Secondary data: Additional considerations

Secondary data defined

In general, secondary data are data that were collected by someone else for some other purpose. Secondary data can be data collected for research or non-research purposes. Examples of secondary data derived from someone else's research include oral histories, national survey data, surveys probing for consumer, health, environmental sentiment, etc. Examples of secondary data derived from non-research purposes include court records, testimonies, trial transcripts, social media posts, etc. Using secondary data can help us advance knowledge by using existing data to answer new research questions. Source: Encyclopedia of Research Design

Advantages of secondary data use in the IG context

• Secondary data offers access to a variety of publicly available informal and novel data sources, like social media, speeches, oral histories, official transcripts, court documents, records of hearings, testimonies, etc., thereby offering different insights into governance dilemmas without the need to obtain Institutional Review Board approval.
• Secondary data are already collected and processed, thus saving time and money, and allowing researchers to focus their efforts on framing questions and conducting analyses.
• Broad scope of available data facilitating analysis of longitudinal studies collected over decades
• Depending on the source, secondary data can be of high quality, such as oral histories and surveys conducted by government agencies or research centers with well-documented research methodologies

Challenges of secondary data use in the IG context

• Secondary data that are not research question-driven, e.g., social media, require adjustments and assumptions before IG-coding can proceed
• Research-driven secondary data sources were generated to answer specific research questions arising out of the context of the originating research studies, like oral histories conducted to document the experiences of influential people, events, or organizations during certain time periods.

These constraints limit secondary data’s applicability to other research designs and can lead to known (missing institutional statements and IG syntactic elements) and unknown (obscure study motivations and research design/questions) information gaps that need to be considered when coding and interpreting the data.

Data saturation

As with primary data, researchers can use a saturation threshold to assess whether the inclusion of additional secondary sources continues to reveal new institutional statements, grammatical elements, actors, or contextual conditions relevant to the research question. (see Primary data for more details)

An alternative approach is to combine different data sources to leverage their strengths and mitigate their limitations. For example, primary data sources can be used to validate institutional statements found in secondary data and provide complementary information about their implementation and underlying motivations.