Sample vs Population
Population and Sample are two important terms in the subject ‘Statistics’. In simple terms, population is the largest collection of items that we are interested to study, and the sample is a subset of a population. In other words, sample should represent the population with fewer but sufficient number of items. One population can have several samples with different sizes.
Sample
A sample may consist of two or more items that have been selected out of the population. The lowest possible size for a sample is two and highest would equals to the size of population. There are several ways to select a sample from a population. Theoretically, selecting a ‘random sample’ is the best way to achieve accurate inferences about the population. This type of samples are also called probability samples, as every item in the population has an equal opportunity to be included in a sample.
‘Simple random sampling’ technique is the most famous random sampling technique. In this case, items to be selected for the sample are chosen randomly from the population. Such a sample is called a ‘Simple Random Sample’ or SRS. Another popular technique is ‘systematic sampling’. In this case, the items for a sample are selected based on a particular systematic order.
Example: Every 10th person of the queue is selected for a sample.
In this case, the systematic order is every 10th person. The statistician is free to define this order in a meaningful way. There are other random sampling techniques such as cluster sampling or stratified sampling, and the method of selections are slightly different from the above two.
For practical purposes, non random samples such as convenience samples, judgment samples, snowball samples and purposive samples can be used. More over, items selected to a non random samples are pertaining to a chance. In fact, every item of the population does not have an equal opportunity to be included in a non random samples. These types of samples are also called non probability samples.
Population
Any collection of entities, which are interesting to investigate is simply defined as ‘population.’ Population is the base for samples. Any set of objects in the universe can be a population, based on the declaration of study. Generally, a population should be comparatively large in size and hard to infer some characteristics by considering its items individually. The measurements to be investigated in the population are called parameters. In practice, the parameters are estimated by using statistics which are the relevant measurements of sample.
Example: When estimating the Average Maths Mark of 30 students in a class from the Average Maths marks of 5 students, the parameter is Average Maths Mark of the Class. The statistic is the Average Maths Mark of 5 students.
Sample vs Population
The interesting relationship between the sample and the population is that the population can exist without a sample, but, sample may not exist without population. This argument further proves that a sample depends on a population, but interestingly, most of the population inferences depend on the sample. The main purpose of a sample is to estimate or infer some measurements of a population as accurate as possible. A higher accuracy can be inferred from the overall result obtained from several samples of the same population rather than from one sample. Another important thing to know is, when selecting more than one sample from a population one item can also be included in another sample. This case is known as ‘samples with replacements’. Further more, investing the relevant measurements of the population from a sample and obtaining almost similar output is a golden opportunity to save the cost and time value.
It is crucial to know that, when the sample size increases, the accuracy of the estimate for the population parameter also increases. Logically, in order to have better estimates for the population, sample size should not be too small. Further, random samples also should be considered to have better estimates. Therefore, it is crucial to pay attention on the size and randomness of the sample to be representative to get best estimates for the population.