1 - 8 of 8 Posts

#### achistikbenny

· Registered
Joined
·
234 Posts
Discussion Starter · ·
A large study of serum folate levels in women aged 16 - 45years reveals that this parameter is normally distributed, with a mean of 5.0 ng/ml and a standard deviation of 0.5 ng/ml. According to the study results, 95% of serum folate observations in these patients will lie between the following limits.

A- 4.0 and 6.0 ng/ml
B- 4.0 and 5.5 ng/ml
C- 4.5 and 5.5 ng/ml
D- 3.5 and 6.0 ng/ml
E- 3.5 and 6.5 ng/ml

#### drmodi44

· Registered
Joined
·
487 Posts
A

The correct answer is A- 4.0 and 6.0 ng/ml...

1SD is .5 ng/ml....so 2SD = 1 ng/ml....
95% CI = mean +/- 2SD....so....5 +/- 1 = 4 - 6 ng/ml....

#### rasheed

· Registered
Joined
·
741 Posts
Correct .. but

The correct answer is A- 4.0 and 6.0 ng/ml...

1SD is .5 ng/ml....so 2SD = 1 ng/ml....
95% CI = mean +/- 2SD....so....5 +/- 1 = 4 - 6 ng/ml....
Your answer is correct. But they don't call it Confidence Interval. This is just 95% of a normally distributed population. Which is as you said mean +/- 2 standard deviations, 95% Confidence Interval of the mean on the other hand is calculated by this formula
mean +/- 2 Z scores X SD/N

#### usmle100

· Registered
Joined
·
18 Posts
1 sd - 68
2 sd - 95
3sd - 99.7

so here its 95..ie 2sd...so 1 on each side of 5

which will be 5-1= 4
and 5+1=6

#### achistikbenny

· Registered
Joined
·
234 Posts
Discussion Starter · ·
You are all correct the ans is A. thanks for the explanations. I was actually confused and didn't know how to go about it. my fear has always been biostatistics but i thank God for this forum. the forum is actually helping me get thru it .thanks again

#### bebix

· Registered
Joined
·
1,357 Posts
Your answer is correct. But they don't call it Confidence Interval. This is just 95% of a normally distributed population. Which is as you said mean +/- 2 standard deviations, 95% Confidence Interval of the mean on the other hand is calculated by this formula
mean +/- 2 Z scores X SD/N
Actually, the formula for a 95% CI is

mean + or - [Z score x sd/sqrt(N)]...and Z is almost 2 (95% confidence Z score = 1.96)...so, you dont have to multiple Z score by two...and dont forget the square root in the denominator.

#### struggle

· Registered
Joined
·
324 Posts
Hi bebix..
Can you please explain me what value should we take for 'N'..??
Because here they are saying large sample??..

#### struggle

· Registered
Joined
·
324 Posts
Hi bebix..
Can you please explain me what value should we take for 'N'..??
Because here they are saying large sample??..
Oh yeah i got it..I didnt looked the question properly..they are asking for 95% population..
:redcheeks;

1 - 8 of 8 Posts