15 Sharing Data
Learning Outcome
By the end of the chapter, learners will:
- Understand why it is important to share data
- Describe best practices in data sharing and apply them to their work
Introduction
The sharing of OER data should reflect the needs and purposes of stakeholders – if possible, target sharing to address the specific goals identified at the beginning of the data collection process. Provide data in an accessible and usable format. If the data is to be shared publicly or used for comparison between institutions, include information about the data definitions and methodology used.
The following sections are adapted from “Calculating and Reporting Student Savings” by Jeff Gallant, The OER Starter Kit for Program Managers, CC BY
Principles
Be able to say “it’s at least this much”
Take every consideration into account when calculating and presenting savings data and report conservatively. An example of this is using only the direct reported savings from OER program grantees in calculations instead of a reported potential for more savings if the entire department adopts the resources. If your actual savings are higher than your reported potential savings, that’s great news for all if it’s revealed later on!
Use transparent methods whenever possible
Some of the numbers you have will be directly taken from surveys, reports, and instructor-reported savings estimates, but others (such as applying an average textbook cost savings estimate per student in one course) may have to be taken from external sources. If possible, use external sources who report their own data and estimates openly, and link to those sources within your reports and shared spreadsheets.
Adapt your methods to the needs of your stakeholders
If you notice an emerging critical question that you did not anticipate in your collection strategy, try to adapt your calculation methods. For example, in its first year, Affordable Learning Georgia started tracking annual savings estimates from its instructional teams; when it was clear that administrators needed semester-by-semester data to answer their questions about OER use on campus, we changed our new grantees’ reporting methods to include semester-by-semester estimates and checked in with our past grantees on this as well.
Be as transparent as possible, whenever helpful
While many of your stakeholders will just need a summary report to keep up-to-date on your program, providing all of the important data in its primary form (spreadsheets, a database) and sharing your methodologies will help in “sharing the work” on how you reached the numbers in your report and assist others in answering their more complex questions. Try to avoid sharing extraneous and/or sensitive data: in your tracking sheets, you may have things like grants office contacts, invoice and purchase order numbers, or personal identifiable information (PII) that does not need to be shared.
Balance this transparency with up-front simplicity
Media outlets and administrators do not have exorbitant amounts of time to parse your data, so while having unprocessed or highly-technical data available is a plus, providing data in only a non-usable or jargon-heavy way is like having a layer of opacity on top of your intended transparency. If your own communications office or upper administration wants to know about your numbers, consider giving a “short version” in reports and communications first, and then put the details below that or in an attachment for reference.
Provide a slow-moving metric for the public
Similar to keeping things simple up-front, media outlets and administrators will be confused by a constantly-moving set of impact numbers. For example, if you are adding student savings being reported by instructors on a rolling basis, and then removing student savings immediately when instructors say they have discontinued using OER, newer reports and articles will start looking “wrong” and creating confusion. In this case, you may want to provide consistent numbers annually in reports and keep per-semester numbers in your shared spreadsheets for those who want to dive deeper into the data.
Reach out with good news
You may have wonderful reports that share great news about your program through your website or a newsletter, but sometimes you need that to go further. While much of this chapter focuses on the creation of reports to suit various stakeholders and media outlets, be sure to use all possible channels to market your program’s successes. Provide appropriate visuals: if you’re sharing savings across multiple institutions or departments, use tables; if you’re tracking student savings across a period of time, use line graphs or bar charts; if you’re showing the percentage of your grant teams that saw positive, neutral, or negative changes in outcomes, use pie charts.
Limitations
Like nearly all data collected for reporting, OER data will come with limitations.
Incomplete Data
By far the most common issue with using open textbook data is getting data reported for 100% of courses. In practice, this is quite difficult to do, depending on the data collection methods, how mandatory it is to report, who reports (instructors, chairs, deans, administrative assistants), and how departments account for last-minute instructor assignments to sections and low-enrollment cancellations.
OER Knowledge
If instructors are asked if they adopted an open textbook and haven’t received any training on what “open” means, it stands to reason that this OER adoption data will only be as accurate as each individual instructor’s own knowledge of what it means for something to be open. Even if textbook adoption data at an institution is at 100%, an instructor knowledge gap on OER will render this data inaccurate.
What’s Actually Required or Optional
If syllabi are being analyzed for data, this can be a more accurate, if more time-consuming, way of collecting textbook information; however, what materials are labeled as “required” and “optional” may differ in practice, and there isn’t a way to determine this from just looking at syllabi.
A Hidden Cost of “Free:” Student Privacy
Just because a resource does not come with a dollar amount does not mean it’s open or even zero-cost. For example, some seemingly zero-cost proprietary platforms might require the student to create an account, and the platform’s company may give this student account data or even behavioral data to a third-party company for monetary compensation. The cost to the student, in this case, would be privacy; truly open educational resources, or even those marked as zero-cost, should not require this.