+ Reply to Thread
Results 1 to 5 of 5

Thread: How to sample a population?

  1. #1
    Points: 1,375, Level: 20
    Level completed: 75%, Points required for next Level: 25

    Posts
    5
    Thanks
    0
    Thanked 0 Times in 0 Posts

    How to sample a population?




    If I were doing a study on a particular disease in a small city, how would I go about sampling the population? Would it be legal to stand outside a hospital and ask each person randomly if they had that disease? Wouldn't this be sort of rude? I mean, I don't even know this person and I surely wouldn't want a perfect stranger asking me personal question like this. So what other techniques could I use to find out what I'm looking for? Thanks

  2. #2
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts

    Re: How to sample a population?

    Quote Originally Posted by 1in10croak View Post
    Would it be legal to stand outside a hospital and ask each person randomly if they had that disease?
    I don't know about the legality of this, but it'd be an ethical minefield - definitely not something to try!

    This is really more a question about applied medical research rather than statistics... My guess would be, though, that you would usually do something this by working together with the hospital or hospitals in the area. They would need to give you permission, as would your university ethics committee, and then there would usually need to be informed consent from participants too (unless you're only accessing anonymised file info). You'd obviously need to be affiliated with a university or formal research lab or something along those lines. What kind of project are you imagining here? What's your main goal?

  3. #3
    Points: 1,375, Level: 20
    Level completed: 75%, Points required for next Level: 25

    Posts
    5
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: How to sample a population?

    I'm actually not going to be conducting the experiment, but am trying to figure out how I could answer this question:

    Suppose that you are conducting a study on the proportion of individuals that have a particular disease. Describe how you may sample the population in a small city to give accurate results for your study.

    This is just one type of sample question that could be on my final coming up, so I am just preparing for it.

    I'm not too sure how specific he wants me to be with this, but he did hint that we would need to perform a 1-sample t-test to do this. So I guess hypothetically what would I do/or could do?

    I know that I need to find the H0 and Ha. I would probably sample about 100 people. Any suggestions?

    Thanks for replying btw.

  4. #4
    TS Contributor
    Points: 18,889, Level: 87
    Level completed: 8%, Points required for next Level: 461
    CowboyBear's Avatar
    Location
    New Zealand
    Posts
    2,062
    Thanks
    121
    Thanked 427 Times in 328 Posts

    Re: How to sample a population?

    Quote Originally Posted by 1in10croak View Post
    Suppose that you are conducting a study on the proportion of individuals that have a particular disease. Describe how you may sample the population in a small city to give accurate results for your study.
    Ok, in that case the key is that you need an (ideally random) sample of the general population in the city. Perhaps via something like phone book or electoral sampling. If it's the type of disease you can more or less reliably determine via survey you could survey people directly, maybe, or get people who agree to participate to visit a medical centre for diagnostic testing. Someone with more of a medical background might have better idea though.

    I'm not too sure how specific he wants me to be with this, but he did hint that we would need to perform a 1-sample t-test to do this. So I guess hypothetically what would I do/or could do?
    I don't see how a t-test would apply here since we're dealing with proportions, but maybe I'm missing something.

    I know that I need to find the H0 and Ha.
    Not necessarily. It depends on the more specific goals of the study. If you're just wanting to make an estimate of the prevalence of the disease in the city, you would want to use a confidence interval. You only need an H0 and HA if there is actually some specific hypothesis about the disease prevalence that you want to test.

  5. #5
    Points: 1,375, Level: 20
    Level completed: 75%, Points required for next Level: 25

    Posts
    5
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: How to sample a population?


    Quote Originally Posted by CowboyBear View Post

    I don't see how a t-test would apply here since we're dealing with proportions, but maybe I'm missing something.
    I probably misheard my instructor. He probably said z-test. This is probably why I was so confused in the first place. Thanks for the clarification.

+ Reply to Thread

           




Similar Threads

  1. Replies: 3
    Last Post: 11-18-2010, 02:39 PM
  2. How does sample size relate to population?
    By combowaffle in forum Statistics
    Replies: 1
    Last Post: 10-27-2010, 09:06 AM
  3. understanding population from sample
    By musicgold in forum Statistics
    Replies: 0
    Last Post: 06-26-2008, 03:49 PM
  4. Population or sample?
    By Sarah20 in forum Statistics
    Replies: 1
    Last Post: 06-11-2008, 01:21 PM
  5. Using the one sample to the population t test?
    By Angela in forum Statistics
    Replies: 0
    Last Post: 09-02-2007, 03:46 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats