The objective of this thesis is to further investigate the base test method and to develop an analytical model which can accurately predict the failure of an uplift-loaded standing seam roof system. Four sets of full-scale tests were conducted and their results were analyzed along with previous results (Anderson 1991). It was found that by normalizing the data, a reduction factor could be established for each test set configuration. This reduction factor applies to both the base test and confirming multi-span test in each test set. Three analytical models were applied to predict the failure of the purlin. It was found that only the LaBoube Method accurately predicted the yield stress for only the simple span tests. It was determined that a variance exists in the failure loads of uplift-loaded standing seam roof systems which need to be investigated through testing. / Master of Science