Tetrachoric correlation is a statistical method used to determine the correlation between two ordinal variables. The spelling of this term can be explained phonetically by breaking down each syllable. "Tetra" is pronounced /ˈtetrə/, with the stress on the first syllable. "Choric" can be pronounced in two ways, either /ˈkɔːrɪk/ or /kəˈrɪk/, with the stress on the second or first syllable, respectively. The word "correlation" is pronounced /ˌkɒrəˈleɪʃən/, with the stress on the second syllable. Together, the correct spelling of the full term is "tetrachoric correlation."
Tetrachoric correlation is a statistical measure used to assess the relationship or association between two categorical variables that have been translated into dichotomous data. It is specifically utilized when the original variables are assumed to have underlying continuous normal distributions, but they have been recoded into two categories. This correlation coefficient is commonly employed in the field of psychometrics and is particularly useful when studying binary data.
The tetrachoric correlation coefficient is calculated by estimating the correlation between two latent continuous variables and then converting it into the bounded scale of the binary variables. It quantifies the strength and direction of the relationship, ranging between -1 and +1. A value of -1 signifies a perfectly negative association, 0 indicates no correlation, and +1 represents a perfect positive correlation.
This correlation measure has several advantages. It allows researchers to explore the relationship between two variables without assuming that they are normally distributed, making it applicable in various settings. Moreover, tetrachoric correlation can be employed to estimate reliability and validity coefficients, such as Cronbach's alpha and factor loadings, respectively.
In conclusion, tetrachoric correlation is a statistical technique used to examine the relationship between two categorical variables that have been recoded into dichotomous data. It provides a measure of correlation on a bounded scale, accommodating non-normal distributions while enabling the estimation of other psychometric properties.
The word "tetrachoric" in "tetrachoric correlation" originates from the Greek root "tetra-" meaning "four" and the Greek suffix "-choric" meaning "space" or "region". The term was coined by the American psychologist Paul K. Kline in the mid-20th century to describe a correlation coefficient used in statistics to measure the strength and direction of the relationship between two binary variables. The prefix "tetra-" signifies the presence of four discrete regions in a cross-tabulation table where the frequency counts are calculated to compute the tetrachoric correlation.