unstructured data


unstructured data

Data that are not in fixed locations. The term refers to free-form text in business documents and reports, news articles and social media. For example, unstructured data are found in word processing files, PDF files, email messages, Internet forums, blogs, Web pages, Twitter feeds and Facebook pages. It is estimated that 80% of all electronic data are unstructured in contrast to highly organized tables in databases. Contrast with structured data. See free-form database and text mining.

unstructured data

data which has been collected without reference to how it might eventually be coded. Some questions in survey questionnaires (and particularly in pilot questionnaires) yield data of this nature. Respondents are asked to give a direct verbal answer to a question for which there has been no precoding. Much data collected by qualitative researchers is unstructured, and content analysts studying newspapers, and historical sociologists studying old manuscript sources make use of unstructured data, because these sources were not compiled with their needs in mind. See also SCALING, QUALITATIVE RESEARCH, CONTENT ANALYSIS, HISTORICAL SOCIOLOGY. Compare STRUCTURED CODING.