Research

Equipped with the team’s research questions (see Get Started for more information on research questions), bring everyone together to determine research methods and tactics.^[1]

For applied research, as a first step you might decide together to develop a survey instrument. Each co-author could take on a project research question and submit survey questions pertaining to the inquiry.

For a literature review, as a first step you might set up a shared citation platform (such as Zotero) and everyone adds sources to it (or you can use a spreadsheet shared on any cloud storage platform to do the same thing). Be mindful of where you source the information the project will cite. If you link to a lot of paywalled resources, your research may not be very useful to a wide audience. The group may commit to selecting only open source resources or to favor open source resources and separate paywalled citations into a subsection of the bibliography.

Before the group starts writing, determine a citation style.

Co-Developing Survey Instruments

Groups need substantial time to collaboratively workshop survey questions. Expect several meetings to get from an initial brainstorm of questions to the wording that will actually retrieve the information you are looking for.

Interviews may be less time-consuming to set up than a survey, but require more than one person supporting them to make effective and useful research. Interviews should be conducted in pairs (or more) with a strong notetaker. Prepare a thoughtful system for taking notes and highlighting particularly useful information or quotations that come up. You can record the interview as a backup or to verify your notes, but rewatching recordings or relying on AI transcripts can take more analysis and time than good notes from a person who has context about the project.

Analyzing Data (Together?)

Analytical tasks are very difficult to complete as a group. No matter how committed everyone is to creating a shared tag library for qualitative analysis, we cannot overcome the reality that our brains interpret broad categories differently.^[2] If you are doing qualitative analysis with a group, leave plenty of sessions for defining and understanding sentiments and groupings you will use. Dividing and conquering the analysis is often the easy part. Bringing together analysis from several different brains is the bigger challenge!

Accessibility

Consider user accessibility when choosing and designing research instruments and methods. Look at platform-determined accessibility attributes such as screen-readability when choosing the platform. Consider user accessibility measures you can take that are not platform-dependent. For example, make survey instruments accessible by deploying plain language for questions and short, direct statements for descriptive context. If the project requires a consent agreement, it should be as short and simple as possible, and can link to longer legalese in a different document if required.

Consider accessibility needs of team members when designing data analysis methods and workflows.

Protecting subjects with IRB

An Institutional Review Board (IRB) is a research ethics committee that reviews the methods proposed for research involving human subjects, to ensure that the projects are ethical. If you work for an academic institution that usually requires approval from the IRB for research, you should review your project to determine if it requires review. Many humanities projects do not, but humanities projects don’t get an automatic exemption. Your institution’s IRB can assist you in administering a review or determining if a review is required.

The Born-Digital Access Working Group brought archivists from several institutions together to research user experiences with born-digital archives. They knew that their project would not likely require IRB, but they still needed to apply for an exemption to follow their university policies. In a collaborative project, one institution can serve as the IRB institution of record. The group chose Yale University as the institution of record. The group chose Harvard University to host the data. The primary concern with an IRB exemption is whether the project involves human subjects research. If you are collecting data from people (human subjects), but no data in the survey is “identifiable,” then the project may be exempt from IRB because it is determined not to involve human subjects research. At that time, Yale confirmed that they did not have to pursue IRB approval for the project because the survey did not include identifiable data. While the group ultimately looked to Yale as the institution of record, several other group members also checked with their institutions to test alignment and got similar responses from their IRBs.

Protecting subjects with a data management plan

It is important to protect research data whether or not you are required to obtain IRB approval. See the Data Management section for tips on creating a data management plan.