Wald confidence intervals for a binomial proportion are known to perform poorly. Adjusted Wald intervals achieve better performance by adding in c successes and c failures (for an appropriate positive constant c) before computing the usual Wald interval. We study the performance of adjusted Wald intervals for a proportion in the context of ranked-set sampling, finding that such intervals outperform the existing intervals in the literature. In the process, we develop new methods for evaluating the performance of confidence intervals for a proportion in the ranked-set sampling context. We describe possible extensions and provide advice on the choice of c.