Yes/No Survey mean?

[rajuthan]

This is the questions I know how to do it but I dont know how I would figure out the mean for this, which is needed to find the confidence level.

Total Sample 1116
Golfers from sample 177 16%
Non Golfers from sample 939 84%


Total Population
75,700,000
36,900,000
28,300,000
27,800,000
21,900,000
17,100,000
207,700,000
32,941,666.67

How many golfers are there in the United States? A survey of American adults (age 18 and above) asked each whether they had played golf at least once a month during the summer. The responses were 2 = yes and 1 = no. The survey also asked respondents to indicate which of the following household income category they fell into:
1. Under $15,000
2. $15,000 to $24,999
3. $25,000 to $34,999
4. $35,000 to $49,999
5. $50,000 to $75,000
6. Over $75,000
The latest census reveals that the number of American households in each of the income categories is as follows:
1. 75.7 million
2. 36.9 million
3. 28.3 million
4. 27.8 million
5. 21.9 million
6. 17.1 million
(Source: Statistical Abstract of the United States, 2006, Tables 685 and 1238)
*The data file is posted on the website ‘Assignment 2 Data file.xls’.
Required: An analysis of the data
1. Estimate with 95% confidence the total number of golfers.

[terzi]

Assuming that each respondent's answer, coded as 0/1 is a Bernoulli trial, then (Sum Xi)/ n, i.e. the proportion, will behave like the mean.

Now, the estimation should be based in the sampling procedure used, so I assume the income information will be used as sampling weights or as strata information.

[daniel7lu]

Originally Posted by terzi

Assuming that each respondent's answer, coded as 0/1 is a Bernoulli trial, then (Sum Xi)/ n, i.e. the proportion, will behave like the mean.

Now, the estimation should be based in the sampling procedure used, so I assume the income information will be used as sampling weights or as strata information.

Could you please show the process in detail? I need to hand in a assignment in four days, and i really have no idea about Statistics?
and what if the second question is like
2. Estimate with 95% confidence the number of golfers who earn at least $75,000.


Please help me, I am waiting online...

Thanks a lot..

[terzi]

Do you know what a sampling weight is? Or a probability weight? It's just a number that how many persons in the population are represented by each observation in your sample.

If you are using a statistical software you should just estimate a total and specify the weights. Perhaps this is new to you, so maybe in your assignment the economic information is for certain post-stratification procedure. It would depend on the topic you are studying in class

[daniel7lu]

Originally Posted by terzi

Do you know what a sampling weight is? Or a probability weight? It's just a number that how many persons in the population are represented by each observation in your sample.

If you are using a statistical software you should just estimate a total and specify the weights. Perhaps this is new to you, so maybe in your assignment the economic information is for certain post-stratification procedure. It would depend on the topic you are studying in class

Could you check my job for me please?
I post it on
http://talkstats.com/showthread.php?p=26753#post26753
I really appreciate for it...

[nuggles]

The confidence interval around a sampled proportion (p_hat) is given by the formula:

p_hat +/- z_alpha/2 * sqrt[ {p_hat * (1 - p_hat)} / n ]

in your example the 95% CI for golfers would be:

0.16 +/- 1.96 * sqrt [ (0.16)(0.84) / 1116 ]

0.16 +/- 0.022