Am targeting all public and private organizations in the country as a population, then from this large and unclear population I will select just those organization who adopt my study’s contracts. as we see it will be difficult to determine the list of the population. Accordingly, How can I determine the population? And which sampling size technique should be use in this case? And how to use it?
To know the population, it is not always should be mentioned by number, such as studying the average of human weight, men and women of age 30 to 40 yrs. You can take one hundred of each males and females of different countries and find that weight. The sample that represent yourvpopulation should be suitable to have a normal distribution curve. Using standard error is so useful in this target. Cheers.
Since this is not a probabilitly sample you can't determine a sample size. Simply do a power calculation and report all of this. Otherwise use a probabilitly sample. See Schaeffer, Mendenhall etc, ELEMENTARY SURVEY SAMPLING for details. Best, David Booth
Murad -
Is this a quantitative survey where you need to estimate ("predict," if a random variable) totals? If so, you will need to establish the population. Even if you look for means, if you don't know your population, then you don't know if your means are for that population, if some of it is hidden.
Besides probability-of-selection-based survey sampling, there is the prediction-based approach, and combined approaches. Here are some references for all three categories:
Thompson, S.K.(2012), Sampling, 3rd ed, John Wiley & Sons.
Cochran, W.G.(1977), Sampling Techniques, 3rd ed, John Wiley & Sons
Blair, E. and Blair, J(2015), Applied Survey Sampling, Sage Publications.
Lohr, S.L(2010), Sampling: Design and Analysis, 2nd ed., Brooks/Cole.
Brewer(2014), "Three controversies in the history of survey sampling," Survey Methodology, Dec 2013 - article Ken Brewer wrote as a consequence of receiving the Waksberg Award:
https://www150.statcan.gc.ca/n1/pub/12-001-x/2013002/article/11883-eng.htm
"The model based (prediction) approach to finite population sampling theory," Institute of Mathematical Statistics Lecture Notes - Monograph Series, Volume 17, 1992, 225-240, Richard M. Royall,
https://www.researchgate.net/publication/254206607_The_model_based_prediction_approach_to_finite_population_sampling_theory/.
The paper is available under Project Euclid, open access:
https://projecteuclid.org/euclid.lnms/1215458849.
Chambers, R, and Clark, R(2012), An Introduction to Model-Based Survey Sampling with Applications, Oxford Statistical Science Series
Valliant, R, Dorfman, A.H., and Royall, R.M.(2000), Finite Population Sampling and Inference: A Prediction Approach, Wiley Series in Probability and Statistics
Brewer, K.R.W.(2002), Combined Survey Sampling Inference: Weighing Basu's Elephants, Arnold: London and Oxford University Press
Särndal, C.-E., Swensson, B., and Wretman, J.(1992), Model Assisted Survey Sampling, Springer-Verlang
In particular, for organizations/establishments, there has been a series of conferences, the ICES (International Conference on Establishment Surveys) which you might want to research.
Here is a project on establishment surveys:
https://www.researchgate.net/project/Cutoff-and-quasi-cutoff-sampling-with-prediction-for-Official-Statistics
Included in that project is the following:
Invited presentation for mathematical statisticians at the US Energy Information Administration:
https://www.researchgate.net/publication/319914742_Quasi-Cutoff_Sampling_and_the_Classical_Ratio_Estimator_-_Application_to_Establishment_Surveys_for_Official_Statistics_at_the_US_Energy_Information_Administration_-_Historical_Development, using prediction
Key to the prediction approach is that you have regressor data on the entire population.
I hope something in the above is of some interest to you.
Cheers - Jim
Murad -
As an example of finding a sample size using a prediction approach, here is a paper regarding sample size requirements for a single population, or subpopulation under a classical ratio estimator (prediction) model:
https://www.researchgate.net/publication/261947825_Projected_Variance_for_the_Model-based_Classical_Ratio_Estimator_Estimating_Sample_Size_Requirements
I compare it to simple random sampling in the chapter on sample size requirements in Cochran, W.G.(1977), Sampling Techniques, 3rd ed, John Wiley & Sons. As with the probability approach, more complex designs have to be accommodated accordingly. But the idea for sample size requirements is always the same:
One first must have bias under control, and then you look at variance. Under the method used, what sample size goes with a given standard deviation to result in a desired standard error? When models are involved, your goal is complicated by different issues, but methodology always provides complications.
Cheers - Jim
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size