I. MAJOR THEMES
A. What is Sampling Risk?
B. Statistical vs. Non-Statistical Sampling
C. What is Attributes Sampling?
Sample size goes up -> Sampling distribution tall and skinny – more specific
Smaller sample size -> Shorter fatter distribution
Variance in the population
Confidence level – Greater confidence wider range gonna have to be
-Less confidence in narrower/specific range
Ex) 100% confident ASU student height between 1-100ft
II. CONCEPTS
1. Why sample?
Economically not viable (possible) to test every single transaction in many instances
2. Effect of Data Analytics on sampling.
A. Task—compare shipping data to revenue date could use a 100% sample.
B. Task—inventory observation. Can’t do a 100% sample.
3. Definition of audit sampling – The application of an audit procedure to less than 100
percent of the items for the purpose of evaluating some characteristic of the balance or
class.
4. What is the risk in using sampling?
Getting a nonrepresentative sample – the sample that belongs too way out to the tail of the
distribution that doesn’t look like a population. It can give us type 1,2 errors
5. Nonrepresentative sampling examples:
Even if you do every right, you’ll always get nonrepresentative samples
A. Over/Under sampling one political part in a national poll/race.
B. Over/Under sampling men/women in a survey of university student height.
6. These procedures typically use sampling:
*A.Inspection of tangible assets (e.g., inventory) - If it’s economically viable for us to
test everything, we usually will.
B. Inspection of records or documents
C. Confirmations
D. Reperformance
7. These procedures typically do NOT use sampling:
100% of everything or not applicable
A. Analytical procedures (Looking at FS)
1
CHAPTER 9 AUDIT SAMPLING
B. Scanning
C. Inquiry (Interviewing people)
D. Observation
