<p>So here is my question:What type of car do you prefer the most?
a)Economy
b)Standard
c)Sport Utility
d)Pick-Up Truck
e)Undecided</p>
<p>So I know the population is all EPHS students.
So is the parameter, all students that answer to the question in a certain way?
Also, how do I eliminate bias? The bias are students going to PSEO, teachers that fail to adminster my survey, open hours, and absent students!
Also, how do I state my null and alternative hypothesis?</p>
<p>eliminate bias by random sampling..</p>
<p>get a student handbook and only sample the students that u randomly select from it..
also get more answer choices if possible...</p>
<p>theres really no Ho and Ha in this study.. its just percentage calculation..
but if u would want to do a better survey; u can compare students' political views (Democrat, Republican, Libertiarian, Populist, Independent, Green) to their choice of car w/ a chi-square.. thats similar to what i did last yr</p>
<p>WARNING.. i got a 4 on the AP Stat exam.. so im not that good at it.. WARNING</p>
<p>Definitely do a random sample. It sounds like you're trying to do a census, which may just not be possible. A random sample would be best. You might want to think about a stratified sample as well, splitting up groups by grade. Freshman and Sophomores (well, at least some sophomores) are not old enough to drive yet, and that may influence their decision. </p>
<p>Other things to think about when designing the survey:
1) Make sure people are clear about what kinds of cars go into each category. I'm not sure I would know the difference between an economy and a standard. And where would a minivan fit in?<br>
2) Think about reponse bias, and how you want to phrase the question. Is it what car people like the most, what car people would be most likely to buy if money weren't an issue...the list could go on. Make sure that people will have the same interpretation of the question.</p>
<p>It definitely sounds like you want to use some sort of a chi-square test, though which one would depend on how you end up conducting the survey and what kind of results you are looking for.</p>
<p>Your Ho and Ha would be something along the lines of:
Ho: No difference in car preference (based on grade/political view/etc, if you choose to add in another factor)
Ha: There is a difference in car preference (based on...if you choose to add in another factor)</p>
<p>Hope this helps you a little</p>