2. Thinking of using IDEA to select a sample?
Have you considered if this is the right
thing to do?
3. Using IDEA to sample efficiently
• Can the client provide data easily?
• Is the data in one report or in several reports?
• Are the reports homogenous?
• Can I easily prove the data population is complete?
• Is the testing more efficient if done in IDEA?
4. Using IDEA to sample Efficiently
• Can the client provide data easily?
• Is the data in one report or in several reports?
• Are the reports homogenous?
• Can I easily prove the data population is complete?
• Is the testing more efficient if done in IDEA?
IF the
answer is yes
to all then use
IDEA
5. Back to basics- sample selection
techniques
• Random Selection
• Haphazard Selection
• Systematic Selection
• Monetary Unit Sampling
• Stratified selection
• Judgemental sampling
6. Where does IDEA have the greatest impact?
• Monetary Unit Sampling
• Stratified sample selection
• Judgemental sampling based on risk
7. Example
Selecting a stratified sample
• Scottish Local Authority
• Data is readily provided by the client, is in excel
and easily imported to IDEA
• Completeness verified by agreement to TB at year
end.
• Year 1 random testing sample of 25.
• Sample generated extremely difficult to trace and
extremely small in value.
8. Example
Selecting a stratified sample
• A key element in picking a judgemental sample is
understanding the population!
• IDEA can help with this almost instantly through
field statistics which give details of:
– maximum
– minimum
– average
– std deviation
• We used this to pick our sample in 2013/14
11. • Focuses the sample on what the auditor defines as
unusual
• Again there is an inherent need to understand the
population
• Design a testing strategy based on the risks you
have identified
Example
Selecting a judgemental sample based on
an area of risk
12. • Areas we typically define as unusual on journals:
- Round sum values
- Values over tolerable error
- Items posted on the weekend/ bank holiday
• The summarisation tool can also be used to
identify irregularities for sampling:
- By amount
- By user
Example
Selecting a judgemental sample based on
an area of risk
13. In conclusion….
• IDEA is a very powerful tool which can lead to
efficiencies in sampling procedures
• We should ensure the use of IDEA complements
our testing strategy
• To ensure we get the most from our sample we
need to understand our clients systems and the
underlying data
• Consider the sample approach you wish to
undertake and use the method which gives
the greatest efficiency.