Correlation coefficient is a statistical measure of the extent to which measures of two variables are related to each other. For example, in ergonomics we know that a tall person is also likely to have long legs and arms. Thus we can say that height of a person is closely correlated to length of legs. The correlation coefficient can vary from -1 to +1. A correlation coefficient of 1 between two variables indicates that these are directly proportional to each other. A correlation coefficient of 0 indicates that the two variables are absolutely independent. A correlation coefficient of -1 indicates that the two variables are inversely proportional.
Anthropometrist often find it more convenient to estimate some anthropometric dimension as combination of more than one independent dimensions, rather than ascertain the dimension by direct field study. Foe example it is better to ascertain the distribution of dimension of the width occupied by two persons siting on a bench it is better estimate it as sum of widths of individual person. In cases like this the we need to add anthropometric dimensions. In other cases the relevant anthropometric measure may be obtain as difference of two dimensions. For example, to design a see-saw the designer will be interested in distribution of difference in weight of two persons sitting at two ends of the see-saw. In such cases the required measurement can be estimated as difference in weight of two persons.