AEIdeas

The public policy blog of the American Enterprise Institute

Subscribe to the blog

Discussion: (31 comments)

  1. PeakTrader

    It’s easy to make false assumptions.

    Someone may note states with a higher minimum wage may have a higher unemployment rate. Therefore, a higher minimum wage causes unemployment.

    However, controlling for other factors, e.g. systematic differences in environmental standards, other labor standards, tax rates, etc., we may find a higher minimum wage doesn’t cause unemployment or causes employment.

    It’s possible, a rise in the minimum wage to $10 may increase employment, or a rise to $12 may decrease employment, etc..

    Also, over time, we can seperate inflation from efficiency in revenues and expenses of firms. Generally, when inflation is high, it’s high in both revenue and expenses, or when it’s low, it’s low in both revenue and expenses.

    1. Peak: “It’s possible, a rise in the minimum wage to $10 may increase employment, or a rise to $12 may decrease employment, etc..

      And if that’s true, how would anyone determine what that optimum minimum wage is, other than through trial and error? An obviously unworkable approach.

      How often would it need to be adjusted considering the dynamics of a large economy and ever changing labor market?

      Who, or what group of whos, could possibly determine what the minimum wage should be next month, without risking disaster?

      And why is increased employment the issue rather than increased prosperity?

      Wait! Come to think of it, there’s already a mechanism for determining the optimum wage not only at the minimum, but at all levels. I think it’s called the *market*.

      Never mind, Peak, we no longer need to concern ourselves about what the minimum wage should be. The problem is solved.

  2. Seattle Sam

    I once saw the State of Minnesota advertise that “Minnesotans live longer than people in most other states”. The statement is true, but fails to point out that the significant variable is the high percentage of people with Scandinavian heritage.

    1. The statement is true, but fails to point out that the significant variable is the high percentage of people with Scandinavian heritage.

      Well yes, that and the fact that Minnesotans are basically refrigerated for half the year which makes everything last longer.

  3. Citizen B.

    “ceteris paribus”

    What incentive is there to do that if it is not good politics?

    1. Aiken_Bob

      The sad thing is that is so true. What is interesting to me – is with all the information available so easily now how do crappy studies get published. I have to believe that as a whole we have become very lazy

  4. Walt Greenway

    How is comparing all male average earnings with all women average earnings without controlling other factors any different from separating all teachers from non-teachers and comparing headcounts without controlling for other factors such as direct student engagement by teacher’s aides, counselors, and librarians?

    1. Citizen B.

      Walt, one is showing the lack of wage disparity between sexes and the other is showing bloated headcount. Control all you want but the results will still show bloated headcount since WWII in public education.

      1. Walt Greenway

        I am not saying there isn’t bloat in schools. I am talking about research methodolgy. I am saying one “study” uses a methodology of using individual factors to find details of earning disparity between genders and the other “study” uses a meat cleaver bifurcation methodology of teacher and non-teacher headcount (which puts even teacher aides who are in the classroom into the administration bloat category). Comparing teacher and non-teacher ratios without controlling for student proximity and interaction is the same exact thing as comparing women and men earnings ratios without controlling for other factors like marriage.

        1. Citizen B.

          “Comparing teacher and non-teacher ratios without controlling for student proximity and interaction is the same exact thing as comparing women and men earnings ratios without controlling for other factors like marriage.”

          No, it is not the same thing.

          Women volunteering to make less money does not create wage disparity.

          Teachers don’t voluntarily give up their positions to increase school district testing monitors.

          Isn’t the voluntary giving up of income vital to the narrative and understanding of male/female wage comparisons?

          1. Walt Greenway

            The narrative of women voluntarily giving up pay is comparable to counting classroom aides and librarians as teachers instead of administrators. You can’t massage data to get the results you want.

          2. Citizen B.

            So Walt, the male/female wage disparity narrative untrue, right?

          3. Walt Greenway

            Citizen B. if you accept the data, the disparity in earnings between the genders is there, but the research shows that factors other than gender probably cause the disparity. It’s a nice piece of research that digs down to find out what is really going on.

    2. For the purposes of a blog post, I created a summary table showing teaching vs. non-teaching personnel in 25 states and the US average for 2010. That table presents one part of the data and findings from two major, comprehensive, thorough and thoughtful (IMHO) studies by Dr. Ben Scafidi, who holds a Ph.D. in economics from the University of Virginia and is a full professor of economics at Georgia State. Links to both of his studies were provided in the blog post.

      The first study from October 2012 “The School Staffing Surge Decades of Employment Growth in America’s Public Schools,” was 32 pages, and contained 11 figures, 9 tables of data, 34 footnotes and 25 references.

      The second study from 2013 was a companion report that contains more state-specific information about public school staffing. That study was 25 pages, contained 9 tables, 2 figures, 19 footnotes and 16 references.

      Also, in Part II of the study, Dr. Scafidi responds in detail to five specific criticisms that have been leveled at his original report, which further supports the seriousness and quality of his research.

      Both of those reports provide a detailed historical analysis of the staffing surge at US public schools, and it would be extremely unfair to refer to these as “studies” (implying it’s not serious research) that “use a meat cleaver bifurcation methodology of teacher and non-teacher headcount.” If you want to criticize my blog post as such, that’s fine, but please don’t characterize Dr. Scafidi’s research, or the Friedman Foundation’s research standards, that way, that’s really not fair or reasonable.

      Now to partially answer your question about controlling for all relevant variables, that is a statistical requirement for a valid multivariate regression analysis, such as testing for wage differences by gender. The wage equation would be: Wages = f (Gender, race, age, experience, education, hours worked, work safety, marital status, number of children, union status, etc. etc.). Then you perform a regression analysis and test for the statistical significant of the estimated coefficient for the Gender variable. If that variable is negative and statistically significant, then you could say that there is a gender wage gap. But to compare average wages by gender without controlling for all of those other variables is not an analysis worthy of the term “study,” which is something students learn in Stats 101 or Econ 101.

      For Dr. Scafidi’s study, he’s not running a regression analysis, but he does compare: a) staffing head counts at public schools over time, b) staffing head counts by state, c) staffing head counts in the US to staffing head counts in OECD countries, etc. He also compares the increase in non-teaching staffing over time to student outcomes (graduation rates and test scores) over time, to assess that relationship. Further, he looks at student-teacher ratios over time, etc.

      In contrast, the IWPR report is only five pages, and presents finding of a “Gender Wage Gap” by only comparing average wage differences by gender in 2012 and previous years. There is no discussion of how age affects the genders wage gap, how marriage and motherhood affect the gender wage gap. Once we control for age, marital status, and children, the gender wage gap either disappears, or reverses in favor of women! To ignore age, marital status and children in a “study” on an important issue like the gender wage gap is such a serious deficiency that the results of such a “study” can’t be taken seriously.

      At the end of both reports from the Friedman Foundation, you’ll find a section on “Commitment to methods & Transparency,” which says, in part:

      “The Friedman Foundation for Educational Choice is committed to research that adheres to high scientific standards, and matters of methodology and transparency are taken seriously at all levels of our organization. We are dedicated to providing high-quality information in a transparent and efficient manner.

      The author welcomes any and all questions related to methods and findings.

      If you’ve read all 50 pages of both reports and still think that it is a “study” (in quotations suggesting that you don’t think it’s a legitimate study) that “uses a meat cleaver bifurcation methodology of teacher and non-teacher headcount,” then I suggest you send your detailed comments and criticisms to The Friedman Foundation and to Dr. Scafifi, and please copy me on those comments.

      1. Walt Greenway

        My “meat cleaver” comment was directed towards your table in your post from the data that split between teachers and everyone else without giving credit to the headcount of people who are directly engaged with the students every day such as classroom aides and librarians. It struck me as just as disenginous as the study that left out factors such as marital status in the gender gap study. Mark, I should have expressed my thoughts in writing better, and I apologize for my very rude remark to you.

  5. Walt Greenway

    Citizen B., you are aware you are showing bias using the term “bloated” instead of increased over time, right? Most academic studies that want to be taken seriously don’t use attention-seeking words in the titles: spin is usually used for advertising and getting elected to office.

    Diving deeply into data one time and simply and quickly splitting the data in two pieces another time could lead someone to think the desired results were predetermined and shaped by how the data are chosen and presented.

    1. Citizen B.

      “Citizen B., you are aware you are showing bias using the term “bloated” instead of increased over time, right?”

      Yes; and I was relunctant to come to this regrettable conclusion.

    2. The title of the post has been changed to: “The staffing surge (‘educratification’) in America’s public schools: ‘Educrats’ now outnumber teachers in 25 states.”

      I have also used the term “administrative and non-teaching bloat” several times in the post to clarify that it’s not just administrative positions that are growing. If you do a Google search for “administrative bloat,” you’ll see that it’s becoming a common term, and there are 26,000 results.

      Here are the facts:

      Since 1950, students attending public schools have increased by 96% (almost doubled), the number of public school teachers has increased by 252% (a 3.5X increase), and “administration and other staff” have increased by 702% (an 8X increase).

      Is “administrative bloat” an inaccurate, or unfair description of the reality? Perhaps “staffing surge” is more acceptable or politically correct, but is that really any different from, or more accurate than, “administrative bloat”?

      1. Walt Greenway

        I think the term bloat is bad unless you are discussing a dog or cow’s medical problem. I put it in the same pile as corporate greed, price gouging, left, right . . . . Some words seem to me to be designed to invoke an emotional response and carry bias even if no bias is meant by the author. Do we really need to read an article that has “Obamacare” in the title to know what is coming?

        1. PeakTrader

          I agree with Walt that using the term “bloat” is improper terminology. However, calling the “Patient Protection and Affordable Care Act” or the Affordable Care Act “Obamacare” isn’t necessarily biased and not inappropriate.

          1. Walt Greenway

            Yeah, even President Obama now gladly accepts the term Obamacare. But tell the truth, when you see that term in a title, don’t you have have a pretty good idea you will be reading an opinion article even if it is not on the op-ed page?

          2. Actually I find the term “Obamacare” to be less biased and closer to an accurate description than “Patient Protection and Affordable Care Act”.

    3. Walt: “Citizen B., you are aware you are showing bias using the term “bloated” instead of increased over time, right?

      Yes. the use of the word “bloat” is indeed biased, and its use, I’m sure, is intentional. That one word instantly conveys a sense of something that is unpleasant, undesirable, and something that needs to be corrected. That would seem to exactly describe an industry that costs ever more over time, while employing ever more people, for whatever purpose, whether in the classroom or not, while the quality of the product of that industry – educational achievement – hasn’t improved at all.

      Using the neutral term “increased over time” might require several pages, instead of one word, to describe the same undesirable condition.

      Using the word bloat may accurately describe most government enterprises, none of which have the correct incentives to promote efficiency and customer satisfaction.

      As you well know, Walt, private enterprises are driven by the profit motive and competition to satisfy customers with ever better products at lower prices. This just isn’t the case with public education.

      Ever higher costs and ever larger numbers of employees to provide the same outcome seems aptly described by the word bloat.

      1. Walt Greenway

        Ron, I’m not so sure educational attainment has not increased over time. My nephew has taken Spanish since 3rd grade and he and his friends can talk to each each as if they are from Mexico. His eighth grade math book has an acceleration physics problem I remember doing in college in the 1970s, and he and all of his friends have more computer skills than most people I know. On the other hand, my college students struggle adding a 6% sales tax on a billing invoice, but I have many students tracked to the vocational fields from high school.

        I hunted, fished, worked at the racetrack, and rode motorcycles when I was a kid. They build computers and design computer games instead now. Maybe we don’t know how to measure what we expect from non-college bound students using standardized tests.

        1. Walt: “Ron, I’m not so sure educational attainment has not increased over time.

          Heh. I’m sure your own personal impression of school achievement carries more weight with scholars than the The School Staffing Surge part 2, but the report addresses that issue with a discussion of the NEAP Long-Term Trend and Main Neap Assesment beginning on page 10. You might find it helpful.

          In any case, even if educational achievement had increased, the question remains as to why school employees per student and costs per student have risen so dramatically, when other industries provide better products at lower prices over time?

          1. Walt Greenway

            Ron, I’ve read the reports. Maybe they should write the standardized tests in the student’s choice of any foreign language the student chooses and have it written in computer language. Students are learning a lot more than reading, writing, and arithmetic today. We need to get them credit for those achievements that are value added and that the standardized tests are not currently measuring. Changing the testing standards, though, would complicate analyzing historical trends.

            I attend the school board meetings in my district, and I am familiar with the budget. A lot of the cost to educate students is in future and current health care cost (it’s 25.4% of our payroll costs). About 75% of our current expenditures are function coded instructional, student support, or teacher support (the other roughly 25% are building and grounds, adminstration, food service, transportation, and community service).

            Yes, I think schools can do better and that competition can be good. I also think that people could help their school districts out be attending meetings and getting involved instead of just saying let’s get rid of public education.

        2. Ron, I’ve read the reports. Maybe they should write the standardized tests in the student’s choice of any foreign language the student chooses and have it written in computer language.

          That would be a great idea as long as the student or his parents are willing to pay the additional costs. There’s no reason for taxpayers to pay for multiple language tests. Mind you, I don’t care what language the tests are written in, as long as there is only one.

          Students are learning a lot more than reading, writing, and arithmetic today. We need to get them credit for those achievements that are value added and that the standardized tests are not currently measuring.

          If you mean computer skills, you should be aware that they are pretty much universal among young people today, are part of our lives, and are expected not only in higher learning but in many job descriptions. They are no more exceptional today than phone skills were years ago, or the ability to program one’s VCR.

          As for being bilingual, I would expect anyone – especially a young person – to be really fluent after nearly 6 years of language classes. While that’s an admirable accomplishment, I’m not sure why you think it’s remarkable, and I presume your nephew got credit for his classes.

          You may be comparing what’s normal for young people with what your own students are capable of. Yours sound like real tough challenges from an education standpoint.

          Changing the testing standards, though, would complicate analyzing historical trends.

          You, like some others who comment here, may be assuming that uniformity and conformity to some third party authority’s standards are more important than the standards of the students and their parents.

          What we may agree on, is that institutions of higher learning and employers are finding the education provided by the public school system inadequate. Apparently the massive increase in school headcount per student isn’t improving students’ preparedness.

          I attend the school board meetings in my district, and I am familiar with the budget. A lot of the cost to educate students is in future and current health care cost (it’s 25.4% of our payroll costs).

          An excellent argument for reducing headcount. thank you, Walt.

          “<i. About 75% of our current expenditures are function coded instructional, student support, or teacher support (the other roughly 25% are building and grounds, adminstration, food service, transportation, and community service).”

          I don’t really care what the percentages are, the total is too high. A private, for profit business would find ways to reduce the cost of education, or their competitors would show them the way.

          Yes, I think schools can do better and that competition can be good. I also think that people could help their school districts out be attending meetings and getting involved instead of just saying let’s get rid of public education.

          No one is suggesting getting rid of public education, but just allowing competition for those student tax dollars. May the best models win. Those who can’t compete will lose customers and go out of business – just like in the real world.

          1. Walt Greenway

            Ron: The foreign language competency is a requirement to obtain a diploma for the class of 2016 and after in Michigan. It will not show up on the standardized test to show students any smarter; however, they are spending time that can’t be spent on testable material.

            The challenges of my students are a result of a college prep curriculum in high school that disregards the possibility of training people for other than employment that requires a bachelor degree. Even if we double the number of students who currently obtain a bachelor degree, four out of ten students still will not get one. Standardized testing needs to include the class time spent learning foreign languages, computer skills and electronic media, and vocational education.

            It’s really difficult to separate the people who want to eliminate teachers’ unions from those who want to make schools better. I think we can make our current public schools much better with the competition you speak about and much more involvement by parents in the public schools we already have. We only have about 800 students and an $8 million annual budget in my school district. I am not so sure what I suggest will work with the enormous school districts such as New York, Los Angeles, Detroit, and Chicago with thousands of students and city-like social and gang problems.

          2. walt: “The challenges of my students are a result of a college prep curriculum in high school that disregards the possibility of training people for other than employment that requires a bachelor degree.

            Well then, that sounds like a perfect opportunity for a system of private vocational and trade schools competing for students and their taxpayer funding.

            Standardized testing needs to include the class time spent learning foreign languages, computer skills and electronic media, and vocational education.

            Certainly, assuming standardized testing is all that important.

            It’s really difficult to separate the people who want to eliminate teachers’ unions from those who want to make schools better.

            I believe people see teacher’s unions as a problem in some cases, and believe that the unions, not so much the teachers, are not particularly interested in either education or children. there’s plenty of evidence to support that view. and of course collective bargaining by government employees is a really stupid idea in any case.

            I think we can make our current public schools much better with the competition you speak about and much more involvement by parents in the public schools we already have.

            Parent involvement is good, but parents voices are not very loud as they aren’t the direct customers in public education. School choice would change all that.

          3. Walt Greenway

            Ron, I suggested standardized testing for what students are actually learning today only because researchers who are publishing papers that students are not learning are citing current standardized tests. We are ready to teach vocational skills to the 40-60% of students who are not academic college bound, but we could use some support in K-12 that they are not second-class students. Have you ever seen any standardized testing to see those students’ achievement in high school metal or auto shop or how well a student presents themselves to an employer such as looking them in the eye when they speak instead of at the floor?

            Competition in education is good, but getting involved is better. And, yes, getting involved includes seeking, supporting, and communicating educational choices other than public schools.

  6. Another factor in the difference between gender earnings is hours worked. On an hourly basis, females are paid 88% of males. On a weekly earnings basis, females are paid 82% of males. Back of the envelope, 6% of the earnings gender gap is explained by fewer hours worked by females.

Comments are closed.

Sort By:

Refine Content:

Scholar

Additional Keywords:

Refine Results

or to save searches.

Open
Refine Content