Attribute Sampling vs Variable Sampling

Please can any of you explain the following with real time example

1)Attribute Sampling vs Variable Sampling

2) precision vs accuracy


In the below URL : Precision and Accuracy are explained very well :

Accuracy : Shooting all bullets anywhere on the target cricle.

Precision :

Shooting all the bullets, at some circle, where all the bullets landed very near (little scatter),  but not on the target circle


Attribute Samplling : IF the height of the person is above 170cm, is eligible to join army, else not eligible.

Yes or No.

Variable Sampling : If the age of the person is between 19 to 30 years, he is eligible for gun shooting, and after 30 years he is not considered as eligible resource and moved to other job.

"how much" or "how bad" or "how good"

 Hi Pkukilla,


Although Sai_murthy has explained most of the concepts correctly, I will disagree with the stated definition of precision. Sai is right that precision is a measure of scatter, but the measurements can be on target as well as off the target. I would like to take this opportunity to explain these concepts in a greater detail.


Precision vs. Accuracy:


PMBOK defines precision as, “the values of repeated measurements are clustered and have little scatter”, while accuracy is defined as, “the measurement value is very close to the true value”.


Consider this; a wire manufacturer is molding copper wires. The wire must be 1.8mm to 2.2mm in diameter. A quality inspector takes 100 measurements of the wire diameter for in two days. Assume that the first 50 measurements (taken on day 1) were in the range of 2.2mm to 2.3mm, while the next 50 measurements (taken on day 2) were in the range of 1.9mm to 2.0mm. What must be concluded regarding the samples taken on these two days?

Day 1: Measurements are precise, but not acceptable since they are not accurate.

Day 2: Measurements are precise and accurate as well.


Assume that the same procedure was repeated on day 3 and further 50 measurements were observed in the range of 1.8mm to 2.2mm. Now what must be concluded regarding the samples taken on day 3?

Day 3: Measurements are less precise, but acceptable since they are within specification limits and are accurate.


In short, precision is a measure of variance in the measurements, regardless whether the measurements were on target or off target. While accuracy is the measure of conformance to specification limits.


Attribute Sampling vs. Variable Sampling


Sai is 100% right with this one, however I would like to elaborate these concepts a little further.


There are two types of data/measurements, ‘variable’ (also called ‘continuous’) and attribute (also called ‘discrete’). Discrete or attribute data can only measured by categories (like yes/no, true/false, pass/fail etc.) or intervals (like absolute rank, educational level, types etc.). Attribute data is always about counting of measurements falling in different categories. Attribute data cannot be further divided, for example if I say I have 10 students who are either taking a math class or a science class, there will be no student who would be taking 50% of the match class and 50% of the science class.


On the other hand, variable or continuous data can be further divided into more classifications and that will still have meaning. For example if I measure temperature for two rooms, 22F and 23F respectively, this does not mean that a temperature of 22.2F or 22.5F cannot be recorded.


Having said that and clearing the concepts of attribute and variable data, answering your question is very simple. If we are taking samples on attribute measurements, we are doing attribute sampling and vice versa.


Another question arises over here, (that you haven’t asked) why we need two different types of sampling techniques. Remember that samples are taken to estimate the population parameters. Statistical formulas for estimating population parameters are different for different probability distributions. To correctly estimate a population parameter, first we have to determine the correct data type, i.e., attribute or variable, and then determine an appropriate probability distribution that fits the sample.


I hope this answers the questions.



Exam Support Team
The PM PrepCast -

 Excellent explanation. Thanks for your time