Preparing a classical variables sample
Concept Information
Before sampling a set of data, you must stratify the table containing the records, and calculate a statistically valid sample size for each stratum.
The CVS Prepare feature in Analytics calculates the required values for you based on input values you provide.
The importance of calculating a sample size
Calculating an appropriate sample size is critical to the validity of the subsequent sample. If the sample is not valid, or representative, you cannot reliably project the results of audit procedures you perform on the sample to the entire population.
Do not skip calculating a sample size, or guess at a sample size.
Most of the input values you use to calculate sample size are based on your professional judgment. Ensure that you fully understand the implications of the values before relying on the results of sampling in a production environment. Consult audit sampling resources, or an audit sampling specialist, if you are in doubt.
Numeric length limitation
Several internal calculations occur during the preparation stage of classical variables sampling. These calculations support numbers with a maximum length of 17 digits. If the result of any calculation exceeds 17 digits, the result is not included in the output, and you cannot continue with the sampling process.
Note that source data numbers of less than 17 digits can produce internal calculation results that exceed 17 digits.
How precision limits work
When you prepare a classical variables sample you must choose one of the options listed below for Precision Limits.
The option you choose dictates the type of estimation range you generate during the CVS Evaluate stage of the sampling process.
You need to choose an option now, during the preparation stage, because the option you choose is one of the requirements for calculating the sample size.
Precision Limits | For this type of estimate during CVS Evaluate: |
---|---|
Both | A two-sided range with a point estimate and an upper and lower limit |
Upper | A one-sided range with a point estimate and an upper limit |
Lower | A one-sided range with a point estimate and a lower limit |
The "sided-ness" of the range
The "sided-ness" of the range is derived from the normal distribution, or bell curve, that forms the basis of classical variables sampling.
A two-sided range
If you are examining an account that as a whole could be either overstated or understated then typically you are interested in whether misstatement in either direction exceeds the amount of misstatement that you judge is tolerable or acceptable.
You need a two-sided range or estimate:
- The lower limit is an estimate of the maximum amount of overstatement that could exist in the account for the confidence level you specify
- The upper limit is an estimate of the maximum amount of understatement that could exist in the account for the confidence level you specify
A one-sided range
If you are examining an account that as a whole is likely to be overstated, or likely to be understated, you may only be interested in whether misstatement in one direction exceeds the amount of misstatement that you judge is tolerable or acceptable.
You can use a one-sided range or estimate:
- A range with only a lower limit is an estimate of the maximum amount of overstatement that could exist in the account for the confidence level you specify
- A range with only an upper limit is an estimate of the maximum amount of understatement that could exist in the account for the confidence level you specify
Should I use a two-sided or one-sided range?
During the CVS Evaluate stage, using a two-sided range is the more prudent choice. A two-sided range allows you to judge whether an account is fairly stated regardless of the overall direction of misstatement in the account.
The benefit of using a one-sided range is that you can reduce the required sample size, which reduces the required labor and cost of analyzing the sample data. The amount of reduction varies, but typically is less than 50%.
The risk of using a one-sided range is that if you are wrong about the overall direction of misstatement in an account, you could miss a material misstatement in the unexamined direction and make an incorrect judgment about the fairness of an account balance.
How input values affect sample size
Input values affect the sample size calculated by Analytics. You can experiment with different input values in the CVS Prepare dialog box to see how they affect the sample size.
Caution
In a production environment, do not manipulate input values solely to achieve a smaller sample size. Input values should be based on your professional judgment about what is most appropriate for the data being sampled and the audit objective.
Increasing this input value: | Decreases sample size | Increases sample size |
---|---|---|
Number of Strata |
After a certain point, increasing Number of Strata has little or no effect on sample size |
|
Number of Cells |
Increases or decreases sample size depending on the nature of the data, but generally decreases sample size After a certain point, increasing Number of Cells has little or no effect on sample size |
|
Minimum Strata Sample Size |
Increases sample size if the minimum threshold applies in one or more strata |
|
Minimum Total Sample Size |
Increases sample size if the minimum threshold applies |
|
Top certainty stratum cutoff |
Increases or decreases sample size depending on the nature of the data If values in the data are relatively evenly distributed, a higher Top certainty stratum cutoff decreases sample size |
|
Bottom certainty stratum cutoff |
|
Increases or decreases sample size depending on the nature of the data If values in the data are relatively evenly distributed, a higher Bottom certainty stratum cutoff increases sample size Note For negative cutoff values, "higher" means closer to zero (0). |
Confidence Level (%) |
|
|
Monetary Precision |
|
|
Number of Expected Errors | Has no effect on sample size | |
Precision Limits | Both requires a larger sample size than Upper or Lower |
Steps
Note
Do not include the thousands separator, or the percentage sign, when you specify values. These characters prevent the command from running, or cause errors.
- Open the table containing the book values that you intend to sample.
- Select Sampling > Classical Variables Sampling (CVS) > Prepare.
Note
The menu option is disabled if a table is not open.
- On the Main tab, select the book value field from the Book Value drop-down list.
- In the Precision Limits drop-down list, select an appropriate option:
- Both
- Upper
- Lower
Note
The options are explained in detail below.
- Enter the input values to use for preparing the sample design:
- Number of Strata
- Number of Cells
- Minimum Strata Sample Size
- Minimum Total Sample Size
- Top certainty stratum cutoff
- Bottom certainty stratum cutoff
- Confidence Level (%)
- Monetary Precision
- Number of Expected Errors
Note
The input values are explained in detail below.
- Optional. If there are records in the current view that you want to exclude from processing, enter a condition in the If text box, or click If to create an IF statement using the Expression Builder.
Caution
If you specify a conditional expression, an identical conditional expression must be used during both the calculation of the sample size, and the drawing of the sample.
If you use a condition at one stage and not the other, or if the two conditions are not identical, the sampling results will probably not be statistically valid.
- On the Output tab:
- In the To panel, select one of the following:
- Screen displays the results in the Analytics display area
Tip
You can click any linked result value in the display area to drill down to the associated record or records in the source table.
- File saves or appends the results to a text file
The file is saved outside Analytics.
- Screen displays the results in the Analytics display area
- If you selected File as the output type, do one of the following:
- Enter a file name in the Name text box.
- Click Name and enter the file name, or select an existing file in the Save or Save File As dialog box to overwrite or append to the file.
If Analytics prefills a file name, you can accept the prefilled name, or change it.
You can also specify an absolute or relative file path, or navigate to a different folder, to save or append the file in a location other than the project location. For example: C:\Results\Output.txt or Results\Output.txt.
Note
ASCII Text File or Unicode Text file (depending on which edition of Analytics you are using) is the only option for File Type.
- In the To panel, select one of the following:
- Click OK.
The CVS preparation output results are displayed on screen, or saved to a file.
Included in the display is a prefilled version of the CVSSAMPLE command.
Note
The output results are explained in detail below.
Save the CVSSAMPLE command for later use (optional)
As a convenience, you can save the CVSSAMPLE command to use once you have prepared a classical variables sample.
- At the bottom of the CVS Prepare display area, click the CVSSAMPLE link to load the command into the command line.
- Copy the entire command from the command line and save it in an Analytics script.
After you have prepared the sample, you can use either the CVS Sample dialog box, or the CVSSAMPLE command, to draw the sample of records.
Note
If you use the CVSSAMPLE command, you need to add the name of the output table to the command, and a seed value (optional).
For more information, see CVSSAMPLE command.
CVS Prepare dialog box inputs and results
The tables below provide detailed information about the input values in the CVS Prepare dialog box, and the output results.
Main tab – input values
Input values – CVS Prepare dialog box |
Description |
---|---|
Book Value | The field that contains the book values you are auditing. |
Precision Limits |
The type of precision limit to use. Both Select this option if:
Upper Select this option if:
Lower Select this option if:
Caution Select Both if you are not sure which option to select. For more information see How precision limits work. |
Number of Strata |
The number of strata (subgroups) to use for numerically stratifying the data set. The minimum number of strata is 1, and the maximum is 256. If you specify a certainty stratum, it is not included in the Number of Strata. For more information, see Stratification. Note The Number of Strata cannot exceed 50% of the Number of Cells. |
Number of Cells |
The number of cells to use for pre-stratifying the data set. The minimum number of cells is 2, and the maximum is 999. For more information, see Stratification. Note The Number of Cells must be at least twice (2 x) the Number of Strata. |
Minimum Strata Sample Size | The minimum number of records to sample from each stratum. |
Minimum Total Sample Size | The minimum number of records to sample from the entire data set. |
Top certainty stratum cutoff |
Optional. A top certainty stratum cutoff value. Amounts in the Book Value field greater than or equal to the cutoff value are automatically selected and included in the sample. If you do not specify a cutoff value, a default cutoff value is used that ensures no records are included in the top certainty stratum: greater than (>) the maximum amount in the Book Value field
For more information, see The certainty strata. |
Bottom certainty stratum cutoff |
Optional. A bottom certainty stratum cutoff value. Amounts in the Book Value field less than or equal to the cutoff value are automatically selected and included in the sample. If you do not specify a cutoff value, a default cutoff value is used that ensures no records are included in the bottom certainty stratum: less than (<) the minimum amount in the Book Value field
For more information, see The certainty strata. |
Confidence Level (%) |
Your desired confidence level that the resulting sample is representative of the entire population. For example, entering 95 means that you want to be confident that 95% of the time the sample will in fact be representative. Confidence is the complement of “sampling risk”. A 95% confidence level is the same as a 5% sampling risk.
|
Monetary Precision |
The monetary amount that is the difference between the tolerable misstatement and the expected misstatement in the account. For example, if the tolerable misstatement is $29,000, and the expected misstatement is $5,800, enter 23200 (a difference of $23,200). The monetary precision establishes the range of acceptability for an account to be considered fairly stated. |
Number of Expected Errors |
Optional. The minimum number of errors you expect in the sample. This value is not used in any of the CVS calculations. Instead, it is used to trigger a notification in either of these situations:
In either of these situations, the only evaluation method available is mean-per-unit. |
Output results
Output results – CVS Prepare | Description |
---|---|
Monetary Precision | The monetary precision that you specified as an input. |
Confidence Level | The confidence level that you specified as an input. |
Stratum Number |
A sequentially incremented number assigned to each stratum. The certainty strata are also assigned numbers, although they are not displayed on this screen:
|
Strata Boundaries |
The upper boundaries of each stratum, and the bottom and top certainty stratum cutoff values. Book values are assigned to a stratum if they are:
Book values are assigned to the bottom certainty stratum if they are less than or equal to the cutoff value. Book values are assigned to the top certainty stratum if they are greater than or equal to the cutoff value. |
Population Items |
The count of the records in the table, broken down by stratum, including the certainty strata. |
Percent of Count | The percentage of the record count contained in each stratum, including the certainty strata. |
Percent of Amount | The percentage of the total book value contained in each stratum, including the certainty strata. |
Population Value | The total book value of the table, broken down by stratum, including the certainty strata. |
Sample Items |
The total required sample size, broken down by stratum. Includes all items in the certainty strata. |
Associated CVSSAMPLE command |
The command for performing the CVS Sample stage, prefilled with values from the CVS Prepare stage. |