Career mobility is a long-standing research topic in social science that investigates the trajectories of individuals in terms of their occupation transitions. It is crucial for the study of social stratification, inequality, and policy-making in multiple disciplines, such as sociology, economics, and political science. With the availability of longitudinal career-related datasets from quantitative history and sociology, traditional statistical approaches are facing three challenges in analyzing these complex data given their sequential, network, and multivariate natures: (1) the lack of analysis from a dynamic longitudinal perspective, (2) the insufficient analysis of potential factors (such as social relations) that may affect careers, and (3) the lack of efficient tools to support ex...[ Read more ]

Career mobility is a long-standing research topic in social science that investigates the trajectories of individuals in terms of their occupation transitions. It is crucial for the study of social stratification, inequality, and policy-making in multiple disciplines, such as sociology, economics, and political science. With the availability of longitudinal career-related datasets from quantitative history and sociology, traditional statistical approaches are facing three challenges in analyzing these complex data given their sequential, network, and multivariate natures: (1) the lack of analysis from a dynamic longitudinal perspective, (2) the insufficient analysis of potential factors (such as social relations) that may affect careers, and (3) the lack of efficient tools to support exploring career mobility from different dimensions flexibly. In the meantime, visual analytics allows feeding domain knowledge into the interactive systems, which has brought new opportunities to solve the above three challenges efficiently. Social scientists can thus verify existing theories and generate new hypotheses under a human-computer collaboration process conveniently.

In this thesis, we focus on designing visual analytics systems for social scientists to address career mobility analytical problems efficiently from three perspectives. In the first work, we aim to obtain career mobility patterns to understand social mobility in different periods. We present CareerLens to explore over 340,000 government officials’ careers in the Qing bureaucracy in China. After obtaining career mobility patterns, a further step is to learn how potential factors that may affect one’s career. In the second work, we expand the scope of our research to academic careers and develop ACSeeker to investigate potential individual (e.g., working domain) and social factors (e.g., social relations) that may affect career mobility. Besides career-related factors, another important perspective that may significantly affect careers is private lives, such as marriage and childbearing. In addition, they may have a cumulative effect on careers. In the third work, we use WLViz to explore and compare work-family dynamics of different social groups (e.g., male and female groups). Finally, we discuss the future research perspectives on building visual analytics systems to facilitate career mobility and social science studies.