Hi, I have just found this forum, and it looks like a great place to get some help (I hope) For my dissertation, which is due way too soon, I am doing a survey, comparing attitudes of 2 independent groups, with 5 scale likert questions. Basically I want to show if they have similar or different attitudes. I am testing 4 hypotheses, and have in total about 20 questions. I have to say my statistic skills are very basic and very rusty, we had some lectures two years ago, where we were introduced to R. I looked through my notes, and back then we did a one sample t-test to analyse likert type questions. I believe I would need to do a 2 sample unpaired t-test. It would be great if someone could give me some feedback if this test is the most suitable one for my purpose, and maybe could explain to me what?s the easiest way to do this in R? You would help me loads!! Many thanks in advance Mona -- View this message in context: http://n4.nabble.com/comparing-attitudes-of-2-groups-likert-scales-tp2015738p2015738.html Sent from the R help mailing list archive at Nabble.com.
Mona_m wrote:> > For my dissertation, which is due way too soon, I am doing a survey, > comparing attitudes of 2 independent groups, with 5 scale likert > questions. > Basically I want to show if they have similar or different attitudes. I am > testing 4 hypotheses, and have in total about 20 questions. > >Using an unpaired (in you case) t-test on Likert scale is a bit risky, because the Gaussian distribution might be severely violated. It might be Ok if your data are reasonable centered around "moderate", but frequently we have responses where all but one subject replied with "very good". If you can create a sum of scores, these are frequently more suitable for being analyzed by some quasi-continuous method, and using a non-parametric Wilcoxon test might avoid reviewer comments in some areas of research. If you only have few levels, something like polr (in MASS) and the plots created from it by Fox/Anderson might be an alternative (google for Fox/polytomous effects). These results are more difficult to interpret and I have seen cases where papers using this where rejected in medical journals (why don't you use Wilcoxon?). Overall, when you have more complex cross-over designs with additional crossed variables, violating Gaussian assumptions for me seems to be the lesser evil compared to violating independence. "Assess independence, equal variance and normality -in that order" (van Bell, Statistical rules of thumb). I remember Douglas Bates mumbling something along the same lines, but he mentioned a 10 level scale. Dieter -- View this message in context: http://n4.nabble.com/comparing-attitudes-of-2-groups-likert-scales-tp2015738p2015812.html Sent from the R help mailing list archive at Nabble.com.
Good luck in your work, The simple solution would be to run many non-paired wilcox on all the 20 questions (the way Dieter suggested). In which case, make sure to adjust for multiple comparisons. Read about it, and see: ?p.adjust If you have some questions you can merge (by a simple mean of them), it will probably do you good (using PCA might be an option, but it could also be an over kill for you). You might also be interested in plotting your data, here is a nice simple hack on how to display the Correlation scatter-plot matrix for your data: http://www.r-statistics.com/2010/04/correlation-scatter-plot-matrix-for-ordered-categorical-data/ And Dieter, thanks for a great quote: "Assess independence, equalvariance and normality -in that order" (van Bell, Statistical rules of thumb)." Tal ----------------Contact Details:------------------------------------------------------- Contact me: Tal.Galili@gmail.com | 972-52-7275845 Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | www.r-statistics.com (English) ---------------------------------------------------------------------------------------------- On Mon, Apr 19, 2010 at 2:07 PM, Mona_m <purplemind@blueyonder.co.uk> wrote:> > Hi, > > I have just found this forum, and it looks like a great place to get some > help (I hope) > For my dissertation, which is due way too soon, I am doing a survey, > comparing attitudes of 2 independent groups, with 5 scale likert questions. > Basically I want to show if they have similar or different attitudes. I am > testing 4 hypotheses, and have in total about 20 questions. > > I have to say my statistic skills are very basic and very rusty, we had > some > lectures two years ago, where we were introduced to R. I looked through my > notes, and back then we did a one sample t-test to analyse likert type > questions. I believe I would need to do a 2 sample unpaired t-test. It > would be great if someone could give me some feedback if this test is the > most suitable one for my purpose, and maybe could explain to me what’s the > easiest way to do this in R? > > You would help me loads!! > Many thanks in advance > Mona > > -- > View this message in context: > http://n4.nabble.com/comparing-attitudes-of-2-groups-likert-scales-tp2015738p2015738.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >[[alternative HTML version deleted]]
David Winsemius
2010-Apr-19 15:09 UTC
[R] comparing attitudes of 2 groups / likert scales?
On Apr 19, 2010, at 10:08 AM, Tal Galili wrote:> Good luck in your work, > > The simple solution would be to run many non-paired wilcox on all > the 20 > questions (the way Dieter suggested). > In which case, make sure to adjust for multiple comparisons. Read > about it, > and see: > ?p.adjust > If you have some questions you can merge (by a simple mean of them), > it will > probably do you good (using PCA might be an option, but it could > also be an > over kill for you). > > > You might also be interested in plotting your data, here is a nice > simple > hack on how to display the Correlation scatter-plot matrix for your > data: > http://www.r-statistics.com/2010/04/correlation-scatter-plot-matrix-for-ordered-categorical-data/ > > > And Dieter, thanks for a great quote: "Assess independence, > equalvariance > and normality -in that order" (van Bell, Statistical rules of thumb)."If you are thinking of using that quote, you might want to check the spelling of his name. My memory is van Belle. -- David.> > > Tal > > > ----------------Contact > Details:------------------------------------------------------- > Contact me: Tal.Galili at gmail.com | 972-52-7275845 > Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il > (Hebrew) | > www.r-statistics.com (English) > ---------------------------------------------------------------------------------------------- > > > > > On Mon, Apr 19, 2010 at 2:07 PM, Mona_m > <purplemind at blueyonder.co.uk> wrote: > >> >> Hi, >> >> I have just found this forum, and it looks like a great place to >> get some >> help (I hope) >> For my dissertation, which is due way too soon, I am doing a survey, >> comparing attitudes of 2 independent groups, with 5 scale likert >> questions. >> Basically I want to show if they have similar or different >> attitudes. I am >> testing 4 hypotheses, and have in total about 20 questions. >> >> I have to say my statistic skills are very basic and very rusty, we >> had >> some >> lectures two years ago, where we were introduced to R. I looked >> through my >> notes, and back then we did a one sample t-test to analyse likert >> type >> questions. I believe I would need to do a 2 sample unpaired t- >> test. It >> would be great if someone could give me some feedback if this test >> is the >> most suitable one for my purpose, and maybe could explain to me >> what?s the >> easiest way to do this in R? >> >> You would help me loads!! >> Many thanks in advance >> Mona >> >> -- >> View this message in context: >> http://n4.nabble.com/comparing-attitudes-of-2-groups-likert-scales-tp2015738p2015738.html >> Sent from the R help mailing list archive at Nabble.com. >> >> ______________________________________________ >> R-help at r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, reproducible code. >> > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.David Winsemius, MD West Hartford, CT
David Winsemius wrote:> > > If you are thinking of using that quote, you might want to check the > spelling of his name. My memory is van Belle. > >Sorry, I thought I had corrected that before mailing. @BOOK{vanBelle2002, title = {Statistical rules of thumb}, publisher = {Wiley series in probability and statistics}, year = {2002}, author = {Gerald van Belle} } -- View this message in context: http://n4.nabble.com/comparing-attitudes-of-2-groups-likert-scales-tp2015738p2016083.html Sent from the R help mailing list archive at Nabble.com.
thanks a lot for your help!! I better get on with reading / working now! -- View this message in context: http://n4.nabble.com/comparing-attitudes-of-2-groups-likert-scales-tp2015738p2016398.html Sent from the R help mailing list archive at Nabble.com.
On 2010-04-19 9:15, Dieter Menne wrote:> > > David Winsemius wrote: >> >> >> If you are thinking of using that quote, you might want to check the >> spelling of his name. My memory is van Belle. >> >> > > Sorry, I thought I had corrected that before mailing. > > @BOOK{vanBelle2002, > title = {Statistical rules of thumb}, > publisher = {Wiley series in probability and statistics}, > year = {2002}, > author = {Gerald van Belle} > } >And a very fine book it is, too. Highly recommended. -- Peter Ehlers University of Calgary