Data in Higher Education Series | Episode 10

Keys to Data-informed Decision Making: Data Readiness, Data Governance and Data Ethics

Published March 12, 2019

Data-informed decision making cannot be a standalone campus project with a start and end date, it should be an ongoing, evolving, strategic priority across an institution. In this podcast, Vice President of Campus Success Annemieke Rice and Vice President of Strategic Initiatives Nicole Melander discuss how institutions can address issues of data readiness, data governance and data ethics when collecting and using data for student success, as well as identify what questions campuses should ask when partnering with external vendors.

Download Podcast

Show notes:

Keys to Data-informed Decision Making: Data Readiness, Data Governance, and Data Ethics

A conversation between Annemieke Rice, Vice President of Campus Success, and Nicole Melander, Vice President of Strategic Initiatives

Campus Labs is hosting its first Analytics & Business Intelligence Institute.

The day-long event will take a campus team through a set of activities and simulations to allow them to think about analytics in a holistic way. Ideally, senior and mid-level leadership across the campus should attend so that participant can explore the data from multiple points of view.

Data readiness:

Data readiness involves both the understanding of what data is available and the business process of what’s happening on campus that allows for the collection and use of that data. It’s about culture change also and getting your colleagues to a place where they are ready to action the data that is now available to them.

The importance of cross-organizational collaboration:

It’s important because it’s the alignment of definitions and understanding what that data really means. With collaboration there is an opportunity to bring forward conversations about definitional consistency and to understand how the same data is being used in different ways in different parts of an institution and how to resolve which definition to use.

The evolution of analytics:

Fifteen years ago, the conversation was about collecting data. As it has evolved, we are looking at historical trends and autopsy data, differences between years and trends and making decisions on those trends. Recent developments in analytics includes both predictive components (for example, identifying students that need help) and behavioral data. This is thinking about the “whole” in analytics.

Building out a holistic and robust predictive analytics data approach:

Having an iterative mindset becomes very important during the process. It’s about what can you do now with the data you have to get you started. At Campus Labs, we have a Student Strengths Inventory (SSI), a noncognitive tool that allows a student to self-evaluate their reaction to a set of questions around their mindset in coming onto campus. It is a lightweight way to collect data and start to connect different cohorts of students with actions that you might take with them. This helps to identify what do you have today, how do you have to enhance that data set and continue to iterate to get stronger.

Why some data goes unused:

It’s about data fluency and comfort with it. There are challenges with transparency and fear of the data because it might make us uncomfortable with the data telling us things we would prefer were not true. Thinking about the culture and the way it is impacted by this sudden high level of transparency that the data brings can be another reason the data doesn’t get actioned.

The ethical use of data:

Our role on campuses is to try to help all the students. The ethics around that often mask that. Ask how that data is being collected, how the algorithms that are being used and providing the visuals and predictives are being created. Are they producing results that are not contextualized? Our role as educators is to think about that data carefully and understand what’s being used to understand the data that’s being given to us and to use the data for good. We should add the human context and our understanding to it, so it’s not about the data telling us what to do. Be “data-informed” so that you’re considering other information that the data does not contain so you can use the data for good.

The role of the data consumer:

That data consumer role should include data fluency through professional training to get comfortable with interpreting data without allowing biases to affect what they think they see in the data. Data consumers should be able to leverage ideas around a continuous improvement cycle where they have the data, they are using and consuming that data, making decisions with that data, and taking action. But that action should result in generation in new data to confirm whether or not that action was the correct action or if they need to refine that action. It is a constantly ongoing and evolving process.

Establishing the right relationship with a software partner:

Avoid thinking about this as a project. Ensure that those from the vendor are thinking about this in the very long term. Ask a lot of very informed questions about how data is being collected, used, secured, processed (internally and externally), and managed by the vendor. Ask what the vendor is going to do to help with professional learning and culture change.