We use the name “Data Analytics” to refer to data analysis (quantitative methods), and their connection to the larger enterprise of data science through other areas such as ethics, code versioning (or, more general, configuration management), solution architecture, reporting, etc. In brief, Data Analytics is the quantitative aspects of data science plus all else that makes that analysis possible.

We have a data science multidisciplinary corpus rooted in tradition. However, that tradition has been upended by the fast development of data availability and AI in the last few years. We went from data paucity to data explosion in a blink (in academic years, that is), and methods have not caught up yet. We have solutions that emerge in niches and tend to propagate organically; this is (many times) effective (but not always) but it is not efficient. The bar to enter the field is too high… or low without foundational knowledge. The latter results in superficial capabilities. Therefore, it might be time to think deeper into what the corpus and the training could look like, and it might not be simply the addition of competencies by discipline. This might sound abstract, but we now have examples. The new science of data is emerging in this landscape as something that surpasses the capabilities of other fields of knowledge established in the past.

The collection of knowledge skills and abilities required to become a competent data analyst is broad and difficult to acquire. It has foundational components related to mathematics, statistics, and computer science. It also has operational components related to information technology, scientific data analysis, and data engineering. There is also a translational dimension related to ethical and legal considerations, as well as interdisciplinary collaborations.

Data analysts employs advanced mathematical, statistical and computational techniques to tackle problems, evaluates alternatives, and executes solutions. They consume extensive data sources to address questions through exploratory data analysis, probabilistic modeling, or predictive analysis. Effectively communicating insights to stakeholders requires statistically rigorous visualizations and narratives. The role on occasion demands algorithm design to unearth data insights. Additionally, quantitative methods and software tools are utilized to construct computer programs, while designing coherent data structures for data access and analytics needs. Ethical and legal considerations are imperative throughout a project’s duration.

There are several several categories in data analytics. Descriptive analytics focuses on shedding light on past events, answering the question of what happened. Diagnostic analytics delves deeper, probing why a particular event occurred. Predictive analytics, as the name suggests, forecasts future events or trends. Prescriptive analytics provides recommendations on actions that should be taken based on the data’s insights.

A competent data analyst never claims to have the true answer; they claim to have a correct answer, that is, a solution and insight generated by following a well-documented and reproducible set of steps. They know that insight and knowledge can change in time as perspectives evolve, new data becomes available, or additional interpretation  informs new routes of analysis. Informed data analysts recognize the pitfalls of introducing one’s values into data analysis or presentation, even when such values are aligned with the mission in which they find themselves in, and they take steps to make such biases explicit.