Number of people to be interviewed during an internal audit?

[dsantos24]

How do you determine how many people should be interviewed during an internal department/process audit? Let say the department has 100 employees?

 

[Guest]

You don't. You define the scope and criteria of the audit first. That will give you an indication. Since the fashionable thing to do its audit a process, that will inform whom you speak with, along with the scope and criteria. Are you planning on auditing a process? Departments have many processes.

 

[Mikey324]

As stated above, this department has 100 people. Are the all performing the same process? I would assume all 100 are not working on the same line, for example. if the department consists of multiple processes (blanking, stamping, milling, etc) you would want to audit those processes within the department.
Each would have their own scope. What are the inputs? What are the planned objectives? Are they being accomplished? If not, what was done? Did it work?

 

[Randy]

Do what gives you a warm fuzzy, 1, 5, 10 whatever.... How much time do you have will many times dictate the number.

Also what do you plan to ask? Don't ask stuff that they can't reasonably answer and you'd better know the ballpark answer before you ask the question to start with.

 

[Sidney Vianna]

How do you determine how many people should be interviewed during an internal department/process audit?

The basic expectation is that the sample selected is representative of the population; however, the sample size and composition should also take into account the criticality of the process being assessed. For more information, you can peruse Annex A of ISO 19011 for additional guidance on judgement based and statistical sampling in management system auditing.

Using your example, if these 100 people were doing all the same thing, my judgement tells me that interviewing 10 would be representative, but, assuming these 100 people are responsible for different sub processes, the number would be different.

 

[Guest]

Since this is an internal audit, it's going to depend on the reason for the audit and hence the scope. The "sample" may be one or two people, even in a department of 100. If, for example, you want to find out about a specific issue which happened in processing, that might only have been touched by 3 people. There would be no "sample". Go and talk to all three.

 

[Illek2.0]

If I was unfamiliar with a particular audit or area I would observe the area from afar for a bit. See who is doing what, which departments were interacting etc. It would sometimes give me a better idea of who to approach especially if something seemed off.

 

[malasuerte]

You don't. You define the scope and criteria of the audit first. That will give you an indication. Since the fashionable thing to do its audit a process, that will inform whom you speak with, along with the scope and criteria. Are you planning on auditing a process? Departments have many processes.

This - determine the scope of the audit, then determine who you need in order to get the answers. For example, I may go to a process area and need a process technician, an equipment technician, an engineer, and their manager if my scope is Product/Processing, preventive maintenance, change control, and management review.

 

[Thee Bouyyy]

I think it's worth asking at least 100 people. This will show the difference of opinion.

Disagree. There is no number fixing to be consider. As @malasuerte said in above post is sufficient. Process owner(s) need to be audited as per the defined scope and criteria. Also, audit is not to check the opinion of difference people.

 

[Zero_yield]

The number of people isn't nearly as important as the purpose behind the people you talk to.

For example, let's say you're auditing a manufacturing line. You'd probably want to talk to technicians on the manufacturing floor and probably technicians who perform different operations. You'd probably want to talk to some of the folks who maintain the equipment, who developed the processes, who package the product, who conduct quality investigations when things go wrong, management, etc.

So even if there are 100 employees who are in some way part of the manufacturing line, there might only be a half dozen people who work in equipment maintenance. If you just talked to them, you'd get a very different view of the process than if you talked with 6 manufacturing floor technicians. If you talk with some people from each operations, group, and function within the organization, you'll get a much better view than if you just hold yourself to a hard number of people to talk to.