Social science research often relies on coding of video or other data for subjective events. This can be very time consuming and labor intensive for a single individual. This project aims to allow researchers and other analysts to quickly and easily interact with large sets of video data to identify and mark subjective events by crowdsourcing the interpretation process. We have shown it is possible to make quick calls to crowd workers, then aggregate their solutions. Ongoing work aims to allow even more robust interactions, enabling users to have a 'conversation' with their data using the crowd as an intermediary.