The effect of ranked set sample (RSS) on regression analysis is investigated. In particular, parameter estimation, testing hypothesis of the simple regression model fit and residual analysis are investigated. It is shown that RSS gives unbiased estimates of the regression model parameters, and these estimators are slightly more efficient than those obtained by using SRS. Some inferences using asymptotic results are given. Numerical analysis, using the blood data from Iowa 65+ Rural Health Study (RHS), is used to study the performance of the estimators and the impact of the sampling schemes on regression analysis in general.