Hi John,How I understood your problem was that you are trying to compute the pooled standard error from the two samples and I have assumed that you know much of the stuffs about this situation.As a result, I have ignored further details like assumptions (normality, equality of variances etc) that must be met in order to apply whichever method that you can and I have just given you one example which is for pooled standard error for two sample means with each sample not differing much in their variances (checked using F-distribution). Also, in my post, I said you should have a look at pooled standard errors. However, from what Robert suggested, I think you need to tell us more about your data and the main objective of your analysis so as to allow us provide relevant response to your question and to avoid speculation.Thanks Kind regards
*****************************************
Justice Moses K. Aheto
PhD Candidate in Medicine (United Kingdom)
MSc Medical Statistics (United Kingdom)
BSc Statistics (Ghana)
HND Statistics (Ghana)
Chief Executive Officer
Statistics and Analytics Consultancy Services Ltd.
E-mail: [log in to unmask]
Skype: jascall12
Mobile: +447417589148.
On Monday, December 29, 2014 11:44 AM, John Mbotwa <[log in to unmask]> wrote:
Hie John,Just as Moses has indicated in his contribution, you can use pooled standard errors. However this will depend on whether the standard deviation for sample 1 does not differ much with that of sample 2. In most books, they use a pooled estimate if standard dev sample 1/standard dev sample 2 is less than 2. Note the bigger standard deviation should be on top when dividing. When the answer is greater than 2 you can't use a pooled estimate of the standard deviation since the pooled estimate will be biased towards the sample with a bigger standard deviation.I hope this will help.
Regards,JohnOn 29 Dec 2014 13:26, "Justice Moses K. Aheto" <[log in to unmask]> wrote:
Hi John,I think you should have a look at pooled standard errors which you can use to estimate the combined standard error for the two samples.Example below: SEx1-x2 = sqrt [ s21 / n + s22 / m ] Hope this helps.Cheers Kind regards
*****************************************
Justice Moses K. Aheto
PhD Candidate in Medicine (United Kingdom)
MSc Medical Statistics (United Kingdom)
BSc Statistics (Ghana)
HND Statistics (Ghana)
Chief Executive Officer
Statistics and Analytics Consultancy Services Ltd.
E-mail: [log in to unmask]
Skype: jascall12
Mobile: +447417589148.
On Monday, December 29, 2014 11:12 AM, John Sorkin <[log in to unmask]> wrote:
How can I compute the SE (standard error) of the concatenation of two samples?
Assume sample 1 has n observations, sample 2 has m observationsThe concatenation of sample 1 with sample 2 would have n+m observations.If I know the mean of sample 1 = m1, and SE sample 1=se1, the mean of sample 2 = m2, and SE sample 2=se2,
can I use n, m1, se1, m, m2, se2 to compute the SE?
Thank you,John
John David Sorkin M.D., Ph.D.
Professor of Medicine
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology and Geriatric Medicine
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)
Confidentiality Statement: This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message. You may leave the list at any time by sending the commandSIGNOFF allstatto [log in to unmask], leaving the subject line blank.
You may leave the list at any time by sending the commandSIGNOFF allstatto [log in to unmask], leaving the subject line blank.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
