Data Analysis & Theoretical Physics: Opportunities Beyond HEP

[barnflakes]

..require the analysis of raw dirty data and theoretical modelling, other than HEP, but do not require you to carry out the experiment yourself? I really want to work with real data because it feels like that's what I should be doing as a scientist/physicist, but it seems there is little opportunity to do data analysis and be a theoretician in this day and age. Any ideas welcome.

 

[Vanadium 50]

What makes you think that HEP doesn't require you to do the experiment yourself? Every HEP experiment I am aware of requires some level of "service work" from the collaboration - constructing the detector, operating the detector, maintaining spare parts, these sorts of things.

Furthermore, what makes you think a HEP experiment would be open to someone who wants to analyze the data but doesn't want to contribute to the experiment by doing service work? Most - probably all - would tell anyone who suggested it to take a hike.

 

[barnflakes]

What makes you think that HEP doesn't require you to do the experiment yourself? Every HEP experiment I am aware of requires some level of "service work" from the collaboration - constructing the detector, operating the detector, maintaining spare parts, these sorts of things.

Furthermore, what makes you think a HEP experiment would be open to someone who wants to analyze the data but doesn't want to contribute to the experiment by doing service work? Most - probably all - would tell anyone who suggested it to take a hike.

Well, thank you for picking up on the most irrelevant part of my post. Basically, forget HEP, what other areas allow the combination of the above?

 

[Division]

Every area of Astrophysics requires you to create theoretical models and use raw data to attempt to verify it.

 

[mal4mac]

Once the detector is up and running don't you then get research students to do pure data analysis? I was (for a short time...) part of a plasma physics research group that had one researcher devoted to computer modelling, and poor fool me twiddling the apparatus with my two left thumbs (I soon left for one of those cushy computer modelling jobs...)

In my theoretical astrophysics phase I had the impression that the guys in the office next to me did nothing but computer modelling & data analysis. NASA wouldn't fund them to go into space and twiddle with the detectors.

My last data analysis job was *all* computers - analysing reams of user interaction data. No chance of me being forced to mess with wires there :-) If, like me, and Max Planck, experiments start going wrong just by you entering the building, I recommend a change of field if you can't get that data analysis job in physics. Doing data analysis elsewhere is a much better bet than forcing yourself to get out the soldering iron if it just isn't your thing.

 

[cosmogirl]

mal4mac already mentioned theoretical astrophysics, but in fact observational astro is all about sitting in your office doing data analysis. This is what all the observers in my department do, especially if they work on extragalactic astrophysics: they get the data from big terrestial ot space telescopes, and basically sit and fiddle with it. Occasionaly (rarely, in fact) they go to different sites to observe themselves, but this is really cool, I think, and in any case doesn't involve soldering anything.

 

[barnflakes]

mal4mac already mentioned theoretical astrophysics, but in fact observational astro is all about sitting in your office doing data analysis. This is what all the observers in my department do, especially if they work on extragalactic astrophysics: they get the data from big terrestial ot space telescopes, and basically sit and fiddle with it. Occasionaly (rarely, in fact) they go to different sites to observe themselves, but this is really cool, I think, and in any case doesn't involve soldering anything.

Do you get to do any theory though? Build any models/do some simulations?

 

[cosmogirl]

Do you get to do any theory though? Build any models/do some simulations?

Well, I'm actually doing theoretical work, but from what I gather, the observers get to test theoretical models against their results, find reasons why the models don't work (and they usually don't :) ), and decide between competing models. I suppose most observers use models the theoreticians have built. This depends on the specific project though.

 

[Vanadium 50]

The problem with "just doing data analysis" even in astrophysics is that large experiments tend to hold on to the data for a period of time - like a year - before making it public. The feeling is that they worked hard to build and calibrate the instrument, so they should get first crack at analyzing the data. So exactly the same comments that were initially applied to HEP apply here.

 

[barnflakes]

The problem with "just doing data analysis" even in astrophysics is that large experiments tend to hold on to the data for a period of time - like a year - before making it public. The feeling is that they worked hard to build and calibrate the instrument, so they should get first crack at analyzing the data. So exactly the same comments that were initially applied to HEP apply here.

My main question still stands, are there any areas of physics where one can do solely data analysis and theoretical modelling, instead of hardware and data analysis?

 

[Vanadium 50]

One can't prove a negative, but the fact that nobody has come up with an example probably tells you something.

 

[barnflakes]

One can't prove a negative, but the fact that nobody has come up with an example probably tells you something.

Give them a chance..

 

[cosmogirl]

My main question still stands, are there any areas of physics where one can do solely data analysis and theoretical modelling, instead of hardware and data analysis?

You should wait and see what more experienced people will tell you, but from what I know, you either do theory or experiment. In the former you build models and test them against data others have already processed for you (this is what I do), and in the latter you play with the raw data and test it against theories others have built.
There are some projects that mix the two that I heard of, and sometimes professors work in tandems of theorist-experimentalist and co-advise one student who does a little bit of each thing. I wouldn't recommend this track, though.

(bringing attention to the thread so that people would come up with other ideas)