Since so many days that we have been learning about research we have heard the word “sampling”, “sample” many times. What do these words mean? Why are they so important? Why must sampling be carried out? How can it be carried out? are some of the questions that come to our mind when we hear these words. This blogging experience will help us answer them.
Before we come to how sampling is carried out we need to know what a sample is. The idea of a sample stems from the word population.
Population/Universe – all items in any field of inquiry, Kothari (2005) [see References]
Census inquiry- A complete listing of all items in the population e.g. Population census for a census inquiry it is required that when all items are considered, any possibility of chance or luck is ruled out and hence most accurate results are obtained.
One of the tasks done by the government in a population census is to count the population of India. Here the objective is to count the number of individuals in a nation. Hence listing every individual in a country is possible with the help of large number of people and government support.
- Do you think it will be possible with a single researcher who instead of listing the Indian citizens wants to find their opinion about politicians?
- If your answer is no, why do you think so?
- Could there be a solution to this problem?
While answering these questions keep in mind a person looking to buy a book of 100 stories and browsing through 4 or 5 stories and then buying it. Is there anything similar between the 2 situations? You may write about this on your own blog as a post or comment on this post itself..Look out for the answers in the next post.
Till then happy thinking… :-)
References :
Kothari (2005) - Kothari C.R., (2005) Research Methodology: Methods and Techniques, Second edition , New Age International Publishers, 4835/24, Ansari Rd., Darya ganj, New Delhi - 110 002 (this book is available in our department library)
14 comments:
I do not think it will be possible for a single researcher to find out the opinion of the whole population about polititions. Because the very first thing is opinions about polititions can not be generalized. There might be as much opinions as of individuals. So, as in case of buying a book, we can brows through 4-5 stories out of 100 stories and can generalize and could think about others of them. In short we can generalize. But about opinion, we could not generalize. But definately instead of one researcher, many could be work together in different parts of the countary.
I dont think it is possible for a single researcher to find opionions of people about politicians because as metioned census requires help of large number of people to suppport so if the task is divided among different people to help the researcher to conduct its research will definately help the researcher. As the resercher wants to generalize ie to Indian citizen so he has to select the sample which is representaive then only s/he can genrealise. Similarly the second situation just by browising 4 or 5 book we cannot say all books are similar. Just like the books we cannot say all can be representaive similarly while consucting research just by selecting a population from a part the resulsts cannot be generalise.
I think it will be possible for a single researcher to find out the opinion of the whole population about politicians.
Researcher can randomly select samples on some criteria like age, sex, caste, religion, different states etc.
Then from the results obtained s/he can generalize the results quantitavely for example s/he can conclude sying 80% of the population of India has positive comments towards the politicians or vice versa.
I don't think that it will be possible for a single researcher to find out the opinion about politicians of whole population. Because census inquiry require help of large number of peopleand goverment support. hence in this case,if the researcher divide his/her work among different people then it will be easy for the researcher to collect the opinoion of maximum population and s/he can generalize the opinion about politicians by selecting the sample who will be the representative of the whole population. similarly we cannot say all books are similar just by browsing 4-5 books.
If its about my individual opinion, then I don't think that a single researcher could do it without the help of anybody.
As collecting the data about the opinions of the Indian citizens about the politicians is a very wide task to do.
Selecting a sample for any research is just unseperable part of doing research or collecting data.
In the following situation of a story book, reader who wants to purchase the good story book of 100 stories , before taking any decision about purchasing a book, reader goes through 4 to 5 stories and randomly selects the stories to read, and takes decision about buying of book.Similarly, while doing research it is advisable to do random selection of sample. because taking opinion of every individual citizen is a difficult task and is very time consuming.so the random sampling is good in case selected samples are the representative of the entire population in India.so, the results can be generalized.
so, the possible solution for this problem could be that, the researcher with the help of the some of the selected people can do the data collection about the openions of citizens.the work will get divided among the selected people from different parts of the country.It will definitely help researcher to do the research.
In my opinion this is impossible for the single researcher to find out opinion of the whole citizens about politicians ,even if the task done by the government in a population census is to count the population of india and if while counting we ask them their opinion about indian politicians , some people may not interested , even if people go for voting they dont know why they vote , whome they voted , they have chosen the correct leader or not especialy rural people, many of the rural people are illiterate so they dont know about politics,some people favour the politicians , some are already currupted,so we cant say if they give their opinion will be relavent or not.This is very difficult for single resercher to choose a sample from the population because every individual have their diffrent opinion. such that if a person buying a book of 100 stories and if he/she go through 1 or 2 stories and then decide to bye that book , it may be possible that few stories are same as he/she wanted to buy, but not the all stories .he/she could not generalize his/her opinion about the all stories.a single reseracher can just initiate the research but together people can work on this problem and find out some solutions.
I do not think it is possible for a single researcher to find out the opinion of people about politicians.As mentioned earlier for population census, many groups of people are working in chain from primary level to higher level and they correlate all the data gathered at the end and find out the results.
But in case of personal opinion about politicians researcher can find difficulties.S/he can not personally approach to all the target population means to his samples,because he may have selected the sample randomly. And the views of individuals may vary, as some of them have positive and some of them may have negative experience about politicians.
And s/he can not generalize the result cause there may some threats to his validity.
As person who wants to buy a book of stories browses through 4-5 stories or take a glance at chapters and then he decides whethwr to take or not, researcher can can select sample which should be representative of some area and then s/he can generalize the opinions about politians.
So the possible solution for this problem could be the researcher could take help of some other people (researcher's helper)who are collecting opinion of people and later they would correlate it.
According to me, it is not possible by one researcher to do this work because population for this research is maximum. As population is maximum, census data is also maximum. To gather this maximum information by one person is a very difficult work. Also there is no any chance of random selection. This research is not a work of one person but it required the help by large number of people and researchers.
In case of buying a book. When a book contains 50 to 100 or more than it, that time we read atleast 5 to 6 stories means we just do a random selection between the stories and then we decide whether to buy it or not. so it just a similar case with the research. While doing research it is better to do random selection so the results of the research could be applicable to the entire population.
It is quite difficult for one single researcher to conduct a study for the whole country even if it is by selecting samples. She may face a problem of forming different groups on the basis of age, sex, education, location, caste, etc. from vast population. The sample may be in less quantity, so the result of the study can not be generalised to the whole population. And also the research may consume much time to complete.
Solution - Researcher can appoint a specific number of research assitant to complete the study in less time. She has to take every possible subject as a sample so that the effet can be generalised.
Selecting a book by going through 4-5 pages can be possible but to conduct a study on the opinions about politicians of the whole population is quite difficult and tidius.
i think only one researcher cant find out the opinion of different people of politicians and if he/she is alone then it will need lot of time for research and till that time the opinion might change.
It can be possible that a researcher selects the sample randomly the way a person may buy a book of 100 stories after browsing 4 or 5 stories that is he is randomly selecting the stories and considers it that a book might have good stories, same way researcher can do random selection for research.
Hi. I think that it would be time consuming for a single researcher to carry out this research. It may take years and years to contact each and every individual of the country and even there are chances that by that time the opinions may change as there is a big time gap.So there may be threats in the reliability and validity of the findings. So this would add to the costs, too.
It is thus very tedious for the 'single researcher'. But I also think that it is not 'impossible' for a single researcher.
A possible solution can be that researcher may divide the populations into subgroups- according to
1.) sex- men, women
2.)age - children, adolescents, adults, elderly
3.)caste - hindus, muslims, christians, etc.
4.)job profiles - teacher, government employees, etc.
and so on.....
Then in each of these categories, s/he may select a sample which may be randomly selected within the categories.
Then use this subgroup as a basis for making inferences about the larger populations - the researcher will then be able to generalize about the population based on observations of the sub group.
I dont think so it will be possible for single researcher to find opinions of people about politicians because census requires help of many people.
if task is divided among different people then it will definately help researcher in his/ her research. researcher will select representative sample and will conduct a research but as in 2nd case we can not say anything about the book by refering to just 3-4 stories, here also he/she can not generalize results of research with that representative sample.
It is not possible to find out the opinion of every individual as it is a very very long process and tedious. and one cannot assume that every individual in this country keeps a track on every polotician or may even be aware of what kind of work they do. so if one wants to do a research, s/he may opt the method of selecting samples and generalising, just like the books. however, the validity and reliability will be at risk because one cannot be fully sure that the results they get will be surely and purely their own opinion.
the possible solution may be that the sample they choose is random, i.e. from different parts of the country, they should be grouped according to their age. There can even be polls that can be conducted or surveys that can be done.
Post a Comment