Project Management
PMI Sites
Project Management
• Webinars
• Templates
• Community
• Topics
• Knowledge & Tools
• Events
• Other PMI Sites
• PMI Sites

# Project Management Central

 Statistical Sampling, Attribute and Variable Sampling? How can i choose the right technique to measure the conformance of the deliverable. For example a company manufacturing water bottles 1000 numbers/day has decided to randomly select 100 bottles for inspection to measure the compliance? Does this is a statistical or variable or attribute? Please brief. Posted: Apr 10, 2019 9:14 AM
Lenin,

attribute vs variable sampling is that attributes are present or not (yes if the bottle is heavier than 1000 grams vs no if not) but variables come out as a number in a range (e.g. weight of the bottle in grams). Variable sampling gives you more information (e.g. if there is a bias or a trend) but is more complex.

Both are statistical sampling, but different distributions and methods apply.

There are US military standard covering both MIL-STD-105 for attribute sampling, MIL-STD-414 for variable sampling. I implemented a program to analyse variable sampling for disk flight head testing in 1983, thanks for bringing up that memory.
...
Apr 10, 2019 10:48 AM
Lenin Anbalagan
...
Thankyou Mr.Thomas.
Apr 10, 2019 9:47 AM
...
Lenin,

attribute vs variable sampling is that attributes are present or not (yes if the bottle is heavier than 1000 grams vs no if not) but variables come out as a number in a range (e.g. weight of the bottle in grams). Variable sampling gives you more information (e.g. if there is a bias or a trend) but is more complex.

Both are statistical sampling, but different distributions and methods apply.

There are US military standard covering both MIL-STD-105 for attribute sampling, MIL-STD-414 for variable sampling. I implemented a program to analyse variable sampling for disk flight head testing in 1983, thanks for bringing up that memory.
Thankyou Mr.Thomas.
Lenin,
The first step is determining the attributes that are significant to the quality of your product (size, weight, material composition, etc.). Based on the key attributes, determine what quantifiable variables can you measure to evaluate the attributes. The water bottles are manufactured to some sort of specification, so examine the specification and determine the critical requirements, and then determine how you would measure them. It can be something as simple as a Go/No-Go gauge (does the bottle fit in the hole?) or much more involved such as measuring the composition of the plastic.

The attributes are the qualities of whatever the thing is that you care about. The variables are the quantified measures of the attributes. The statistics are the aggregated results of measuring the variables multiple times to determine the typical values and the variance in the measurements.
...
Apr 11, 2019 11:49 AM
Lenin Anbalagan
...
Thank-you Mr.Keith.
Apr 10, 2019 11:05 AM
...
Lenin,
The first step is determining the attributes that are significant to the quality of your product (size, weight, material composition, etc.). Based on the key attributes, determine what quantifiable variables can you measure to evaluate the attributes. The water bottles are manufactured to some sort of specification, so examine the specification and determine the critical requirements, and then determine how you would measure them. It can be something as simple as a Go/No-Go gauge (does the bottle fit in the hole?) or much more involved such as measuring the composition of the plastic.

The attributes are the qualities of whatever the thing is that you care about. The variables are the quantified measures of the attributes. The statistics are the aggregated results of measuring the variables multiple times to determine the typical values and the variance in the measurements.
Thank-you Mr.Keith.
Thanks for sharing. I can understand that both have some relation to each other.
Attribute Sampling
In attribute sampling, data is in the "attribute" form, and the result either conforms or does not conform.
It is a method of measuring quality that consists of noting the presence or absence of some
characteristic (attribute) in each of the units under consideration.
Attribute sampling checks whether an item is defective or not. It's a yes or no answer.

Variable Sampling
In variable sampling, data is in the "variable" form, and the result is rated on a continuous scale
that measures the degree of conformity.
Variable sampling is about checking "how much", "how good", or "how bad".
You are mixing things. You are talking about sample and the sample size has to be taken with focus on statistics methods. But it is not conformance on deliverable. Be aware that if you are talking about quality you have two components: quality assurance and quality control. You, as project manager, must be define project quality not product quality. Product quality is outside your scope of work. No matter that, when you define project quality in planning phase you define it from product quality.
I am fascinated by the answers on this thread. My brain is greatly expanding, this is the true art of these forums! Thank you all!!
Both sampling methods comes under which PM process?
...
Jul 03, 2020 12:49 PM
Keith Novak
...
I would generally put it under 8. Project Quality Management, although that could also fit under 4.5 Monitor and Control Project Work, which in my mind includes quality.
Jul 03, 2020 11:30 AM