How Can Big Data Be Used in Social Research?

BIG DATA: HOW CAN BIG DATA BE USED IN SOCIAL RESEARCH?
by (Name)
The Name of the Class (Course)
Professor (Tutor)
The Name of the School (University)
The City and State where it is located
The Date
Big Data: How Can Big Data Be Used in Social Research?
1 Introduction
As big data continues to descend on different sectors, ranging from e-commerce to the health industry and governments to non-governmental organizations, information system designers and analysts are facing an array of opportunities that come with the technology as well as challenges (Tian and Liu, 2017). Howe and Elenberg (2020) have defined big data as humongous amounts of data sets that can only be analyzed using sophisticated computing techniques to reveal patterns, associations, and trends. Josh and Ralph (2015) have noted that while the emergence of big data has served as a potential boon for social research, this new area of research also introduces distinct epistemological encounters that sociologists have sought to resolve. Tian and Liu (2017) have particularly identified challenges among information system designers in developing a holistic approach for big data research that the modern analytics savvy generation can adopt. By utilizing evidence-based literature on the existing and emerging applications, Tian and Liu (2017) have assessed the significance of big data analytics in leveraging opportunities from domain-specific analytics and the abundance of data in diverse critical scopes. In practice, peer-to-peer and leader-based social information channels have traditionally influenced consumer buying decisions on common social e-commerce websites. However, combing theoretical models and big data analytics has empirically been shown to produce more effective and robust consumer buying decisions, thus serving as complements (Tian and Liu, 2017). Josh and Ralph (2015) have focused on the use of inductive approaches, the questions on causal versus correlational research, and the application of theoretical frameworks in the age of big data. Drawing from Tian and Liu’s study, this paper explores how big data, one of the powerful tools in the modern analytics age, can be applied in social research while complementing it with the existing theoretical models.
2.0 Methods, Data, and Findings
Tian and Liu (2017) have delved into the question of the superiority of big data and arrived at the conclusion that for social research to be meaningful in producing more effective and sound outcomes, it requires an investigator to combine it with existing theoretical frameworks. This similar problem has been dealt with by Halford and Savage (2017), who have opined that numbers do not speak for themselves but will demand theoretical backing for sociological research to become expressive (p. 1142). In their study, Tian and Liu (2017) have employed BigML in assessing the interaction of leader-based and friend-based social information channels in influencing consumer buying behaviors and outcomes on social e-commerce websites. The authors integrate sociological theoretical models and big data analytics through an empirical study to demonstrate if the two can be merged to produce better results to inform consumer buying decisions. The researchers used one Asian website as a sample, and BigML was used to create machine learning models of the site with 10,097 products belonging to 60 different brands. Of the 60 brands, investigators identified 163,845 customers associated with the brands during the study that took place in November 2012 (Tian and Liu, 2017). In particular, the authors were interested in data about customers’ “following” and “followers” to build an egocentric network for every customer.
Results from the study offer invaluable insights into information systems research and practice and demystify claims that big data alone can be used in providing unmatched accuracy in predicting consumer behaviors (Tian and Liu, 2017). Researchers in the field of sociology, particularly referred to as social scientists, have come up with the concept of symphonic social science concerning the application of the two models in the modern big data era. Symphonic social science explored by modern social researchers has phased out traditional theories that were celebrated before the emergence of big data analytics (Halford and Savage, 2017). Tian and Liu’s (2017) results disprove hyperbolic claims that big data will replace linguistic and sociological theories that have for centuries and millennia studied human behaviors.
3.0 Limitations of the Study concerning Representation and Inference
While Tian and Liu’s (2017) study has provided important insights into understanding the significance of integrating sociological theories and big data analytics in predicting consumer purchasing behaviors, it is not devoid of limitations. The first important limitation is the sample size, where the researchers used one e-commerce website to prove their hypothesis. Sample size in statistical analysis is an important factor when investigating a hypothesis with different factors that could influence the outcomes or observations. By selecting a single popular website, the investigators assumed that one e-commerce website represented all other sites around the world, and therefore, the inference was largely biased. Secondly, picking samples from one region, Asia, would introduce another bias in the results. Statistics aims to infer information about large populations from a relatively small sample because it is impractical to receive measurement for every entity in a population, but selecting a sizable sample would avoid chances of bias (Krautenbacher, Theis and Fuchs, 2017). Third, online behaviors might not be similar to real-life behaviors when customers want to purchase a product or service. The study relied on tracking the “followers” and the “following” trends in the crawled sites to gain insights into con...