Applications to disability insurance (DI) have declined in recent years but extant research sheds little light on what is driving these trends. Research surveys and interviews based on self-reported data may not reveal more personal situations or include financially vulnerable populations. This study will help address these limitations by using a text analytics and text analysis approach to explore how individuals communicate with each other about DI on internet forums. Online forums and other social media platforms facilitate online communication in an open context. These communication platforms enable users to share their feelings, experiences, and advice in an informal and nonthreatening environment; as a result, they may provide information that is not available from formal surveys.
We collected data on online conversations that mentioned SSDI from seven online forums for the period 2004–2019. We then created a master data set of approximately 150,000 posts in roughly 15,000 unique threads. We conducted text analytics and text analysis to identify term and word frequencies, as well as topics modeling using unsupervised machine learning. We also developed preliminary collective associative networks (CANs) to delve deeper into the data. After describing our methods, we provide a summary of the findings and recommendations based on the outcomes.