Thanks for the suggestions. Seeing as I'm reasonably comfortable with a spreadsheet and these statistics packages look daunting, a spreadsheet is looking like the best way to start off. I'll have to do some more research on OOo Calc's features. As you said, some of these other packages support import via csv in case I need more functionality.
In regards to collecting data, I don't have access to slashdot's database and some of my variables are subjective (such as ideology), which will involve coding each individual comment. Depending on how much time this takes, I'm probably looking at anywhere between 200-2000 comments. Thanks again for all the suggestions, if anyone has any more advice on how to go about collecting or organizing this kind of data, I'm all ears. Jesse On Mon, 2007-01-10 at 19:06 -0700, Johnny Stork wrote: > Well I certainly wouldnt suggest matlab or R for someone who clearly > states they are not that good/comfortable with statistics. > > Depending on the complexity of your design, I would suggest a spread > sheet to start. You didnt state which package you would be using for the > analysis so I am guessing you may may not have one (SPSS, Minitab, > Systat etc). > > Spreadsheets are a good starting point since you can do your organizing > and data cleaning in a familiar interface. Once you collect, clean and > organize (grouping etc) your daya you can then export as CSV to import > into any stats package, or, you can do some basic stats in Excel (Open > Office Calc). I am also guessing you are only looking at doing some > descriptive stats (counts, means, variance, sd etc) and so these are > easoly done in Excel. You can also do many different basic means tests > (is this mean/average statistically larger than the other mean), in > Excel (T-Tests) > > With regards to gathering the data, your best bet is to get someone to > write some custom sql to pull out the data, grouped and structured, the > way you want. If you dont have access to the systems/db's you want to > collect data from, then you certainly have some work ahead of you. > > > ps: I have been a statistician for more than 20 years, mostly in > academic research, psychology etc, but some marketing and forestry research > > Richard Carter wrote: > > You could try gretl > > > > Robin > > > > On 10/1/07, *Jesse Kline* <[EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]>> > > wrote: > > > > I'm looking into doing some research on the level of discussion on > > a web > > forum such as slashdot. My analysis would keep track of variables such > > as User ID (is the discussion dominated by a few users), moderation > > score (what is the level of discourse on the board), ideology (is it > > dominated by people on the left or the right), patterns of > > agreement and > > disagreement (do posts generally agree or disagree with the > > parent), and > > possibly a few other variables. > > > > I'm not looking at doing too much statistical analysis of the data, as > > I'm really not good at statistics, but I would like to find a good way > > of gathering and analyzing the data. I know that this can be done > > using > > a spreadsheet, but I don't think that is the best software in this > > situation. I'm also not much of a database guy and I would like to > > find > > some software that can help me analyze the results. Does anyone > > have any > > suggestions? > > > > Thanks, > > > > Jesse > > > > > > _______________________________________________ > > clug-talk mailing list > > [email protected] <mailto:[email protected]> > > http://clug.ca/mailman/listinfo/clug-talk_clug.ca > > Mailing List Guidelines (http://clug.ca/ml_guidelines.php) > > **Please remove these lines when replying > > > > > > > > -- > > This message has been scanned for viruses and > > dangerous content by *MailScanner* <http://www.mailscanner.info/>, and is > > believed to be clean. > > ------------------------------------------------------------------------ > > > > _______________________________________________ > > clug-talk mailing list > > [email protected] > > http://clug.ca/mailman/listinfo/clug-talk_clug.ca > > Mailing List Guidelines (http://clug.ca/ml_guidelines.php) > > **Please remove these lines when replying > _______________________________________________ clug-talk mailing list [email protected] http://clug.ca/mailman/listinfo/clug-talk_clug.ca Mailing List Guidelines (http://clug.ca/ml_guidelines.php) **Please remove these lines when replying
