Is the survey biased?

W

WlND

New member
Joined
Dec 12, 2011
Messages
38
In a given population, 12% of likely voters are visible minorities. A survey using simple random sample of 700 telephone numbers finds 5% visible minorities. Is there evidence that the survey is biased?

Ho: the survey is not biased
H1: not H0

\(\displaystyle \frac{0.05-0.12}{\frac{p(1-p)}{n}}\)

\(\displaystyle \frac{0.05-0.12}{\frac{0.12(1-0.12)}{700}}\)=466.67

which can't be right! Where am I going wrong?

Thanks
 
WlND said:
In a given population, 12% of likely voters are visible minorities. A survey using simple random sample of 700 telephone numbers finds 5% visible minorities. Is there evidence that the survey is biased?

Ho: the survey is not biased
H1: not H0

\(\displaystyle \frac{0.05-0.12}{\frac{p(1-p)}{n}}\)

\(\displaystyle \frac{0.05-0.12}{\frac{0.12(1-0.12)}{700}}\)=466.67
Shouldn't there be a square root in this expression?

which can't be right! Where am I going wrong?

Thanks
Click to expand...
The expected number is 0.12×700 = 84
The observed number is 0.05×700 = 35

The standard deviation of a binomial distribution with p=0.12 and n=700 is

\(\displaystyle \sqrt{p\ (1-p)\ n} = 8.6\) people [or 8.6/700 = 1.22%]
 
You must log in or register to reply here.
Top