 in this presentation we will take a look at classical variable sampling within classical variable sampling normal distribution theory is used to evaluate the characteristics of a population based on sample data first a word from our sponsor well actually these are just items that we picked from the youtube shopping affiliate program but that's actually good for you because these aren't things that were just given to us from some large corporation which we don't even use in exchange for us selling them to you these are things that we actually researched purchase and use ourselves here we have a western digital wd elements 20 terabyte usb 3.0 desktop external hard drive we use as part of our backup system noting that if you lower the number of terabytes of storage the price will lower dramatically as well when you're thinking about a backup system you usually think about an online system or an external hard drive system like this or ideally some combination between the two given you some redundancy you can also work directly from an external hard drive like this but there are some drawbacks to doing that one being if you use this as your primary drive you're working from it's no longer a backup drive and you're going to need a backup system possibly another external hard drive and or some kind of cloud backup system and if you're working on something that takes up a lot of short-term memory a lot of ram as you're working on it such as video editing the external hard drive can slow up the system so you might want to come up with some kind of system where you download the project you're working on to your computer to your c drive or possibly to a solid state drive which is a much more expensive external hard drive as you do the work once the work is done then save the project to an external hard drive such as this if you would like a commercial free experience consider subscribing to our website at accounting instruction dot com or accounting instruction dot think of it dot com where we have many different courses you can purchase one at a time or have a subscription model given you access to all the courses courses which are well organized have other resources like excel files and pdf files to download and no commercials auditors generally use classical variable sampling to estimate the size of misstatements so we're going to estimate the size of a misstatement typically with the use of classical variable sampling or when classical variable sampling is applied it's typically used to estimate the size of misstatement sampling distributions are formed by plotting the projected misstatements yielded by an infinite number of audit samples of the same size taken from the same underlying population so this is going to be our standard distribution type curve that we would expect to see if we have a larger population the curve will typically tend to a shape more like this if we have a small if we have a small sample then we'll typically have a curve a curve that will look similar to this obviously the larger sample giving us more precise data around the mean which would be right here a sampling distribution permits us to estimate the probability of observing a single sample result the sample mean is the best estimate of the population mean so here's the population mean and then we can use basically statistical theory a sampling distribution permits us to estimate the probability so we can estimate the probability using our standard deviation methods of observing any single sample result so we can consider what the single sample result would be or what's the probability of a sample result to occur now we're going to take a look at advantages of the use of classical variable sampling we're going to be comparing and contrasting as we consider the advantages you're thinking of comparing it to the monetary unit sampling so this method will result in a smaller sample size than monetary unit sampling when the auditor expects a relatively large number of differences between book and audited value so remember our goal typically we would like to have a smaller sample size because that means less work for us less actual testing for us so this method will result in a smaller sample size so hopefully we get good results with a smaller sample size then mus monetary unit sampling when the auditor expects a relatively large number of differences between book and audited values the techniques will work well for both over statements and under statements as opposed to monetary unit sampling and the selection of zero balances generally does not require specials sample design considerations and again that was one of the problems with the monetary unit sampling not a problem here with the zero balances classical variable sampling disadvantages some of the problems with classical variable sampling when little or no misstatement is expected it does not work well so if we're going to have little to no misstatement this design isn't going to be working well for us to determine sample size the auditor must estimate the standard deviation of the audit differences so we have to include then an estimate and anytime we include an estimate we clue we increase the complication of the likelihood of some type of problem as a result from it if few misstatements are detected in the sample data so we have few misstatements the estimated variance used for the evaluation can underestimate the true variance so once again we have these little these few misstatements and it causes a problem the projection resulting of the misstatements and the related confidence limits are not likely to be reliable under those conditions sampling unit the sampling unit may be a customer account an individual transaction or a line item for example when auditing accounts receivable the auditor may define the sample unit to be the customer's account balance or an individual sale invoice included in the account balance the sample size determination here's the formula for the sample size noting that cc is the confidence coefficient sd is going to be the estimated standard deviation we're going to take the population times times the cc confidence coefficient times the sd standard uh estimated standard deviation divided by the tolerable misstatement minus the estimated misstatement to the power of two