In the first stage, literature in the subjects of diversity, equity, inclusion, reparative descriptive practices, social justice, and historical empathy was reviewed. From this review, four main categories emerged: Original Content; Locally Supplied Metadata; Historical Empathy & Social Justice; and Curriculum, Research, and Community-Building.

For all these categories, there were 104 statements developed, such as: Based on the content I was assigned to review, I think this digital collection has the potential to “foster a sense of belonging among all individuals/groups when this content was created” and to “portray all individuals/groups from the past in dignifying and respectful ways; or help me “recognize how the past context was different than my own” and “better understand Furman’s past”. A seven-point Likert scale was developed and anchored with notations: 7=strongly agree, 6=agree, 5=somewhat agree, 4=neutral or no evidence, 3=somewhat disagree, 2=disagree, 1=strongly disagree (see Appendix A). The instrument was field-tested with the primary team (n=8) through a norming session to achieve interrater reliability, including library faculty and staff (See Appendix B: Norming Session).

The survey went through several revisions before it was ready for implementation. After rounds of edits from the project leader, and the norming session, the instrument was then modified to its current 40-item version. Following the initial norming activity, 90 participants representing FU students, faculty, and staff members were instructed to indicate their level of agreement to each statement along the 7-point Likert scale, and to provide comments and examples were appropriate.

An electronic version was created with Qualtrics. This version was implemented in one classroom. A modified version was drafted based on what the learned from this initial run. The survey was implemented both in paper version and electronically (via Qualtrics). Responses provided in print were input in Qualtrics (via transcription by team leader). Responses provided through the initial Qualtrics version were moved (via transcription by team leader) to the latest Qualtrics version.