Ecological Fallacy - Group and Total Average

Group and Total Average

Ecological fallacy also happens when the average for a group is approximated by the average in the total population divided by the group size. Suppose one knows the number of Protestants and the suicide rate in the USA, but one does not have data linking religion and suicide at the individual level. If one is interested in the suicide rate of Protestants, it is a mistake to consider as an unbiased estimate the total suicide rate divided by the number of Protestants. This estimate implicitly assumes that the suicidal rate of the other religions is zero.

Formally, denote the variable of interest and the mean of the group, we generally have:


\begin{align}
E \neq \frac{E}{P(\text{Protestant})}\\
\end{align}

However, the law of total expectation gives


\begin{align}
E= {\color{Blue}E} P(\text{Protestant})+ {\color{Blue}E}(1-P(\text{Protestant}))\\
\end{align}

This equation must be used to compute a more congent estimate. In this equation, the only things we don't know how to estimate are in blue. We know that, as a probability of suicide, is between 0 and 1. We can plug in this bound and our total estimate in the equation above to obtain an estimate of .

Read more about this topic:  Ecological Fallacy

Famous quotes containing the words group, total and/or average:

    The virtue of dress rehearsals is that they are a free show for a select group of artists and friends of the author, and where for one unique evening the audience is almost expurgated of idiots.
    Alfred Jarry (1873–1907)

    Parenting is the one area of my life where I can feel incompetent, out of control and like a total failure all of the time.
    —Attorney Father. As quoted in Reviving Ophelia, by Mary Pipher, ch. 4 (1994)

    The idea that leisure is of value in itself is only conditionally true.... The average man simply spends his leisure as a dog spends it. His recreations are all puerile, and the time supposed to benefit him really only stupefies him.
    —H.L. (Henry Lewis)