Nottingham Trent University (NTU) has a student population of 26,000 and positions itself as one of the leading 'new' universities in the UK. In its 2004 Institutional Audit, the Quality Assurance Agency stated that: "Broad confidence can be placed in the soundness of the university’s current and future management of the quality of its academic courses and the academic standards of its awards". NTU has been keen to underscore this success with a robust electronic tracking and management system for its academic courses. XCRI's research assistant, Julie Hardman, and I met with Professor Paul Periton, Head of the Centre for Academic Standards and Quality (CASQ) and Francis Lowry, Development Director for IT Operations, to learn how this was to be achieved and discover where XCRI's work might contribute.

Earlier XCRI work had highlighted three archetypal solutions for maintaining a definitive repository of curriculum information:

• populating a relational database with online forms
• populating a relational database with content extracted from word processed documents (London Metropolitan University)
• using the on/off-line form features of Microsoft InfoPath to enter and validate XML content (Fife College)

• Test the mapping of our emerging XML schema to NTU's programme information requirements
• Clarify the value of a standard XML schema to an institution that already had its own database-driven approach

Mapping the XML schema

NTU modelled the approvals business process, identified the information requirements for each step, mapped them to existing information systems and identified gaps to be closed by extending those systems. The result of that exercise was field-by-field breakdown of programme information requirements, which can be mapped to our emerging XML.

The mapping exercise aboves suggests that our emerging XML schema has the flexibility and scope to express most of NTU's programme information as it stands. In most areas there is an obvious one-to-one mapping between the text field used to hold the information in NTU's extended banner student records system, and a curriculum XML element. In some areas,the XML has a finer level of resolution than the single text field, which might necessitate some text processing or compromise about the resolution of data stored. Work continues on a recursive structure for representing the structural composition of curriculum components so the core module programme structure field is flagged with a question mark, but the other question marks highlight areas in which the draft curriculum XML could extend its coverage:

• Details of placement experience - possibly Specification/structure[@type="placement"]
• Investment in employability - possibly Specification/outcome[@type="career"]
• Graduate destinations - possibly Specification/careers
• Other details - possibly Specification/other

Value of a national XML schema

For an institution that has invested in a custom-built database solution to manage the data entry and publication of programme information, it is important to articulate the value of XCRI's work. The contribution comes in two ways:

1. Inspiring reflection about information coverage within the in-house solution by presenting a set of elements devised to accomodate national need
2. Presenting a national XML standard that could be embraced by UCAS for receiving entry profile programme information

Discussion points:

Different programme descriptions for different audiences
NTU's approach centres on a definitive data source (the extended banner student records system) controlled by a designated custodian (CASQ). Whilst most users' information needs would be met directly from the definitive source, in order to market courses effectively it may be necessary to re-write some of the text to meet the needs of the intended audience (potential applicants). Whilst NTU did not intend (initially at least) to bring the re-written fields back into the banner system, it would seem prudent to allow for the concept of multiple versions of XML curriculum elements to satisfy different audience requirements. Schemas like Dublin Core already accomodate different language versions using the xml:lang attribute:

• <dc:subject xml:lang="en">seafood</dc:subject>
• <dc:subject xml:lang="fr">fruits de mer</dc:subject>

• <dc:description ntu:audience="internal">This programme provides an introduction to basic IT systems</dc:description>
• <dc:description ntu:audience="prospectus">Designed in close consultation with leading businesses, this course lays the foundations for a career in the fast-moving world of Information Technology</dc:description>

Relationship between curriculum specifications and offerings
NTU's system intends to create enrolment records from the programme specifications approved to run in each academic year. Most programmes are delivered in a single mode, once a year, based at a particular location, which would imply a one-to-one mapping between curriculum specifications and offerings: each programme specification would have a single offering on to which the student would enrol. A minority of programmes are offered in multiple modes. NTU's student records systems enforces the rule that a live programme can only be offered in a single mode. In such cases the single programme specification would have multiple offerings, one for each mode of study. Although only required by a minority of NTU's programmes, this lends support to XCRI's decision to separate curriculum specification from offering (see partners' feedback on CDM).

Curriculum hierarchy
Many NTU programme specifications have a one-to-one mapping with courses advertised through UCAS. An NTU programme typically has many modules grouped in terms of levels. In some cases, however, a validation event will be conducted at a subject, rather than programme level, and a portfolio of cognate programmes will be approved at a single event. NTU's requirement to capture specifications for subjects, programmes and modules lends further support to XCRI's decision to extend CDM's two-level (programme/course) hierarchy to include multiple, institutionally-specified levels (see partners' feedback on CDM).

Conclusions

The meeting with Paul and Francis from NTU proved extremely useful. It provided:

• a case-study of the online forms approach to maintaining curriculum information;
• an opportunity to test the mapping of our emerging XML to a real-world situation;
• an opportunity to reflect on the value of a national XML schema for an institution that had already committed to substantial in-house development; and
• it raised important discussion points about ways to satisfy multiple audiences, the split between curriculum specifications and offerings, and the need to represent a multi-level curriculum hierarchy.