Early ICC Definition: Unbiased But Complex Formula
The earliest work on intraclass correlations focused on the case of paired measurements, and the first intraclass correlation (ICC) statistics to be proposed were modifications of the interclass correlation (Pearson correlation).
Consider a data set consisting of N paired data values (xn,1, xn,2), for n = 1, ..., N. The intraclass correlation r originally proposed by Ronald Fisher is
- ,
- ,
- .
Later versions of this statistic used the proper degrees of freedom 2N −1 in the denominator for calculating s2 and N −1 in the denominator for calculating r, so that s2 becomes unbiased, and r becomes unbiased if s is known.
The key difference between this ICC and the interclass (Pearson) correlation is that the data are pooled to estimate the mean and variance. The reason for this is that in the setting where an intraclass correlation is desired, the pairs are considered to be unordered. For example, if we are studying the resemblance of twins, there is usually no meaningful way to order the values for the two individuals within a twin pair. Like the interclass correlation, the intraclass correlation for paired data will be confined to the interval .
The intraclass correlation is also defined for data sets with groups having more than two values. For groups consisting of 3 values, it is defined as
- ,
- ,
- .
As the number of values per groups grows, the number of cross-product terms in this expression grows rapidly. The equivalent form
where K is the number of data values per group, and is the sample mean of the nth group, is simpler to calculate. This form is usually attributed to Harris. The left term is non-negative, consequently the intraclass correlation must satisfy
- .
For large K, this ICC is nearly equal to
which can be interpreted as the fraction of the total variance that is due to variation between groups. Ronald Fisher devotes an entire chapter to Intraclass correlation in his classic book Statistical Methods for Research Workers.
For data from a population that is completely noise, Fisher's formula produces ICC values that are distributed about 0, i.e. sometimes being negative. This is because Fisher designed the formula to be unbiased, and therefore its estimates are sometimes overestimates and sometimes underestimates. For small or 0 underlying values in the population, the ICC calculated from a sample may be negative.
Read more about this topic: Intraclass Correlation
Famous quotes containing the words early, unbiased, complex and/or formula:
“The shift from the perception of the child as innocent to the perception of the child as competent has greatly increased the demands on contemporary children for maturity, for participating in competitive sports, for early academic achievement, and for protecting themselves against adults who might do them harm. While children might be able to cope with any one of those demands taken singly, taken together they often exceed children’s adaptive capacity.”
—David Elkind (20th century)
“There is not a more disgusting spectacle under the sun than our subserviency to British criticism. It is disgusting, first, because it is truckling, servile, pusillanimous—secondly, because of its gross irrationality. We know the British to bear us little but ill will—we know that, in no case do they utter unbiased opinions of American books ... we know all this, and yet, day after day, submit our necks to the degrading yoke of the crudest opinion that emanates from the fatherland.”
—Edgar Allan Poe (1809–1845)
“All of life and human relations have become so incomprehensibly complex that, when you think about it, it becomes terrifying and your heart stands still.”
—Anton Pavlovich Chekhov (1860–1904)
“But suppose, asks the student of the professor, we follow all your structural rules for writing, what about that “something else” that brings the book alive? What is the formula for that? The formula for that is not included in the curriculum.”
—Fannie Hurst (1889–1968)