Goodness of Fit (GOF) tests for non-Location/Scale families of distributions have not been studied extensively. The primary challenge in this area is that the asymptotic distribution either does not have a tractable form or have a tractable form that is depended on the unknown parameters. Thus the determination of the critical values and rejection regions becomes impossible. Recently the rapid development of bootstrap theories and applications provides a new channel to explore this problem. The advantage of bootstrap is that we can use the bootstrap distribution instead of the asymptotic distribution to determine the critical values and rejection regions. Application of bootstrap usually requires extremely heavy computation. With the help of modern computers, this issue becomes less and less noticeable. This dissertation focuses on this subject. It intends to develop a bootstrap based GOF test. procedure that can apply to general non-Location/Scale families of distributions. A Monte-Carlo study was performed to evaluate its suitability for use with finite sample sizes and compare its power with two existing procedures. The results indicate that it is a reasonable procedure usually with power greater than that of its competitors.