In: Statistics and Probability
An investigator takes a random sample of n people from a certain population to obtain the proportion p (hitherto unknown to the researcher) of smokers. Calculate the sample size you should take to ensure that the proportion of smokers does not differ from the true p-value by more than 0.01 with a probability of at least 0.95 using:
a) Chebyshev´s inequality
b) Central limit theorem
Compare both values of n obtained, what can you conclude about it?
a)
sample proportion , p̂ =
0.5
sampling error , E = 0.01
Confidence Level , CL= 0.95
alpha = 1-CL = 0.05
Z value = Zα/2 = 2.000 [excel
formula =normsinv(α/2)]
Sample Size,n = (Z / E)² * p̂ * (1-p̂) = (
2.000 / 0.01 ) ² *
0.50 * ( 1 - 0.50 ) =
10000.0
so,Sample Size required=
10000
b)
sample proportion , p̂ =
0.5
sampling error , E = 0.01
Confidence Level , CL= 0.95
alpha = 1-CL = 0.05
Z value = Zα/2 = 1.960 [excel
formula =normsinv(α/2)]
Sample Size,n = (Z / E)² * p̂ * (1-p̂) = (
1.960 / 0.01 ) ² *
0.50 * ( 1 - 0.50 ) =
9603.6
so,Sample Size required=
9604
Sample size for a is greater than b
THANKS
revert back for doubt
please upvote