Research study to assess to what extent data from the Single Application Form could be used to meet the statistical requirements of the June Agricultural Census

Group 1

Strengths revealed by the comparative analysis

Coverage - JAC = population
Expertise - two groups with skills and understandings of the datasets.
Overlap - 89% of area is common
Specificity - JAC = holding level livestock data of use for secondary functions such as disease modelling etc.
Accuracy - payments and inspections mean that SAF may be more accurate for some key items.
Complementarity - can missing items be provided by the other dataset
Willingness to share data is there (providing all data sharing legalities have been resolved) .

Weaknesses revealed by the comparative analysis

Problems of resolving Businesses, Holdings and other entities - what are the definitions?
Mismatch in dates on which the data are captured
Difficulties of exchanging data
Response rate of JAC c. 70% - how effective is imputation
Not all detail needed for claim processing in SAF so not checked is it accurate.
SAF not a full population
SAF is a changing population so annual trends do not represent actual trends within the industry
Livestock numbers too far apart.
How to resolve the differences - which is correct/more correct?
How to even up the data elements.

Opportunities - of combining the two datasets

Efficiency - for farmers and for streamlining processing by SG, possible cost savings, or opportunities for staff to take on additional tasks.
Opportunity to cross check the data - more complete coverage
Savings on IT support if only one dataset
Simplified SAF - if drop all questions that appear in JAC

Threats - of combining the two datasets

Need to meet payment dates - high profile failure if not met. Claim processing must not be affected.
Need to resolve requirements - what is needed and by whom - statutory and highly desirable.
Loss of detail from JAC if just the SAF classes are used
One form could be too large
Confusion from the mixture of needs
Partial integration the worst case - would leave a complex situation
Different requirements (esp in the future) may mean that can't amalgamate - what will the SAF requirements be in the future - uncertain.

Group 2

Strengths revealed by the comparative analysis

Weaknesses revealed by the comparative analysis

Different definitions
Different dates
JAC has 30% non-completion (and therefore imputed values) may be way out for individual holdings but accurate at aggregate level
JAC will the requirements under the 1947 Act ever change, i.e. is there a resistance to change within the JAC.
Too many farm activity codes
Different populations for each
Different responsibilities for those completing each
Lack of common identifiers, i.e. BRNS - need for a common identifiable reference to hang on otherwise real problems integrating from an IT point of view

Opportunities - of combining the two datasets

Rationalise number of farm activity codes as far too many in existence
SAF livestock categories are simplified as little as possible
JAC figures and categories equally useful
Simplification for the farmer
Serious look at forms and questions as asking the same questions
Why fill out 2 forms in May/June when one would do. Should be part of SEARS.
SAF livestock figures could be gathered at holding level
SAF dates and sheep inventory dates are flexible
Streamline and reduce perceived bureaucracy
Comparisons with SAF should allow evaluation and improvement of the JAC imputation, which could then feed into a fresh comparison

Threats - of combining the two datasets

Increased complexity of payment process could lead to system difficulties and lead to payment delays.
Could actually become less efficient and more costly to administer than current.
Introduction of SRDP onto SAF from 20008-9 onwards may make SAF too complicated
How serious is the business to change (is there any scope for change) will either one ( JAC/ SAF) ever do the job of the other

The Scottish Government