8 Ways Universities are Making an Impact with Data (Shawn Pfaff, Tableau Software)
Universities and colleges strive to grow and fulfill their mission of educating their communities. Communicating the data around that mission—how many students are graduating? What does the student population look like? Is the University managing its finances?— is an important component of any institution’s daily life. We present eight ways that higher education is using analytics and data visualization.
Adapting strategies in a dynamic environment (Tina Smidt & Ryan Fellers, University of North Texas)
This presentation will demonstrate how the University of North Texas utilizes multiple strategies to implement a new data warehouse in a dynamic environment. All higher education organizations change regularly, but once the project becomes sponsored, it is critical to deliver within the scope of the charter. These strategies for implementation can be influenced by several factors including governmental guidelines, organizational culture, stakeholder selection, and competing strategic goals.
Agile BI: Increasing adoption & delivering value, 10 days at a time (Jelena Rolijevic & Margaret Roldan, George Washington University)
In 2013, the GWU BI team had two firsts: pulling data from its financial ERP into the warehouse for the first time, and using an Agile Scrum methodology. The objective was to build an interactive dashboard for Principal Investigators to help them manage research grants more efficiently. This presentation will focus on how Agile helped the BI team deliver a solution that increased user adoption through constant engagement and timely feedback from end users. Lessons learned will be shared as well as the future of Agile at GWU.
Analytics — Really? (Suneetha Vaitheswaran, University of Chicago)
It’s all the buzz, but what is it – really? Isn’t analytics what we’ve been doing for years with DW/BI? Isn’t it the domain of IR? What are my leaders hearing that creates so much pressure for this? Where is the magic of “Analytics”? Most professionals in the information management space are fast encountering these questions. At the University of Chicago, we are working to bridge a DW/BI-based reporting strategy into an Analytic-focused one – impacting platforms, staff skills, and culture!
Analytics to support performance based funding initiatives (Natalie Kellner, New Mexico State University)
As higher education institutions are increasingly asked to demonstrate value and forced to compete more effectively for funding, analytics provide the basis for case-building. Whether it be a state legislature, governing board, accrediting body, or the campus community, the information and tools now at our disposal provide an excellent environment for responding to external entities, as well as for driving internal stakeholders to performance targets critical to our institution’s future success.
Ask a Guru
An opportunity to ask those who have gone before for advice, insight and BI/DW predictions. Panelists: Aaron Walz, Anja Canfield-Budde, Hank Childers, Beth Ladd, Ted Bross, Suneetha Vaitheswaran
Big Data: What does it mean to you? (Greg McGinn, Technology Dynamics)
This session will explore the meaning of Big Data and how it fits into the Higher Education Analytics Journey. Participants will gain an understanding of key factors in the market place that are driving IBM’s Big Data Platform and the need for this type of solution. Participants will leave the session with a greater sense of the solutions that can be provided by the IBM Platform through Use Cases from successful implementations at other institutions.
BINGO (Christina Rouse, Incisive Analytics)
This interactive session covers 6 hot BI topics in 10 minutes each. The audience will engage in conversation about creating a BI needs and vision statement, understanding the star schema design, calculating and addressing data quality in BI, discovering measures and how to rationalize each one and the knowledge and skill set required by BI professionals.
Birds of a Feather Gatherings:
- Faculty Activity – teaching & research approaches and solutions
- Student Enrollment – approaches and solutions
Case Study: Collaboratively delivering a Student retention & success model with Agile style (Phyllis Wykoff, Miami (OH) University)
This case study focuses on:
- Miami’s strategic need to increase the six year graduation rate to 85% by 2020. To accomplish this, the first year retention rate must be raised to 94% beginning with students that enter Miami University in the fall of 2014.
- The building of a trusting collaborative team of users and developers to design, build and deploy the solution
- The process of deploying an agile development methodology while this project was underway.
A case study update: BI at University of Arizona (Hank Childers, University of Arizona)
Starting in 2008 the University set out to replace all its legacy systems and build a new BI capability. This provided a great opportunity to re-conceive its approach to BI and to reporting in general in a compressed period of time. This presentation will cover most aspects of BI implementation including budgeting, team composition, methodology, sample dashboards, problems encountered, lessons learned, and emerging issues. It will be an updated version of last year’s presentation, noting that not all things have gone according to plan!
Closing the gap between IT & business; building & delivering BI solutions to enhance user experience (Bart Pietrzak & Pieter Visser, University of Washington)
Success of EDW/BI initiative depends heavily on two factors: how far IT is from the Business and the maturity of the user community. In the presentation, we will use two successful projects that were fully delivered in the period of ten months to address the premise of closing the gap between IT and business and responding to growing user community. Specifically, we will illustrate how a team of four was able to build a BI Portal and 23 institutional dashboards using Tableau to enhance user experience to view institutional data.
Complement your current data environment by creating interactive, dynamic dashboards in hours (that Execs will actually use) (Brian Stevens, iDashboards)
The presentation will focus on best practices as to how institutions of all types and sizes throughout the country are using iDashboards from a decision support, performance management, transparency, and peer benchmarking perspective. Actual examples of iDashboards interactive customer website dashboards, as well as example dashboards that institutions are using from more of an internal decision support perspective to most effectively allocate resources will be shown.
Computer aided mapping of textual examination regulations into HEDW applications (Katharina Kroeger, Bodo Rieger & Sonja Schulze, Osnabruck University)
This presentation will discuss a concept to realize benefits for HE data warehousing and related applications by information extracted and mapped from textual resources, e.g. examination regulations. Text mining, ontologies and conceptual modeling are to be applied in a multi-layered architecture to comprehensively support HE applications, ranging from the configuration of exam administration systems and data quality management of DW, up to model-driven decision support for student lifecycle.
Dashboards for Student Success (Sri Sitharaman & Bob Diveley, Columbus State University)
The presentation details the BI journey of Columbus State University. The whole concept of BI at CSU was in the infancy stage when the presenter attended the HEDW conference in 2011 for the first time with support from Cathy Lester Attendance Grant. CSU then partnered with Oracle and developed student retention, progression and graduation, financial aid, and financial metrics dashboards. CSU is now in the process of using Endeca to capture unstructured data from surveys, and social media.
Data Discovery: Big Data for Student Success featuring Valdosta State University (Brian Haugabrook, Valdosta State University)
Data Discovery allows organizations to discover patterns and join data from any source including social media and surveys. Analyzing unstructured data and incorporating sentiment analysis can unlock hidden secrets on your campus. Brian Haugabrook, the interim CIO at Valdosta State University will present how leveraging non-traditional BI technology can transform your organization’s decision-making process, including:
- STUDENT & COURSE STATISTICS: with at-risk indicators and actionable intelligence.
- DASHBOARDS: combine quantitative, qualitative, and sentiment analysis in one dashboard.
- SELF-SERVICE: no more waiting weeks or months to get data, create analysis in less than 5 minutes.
- ACCELERATING OUTCOME ANALYSIS: get to the root of personalized learning models and improve outcomes.
- UNLOCKING INSIGHT FROM ANY DATA SOURCE: immediate access to both structured and non-structured data, in a zero training user interface.
Data Governance Technology — A game changer (Ronald Layne, George Washington University)
This presentation will focus on how technology is a game changer helping to mature the Data Governance process. Topics will include:
- Acquiring the technology
- Developing requirements
- RFI and RFP preparation and evaluation
- Conducting a proof of concept
- Installation and Pilot phase
- Going live
- Features
- Benefits
- Meta Data Repository
- Business Glossary
- Report Catalog
- Approval Workflows
- Policies, Standards and Process
- Data Lineage
- Data Issue Tracking
- Metrics
- Lessons learned
Data Warehouse automation – case study (Douglas Barrett, WhereScape)
WVNet were able to build their data warehouse with two developers in less than 3 months using WhereScape RED – a data warehouse automation tool. WVNet are the shared IT Services organization of 22 colleges in West Virginia. WVNet built a BI solution consisting of a data warehouse sourced from a Banner student system to support their P20 initiative. Data warehouse automation is a new category of tool that provides speed and agility by multiplying the productivity of your team.
Data warehouse QA in an agile environment — a practical approach (Bomani Siwatu, University of Washington)
Organizations find benefits updating their data warehouses by integrating user feedback into an iterative development process. This presentation will discuss leveraging our Quality Assurance program within our agile DW Dev Life Cycle (incl. fast prototyping) to:
- Verify ETL meets spec
- Assure fitness by end-users
- Provide automated, reusable, repeatable tests
- Rapid test development & execution – un-bottlenecking QA (90% of QA off critical path)
- Understand what has and has not been tested
A deep dive into portal reporting at Iowa State (Lynn Miller & Kevin DeRoos, Iowa State University)
Attend this live demo session and learn how Iowa State University (ISU) designed, implemented, and deployed an institutional reporting tool. Attendees will learn deployment best practices, see how ISU standardized its report development to increase user acceptance, and view the flexible report delivery portals that allow individuals to run a number of reports from a single interface driven by a comprehensive set of parameters. This session will also review the lessons learned at ISU.
Wednesday 4/30/2014 9:15am, William Penn Suite
Deliver business values via MDM efforts in enterprise analytics: a multi-factor analysis of our practices with two master data scenarios (BK Chen, University of Washington)
MDM for Enterprise Analytics is explored in this presentation to see the factors that impact the success of a MDM effort, using two MDM scenarios from UW. Seven factors analyzed are:
- MDM as a culture change
- “The master(s)” of the master data
- The forces behind the MDM change
- “Agile” or “waterfall”
- “Make it happen first”
- Design sophistication for complexity and changing environments
- Effective training for a broad acceptance.
Effective practices for data governance & lessons learned during deployment (August Freda, University of Notre Dame)
When Notre Dame launched a Data Driven Decision-Making (“D3M”) initiative well over a year ago, it was clear that a solid foundation of data governance and stewardship needed to be established. Over that time, we have established and continually refined an approach and practices to create a robust data dictionary upon which analytic tools can be built and deployed. With 200+ terms defined and another 200 in process, this presentation will outline these practices and lessons learned.
Evolving beyond BI (Rainbow Di Benedetto & Louise Nelson, University of Texas at Austin)
IQ began narrowly scoped to deliver summary-level answers to executives via cubes and reports. But campus clamored for access to the valuable transactional data in IQ. We will talk about how and why IQ has been opened up to developers coding in Python and data analysts using stats packages, while still supporting use of traditional BI tools, addressing the challenges and pain points of this transition, as well as the successes it has resulted in and the long-term benefits it will yield.
From Months to Minutes — Data Warehouses at Iowa State, Loyola and the Maine System (Yiorgos Marathias and Robert Silcher, Phytorion; Lynn Miller, Iowa State University)
Iowa State University needed to deliver enterprise Student, Staff and Financial information to its many users but its Student and HR data lived in Legacy systems, and its Finance data was about to be converted to Kuali Finance and Coeus. The IR department needed months of analysis before it could publish the Student Credit Hours numbers.
Loyola University Chicago had data that was manually maintained, siloed, lacked history and yielded inconsistent results. Its Faculty Instructional Activity that sourced five systems took weeks to complete, and the Financial Aid Discount Rate could only be done on an occasional basis. A data warehouse was on the agenda for many years.
The University of Maine System, made up of the seven public universities in the State, had a pressing need to make better use of data to determine recruitment, financial aid and retention strategies.
These three different institutions required different approaches ranging from a mostly vanilla Data Warehouse implementation to full custom work. In the end, Student Credit Hours went from months to minutes, Loyola’ s Faculty Instructional Activity not only produces the data for IPEDS, the Delaware Study and various surveys, but it can be delivered on demand and so can the Discount Rate; and Maine is able to improve recruitment and retention at the strategic level.
In this session, we will focus on key functionality from different modules, we will review the tailored implementation approaches each school followed and discuss how the partnership overcame challenges.
Getting on the same page: Using the data cookbook to collaborate & communicate definitions (Elizabeth Wallace & Leslie Sine, American Public University System)
To improve consistency and data literacy, APUS needed a common data dictionary that was accessible and clear to people at all levels of the university. IR and IT joined forces to create functional and technical definitions using the Data Cookbook. Since it comes with terms from IPEDs, PAR and other schools, we can map our functional definitions to those used by others. This session will describe implementation, data governance issues and plans for cross-institutional collaboration.
Going Agile at University of Toronto (Christine Beckman & Alexandra Agostino, University of Toronto)
The UofT BI team began using Agile PM methodology in the spring of 2013. Unsure whether this would work well for our BI projects, we decided to embrace the approach and experiment with it. Now into our second agile-run project, this session would look at:
- Defining user stories for BI
- Agile data modeling while planning for future projects
- Experience of stakeholders and business leads
- Shifting the BI team culture to self-organizing
- Managing the projects using JIRA and Greenhopper
How did Berkeley’s curriculum project win Oracle’s highest award? You can use the same process (Greg Hamilton, University of California, Berkeley)
UC Berkeley’s Student Curriculum project won Oracle’s highest Excellence Award. Come listen to how UC Berkeley utilized Agile Development, Prototyping and off-shore resources to reduce development time by 25%, fully engage users and deliver a product that was cheered by faculty. This is NOT an Oracle specific approach.
Is the data vault methodology the best way to structure your university data? (Ed Razon & Eric Elkins, University of Washington)
Data Vault is not a new concept, many papers have been written and books published about the Data Vault technique, but none has mentioned of its application in Higher Education data. This presentation will cover the basics of data vault modeling as it pertains to student data. This presentation will demonstrate why Data Vault is the most appropriate data model for the Integration Layer. We will show how Data Vault solves the most common problems in data warehouse design.
Journey in the development of a hybrid BI reporting system (Michael Hansen & Jeff Merth, Oregon State University)
This presentation will inform the audience about the journey Oregon State University took to develop a hybrid BI solution, moving away from a typical single-vendor / vendor-dependent solution to an integrated solution that leverages current site license agreements, uses open-source software when possible, is easily adapted to ongoing needs, and allows for the use of best-of-breed solutions. OSU’ s BI reporting solution is more than a reporting system.
“Lead Faculty”; Faculty supported QA in the online classroom (Kimberly Ford, Walden University)
This session provides a model for faculty supported quality assurance in the online classroom. Faculty peers are trained to mentor instructional faculty in key issues related to the course, provide peer reviews of the classroom, and support ongoing improvement to course content and format. Session learning outcomes are to provide an overview of the “Lead Faculty Model,” including challenges during development and implementation, ongoing refinement, and demonstrated outcomes.
Mizzou Data Source: A collaborative effort between institutional research and IT (Kathy Felts & John Harding, University of Missouri)
It is important to understand and define the roles of Institutional Research (IR) and Information Technology (IT) in data management, data governance, and decision support efforts. This presentation will provide an example of the initial stages of developing a data warehouse with governed data for the purpose of providing institutional information through a collaborative effort between IR and IT with the intent of having a conversation about the roles of IR and IT in other similar efforts.
The most complicated security they have ever seen: modeling metadata for maintaining fine grained row-level security (William Brickman, Harvard University)
The Harvard Data Warehouse team has implemented many distinct types of very complex row-level security for its reporting environments: for GL, HR, Grants, and others. Maintenance and metadata reporting needs drove us to create a custom abstract MDM model for row-filtering security that worked across all. This talk will review the history and explain the models built, and how they increased efficiency of security data builds and user metadata reviews of the row filtering rules.
Moving from Reports to Business Intelligence and Advanced Analytics: Case Studies and Lessons Learned (Kirby Lunger, Performance Architects)
Learn about other institutions’ ongoing processes to transform their reporting and analysis processes and environments. Discover what drove decision-making processes, the effort that goes into these kinds of initiatives, and the value that institutions are gaining from these efforts. Discussion topics will also include the different types of business intelligence / analytics capabilities currently available and our recommendations for how you can institute a similar initiative in your institution.
A natural alignment: IR and BI support student success (Christina Drum & Mike Ellison, University of Nevada, Las Vegas)
We will address our strategy for delivering retention, progression and completion (RPC) information in support of UNLV’s latest RPC initiative, including the creation of a data mart to enable longitudinal reporting and analysis of student cohorts. We will speak to architecture and processes, and share RPC benchmark dashboards. We will also describe a new dashboard presenting student risk factors for attrition, initially based on descriptive analysis and later refined with predictive modeling.
Networking Bingo: Bring a stack of business cards and a sense of adventure to network with your peers (two session)
Performance measurement through agile use of BI/Dashboards (Ted Curran, Carnegie Mellon University, Tepper School of Business)
This presentation will focus on performance management and aligning metrics to key strategic plans through Business Intelligence/Dashboards. Carnegie Mellon and the Tepper School of Business have implemented Tableau, which has become the data hub for reporting and analyzing key information across more than 15 best of breed systems. The session will cover strategies for those creating dashboards versus those consuming the information along with partner strategies for business and IT resources.
Roundtable discussion: Role of BI in medical school cross-mission reporting (Vicki Croddy, Indiana University, School of Medicine)
Medical schools have a unique challenge of balancing the education of future doctors, performing research to advance medicine, and providing healthcare to communities. With this challenge comes another secondary challenge of how a medical school can determine whether it is meeting its’ objectives of the three missions; education, research, and patient care. This roundtable discussion would provide a forum for discussion on the use of BI to meet this challenge.
Self Service, Agile & Metadata: A roadmap to the implementation of the MS BI toolset (Kristin Kennedy, Arizona State)
This session will talk about how Arizona State University took the opportunity of bringing in a new toolset to improve the great things they were already doing. This session will focus on how ASU did this using an Agile Approach to bring about BI Self Service, along with a sustainable model of metadata that is flexible and user friendly. This session will focus on the Microsoft Toolset and how it fits into a Self Service, Agile, and Metadata BI world of today.
Taking the Drexel DW to another level with IR (Insiyah Jamal, Drexel University)
IR is partnering with IT to build a DWH to be used by all departments at the university. This project involves migration of prior frozen census data, validation/data clean-up & decisions regarding definitions for various measures and indicators. The data model is also being extended to include items to support the university Fact Book. Key members of the project team also include Academic Information & Systems, who provide end-user reports and support/oversee the Student ERP, Banner, and, related census data process, being merged into the central DWH. Meanwhile new initiatives lend color & structure to the shifting project landscape: enter the DWH Steering Committee, and the rollout of a formal PM process.
Understanding Data Warehousing: from architecture to design to rollout (Kevin Drury, Velaris)
Using the past 18 years as experience, Velaris has worked with many Higher Ed Institutions to create and renovate their BI Dashboards/Reporting as well as create trend-spotting Analytics. In this presentation, Velaris will break down the components of different Reporting and Analytics alternatives for those who desire to derive an informed path to deliver Higher Ed Dashboards and Analytics.
The presentation will review common problems and identify pitfalls with building a Business Intelligence solution and how to avoid them. We will discuss “what are they really trying to build” and examine costs, type of staff needed, architecture, design, programming and delivery.
The audience should walk away with an understanding of what to expect if they are trying to build a data warehouse and/or a reporting system. This would include realistic timeframes, cost and technology.
University Performance Management (Darren Catalano, University of Maryland)
Learn how UMUC is making data driven decisions in areas ranging from enrollment management to student success. Technology and tools are not the drivers of great business intelligence programs. At UMUC, we are building a rigorous performance management framework to improve collaboration among teams and measure results. By establishing regular meeting cadences with key stakeholders, the business intelligence team facilitates conversations using data to answer questions and take proactive actions.
Unleash the power of your meta-data; graph databases (Chris Frederick, University of Notre Dame)
Many applications handle data that is deeply associative, i.e. structured as graphs. Meta-data management systems are a text book example of this as they deal with strongly linked data. These types of recursive data structures in a traditional RDBMS are difficult to deal with as in essence, each traversal along a link in a graph is a join, and joins are known to be very expensive. This presentation will demonstrate the Neo4j powered graph database behind ND’s meta-data driven BI Portal.
The value of data documentation when implementing a data warehouse (Allison Weingarten & Jacob Campbell, The New School)
This session discusses The New School’s experience with creating and maintaining data dictionary and report definitions concurrently with the implementation of our data warehouse. It will cover why documenting data and reports is essential to this process, the tool we chose to document our data and manage our report request queue and how we’re using it, some of the challenges we’ve faced and how we’ve overcome them, tips for collaborating with other departments to define data they use, and the benefits we have gained so far from the process. We will also provide a demonstration of the integration between our documentation tool (iData’s Data Cookbook) and our reporting tool (Evisions’ Argos).
Visualizing data using mapping tools (Laura Diefenderfer, Eastern University)
Mapping tools create a new way for institutions to visualize data. Both web-based solutions and Microsoft Excel will be used to show how institutions can map data to see trends. Examples of data that has been mapped include locations of competitors for benchmarking; internal student data to show density of students; and external data such as census data for projecting growth opportunities. Attendees will leave with tools that will enhance their ability to summarize, analyze and present data.
A Walk-through of Academic Analytics’ Data Model (Mike Rohlinger, Academic Analytics)
Academic Analytics maintains a data model based on research scholarship for research universities in the US and UK. We will walk through this university data model during our presentation. The model will display our use of person identifiers, such as (e.g. ORCID ID). The scholarship data includes: conference proceedings, articles, grants, awards, books, and book chapters. We will also look at the Academic Analytics tools that take unstructured scholarship data as input, and output structured data with a primary key on doi and person identifier. We will demonstrate how your stakeholders may use the data to compare themselves to other faculty, departments, or universities.
Warehouse Analytics — What do you know about your warehouse? (Kevin Joseph, University of Maryland)
Our warehouses provide insight into the workings of our institutions, but how much do we know about the workings of the warehouse itself? This presentation will explore how UMBC has developed reports to help us manage and understand how the warehouse is used at the user level as well as explore how well the ETL process functions.
Why is Data Management so important? (Scott Flory, iData Incorporated)
If your school continues to struggle with institutional reporting, then come join our community. IData is leading a growing band of schools towards better data management practices. The time is now to change how you approach institutional reporting. Do not let that reporting project start without establishing some new processes. Do not roll-out that new reporting tool without having a way to train people on the data. These and other tips and techniques are discussed in this presentation.