2.2: DETERMINING THE SAMPLE SIZE

2.2: DETERMINING THE SAMPLE SIZE

In research, the entire population is seldom used because of the cost and time involved. Most researches use only a small representative of a population called the sample. The sample is used to know and/or describe the characteristics of a population.

To determine the ideal sample size for a population, the Slovin’s formula is used:

sample_size1.jpg

where:
n = sample size
N = population size
e = margin of error

note:
* When only a sample (instead of the entire population) is observed or studied, we do not get the actual value but just an estimate of the actual value of the parameter. Therefore, it is important to consider the margin of error when using the sample.

Sample Calculation:
A group of students want to know the age of students in a high school but do not have the resources to survey an entire population of 2,500. If they want to use a sample with a 5% margin of error, what should their sample size be?

    Given:
    N = 2,500
    e = 5% = 0.05

    Required: n = ?

    Solution:
    sample_size2.jpg

10 Responses to “2.2: DETERMINING THE SAMPLE SIZE”

  1. Jenny Says:

    I’m not gonna leave a comment…i just would like to ask something regarding the slovi’s formula…you see, I really am finding it hard to derive a formula when its the margin of error that’s missing. How can i compute for the margin of error having the sample(n) and population (N) as the given data? Can I still use the slovin’s formula for this type of problem? I’d really appreciate it if you mail back soon..please??! Thanks…

  2. riajose Says:

    Jenny, when the margin of error is not given, we usually use 0.10 or 10% as the margin of error. I’m not sure though if that is by convention or just something my professor made up. In my opinion, I think 10% is safe to use since in most scenarios / situations / studies, it is the most cceptable margin of error. Anyway, you (as the researcher) can set (or determine) the margin of error while determining sample size since it is still the early stage of the study. Your set margin of error should be consistent throughout the experiment though (ie in computing for statistical information, explaining results, etc).

    I don’t know of a derivation or substitute for Slovin’s formula which can be used to determine the sample size if the margin of error is not given.

    I hope I was able to help you.

  3. tim Says:

    sloven’s formula is accordingly baseless equation, meaning , there is no available derivation of this equation in any publication.

    if you do believe that this equation is valid, can u pls cite the underlying assumptions when using this equation?

  4. Anna Says:

    Much thanks for your user-friendly presentation.

  5. mark Says:

    what about if the population size is missing or “N” how could you solve , cause im trying to derive multiple equations to get N but the answer seemed wrong please help

  6. Ed See Says:

    My apprehension in using this formula is that it does not distinguish between the parameters we are measuring. For example, the margin of error may be applicable to the income but not to age. The heterogeneity or homogeneity of the different parameters may not be the same.

  7. jc ruse Says:

    can you help me! i don’t know how to derived the population, if the given are the margin of error and the sample. its our assignment and i don’t know how to derived it….
    thanks

  8. 117 Says:

    are there any derivations on how to find either the sample siza or population if they are not given while the me is given??

  9. Sue Says:

    Hi! I am doing my thesis and I want to know if you have literature or any sources to defend that this formula is valid. Please please! :D

  10. hershey Says:

    what could be the possible explanation why there are times that you need to exceed the computed population. example there are 306 teachers, the the computed population using this formula is 173..then suddenly it was round up to 200.what will be the explanation in this matter..
    i’ll be waitingfo you reply…i need so bad… thanks

Leave a Reply