to her word. In the end, the book is little more than a compilation of the same tired arguments that unions use to oppose change.
Perhaps I simply expected too much. I wanted a book that would take the field of education reform and tie it together like The Origin of Species did for biology. I wanted theory and evidence. I got stories and platitudes.
It is hard for me to give the book a simple thumbs up or thumbs down. It engaged me with a number of interesting stories about different attempts at reform and the mixed results. It encouraged me to learn more about curriculum reform and to question my beliefs about choice and accountability. But I am left disappointed. She states at the outset that she isn't going to offer a "silver bullet," and she certainly stays true to her word. In the end, the book is little more than a compilation of the same tired arguments that unions use to oppose change.
Perhaps I simply expected too much. I wanted a book that would take the field of education reform and tie it together like The Origin of Species did for biology. I wanted theory and evidence. I got stories and platitudes.
Monday, February 21, 2011
Against State Sanction of the Sacredness of Marriage
Recently I read an argument for restricting marriage to heterosexual couples that I interpreted to depend on the following logic:
- If the state doesn't sanction a restricted view of marriage, it will undermine many people's view of the sacredness of the institution.
- If people do not think that marriage is sacred, they will be less likely to have stable families.
- If families are not stable society will suffer from increased rates of state dependence.
This is my view of how people justify the claim that extending marriage rights to a segment of society that is generally wealthy and functional (the subset of the gay population who are in committed relationships) will result in more dysfunction among a totally separate population of lower socio-economic status.
Although I agree with the last point, the first two are dubious. Even if they are all true, in order to justify state recognition of exclusively heterosexual marriage for this reason we must balance the harm against the presumption in our country that state interference in religious liberty does harm to individual liberty. Let me point out that there are arguments in support of restricting marriage that don't depend on the "sacredness" of the institution, but those that do run into this 1st Amendment problem.
Before I point out why I think the first two points are wrong, let me try to define what is meant by "sacred." Although the argument is not mine, I think that the term "sacred" in this context can be understood to mean something that is instituted by God and should be done according to his will. In the words of Focus on the Family, "God, the mastermind of marriage, also has a plan for how marriage works best."
So we can read the first point as saying that if the government expands its definition of marriage, it will undermine the belief that marriage serves an important religious function. My reason for doubting this is that the United States serves as an important example of how religious beliefs can thrive in a society where the state does not attempt to sanction religious views. Nations in Europe generally grant official recognition and subsidies to state religions. Yet compare religiosity in the US to Europe (from a Pew Research center report):
Outside of the US, religious belief is negatively correlated with national wealth. Yet in the wealthiest and most powerful country, people still believe. Why? One possible explanation is that in the absence of state interference religious freedom has led to increased competition, which in turn leads to religion that is more relevant to our lives. In any case, state support of religious institutions or beliefs is not critical for their survival. I think that whether or not people believe that their marriage is sacred depends more on the effectiveness of their church than on legal distinctions. One could even argue that state support of religion tends to weaken belief.
It is true that while we claim religious freedom in the US, many of our laws reflect the Christian values of our lawmakers. So perhaps the vitality of religious belief is best served by an environment where government claims to tolerate diversity, but still maintains a paternalistic grip on behavior through legislation. But that discussion will take us too far afield. For now let me just state that I have no real reason to believe that people can't maintain a sense of sacredness without government sanction of their beliefs.
With regard to the second point, I do not doubt that a sense of sacredness provides many people with a greater sense of solidarity within their family. However, I just don't think that this always results in significantly different behavior on a societal scale. Evidence of this can be found in the fact that the "Bible Belt" states have the highest divorce rates in the US. Catholic families are less likely to get divorced because that religion does not recognize the termination of marriages. In that case it is social pressure not sacredness that keeps marriages together, and government recognition of gay marriage is unlikely to impact Catholic policy on divorce.
Family stability is driven by factors like income and education more than religious belief, and the impact of religion depends mostly on whether divorce is tolerated. Therefore, even if state sanction of gay marriage did undermine the sense of sacredness associated with marriage, it would likely only have a minor impact on family stability.
I think that the argument I outlined ultimately stems from the very conservative belief that any change to traditional ways of doing things will likely result in unanticipated negative consequences. This Burkean view rests on the idea that our traditions represent the accumulated knowledge of generations, and reform is often based on untested theory. While I think the perspective has some merit, it does not provide us any means of deciding which reforms are truly dangerous and which are will lead to real progress. In this case I am not convinced by the sky-is-falling attitude regarding the decay of the family, and I don't think that leaving sacredness to religion will make the situation any worse.
Wednesday, February 16, 2011
Utah is not DC
A few days ago I posted on how the intended effect of most education reforms is to either improve teaching quality or increase the amount of time spent on the desired curriculum. One thing I didn't mention was cost.
Recently an interesting report has come out from the Center for American Progress stating that different school districts in the country have widely varying return on investment. That is, the ones that spend the most aren't necessarily the highest performing. Here is a nice interactive map that shows the disparities. And here is a chart from the New America foundation:
Notice that Utah is on the far left in terms of spending, but that our graduation rate is a respectable 75%. Compare to DC, which spends almost 3 times as much and has a much lower graduation rate. Now, graduation rates are questionable since different states might have different criteria. This report tries to standardize things. Utah actually reports its own graduation rate as closer to 90% because it doesn't count students that the system can't track down. Also, states have different graduation requirements.
Another factor is that aggregate data like state graduation rates covers up differences among different socioeconomic groups. For example, the chart above makes it look like Wisconsin has a pretty dang good education system. And it probably does. But Wisconsin is also very white, and their minority groups dont' have very high graduation rates. If we just counted Native Americans, Hispanics and African Americans, Wisconsin's rate would be more like DC's.
In any case, I still think it is clear that some states are spending their money more efficiently than others. It is also clear that spending money by itself does little to improve education unless you spend it cost effectively. The scant relationship between spending and educational outcomes has led many to argue that Utah does not need to spend more money on education. But this is not necessarily true. It is probably true that DC does not need to spend more money on education. A lot of education money there is probably wasted. There is probably corruption and inefficiency.
Utah is not DC. I don't our education spending thus far has really provided many opportunities for corruption and inefficiency. Sure, there is some wasted money as there is in any large organization. But it can't be too much because we don't spend that much in the first place. DC probably wastes more money per student than Utah spends in the first place.
One way to test how well Utah students are doing that relies on more standardized data than graduation rates is to compare average ACT scores. Here is a chart comparing different states. Basically, Utah does better than most southern states and better than all states that currently test all high school students. But we do significantly worse than several states with comparable testing rates such as Nebraska, Minnesota and Wisconsin. Those states spend about $10,000 per student compared to our $6,000.
This suggests to me that while some states can improve education outcomes by reducing waste, it is probably not the case in Utah. We may have no choice but to spend more, and not just a little bit more. Sure, the cost of living might be low in Utah compared to other states, but is it that much lower than Nebraska? According to this data gleaned from the Bureau of Labor and Statistics, Nebraska actually has a lower cost of living than Utah.
It has often been pointed out that while Utah spends very little per student, we spend a lot per capita. The implication is that because we have a lot of children, it is okay that we value each individual child less. I think the truth is something more like many Utahn's are willing to sacrifice worldly success (both their own and their children's) in favor of large families. Those who don't have large families, on the other hand, are not willing to sacrifice too much to of their hard earned money to improve the prospects of everyone else's children.
In any case, unlike DC, Utah's problem is a funding problem not a waste problem. The implication is that indirect reforms which primarily address inefficiencies instead of directly improving teaching quality or teaching time are less likely to work in Utah than in more inefficient states. That includes reforms such as vouchers*, charter schools, and accountability measures. While these may be more popular here than in many other places, they aren't likely to get us far. We have to spend more money. We have to increase the length of the school year and increase wages enough to attract more qualified teachers. Until we do this we will have mediocre education outcomes (not bad ones, because of the socio-economic status of most of our students).
*I should comment that vouchers might actually do more than encourage competition among schools (which theoretically could increase efficiency). They might also encourage more middle class parents to send their kids to private schools that spend (and cost) more per student than the local public school. If so, vouchers might be the only politically viable way to increase education spending.
Recently an interesting report has come out from the Center for American Progress stating that different school districts in the country have widely varying return on investment. That is, the ones that spend the most aren't necessarily the highest performing. Here is a nice interactive map that shows the disparities. And here is a chart from the New America foundation:
Notice that Utah is on the far left in terms of spending, but that our graduation rate is a respectable 75%. Compare to DC, which spends almost 3 times as much and has a much lower graduation rate. Now, graduation rates are questionable since different states might have different criteria. This report tries to standardize things. Utah actually reports its own graduation rate as closer to 90% because it doesn't count students that the system can't track down. Also, states have different graduation requirements.
Another factor is that aggregate data like state graduation rates covers up differences among different socioeconomic groups. For example, the chart above makes it look like Wisconsin has a pretty dang good education system. And it probably does. But Wisconsin is also very white, and their minority groups dont' have very high graduation rates. If we just counted Native Americans, Hispanics and African Americans, Wisconsin's rate would be more like DC's.
In any case, I still think it is clear that some states are spending their money more efficiently than others. It is also clear that spending money by itself does little to improve education unless you spend it cost effectively. The scant relationship between spending and educational outcomes has led many to argue that Utah does not need to spend more money on education. But this is not necessarily true. It is probably true that DC does not need to spend more money on education. A lot of education money there is probably wasted. There is probably corruption and inefficiency.
Utah is not DC. I don't our education spending thus far has really provided many opportunities for corruption and inefficiency. Sure, there is some wasted money as there is in any large organization. But it can't be too much because we don't spend that much in the first place. DC probably wastes more money per student than Utah spends in the first place.
One way to test how well Utah students are doing that relies on more standardized data than graduation rates is to compare average ACT scores. Here is a chart comparing different states. Basically, Utah does better than most southern states and better than all states that currently test all high school students. But we do significantly worse than several states with comparable testing rates such as Nebraska, Minnesota and Wisconsin. Those states spend about $10,000 per student compared to our $6,000.
This suggests to me that while some states can improve education outcomes by reducing waste, it is probably not the case in Utah. We may have no choice but to spend more, and not just a little bit more. Sure, the cost of living might be low in Utah compared to other states, but is it that much lower than Nebraska? According to this data gleaned from the Bureau of Labor and Statistics, Nebraska actually has a lower cost of living than Utah.
It has often been pointed out that while Utah spends very little per student, we spend a lot per capita. The implication is that because we have a lot of children, it is okay that we value each individual child less. I think the truth is something more like many Utahn's are willing to sacrifice worldly success (both their own and their children's) in favor of large families. Those who don't have large families, on the other hand, are not willing to sacrifice too much to of their hard earned money to improve the prospects of everyone else's children.
In any case, unlike DC, Utah's problem is a funding problem not a waste problem. The implication is that indirect reforms which primarily address inefficiencies instead of directly improving teaching quality or teaching time are less likely to work in Utah than in more inefficient states. That includes reforms such as vouchers*, charter schools, and accountability measures. While these may be more popular here than in many other places, they aren't likely to get us far. We have to spend more money. We have to increase the length of the school year and increase wages enough to attract more qualified teachers. Until we do this we will have mediocre education outcomes (not bad ones, because of the socio-economic status of most of our students).
*I should comment that vouchers might actually do more than encourage competition among schools (which theoretically could increase efficiency). They might also encourage more middle class parents to send their kids to private schools that spend (and cost) more per student than the local public school. If so, vouchers might be the only politically viable way to increase education spending.
Monday, February 14, 2011
Testing and Intrinsic Motivation
At the end of my last post I mentioned that student motivation did not fit neatly into my model. That doesn't mean I am not thinking about it. In fact, in my own classes it is one of the main things I think about. Basically, I wonder whether my teaching method leaves students with a higher or lower level of intrinsic motivation to study math.
There are a lot of factors that affect whether students are motivated in school. In response to my questioning, my students have stated that the main reason they try to do well in my class is fear of punishment from parents or administrators. The only reason the do homework or study for a test is to get a decent grade, and fear of immediate consequences is the main reason they want to get good grades.
This is not a great situation for a math class. Math is a very abstract subject that requires a considerable amount of mental effort to master. If you are just scraping by to avoid punishment, your understanding of math will almost certainly be very superficial. One result of this is that the students have a very hard time pushing themselves to solve problems that aren't exactly like something that has been demonstrated for them numerous times in class. They tend to get easily frustrated and assume that they can't do problems when they seem complex.
So I have been thinking a lot about how to help them develop a little more intrinsic motivation, and I worry that the standard approach (which I use) of focusing on homework and testing may undermine my efforts.
This question also applies to one of my favorite reform proposals: higher graduation standards. I have maintained that students should be required to pass a statewide exam after each class that is required to graduate. As it is, students can graduate with a certain number of credits without ever having mastered any specific material. Schools are then judged based on student performance on exams that mean nothing to the students. I think we should do away with standardized tests that are designed to measure schools, and replace them with meaningful graduation requirements. The performance of schools can be determined by graduation rates. This method may necessitate creating more than one type of diploma.
One reason for advocating this method is that it gives students more incentive to learn material rather than just get "credit" for a class. Credit is meaningless. It does not necessarily communicate to students the expectation that they learn anything. The current system leads students to judge classes based on how easy it is to pass rather than on how much they learn.
However, while focusing on testing may create an external incentive for students to learn material, it will not necessarily improve their intrinsic motivation. It might even be more harmful than what we do now (although it is hard to imagine it being worse than the status quo). In any case, the question is how to best combine extrinsic and intrinsic motivating factors. I tend to think that requiring students to pass a criterion referenced test for each required course in order to graduate will at least shift the focus from passing to content mastery. But it won't take us the whole way. It won't make kids appreciate the power of mathematical reasoning.
There are a lot of factors that affect whether students are motivated in school. In response to my questioning, my students have stated that the main reason they try to do well in my class is fear of punishment from parents or administrators. The only reason the do homework or study for a test is to get a decent grade, and fear of immediate consequences is the main reason they want to get good grades.
This is not a great situation for a math class. Math is a very abstract subject that requires a considerable amount of mental effort to master. If you are just scraping by to avoid punishment, your understanding of math will almost certainly be very superficial. One result of this is that the students have a very hard time pushing themselves to solve problems that aren't exactly like something that has been demonstrated for them numerous times in class. They tend to get easily frustrated and assume that they can't do problems when they seem complex.
So I have been thinking a lot about how to help them develop a little more intrinsic motivation, and I worry that the standard approach (which I use) of focusing on homework and testing may undermine my efforts.
This question also applies to one of my favorite reform proposals: higher graduation standards. I have maintained that students should be required to pass a statewide exam after each class that is required to graduate. As it is, students can graduate with a certain number of credits without ever having mastered any specific material. Schools are then judged based on student performance on exams that mean nothing to the students. I think we should do away with standardized tests that are designed to measure schools, and replace them with meaningful graduation requirements. The performance of schools can be determined by graduation rates. This method may necessitate creating more than one type of diploma.
One reason for advocating this method is that it gives students more incentive to learn material rather than just get "credit" for a class. Credit is meaningless. It does not necessarily communicate to students the expectation that they learn anything. The current system leads students to judge classes based on how easy it is to pass rather than on how much they learn.
However, while focusing on testing may create an external incentive for students to learn material, it will not necessarily improve their intrinsic motivation. It might even be more harmful than what we do now (although it is hard to imagine it being worse than the status quo). In any case, the question is how to best combine extrinsic and intrinsic motivating factors. I tend to think that requiring students to pass a criterion referenced test for each required course in order to graduate will at least shift the focus from passing to content mastery. But it won't take us the whole way. It won't make kids appreciate the power of mathematical reasoning.
Thursday, February 10, 2011
A Framework for Understanding Reform
As I think about education reform, I have decided it might be useful to develop a simple framework for understanding how different education reform proposals relate. To this end, imagine that there is some platonic Curriculum that encompasses everything that a student ought to learn before they enter college or the workforce (and assume for now that it is the same for all students). Then we can construct a simplistic model to explain how much of this Curriculum a student actually learns:
L = T x E x A
where L is the amount learned, T is the time spent on the Curriculum, E is the effectiveness of the teaching, and A is the innate ability of the student. The effectiveness of the school system can be measured by taking some aggregate measure of the L's for all of the students in the system.
There are a lot of school reform ideas out there, but they can be roughly categorized into two groups: those that seek to directly increase T or E (assuming A is fixed), and those that seek to improve them indirectly by making systems more efficient or removing obstacles to improvement.
Reforms that Improve T
There are several ways to increase T. The most obvious is to increase the number of hours spent at school either by increasing the number of days in the school year or the number of hours in the school day. I think this is perhaps the most straightforward of all school reform proposals, and I think it is one of the most likely to work. However, it is not cheap.
Another method of increasing T is to implement a curriculum that more closely resembles the ideal platonic Curriculum. This is the approach of the national standards movement. Many people think that the ideal curriculum should focus more on basics like math and reading. If this is true, we can increase T without spending much more money by cutting out programs that aren't necessary.
A third possible way to increase T is to improve attendance. Finally, T can be increased if students spend more time learning the material at home. If you figure out how to do this, let me know.
Reforms that Improve E
Three of the most important proposals that are aimed at improving E are reducing class size, improving teacher quality, and introducing better education technology.
My experience in the classroom hasn't convinced me that class size is the most critical variable in determining E. Research tends to suggest that the identity of the teacher is more important than the student-teacher ratio. So how do you improve the teacher? Well, you can hire better teachers, fire bad teachers, or improve the ones you have. I think firing bad teachers is a good place to start because it is probably easier to identify a bad teacher that you have than to identify a good teacher that hasn't yet started teaching. Given the sheer number of teachers, however, I think improving the existing cadre is necessary to make any improvements after picking the low hanging fruit.
I am not exactly sure about the prospects of improving education with technology. On some level it seems that if the advances of the information age were fully utilized, education would be entirely transformed. Yet my experience with technology in the classroom suggests that most of the existing options are mostly cosmetic. I say this despite having great respect for the smart board.
Indirect Improvements
Some of the more popular indirect reforms include school choice (e.g., vouchers and charters), merit pay, and and accountability.
The theory behind school choice is that school administrators are not choosing the most effective ways to use their resources. This means that either they are misusing their time or misspending their money. Perhaps the main problem is that they aren't hiring the right teachers, firing the wrong ones, and training the ones they have. Maybe they would do better if they were subject to some competitive pressures. Vouchers are one way of providing this choice within a publicly funded system, publicly funded charters are another.
Merit pay is an indirect way of trying to improve teacher quality. I agree that paying teachers based almost entirely according to seniority isn't the best way to inspire excellence. But most teachers would probably improve their teaching if they knew how. Perhaps a bigger issue is not the level of teachers pay relative to other teachers, but their pay relative to other professions. In some places, teachers are paid a decent salary (although not in Utah) but it is still not comparable to the pay of other professional options such as medicine, law, or even military.
Accountability is major trend in school reform (anyone ever heard of No Child Left Behind? It's this new government program...). Like school choice, it is basically an indirect way of improving teaching quality (and maybe even T). But unlike school choice, accountability schemes rely on the education bureaucracy rather than consumer choice to put pressure on administrators to find the right teachers.
So there you have it, my quick and dirty classification of school reform proposals. One thing that doesn't fit easily into this classification scheme is student motivation. Perhaps the part of student motivation that can be impacted by educators should be considered part of E.
L = T x E x A
where L is the amount learned, T is the time spent on the Curriculum, E is the effectiveness of the teaching, and A is the innate ability of the student. The effectiveness of the school system can be measured by taking some aggregate measure of the L's for all of the students in the system.
There are a lot of school reform ideas out there, but they can be roughly categorized into two groups: those that seek to directly increase T or E (assuming A is fixed), and those that seek to improve them indirectly by making systems more efficient or removing obstacles to improvement.
Reforms that Improve T
There are several ways to increase T. The most obvious is to increase the number of hours spent at school either by increasing the number of days in the school year or the number of hours in the school day. I think this is perhaps the most straightforward of all school reform proposals, and I think it is one of the most likely to work. However, it is not cheap.
Another method of increasing T is to implement a curriculum that more closely resembles the ideal platonic Curriculum. This is the approach of the national standards movement. Many people think that the ideal curriculum should focus more on basics like math and reading. If this is true, we can increase T without spending much more money by cutting out programs that aren't necessary.
A third possible way to increase T is to improve attendance. Finally, T can be increased if students spend more time learning the material at home. If you figure out how to do this, let me know.
Reforms that Improve E
Three of the most important proposals that are aimed at improving E are reducing class size, improving teacher quality, and introducing better education technology.
My experience in the classroom hasn't convinced me that class size is the most critical variable in determining E. Research tends to suggest that the identity of the teacher is more important than the student-teacher ratio. So how do you improve the teacher? Well, you can hire better teachers, fire bad teachers, or improve the ones you have. I think firing bad teachers is a good place to start because it is probably easier to identify a bad teacher that you have than to identify a good teacher that hasn't yet started teaching. Given the sheer number of teachers, however, I think improving the existing cadre is necessary to make any improvements after picking the low hanging fruit.
I am not exactly sure about the prospects of improving education with technology. On some level it seems that if the advances of the information age were fully utilized, education would be entirely transformed. Yet my experience with technology in the classroom suggests that most of the existing options are mostly cosmetic. I say this despite having great respect for the smart board.
Indirect Improvements
Some of the more popular indirect reforms include school choice (e.g., vouchers and charters), merit pay, and and accountability.
The theory behind school choice is that school administrators are not choosing the most effective ways to use their resources. This means that either they are misusing their time or misspending their money. Perhaps the main problem is that they aren't hiring the right teachers, firing the wrong ones, and training the ones they have. Maybe they would do better if they were subject to some competitive pressures. Vouchers are one way of providing this choice within a publicly funded system, publicly funded charters are another.
Merit pay is an indirect way of trying to improve teacher quality. I agree that paying teachers based almost entirely according to seniority isn't the best way to inspire excellence. But most teachers would probably improve their teaching if they knew how. Perhaps a bigger issue is not the level of teachers pay relative to other teachers, but their pay relative to other professions. In some places, teachers are paid a decent salary (although not in Utah) but it is still not comparable to the pay of other professional options such as medicine, law, or even military.
Accountability is major trend in school reform (anyone ever heard of No Child Left Behind? It's this new government program...). Like school choice, it is basically an indirect way of improving teaching quality (and maybe even T). But unlike school choice, accountability schemes rely on the education bureaucracy rather than consumer choice to put pressure on administrators to find the right teachers.
So there you have it, my quick and dirty classification of school reform proposals. One thing that doesn't fit easily into this classification scheme is student motivation. Perhaps the part of student motivation that can be impacted by educators should be considered part of E.
Subscribe to:
Posts (Atom)