Getting Started with Educational Intelligence: Typical Impediments and a Suggested Approach
If you’ve just started exploring, planning or implementing an initial analytics project, you are not alone! Leveraging analytics to provide Educational Intelligence (EI) insights across the student life cycle is very much in it’s early yet exciting and rapidly evolving stages. There has certainly been an abundance of positive hype and discussion amidst the education community at conferences, in social media, articles and blogs, and a growing number of institutions about analytics in education. In addition, the education focused analytics solution providers and products, standards, and data analytics technology in general have emerged within the edtech community to supply some concrete initial analytics capabilities to build upon.
With all of this excitement and early stage activity, it is easy to feel overwhelmed, and possibly already in the back of the analytics pack in terms of getting started. The reality is, actual implementations of EI analytics have been slow to get started and still very much in its infancy. However, there is great potential and value to be realized.
So What Can Slow Down Getting Started?
Over the past few years, there has been an awful lot discussion and churn on relevant topics related to getting started with analytics in education. This churn has impacted the pace of getting analytics initiatives underway. Some of the more common topics include:
- Wrestling with the challenges of collecting and unifying data collection from multiple learning platforms and applications
- Privacy and security
- Institutional governance over and leverage of student data
- What and how much data is really needed to be meaningful
- What insights would be most useful and valuable for the key stakeholders to improve teaching and learning
- Advanced algorithms that potentially have a “black box’ effect or bias when impacting the learning experience
All of these topics have merit, but are not insurmountable and certainly should not inhibit moving forward in order to reap the positive impact that EI can offer as noted in our prior posts. Many of these topics are being addressed as educational analytics continues to rapidly evolve and mature. For example:
- Emerging data interoperability standards and integration APIs such as IMS Caliper and xAPI are making the collection of data across multiple learning applications more consistent
- Institutions, government and individual solution providers are revising and defining governing privacy and security policies to specify data access and usage guidelines
- Learning platforms, applications and solutions are increasingly enabling strict role based features and access to analytics, working closely with the stakeholders to tailor the EI insights specifically needed for individual roles
- Emerging advanced analytics solutions can facilitate more transparent, modular and plug-n-play algorithmic driven EI to mitigate the bias and “black box” effect
So How Best to Get Started with EI?
Define a Roadmap
Many institutions have struggled with how best to define, prioritize and approach their analytics roadmap and implementation plan. Some institutions have prioritized pursuing analytics to gain a better understanding of the student life cycle data as it relates to enhancing the marketing, recruitment and enrollment of students. Others have prioritized analytics to help optimize the retention and success of their student’s completing their degree based programs. Those with primarily academic focused priorities have prioritized learning analytics in order to gain deeper insights into the teaching and learning academic experience as student’s traverse through their coursework. Academic focused EI fosters more informed real-time interventions with students based on their progress, as well as to inform any adjustments needed to optimize the curriculum itself, to ensure improved outcomes.
To be most effective, first and foremost establish the priority and sequence of which student life cycle scenarios will be included on your overall roadmap and a well defined scope for each. This scope should include what specific stakeholder EI insights and value will be delivered, what institutional process or governance may need adjustments to proceed, what platform and application data will be required, and the implementation requirements as it relates to in-house and/or any 3rd party resources needed.
Adopt a Staged Implementation Approach
To proceed more effectively, institutions should seriously consider and adopt a more staged, incremental approach to implement their analytics roadmap. Each stage should be focused on delivering EI insights within a specific student life cycle scenario (i.e. instructional, marketing, retention etc) AND most importantly to a well defined set of stakeholders. It is important to get stakeholder requirements and input factored into the specific EI insight dashboards and other intelligence driven functionality planned. This approach can help mitigate the time and complexity of implementing what might be a very comprehensive roadmap. By breaking the implementation into smaller, more manageable stages, each implementation milestone will very likely have a lower risk threshold and a higher probability for meeting or exceeding requirements. In addition, this approach fosters iteratively learning from prior experience and refining each subsequent stage of implementation continuously improve on each milestone as well as achieve the best overall solution..
Establish an Analytics Foundation
Understanding what analytics technologies, products and solutions are available, and how they can factor into the institution’s analytics implementation, has been a challenge. Some basic analytics capabilities have been added as features to the LMS and other learning applications to improve upon descriptive reporting or personalization. However, these are specific to each individual application and although they do provide some incremental value, they should not be mistaken as an analytics foundation. They fall far short in meeting the requirements for, and full value derived from, a more overarching and comprehensive analytics enabling baseline.
Unifying all of the invaluable student generated academic and operational data for the institution to own, access and benefit from is a foundational prerequisite to any EI analytics implementation planned. The reality is most of this data is silo’ed within in each platform and application for its own use and specific reporting or analytics capabilities. Therefore this data is basically unusable for the institution’s overarching analytics initiatives and benefit. This shortcoming is resolved by deploying a highly scalable data management and analytics enabling service as a foundational “platform” integrated with all learning to collect, aggregate, store, correlate, analyze and derive EI insights via stakeholder dashboards and more.
Leveraging this platform for each stage of your implementation will allow you to more effectively manage the phased delivery of a range of capabilities from the most basic to more advanced EI insights. This is because the platform enables incremental reuse and expansion of the data collection, storage and analysis from existing or new learning platforms and applications as needed to implement added analytics capabilities as needed. This can take the form of new or revised EI dashboards to meet the next set of specific stakeholder requirements on your roadmap, or more advanced EI insights that require more predictive or prescriptive analytics. Basically, the platform enables the additional data needed as inputs to the modular computational processing needed for the more advanced analysis to deliver more advanced EI.
Now that we’ve defined EI and discussed an approach to more effectively getting started, future posts will isolate and cover some of the key elements of EI to consider and fully understand in more detail before factoring them into your EI roadmap and implementation.