I am not familiar with the "difference-in-difference" terminology. Please describe and/or provide an example.

Dear Gordon, thanks for your answer! I am talking about estimates (often used in econometrics) of quasi-experimental designs. DID makes use of longitudinal data from treatment and control groups to obtain an appropriate counterfactual to estimate a causal effect. DID is typically used to estimate the effect of a specific intervention/treatment/event by comparing the changes in outcomes over time between a population that is exposed to the treatment (the intervention group) and a population that is not (the control group). See: https://www.mailman.columbia.edu/research/population-health-methods/difference-difference-estimation

However, I do not have means and SD's but only the DID-coefficient. And, I fail to find this in CMA's list of options.

OK, I for the most part understand what you are talking about. I was just calculating effect sizes for such data yesterday. I don't know what you mean by DID-coefficient though.

That depends on the type of your data

if your data are of the discrete type then total and disease number for both control and study groups are entered

If your data are continuous or scale type you enter total sample size, mean and SD for both study and control group

Thanks! That’s indeed the problem: this data is not given (only the final DID-estimate). But I’ll contact the authors for the numbers (it’s descrete data)!

Dear Colleague

I am so acknowledged by your kind response, Yet far from DID? concept?

In Metanalysis there exist no such and I hope you make it simple for me so i will do all the job for you and for free:

are your data of the kind total number and positively identified number presented as %? like 10% out of 55 patients?

Or your data are of the mean+Standard deviation for a total N1 in the study group and N2 for the control group?

-----------------

In the first example the random and fixed effect of Odd ratio are calculated first and about which one to choose? depends on the test of homogeneity

While in the case of mean and SD group first the SMD standardized mean difference is calculated still for the fixed and random effect and the choice between them depends on the test of homogeneity!

-----------

Above all how many paper you would like to undertake for your metanalysis?

I shall be more than happy to help you even by giving a theoretical example so you understand the problem in better way

Did you figure it out? I came across the same issue. I tried to contact authors for my project and no response from the them....