Below is the alphabetical listing of the member presentations that were given at the 2018 HEDW Conference in Corvallis, Oregon.
Advancing Data Governance at Purdue by Partnering with Academic Resources
Sarah Bauer & Kendal Kosta-Mikel, Purdue University
We need data governance to define data governance! It has many meanings and implementation paths. It continues to be a hot 
topic across all industries. In higher education, it’s a priority to ensure that our critical asset, data, is available to make impactful, far-reaching decisions. This update on Purdue’s data governance journey includes the use of Data Cookbook. The differentiation of data governance from business intelligence is underway with an increased focus on data quality. University collaboration re: standard definitions, reports and processes, from central and academic areas will determine success. Speakers represent a central office and an academic college.
Alexa Goes to College: Arizona’s Innovative Use of Voice Technology
John Rome, Arizona State University
Arizona State University (ASU) is re-imagining data access and higher education with voice user interfaces (VUIs) based on Amazon’s Alexa Voice Service technology. This presentation will discuss and showcase the work ASU has done with the Amazon Echo, including initiatives that integrate voice technology into the everyday life of faculty, staff and students (including in the curriculum.) With over 30 million “smart speakers” sold in 2017, and estimates that over 50% of households with have a smart speaker in their home by 2020, it’s inevitable that this form factor will become more prevalent in the work of business intelligence over time.
An Agile BI Triptych: How Agile Works for Three Universities
(A presentation and panel discussion)
Joanne Wilhelm – Indiana University, Robert McDade – University of Washington & Kristin Kennedy – Arizona State University
We all want to provide high-value BI products to our clients with fast delivery and short development cycles, accurately meeting the information needs of our universities with high-performing teams. So how do we organize our work to achieve this challenging goal? Join us in a presentation and spirited panel discussion with colleagues who employ Agile BI methods for this purpose. Panelists will address a range of Agile methods and architectures, from Scrum to innovative agile-like hybrids supported by equally diverse technical architectures. We will highlight what’s working and what’s not, using practical examples.
Building an Enterprise Data Environment in the Cloud
Steve Fischer & Nate Polek, The Ohio State University
The Ohio State University is moving their enterprise data environment to the cloud (AWS). This new environment includes an enterprise data lake, data warehouse, and Tableau environment. Hear the strategies that drove the decision, the process for gaining security approval, and the new opportunities to better serve the university community. We will provide an overview of the architecture and specific components. Learn the skills and technologies that have carried over and the new ones gained. Hear the challenges we faced, and the newfound advantages gained through the cloud and modern technologies.
Building the Yellow Brick Road: How the University of Michigan is Planning for an Information Architecture in The Cloud
Michael Sheppard & Amber Madden, University of Michigan
What is your cloud data integration strategy? With Higher Ed IT moving more applications and services to the cloud, Data Integration (and Business Intelligence) is more challenging than ever before. Learn how University of Michigan is moving interdependent systems to cloud. Hear UM’s developing strategies for data warehouse, ETL, BI and data governance in response to a constantly evolving ecosystem of disparate data sources.
Business Intelligence and DevOps or The Road to Hell is paved with Good Intentions
Kyle Quass, Indiana University
Indiana University has a charge to adopt a disciplined approach to software development and delivery in which operations and development engineers participate together in the entire service life-cycle, from design through the development process to production support. It consists of a set of practices and cultural values that enable an organization to deliver valuable software and services at a high velocity. This presentation will examine the long and winding road to adopting this practice within the Enterprise Business Intelligence team at the University.
Campus Mapping Challenges Solved!
Andrew Cluff, Brigham Young University
Do you need to map data on campus? We will use Tableau and other tools to solve 3 main challenges with campus mapping. First, we will demonstrate how to get a more detailed map (or a satellite view) of campus areas. Next, we will explore ways to deal with duplicates without data loss. Finally, we will use Python to determine inclusion or exclusion of a data point on the map.
Closing the Gap between Resourcing and Project Delivery with Data Warehouse Automation
Kristen Handley, Whitman College for WhereScape
To unite data across systems and create a single version of data truth for college leaders, the Whitman College Enterprise Technology team began to build the college’s first data warehouse on its own. It soon discovered that with limited resourcing and a steep learning curve ahead, the challenge and project timeline would be much greater than anticipated. In this session, hear from project lead, Kristen Handley, how Whitman College elected to use automation software to close the gap and the results they have seen to date as they begin development. Kristen will also share lessons learned helpful to any team beginning its first data warehousing project.
Cloudy with a Chance of Success: Arizona State University’s Journey to the Public Cloud
Kristin Kennedy & Jason Green, Arizona State University
Arizona State University has always embraced the concept of Managed Self Service so it only makes sense that moving to the public cloud would be the right choice. Moving to the public cloud can be challenging for many reasons, but if embraced fully can really benefit those in higher education. We will discuss our journey, where we are, what we have learned and where we are going in the future. At the end of this session hopefully we can give you a road map to help you plan for the cloud should you choose to do so.
Cognitive Course Scheduling with Minerva CS
David Pacific & David Doucette, Lighthouse Computer Services
With cognitive computing, universities can optimize the course selection process to help students make effective decisions. That is the power of Minerva Cognitive Solutions (CS). It provides intelligent course selection decision support, taking advantage of IBM® Watson™ APIs, to find courses that meet many criteria, drawn from both structured and unstructured data. It enables: Improved student decision-making effectiveness and confidence: Provides students with greater visibility and insights about the courses available and which align best to their persona and goals and Accelerates the discovery and discussion process between students and advisors to support the student’s need for personal attention and the advisor’s need for efficiency in their meetings.
Data Governance – The Saga Continues
Augie Freda, University of Notre Dame
Since 2012, Notre Dame has been executing a data governance program. Initially focused on the University’s BI initiative, data governance efforts have expanded and impacted the use of information well beyond. Learn how data governance efforts act like a flashlight, illuminating dark corners long unexplored and cultural norms long unchallenged. Go beyond definitions and metadata into how data is shared, how jobs and positions are defined and managed and how the University’s organizational structure is defined. Along the way, see how data matters in ways unforeseen and how the true value of a BI is realized as the program matures.
Data Modeling & Decision Support for a more Diverse Faculty
Heidi Hiemstra, PhD, University of Kentucky
At the University of Kentucky, the offices of faculty advancement and diversity have joined forces with institutional research to identify and measure roadblocks in the hiring, tenure, promotion, and leadership of female and under-represented minority faculty. The definition and operationalization of key faculty career milestones will be discussed, along with the new data models and visualizations developed to answer questions about faculty cohort progression. The resulting visualizations will be presented, as will the ways in which these results have been used to challenge practice and procedure around faculty advancement at UK.
Data Warehouse of Dreams: If we build it, will they come?
Theresa Sherwood, Bowling Green State University
When BGSU started its warehouse, we built complex reports and dashboards that we thought our users would love based on input from user interviews. However, when we rolled out our content, we found that few people logged in to view it. In follow up interviews, we learned that the content was too confusing and not useful in day to day operations. Join us in our presentation to learn steps we’ve taken since to encourage user adoption and to write reports that people actually use!
Effective Partnerships between IR & BI for Improving the Institution’s Data-Informed Decision Making Capabilities
Yuko Mulugetta & Vanessa Brown, Ithaca College
Creating a data-informed decision making environment is a complex undertaking. The central question is Are we really delivering accessible, understandable, and actionable analytics to the campus community in a timely fashion? This session discusses how a mid-sized institution in New York leveraged an effective partnership between IR and BI to create a data-informed environment by focusing on the technology, analytics, talent and culture in a time of limited resources.
Empowering Agile BI with DW Testing Automation
Adithya Buddhavara & Narendar Yalamanchilli, Datagaps
Our products (ETL Validator and BI Validator) can help with end-to-testing automation to help customers deliver large BI projects on time and in budget. During the presentation, we will walk the users through few use cases from our existing customers in the higher education space.
Exploring Text Classification to Automate Document Categorization
Dimuthu Tilakaratne and Archana Mandala, University of Illinois
Unstructured documents are a wealth of information for data analysis, but classifying text to easily categorize the data is not a trivial task. With our Sponsored Projects reporting and analysis solution we wanted to automate classification of project abstracts to make it easy for clients to find data about grants of similar type without manual intervention to categorize them. This presentation will describe our journey through this analysis.
Finding Data Governance in Your Data Warehouse
Scott Flory – ASR Analytics
Are you in the process of building a data warehouse? Establishing data governance? Both? At ASR, we see these two efforts as having a symbiotic relationship; one without the other can result in projects with no direction, and disappointment in the outcomes. Data governance provides the structure necessary for institutional agreement on data definitions and business logic when creating or expanding a data warehouse. Similarly, a data warehouse project can support your data governance efforts by providing a tangible purpose and fostering buy-in. We will discuss ASR’s comprehensive approach to leveraging the warehouse development process to formalize your data governance strategy.
From Rags to Riches in 5 years
Eady Broscheit, Regis University
Within five years, Regis University went from a non-existent, decentralized reporting, and non-data driven decision making process, to a fully-functioning warehouse, centralized reporting, and data-driven decision making University. We will explore this University’s approach, tips and tricks, to going from data rags to data riches.
Getting BI Development Projects over the Finish Line: A panel discussion on Release Management for BI Development Teams
Dawn Hemminger – University of Washington, Nick Roberts – Davidson College, Brian Lawton – University of Washington St. Louis & Shelly Turner -University of Michigan
Many BI Development teams are not getting their projects to the finish line on time. Do you know how your team is doing that? Hear from panelists from four institutions who have been tackling the age old question, “How do you deliver projects that are of value to our customers, on time and within budget?” Our panelists are from big and small schools, at different levels of maturity for implementation, and applying different methodologies that fit their unique culture. We’ll be the first to acknowledge that Release Management isn’t sexy, but it’s a critical component to getting good work out the door!
Getting Workday Financial data into the Data Warehouse – Jeffrey Meteyer , University of Rochester
With the inclusion of a new cloud based financials reporting system (Workday), all reporting was expected to be sourced from the application, but different business needs, an existing and trained workforce using BI tools, needed access to the data for various projects and business processes. The University will explain how it developed structures and ETL processes to enable Workday Financials in its Enterprise Data Warehouse for consumption, while being in sync with the source system.
Govern your Reporting Environment: create a clear path from question to answer
Brenda Reeb, IData Incorporated
Successful institutional reporting is based on good reporting processes and data governance. In this presentation, we will share our perspective on the best practices of data management. We will discuss the iterative lifecycle of data requests from questions to answers and we will also introduce the Data Cookbook, the data management tool for higher education. The Data Cookbook provides workflows to manage the process of reporting and to govern the knowledge that is shared through your reports.
How Notre Dame made its Data Warehouse Accessible & Understandable to Users across Campus-Without IT becoming the Bottleneck
Steve Sporinsky, University of Notre Dame
For Notre Dame to embrace data driven decision making, it wasn’t enough to have an enterprise data warehouse; we needed people to want to use it. To achieve this, we recognized the need for a full service model, including training, designed to make the EDW accessible by tools people use in their day-to-day work such as Excel and Tableau. We also needed to protect sensitive data. The vision: Make our data reliable, reachable and relatable. Steve Sporinsky describes how Notre Dame is tackling these challenges.
How One Report Can Deliver A Thousand
Robert Silcher & Yiorgos Marathias, Phytorion
When a report is developed, it rarely goes unchanged. Users want and need changes. A common way to address this issue is through ad-hoc querying delivered via a BI tool, but BI tools do not always provide an ideal solution for all users and certainly do not solve the underlying data problems. In this session, we will discuss best practices and software solutions to address the difficult challenge of changing report requirements. We will also review analytical development methodologies such as the logical data warehouse, the classic data warehouse and data lakes.
Human Resources: How the Right BI Architecture & Visualizations Create Strategic Insight
Dawn Moore & Brenda Ulin, University of Iowa
University Human Resources and the BI Shared Service Center collaborated to produce BI solutions that allow university HR professionals to make informed workforce planning decisions, track key demographics, understand turnover, and glean insights on how their departments compare to university norms. This presentation will focus on the importance of a hybrid data architecture, strategies to leverage complex effective-dated ERP data, and how visualizations were chosen specifically to enhance the ability of HR professionals to extract key insights. HR Diversity, retention, turnover, and succession planning case studies demonstrate the value of the HR BI solution.
Introduction to Graph Analytics
Pieter Visser, University of Washington
Most people are familiar with the concept of a “friend of a friend” query, and have seen Visual network graphs, but few people realize how easy it is to get started with graph analytics and derive new insights into their own data sets. This presentation gives an overview of the field of graph analytics, how to do exploratory data analysis, and find insights into networked data sets. We will look at several real-world use-cases and perform a wide range of analyses on a variety of graph networks.
It Takes a Village: Ideas on Building a BI Community on your Campus
(Panel Discussion)
Shelly Turner, University of Michigan, Michael Hansen, Oregon State, Nick Barbulesco, UC Davis, Susan Schaefer, University of Utah, Nick Chaviano, Georgia Institute of Technology
Community of Practice, Competency Center, Community of Experts & User Groups – the central BI team cannot perform all the BI for all of campus needs. We need to spread our knowledge, expertise and guidance as well as provide passion and support in order to build a campus-wide ‘data village’. Attend this lively panel discussion to hear success, started strong then fizzled and recovering from failed attempts stories. Learn from the panel’s experiences and as part of the audience, share your own and plan to ask questions!
Lessons Learned in a 15 Year Journey
Ted Bross, Princeton University, Amy Miller, University of Pennsylvania, Suneetha Vaitheswaran, University of Chicago, Aaron Walz, Purdue University
Commencing in 2003, HEDW has grown from an informal group of 30 to a current membership surpassing 3000. Some things have changed during that time while others have not. This has happened on both a technical level as well as organizationally in each of our institutions. This panel discussion features 4 past presidents of HEDW, each of whom will provide his or her unique comments and perspectives.
MacGyver ETL: Implementing an Open Source ETL while the Clock’s Ticking
Nick Kimmel & Steve Letzring, Bowling Green State University
At Bowling Green State University, database and desktop operating system changes require an upgrade to the existing ETL tool. These changes are happening in conjunction with a move of the datacenter to an offsite location, driving constraints in both resources and time. After conducting an analysis of potential ETL tools, we have chosen to implement an open source product. Join our presentation to learn how we selected a new compatible ETL tool and our plan to upgrade and move to the offsite data center within a six-month timeframe.
Modeling NSF Higher Ed Research Data (HERD) for Analytic Dashboards Jason Morales – Microsoft, Ian Czarnezki & Miriam Clark – Kansas State University
The National Science Foundation (NSF) annually collects data from over 600 institutions on research and development (R&D) activities at higher education institutions. This serves as the primary source of information gauging the national level R&D activities. Kansas State University collaborated with Microsoft in the development of a Power BI solution that will provide greater insight into any institution’s research activity compared to the others, including the ability to compare expenditures to other institutions, by principal investigator, by discipline, and by source of funds. In this session, we will demonstrate both the dashboard/reports and the data model behind it. Due to the complex format of the data, we will demonstrate how you can use the data transformation capabilities of Power BI to create a structured data model that will facilitate analysis. We welcome your questions and input during this interactive Data Community session.
One Bird at a Time: Tackling Monumental Tasks at Your Institution
Melissa Hartz, Colby College
“I need you to build a data warehouse.” “Establish data governance.” If only it were that easy! Tasks like these are so vast and overwhelming, it can be hard to even figure out where to start. Maybe you’ve been tasked with a project that feels Herculean – the good news is, you’re not alone. Hear how the IR Office at Colby College is taking on tasks around building a data-driven culture on campus. You will learn how to develop a strategy, what’s working, and lessons learned so far in these monumental projects.
Oregon SLDS Project: The Devil is in the Data
Rebecca Ator – Informatica, Michael Rebar, Ph.D.- CEdO of Oregon, Mark Richards & Elizabeth Snow-Trenkle – EKS&H
The Chief Education Office (CEdO) of Oregon was created in 2015 with the mission of coordinating and removing barriers for school students by overseeing a unified public education system that starts with early childhood services and continues throughout public education from kindergarten to post-secondary education. The state’s educational outcomes are defined by the Governor’s “40-40-20” outcome goals:
- 40% of Oregon students will achieve a 4-year college degree or better
- 40% of Oregon students will achieve an Associate’s degree or professional/technical certification
- 20% of Oregon students will finish their education with a high school diploma or GED
Join staff from the CEdO, EKS&H and Informatica as we explore the vision, architecture, challenges and successes faced in collecting data from multiple institutions. We’ll also discuss transforming, profiling, standardizing, matching, mastering and dimensionalizing data and the reporting and visualizations needed to support and drive education initiatives.
Panel Discussion: Path to Next Generation BI
Greg Siino, University of California, Davis
Join this panel discussion to explore next-generation BI from the perspective of those institutions just getting started. What is next-generation BI? Should institutions develop a traditional star-schema data warehouse and reporting solution or possibly leapfrog to data virtualization, data lakes, Hadoop, Hana, self-service analytics, cloud, etc. Which BI components are foundational to both “traditional” and “next generation” and which components can be skipped in favor of new approaches? Even if you must go through traditional stages of maturity, are there ways to fast track the foundational work to quickly get to leading-edge BI? What role does governance and executive sponsorship play? How does make vs. buy factor into the approach? We’ll engage with BI leaders from institutions with varying levels of BI maturity to gain their insights. This will be a highly interactive session with opportunities for those joining the session to share their expertise too.
Partnerships in Technology, Predictive Indexes & Process to Influence Student Success & Retention: A Data Science Perspective
Uche Nwoke, Brenda Ulin & Grant Brown, University of Iowa
Predicting student success is an important tool for educational institutions; however, such efforts are subject to competing requirements for student, collegiate and institutional specific objectives. Analytics must provide accurate student-level predictions, reliably forecast aggregate outcomes, and provide informative and actionable qualitative business intelligence. We discuss our collaborative approach to these issues, focusing on three components of data science that have been key to our success: data architecture, machine learning and statistical modeling, and integration of analytics into campus practice. We share strategies to navigate the complex balance between individual and aggregate objectives and the impact of identifying influencing factors.
Partnerships in Technology, Predictive Indexes & Process to influence Student Success & Retention: A Process, Data Architecture & BI Perspective
Brenda Ulin & Danielle Martinez, University of Iowa
University of Iowa utilizes data scientists, student success staff, and IT resources to support student success and retention. Through this collaboration, we have created student success predictive indices and identified associated key positive and negative influencing factors. Incorporating these indices and influencing factors into our institutional intervention systems provides coordinated student success outreach and messaging. We discuss our collaborative approach to utilizing predictive indices and influencing factors for modifying student success policies, programs, and practice. We will share our approach to holistic, quantitative and qualitative, student attribute identification and collection; data storage strategies; process integration; outreach correlation; and BI solution demonstrations.
PASSHE and Performance Architects Case Study: Student Enrollment Analytics Modernization Across 14 Institutions in the PA State System of Higher Education
John McGale, Performance Architects and Jeff Montgomery, IT Services Coordinator of Application Development, Indiana University of Pennsylvania (IUP)
Indiana State University of Pennsylvania (IUP) was selected to drive student enrollment reporting modernization across 14 state institutions in the Pennsylvania State System of Higher Education (PASSHE), including IUP. This effort resulted in an unprecedented new system and process for the state, allowing the team to seamlessly view data within and across all universities. Join us to learn how IUP partnered with Performance Architects to develop internal staff skills in order to foster a new analytics-based culture and to create a strategic multi-university data warehouse environment. This presentation discusses our recommendations on how you could address challenges you might face along the way that we faced, such as enrollment, budgetary constraints, and legacy system issues, all of which required a new approach to capturing data in a governed manner while still being agile and focusing on delivery.
Predicting Student Success: Insights and Lessons Learned
Heather Chapman, Weber State University
Enrollment, retention and graduation are some of the most important metrics used to determine the health of universities. As competition for enrollments and threats to the stability of these institutions continue to increase, universities are implementing a diverse set of strategies to combat the threats. Predictive analytics can be used as a monitoring and accountability tool to inform the discussion. This presentation addresses methods, pitfalls encountered, outcomes experienced and next steps associated with two large-scale predictive analytics projects. Suggestions and “lessons learned” will be provided in order to inform others attempting to introduce predictive analytics at their institutions.
Quality Data In, Quality Analysis Out
Kimberly Griffin & Sanish John, Northwestern University
Embedding data warehouse content in a website to provide on-the-spot access to data without the need for logging into complex systems was the vision, and Northwestern’s Enterprise Reporting & Analytics team partnered with the Office for Sponsored Research to make it a reality. Through a combination of Tableau Server and web service calls to Cognos reports, the ERA and OSR teams succeeded in deploying enhancements to the OSR website, providing up-to-date information to campus about the status of sponsored research transactions along with a year-to-date look at proposal, award and contracting activity.
Redefining Data Warehouse and Analytics Operations in a New Era – Panel Discussion
Jack Neill, HelioCampus (Moderator), Dr. Yuko Mulugetta – Ithaca College, Dr. Craig Rudick – University of Kentucky & Kevin Joseph – UMBC
As the data warehousing community enters the two decades of implementation, the main question is no longer how to build a data warehouse. The main question is Are we leveraging our data warehouses to deliver accessible, understandable, and actionable analytics that are timely to the campus community? In contrast to the earlier “how to build” IT-centric question, the “how to leverage” question addresses the user-centric perspective and the value to the enterprise. This shift brings to the forefront a new set of issues. This panel explores a new approach to data warehousing and analytics from the user-centric perspective.
Scanning Success – How Student Affairs and IT grew a Participation Tracking System
Ken Schreihofer & Kevin Joseph, University of Maryland, Baltimore County (UMBC)
UMBC’s Division of Student Affairs partnered with IT to solve a participation tracking problem for the Student Life office – and the successful venture spread university-wide. By building the feature into the campus portal, usage has grown across the university. Data is then stored and analyzed in the Data Warehouse.
Securing Business Intelligence with the Easy Button
Julie Parmenter, Indiana University
Tableau and other visualization tools have been game-changers in that they allow non-technical users to display data in meaningful ways. However that data is still an important asset and must be managed and protected appropriately. At Indiana University, we have developed new software that will create, route for approval and apply security groups to Tableau server. Security groups can be developed using Role-based access or by individual user names. The software uses Tableau APIs to perform these actions. Join us as we share details and demo our software.
Self-service overhaul: Three Lessons Learned at Prince George’s Community College
James Dick, Prince George’s Community College
In spring of 2014, Prince George’s Community College embarked on a process to allow individual department’s access to enterprise data though our newly built data warehouse and business intelligence system. One year later, the system was severely underutilized and frustration was high. This presentation will discuss Prince George’s Community College’s self-service implementation, our three largest challenges, and what we are doing to address them.
The Art of Analysis: Using Dashboards that Tell the Right Story
Kimberly Ford, Walden University
Creating effective dashboards that tell the right story can help users of the data make the best decisions. Analyzing data in a way that help users make use of dashboards and data can aid in faculty performance management, programmatic decision making and ensure student success. “
The Data is Flat: Enabling Learning Analytics Research using Institutional Student Data
Steven Lonn & Glenn Auerbach, University of Michigan
Hear about a collaborative project at the University of Michigan to transform institutional reporting data into a centralized dataset for institutional research and learning analytics. They will describe how the team transformed admissions, registrar, and other institutional data into a simplified, normalized, flat, and ready-to-analyze format. The rationale, process, reasoning, and development processes completed over 18 months will be discussed. Session attendees will learn how they might build such a dataset at their own institution and how this project can be a model for partnerships between IT, faculty, students, and staff across the institution.
The Flying V Strategy: Finding the Right Formula for BI Success
Nick Chaviano & Rodney Pacis, Georgia Institute of Technology
Georgia Institute of Technology has attempted to institutionalize and transform campus operations through business intelligence over the past six years, but failed every attempt. Through these multiple failures came the winning “Flying V” strategy: the creation of Enterprise Data Management. On this third attempt came effective leadership, a clear, goal-oriented strategy, executive support, proper staffing, and ultimately success. This presentation will examine the past landscape at Georgia Tech and why prior attempts failed, discuss the formula for success on the third attempt, explain the current state including our BI, data warehouse, and governance strategies, and finally explore our future.
The Promise and the Pitfalls of Predictive Analytics
Craig Rudick, University of Kentucky
Predictive analytics methods can be a powerful set of tools to analyze institutional data, however their implementation too often fails to live up to their promise. In this session I will discuss several successful predictive analytics projects we have undertaken at the University of Kentucky. I will describe not only the methodological details and quantitative results of these analyses, but also the project management, communication, and implementation strategies that led to successful project completion. I will also share a number of lessons learned and pitfalls to avoid, both from our own experiences at UK and my observations of other institutions.
Unraveling and Defining Student Success: Analytics Guided by the Student Experience
Daniel Newhart, Oregon State University
In the past couple of years, we have been interested in learning more about student success for specific populations and groups, hypothesizing that pathways to success might look different for various groups of students within higher education. Using our participation tracking systems, we are beginning to learn more about how participation in campus services outside the classroom may be contributing to student success and how this might inform our analytic models. This presentation will discuss important considerations as campuses seek to understand student success more holistically, and how technology can assist in this understanding.
Using SQL to Write SQL – What is in Your SQL Toolbox?
Neil Belcher, Cornell University
Using SQL to generate SQL scripts can be a very powerful tool for doing many mundane data tasks. This technical session will walk through example SQL statements that can make your life much easier. For example: How about SQL that generates a list of CREATE INDEX statements for Dimension and Fact tables, a list of ALTER TABLE statements to create PK/FK’s or “Uber Minus” statements that compare the data between two tables? What is in your SQL toolbox? This session will include Oracle and Transact-SQL examples. Attendees are encouraged to bring their favorite SQL snippets to share as well.
Visualizing the Cost & Benefits of Student Incentives. Is it Sustainable?
Susan Schaefer, University of Utah
Can student incentives exceed revenue when departments or programs try to get students to enroll? Do we give too many scholarship and/or waivers to our out of state students or other student populations? In order to try and answer these and other questions, the budget and analysis department worked on combining student data with financial records to create a dashboard for departments and executives. This presentation will look at the process taken to combine student and financial data and create visualization that answers questions and starts a dialogue.
What Good is Data if Leadership doesn’t see it when it Counts?
Jason Simon, PhD & Dan Hubbard, Ph.D., University of North Texas
We often work in isolation and wonder if our efforts are applied to key business/policy decisions. In order to best leverage the power of analytics to inform institutional policy, procedure, and financial decision making institutions need a new way to put data in the hands of executives to respond to rapid-fire questions. Using the metaphor of moving from an untamed data jungle to a well-planned data arboretum, this session will highlight practical approaches to surfacing vitally important information during high-stakes gatherings of executives and regents/trustees. Elevating data is key to ensuring a sound analytical culture of data-driven decision making.
What We Know About the Impact of Student Engagement
Heather Chapman & Jessica Oyler, Weber State University
Retention of students is very important to institutions of higher education. Retaining existing students is cheaper and leads to improved outcomes for students. To potentially improve retention rates and help students persist, universities often have a variety of services and engagement opportunities available. Data on how well these services work to retain students is scarce. This session will report results from both descriptive and predictive data regarding the importance of student engagement and services, and will provide next steps and lessons learned.
When Data Custodians Go Rogue
Keith Van Eaton, University of Washington
Within a maturing data governance program, there was a strong and immediate need across the UW campus to develop and approve institutional business terms. Without a fully functioning data governance program, terms could not be added or updated to support the UW’s growing reporting needs. To solve this issue, a business term vetting and approval process was created from scratch, validated through a proof-of-concept, and is successfully used today to manage approvals. In this session, you will learn: Why and how the process was built, how the proof of concept played out, and how this process is being used today.
Where Should All this Business Logic Go? A Panel Debate
Daniel Riehs – Boston College, Ravindra Harve, Boston College, Lauren Himml – Boston University, Kyle Quass – Indiana University, & David Ricker – Dartmouth College
Data warehouse reports can incorporate a great deal of business logic, stored in the database, meta-data layer, and report specification. Best practices suggest that all three have a place, but people have strong, differing opinions depending on their place in the BI organization. An end-user might want all logic in reports so he can examine and modify it, while an ETL developer might insist that logic belongs in the database to speed up the system and allow for easy re-use. This panel session will feature a variety of opinions, and attempt to answer: Where should all this business logic go?
Workday in the Warehouse – Panel Participation
Steve Fischer – Ohio State University, Ryan Schlagheck -Yale University & Jeff Meteyer – University of Rochester
Listen and ask questions with three individuals that have implemented Workday solutions at their institutions, the challenges they’ve experienced, and how they have enabled new feature/functionality with their data warehouse offerings for BI and reporting objectives.
BONUS: Ask a GURU!
Amy Miller ( UPenn), Hank Childers ( Univ of Arizona ), Dipti Desai (Univ of Chicago), Vanessa Brown (Ithaca College), Suneetha Vaitheswaran (Univ of Chicago)
DOUBLE BONUS: 2018 General Assembly and Research Findings – ” The Last Hank and Aaron Show”
2018 Research findings and interpretation provided by Hank Childers ( University of Arizona) and Aaron Walz (Purdue University).