12.02.2013 Views

Environmental Health Criteria 214

Environmental Health Criteria 214

Environmental Health Criteria 214

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

HUMAN EXPOSURE ASSESSMENT<br />

binary responses. The possible responses can be generally labelled as<br />

success or failure. Often we are not interested in a single outcome,<br />

but rather in the number of successes (k) and failures (n - k) for<br />

a specific number (n) of repeated independent trials for the<br />

outcome. The probability of exactly k successes in n independent<br />

trials, given a probability of success (p) in a single trial, is<br />

given by the binomial probability distribution ( P k ) in Table 10.<br />

For example, assume daily exceedances of an ozone air quality<br />

standard are independent events in a study of 1-year and 3-year time<br />

periods. Let k be a random variable describing the total number of<br />

exceedances encountered in a 1-year period ( n = 365 days). Further<br />

assume from historical data that the expected number of exceedance<br />

days each year is 1, thus p = 1/365 = 0.00274. The calculated<br />

probabilities of k days of exceedance per year are shown in Table<br />

12. Examination of the resulting probabilities in this example reveals<br />

a right-skewed distribution with the greatest probability occurring<br />

between k = 0 and k = 4 days.<br />

4.3.4 Poisson distribution<br />

Some exposure-related measurements are expressed as a rate of<br />

discrete events, i.e., the number of times an event occurs per unit<br />

time, such as the frequency (times per week) that a person consumes an<br />

ocean fish containing a methylmercury concentration greater than<br />

5.0 ppm. The Poisson distribution is used for describing potentially<br />

unlimited counts or events that take place during a fixed period of<br />

time (i.e., a rate), where the individual events are independent of<br />

one and other. The Greek letter lambda is typically used to denote the<br />

average or expected number of counts per unit time. In a Poisson<br />

distribution, the parameter lambda also describes the variance of the<br />

random variable. We can think about this intuitively by noting that as<br />

the expected number of counts or events increases (i.e., the rate of<br />

http://www.inchem.org/documents/ehc/ehc/ehc<strong>214</strong>.htm<br />

Page 66 of 284<br />

6/1/2007

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!