VARIABLE SAMPLING VARIABLE SAMPLING OR QUANTITATIVE ESTIMATION IS A STATISTICAL TECHNIQUE USED TO ESTIMATE THE DOLLAR AMOUNT OF AN ACCOUNT BALANCE. WHEN VARIABLE SAMPLING IS USED THE RESULT IS USUALLY STATED IN TERMS OF A RANGE OF NUMBERS (E.G., $2,100,000 TO $2,700,000) WHICH THE AUDITOR IS X% (E.G., 95%) CONFIDENT INCLUDES THE TRUE VALUE. UNLIKE ATTRIBUTE SAMPLING WHERE THE CONCERN IS WITH THE UPPER PRECISION LIMIT, VARIABLE ESTIMATION USES BOTH AN UPPER AND LOWER PRECISION LIMIT, I.E., AN ACCOUNT BALANCE CAN BE EITHER UNDERSTATED OR OVERSTATED. THERE ARE THREE BASIC VARIABLE SAMPLING PLANS AVAILABLE TO THE AUDITOR. THESE THREE METHODS ARE DEFINED BELOW. HOWEVER, WE WILL NOT USE THEM IN THIS COURSE. THE METHOD MOST OFTEN USED BY AUDITORS IN ESTIMATING DOLLAR AMOUNTS IS REFERRED TO AS PROBABILITY PROPORTIONAL TO SIZE (PPS) OR MONETARY UNIT (MUS) SAMPLING. PPS SAMPLING IS COVERED THIS COURSE. THE STATISTICAL ACCOUNTING ESTIMATION MODELS ARE: 1. UNSTRATIFIED MEAN PER UNIT (UNSTRATIFIED MPU) AFTER A SAMPLE IS SELECTED AND A VALUE IS DETERMINED FOR EACH SAMPLE ITEM, THE SAMPLE MEAN OF SAMPLE VALUES IS MULTIPLIED BY THE NUMBER OF ITEMS IN THE POPULATION TO PRODUCE AN ESTIMATE OF THE TOTAL DOLLAR VALUE OF THE SAMPLED POPULATION. THE REQUIRED STATISTICAL TECHNIQUES REQUIRES LARGE SAMPLE SIZES BECAUSE OF THE WIDE POSSIBLE RANGE OF VALUES. IN OTHER WORDS, THE USE OF THIS METHOD IS COSTLY BECAUSE OF THE REQUIRED LARGE SAMPLE SIZE. IN ORDER TO REDUCE THE SAMPLE SIZE, THE AUDITOR MAY USE THE STRATIFIED MEAN PER UNIT MODEL. 2. STRATIFIED MEAN PER UNIT (STRATIFIED MPU) AS NOTED ABOVE, WHEN A POPULATION IS HIGHLY VARIABLE (HAS A LARGE STANDARD DEVIATION), UNSTRATIFIED MPU MAY PRODUCE VERY LARGE SAMPLE SIZES. IN MANY SUCH CASES, STRATIFICATION Of THE POPULATION PRODUCES AN ESTIMATE THAT HAS A DESIRED LEVEL OF PRECISION WITH A REDUCED SAMPLE SIZE. STRATIFIED MPU MAY BE APPLIED TO POPULATIONS WHERE NO MONETARY MISSTATEMENT S ARE EXPECTED, OR WHERE A MODERATE RATE OR AMOUNT OF MONETARY MISSTATEMENTS ARE EXPECTED.