Despite broad use of surveys and survey experiments by political science, the vast majority of survey analysis deals with responses to options along a scale or from pre-established categories. Yet, in most areas of life individuals communicate either by writing or by speaking, a fact reflected in earlier debates about open and closed ended survey questions. Despite good reasons to collect and analyze open ended data, it is relatively rare in the discipline and almost exclusively done through a process involving human coding of survey responses. We present an alternative, semi-automated approach, the Structural Topic Model (STM) (Roberts et al. 2013), that draws on recent developments in machine learning based analysis of textual data. A crucial contribution of the method is that it incorporates information about the text, such as the author's gender, country of origin, treatment status, or when something was written. This paper focuses on how the STM is extremely helpful for descriptive, exploratory, or inferential purposes for survey researchers and experimentalists. The STM makes analyzing open ended responses easier, more revealing, and capable of being used to estimate treatment effects. We illustrate these innovations with several experiments.