If this is for a statistics course assignment, you should read a little bit
about convenience or accidental sampling in a research design text at your
school's library. Nonprobability sampling provides no way to forecast that
each element in your target pop. will be adequately estimated or represented.
To put a good spin on it, you could say the sample consisted of 1000
individuals who either self-selected themselves or their friends. Before
wasting time collecting and tabulating these data, why not do just a little
homework and do it right...the first time. Secondly, press your statistics
professor/advisor about probability vs. nonprobability sampling. How samples
are drawn and selected are the bedrock of survey research. Running
crossbreaks and contrived descriptive statistical analyses mean nothing if the
sample does not represent the population. Good luck.
In article <81a1pf$tvr$[EMAIL PROTECTED]>, [EMAIL PROTECTED] wrote:
>We are college students who are conducting a survey about the types of
>internet browser people use. Our target size is 1000. Please
>participate by filling out the following six questions and pass this to
>your friends.
>
>Your help is much appreciated!
>
>------------------------------------------------------
>Gender (M/F):
>
>Age:
>
>Most frequently used browser (Netscape/IE/AOL/Others):
>
>Type of Internet Service Provider (ADSL/Cable/Dial Up/Network Others):
>
>Average hours of surfing per week:
>
>Browsing experience (in months/years):
>
>------------------------------------------------------
>When you're done, please send this back to [EMAIL PROTECTED]
>
>
>Sent via Deja.com http://www.deja.com/
>Before you buy.
