Analysis of covariance based on a general linear regression model will then be used to determine whether any possible group differences remain after controlling for other factors such as race, sex, estimated dietary intake of Vitamin D and sunlight exposure (using month of study as surrogate for sunlight exposure).