“
While there are some bureaucratic hurdles at the EPA, it is probably just like any other
job in industry, government, or academe. But, the rewards are many
”
.
I Thought Statistics Is a Science; Does It Really
and maintained properly whose emissions represent those of other
Matter in Policy?
such vehicles is a difficult problem and quite a bit different from
Oh, yes. There are many instances in which the statistician at the
selecting a random sample of six marbles out of an urn containing
EPA plays a pivotal role in assisting in the formation of environ-
100 marbles. Reflect upon this. Most statistical techniques pre-
mental policy. Let me give two instances in which it was quite
scribe what to do with the data from a random sample. Selecting
crucial: in an enforcement case and in a decision made by the presi-
that random, representative sample might be the toughest part.
dent of the United States.
How Big May I Think?
The legal case involved the United States vs. Chrysler Motors.
Based on random sample testing, the EPA had determined that
Think as big as you like, but, in one area, think small. Sampling is
properly maintained vehicles were violating the federal carbon
expensive, and environmental sampling is usually quite expensive.
monoxide standards. The EPA chose to consider the emissions
At the EPA, we have to do the best we can with small samples or
results in a binomial manner. That is, either a sampled car passed
develop models. I mentioned the enforcement case earlier. We won
the standard or failed it. We did not use the actual score in some
the case and Chrysler had to recall 208,000 vehicles. What was the
sort of parametric distribution. Part of the reason was to conform
sample size required to affect such a recall? The answer is a mere 10.
to the legislation, which just mentioned meeting or failing stan-
It is both an affirmation of the power of inferential statistics and a
dards (no discussion of margin) and part was to avoid distribu-
challenge to explain how such a sample could possibly suffice.
tional assumptions that could be challenged in court. So, for policy
So, Is It Rewarding, or Just Bureaucratic?
and legal reasons, we simply used pass/fail.
The second example was actually quite simple, but revolved
While there are some bureaucratic hurdles at the EPA, it is probably
around a judicious choice of stratification. In the mid-1970s, the
just like any other job in industry, government, or academe. But,
United States suffered gas lines as motorists tried to fill up. As the
the rewards are many. I think back on my career and realize how I
lines were due to a foreign embargo on fuel, one policy consider-
was an integral part in reducing and ultimately removing lead from
ation was that by allowing more lead in gasoline (yes, it existed in
gasoline, how I was a part of defending a major vehicle recall that
every gallon at that time), the crude oil usage could be stretched
led many other automobile manufacturers to strengthen their qual-
further to produce more gallons of suitable gasoline. Lead is a seri-
ity procedures to produce cleaner cars, how I was the developer of a
ous neurotoxin, and EPA was busily phasing it out of gasoline. On
motor vehicle tampering survey that assessed the degree of remov-
the other hand, gas lines were causing all sorts of social and eco-
ing or disabling emission control equipment, which led to new
nomic disruptions. The country was looking for a solution.
programs to outlaw and deter this. And, that is just me. I certainly
At a White House meeting, the EPA administrator presented
believe it is worth the occasional hassle. ■
President Jimmy Carter with a graph showing blood lead levels
and gasoline lead levels over time. They both showed amazing cor-
relation in that both were going down, but they exhibited the same
seasonality. However, what really caught the president’s eye was not
just the general correlation. The EPA had prepared the graph by Professional Accreditation
stratifying the sample into black and Hispanic populations. Blood
lead levels were decreasing for both groups, but not as quickly as for
What is it?
whites. The president did not want a policy to be disproportion-
ately adverse to minority populations. He held firm on the con-
Is it for me?
tinued reduction of lead. An astute stratification strongly helped
formulate policy!
Why would the ASA want to support it?
These are only two examples of the applied use of statistics in
major decisions. I was fortunate enough to be directly involved in The ad hoc Committee on Professional Accreditation
both. Multiply this by many areas of the EPA and it is easy to see the
for Individual Members has set up a web site where
value of statistical analysis, and the power to make a difference.
you can find answers to these questions and more.
Is Statistics in the Real World Very Difficult?
You also can link to recent articles about accreditation
Frequently, it is what you make it. We want statistical arguments to be
and a set of frequently asked questions, provide your
accurate, applicable, understandable, and defensible. After all, the EPA
comments (anonymously, if you wish), and post
is sued a lot from all sides, so it pays to have done the work correctly.
questions to committee members.
There are frequently needs for innovative answers that might
require collaboration with university professors. While these may
not be “hard,” they are certainly new and challenging. You might
Go to www.amstat.org/comm/accreditation.
also find that even the “easy” stuff is not so easy. For instance,
selecting a random sample of motor vehicles that have been driven
MAY 2008 AMSTAT NEWS 3
Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80  |  Page 81  |  Page 82  |  Page 83  |  Page 84  |  Page 85  |  Page 86  |  Page 87  |  Page 88  |  Page 89  |  Page 90  |  Page 91  |  Page 92