Variation in the population, the risk of incorrect acceptance, and tolerable and
expected misstatement affect sample size in the following way:
Desired confidence level
: Direct, as the desired confidence level increases, the
required sample size increases.
Risk of material misstatement
: Direct, as the risk of material misstatement
increases the required sample size increases.
: Inverse, as tolerable misstatement increases, sample
: Direct, an increase in expected misstatement results in
an increase in sample size.
The advantages of MUS over classical variables sampling are as follows:
MUS sampling is generally easier to use than is classical variables
The calculation of sample size in a MUS sample is not based on an
estimate of the standard deviation in the population.
MUS sampling in conjunction with probability-proportional-to-size
selection results in a stratified sample.
Individually significant items are automatically identified.
If no misstatements are expected, MUS will usually result in a smaller
sample size than classical variables sampling.
Using Table 8-5 in the text with a desired confidence level = 95%, tolerable
misstatement = 5% ($15,000
$300,000), and expected misstatement = 2%,
$300,000) sample size is equal to 181 items.
The sampling interval
is $1,657 ($300,000
Using ACL with a desired confidence level = 95%; population = $300,000;
tolerable misstatement = $15,000; and expected misstatement = $6,000; the
sample size is equal to 161 items.
The sampling interval is $1,853.93.
The total projected misstatement for the three misstatements identified is
calculated by first computing the tainting factor as follows:
Not applicable, since book value