# [R-meta] measurement conversion is possible ?

Gerta Ruecker ruecker @end|ng |rom |mb|@un|-|re|burg@de
Fri Nov 8 10:37:54 CET 2019

```Dear Martin,

Of course it would be possible to impute some correlation. For example,
if you assume no correlation between the pre and post average
measurements (x1 and x2, say), the variance of the mean difference is

Var(x1 - x2) = Var(x1) + Var(x2) - 2* Cov(x1, x2) = Var(x1) + Var(x2)

and the standard error is

SE(x1 - x2) = sqrt(Var(x1) + Var(x2))

where sqrt is the square root.

However, you should have some motivation for assuming no correlation.

Another example: If you assume a positive correlation, say r = 0.5, then
you have

r = Cov(x1, x2)/(SE(x1) * SE(x2)) = 0.5

=> Cov(x1, x2) = 0.5 * SEr(x1) * SE(x2)

=> Var(x1 - x2) = Var(x1) + Var(x2) - 2* Cov(x1, x2) = Var(x1) + Var(x2)
- SE(x1) * SE(x2) = SE(x1)^2 + SE(x2)^2 - SE(x1) * SE(x2)

Again there must be some motivation for assuming this. For example, if
you have one or two studies with known standard errors for the
differences, you may use these to impute correlations/standard errors
for the other studies.

Best,

Gerta

Am 07.11.2019 um 22:54 schrieb Martin Lobo:
> Thank you very much for the example. then it is impossible since the
> correlation between the variables is never reported. At least in the
> data I'm looking for. It should have a minimum correlation between the
> baseline variable and the varable post. is that so?
>
> best,
>
>
>
> */
> /*
> */Lorenzo Martín Lobo /**/^MTSAC, FACC, FESC /*
> /*/*Jefe de Cardiología *//*Hospital Militar Campo de Mayo*/
> */
> /*Ex Jefe de Unidad Coronaria *//*Hospital Militar Campo de Mayo*/
> /
> /*Miembro Titular de la Sociedad Argentina de Cardiología*/
> /
> /
> /*Fellow American College of Cardiology*/
> /
> /
> /*Fellow European Society of Cardiology*/
> /
> /
> /*/*Miembro del Area de Investigación de la SAC*/*/
> /
> /*Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC*/
> /*Miembro Asesor /*del Consejo de Aterosclerosis y Trombosis de la SAC*/*/
> /*/
> /*/
> /*Ex Director del Consejo de Epidemiología y Prevención Cardiovascular
> de la SAC*/
> /*/
> /*/
> /*/
> /*//*/
> /*/
> /*Miembro Asesor del Consejo de Epidemiología y Prevención
> Cardiovascular de la SAC*/
> /*/
> /*/
> /*/
> /*/
> /
> /*Instructor de ACLS de la American Heart Association*/
> /
>
>
> ------------------------------------------------------------------------
> *De:* Gerta Ruecker <ruecker using imbi.uni-freiburg.de>
> *Enviado:* jueves, 7 de noviembre de 2019 13:53
> *Para:* Martin Lobo <mlobo4370 using hotmail.com>; Viechtbauer, Wolfgang
> (SP) <wolfgang.viechtbauer using maastrichtuniversity.nl>;
> r-sig-meta-analysis using r-project.org <r-sig-meta-analysis using r-project.org>
> *Asunto:* Re: [R-meta] measurement conversion is possible ?
>
> Dear Martin,
>
>
> one cannot conclude from the variances of pre/post values on the
> variance of the pre-post difference without knowing/making assumptions
>
>
> See the attached R code for a hypothetical example.
>
>
> Best,
>
> Gerta
>
>
>
> Am 07.11.2019 um 16:52 schrieb Martin Lobo:
>> Thank you very much !!
>>
>> Do you know any method to impute the standard deviation of a
>> difference of means (that is, I have the mean 1 with its deviation
>> and the mean 2 with its deviation, I need to estimate the deviation
>> of m1-m2)
>>
>> Best
>>
>>
>>
>> */
>> /*
>> */Lorenzo Martín Lobo /**/^MTSAC, FACC, FESC /*
>> /*/*Jefe de Cardiología *//*Hospital Militar Campo de Mayo*/
>> */
>> /*Ex Jefe de Unidad Coronaria *//*Hospital Militar Campo de Mayo*/
>> /
>> /*Miembro Titular de la Sociedad Argentina de Cardiología*/
>> /
>> /
>> /*Fellow American College of Cardiology*/
>> /
>> /
>> /*Fellow European Society of Cardiology*/
>> /
>> /
>> /*/*Miembro del Area de Investigación de la SAC*/*/
>> /
>> /*Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC*/
>> /*Miembro Asesor /*del Consejo de Aterosclerosis y Trombosis de la
>> SAC*/*/
>> /*/
>> /*/
>> /*Ex Director del Consejo de Epidemiología y Prevención
>> Cardiovascular de la SAC*/
>> /*/
>> /*/
>> /*/
>> /*//*/
>> /*/
>> /*Miembro Asesor del Consejo de Epidemiología y Prevención
>> Cardiovascular de la SAC*/
>> /*/
>> /*/
>> /*/
>> /*/
>> /
>> /*Instructor de ACLS de la American Heart Association*/
>> /
>>
>>
>> ------------------------------------------------------------------------
>> *De:* Gerta Ruecker <ruecker using imbi.uni-freiburg.de>
>> <mailto:ruecker using imbi.uni-freiburg.de>
>> *Enviado:* jueves, 7 de noviembre de 2019 11:41
>> *Para:* Martin Lobo <mlobo4370 using hotmail.com>
>> <mailto:mlobo4370 using hotmail.com>; Viechtbauer, Wolfgang (SP)
>> <wolfgang.viechtbauer using maastrichtuniversity.nl>
>> <mailto:wolfgang.viechtbauer using maastrichtuniversity.nl>;
>> r-sig-meta-analysis using r-project.org
>> <mailto:r-sig-meta-analysis using r-project.org>
>> <r-sig-meta-analysis using r-project.org>
>> <mailto:r-sig-meta-analysis using r-project.org>
>> *Asunto:* Re: [R-meta] measurement conversion is possible ?
>>
>> Dear Martin,
>>
>> There is some literature on this problem, see for example
>> https://bmcmedresmethodol.biomedcentral.com/articles/10.1186/1471-2288-14-135
>>
>> Best,
>>
>> Gerta
>>
>> Am 07.11.2019 um 15:27 schrieb Martin Lobo:
>>> Hi everyones
>>> Several times it happens that we do not have the measure we need or its dispersion.
>>>
>>> for example:
>>> median (inter quartile range)
>>> median (confidence interval)
>>> or we have mean (standard deviation) pre test and post test, but we don't have the standard deviation of the difference.
>>> Is there any way with this data to estimate the mean and standard deviation without too much error (or that it is accepted methodologically)?
>>>
>>> Thank you
>>>
>>>
>>>
>>>
>>> Lorenzo Mart�n Lobo MTSAC, FACC, FESC
>>> Jefe de Cardiolog�a Hospital Militar Campo de Mayo
>>> Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo
>>> Miembro Titular de la Sociedad Argentina de Cardiolog�a
>>> Fellow American College of Cardiology
>>> Fellow European Society of Cardiology
>>> Miembro del Area de Investigaci�n de la SAC
>>> Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC
>>> Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC
>>> Ex Director del Consejo de Epidemiolog�a y Prevenci�n Cardiovascular de la SAC
>>>
>>> Miembro Asesor del Consejo de Epidemiolog�a y Prevenci�n Cardiovascular de la SAC
>>>
>>>
>>> Instructor de ACLS de la American Heart Association
>>>
>>>
>>> ________________________________
>>> e la American Heart Association
>>>
>>> 	[[alternative HTML version deleted]]
>>>
>>>
>>> _______________________________________________
>>> R-sig-meta-analysis mailing list
>>> R-sig-meta-analysis using r-project.org  <mailto:R-sig-meta-analysis using r-project.org>
>> --
>>
>> Dr. rer. nat. Gerta Rücker, Dipl.-Math.
>>
>> Institute of Medical Biometry and Statistics,
>> Faculty of Medicine and Medical Center - University of Freiburg
>>
>> Stefan-Meier-Str. 26, D-79104 Freiburg, Germany
>>
>> Phone:    +49/761/203-6673
>> Fax:      +49/761/203-6680
>> Mail:ruecker using imbi.uni-freiburg.de  <mailto:ruecker using imbi.uni-freiburg.de>
> --
>
> Dr. rer. nat. Gerta Rücker, Dipl.-Math.
>
> Institute of Medical Biometry and Statistics,
> Faculty of Medicine and Medical Center - University of Freiburg
>
> Stefan-Meier-Str. 26, D-79104 Freiburg, Germany
>
> Phone:    +49/761/203-6673
> Fax:      +49/761/203-6680
> Mail:ruecker using imbi.uni-freiburg.de  <mailto:ruecker using imbi.uni-freiburg.de>

--

Dr. rer. nat. Gerta Rücker, Dipl.-Math.

Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center - University of Freiburg

Stefan-Meier-Str. 26, D-79104 Freiburg, Germany

Phone:    +49/761/203-6673
Fax:      +49/761/203-6680
Mail:     ruecker using imbi.uni-freiburg.de
Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view

[[alternative HTML version deleted]]

```