Biserial correlation

Uses your system voice; voice and rate in settings.

Snapshot

≈ 3 min read

Names in other languages

No names in other languages are held for this entry yet.

Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.

The biserial correlation coefficient is a measure of association between a continuous variable and a binary (dichotomous) variable. It is used when the binary variable is assumed to have been created by dividing an underlying continuous latent variable into two categories. The underlying variable is assumed to follow a normal distribution.

For example, an examination result recorded as "pass" or "fail" may be treated as a dichotomized version of an underlying continuous variable representing a student's level of knowledge.

The biserial correlation coefficient is a measure of association between a continuous variable and a binary (dichotomous) variable. It is used when the binary variable is assumed to have been created by dividing an underlying continuous latent variable into two categories. The underlying variable is assumed to follow a normal distribution.

For example, an examination result recorded as "pass" or "fail" may be treated as a dichotomized version of an underlying continuous variable representing a student's level of knowledge. r_b = \left(\frac{\bar{x}_{1}-\bar{x}_0}{s_x}\right) \left(\frac{n_1 n_0}{n^2\phi\left(\Phi^{-1}\left(n_1/n\right)\right)}\right) {\sqrt{\frac{n}{n-1}}} , where \bar{x}_1 is the mean value of the quantitative variable in the first group defined by the dichotomous variable, \bar{x}_0 is the mean value in the second group, s_x is the sample standard deviation of the quantitative variable calculated with Bessel's correction (using n-1 in the denominator), n_1 and n_0 are the sample sizes of the first and second groups respectively, n is the total sample size, \phi is the probability density function of the standard normal distribution, and \Phi^{-1} is the inverse cumulative distribution function of the standard normal distribution. An alternative but equivalent formula is: r_b = \left(\frac{\bar{x}_{1}-\bar{x}}{\widehat{\sigma}_x }\right)\left(\frac{n_1/n}{\phi\left(\Phi^{-1}\left(n_1/n\right)\right)}\right) = \left(\frac{\bar{x}_1-\bar{x}}{s_x }\right)\left(\frac{n_1/n}{\phi\left(\Phi^{-1}\left(n_1/n\right)\right)}\right) {\sqrt{\frac{n}{n-1}}} , where \bar{x} is the mean value of the quantitative variable in the entire sample (both groups combined), and \widehat{\sigma}_x is the standard deviation of the quantitative variable calculated without Bessel's correction: \widehat{\sigma}_x = \sqrt{\frac {\sum_i(x_i-\bar{x})^2}{n}} = s_x \sqrt{\frac{n-1}{n}} .

See also

Tetrachoric correlation Polychoric correlation Point-biserial correlation coefficient

Text derived from https://en.wikipedia.org/w/index.php?oldid=1373675964 under CC BY-SA 4.0; modifications have been made.

Provenance

snapshot enwiki-001373675964 · captured 1788757116.0 · version history

Cite this page as: encyclopegia.org, “Biserial correlation”, snapshot enwiki-001373675964. https://encyclopegia.org/a/biserial-correlation

Sources cited in this revision

No external citation was extracted for this revision. That is a statement about the source, not a claim that none exists.