Testing interlanguage pragmatics has only recently caught the attention of the scholars in the field (Brown and Ahn, 2011;Liu, 2010) and currently there exist a multiplicity of uncharted areas and unanswered questions. This study is an attempt to develop a test of pragmatics in the Iranian context in relation to the speech acts of request and apology. Test development constituted five stages of exemplar generation, likelihood investigation, scenario generation, initial multiple-choice discourse completion (MDCT) piloting and the final MDCT administration. A total of 245 male and female participants from Sharif University of Technology (SUT) and Iran Language Institute (ILI) participated in different stages of test development. Rasch analysis was conducted for statistical analysis. The results indicate that the test is fairly unidimensional with one misfitting item and no misfittings persons. The test has local independence. Moreover, item reliability is found to be .88 whereas person reliability is found to be .64. High item reliability is attributable to the large number of participants.