Empirical analysis is an evidence-based approach to the study and interpretation of information. The empirical approach relies on real-world data, metrics and results rather than theories and concepts.
Empiricism is the idea that knowledge is primarily received through experience and attained through the five senses. Empiricism contrasts with rationalism, the idea that knowledge is largely attained the through exploration of concepts, deduction, intuition and revelation.
Empirical analysis is integral to the scientific method and is the usual approach used to study subjects for a probable answer through quantified observations of empirical evidence. Empirical analysis never gives an absolute answer, however, only a most likely answer based on probability.
In IT, empirical analysis is performed in market research, software development, data analytics and project management. In machine learning, empirical data analysis can be used as a data-driven approach that is free from potentially restrictive strong initial assumptions, unlike other probability-driven models.
Empirical research often begins with a question such as: Does talking on a phone impair driving ability? From this initial question, a hypothesis for research can be proposed: Speaking on a cell phone will impair driving. That hypothesis can then be tested by examining primary data gathered by the researcher for that particular study or existing secondary data that has already been gathered by others. For example, empirical data might be gathered from correlating police records or speaking to a representative of the police department as primary research or from examining previously compiled studies as secondary research. From the gathered data, it can be decided if the hypothesis is supported or not and work towards the conclusion.
The term empiricism comes from the Greek word for experience: empeiria.