How Many Billions Are We REALLY Spending on Testing?
The Brookings Institute released a report last week that purports to calculate the amount of money being spent on standardized tests.
The number researcher Matthew Chingos comes up with is $1.7 billion, which, when we consider that we spend more than $600 billion a year on education, does not sound like all that much. However, it turns out that all he is actually counting is the contract cost that states pay directly to test vendors.
This report focuses on the costs of contracts between states and test-making vendors because they constitute the lion's share of state-level expenditures on testing. According to assessment cost data gathered by PARCC from its member states, of 21 states that provided both total assessment cost and contract cost data, 18 states reported contract costs making up more than 85% of total costs. Other state-level costs are surely important, such as the salaries paid to state assessment officials who play a vital role in selecting contractors and overseeing the vendors through test development, administration, and scoring. But such costs are difficult to track consistently across states, and usually represent a small fraction of the testing budget.
The roles played by school and district employees who aid in test administration and scoring are important as well, but the cost of this work is challenging to measure. Calculating such costs requires information on which employees have these responsibilities, their compensation levels, how much time they devote to test-related activities, and what work they would be doing if they weren't involved in testing. Future research should attempt to measure how significant these costs are, how they vary across different types of tests, and whether there are efficiencies to be gained by outsourcing more of the responsibilities currently delegated to teachers and administrators.
Indeed future research SHOULD look at all these costs, and until we have such research, the number that Mr. Chingos provides us is really meaningless. It is like calculating the cost of raising a child by looking at the hospital bills associated with its birth.
Meanwhile, the experts over at Pearson publishing, which makes its money from tests and associated products, believe that we should not worry too much about the money and time spent on testing. Steve Ferrara writes,
Let's estimate that students in grades 3-8 spend about ten hours on end of year tests. In a 180 day, six hours per day school year, that amounts to just under one percent of the school year (i.e., 6 x 180 = 1,080 hours; 10/1080 = 0.93 percent), or about a day and a half of school per year. Compare that to the time that may be lost on the day before school vacations, the last week of the school year, assemblies, organizing at the beginning and end of the school day, disruptions, etc. Ten hours a year seems like time well spent, to me.
Once again, I think these estimates are rather remarkable. When high stakes are applied to tests they consume far more time than the minimal figures cited here. As one commenter, Katie Stafford Strom points out:
Students DO NOT spend "ten hours" on testing. Only someone who has never spent time in a classroom would make this claim. The average school spends two weeks on the state's standardized tests--because if you know anything about children, it's that their natural inclination is not to sit perfectly quietly and fill in bubbles for six hours straight. To ensure that students stay focused, schools test for a few hours in the morning-- they often have abbreviated days on testing, or they have students go to one other class and watch movies (any teacher will tell you that little actual learning happens, if any, during testing weeks). Add to this other testing days- in New Jersey, there are a whole assortment of tests that kids in different grades get, and in Newark (where my work is centered) there are district-mandated standardized assessments for core subjects. And accountability mania- all centered on standardized tests as the gold standard of measurement- means the time spent on testing is time not spent on learning (perhaps the one thing I agree with you about). So students don't lose HOURS of learning, they lose WEEKS of learning.
Another educator, Mary Beiger, comments:
I think you do not realize what the weight of these tests is doing to our education system. It's not just the time with the students. Staff meetings and Staff Development Days have become centered around the results of these tests. We analyze data, try to find the root cause for what went wrong with certain questions, and then bend over backward to find a better way to teach this concept. Students must take benchmark tests, practice tests, so we can analyze more data, discover if there is growth or if we are still in danger of getting the answer wrong on the following year's test. Of course, the questions change from year to year, and inevitably it becomes impossible to compare separate tests. Teachers will be evaluated on the ratings, nonetheless. Teachers no longer have the time to communicate and collaborate with each other. We are too busy crunching numbers, filling out forms, trying to keep up with the many new incentives that have been forced upon us.
Last summer at the National Education Association's Representative Assembly,
9,000 teachers voted to approve New Business Item (NBI) #82 which calls on the NEA leadership to make available a toolkit and assist and encourage Union locals to collect and publish all manner of data relating to testing and test prep. Through a battery of surveys and tools accessible through the NEA web resources, rank-and-file members will be able to get a comprehensive analysis of precisely how much taxpayer money and instructional time schools, students, and teachers are being spent on testing and test-prep in their municipalities, states, and ultimately in the nation.
We have not heard any results from this investigation yet, but this is obviously critically important information that we may not get from any other source.
What do you think? Are we getting accurate estimates of costs and time spent from these experts? Or is it time for those with direct expertise to weigh in?
Continue the dialogue with me on Twitter at @AnthonyCody