Week 4- Case Study and Two discussions
Comment #1 to JP
In auditing, quantitative measures could be "generic empirical information (e.g. physical metrics) or monetary values (e.g. financial metrics) that signal a certain magnitude of financial effect on the reporting organization" ("Qualitative vs Quantitative"). It is very common that audit firms will develop guidelines to test out the quantitative aspects of a company. This has the advantage that it tends to be more consistent since they are using the guidelines across the board. Another advantage to this approach is that there is less probability of the auditor loosing their independence since they are working strictly with the numbers to come up with their opinion. The disadvantage of this approach would be that since the analysis will be uniform, it won't allow for any special attention to be given to any specific areas.
The individual partner approach to the audit is good because it allows the auditor to look at certain areas that they already know need special attention to. This means that certain areas that already have problems or that are more prone to problems could be looked at closer. However, the disadvantage with this approach would be that it could be easy for auditor bias to slip in. Because of this, I would favor a combination of both approaches. To perform the quantitative tests across the board just to get a good baseline to work with and then take a closer look at the areas that I know need more attention.
A simple random sample is simply defined as “A simple random sample is a subset of a statistical population in which each member of the subset has an equal probability of being chosen.” (Hayes, 2019). This means that in a set of 100 population and a desired sample of 25 each member of the population has a 25% chance of being chosen. This is the most common form of sampling and is used when there is low risk and an auditor needs to test a few accounts to gain assurance.
Systematic sampling is also known as interval sampling where “sample members from a larger population are selected according to a random starting point but with a fixed, periodic interval.” (Hayes, 2019). This means in the same sample as above where we test a population of 100, we pick a random number, say member 7, and test every member that is an interval away from that. If the interval is 10 in the same example, we would test member 7, 17, 27, 37, etc. This is most common with invoice testing when there is moderate to low risk.
Systematic random sampling is when the starting point referred to with systematic sampling is chosen not by humans but by the simple random sampling technique. This is used when there is moderate risk over the same accounts that are tested by systematic sampling, by using random sampling techniques it takes out any bias or human error caused by just systematic sampling.
Haphazard sampling is similar to monkeys throwing darts. Haphazard sampling “though it is nonstatistical in nature, the intent is to approximate a random selection by picking items without any conscious bias, which the auditor intends to be representative of the population” (Bragg, 2019). This is explained as when a population is presented and an induvial randomly picks a sample without rhyme or reason. This is used in situations when there is low risk, as this is not a statistical measure it has to be used with skepticism.
Block sampling is “where a sequential series of selections is made.” (Bragg, 2018). This means that to find a sample of 25 in a population of 100 the auditor will pull all members numbered 50 to 74. This is used in situations when there is moderate to high risk over accounts with large populations. If the auditor is worried that there is a series of inaccurate transactions then this is the one to use.
There are a few situations where you would use multiple techniques. The one I saw at PwC was a mix of systematic random sampling and block sampling where we would determine a large interval and would pick a block of members after the chosen intervals. We used this for an account that had a massive amount of transactions that we had a low level of comfort over and believed there was a number of transactions around the same time frame that were entered incorrectly.