At its simplest, a primary source is an account of something that happened by the people who were there, such as participants or witnesses. A secondary source is an account of something that happened by people who were NOT there, often framed as a review, summary, or analysis. Technically, a secondary source is a review or analysis of primary sources—there’s also what’s called a tertiary source, that analyzes secondary sources, and so on.
The most common type of primary source used at Illinois Tech is the research paper written by the researcher(s) who actually carried out the work. These papers are typically published as articles in peer-reviewed journals but could also be in the form of a thesis or dissertation, research report, case study, clinical trial, etc. In addition to written reports, various ancillary materials can be primary sources. These include data, surveys, questionnaires, interviews, computer code, images, and other supporting materials that were generated or collected as part of the work.
Secondary sources may be published in peer-reviewed journals as well but most often occur in popular media, like websites, blogs, newspapers, etc. Secondary sources in peer-reviewed journals are easy to identify because they use the word “review” in the title or abstract and don’t present any new research. Also considered as secondary sources are any ancillary materials that were re-used or repurposed from other research.
Interestingly, primary source research papers almost always include a review of prior research as part of the introduction or as a “literature review” section. The primary source material only includes those parts that talk about the new research: the methodology, results, discussion of results, conclusions, or other similar sections.
Simply put, people make mistakes. There’s an old party game called Telephone where a phrase is whispered from one person to the next around the room and at the end of the game, everybody is amused at how the phrase or its meaning has changed. Using secondary, tertiary, or other sources is like playing Telephone with your research. Reviews and other secondary accounts are summaries, so even at their best they omit parts of the original research and lack the detail and nuance of the original paper. At worst, a review author could entirely misunderstand or misrepresent the original research.
No, not at all. For older, well-established research that’s had ample time to be reviewed and consolidated into the general knowledge of the field, there’s no need to go back to primary source material unless you’re challenging the conventional interpretation.