University Data Mart
Project Background
Purpose
The purpose of this project is to develop a data mart, dashboard, and query tools that will provide easily accessible, decisionsupport information for NC State’s executive officers, deans, and department heads.
Overview
The data mart will reside in a centralized database with consistent data and a userfriendly interface. The dashboard will include selected measures of university, college, and departmental quality, effectiveness, and efficiency. The data and constructed measures will be consistent across multiple years, requiring standard census dates and definitions.
With the implementation of compatible student information, HR, and financial systems, NC State can now more easily integrate them into a unified data mart, develop a common set of performance measures, and develop retrieval tools for decision makers across the university.
Goals & Objectives
- Identify and define key data elements and measures that executive officers, deans, and department heads need to use for tracking performance, planning, and other business functions
- Build a data mart to house these data elements and construct measures
- Build a highlevel dashboard to display and provide drill down capacity for these measures
- Provide an appropriate query tool for fixed, adaptable and ad hoc reports from the data mart. Configure the tool to guide clients to selected appropriate comparisons
- Design the data mart, dashboard and query tool to include as much of a unified view of all university data as possible. Implementation will be phased
- Recommend a continuing process for monitoring the effectiveness of the data mart, dashboard, and query tools, and continue development as needed
This project will support all five OIT operational goals.
Scope
The definition phase will identify the data elements and measures to be included in the data mart. Users of the data mart will be consulted for their needs.
Data definitions, reporting periods, and organizational structures will be addressed. The data areas for the first phase will include information that is already reported in a consistent format, such as enrollment, credit hour, degrees awarded, employee, and year-end financial data. Others will be added as needed and when resources are available.
The project teams will search for an appropriate set of tools for creating the data mart and dashboard, and for ad hoc queries. The teams will implement the approved plan.
Authority
The project will be formed as a task force with a steering team that makes recommendations about data mart and dashboard content to the Provost and that makes recommendations about data structures and retrieval tools to the Vice Chancellor for Information Technology.
Project Managers:
| Karen Helm | karen_helm@ncsu.edu | (919) 515-6648 | Director, University Planning & Analysis |
| Lewis Carson | lewis_carson@ncsu.edu | (919) 515-6208 | Associate Director for Institutional Research, University Planning & Analysis |
| Connie Reitfort | connie_reitfort@ncsu.edu | (919) 513-1403 | Assistant Director, Enterprise Application Services |
Major Constraints and Assumptions
The assumption is that this project can be accomplished with existing personnel, data holdings, and knowledge. The major constraint is resources for development: analyst and programming time, specific tools, and infrastructure such as databases. There will be a search for appropriate software tools, beginning with tools currently under license. If currently licensed tools and hardware are not adequate, funding for an appropriate tool and additional hardware may be an additional constraint.
Consistency between internal and external reporting will be enforced, which will facilitate communication between the university and external agencies like UNCGA. Structural changes to the data will be made to suit customer requirements. For example, the current organizational structure will be applied to history so that measures look like the current structure has always been in effect. Measures and ratios will be precalculated so inappropriate constructs are not possible.
