Ok, now this sounds a bit silly, but I'm a bit concerned about the positive feedback I got on an article I submitted to a peer-reviewed journal. At first I was just happy to get a revise+resubmit, but now that I'm starting to make the changes I realise that the comments of the reviewer are rather superficial, and I'm getting somehow scared that my stats didn't get scrutinized properly.

The thought of publishing something that bears my name and then turns out to be flawed is rather scary right now. I initially submitted the article to get some more substantial feedback (especially on one issue of my use of the statistical method) as they are not too stats-savy at my department, and I wanted some external feedback whether what I'm doing is ok. But the reviewer's superficial comments don't reassure me. I feel thatif the analysis is flawed but published I can only blame myself.. thestats department at my uni is not very approachable and so I've beenmainly teaching myself the details of the particular method used.

I'm kind of getting suspicious now, but I don't know if that's just my paranoia of being a fraud or if I have a point, and if the latter, what to do.

Any words of wisdom?

There should always be one statistician hiding in your department (somewhere), ferret them out. Failing that, bite the bullet and ask the maths department.

Peer review can be an absolute joke! It's one of those things that looks like a good idea on paper, but in practice it is utterly useless.

Sorry this isn't actually helpful apple, i just fancied a rant :p

I wouldn't rely on peer review for stats advice. It depends which peers it gets sent to. I had a recent article that was reviewed by someone clearly obsessed with one particular technique and obviously had no idea about the field we were actually looking at - I was tagging on to my sups publication, so she knows what she's on about!

Basically you can end up getting reviewed by 2 people who love e.g. factor analysis and another who loves structural equational modelling and get slated by one or all of them.

Hi all,

thanks for your comments..

The journal wrote me a letter last autumn saying that they forwarded the paper to someone to specifically look at the statistical bits in it.. so in a way that should reassure me that they now say it's ok but I'm still somewhat puzzled (that someone who's supposed to give stats advice to the journal is simply ok without commenting much on the stats at all).

So yes, maybe the strategy to get free stats advice from a journal was a bit mistaken. I feel a bit silly for getting so worried now about this issue when they actually said it's ok, but some suspicion/paranoia remains.

I have, however, heeding Hypothesis' advice, now tried again to contact someone from the stats department of my university.. so hopefully will have another opinion on the matter, though so far haven't gotten any reply to my e-mail. But then I guess the matter will be more complicated even, what if the stats person says it's indeed rubbish when the journal wants to have the original rubbish?

Argh, I'm getting on my own nerves here fretting about this issue, but thanks a lot for letting me vent, and thanks for your comments.