This paper was presented at CUMREC '99, The College and University Information Services Conference. It is the intellectual property of the author(s). Permission to print out copies of this paper is granted provided that the copies are not made or distributed for commercial advantage and that the title and authors of the paper appear on the copies. To copy or disseminate otherwise, or to republish in any form, print or electronic, requires written permission from the authors.

Expanded Grade Context Record
An innovative and comprehensive approach to grade reporting at Indiana University

Abstract

In 1994, the Indiana University Bloomington Faculty Council (faculty senate) passed a resolution calling for the development of an indexed grading system. The mandate led to a discussion of grade inflation that culminated in a decision to present course grades so they were understandable and meaningful to the student, advisor, and all external recipients of transcripts. This challenge led to the development of the Grade Context Record that includes elements from a traditional transcript (e.g., course title, academic department, grade, etc.) as well as additional elements that place the grade into a broader context. This information is distributed via standard reports (e.g., student grade mailers, transcripts) and made available for inquiry via the World Wide Web. The Expanded Grade Context Record was jointly developed by members of the user office (Office of the Registrar) and the computing center (University Information Technology Services [UITS]). Registrar personnel developed all of the presentation (client) components (e.g., On-line transcript display, print, and web access) while UITS personnel developed the business rule components. The Expanded Grade Context Record was implemented in the spring of 1998.

Project development and implementation was complex for a number of reasons.

• a grade distribution database was required,
• the grade context transcript was built by adding a set of business rules to a newly developed academic history system using object oriented design principles,
• all presentation components were developed in the user office in a client/server architecture, and
• general inquiry was placed in both character-based and web environments.

The presentation will provide a description of the Grade Context Record, the general design of the record and new academic history system , the application architecture, the partnership between the user office and the computing organization and how the overall team was structured. The presentation will also include a description of how the Grade Context Record was incorporated into our web- based self-service environment (Insite).

EXPANDED GRADE CONTEXT RECORD

Indiana University

An innovative and comprehensive approach to grade reporting at Indiana University

Roland Coté, Associate Registrar, Services
[email protected]

Mark McConahay, Associate Registrar, Systems
[email protected]

Indiana University

I. The Expanded Grade Context Record

In 1994, the Indiana University Bloomington Faculty Council (faculty senate) passed a resolution calling for the development of an indexed grading system. This marked the beginning of what has been viewed as one of the most innovative and objective approaches to grade reporting. On March 3, 1998, Indiana University introduced the Expanded Grade Context Record.

The awarding of grades and their significance have been topics of discussion and concern in the higher education community since the beginning of record keeping. Readers of transcripts from outside the education community as well as many within, are confused and bewildered by the variety of grading schemes, and they are often unable to determine how to evaluate the academic performance of individual students. The issue of grade inflation has also been debated widely and openly on campuses and in the media for the past 25 years. Several institutions have reviewed their methodologies for awarding grades and reporting them on institutional transcripts. This discussion in the mid-1970's at Indiana University resulted in the Faculty Council mandating that the registrar produce reports that show the distribution of grades awarded to undergraduate students in undergraduate courses. Since 1978, the registrar has provided a copy of grade distribution reports to each academic school and department.

In 1993, the Faculty Council took up the discussion again, this time focusing on how to best present grades so that they will be more understandable and meaningful to the student, advisers, and external recipients of transcripts. The Faculty Council charged its Educational Policies Committee (EPC) to develop a grade indexing scheme that would put individual student grades into a context that would be meaningful to the student as well as to all others who viewed the information.

In 1994, the EPC and the registrar presented a recommendation to the Faculty Council that called for the development of an Expanded Grade Context Record to be incorporated into the student's semester grade report. In addition, a new optional Context Transcript would be developed that could be produced at the student's request instead of the Traditional Transcript. The committee determined that the grade distribution information that is used to generate the context record should also be made available to anyone who requests it.

The following products were developed and implemented at Indiana University Bloomington:

1. Expanded Grade Context Record (Context Transcript attached)
2. Final Grade Notification with expanded grade context (Student Notification attached)
3. Grade Distribution Reports
4. Grade Distribution Query on the World Wide Web

The data elements selected by the faculty to be displayed on the Context Transcript include only those which originated at, and are already being maintained by, the University. For this reason, test scores from external agencies (e.g., SAT, ACT, AP scores) and grades awarded at other institutions are not included. For each course in which the student was enrolled, the traditional transcript already displays the following elements: course title, academic department, course number, credit hours, and grade. The Context Transcript includes those existing elements and also the following additional elements for each course entry:

• Index The number of students in the course section who received either the same grade as the student or a higher grade than the student, followed by the total number of students in the section who received GPA grades.

• Instructor name The name of the instructor who taught the course and who submitted the grades.

• Grade distribution The number of students who received each of the possible grades, including those that indicate student withdrawals during the first week of classes and after the first week.

• Class GPA The average of all GPA grades awarded in the course section (i.e. the average grade expressed in n.nn format).

• Average student GPA The average of the academic record GPAs of all students enrolled in the course section at the time the context information is compiled. The GPA level used for each student is the student's current GPA level (undergraduate or graduate). Grades awarded for the semester displayed are included in the Average Student GPA calculation.

• Majors The percentage of students in the class whose major school (or department for students in the College of Arts and Sciences and University Graduate School) matches the school or department offering the course.

• Context Effective Date The date when context information for the semester was compiled. Context information is compiled twice for each semester; once at the end of the semester when reported final grades become official transcript grades, and again approximately one month later to reflect grades received after the deadline and changes submitted by instructors.

In order to comply with FERPA and the University Policy on the Release of Student Information, neither the index nor the context information appear when fewer than five students were enrolled in the course section or if GPA grades were awarded and fewer than five students received a GPA grade.

II. The Expanded Grade Context Record Information System

A. Introduction

In order to deliver a grade context record, the project team established a set of project deliverables that comprised the project. They were:

Grade Context Information Database

A DB2 database was created to store and retrieve grade distribution and context information. The database serves as the primary source of grade context information for all information retrieval, inquiry and transcript production applications. In addition, since the data is relatively stable (frozen at the end of the term and after grade processing is complete) the data can be used to support academic managers and longitudinal academic studies. The database will also serve to create aggregate grade distribution reports (e.g., by course, academic school, academic department, etc.).

The Expanded Grade Context Record

The Expanded Grade Context Record merges a student's academic course history (e.g., semester of enrollment, course, course title, hours, grade, etc.) with grade context information (e.g., class GPA). The record is available to students and academic staff in hard copy (as a transcript) and for network inquiry.

The Final Grade Report with Context Information

At the conclusion of each academic semester, all students receive a printed grade report that includes a listing of their enrolled courses, credit hours and the grades received. The new Expanded Grade Context Report includes this traditional information, but integrates it with grade context information. Two reports were defined:

• Student Grade Context Report (Mailer)
• Academic Department Grade Context Record (a.k.a. "Deans" copies)

B. Expanded Grade Context Information System - Development Methodology

In order to support the applications and reports defined above, the project team had to define the individual components that would comprise the Expanded Grade Context Information System. Many components had to be developed anew, while others existed in current operational systems (or systems that were being built). Since late 1996, systems development at Indiana University has been governed by the following set of guiding principles:

• Limit the replication of business logic.
• Control the redundancy of data to ensure data integrity and reduce maintenance.
• Insure university-wide (all eight campus) functionality
• Create standard interface boundaries
• Enable local customization at point of service

The Expanded Grade Context Information System had to be developed in compliance with this set of principles. Although constraining, the guidelines provided a good framework from which to design the new information system.

The primary components required to operate the Grade Context Information System are:

Grade Context Information Database

A database (keyed by semester, department, course and section) was defined to include the data elements desired by the EPC. In order to populate this database, a batch program was developed using the data file that drives our grade reporting process (see figure below). This file is derived from our Course History database at the conclusion of a grade processing cycle and includes standard grade information (e.g., grade, summary GPA). The program captures this information and performs the calculations necessary to populate the section context fields (e.g., number of majors, Class GPA, Average Student GPA, etc.) and retrieves additional information from the Schedule of Classes file and the Faculty Database. The program also calculates the total number of GPA grades and the date the file was updated which is used as the Context Effective Date.

General inquiry to the Grade Context Information Database was made available in two modes:

1) for analysis - the database is a relational DB2 table available for ad-hoc inquiry using data retrieval tools (e.g., FOCUS), and,

2) for detail inquiry - a World Wide Web application that retrieves grade distribution information by section number or department/course combination. The Grade Distribution Inquiry application is available by pointing your browser to "http://wwwreg.indiana.edu/GradeDistribution" or from links from the Office of the Registrar homepage (http://www.indiana.edu/~registra) and the Indiana Student Transaction Environment (insite � "http://insite.indiana.edu"), our student direct access service facility on the web.

A set of Grade Distribution reports was also developed using the database as the primary data source. These reports create grade distribution profiles by course, by course department, and by course school. In addition, the grade distribution data is summarized by school, department, and course level. All of these reports are considered public and are shared with each of the academic schools. Note that these reports only include sections with a student enrollment of five or greater; reports for sections with less than this number are produced, but are considered confidential.

Grade Context Record - Transcript

General Design

The creation of the Expanded Grade Context Record, specifically a transcript, was by far the most complicated component of the system. We wanted to leverage existing/developing components of the Transcript Production System (deployed on an OS/390 mainframe - Amdahl Millenium 535 - running CICS 4.1) and concentrate development resources only on pieces that were unique to the Grade Context Record. In addition, the system affected only the Bloomington Campus, not the entire Indiana University system, and the campus wanted printing flexibility (fonts, layout, etc.). These factors led to a decision to develop the user interface and information delivery components using a client/server architecture and client development tools. The central business rules were encapsulated and made accessible to any calling program using the correct protocol. These business rule modules were separated from data overhead activities (I/O) and user interface and presentation modules. A partnership was formed: University Information Technology Services (UITS) was responsible for coding the business rules and the overhead processing modules, while the Office of the Registrar was responsible for developing presentation modules on client platforms. The development principles (stated above) guided this design.

Transcript Production System

The Transcript Production system used at all campuses of Indiana University was originally implemented in 1977. This system served the university well but had been developed in a language (Macro Level COBOL) that is incompatible with current releases of our mainframe transaction system (CICS). In fact, IBM discontinued support for a Macro Level COBOL compatible version of CICS in January of 1997, and thus the need to replace the system (as well as our venerable Records Maintenance system) was acute. In addition, the transcript production system was difficult to manage with multiple programs performing many of the same tasks and/or delivering the same product (e.g., official transcript).

The new Transcript Production system was developed using an Object Oriented approach (see figure below). The main idea was to isolate two primary sets of business rules: 1) The Transcript Build process gathers and calculates the data required for the construction of all transcript types, and 2) Selection Modules where rules defining a transcript type are stored. The Transcript Build program identifies, assembles and marks assembled records as a "block type". For example, a student with a degree will have records that constitute a degree block, a student with transfer work will have a "transfer" block, and so on. Once these records are assembled and marked, they are passed back to the calling Selection Module. Each Selection Module can be invoked by either a standard mainframe CICS transaction or a remote program call (RPC) from a remote client (e.g., Web, Visual Basic, PERL, etc.). The Transcript Build module is the only program that collects information and, as a result, is the only module that interacts with student data files. The Selection Modules call the Transcript Build module, but are not responsible for presenting information to the user. Selection Modules also use a "layout" module to determine format type (one or two column display. (N.B.: Although they are not included in the above figure, additional modules were developed for student identification and transcript eligibility, transcript journalling (logging) and for billing.)

The architecture that supports remote client access to our institutional data is shown in the figure to the right. Client programs invoke stored procedures on the Sybase database server that direct an inquiry to the Sybase network gateway. The network gateway, in turn, performs the necessary protocol/SQL conversion that enables remote access to RPCs on our OS/390 mainframe. RPCs are programs that communicate directly with a network gateway and can call or execute inquiry or update transactions in CICS. Currently, RDBMS (Sybase) security is used to control access to the stored procedures. In addition, users must also have access to the transactions on the mainframe in order to successfully complete an inquiry or update.

Since the Expanded Grade Context Record requires information identical to a traditional transcript, but with the addition of the section grade context information, the application to construct the context record is built upon the basic components of the Transcript Production system. In the figure below, the Transcript Build and the Student I/O modules (as well as the Identification/Eligibility and Billing modules) were developed as part of the base transcript system. A new Grade Context Selection Module was then developed to:

• invoke and select the appropriate information from the transcript build routine

• retrieve and integrate the grade context information for each course section (the routine will accept a request for the student’s entire course history or for a specific semester)

• apply the grade context privacy rule (no context information for sections with less than five students)

• calculate the student’s grade index for the section

Once assembled, the Grade Context information is passed back to the calling program for presentation to the user. A standard grade context inquiry application (CICS) was written for academic staff; a similar query for students, but deployed on the World Wide Web, will be developed in the near future.

The Grade Context Record is printed via a Visual Basic client application developed by the Office of the Registrar. This client application has three primary purposes:

• to identify the student,
• to display the student’s transcript eligibility status (e.g., financial holds), and
• to print the Expanded Grade Context Record .

The client application is responsible for establishing connectivity, handling security, calling the appropriate program, and presenting the information to the user. Like all of the transcript system, it was built using Object Oriented design principles to reduce future maintenance costs and to provide for extensibility in the future. Functions such as printing (developed as a dynamic link library object), database access and security have all been isolated in separate modules. The model also minimizes data redundancy and business rule replication.

Development of the Grade Context Student Grade-Reports was completed using the same design philosophy as the Grade Context Record. We leveraged our existing grade processing system to identify, gather, and format the basic information. We then developed a new batch job that integrated this information with the grade context information. This program must be sequenced to run after the creation of the Grade Context Information database. This information is then formatted to fit onto a specially designed form.

Two types of notifications were developed:

• Final Grade Report that is mailed directly to the student, and
• Academic Department copy of the Student Final Grade Report.

III. Benefits/Costs of the Grade Context Record

A. Benefits

The Bloomington Faculty Council resolution reflected the University’s desire to provide the reader of the final grade report and the transcript with sufficient information to place individual grades awarded within the context of the class. By including not only the index and grade distribution, but also some of the characteristics of the students in the class, the reader is in a better position to evaluate the individual grade awarded to the student and his or her academic performance. Advisers are in a better position to evaluate the progress of their students and to help them in future course selections. The faculty receives feedback concerning the student academic performance and their own grading practices.

The Grade Context Information Database also serves as an information resource and additional management tool for academic deans and campus administration. Those who are interested in the detail information are well served by the Grade Distribution Inquiry application on the World Wide Web.

2. Development Costs

• Development Timeline

The development of the Expanded Grade Context Record was a joint effort of the Office of the Registrar and University Information Technology Services. Since the design relied heavily upon the development of Transcript Production modules, many components were developed early in the project life cycle, while others were delivered later. The timeline for the delivery of the major components is listed below:

April 1994Bloomington Faculty Council passes Grade Indexing policy

Summer 1994 - Fall 1996Educational Policies Committee defines the Expanded Grade Context Record

Winter 1996Expanded Grade Context Specifications Developed

Summer 1997Grade Context Information Database created
Identification/Transcript Eligibility Module completed
Remote Program Call for Identification/Eligibility completed
Transcript Build Module coding completed - testing begins
Grade Context Selection Module coding begins
Grade-Report Coding begins

Fall 1997Student Information Client begins
Grade Context Record Printing Module coding begins
Grade-Report coding completed

Winter 1997Grade Context Selection Module completed
Transcript Build Module completed
Student Information Client completed
Grade Context Record Printing module completed
Grade Context Selection Module completed
Web Grade Distribution application completed

March 1998Grade Context Record project complete
Status reported to Bloomington Faculty Council
Student Grade Reports sent to students

The original intent of the Bloomington Faculty Council was to present a student grade in a more meaningful and understandable manner. The council's concept of a grade index was expanded by the Educational Policies Committee to include a more complete picture of the grade awarded in the individual course. The presentation of such a complete picture to students and those who review the student's academic record is unique in academic record keeping.

The design and implementation of the project, using object oriented design principles and a combination of developers from different units (the computing organization and the user office) proved to be a very effective and productive use of development resources.

The application architecture allowed central business rules to be re-used as needed, and enabled the user to tailor the display of information to the needs of the consumer. Development using these design concepts requires a larger investment during construction of the system. However, this capital cost is offset by the increased flexibility of the user interface and by minimizing the cost future application maintenance.

Finally, the project completely fulfilled our obligation to the faculty of the Bloomington Campus.

V. Expanded Grade Context Record - Development and Implementation Teams

Project Management Team
R. Gerald Pugh, Registrar
Roland Coté, Associate Registrar, Services
Mark McConahay, Associate Registrar, Systems
Elizabeth McCullough, Assistant to the Registrar
Ulrik Knudsen, Assistant Registrar

Application Development Team (UITS)
Christopher Reising, Analyst/Programmer
Gary Weideman, Analyst/Programmer

Client Development Team (Office of the Registrar)
Andy AdamsSr., Analyst/Programmer
Jean TerretSr., Analyst/Programmer

Grade Distribution Reporting Team (Office of the Registrar)
Susan Langsdale, Associate Registrar, Data Management and Administration
Tina Hertel, Statistical Analyst
John Scully, Sr. Analyst/Programmer
P. Bruce Stephenson, Sr. Systems Consultant

Implementation Team (Office of the Registrar)

Pat Wheeler, Associate Registrar, Processing
Gary McCullough, Manager, Departmental Computing
Mary Beth Myers, Sr. Assistant Registrar