Back to [Archive] Post-review discussions

[ODP]Reassessment of Jewish Cognitive Ability:
The terms of use of the ISPCR are: "You agree not to redistribute data or other materials without the written agreement of ICPSR". I don't think they would grant someone the ability to upload the data elsewhere.

I understand and condone the policy of the journal. However, I forsee others running into the same problem. Many of the data files used by researchers are under similar, often more limiting, restrictions (e.g., ADD Health). In thge case of Project Talent the data is available just three clicks away.

I appreciate the comments and especially editing and help with the MCV. Curt
It's sad thing if it cannot be published here. It's a known fact that this kind of data are not destined to be shared. The problem is with the institution.

But I can see another alternative. If publication of raw data is not possible (even only for the variables considered in the study) why not publish the correlation and covariance/variance matrix, along with the mean scores of those variables instead ?

If you only publish covariance, even this can be quite useful. For example, with covariance/variance matrix (+mean and SD) you can perform MGCFA, but also SEM and other path analyses. That can be very helpful.

EDIT : that's with the use of covariance matrix that some people like Briley and Beaujean succeed to publish the following papers :

.Explaining the Increasing Heritability of Cognitive Ability Across Development: A Meta-Analysis of Longitudinal Twin and Adoption Studies.

.Heritability of cognitive abilities as measured by mental chronometric tasks: A meta-analysis.

EDIT (2) :

The data is just there

I look at it and found the variables Dunkel used, for example, the question "HOW WELL DOES PARENT SPEAK HEBREW OR YIDDISH" has the variable label "BY_SI062". I can see it's a big data.
It looks like anyone can download the data just by signing up. This means that the data is de facto public and the paper can be published. I am downloading it right now to verify. It is a huge file, 3.5 GB.
I know. But I just wanted to ask if the author can share the covariance/variance matrix (+ mean and SD) instead of the raw data if the latter is forbidden. It is feasible or not ?
It is not a problem. I will share the data. I'm not a US citizen, the webserver is not hosted in the US, and data cannot be copyrighted anyway (

I have uploaded the datafile here:

The paper can now be published if reviewers agree on publication.
Here is the updated version. It is not substantially different from the last uploaded version, the only differences is that a few typos were corrected.
I think it is publishable.

Can the other reviewers please state whether they consent to publication of this submission?
Three reviewers have given their consent. Can the author please upload a final edit. Remember to put the publication date in the top for the day you want it to be published.
Here is the final edited version.
Slight edit to the first page. Final version.