2 replies to this topic

[nubcke]

Okay, this is the first thing I have no idea how to do. Finding the confidence interval in the problem in the attached image. Help? D;

Attached Files

•  statshelp1.jpg   38.1KB   28 downloads

[luvsmyncis]

You need to find the confidence interval of your LIFE first.

[Scot]

First you need to find p1 and p2.

-Let Women who find female bosses more harsh(26%, 8525) be x1 and 32787 be n1 since it's half of the sample size.
-Let Men who find female bosses more harsh(24%, 7869) be x2 and 32787 is n2.

-p1 is given as x1/n1 and p2 is x2/n2.

p1 = .26
p2 = .24
alpha = 1 - .95 = .05
So alpha/2 = .025

In general, a confidence interval is created using a value plus or minus the Z-score times the Standard Error. In the case of a proportion, the equation is

(p1-p2) +/- Za/2 * SQRT((p1(1-p1)/n1 + p2(1-p2)/n2)

Looking at the cumulative normal distribution chart, the z score for .025 is -1.96, if you don't have the chart, in excel do =NORMSINV(.025).

The standard error is SQRT((.26*.74)/32787 + (.24*.76)/32787) = .00338

So plug all that good shit in and the answer is

(p1-p2) + Za/2 * SQRT((p1(1-p1)/n1 + p2(1-p2)/n2)
.02 + (-1.96 )* .00338 = 0.0134

(p1-p2) - Za/2 * SQRT((p1(1-p1)/n1 + p2(1-p2)/n2)
.02 - (-1.96) * .00338 = 0.0266