Transcript Derived Statistics Launch

Monitoring Using HESA Data
Ben Grassby
Two main parts
• A Derived Statistics Monitoring Exercise
• A Web Facility to Prepare For The Above
Purpose of the monitoring exercise
• Monitor the allocation of funds
• Improve data quality and usability of
HEFCE and HESA returns
• Increase our understanding of these
returns
The basics – The derived statistics
exercise
• HEFCE receive data from HESA.
• Re-creations produced and compared to
original returns. (SAS algorithms)
• Select institutions to respond and
reconcile.
– HESA data errors
– HESES errors
– Algorithms
• RAS, HESES, Cost Centres
The basics – The derived statistics
exercise
•
•
•
•
•
HESA data amended (at cost).
Re-creation supersedes original return.
WP Allocations calculated on HESA data.
Grant adjustment affect multiple years.
Typically takes several months to conclude
Outputs from the exercise and web
facility (wf)
• Basics of the derived statistics exercise.
• Introduction to the web facility.
• Typical outputs
–
–
–
–
–
–
–
–
HESES re-creation
Re-creation by cost centre (TRAC)
Cost centre sector norm.
RAS re-creation
Derived statistics for WP
Research degree rates of qualification (wf)
Derived statistics for regional statistics (wf)
Non completion toolkit
• The individualised file.
Selection
•
•
•
•
Based on thresholds
Two distinct parts to the exercise
Thresholds are not points to reconcile to
Separate exercise to other assurance
audit functions. Inform each other.
Typical Calendar
•
•
•
•
•
•
June October December/Jan
February
March
May
Web facility launched
HESA data signed off
Exercise launched
Action plans
Amendments
Interim Grant Adj.
HESES re-creation
• Presents original HESES and re-creation
– FTS (Table 1a) Medical and Dental (1b) SWOUT
(Table 2) PT (Table 3)
– Grant Adjustments and calculation sheets.
• Summary information.
• Differences to be investigated using
troubleshooting guidance + individualised file
• Area of difference
– HESA
– HESES
– Algorithms
Amendment before submission
Investigation and preparation
Problems of fit can be informed
HESES re-creation based on cost
sector norm.
• Same outputs as the main HESES05 recreation
• activity allocated to cost centres based
upon the member of staff most directly
associated with the subject.
• We use a norm mapping of activity rather
than an individual institution’s mapping.
RAS re-creation
• Re-create fundable home and EC fee.
paying tables for comparison.
• Re-create the supervision funding report.
• Individualised file.
• Summary reports.
• Approximations including mapping of UofA
to subjects.
WP allocations
• HESA2005-06 informs 07-08 allocations
– Widening access for students from
disadvantaged backgrounds. (Postcodes)
– Improving retention for FT students (age and
entry qualifications)
– Widening access for disabled students (DSA)
Derived statistics web facility
• Normally released in Summer before
HESA data submitted
• Chance to ‘dry run the exercise’ and make
corrections.
• Non-mandatory
• Non use often leads to selection for main
exercise.
HESES05 re-creation as an example of
advantages to web facility
Web facility is not mandatory.
Data submitted is anonymous.
Can be re-submitted many times (quickly).
Allows identification of HESES errors far in
advance of exercise.
No funding adjustments.
Previous exercises, correlation between
selection and non-use.
Regional Statistics
• Web facility opportunity to verify
underlying data.
• HESA 2006-07 student data produces:
– Franchised data
– Campus data
– Distance learning data
– Provision by location
• Individualised file
Non-completion toolkit
•
•
•
•
Excel pivot table.
Select influential forecast variables.
Estimates of non-completion created.
Only an aide, institutions should use own
judgement.
• Errors in HESA data or small sample sizes
need to be considered.
Research Degree Rates of
Qualification
•
•
•
•
Time taken to obtain qualification
Based on HESA student data
Links across multiple years
99/00 FT cohort to be published in 2007
Using the web facility
• https://extranet.hedata.ac.uk
• Essentially upload your HESA file and
download the results.
• Data does not need to have passed all of
HESA’s validation
• Retain Leading zero’s! (e.g. RECID)
• Can process ZIP files