Loading ...
Sorry, an error occurred while loading the content.
 

Re: The accuracy of a survey

Expand Messages
  • Mario Ratnakar Nagathota
    hi joshi, you are right if some people are reluctant to answer/give wrong answers then you get non-response/wrong response bias which makes the estimates very
    Message 1 of 4 , Jul 23 1:09 AM
      hi joshi,
      you are right
      if some people are reluctant to answer/give wrong answers then you get
      non-response/wrong response bias which makes the estimates very
      unreliable, however there are a few procedures like trimmed mean but this
      is used mostly when outliers are present, same with any robust method,
      however you might want to try to take a completely simple random sample,
      completely as in follow the random procedures, its not enough if you think
      that it is random enough.

      the best way would be if you could get grades from the school, ofcourse
      the confidentiality issue is there, but you do not need the names, all you
      need is the GPAs, so if you could convince them that all you need is the
      numbers not names they might be willing to give you the information,
      otherwise there is always a bias, all you can do is try to reduce it as
      much as possible using simple random sampling.....

      good luck...

      --- "Joshi, Adarsh" <adarsh@...> wrote:
      >
      > Dear Colleagues,
      >
      > I am trying to design a survey for college students. One of the critical
      >
      > questions of the survey would be the respective GPAs of the concerned
      > students. However, I am skeptical if people, especially those with GPAs
      > in
      > the lower range, would choose to answer this question. And there's a
      > possibility that many people may answer the questions wrongly. This
      > might
      > lead to a sample bias and the conclusions of the survey might be all
      > wrong.
      > Hence, I seek your help on tackling this problem. Please let me know
      > what
      > you feel about the situation. Are there any books or published articles
      > on
      > the topic that you know of?
      >
      > Awaiting your response,
      > Adarsh Joshi
      >
      >


      =====

      +18595767579 cell

      +18593233553 home





      __________________________________
      Do you Yahoo!?
      Vote for the stars of Yahoo!'s next ad campaign!
      http://advision.webevents.yahoo.com/yahoo/votelifeengine/
    • Saman Riaz
      Hi Adarsh I conducted a similar survey for my stats course in MBA. Trust me it was not all that bad. Though the advantage i had was that my sample market was a
      Message 2 of 4 , Aug 10, 2004
        Hi Adarsh

        I conducted a similar survey for my stats course in
        MBA. Trust me it was not all that bad. Though the
        advantage i had was that my sample market was a bit
        small i mean the Universities that offer MBA
        programme.

        I realized that esp. s i was not concerned with their
        names, they were not very reluctant abt letting me
        know their GPA's, though i had to make them understand
        the sensitivity of the survey.

        Like already suggested, checking it from the
        University is a good post check and again universities
        are now a bit open to it.

        well m suggesting so many things, m still not clear s
        to yr purpose of collecting this information, coz that
        does matter.

        Take care

        Saman
        --- "Joshi, Adarsh" <adarsh@...> wrote:

        >
        > Dear Colleagues,
        >
        > I am trying to design a survey for college students.
        > One of the critical
        > questions of the survey would be the respective GPAs
        > of the concerned
        > students. However, I am skeptical if people,
        > especially those with GPAs in
        > the lower range, would choose to answer this
        > question. And there's a
        > possibility that many people may answer the
        > questions wrongly. This might
        > lead to a sample bias and the conclusions of the
        > survey might be all wrong.
        > Hence, I seek your help on tackling this problem.
        > Please let me know what
        > you feel about the situation. Are there any books or
        > published articles on
        > the topic that you know of?
        >
        > Awaiting your response,
        > Adarsh Joshi
        >
        >




        __________________________________
        Do you Yahoo!?
        Yahoo! Mail - 50x more storage than other providers!
        http://promotions.yahoo.com/new_mail
      Your message has been successfully submitted and would be delivered to recipients shortly.