Systematic Sampling
Systematic Sampling is a probability sampling method in which sample units are selected
from a larger population- with random starting point with fixed periodic interval.
(Linear) systematic sampling
Circular systematic sampling
• It is simple and convenient to use.
• Researchers can create, analyze, and conduct sampling easily
when using this method because of its structure.
• Use systematic sampling when there’s low risk of data manipulation.
•
A sampling technique in which the first unit is selected with the help of random numbers
and the rest units are selected automatically according to some pre-assigned fashion is
known as Systematic sampling. N the number of units in the population are serially
numbered from 1 to N and N=nk.
Requirement:
• Sampling frame - to be updated
• Fixed intervals from the larger population are used to create samples in
systematic sampling.
• A random starting point is selected from the population in systematic
sampling. Then the sample is taken from the fixed intervals of the
population. It is dependent on size.
(Linear) systematic sampling
Procedure: Suppose the N units in the population are serially numbered 1 to N . Suppose
further that N is expressible as a product of two integers n and k , so that N =nk. (n is the
sample size, k is the sampling interval)
To draw a sample size n-
• Select a random number i , 1ik
• Select the unit whose serial number is i
• Select every k th unit thereafter
• Sample contains units with serial numbers i,i+k,i+2k,.....,i+(n-1)k
Example: Let N = 50 and n = 5. So k =10. Suppose the first selected number between 1
and 10 is 4. Then the systematic sample consists of units with the serial number 4, 14, 24,
34, 44.
When the population is periodic, the systematic sampling is not better than the simple
random sampling since then the elements sampled may be too similar to each other.
Example1: A local NGO is seeking to form a systematic sample of 100 volunteers from a
population of 1500, they can select every 15th person in the population to build a sample
systematically.( N=1500, n=100, k=15).
Example 2: If a college wants to form a systematic sample of 50 students for a volunteer
event from a population of 2500 students, then they can select every 50th student from the
entire student population to build a systematic sample.
_____________________________________________________________
Circular systematic sampling
When N nk, the systematic samples are not of same size and consequently sample mean is
not an unbiased estimator of population mean. To overcome this disadvantage of systematic
sampling, the procedure is modified and is known as Circular systematic sampling.
The procedure consists of selecting a random number from 1 to N and then selecting the
unit corresponding to this random number. After this selecting every k th unit ( k Nn) , k is
integer), in a circular manner until a sample of n is obtained.
Illustration : Let N=11 , n= 4 then k=3
The possible samples are
Sl No Samples Sl No Samples
1 1,4,7,10 7 7,10,2,5
2 2.5,8,11 8 8,11,3,6
3 3,6,9,1 9 9,1.4.7
4 4,7,10,2 10 10,2,5,8
5 5,8,11,3 11 11,3,6,9
6 6,9,1,4 -- --
In other words, if i is a random number selected at random from 1 to N , then the circular
systematic samples consists of units with serial number
I + jk if i + jk N
I + jk − N if i+ jk ＞ N for j= 0,1,2,....., (n− 1)
• The systematic sampling spread more evenly over the population hence no large part
will fail to represent the sample. The systematic sampling fails in case of too many
blanks.
• Cost is low. Much less training is needed for surveyors to collect units through
systematic sampling.
• Disadvantages: The process requires : population is available or can be
reasonably approximated.
• The method is less random than simple random sampling
• if an entire population list is available. If that is not possible, then this method is no
longer useful.
• The researcher must be certain that the chosen constant interval between subjects
does not reflect a certain pattern of traits present in the population. If a pattern in
the population exists and it coincides with the interval set by the researcher,
randomness of the sampling technique is compromised. If patterns are present,
avoid systematic sampling.
• If an entire population list is available. If that is not possible, then this
method is no longer useful.
If the population is ordered cyclically or periodically, then the resultant sample will not be
representative.
Illustration: In periodic lists, consider the population consists of alternating boys and girls,
and if one has to choose after an interval of 10 members, then the sample will consist only
of boys or only of girls.