As the auditor can't test everything, only samples are used for substantive testing
The tests of controls which we have looked at will establish for the auditor how much reliance he can place that the information generated by the system is free from error.
The results of tests of control will therefore determine how much substantive testing is required (tests performed on individual figures in the financial statements).
The amount of substantive testing undertaken can therefore be varied by using different sample sizes.
This is one of the reasons the auditor cannot give absolute assurance over figures in the financial statements – the audit has been carried out on a sample basis.
ISA 530 states
All sampling units should have a chance of selection
Testing the sample gives evidence which helps form a conclusion for the whole population
Either a statistical or a non-statistical approach can be used
This is telling the auditor that they can use a sample to draw conclusions about some aspect of the transactions (e.g. were they authorised?) rather than looking at every transaction.
Material items in the population must be tested. This means that 100% of transactions may be tested if they are all material.
The ISA’s do not require sampling to be used.