not sure if this is "calculus" but need help pleas

[wato]

(a prob I could use some help on)

I have to test the claim that 30% or less of country a's prison population is made up of political prisoners (that's their claim). Amnesty's sample demonstrates that it is 100 out of 500. My hunch is that the we will fail to reject the claim because it seems obvious that there is sufficient evidence behind it if Amnesty's sample shows that the percentage is even smaller. Of course the country is not going to over estimate how many political prisoners they have because that would make them look bad and they are trying to attract FDI.


Sooooo...

Since our claim for our population is 1/3, that would mean that P= .33

our sample is 1/5, so P^= .2

Z = P^ - P divided by the square root of [P(1-P) divided by N]

Z = -.13 divided by the square root of [.33(.67) divided by 300]

Z = -.13 divided by the square root of .22 divided by 300

Z = -.13 divided by the square root of .00073

Z= -.13 divided by .027

Z = -4.81

Right...?

Now, do I do a left tailed test because I want to find out that it's less than or equal to 1/3?

What do I do next? I look up 4.81 on that z table? It seems too big. Is there a prob with my math?

 

[tkhunny]

Where did you get P = 1/3 = 0.33? It looks like P = 0.3 to me.

Where did you get n = 300? It looks like N = 500 to me.

You have not selected a level of significance. Do that and pick a critical value before you do the arithmetic on the sample.