Presentations

Exploring various AAUDE related topics including: Structured vs Unstructured Data; How to search the AAUDE Institution Websites; Searching the AAUDE listserv.

        April 14, 2015 - Joe Mantione

This presentation demonstrates three ways an institution can use faculty detail data.  Methods demonstrated include a searchable database documenting all faculty honors, comparative charts for faculty within units using performance based metrics, and a scatter chart with quadrants showing publication and citation rates per discipline for US and Canadian institutions.

        April 9, 2014 - Craig Abbey

Forecasting is essential for university planning process.  In this session, the presenter will demonstrate how to create a ratio-driven historical enrollment model for predicting returning student enrollment.  Tableau visualizations will be presented that demonstrate forecasting tools, such as sliding scales for retention percentages that will show immediate impact on predicted student enrollments.  Additionally, the presenter will forecast graduation rates and tuition revenue using the previously created enrollment prediction models.

        May 31, 2018 – Rachel Link

Conducting a faculty salary equity study does not need to be difficult.  This presentation will guide researchers through the process, from literature review through study design, completion, delivering results to administrators, to suggesting communication strategies.  The presenters will share experiences from conducting multiple salary equity studies to help simplify the process, while maintaining positive relationship with both faculty and administrations.  This session will provide recommendations for statistical modeling, focusing on variable selection and process.

        May 31, 2018 – Melinda Whitford

Accurate prediction of retention numbers is essential for university planning processes. The presenters will demonstrate how to create one type of returning student enrollment prediction model using SPSS and perform simulations using the Monte Carlo method to quickly provide estimates of student returning population. Participants will learn how to create their own models and how to perform simulations which would be useful for a range of potential outcomes, resulting in improved decision support information for university administrators.

        May 31, 2017 – Rachel Link, Mike Randall

Effective data visualization, a key tool in any institutional researcher’s toolbox, depends on a match between the questions asked and the types of visualizations produced. The right match makes data highlights leap off the page, but the wrong visualization can obscure key points. Our presentation will outline the main types of questions asked of institutional researchers, identify the best visualizations for each type, and explain the aspects of each visualization that make it the right tool for a particular job.

        May 31 - Craig Abbey, Lauren Young

Bibliometric analysis is an important part of research evaluation. This session provides an overview of how bibliometric analysis was used to support a university’s hiring and investment plans. Using the Global Research Benchmarking System and Scopus data, techniques to identify strength areas and collaboration patterns using Tableau, Circos, and NodeXL are discussed and demonstrated. The resulting graphics take a large volume of data and create visualizations that are easy to understand and highlight relative strengths and collaborations patterns.

        (Session Id: 1578)
        May 28, 2014 - Craig Abbey,  Rachel Link

Finish in 4, a new retention initiative of the University at Buffalo, provides an institutional commitment of resources and guidance to help undergraduates complete the bachelor’s degree in four years. What is the appropriate reach of such a program? This presentation describes a cluster analytic approach to identifying students who are more likely to pledge the Finish in 4 program and to evaluating which of these students are more likely to succeed when armed with the program’s resources. Adding this approach to descriptive program assessments can provide rich context for discussions of program success and planned growth.

        (Session Id: 1691)
        May 28, 2014 - Lauren Young

A salary equity analysis of faculty and staff was developed at a large research university in keeping with federal evaluation guidelines.  Analyses of employees working in similar capacities suggested that neither gender nor minority status were generally associated with salary discrepancies over and above the effects of background and experience.  Nonetheless, regression models differed greatly in their predictive power, and these differences were often attributable to imprecision in employee group definitions.  Further discussion focuses dually on the practical choices that led to model refinement and the data needs that hindered analysis of specific employee groups.

        June 5, 2012 - Lauren Young

Once the big thinkers have blessed the 50,000 foot concept views, the checks have been cut, and the software has been installed, someone has to actually sweat the messy details of making campus Business Intelligence (BI) a reality. At the University at Buffalo (UB), the Office of Institutional Analysis (OIA) is often called upon to step into the breach to get the job done. This presentation will focus on identifying how UB OIA leverages both its position within the campus processing infrastructure as well as its relationships with personnel down in the trenches to help create workable BI solutions. Case studies involving implementation of a data quality initiative to enhance the accuracy of UB's enrollment information, the creation of a Tableau based FactBook dashboard based on these data, and our efforts to "hack" UB's PeopleSoft Campus Solutions system to identify on-line majors will be presented to inform the discussion.

        June 9, 2016 - Jon Havey,  Mark Molnar, Michele Sedor, Laura Szefel

The presentation helps show survey response results that are interactive, user friendly, and in common formats (e.g. same Likert scales or other coding schemes when possible).  The interactive search tool allows for finding topical area questions from various surveys much easier.

        June 5, 2014 - Amy Guthrie, Joseph Mantione, Michael Randall

How The University at Buffalo is using Tableau, touching upon: Data, Schedule, Delivery, Format, Audience and Content.

        June 6, 2013 - Rachel Link, Joseph Mantione

Snapshot data tied to census dates is critical to offices of institutional research. Without it, volatile transactional data becomes a source of endless disagreement over numbers. However, snapshot/census data can be limiting as well. Oftentimes, there is rich dimensionality in our transactional systems that is not captured in census snapshots because it is not needed by the intended recipients. Given the many priorities being juggled when a new ERP is being developed, producing a serviceable census dataset represents a considerable challenge by itself. This presentation will cover various techniques employed at SUNY Buffalo for adding synchronized dimensionality to our official reporting data set.

        April 5, 2016 - Jonathan Havey

The University at Buffalo (UB) implemented PeopleSoft Campus Solutions in 2011. As part of this effort, UB also implemented iStrategy's HigherEd Analytics (now BlackBoard Analytics) in a Data Access and Reporting initiative that went live at the same time. The first phase of the Data Access and Reporting initiative was intended to provide data to reporting super users and to create a suite of basic, web-accessible operational reports for use by departmental administrators. To this end, the pre-existing data-mart (branded "Info Source") was initially populated using BlackBoard Analytics data models, and a pre-existing OBIEE instance containing human resource and financial reports was leveraged to deliver operational reports in the student realm. The rollout, following an extremely aggressive schedule for implementing Campus Solutions (which we branded as “HUB”), was successful in the sense that there was no interruption of service in providing data to the university community. Departmental administrators were able to run basic reports on majors, classes, enrollment, student groups, and service indicators (i.e., holds), and reporting super users were able to start rebuilding their local reporting systems. As time went on, however, certain gaps emerged that were difficult to reconcile with the data in the HUB. All of these challenges originated in the attempt to generate a student-term data model using effective-dated data that lacked term dimensionality. This presentation will discuss how the University at Buffalo responded to these gaps through the following: Customizing views, constructing supplemental views to fill the gaps, reviewing business process in assigning effective dates, and updating the OBIEE .rpd file to capture some of the lost data.

        April 21, 2015 - Jonathan Havey

The analytics team in the UB College of Arts and Sciences developed an information system that tracks the path that each individual student takes during their career, then delivers visualizations to help understand key changes that occur. The system includes a custom dataset with tools for analytics and visualization. Twenty years of data are combined from disparate sources to build the dataset. Analytics tools combine this extensive dataset in countless ways to gain detailed understanding of relationships among various metrics. Most critically, the visualization tools then allow for dynamic viewing of patterns in the raw flow or analytics data to "see" what has happened and help plan for pre-emptive actions to reverse problematic patterns.

        November 14, 2016 - Brian J. O'Connor

Tableau will be partnering with the University at Buffalo to discuss how to use Tableau and how to share survey results, university facts and other useful data with a wide audience.  The presentations will cover some real life examples at the University then go into detail as to how to create and share visualizations.

        November 2, 2015 - Jon Havey, Rachel Link

Data quality is a frequent concern of IR staff. We all have our pet examples: High school GPAs of 980.25, a combined SAT Reading and Math score of 390, and a graduate who was never in the degree major. Frustrations on the part of analysts are understandable, but they do not point to a solution because the main purpose of a transactional system is to provide functionality to the university community. Like at many universities, the areas responsible for our system--the technical support office, business offices, and IR office--report through three separate tracks. A make it so approach would have been impossible. This presentation will recount how we were able to launch a data quality initiative that was a volunteer, bottom-up effort.

        November 3, 2015 - Jon Havey

Institutional researchers are often expected to produce definitive summary data to be posted in the public domain for use by external policymakers, accreditors, peers, and other outside constituents. It is imperative that these summaries be correct and consistent, but institutional data may be fraught with pitfalls. Given that raw data are rarely perfectly clean and consistently structured, how can you be assured that a summary number is the right one? We will discuss examples of problems that may arise from missing data, incorrect data entry, definitional issues, multiple records or constituencies, and policy changes and offer diagnostic and analytic solutions.

        November 10, 2014 - Rachel Link, Lauren Young

Recently, the University at Buffalo began to review its policies on course resignations in the hops of reducing undergraduate time to degree. A key component of this review is the assessment of a recent policy initiative that limited repeats for popular introductory and general education courses. We expect that patterns of enrollment and student success following the prior policy change will help to support or refute the anticipated benefits of disincentivizing course resignation. This presentation will review bivariate and multivariate analyses that highlight the impact of course resignation under old and new course repeat policies.

        November 10, 2014 - Lauren Young

Institutional research often leverages alumni and advancement data for college rankings and other reporting requests. Understanding common terminology used in alumni and advancement offices can make the task easier for the researcher and the end result more useful for the advancement and alumni offices, and a better understanding of this terminology can lead to better reporting and more positive outcomes in college ranking. Additionally, learning common giving terms and statistics helps with financial reporting for the institution. Finally, many institutions do not employ a development analyst who can perform statistical analysis on fundraising data in order to segment mailings or perform ROI analysis, and this task may fall to institutional research. This poster will provide these IR professionals with a strong understanding of the terminology, common beliefs, and accepted methodology to make analysis and reporting easier.

        November 11, 2013 - Rachel Link 

This presentation shows how a large public institution has used geographic analysis to better visualize its enrollment pipeline from high school graduates and college inquiries thru application/admission/enrollment and retention. Several dimensions of the potential applicant pool is graphically portrayed in layers as we uncover more understanding about where our applicants come from, who applies, who gets admitted, and who ultimately enrolls.

        November 11, 2013 - Mike Randall

This presentation shows how a large public institution has used geographic analysis to better visualize its enrollment pipeline from high school graduates and college inquiries thru application/admission/enrollment and retention. Several dimensions of the potential applicant pool is graphically portrayed in layers as we uncover more understanding about where our applicants come from, who applies, who gets admitted, and who ultimately enrolls.

        November 5, 2012 -    Craig Abbey, Mike Randall

        May 14, 2014 - Leah Feroleto, Amy Guthrie, Troy Joseph, Joseph Mantione

Presentation reviews why analytics are so important to Higher Education.    Includes Tableau reports illustrating the Admission’s funnel, the enrollment process, benchmarking donor rates, and a review of publication and citations.

        September 9, 2013 - Leah Feroleto, Joseph Mantione, Michele Sedor

For further information, please contact the presenter or