In my study, I chose a probability sampling method to draw a sample X of a population. Before beginning the survey, I try to validate the survey questionnaire before using it.
For this,is true to take sample exp 300 subjects, of the sample X already selected? I apply a sampling method to select these 300 subjects?
Help me to find the solution
I am not clear on your question. Are you saying that you already have 300 subjects to answer the survey? Is this survey already developed and validated, or are you in the beginning stage of development?
No... my question is in the case, of the survey which has not yet started. theoretically I did the sampling plan. and I calculated the sample size of the survey. which I note it here with "A"
Before beginning the survey I have to validate the questionnaire, so I need a little sample (I note here "B") to perform the validation. My question is how to select this sample "B"? I have to apply for this, a sampling methods?
I Selected the sample "B" from the sample "A" !!! I have not found the solution
This is still not clear. Is your survey already developed or are you developing it? How did you determine you needed N = 300? What was based on? I am not clear what type of validity you are trying to test (construct, criterion, etc.)? Please provide more information so that I can answer your question accordingly.
Hi, Ariel Linden
N = 300 is just an exemple. I will try to put the question differently. For the validation of a questionnaire (construct validity, concurrent validity and reliability) the sample selected for this
1-is there any conditions to be followed to select this sample?
2-Is the sample must be representative of the population?
I see. Usually the first iteration of tool is for testing the conceptual underpinning and understanding of the question set. As such, a sample size calculation is not typically performed at this stage. You could likely go through several iterations of the development process of the tool before you're read for "field testing". When you get to that point, you want to maximize your sample size so that your factor analysis (or whatever statistical tools you'll use) will have the best opportunity for differentiating between responders along the spectrum of response and question in the set.
Buenos días. En cuanto a tu pregunta, Zed Bib, la submuestra de una muestra podría ser representativa, pero los estimadores insesgados de la media poblacional y de la varianza poblacional pueden quedar afectados por el proceso de submuestreo (subsampling), y por lo general deben ser corregidos, al menos uno de ellos (the unbiased variance estimator).
Mario, this is an English language service, so responding in another language does not help anyone who does not speak that language. Perhaps you should consider finding a similar service in Spanish
Ariel, thank you for your answer, but I can write perfectly in Spanish and so responding me in my language I would not have difficulties in understanding you. Perhaps you would must answering only to English fluent people if you prefer this language.
Zed, you could select your sample for testing questionnaires as a subsample of the finite population consisting in the first selected sample. The estimators could be corrected to remain unbiased.
Hi,
Good Afternoon,
Helloooooooo, you are having 300 subjects, you want how many sample size for your study? based on that you have to use the sampling method.
Case:1 if you are using simple randome method, then you have to follow
(a). Lottory method or (b). Random number method.
Otherwise, you have to follow the systematic random sampling method.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude