Does exposure to lead cause crime?

While driving to change the oil in my car yesterday, I listened to Philip and Alice Shabecoff discuss their book Poisoned Profits: The Toxic Assault on Our Children on the Diane Rehm show.

I haven't read it (don't plan to), but if the title and the interview are any indication, the book is polemic. The Shabecoffs have concluded, based on some vague anecdotal and correlational research, that a whole bunch of chemicals are toxic and should be banned.

Don't get me wrong -- I'm not a big fan of dumping/consuming random chemicals. But I'm also wary of beating up corporations just because they're easy political targets. And the evidence here is ... shaky. The problem is a statistical classic: correlation does not imply causation.

Case in point: a recent study on the effect of childhood lead exposure on adult criminal activity. The researchers targeted pregnant mothers in high-lead areas, took detailed readings of lead levels for the mothers and (later) their children, and then looked at the criminal histories of those children once they reached adulthood. They concluded that a little lead as a child makes a person much more likely to be arrested later in life.

The problem: which children are most likely to grow up in the homes with the worst lead problems? Those with the most impoverished, disadvantaged (and least health-conscious) parents. So when the researchers look at the correlation between lead exposure and arrests, it's impossible to tell whether lead or poverty is the real cause of crime.

So the study, like so many others is inconclusive. What do you think the standard of evidence for concluding that a chemical is harmful should be?

Michelle Rhee's merit pay plan: Out of control

Until recently, merit pay for teachers was a debate about control. Those in favor (often business/economics types) argued that merit pay motivates teachers and improves achievement by "aligning incentives." Those opposed (including, notably, the teachers unions) argued that merit pay debases teachers by subjecting them further to uncaring administrators and biased tests.

Now there's a new wrinkle in the debate: Michelle Rhee, Teach for America alum and current chancellor of D.C. public schools, is proposing a voluntary merit pay system.

(Check out blog reactions at The Quick and the Ed and DC Teacher Chic. These, and most of the reactions I've seen online have been positive.) (Don't be surprised -- think about who blogs.)

I haven't sorted through the pros and cons of Rhee's proposal, so I can't tell you if it's good policy. But politically speaking, voluntary merit pay is a real innovation, because it breaks the status quo perception of merit pay as a power grab by district administrators. It's pretty hard to argue that a plan is about control when entry is voluntary.

Rindskopf’s Rules for Statistical Consulting

(I stole this post wholesale from this site, mostly because I want to be able to find this text if I ever look for it again.)

Some of these rules are universal, while others apply only in particular situations: Informal academic consulting, formal academic consulting, or professional consulting. Hopefully the context will be apparent for each rule.

Communication with the Client:

(1) In the beginning, mostly (i) listen and (ii) ask questions that guide the discussion.

(2) Your biggest task is to get the client to discuss the research aims clearly; next is design, then measurement, and finally statistical analysis.

(3) Don’t give recommendations until you know what the problem is. Premature evaluation of a consulting situation is a nasty disease with unpleasant consequences.

(4) Don’t believe the client about what the problem is. Example: If the client starts by asking “How do I do a Hotelling’s T?” (or any other procedure), never believe (without strong evidence) that he/she really needs to do a Hotelling’s T.

Exception: If a person stops you in the hall and says “Have you got a minute?” and asks how to do Hotelling’s T, tell them and hope they’ll go away quickly and not be able to find you later. (I’ve had this happen, and if I ask enough questions I inevitably find that it’s the wrong test, answers the wrong question, and is for the wrong type of data.)

Adapting to the Client and His/Her Field

(5) Assess the client’s level of knowledge of measurement, research design, and statistics, and talk at an appropriate level. Make adjustments as you gain more information about your client.

(6) Sometimes the “best” or “right” statistical procedure isn’t really the best for a particular situation. The client may not be able to do a complicated analysis, or understand and write up the results correctly. Journals may reject papers with newer methods (I know it’s hard to believe, but it happens in many substantive journals). In these cases you have to be prepared to do more “traditional” analyses, or use methods that closely approximate the “right” ones. (Turning lemons into lemonade: Use this as an opportunity to write a tutorial for the best journal in their field. The next study can then use this method.) A similar perspective is represented in the report of the APA Task Force on Statistical Significance; see their report: Wilkinson, L., & APA Task Force on Statistical Inference. (1999). Statistical methods in psychology journals: Guidelines and explanations. American Psychologist, 54, 594-604.

Professionalism (and self-protection)

(7) If you MUST do the right (complicated) analysis, be prepared to do it, write a few tutorial paragraphs on it for the journal (and the client), and write up the results section.

(8) Your goal is to solve your client’s problems, not to criticize. You can gently note issues that might prevent you from giving as complete a solution as desired. Corollary: Your purpose is NOT to show how brilliant you are; keep your ego in check.

Time Estimation, Charging for Your Time, etc.

(9) If a person stops you in the hall and asks if you have a minute, make him/her stand on one leg while asking the question and listening to your answer. If they ask for five minutes, it’s really a half-hour they need (or more).

(10) Corollary: Don’t charge by the job unless you really know what you’re doing or are really desperate. Not only do people (including you) underestimate how long it will take, but (a la Parkinson’s Law) the job will expand to include everything that comes into the client’s mind as the job progresses. If you think you know enough, write down all of the tasks, estimate how much time each will take, and double it. Also let the client know that if they make changes they’ll pay extra (Examples: “Whoops, I left out some data; can you redo the analyses?”, or “Let’s try a crosstab by astrological sign, and favorite lotto number, and...”)

(11) Charge annoying people a higher hourly rate. If you don’t want to work for them at all, charge them twice your usual rate to discourage them from hiring you (at least if they do hire you, you’ll be rewarded well.)

Resources
http://www.amstat.org/sections/cnsl/index.html ASA section on consulting
http://www.amstat.org/sections/cnsl/BooksJournals.html Their guide to books and journals on statistics
Boen, J.R. and Zahn, D.A. (1982) The Human Side of Statistical Consulting, Lifetime Learning Publications.
Javier Cabrera and Andrew McDougall. (2002). Statistical Consulting. Springer-Verlag.
Janice Derr. (2000). Statistical Consulting: A Guide to Effective Communication.. Pacific Grove CA: Duxbury Press, 200 pages, ISBN:0-534-36228-1.
Christopher Chatfield (1988). Problem solving: A statistician's guide, Chapman & Hall.
Taplin R.H. (2003). Teaching statistical consulting before statistical methodology. Australian & New Zealand Journal of Statistics, Volume 45, Number 2, June 2003, 141-152. Contains a good reference list on statistical consulting.

50 Things New Teachers Need to Know...

This blog post is best slice-of-life I've seen into the world of teaching in a long time. The anonymous author strikes me as a pretty good teacher. His advice is practical, and also a little disillusioning. Did my teachers see themselves as mentors cum bureaucrats cum parole officers this way?

PS: I found this link here, at Joanne Jacobs' blog. For anyone interested in education, this is a blog worth keeping an eye on.

Looking for a flashlight in the dark

Today, we bring you a true story...

Two years of looking for a flashlight in the dark: An astounding case study in government decision making

In 2006, the Michigan legislature passed the Michigan Merit Curriculum (MMC), a tough new set of high school graduation standards. Among other things, the standards called for every student to graduate with four years of math. Previously, students needed about three years to graduate.

Unintended consequence: in one blow, the MMC increased statewide demand for math teachers by a third. More math classes means more math teachers. Since math teachers were already in short supply and teacher accreditation (going through ed school, student teaching, etc.) usually takes years, this created a huge pipeline problem for Michigan's school systems. Apparently, nobody -- not the legislature, the Michigan Department of Education (MDE), or any of the research groups urging the curriculum change -- had thought through this part of implementation. MDE worried about it, but didn't have the capacity to really come to grips with the problem.

So after the legislation passed, MDE commissioned a study of "teacher supply and demand" from a group of policy students at U of M. I was one of the students, and -- I can say this now -- we never really understood what our task was. The research question was very broad and vague, so we did the study that seemed most interesting: an investigation of the distribution of "high quality" teachers across the state. We based our work on the registry of education personnel (REP), a statewide database that tracks all Michigan teachers. Our statistics were decent, but didn't speak directly to the question MDE needed answered. This turned out to be the first in a long sequence of research projects trying to figure out where the extra teachers would come from.

The next year, another U of M student group picked up where we had left off. They had somewhat better data and a clearer understanding of their goal, but were still overwhelmed. They arrived at one major conclusion: the REP doesn't contain the information necessary to answer the real questions about teacher supply and demand. The REP can only tell us about teachers currently teaching, not the pool of potential teachers. The team's recommendation was a statewide survey of building administrators to assess real needs.

At the same time, a research team at MSU was working on getting deeper into the REP to see what it *could* tell us. They came up with a set of three reports, gradually decoding the data structures of the REP. Major conclusion: Michigan has almost no reserve pool of practicing math teachers. Four out of five teachers currently teaching and certified to teach math are already teaching math. Again, the conclusion was that Michigan would need to bring in more math teachers from outside the pool of currently practicing teachers.

Bear in mind that all of this analysis so far has been based on the REP -- data that the state has collected routinely for years. Two years and five report later, MDE still doesn't know anything about potential teachers *not* captured in the REP.

By now it's 2008. The Merit Curriculum has been in place for two years and the first cohort of high school freshmen have gone through their first year of math. Newspapers are starting to run articles about failure rates. The most extreme claim that over 20% of high school freshman failed algebra I, the first class in the sequence. Editorial pages are running apocalyptic predictions about drop-out rates and missed graduations. The legislature is starting to discuss repealing MMC.

MDE is desperate to increase the number of competent certified math teachers, but still doesn't know where to look. The next school year starts within a month. The clock is ticking...

Moral of the story?
Beats me. "Reform is messy." / "Reform is hard." / "Reform takes time." "Unintended consequences dominate good intentions." ?

Externalities dominate...

Check out this NYTimes Freakonomics blogpost on the recent NBER paper "Externalities in the Classroom: How Children Exposed to Domestic Violence Affect Everyone's Kids." I love the paper, hate the title.

The paper... is a rare quantitative glimpse into the impact that home life has on schools. I skimmed it; it isn't perfect, but it's credible. The basic conclusion:

we estimate that adding one more troubled boy peer to a classroom of 20 students reduces boys’ student test scores by nearly two percentile points (one-fifteenth of a standard deviation) and increases the probability that boys commit a disciplinary infraction by 17 percent (4.4 percentage points).

The title... makes me xenophobic. ("Are there kids like that in my brother's classes? How could we get rid of them?") It turns abused kids into damaged goods. It's a fair characterization of the evidence presented in the paper, but it's not the angle I would have taken.

In my mind, the real story is about the cost of broken relationships. It's more sad than surprising to see that really bad parenting can influence not just their own kids, but all the kids around them enough to show up on test scores and administrative records.

Key takeaway: People are remarkably interconnected. Externalities dominate any place where people are learning.

A PS -- I found the comments on the blog funny in a dark way. It's a mixed bag of snarky, offensive, shallow, and pompous. A few are insightful. I suppose we're all better for this kind of discussion, but the process is pretty messy.

Postracial indeed -- Obama's take on affirmative action

Debates about affirmative action have always made me uncomfortable. I believe in fairness, mainly in the form of meritocracy -- equality of opportunity, not outcome. However, it's also obvious that opportunities are not equal in our society, and a lot of those inequalities are racially distributed. Our massive achievement gaps are proof of that.

My (limited) experience suggests that in most parts of America, soft discrimination is a much bigger problem that overt KKK-style bigotry. I've always worried that racial affirmative action perpetuates the stereotypes it is meant to compensate for. Also, many black kids are born into well-educated, affluent families and it seems unfair to give them a bonus intended for the disadvantaged. So in the big pro/con debate of affirmative action, I've leaned con.

Of course, as a half-white, half-Asian child of upper-middle class parents with advanced degrees, it's simply not PC for me to speak against affirmative action. What do I know about race and discrimination?

So three cheers to Barack Obama for saying what I couldn't, as reported in Sunday's NY Times.

During a presidential debate in April, Mr. Obama said his two daughters, Malia, 10, and Sasha, 7, “who have had a pretty good deal” in life, should not benefit from affirmative action when they apply to college, particularly if they were competing for admission with poor white students.

Read the whole article for details; the main idea is that Obama is on record as saying that affirmative action should be based on socioeconomic status instead of race. His daughters have grown up privileged; they don't need or deserve affirmative action.

This take on affirmative action sits a lot better with my conscience. It shares opportunity fairly and broadly, but doesn't reinforce racial stereotypes. I don't know the details, but I like it.

I don't know how this idea will play out in public opinion, but I'm hopeful. Two years ago, Michigan went through a similar debate on Proposal 2, a constitutional initiative to make "preferential treament on the basis of race, sex, color, ethnicity, or national origin" illegal in the state. Campaigns on both sides fought angry and dirty. I quietly supported the Prop 2 for the reasons above. Liberal newspapers around the state castigated moderates making essentially the same argument that Obama is making. Affirmative action played out as a divisive polarizing issue.

This time, the combination of messenger and message seems to have some resonance. (See the same NY Times article for some details.) If Obama pushes, he may have the reach and credibility to reshape take the affirmative action debate beyond race -- a real milestone for American politics.

We'll see how it goes, but so far on this issue I think Obama is living up to his potential for running a truly post-racial campaign.

PS - The picture in this post is borrowed from http://lib.colostate.edu/research/divandarea/bif/archive/bif03/affirm.jpg. I have no idea what their content is like. Go Google images!