Standard deviation and confidence level: how to interpret and evaluate the results Apologies if this may sounds a dumb question, but the more I'm trying to understand how to interpret and evaluate the results, the more I can't find a proper answer.
I've been trying to understand the confidence level and the standard deviation topics.
Given a single column of values (whether time, votes, points ... you choose), I got the mean, the standard deviation and the 95% confidence level.
I've been doing this in Excel for the sake of practicality

Given the above, the CI lays between 3 (5.74-2.74) and 8.48 (5.74+2.74)
Questions

*

*In looking at the Confidence interval, I assume the interval has to be taken in consideration against the individual results. Is this the case?

*The 95% confidence or 5% alpha, and the results next to it, what does exactly tell? And how am I supposed to use that figure?

*Should I want to find what are the results in my 95%, what should I do?
Very simplistically, and perhaps incorrectly, I did evaluate whether the results are in the proposed range, but here's the catch. The CI proposed once evaluated against the dataset shows me only 14 returns are in the range. 14 out of 23 is 60% and not a 95%. What am I doing wrong?

Or should I just take the average of 21.85 (so 22) of those results and compare against the mean previously calculated?
Thanks for your patience and help.
UPDATE:
Have been trying to get this right somehow via the help provided, but I'm not getting there.
Whether I calculate the CI adding/subtracting the confidence value to the mean (which I understand it's the correct value) or the value of the standard deviation (incorrect), the people value spanning in between that CI is either 7 or 14.
And those are not next to the 95% of the confidence level I have considered.
That's the part that I cannot understand. If this number has to give me confidence that the 95% of the people in the subset has to be in a range, I would expect a counter-proof.
What am I doing wrong?

 A: You are confounding two different notions, lets split these apart.
First there is the normal distribution, which can be used for example to model the heights of people. The distribution is characterized by the mean and the standard deviation. Most people will be close to the mean.
Second there are confidence intervals. Confidence intervals serve to indicate the confidence ("precision") of a statistic or parameter. An example of a statistic or parameter is for example the mean. A confidence interval for the mean would actually indicate how much confidence you have in the calculation of the mean. Note that this mean is calculated using a sample, and the confidence interval then can give information on the possible values of the "real" population mean. The field of statistics in general is about how to generalize from samples to populations.
What you are interested in is the range of lengths which contains 95% of the people, which is the first case. For this one would calculate the interval around the mean of the observations/values by adding and subtracting 1.96 * standard deviation. The Z value for 95% confidence is Z=1.96. The Z value is a property of the normal distribution.
HTH.
