Page 116 - DCOM509_ADVANCED_AUDITING

Unit 7: Audit Sampling

7. Risk that procedures will not find errors is known as …………………… . Notes
8. …………………… is the probability that the sample results are not representative of the
entire population.

7.3 Meaning and Definition of Sampling Error

Sampling error, generally, refers to a statistical error to which an analyst exposes a model only
because he/she is working with sample data instead of population or census data. However,
using sampling data involves the risk that results found in an analysis might not represent the
results that would be acquired by using data involving the whole population from which the
sample was derived.
The use of a sample comparative to a whole population often becomes necessary for various
practical and/or financial reasons. Although there are possibilities of some differences to occur
between sample analysis results and population analysis results, yet the extent to which these
can differ is not projected to be substantial.

The methods of mitigating sampling error include increasing the size of the sample in addition
to ensuring that the sample adequately represents the whole population.

7.3.1 Types of Sampling Errors

Following are the types of sampling errors:

(a) Random Sampling Errors: In statistics, the sampling error can be found by deducting the
value of a parameter from the value of a statistic. This type of sampling error occurs where
an estimate of quantity of interest, for example an average or percentage, will generally
be subject to sample-to-sample variation. An example of the sampling error in evolution
would be a genetic drift – a change in population’s allele frequencies due to chance. The
bottleneck effect and the founder effect can be considered as an example of random sampling
error.
 The Population Bottleneck: Disasters such as a fire, hurricane, or earthquake reduce the
size of a population drastically, killing many unselectively. The resulting genetic
makeup of the left over population is not representative of the original population,
leading to a bottleneck effect.

 Founder Effect: Founder effect occurs when a new colony is started by a few members
of the original population. This small population size means that the colony may
have:
 Reduced genetic variation from the original population.
 A non-random sample of the genes in the original population.
 Bias Problems: Sampling bias is likely to be a source of sampling errors. The bias
problems lead to sampling errors which have a prevalence to be either positive or
negative. These types of errors are also considered as systematic errors.

(b) Non-sampling Error: Sampling errors can be contrasted to non-sampling errors. The non-
sampling error is a catch-all term for the variations from the true value that are not a
function of the selected sample, counting different systematic errors as well as any random
errors which are not accrued to sampling. Moreover, it is much more difficult to quantify
non-sampling errors than the sampling errors.

LOVELY PROFESSIONAL UNIVERSITY 111