1. Sampling versus Census Method

Transcript

Welcome to week 2. statistics for agriculturist. In this lecture we will try to cover Sampling versus Census Methods. In the last one week we have covered the Role for statistics in agriculture, Why do we use statistics in agriculture, How do we use statistics in agriculture, Basics of statistics and Statistical variables.

So now we will keep moving further and understand more on Sampling there are four different types of sampling. So, when you try to do sampling, we have to follow any one of the four Sampling techniques before we understand the four different types of sampling, let us understand what is Sampling. Sampling is nothing but a subset of a target population from which the data is collected. Trying to go around the city collect all the people data that is population and trying to collect only male data is a sample ok. So, the best way to fetch samples is to choose randomly as biased sample would be useless or will try to camouflage lot of information. So, what we are trying to say is you randomly pick people rather than picking them in one sequence. For example if you have a field in that field, you have 10 rows of banana plants planted so, if you try to take the yield of the banana and if you want to use it for some checking try to choose the banana tree randomly from different different columns rather than choosing everything from one column if you are trying to choose banana all across Africa try to choose from every corner of Africa banana and then try to evaluate for their output.

So, randomness is one thing which is a must while taking sampling ok. So, the first sampling technique is random sampling you have a huge population in that huge population you try to extract samples, and these are the things which you extract and then you try to choose three or four of them for measurements. So, these things are called as random samplings. What is Systematic sampling? Systematic sampling is one in three I will select along a particular row or a column that is called as systematic sampling I try to pick 1, 3, 5, 7 and here you can also assume that 1, 3, 5 and 7 can have a particular influence which was not there in the random sampling. So, systematic sampling can be done random sampling can be done. Then Stratified sampling is there and Clustered sampling is there. Stratified sampling is the samples depend on independence and lack of bias of simple random sample only. So, here what we do is, we try to choose good bananas, bad bananas and amongst all the bad bananas or all the good bananas we try to pick one and then do it that is called as stratified.

The next one is clustering you try to form small clusters and try to randomly pick clusters 100 bananas 10 bins and what you do is you completely select that bin that is called as clustered sampling. So, you have to do any one of the sampling techniques to extract information from the population. So, what is Census? Census means collection of data from every member of the population is called as senses. Rather than focusing only male only female, only children, only adult, only transgender, only animal, only dog only peacock, you try to take all the informations together and from there you try to pick up sample census means a collection of data from every member of the population is census. So, generally in an agricultural survey we try to have crop production we try to have livestock production cost of production and production prices then post harvesting losses crop marketing and production methods all these things are agricultural senses. So, from this what do we extract is a population census. So, when we try to compare sample versus census which is good you will always have this question. When we try to do census, it is more accurate, it is primary data, it is conducted by a bigger body, it has less frequent, because it one is time consuming and when you try to take continuous data you will never get two datas in the same thing. When you try to take sample, sample is an extraction from sensors, so, it is less accurate as compared to that of census it can be conducted by anyone it has more frequent and then census. In population census in India, it happens once in 10 years. Whereas, when you do sampling of particular disease it happens once in a week, or once and 10 days. During pandemic it was almost on daily basis. So, this is the difference between Sample and census. Census is time consuming, it is monetary intensive, but you get true data. Sample is done cost effectively and there is a science behind the sampling also I said four, you should have any one of the four to talk about your output.

So, going towards a small story along with the census and sampling. So, here is a situation on one fine day on a coconut island mr x and mr y were trying to talk to each other and they were curious to know how many people in this island are graduates. So, then they figure out what is the population of this they said that there are more than ten thousand people in this island. So, then they thought if I wanted to ask the question how many of you are literate or illiterate, or agriculturist non-agriculturist, it is going to take lot of time and it is going to take lot of money. Then they both decide that why don’t we do random sampling. Then, how to do the random sampling? We have studied four different ways in which they can do. They picked up the random sampling and then they try to understand amongst the population how many of them were graduates that was decided based upon the sample size what they did. So, this makes a big difference between sample and census. So, the verdict from this lecture what do we learn is these two points. Sampling is better when there is a very large population, when you have hundreds of data, thousands of data, millions of data, do sampling. When you have a small set of population then try to do Census. Then try to capture every variable and then do it. People try to do 27 experiments, in the 27 experiments what we do is we collect all the output, try to make a census data and then on this we try to apply statistical techniques to understand about the population.

Thank you.