Hi everyone,

I'm wondering if someone can help with this. I have lots of samples which I'm analysing for levels for elements. I want to see whether there are correlations between the element levels across the samples, but is there a way to present data for these correlations (there are many elements I'm looking at, which creates a rather large number of combinations), and is there anyway to look at correlation between more than two elements, like a multiple correlation?

I'm using SigmaPlot, but could use something else for this is needed...

Thanks in advance!

Matt

I'm currently 9 months in to my PhD proper (1+3), and I'm thinking of a switch from Word to Latex. The only things that are holding me back are reference manager/referencing, and worrying about doing something wrong which can't easily be rectified later.

This may be a stupid question, but was there any link between the authors of the paper and the author of the thesis which may be a reason for this? Or was there somewhere in the thesis (possibly hidden away) where mitigation for this was given?

After much thinking, I've decided that, after my transfer (or upgrade, if you like) in September, which will be 12 months' in, I'm going to switch from Word to LaTeX. I've already installed TeXniccentre and MikTeX, and have been having a play around with them, but the one thing which still worries me is the referencing. I've downloaded JabRef, but I can't get it to work with my outputs. I'm sure I'm doing something wrong (generally, the output just contains the name of the key, not the actual reference, and no bibliography is built at the bottom), but this is the one area in which, at the moment, Word/Endnote still wins hands down. Can anyone help with this, or point me in the right direction?

Equally, if anyone knows of any better options that TeXniccenter, MikTeX, or JabRef, feel free to suggest. The problem is that I have zero experience in programming, and lots of the internet-based info is written by people who know a lot more about this kind of thing than me, even to the extent where I'm confised by terms like "class" and "argument".

Thanks for the help in advance! :-)

Hi Cmml100,

As far as I know, it is pretty much unheard of to do two PhDs. The point of the PhD isn't really so much about your subject of research, but the fact that you're doing research in itself. It's a mark of being able to do independent, academically sound research. Therefore, even if you went into another, completely different, field, you wouldn't do another PhD. The fact that the area you want to move into is overlapping with your current work is even more reason for you not to entertain this idea. Yes, you'd have to get to grips with the new area - the knowledge, the research methods, and so so - but you wouldn't have to learn how to do research over again.

I'd suggest looking for a postdoc in this...

Sneaks, buffer/bugger is a mix-up which has happened to me a few times in the past...

I once saw one, albeit not in a scientific journal, claiming that the search for the Higgs boson was taking place at the "Large Hardon Collider"...

Ok, well that 5 in favour of home here, and three friends, so I'm off with my parents now. Thank you for all your replies. The clincher was my housemate telling me that it looked like I'd been hit in the side of my face with a baseball bat!

Well, I have loads of papers with me so I can do a bit... I get told I'm a workaholic by most of my peers, but I suppose that's just me!

Hi everyone,

I've just had abadly impacted wisdom tooth removed, and it wasn't the most pleasant operation ever - I've had a tooth pulled before, but (without wanting to give you details) this wasn't as straightforward, even according to the consultant, and it took about 45 mins. This was three days ago and I'm in a lot of pain with bad swelling.

Anyway, my question is, has anyone else had anything like this, and how much time did you have off? I'm in my second year, and have only been off sick once before (a week in January this year with tonsilitis). My parents came to my city for the op, but I'm tempted to go home as I can't really eat anything, walk far, etc etc, but then I'm feeling really guilty about taking time off.

Any opinions? Much appreciated...

Bit of a stats question here. I have lots of samples from which I've measured lots of things. I'm wondering, is there some way I can probe the data for correlations between these things, for instance between levels of one chemical and another, or pH with potassium content?

Obviously, I don't want to have to test every single permuation separately, manually.

Well, I started reading about a year ago. All my "readings" are journal articles, mostly printed from online journals, some photocopied. I highlight and make notes on the paper itself, and then summarise the notes at the top of the paper, on the first page. I then enter the paper details into Endnote, and enter the summarised notes in the "research notes" field - that way, I can search for any papers containing a word, phrase, or concept I may have foudn interesting or potentially useful.

After about 70 papers, I began to write, and wrote a rough lit review. Now that I have the framework for it, whenever I read stuff now, I can hang it on those bare bones to give the review more substance.

In terms of keeping them, I have about six sections in my lit review, so I have a box file for each section, and within each one, I try to keep the papers in alphabetical order.

Once I've read a paper, it goes to my "to be written about" pile. Then, once it's added to the lit review (if necessary), it goes to the "for filing" pile, and then to the box files. I also have other magazine holder-type files where I keep papers which I want to read but haven't yet go around to reading, and these are also put under one of the six sections in my lit review, just so that I know roughly where to look for them.

Hope that helps...

Thanks for all the replies so far, but more are still welcome! I'm going to be in my room at about midnight I think, and I can lock from the inside (just not from the outside), so that's ok. I suppose there won't be to much drunkenness by then, so it should be ok. I hope.

Hi all,

One of my housemates is having a party at our house in a week's time. To be fair, he asked us first, but I'm a little concerned about the security of my room, with the door not having a lock fitted. Can anyone recommend a way of keeping my room secure (locking chain, hasp, or something) which can easily be removed when I leave? It's a hollow door, which means that I can't actually put a lock barrel inside the door, and it also means that, whatever the security measure, someoe could kick it down if they really wanted. Still, it's just something to show that people aren't meant to go in there...

The alternative is staying in all night (I won't go to most of the party as I never drink and don't know any of his friends, and I usually find that sober me, plus boozy strangers, don't mix that well :( )

Thanks!

Melsie, thanks for your reply. Most of it's above me (I'm in biomedical science and we generally avoid stats where possible!). However, I'm not sure I follow one of your points. You say that I should include time as a variable. However, if I pick just two time points (say an "early" one at 3h and a "late" one at 24h), then when do I need to correct for time? If I understand what you're saying, it's almost as if the act of taking a measurement at other time points alters the way I have to analyse them, which seems very strange...