susannah said:

it is 11am

ellieadore said:

I WAS HERE TIL 11PM LAST NIGHT!


Dont shout at me

This thread is really... unfortunate.

But I say the bar graph!

susannah said:

almost done Think of the piss up u will have on Friday night

Not long to go now until we

ellieadore said:

ok I am going back to work now

The first graph is clearest to understand.

Good luck sweety. Yeah, the first graph.


M

It depends what you are trying to conclude.

Each is an independent analysis so by placing them in some kind of order in graph one you are implying importance to each end of the graph. In which case the area graph (graph 2) is the best for the job (despite looking hideous BTW). If your message is as I expect it to be and warranting graph one, then put the cities in alphabetical order. You can then discuss the implications on London and New York as "most necessary" in your discussion. This gives a more non bias approach to presenting the data and at the same time give you scope for discussion (without “giving it all away”)

I would also tone down the colours, it is possibly better if you let the data speak for itself and it is good data This can depend on your field but I am guessing plain and simple factual presentation is key.

Please excuse my comments if you like but you knew I wouldn't be able to stay away from this one! Let me know if you have any other Q's

There are other options too for the graphs such as individual pie charts for each city but that could look messy.
[Edited 5/1/07 8:06am]

PREDOMINANT said:

I was wondering when you'd show up!

I did originally have them in random order, in the order I asked them in the question, but apparently that was difficult to read, and to make my [slightly biased, but then so is the data] point, I put them in order of replies.

As for the colours, all my other graphs are simple and look rather snazzy with their blue background and blue data - but there are just not enough shades of blue for this graph! So I went for some sort of "danger" theme

I did think of individual pie charts, but there would be too many, and take up too much room, and as it is, the text I have explaining it does a pretty good job, I think...

I tried a line graph but that just looked mad. Hang on I will get the other ones on here, just for you Professor Predom

Nothing to do with the fact that youre the only person in this thread who's ever marked anything close to a BSc Dissertation!

Thanks Steve

The line graph



The original 2nd graph



How about this then? Change order?

Aaaaand again.

Don't you think the biased one, in order, is clearer than this?

susannah said:

the first one was a lot clearer.

evenstar3 said:

the blue-greens are better though aren't they?

Thanks!

susannah said:

well, a higher bar in each city translates to more protection needed against terrorism, right? on a basic level i think red = danger, so the other colors might be better. the blues are more matchy though.

evenstar3 said:

Thats what I thought!

I should have just kept my mouth shut

susannah said:

Stick the the original red-coloured graph. So clear. Clarity over looks, darling.

Spookymuffin said:

The first one from the first post?

Ok, but I'm waiting for Predom to come back though....

susannah said:

I like stars.

Spookymuffin said:

Eh?




nvm

do I get a free 50" plasma TV?

NastyPig said:

What d'yo think I am, Comet?!

Sorry, got caught up in the footie last night and didn't log on again.

Blue green is definitely better. All results, tables and graphs should (in general) stick to the same colour scheme throughout your piece of work. This will give your dissertation clarity and continuity and convey to the reader that you understand the topic and your results.

With regard to the order, this is difficult because we are in different fields. In science data should be presented in a random, non-bias fashion. The results you portray are then seen to be more "real" and not messed with. By re-ordering the cities you are almost implying that this was implicit from the start (and therefore why study it?). The order as determined from the results (not by the results) can then be explained in words or, if you like, further figures.

There is no justification for the Tokyo first example, like I said yesterday go with alphabetical if you decide to go with and out of order option. As you discuss your data put in another figure just showing the "highly necessary" data in order to illustrate your point. In fact this will highlight the fact that some believed that Edinburgh is in need of protection. Sometimes these smaller points can become lost in a "full" figure. Also by breaking the figure down it will give you much more mileage with your data.

One thing I forgot yesterday was to mention that there are no units on your "y" axis, this should be

necessity in number of responses (for example, if this is the case)

The other graphs are pants!

Now can I get on with marking the pile of dissertations I have on my desk? These will fail you will do just fine

PREDOMINANT said:

THANKYOU Predom

Taken it all on board

Now, back to work for both of us!

susannah said:

PREDOMINANT said:

I'm going for a fry-up now

susannah said:

Y'see now thats where we differ. I have a cold cup of coffee and some bulshit amin to do.

I wanna new job

PREDOMINANT said:

Oh don't worry dear, everyone wants my job (student)

susannah said:

What i would give to be a student again!

PREDOMINANT said:

Don't, I will be in that sorry state soon enough