a) The auditor can use each of the following methods of sample selection when carrying out his audit assignment as follows
i) Stratified sampling
Stratification is used to reduce the degree of variation between items within a population and enables auditor to direct attention to those areas where there is the greatest potential monetary error. It has the effect of reducing sample sizes. Where auditors are concerned with discovery of overstatement errors and consider that the largest monetary errors are likely to occur in the largest individual populations, they will stratify population by value They may direct major part of their audit effort to those items with highest individual value.
Example;
Value Number of items in stratum Value of Stratum Test size
Above Ksh.1m 10 Shs.36 million 10
Shs.250,000 to Shs.1m 150 Shs.95 million 50
Shs.50,000 to Shs.250,000 1000 Shs.93 million 50
Shs.10,000 to Shs.50,000 3000 Shs.60 million 35
Less than Shs.10,000 7000 Shs.15 million 15
ii) Cluster Sampling
It is useful when data is maintained in clusters (groups or bunches) as wage records are kept in weeks or safes in months. The idea is to select a cluster randomly and then to examine all the items in the cluster. The problem with the method is that the sample may not be representative.
iii) Horizontal sampling
It is simply choosing items subjectively but avoiding bias. Bias might come in by tendering to favour items in a particular location or in an accessible file or conversely in picking items because they appear unusual. The method is acceptable for non-statistical sampling but is insufficiently rigorous for statistical sampling.
iv) Block sampling
It is simply choosing at random one block of items e.g. all June invoices. It consists of selecting a number of adjacent transactions or items e.g. all sales in a particular week or all credit customers with a name beginning with a particular letter. It is not an appropriate selection method since populations might be expected to be structured in such a way that items in a sequence have similar characteristics to each other but different characteristics to items elsewhere in the population.