 sampling we're recognizing that we're at a 90 to 95 percent statistical sampling which means we're willing to accept that the sample is incorrect or the results from it are incorrect at 5 percent or 10 percent in this example. Tolerable deviation rate, the maximum deviation rate from a prescribed control that the auditor is willing to accept while still considering the control effective. For example if a control is highly important the tolerable deviation may be set to 5 percent but if the if it's only moderately important we may set it to 10 percent. So once again tolerable deviation what is going to be tolerable within the deviation the maximum deviation rate. So with the deviation rate now that's going to be the maximum deviation rate from a prescribed control that the auditor is willing to accept while still considering the control effective. So we're going to say we accept these deviations and we're still going to basically come to the conclusion that the control is effective. Expected population deviation, this is the expected population deviation rate is the rate the auditor expects actually exists in the population. The larger the expected population deviation so the deviation that the basically deviation from what we're testing the larger the sample size must be. So if we're considering that there that if we look at it we consider that there's going to be a larger deviation then we would want to increase the sample size because there's going to be more risk. Not considering any other factors. Attribute sampling population size. Although it would seem so population size is not an important factor in determining sample size for attributes sampling. So let's read that again because this is a bit counter intuitive if we're not a statistician here. The population size is not an important factor in determining the sample size and if you're not a statistician you would probably think that it would be. You'd probably say well how do we know what the sample size would be? Well wouldn't we first need to know how large the population is? In other words if you're talking about basically trying to pull the entire country and determine what the what their opinion is about a certain who they're going to vote for or something like that you would think that you'd have a bigger poll just based on how big the country is and as since the population size being the thing that's going to determine what your sample size would be you need a bigger sample you would think in order to represent a larger population. But that's not the case generally when you're talking about large number type of items. Now if you're talking about small populations that may well be the case. The population size has little or no effect on sample size except when the population is relatively small say less than a thousand. So if you're talking about you want to basically take a sample of something that's less than a thousand and and think about how big the sample size should be well then you you might adjust it based on how big the sample the population is if it's something under than under a thousand you might adjust then your sample to coincide with the population that was 500 600 up to a thousand. But if you're talking about large populations then that no longer is necessarily the case. We do have these relationships below so we're going to have these relationship factors into this table expected population deviation rate. So if we want the expected population deviation rate to go lower now this is the expected population deviation rate then the effect on the sample size would be to decrease. So if we want the expected population deviation rate to increase then we're going to increase the sample size. So these are the factors we're taking into consideration. Then we have the tolerable deviation rate. The tolerable deviation rate if we want it to go lower we're actually going to increase the sample size. If we want the tolerable deviation rate to go higher we're going to decrease the sample size. So in other words the tolerable deviation rate has that inverse relationship and then we have the desired confidence level and that's going to be a straight relationship again. If the desired confidence level we wanted to go lower then the effect on the sample size is to decrease. The desired confident level if we wanted to go higher the effect on the sample size is to increase and then again the population size decrease size only when the population is small say less than something like a thousand.