Best way to tell how many (and how important) papers a PI is publishing?

[quanto]

Anyone know a good way to find out not just how many papers a professor is publishing, but also how well referenced the papers are? I know a metric exists that measures these two criteria together, but not sure if a student like me would have access to it. Where would I find this information, or a similar statistic? I know I can go on Pubmed to look at the amount of pubs. Also, does anyone know what's considered a 'good' and 'average' rate of publications?

I'm trying to choose between 2 PI's, both whom I really like..

[hippyscientist]

It's less about what the PI is publishing and more about what their current students are publishing under them. Research Gate is quite useful for that, as is just searching on PubMed or an equivalent to find papers each PI have published and see the other authors. 

Remember to take into account other factors like lab environment, location, cost of living, and the arbitrary "where do you think you'll be happier?" as well as productivity and reputation. Good luck  

[St Andrews Lynx]

Google Scholar shows the number of citations a paper has, and you can search by PI name. That you don't need any kind of institutional access for. If you're enrolled in an institution then you can use SciFinder and Web of Science - both of which will give you citations and allow you to analyse the publication list by year, journal name, etc. You can also look for an individual PI's "H-index".

Chemistry's my field, but publication rate can vary a huge amount between subfields, and depending on the size of a group (e.g. a 10-person lab publishing 15 papers per year vs. a 40-person lab publishing 20). And as hippy scientist has said, there's a lot of other factors involved in choosing a lab/being happy in a lab. 

[TakeruK]

Google Scholar is a good resource as others said, but be warned: it doesn't actually count citations! It makes an estimate based on Google's (very good) algorithms. However, I find that this overcounts my citations---I like seeing the inflated numbers for my own self-esteem but it's not actually true! It might be okay to compare Google Scholar results within themselves though, assuming the inflation is the same.

In my field, there are 2 or 3 major journals and you can always search individual journals without library access. And you can always get some basic information on each article for free, such as the abstract, author list, and metrics such as "how many papers cited this one". Usually you just have to pay for the full text.

Also, we have a abstract database service that is pretty good at keeping track of stats. Here's an example page for an important paper in my field: http://adsabs.harvard.edu/abs/2010ApJ...718L.145W There is a count of total citations next to "Citation History" and you can even see histograms of the breakdown by year etc. Maybe you can find something similar for your field?

Finally, if you are just choosing between 2 PIs, then in addition to what others said about the non-paper related factors, I want to make another suggestion. Instead of looking for papers written by the PI, instead, find the papers that are written by the student. I know that in some fields, the student isn't the first author, so I mean look for the papers that are produced by the student's work. Because chances are, papers like these would be what you will write in a few years. You can find these papers by finding out what the students have been working on and matching them up to papers produced by the lab.

When you look at these papers, you can do things like:

1. Compare papers of different students/projects with each other. Are they of similar writing style/quality? This will give you a sense of whether or not the PI and the lab are hands-on with writing and you can see if there is a similar style across all papers that the PI is involved in, or if students generally write everything themselves and the PI is just signing off on them.

2. Are the papers introducing novel research and results? Or, are the students just part of a big research machine and they are just the gears in a research protocol that the PI already established? That is, are students thinking up their own ideas and publishing them or are they just doing what the PI tells them to do? 

3. Are the students producing interesting research---are the papers being cited by people outside of their lab/collaboration?

In my opinion, I think the quality of the papers produced by the lab and the students is more important than just the quantity. And I think you especially want to look at papers produced from student projects because that's what your papers may look like in a few years. 

[Eigen]

Rates of publication are also going to vary hugely between subfields of chemistry- even within a subfield, depending on the project. 

[quanto]

9 hours ago, St Andrews Lynx said:

You can also look for an individual PI's "H-index".

Yeah, the H-index is exactly what I was talking about! Thanks for the tip!

[TakeruK]

11 hours ago, quanto said:

Yeah, the H-index is exactly what I was talking about! Thanks for the tip!

Just remember that this is just one metric, and there's a lot of reasons why people don't like this metric. For example, h-index doesn't include information about where the author appears in the author list. Maybe this doesn't matter as much for established professors because most of their papers will have them in whatever author position is typical for the advisor in their field (in mine, it's 2nd author, in others it's last author). However, this still does not distinguish between papers where the prof writes with their lab members and other papers where the prof might appear as a collaborator but the work is actually done by a different lab (i.e. not representative of what your experience in the lab could be). 

Also, the h-index doesn't account for time. Someone who used to be really active but have slowed down a lot in the last 10 years would still have a high h-index but this doesn't represent the current impact of their lab group to the field, and this is something a new student would also be interested in knowing.

Like any other numerical index, it's important to know the limitations