The National Education Policy Center (NEPC), out of the School of Education at the University of Colorado, Boulder, just put out two short research papers that conclude, class size matters, and money in education matters. I believe the papers are right on both counts, but as always when I site research, whether I agree or not, I have to add that no conclusions in education research are conclusive. Education has so many moving parts, it’s impossible to create perfect control groups or isolated variables. That being said . . .
The class size reduction study looks at data and research dating back to 1979, including the much-discussed Tennessee STAR (Student/Teacher Achievement Ratio) study where elementary students in a number of schools were randomly assigned to small classes of 13-15 students and larger classes of up to 25 students. According to the author,
The smaller classes performed substantially better by the end of second grade in test scores, grades, and fewer disciplinary referrals.
The gains lasted. The students that had been assigned to smaller classes were more likely to graduate in four years, more likely to go to college, and more likely to get a degree in a STEM field. The positive effect was twice as large for poor and minority students, and thus narrowed the achievement gap.
The finding that small class sizes most benefit poor and minority students isn’t surprising. Students who are less likely to succeed in school due to socioeconomic factors are more likely to benefit from increased academic and emotional attention from teachers than students who have stronger economic and educational support systems in their homes and communities.
According to one researcher, the improvements are both significant and cost effective.
[Alan] Krueger noted, as have many others, that class size reduction most benefits minority and disadvantaged students, and would be expected to narrow the racial achievement gap by about one-third. He also estimated that the economic gains of smaller classes in the early grades outweighed the costs two to one.
Class size in upper grades haven’t been studied as closely as in the lower grades, but indications are that smaller classes lead to short term and long term gains there as well.
The research paper discussing the importance of money in education is linked to the class size study, since increased funding usually leads to smaller class sizes. One group of researchers found especially strong correlation between funding and achievement.
[The researchers examined] data in 28 states that had implemented finance reforms between 1970 and 2010. In addition to measuring short-term outcomes, they followed up on long-term outcomes in students’ lives. The results were significant and meaningful. A 20% “increase in per-pupil spending each year for all 12 years of public school leads to 0.9 more completed years of education, 25 percent higher earnings, and a 20 percentage-point reduction in the annual incidence of adult poverty.” The gains were achieved primarily by lower student-to-teacher ratios, increases in teacher salaries, and longer school years. Gains were strongest for economically deprived children and were strong enough to eliminate from two-thirds to all of the adult outcome gaps between those raised in poor and non-poor families.
Other studies by other researchers have found less correlation between class size, money and achievement than the studies emphasized in these two reports, though the report says those studies tend to lump together good and bad data as well as good and bad studies, which muddies their results. But even the class size/money nay-sayers don’t deny that class size and money have a positive effect. They just say it’s minimal, not enough to justify the expense.
(I should add, to save commenters the trouble, that in recent scores on the national NAEP tests, most states’ scores fell while many Arizona scores remained stable or increased. Given Arizona’s low education budget and high class sizes, that seems to give the lie to the conclusions I’ve noted. It’s definitely something worth looking at, though one set of scores don’t necessarily indicate a trend, and Arizona’s changing demographics need to be considered to see how they contributed to the scores. I’m sure people will be looking at the next few rounds of NAEP testing to see if the Arizona trend holds up and if so, what it means.)
This article appears in Jun 16-22, 2016.
Probably true. But as long as the powers that be insist on using taxpayer dollars to educate children of illegal aliens and use schools as social engineering projects, the people with the money are going to continue to refuse to fork it over.
Why in the world would you not want “children of illegal aliens” to be well educated? Yes, it will cost something but they are going to be your neighbors, your employees (and maybe your employers), the taxpayers who support your Social Security, your first-responders, your national guardians, the people who vote, etc.
I also don’t understand why you would imply “schools as social engineering” is a negative concept. Of course schools are part of creating a better society.
Here we go again, class size versus money, money versus class size and the so called “experts”.
What doesn’t take a study, is to look at the effect of federal government intervention in education. Jimmy Carter implemented the Department of Education in 1979. After almost 40 years and billions and billions of dollars what exactly have we received thanks to federal involvement? Nothing. Except I would say it’s gotten worse. Of course, with any government bureaucracy the answer ALWAYS is: more money. How much longer do we keep pouring money into this black hole expecting better results? In what other investment would one accept such a poor return? When is it time to admit the Dept of Education is a failed idea, and try something else? Will it take 50 years? 75 years? 100 years? Where does this faith in big government come from? I don’t understand it.
The STAR study is so discredited. Their random assignment was completely busted by numerous high performance parents who weren’t going to allow their student to be randomly assigned to a large class. The first year of the study assumed that all students started at the same level, not measuring a baseline, because of the “random” assignment assumption. All of the gains were first year gains.
The Hubble telescope of education studies, The Early Childhood Longitudinal Study, which spent $140 million just collecting its data, had a random sample of 20,000 students, following students from the start of kindergarten to the end of 8th grade found something completely different. They found that class size did matter but in a more complex fashion. They found no class size benefit below 25 students but they found a sizable penalty as class sizes moved above 25. Our student to teacher ratio in Arizona is less than 20, so we are well in the game, if we can get almost all of those teachers in front of students.
David, on what basis do you assert “increased funding usually leads to smaller class size”? What are the studies showing this and who did them? Please provide citations.
In some poorly managed public districts serving poor and vulnerable populations where constituents don’t have the tools needed to hold district administration accountable, increased funding can result in more money going into the pockets of contractors and corrupt administrators, and anyone who has taken the trouble to study problems in urban education knows this very well.
The missing link between more money and improved results is a constituency active and vigilant enough to hold governance and administration accountable, yet you never advocate for that, either in theory or in local practice. I wonder why. Could it have something to do with your cozy relationship with the TUSD board majority?
Keep summarizing “research” in ways that supports increasing the money flowing into TUSD coffers. Support 123 while “holding your nose,” then immediately turn around and try to convince us you are a noble warrior combatting the Ducey/Koch forces of darkness. You’re not convincing many readers here…if you take a look at the likes/dislikes when you comment in your own comment streams, you may get the impression, as I have, that most readers see through your disreputable capitulation to Ducey and your related transparently propagandistic blog-scamming on behalf of the malfeasant leadership of TUSD.
jhupp…. above is right to refer us to the Early Childhood Longitudinal Study for some better understanding of the issues discussed here but he seems to be reading more into the study’s conclusions than are justified. Perhaps his biggest error is to extrapolate well beyond the study’s scope. The oldest children in the study are barely out of eighth grade.
I suggest that anyone who wants to really understand what’s going on spend the time to read at least some of the summaries of the study.
In response to Provide citations to back assumptions, please: Here is the quotation in the study I was referring to when I connected more funding to class size. It’s in one of the passages I quoted in my post:
“A 20% “increase in per-pupil spending each year for all 12 years of public school leads to 0.9 more completed years of education, 25 percent higher earnings, and a 20 percentage-point reduction in the annual incidence of adult poverty.” The gains were achieved primarily by lower student-to-teacher ratios, increases in teacher salaries, and longer school years.”
So that study had to be 13 years old or older to be true. 20% increase? That is insane.
Nope, sorry, David – the quote you provided in response to “Provide citations to back assumptions, please” — a quote you provided BOTH in the article and AGAIN in the comment stream does NOT provide evidence for your assertion that “increased funding usually leads to smaller class sizes.”
Your quotes say that during a 40 year period with data sampled from 28 states that had implemented “finance reforms,” there were indications that increased funding leads to more completed years of education, higher earnings, and reduction in incidence of adult poverty. The quotes say these gains were achieved “primarily” (?) by an unspecified combination of lower student to teacher ratios, increases in teacher salaries, and longer school years.
This is NOT the same as saying “increased funding WHEREVER AND WHENEVER IT OCCURS usually (i.e. IN THE MAJORITY OF ALL CASES IN WHICH IT HAS EVER OCCURRED) leads to smaller class sizes.”
Your ability to read and understand what study results can and cannot be said to entail is disappointingly weak, for a former English teacher.
The Early Childhood Longitudinal Study was completed many years ago. You are talking about the second phase. I am quoting from the first phase in which 8th graders completed in 2006.
20,000 students were randomly sampled from 2,000 different classrooms and even randomly sampled within classrooms. The tests were the best ever designed to measure the achievement of young children, using an adaptive design for great accuracy. Data was collected not only on academic achievement but also on motivation, pro-social and antisocial behavior. The sample sizes were so huge that subsets of subsets of subsets are statistically significant with small standard errors. It still stands as the best study ever conducted in education.
JHupp,
You say, “The Early Childhood Longitudinal Study was completed many years ago.” I’m not having any luck tracking down any such study. Can you give me some clues for finding the original study you refer too?
early childhood longitudinal study kindergarten class of 1998-99
that is the data set. there are probably thousands of studies done using that data set. you have to search in combination with the data set and what aspect of education policy you want to examine.
For example on the all-day kindergarten versus half-day kindergarten issue I had to go into the bowels of the Institute for Education Sciences to find out that all day kindergarteners were 10% of a standard deviation behind half-day kindergarteners at the end of 8th grade.
The problem with David’s research and “meta” studies in general is the poor quality of data collection and research in education. The National Reading Panel spent 10 million dollars examining the best 10,000 studies in reading and found out that 96.5% of them were not worth the paper they were printed on and the remaining 400 studies were of limited value.
In general, research in education serves to obscure the truth not to enlighten it. That is why the Early Childhood Longitudinal Study is so important. The data collection was done by outsiders without a dog in the fight, they had a fortune to collect the data in a quality manner and their sample size was unbelievably huge and they followed the same students for 9 years.
Money matters, but it is equally–if not more–important HOW money is spent. The Tucson Unified School District, for example, receives at least $340 million dollars per year in funding for approx. 50K students. However, the money is misspent in opaque and suspicious processes, which is the case in most failing or suspiciously opaque districts. The result is low or stagnant achievement levels.
When, for example, a district decides to buy e-textbooks before ensuring that all students have access to a computer, money is wasted and students suffer. But the administration remains popular, well-liked. When classroom aids are fired and long-term substitute pay is cut so the superintendent can take 6 figures home each year, money is wasted and students suffer. But the school board can say they have strived hard to stabilize the district by keeping one superintendent in place and they can pat themselves on the back in front of their constituencies. Students continue to suffer.
The problem in TUSD is not money (or overly large classes). The problem here is a combination of the school board and the superintendent and the faulty way they spend money on things that do not benefit the students. Indeed, the students suffer.