Funded by the: Jisc e-Learning programme.

Lead Institution: University of Hertfordshire.

Learner Provider Type: Higher Education

Project Duration: January 2012 - March 2013

Key Words: Course Data

Case study tags: course data, process improvement, enterprise architecture (ea), kis, hear, course information, change management, university of hertfordshire

Note: This is an abridged version of this project's final report.  The full version is available here.

 

 

Project Summary

The University of Hertfordshire (UH) recognises the challenges inherent in the increased demands on course data that the sector must respond to, specifically the need to provide up to date, transparent data to prospective students.  The Jisc Course Data project presented an opportunity to address these issues, particularly in relation to course marketing aggregation web sites including the sample aggregator designed by Jisc and the Key Information Set (KIS).

 

This project sought to deliver a system-generated standard XCRI-CAP course feed using University course data for a sample of courses representing a cross-section of UH provision and provide this feed to support the Jisc in its development of a proof of concept’ aggregator and discovery service for prospective students.  This included a range of courses that applicants typically have difficulties finding and the development of standards for the use of cool URIs. 

 

It also sought to examine synergies between XCRI-CAP and work on the KIS and Higher Education Achievement Report (HEAR).  This work was supported by a process of mapping and reviewing policies, processes and data flows for using agreed authoritative sources of course information (Student Record System and Course Marketing Module) to capture course information and enhance the course data life cycle at the University.

 

Throughout the project the team applied ‘XCRI-like’ thinking to the systems involved in the creation and marketing of courses to ensure data was entered once only and used multiple times.  This removed replication and enhanced accuracy.  The application of this thinking has meant the University is now XCRI-ready, as well as improving other processes and data flows beyond the XCRI-CAP remit.  It has also enhanced communication and understanding between the different areas involved in these processes.

 

 

What could have been improved?  What lessons have been learned?

Consideration of possible uses of XCRI-CAP

Having produced the feed, the University sought to consider possible uses for this feed.  One consideration when conducting the project was that no aggregator had picked up the XCRI-CAP format.  We wanted to have a system of moving data around that was robust and that would not only support XCRI-CAP, but also other requirements.  The development of the processes behind the XCRI-CAP feed has also supported the development of the UH website.

 

Towards the end of the programme, we identified three possible aggregators:

• Prospects
• Facebook course listings
• Enhanced Learning Credit Administration System (ELCAS)

 

Prospects has announced they are picking up XCRI-CAP and will allow Universities to provide course listings to them in this format.  Jisc has produced a Facebook tool that allows course listings on Facebook.  This tool is being looked at by the University Marketing Team.  ELCAS currently accepts submissions via an Excel spreadsheet upload but this means that their listings need to be updated often to be kept up to date.  It is certainly not possible to re-upload the spreadsheet as often as it is possible to update an XCRI-CAP feed.  The spreadsheet is, however, produced out of the CMM and so is consistent with our website and other marketing material.

 

In order to identify whether we would make use of any of these aggregators we are considering a series of questions we would ask of any potential implementation of XCRI-CAP:

 

• Do we currently provide data to the aggregator?
• If not, what would be the benefit of providing data to the aggregator?
• If so, what method is currently used to provide data to the aggregator?
• What data are they asking for?
• What differs about this implementation of XCRI-CAP compared to other implementations that have already been made? Consider all differences and additional data.
• What work would be required to be able to meet the needs of this implementation of XCRI-CAP?  Would this work allow us to meet the needs of other implementations already identified?
• How do the benefits of using this aggregator compare to the work required to set up this implementation of XCRI-CAP?
• Are there any other concerns about this implementation of XCRI-CAP and could these be overcome?
• Will the aggregator give the necessary context to the data it has requested?

 

In order for XCRI-CAP to be useful to the University, it needs to provide benefits compared to current practice.  Those benefits might be:

 

• Time saved by not having to manually aggregate/input data to an aggregator on a regular basis.
• Increased accuracy and consistency from having data fed directly from the system.
• Increased recruitment from course information being available in a new aggregator.
• The aggregator paying the University for the administrative and development resource of populating the aggregator.

 

However, these need to be weighed up against the costs of providing the data feeds.  If a specific data request would require significant new work to provide it, then  the cost of that development might outweigh the benefits to the University.  Also, the University might feel that data being asked for is inappropriate to be displayed in the requested location, or that the aggregator’s interpretation of data is misleading.

 

The questions outlined above have been drafted into a form that will manage the decision-making and process for implementation.  If the University determines that the feed to another aggregator would be beneficial, then Marketing, the Web Team and Registry will examine the detailed requirements of the aggregator and its data definitions and scope and cost the work involved.  Following this the completed form will require approval by the Head of Marketing and the Chief Information Officer before implementation takes place.  This is an initial process and it is likely to be refined as good practice develops. At present it has been developed to address the known potential aggregators.

 

We have also considered XCRI-CAP as it might relate to the Key Information Sets (KIS) and the Higher Education Achievement Report (HEAR).  KIS has been discussed in a previous section.  The HEAR has an XML specification related to XCRI-CAP that has been developed by Jisc.  The University has looked at this specification.  Whether this particular XML specification is used or not, it is again likely that similar principles will apply to our HEAR solution and similar methodologies would be employed to feed this out to external providers if required.  The outcomes of this project will be fed into the HEAR project.

 

Revised Self Assessment Framework

An XCRI-CAP Self Assessment Framework was made available to all participants in the project.  All projects completed this as a part of their Implementation Plan.  This tool was particularly useful in helping us assess our progress towards an implementation of XCRI-CAP.  The tool asked a series of questions and rated responses as red, amber and green, with the aim being to be able to answer all of the questions and achieve green in every area.  As the project progressed, we completed the assessment again, using it to help identify gaps and demonstrate progress.  We did this by producing a Self Assessment Framework Gap Analysis which identified our original status, our revised status and our target status.

 

Understanding your Data

The application of XCRI-like thinking has immediately benefited the University.  However the changes it has brought have been difficult to communicate to the University as a whole.  For many staff their activity had not changed, but the use of the outputs of their activity had.  This was evidenced by an increased frequency in queries and meetings to explain ‘what was going on?’

 

A good example of this might be how details of modules end up appearing on the UH website.  The website now takes module details directly from the SRS and publishes them.  Previously, Marketing manually listed modules on the website. 

 

If a member of staff reviewed the modules on the website and thought they were wrong under the old system, they contacted their Marketing Officer, who changed them.  This led to two opportunities for inaccuracy.  Firstly, the change might have been misunderstood by the requesting staff member. Secondly, there are possible errors in the transcribing of modules.  Added to this, the Programme Specification is published on the course web page.  If the requesting member of staff requested an update from their Marketing Officer, they might not also check the Programme Specification and ensure this is also up to date.  The reverse could also happen – a change is made to the programme specification and not picked up quickly by Marketing and therefore be out of date.  Processes were in place to review these, but changes outside of those processes might take time to be communicated.

 

In the new system, a different set of staff need to understand the implications of their local decisions and data entry practices and their work underpins the subsequent re-use of the data through the end to end processes now in place.  For example the naming of modules on the SRS, had mainly been focused around identifying them internally, but now the naming conventions need to take account of the use of this data externally.  Similarly, internal jargon needs to be replaced with generic notation that can also be understood by an external audience. 

 

Similar issues exist around other data sets and have been dealt with on an ad hoc basis, but all the changes to processes in the last year have made the matter more obvious.  The University is taking the opportunity to carry out more work in this area.  The process maps will form a basis for the production of information and training materials.  These will be delivered largely through existing groups and opportunities and cascaded out.  Recent changes in the Higher Education landscape – the KIS, the HEAR and the Quality Assurance Agency now looking at the quality of published material – make this understanding more important than it has ever been.

 

What is a course?

The definition of a course is increasingly complex.  An added layer of complexity we faced were the changes to courses as a result of the requirements for the new Student Record System (SRS)… The programme has been mapped to the SRS course by use of a ‘Validated Programme Code’ being applied to all SRS courses.  This allows to report at programme level where needed. The particular challenge has been to map the Marketing, XCRI-CAP and KIS courses against the SRS course so that these can be created automatically.

 

This kind of mapping exercise is important, as is a process for checking any automated mapping.  Understanding these differences is another part of the awareness raising work.  Understanding the relationship between different types of course was vital to both the production of the XCRI-CAP feed and the KIS.  In order to apply XCRI-like thinking, these different entities had to be linked such that information for one of them would be able to be identified to serve another of them.

 

Programme Specifications

From early in the project, the disadvantages of not generating programme specifications using structured data became apparent.  Some key pieces of data only existed within the programme specification, but as Word documents it was not possible to extract this data efficiently for inclusion in data feeds.  Whilst it is possible to provide a link to the programme specification for this particular information, this can only be done if the aggregator allows such links and a potential student has the motivation to click through.  We also identified the potential disadvantage of the use of disparate terminology between the programme specification document and that used by the aggregator.

 

Immediate Impact

University of Hertfordshire Website

The project to update the University website has benefited from the work being done on both the KIS and XCRI-CAP.  This was a piece of work already in the pipeline with the intention being to move to having a website driven by the University’s authoritative data systems, enhancing both efficiency and accuracy.   As the XCRI-CAP project developed, joint working between central Registry, IT and Marketing teams supported both initiatives.

Processes

The changes that have been made to processes as a result of the implementation of the new SRS were significant.  The opportunity to feed into these, and to apply XCRI type thinking to these, was invaluable.  Good examples might be the addition of awards notes so that certificates can be printed directly from the SRS (previously they had been processed through a database in order to allow this to happen), and the consideration of data and business intelligence right from the initial creation of a course.

 

One significant change to come out of this work was the move to improve the speed with which new courses could be advertised.  Previously new courses were created on the old SRS following validation of the course.  This had less impact on the marketing of new courses as the marketing information was created separately for the website and for hard copy materials.

 

With the move to driving marketing data for both the website and XCRI-CAP from the Student Record System, this issues of disparity were resolved but replaced with another problem – the timeliness of courses being entered into the student system in order that a web page could be created based on this data.  After course validation was too late.  It became apparent during process mapping and discussions around it that marketing needed to begin as early as possible.  The first identified point at which courses could be entered into the SRS was following initial approval by a central committee.  Prior to this the course has no authority to be advertised.  A review of the data collected for presentation against the data needed to create a course was conducted and it seems that most information would be present.  This has been implemented with a view to suggesting additional improvements after the next round of approvals.

 

Relationships

The project brought together a variety of different departments and enhanced relationships between them.  Some of the changes have not just encouraged a reduction in silo working, but have forced such a reduction.   The staff involved have found better ways to work together and developed an understanding of how to help one another.  In particular, XCRI-CAP has brought together central Registry, Marketing and IT to work collaboratively in a focused way which has enhanced other work around it.

 

Efficiencies and accuracy

The enter once, use multiple times approach enhanced both efficiency and accuracy.  In addition, there is potential, if XCRI-CAP is picked up by other aggregators, for a significant time saving here.  It did, however, also create work.  Whilst the work is funded for this year, the ongoing work is not.  This is why assessment of the work, costs and benefits have been added into the process for considering each new XCRI-CAP feed.  The KIS is also a significant additional burden which has been absorbed into existing business provision.

 

Opportunities for error have been reduced through this work.  In the old system, opportunities arose with each point of entry and each intervention.  In the new system, these inaccuracies are likely to occur only at the single point of entry or around the Programme Specification.

 

Future Impact

Information & training

This piece of work will be about promoting the changes the application of XCRI-like thinking has brought about and why they are important.  Examples might be:

• Maintaining process maps clearly stating the stages by which the data moves through the system, broken down into a series of clear procedure notes
• Data mapping alongside this clearly stating who is responsible for which data items and where these data items end up
• A set of presentations bringing this together in an accessible format, possibly to be recorded as an online training session
• A formal guidance note explaining the detailed steps in the process in the in-house SRS style
• A set of quick guides for ‘what to do when …’ for the most common queries around this process

The hope is that by explaining these things in different ways and different formats and with differing levels of detail, people will find something that will suit them and the time they have.

 

Conclusions 

XCRI-CAP

XCRI-CAP is a positive step forward in creating consistent and efficient information feeds both inside and outside of Higher Education institutions.  If it is picked up by aggregators, it is something that should provide benefits to both students and institutions.  In particular, establishing its status as a national standard is important here.

 

The caveat we add to this is that each aggregator must provide sufficient guidance and context to give meaning to the data requested.  This may mean that each aggregator must either write its own specification or adopt the specification of another aggregator.  Whilst the Jisc specification formed an excellent template for this project and the ‘proof of concept’ aggregator, it does not give sufficient guidance for use in practice in other contexts given the scope for different interpretations of a number of the data requirements. .  This is also true of the current KIS data specification, which we assume will similarly develop further over time.  It is also important that potential aggregators understand that universities will only provide feeds if those feeds are going to be accurately and fairly represented in comparing courses and institutions. 

 

Cool URIs

Cool URIs are particularly valuable as a support to XCRI-CAP and similar feeds and returns (e.g. KIS) where data must remain similar over time.  Whilst the full extent of the Linking You toolkit will not be implemented, we have approached the new University website with what might be called cool-thinking and the consistency of links, in particular, is a high priority.

 

Processes

The processes at the University have changed over the last year with the implementation of a new Student Record System.  The value of mapping these processes is not simply in introducing changes, though this was useful.  It is also in helping people to place themselves within these processes.  We will continue to use process mapping (particularly through an Enterprise Architecture approach) to record and map our processes and to support guidance and education around these.

 

Information

We have applied XRCI-type thinking to many of our data processes during the course of this project.  This has been useful and has given us gains in both efficiency and accuracy.  It has, however, been difficult to ensure that all the right people know about these changes which are not always visible.  The University is going to attempt to train and document these processes such that information is available for everyone who needs it.

 

Moving onwards, we will apply more XCRI-cap like thinking to processes as they are reviewed.  There is, for example, a large amount of duplication and manual creation around current validation processes which could benefit from an XCRI-CAP type review. Extending XCRI-CAP thinking more widely is clearly a major and systematic undertaking, rather than a quick fix.  For some teams it may represent a considerable culture change and needs to be planned in from the outset as part of any review of business processes, workflow and data management.

 

Recommendations

Recommendations for the University

• Consider ways of generating programme specifications and implementing validation processes online from underpinning structured data sources for increased accuracy, efficiency and availability.
• Consider ways of generating wider understanding amongst staff about data and processes using process mapping and documentation as well as training.
• Look at business information tools with a view to being able to combine guidance and output in an interactive way.
• Adopt cool URIs
• Use the Self-Assessment Framework for XCRI-CAP and applying XCRI-type thinking.
• Ensure when making changes to your processes that not just the people directly affected but also those indirectly affected are aware of process maps and to keep maps up to date as issues occur.

 

Recommendations for Jisc

• Maintain an overview of XCRI-CAP and the validators that pick it up.
• Continue to work with UCAS in the hope of achieving an XCRI or XCRI-like feed to provide them with data.
• There is a skills gap regarding the understanding of the way aggregators make use of feeds.  Jisc should document how its aggregator interacts with XCRI-CAP feeds so that institutions understand what happens once they have produced them.

Recommendations for the sector

Consider making Key Information Set data available in more useful formats.  Currently each University develops its own system to make this data usable.  It would be more efficient if this was done once at a central level.

 

Further details: email and contact names etc

Project Director         Professor Di Martin

Project Manager        Ms Helen Bennett

Contact email             h.1.bennett@herts.ac.uk

Project Web URL       www.herts.ac.uk/XCRIProject