advice wanted please on descriptive stats

[MAVirgin]

As my username would suggest, I'm doing my first dissertation. Due to family illness, I'm already far behind and trying to make it simple. So my question is, I've done a SurveyMonkey questionnaire, got a reasonable number of replies and SurveyMonkey itself gives me the data eg percentages who said X or Y and pretty graphs if I want them. But do I also need to provide means and standard deviations? My institution gives me access to SPSS but I'm not sure if I have the time or ability to get to grips with it.

So for descriptive stats, can I just get away with reporting basically just what SurveyMonkey says, or would that look really crap?

[Summerswallow]

I'm guessing from your name this is Masters level. I can't say for sure, I would expect means and standard deviations, however, the point is that it doesn't matter what I would expect- it's all about what your institution and your discipline and your marker would expect! That's what you have to clarify with your dissertation supervisor, even if they are busy/don't reply straight away etc. The requirements for a psychology Masters might be much more stringent in terms of stats (going beyond description) than for a subject which is less stats oriented, also depends on the supervisor marker. Make it a priority to contact them immediately and ask them the above question and get a written answer and/or ask to look at a quant Masters dissertation sample for guidance (if they do this).

We just can't tell you, sorry!

[Summerswallow]

And- if you are behind due to family illness, look into the procedure to get an extension now and get one, no point in trying to cover it up/bash on, perhaps you can, perhaps you can't, but this needs clarifying- the supervisor may not be the person who knows this, the administrator/webpages with the handbook is probably the best place to look for this, ours would require evidence usually though in the form of a sick note but not all do.

[MAVirgin]

I have an extension already so can't get another! It's a mixed methods research, so the survey is only a third of my research, so not that crucial. I'm rally near my deadline which is why I don't want to ask my supervisor now! If I do need the means/SD then will have to ask a friend for help, as I won't be able to figure out how to work them out before then on my own...but I'd rather not do this if it's OK as it is.

Any more thoughts?

[geekaMaxima]

Oh ffs, grow a spine and ask your supervisor! No one else can tell you how important it is

As questions go, it's a long way from the silliest ones that supervisors get. Checking expectations / marking weights = sensible.

[MAVirgin]

Agreed - but it's due in this week. So a bit late!

I'd rather leave it out than make it obvious it was that last minute... (My supervisor does know about the family illness though.)

[geekaMaxima]

As a supervisor, I honestly wouldn't care that it was last minute as I only mark the submission in front of me.

Straight up, just say that you're having some trouble with figuring out what's appropriate for descriptive stats and would like to know if means and st devs are (a) required/expected, or (b) nice but not essential. You have nothing to lose.

[MAVirgin]

Thanks!

[Veterinari]

Means are very easy - add up each individual value and then divide by the number of values - you can do those st least - its gcse maths level.

[giggleshizz]

For my MA dissertation I had to do advanced statistics. What subject? If they've given you access to spss they will probably be expecting you to use it. If you are this far behind due to family illness I would seriously look at deferring and try and get your head around the stats. Good luck.

[giggleshizz]

Sorry just saw you can't get another extension.

[inkydinky]

"I'm rally near my deadline which is why I don't want to ask my supervisor now!"

This is EXACTLY the time to ask your supervisor. What was the survey done for if you aren't going to analyse the data? Was there actally a research question? If not, your research was pointless (and unethical).

At M level I would expect to see a minumum of basic univariate statistics (i.e. to go beyond the descriptives) and actually would expect advanced statistics for everything other than qualitaitve studies or case studies. You should have had a data analysis plan formulated when you constructed the survey and got ethical approval for it. What did it say you would do?