A readily applicable procedure for point and interval estimation of coefficient alpha is outlined for complex sample designs, in cases where this coefficient is close to composite reliability. As a by-product, point and interval estimation of the change in alpha due to measuring instrument revision is also possible. This approach can be used as an aid for scale construction and development in empirical investigations involving analyses of complex sample survey data, in particular for secondary data analyses from large-scale studies. The procedure is illustrated with data from an educational survey.; A readily applicable procedure for point and interval estimation of coefficient alpha is outlined for complex sample designs, in cases where this coefficient is close to composite reliability. As a by-product, point and interval estimation of the change in alpha due to measuring instrument revision is also possible. This approach can be used as an aid for scale construction and development in empirical investigations involving analyses of complex sample survey data, in particular for secondary data analyses from large-scale studies. The procedure is illustrated with data from an educational survey.