confidence limits

Prep for USMLE Forum

» Log in | Register

Forum

Step 1

Match

Prep4USMLE Forum » USMLE Step 1 Forum » Behavioral Science Forum

confidence limits

Author

15 Posts

msyamp
Forum Fanatic

Topics: 60
Posts: 1,462

01/03/06 - 11:02 PM

suppose the confidence limits of the mean of a variable are 14.55 and 14.65 the limits are:________________ than 14.20 and 15.00

1. less pricise but have higher confidence
2. more precise and lower confidence
3.less precise and lless conf
4.more precise and more conf
intermediate as conf level is not told

___________________
If you think you can You can! If you think you cant you are right again!!

Geroo
Forum Guru

Topics: 114
Posts: 799

01/03/06 - 11:14 PM

1. less pricise but have higher confidence

msyamp
Forum Fanatic

Topics: 60
Posts: 1,462

01/03/06 - 11:16 PM

how?

___________________
If you think you can You can! If you think you cant you are right again!!

Geroo
Forum Guru

Topics: 114
Posts: 799

01/03/06 - 11:29 PM

sorry lower confidence interval and hence more precise.answe is 2

msyamp
Forum Fanatic

Topics: 60
Posts: 1,462

01/03/06 - 11:41 PM

explain your answer.how can it be lower conf interval?

___________________
If you think you can You can! If you think you cant you are right again!!

robin082006
Forum Hero
avatar

Topics: 471
Posts: 5,125

01/04/06 - 12:18 AM

Ans is 2.

We see that the confidence interval of 14.55---14.65 is narrower that that of 14.20---15.00 (14.65-14.55=0.10 compared with 15.00-14.20=0.80). This means that the standard error of the former is smaller that that of the latter. Thus, the former confidence interval is more precise. (recall the formula CONFIDENCE INTERVAL = MEAN +/- ALPHA*STANDARD ERROR

Because the first confidence has lower values, maximum is 14.65 in comparison with 15.00 of the second, it is said that the first confidence interval is lower confidence interval.

Hope it clear.

___________________
The Key to Succeed is Patience.

msyamp
Forum Fanatic

Topics: 60
Posts: 1,462

01/04/06 - 01:09 AM

hey the values are not the amount of confidence but they are the range of mean of the variable measured.

___________________
If you think you can You can! If you think you cant you are right again!!

robin082006
Forum Hero
avatar

Topics: 471
Posts: 5,125

01/04/06 - 01:20 AM

yeah we know that the confidence interval is the range, not values, that a proportion (e.g. 95%) of samples lie somewhere in this range.

the smaller the standard error, the more precise the confidence.

___________________
The Key to Succeed is Patience.

satyaking
Forum Junior

Topics: 5
Posts: 58

01/06/06 - 09:11 PM

yes robin is correct,the answer is 2

vamsi2004
Forum Elite

Topics: 7
Posts: 144

08/10/06 - 05:49 PM

#10

msyamp what is the correct ans? plz explain it..

study_ing
Forum Fanatic
avatar

Topics: 180
Posts: 2,511

08/10/06 - 06:20 PM

#11

it is 2 for sure..and robins explanation applies.

___________________
If you yourself are at peace, then there is at least some peace in the world.

equinoxe

Topics: 15
Posts: 77

09/10/06 - 06:32 AM

#12

Because the first confidence has lower values, maximum is 14.65 in comparison with 15.00 of the second, it is said that the first confidence interval is lower confidence interval.

I understand that is more precise, as directed by SEM - but _lower confidence_ means lower value for maximal value of the confidence interval?

I imagined confidence as being a measure of how wide the interval of confidence is - 0.10 compared to 0.80 as mentioned. It this not true?

Thank you for your answers, they help a lot.

___________________
waiting gives the devil time

jsknetsdude
Forum Junior

Topics: 4
Posts: 63

09/17/06 - 02:35 AM

#13

Confidence INterval = Mean +- Z-ScroeXStandard error .. Aussuming Mean and Standard deviations are the same for both .. Well if the interval is large tha means that i am more certain (i.e 99%) that ther real mean would lie some where in b/w so i will have greater confindence .. but it would be less preceisce on the other hand if lnterval is smaill then i would be less certain (i.e 95%) the the real mean would like somw where in b/w so i will have less confindence .. but since the interval is small (less, wide, less scattered) it would be grouped more closer together therefore more precise

___________________
JSK .. O Lord !! Increase my capacity to Absorb knowledge ..

Slacker
Forum Elite

Topics: 29
Posts: 141

09/21/06 - 09:03 AM

#14

Nice explaination jsknetsdude. Really helpfull!!! sticking out tongue

MRMAVERICK
Forum Guru
avatar

Topics: 40
Posts: 713

10/17/06 - 02:37 PM

#15

yup nice explanation-cleared many thngs

You don't have permission to post.