# Help with figuring out confidence interval level

#### asus30

##### New Member
Hello. My group was conducting a statistics study and we had all our data in an Excel file. We held a pretest (asked survey to nonmembers of the population but had the same characteristics). The results that we got from there was what helped us determine the needed sample size from the actual population for our study. Other group members calculated all those before in the Excel file using Excel functions. We are nearly done with our study on the population itself, but I have a problem. I have forgotten what confidence interval level (95% or 99%) did we use to calculate or sample size. I see in our Excel file that our pretest sample had a standard deviation of 0.1860095. Now, here is where I get confused. My groupmates used the T.INV function in this way: T.INV(0.975, 7) and right below that they calculated this expression

(((standard deviation of pretest which was 0.1860095) x (the result of the T.INV (0.975, 7)) all divided by 0.04) ^2

and the result of the above expression was 120.9088. Given that our study's pop'n is 1186 people, how can I find out what confidence interval level did we use for the study?

I know this question seems very stupid and I apologize for it. Please, I really need help right now. Thank you.

#### rogojel

##### TS Contributor
Hi,
without really understanding the formula, the number 0.975 means that you were using the 95% confidence 0.975=1-0.05/2

regards

#### hlsmith

##### Less is more. Stay pure. Stay poor.
As rogojel wrote, the 0.97 would be associated with a two-sided 95% confidence interval. So does it make sense the 7 represent n-1, so the initial sample just had 8 people?

The formula for confidence intervals is: value of interest plus or minus (critical value (t-value for level of significance and given sample size) * standard error which approximates the standard deviation in the population (standard deviation divided by the square root of the sample size)).

The confidence interval formula for t-distribution is easy found when doing using an internet search engine.