Wolfgang Viechtbauer, Ph.D., Statistician | Department of Psychiatry and
Neuropsychology | Maastricht University | P.O. Box 616 (VIJV1) | 6200 MD
Maastricht, The Netherlands | +31 (43) 388-4170 | <http://www.wvbauer.com/>
http://www.wvbauer.com
>-----Original Message-----
>From: R-help [ <mailto:r-help-boun... at r-project.org>
mailto:r-help-boun... at r-project.org] On Behalf Of K Amoatwi
>Sent: Tuesday, 16 June, 2020 4:50
>To: r-help at r-project.org
>Subject: [R] Error message in meta-analysis package Metafor-weights =""
>
>Dear All,
>I am using the example from one of the tutorial about "Metafor" package and
>"escalc" function, to learn how this package can be applied to do
>meta-analysi; the code and the data is directly from the tutorials but
>"weights=freq" option in the escalc function is given me error message
>This is the code below:
>
>library(metafor) # Load package
>#####DATASET 1: BCG Vaccine Trials
>data(dat.bcg) # BCG meta-analytic dataset
>
>##Formula based Specification
>##That is, what if I have multiple rows per study, corresponding to
>difference treatment groups?
>
>library(reshape2) # Load package for data reshaping
>
>bcg.long <- melt(dat.bcg[, c("trial", "tpos", "tneg", "cpos", "cneg")], id
>= "trial")
>bcg.long$pos <- ifelse(bcg.long$var == "tpos" | bcg.long$var == "cpos", 1,
>0)
>bcg.long$group <- ifelse(bcg.long$var == "tpos" | bcg.long$var == "tneg",
>1, 0)
>
>##sample of the data, the first 6 rows
>head(bcg.long)
> trial variable value pos group
>1 1 tpos 4 1 1
>2 2 tpos 6 1 1
>3 3 tpos 3 1 1
>4 4 tpos 62 1 1
>5 5 tpos 33 1 1
>6 6 tpos 180 1 1
>
>##Now applying the " escalc " function
>
>escalc(factor(pos)~factor(group)| factor(trial),weights = value,data =
>bcg.long, measure = "OR")
>
>##Then I got this error message
>Error in escalc(factor(pos) ~ factor(group) | factor(trial), weights =
>value, :
> object 'value' not found
>
>I used the same data with different example from another author and got a
>similar error message
>Second code with the same data but different coding
>Sample data
>
>with the first 6 rows of the rearranged data shown below. (T=treatment,
>C=Control group, Out=outcome whether positive or negative, and then
>frequency)
> study grp out freq
>1 1 T + 4
>2 1 T - 119
>3 1 C + 11
>4 1 C - 128
>5 2 T + 6
>6 2 T - 300
>
>>escalc(out ~ grp | study, weights = freq, data = dat.fm, measure = "OR")
>
>Error in escalc(out ~ grp | study, weights = freq, data = dat.fm, measure =
>"OR") :
> object 'freq' not found
>
>I am not sure what I am doing wrong since both authors were able to get
>their results while I am getting error messages.
>
>Any help will be very much appreciated
>
>Amoatwi
______________________________________________
[[alternative HTML version deleted]]