SIGNATURE WORK
CONFERENCE & EXHIBITION 2022

In the modern age of information, a variety of new types of emerging data are generated by individuals. Data analysts, therefore, have more choice than ever before about what data to look at and deal with. Yet novices who want to learn data analysis remain poorly informed, sometimes even misinformed, about the choice of data to start with. What are the different types of emerging data? What are the features of such data? Where to get such data? What are the tips when trying to get such data? What to pay attention to when cleaning or pre-processing such data? As these freshly generated data depict individuals’ real behaviors and decisions in their daily life, knowing the answers to the series of questions concerning these data is becoming tremendously essential to study human behaviors (Box-Steffensmeier et al., 2022). In this paper, I will categorize these new data into various types based on their sources, i.e., publicly available data sets, survey data, experimental data, publicly available data online (but needs to be scraped by researchers), privately available data online (but needs to be scraped by researchers with consent of participants), internal data within institutions (i.e., companies, universities etc.). In the following sections, I will analyze the features of these data, how to get such data, things to pay attention to when getting such data, as well as tips on cleaning or pre-processing such data based on my experience applying each type of these data to studying computational social science for two years. I will also briefly summarize the two projects I have done, which have equipped me with a large substantial amount of knowledge about data cleaning and data analysis.