TYPE II ERROR BIAS IN GOVERNMENT DECISION MAKING: THE CASE OF KATRINA*

Type II error bias is not limited to just the FDA. In the April 2006 journal Public Choice Russel Sobell and Peter Leeson explain that type II error bias played a role in the delayed response of the federal government after hurricane Katrina.

Let’s say our government leader finds them self in the following predicament. After the destruction and the levees have broken someone has to make the decision to send in relief workers. However there are risks. Disease infested water, collapsing buildings and roads, roaming bandits, toxic chemical exposure etc. If our leader sends in workers and they meet a terrible fate, the consequences are on his shoulders. Later the media would hang him over sending our brave heroes in harms way.

Given limited information, here is the model in the context of hypothesis testing:

Ho: X= Xo ‘it is too harmful to send in relief ASAP’
Ha: X= Xa: ‘the harm is trivial, and justified to send in relief ASAP’


Type I Error: sending relief workers into unnecessary danger

If on the other hand, our government official decides to wait, until he has better information he could avoid this. The consequences may be the lives of hurricane victims, but the blame can be shared with nature.

Type II Error: Over cautiousness that prevents a quick response by relief authorities

In general this is one of the explanations given in the Public Choice article explaining the federal government’s response to Hurricane Katrina. It is also one of the reasons in general that many of our leaders at the federal, state, local, levels fail to make timely decisions or provide leadership when needed.

TYPE II ERROR BIAS: THE FDA

Let’s assume that FDA researchers are reviewing a drug for possible approval. They are reviewing results and data from clinical trials etc. Of course there are some side effects , but only from a small percentage of the samples. Let’s look at a model of their decision making process in the context of hypothesis testing. Let’s assume:

Ho: X = Xo ‘drug is harmful’
Ha: X = Xa ‘drug is safe’

If the decision makers make the mistake of releasing a harmful drug, the consequences would be easily identifiable, and could be visibly traced to their decision. They would want to avoid this at all cost. In essence, they want to avoid making a type I error.

Type I Error: = releasing a harmful drug

From my previous discussion on hypothesis testing, we know that to decrease the probability of committing a type I error, we choose an alpha level that is lower. In a t-test, if we are really afraid of making a type I error, we might set alpha ( the significance level) at .05, .01, or to go overboard.001 etc.

As previously discussed, setting alpha lower and lower increases beta, or the probability of committing a type II error. Recall, a type II error is accepting a false Ho. In this case that would be equivalent to falsely concluding that the ‘drug is harmful’ when it could actually be released and improve the lives of millions.

Type II error bias = over precaution; setting alpha so low, or setting the standard of proof so high as to almost always reject Ho, and biasing the decision in such a way that the probability of a type two error (beta) is greatly increased.

As a result of type II error bias, many life improving drugs never make it to the market.

TRICKLEDOWN TAXES

This is in response to the article in the Daily News ‘New tax OKed on first reading’

http://bgdailynews.com/articles/2007/03/14/local_news/news/news3.txt

How much of the money in the county budget shortfall is related to illegal immigrants, or refugees that are brought into the county? If either of these are reasons contributing to our budget shortfall then we have a serious case of unfunded mandates.

First, is it not the job of the federal government to control immigration? Truly, illegal immigration is only a problem if it contributes to increased levels of crime strains both federal and local resources, in cases like we may have now with Warren County. If the federal government cannot resolve the illegal immigration problem, why should we have to pay for it locally?

Even with legal immigrants that are refugees, why should we have to pay for federal programs that are unconstitutional?

In 1794 Congress appropriated $15,000 for relief of French refugees. James Madison objected stating "I cannot undertake to lay my finger on that article of the Constitution which granted a right to Congress of expending, on objects of benevolence, the money of their constituents."

Have we amended the constitution since 1794 to allow for this? If this is something that the people of the United States would like to do, then I do not have as much of a problem with it. It is in fact a very charitable act. But, it must be done with our consent. In this case, expanding the role of government beyond the powers specifically enumerated in the constitution requires consent in the form of a constitutional amendment. That requires a vote of two-thirds of both the House of Representatives and the Senate, followed by a ratification of three-fourths of the state legislatures.
This may seem like a very drawn out process, but is was designed by our founders to ensure the ‘consent of the governed.’ According to economist Walter E. Williams of George Mason University, two-thirds of federal spending has by-passed these criteria for consent. It is what economist Thomas Sowell has coined ‘the quiet repeal of the American Revolution’ in his book ‘The Quest for Cosmic Justice.’

So now, we are paying higher taxes at the federal level without our consent. As a result of these federal programs and failure to secure our borders we are having resource problems at the local level. Is that why they are proposing a tax on insurance premiums? Is this a ‘trickle down’ effect? Are these ‘trickle down tax increases’?

TYPE I AND TYPE II ERRORS

In an experimental setting where a statistical test of a hypothesis is conducted one may either reject the null hypothesis ‘Ho’, or fail to reject Ho. Let’s assume that the true population parameter we are testing is X. Our null hypothesis may be Ho: X=Xo.

The probability of rejecting the null hypothesis when it is true can be defined by:

Alpha = P (reject Ho X = Xo)

Rejecting the null hypothesis when it is actually true is referred to in statistics as a type I error. Therefore alpha is the probability of ‘committing’ a type 1 error. In a basic statistics, when you conduct a basic t- test ( i.e. reject Ho if t > t-critical) at the 5% level of significance, you are establishing a 5% chance of committing a type one error.

Of course, you may fail to reject Ho, or loosely speaking ‘accept’ Ho.

The probability of ‘accepting’ Ho when it is false can be defined by:

Beta = P (accept Ho X = Xa)

Speaking heuristically, accepting a false Ho is referred to as a type II error. Beta is therefore the probability of committing a type 2 error.

It turns out, that as you increase the significance level of a test ( by making alpha lower and decreasing the probability of a type I error), the probability of a type II error (beta) increases. This is the theoretical basis for ‘type II error bias.’

GOVERNMENT DECISION MAKING AND TYPE II ERROR BIAS

It has long been affirmed by economists that the FDA’s drug approval process suffers from type II error bias. This results in fewer drugs coming to the market that could improve the well being of millions. In the April 2006 journal Public Choice Russel Sobell and Peter Leeson explain that type II error bias played a role in the delayed response of the federal government after hurricane Katrina. In the next few posts I plan to explain type I and type II errors, and type II error bias in government decision making.

Bt COTTON AND ENVIRONMENTAL HEALTH

Because Bt cotton and Bt corn provide pest protection in the plant( a gene is expressed that is toxic to lepidoptera species of insects) there is no need for the less effective broad spectrum application of insecticides. As a result only insects that feed directly on the plant are affected. This is a great stride in protecting diversity and beneficial insect populations. In addition it reduces worker exposure to pesticides (with regards to both conventional and organic production). For example, in China, 400-500 deaths per year are attributed to pesticide poisoning. Since the introduction of Bt crops in China, pesticide applications have decreased by 75%.

The Bt plant genetics essentially displace the manufactured synthetics. According to research from Dr. Roger Leonard at the LSU Agricultural center and Dr. Ronald Smith at Auburn University, 1.04 million fewer pounds of insecticide are applied each year as a result of the reliance on Bt genetics in cotton alone. As a result about 2 million gallons of fuel are saved each year from decreased manufacturing and distribution of chemical pesticides. In addition, 2.41 million gallons of fuel and 93.7 million gallons of water are saved on the farm from fewer insecticide applications.

Sources: Dr. Roger Leonard, LSU Agricultural Center. Dr. Ronald Smith, Auburn University. Gregory Conko “The Benefits of Biotech” Regulation Spring 2003.
Gerald C. Nelson Genetically Modified Organisms in Agriculture: Economics and Politics San Diego Academic Press 2001.