[R] Statistical Test

Jim Lemon drjimlemon at gmail.com
Mon Jul 11 03:49:51 CEST 2016


Hi Julia,
You seem to be looking for a test for trend in proportions in the
first question. Have a look at this page:

http://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/R/R6_CategoricalDataAnalysis/R6_CategoricalDataAnalysis6.html

The second question may require GLMs using experimental condition as a
predictor and proportion of each type of error as the response. Are
the groups balanced?

Jim


On Sun, Jul 10, 2016 at 10:34 PM, Julia Edeleva
<psychexpert1992 at gmail.com> wrote:
> Dear R-community,
>
> Thanks for replying to my previous post. I would need some more help,
> thoug.
>
> I am performing statistical analysis on chidlren's accuracy rates  as a
> dependent variable and two predictor variables with two levels each (syntax
> - subject vs object; internal NP position - pre vs post).
>
> As an outcome of my study, children committed 3 types of errors. I want to
> compare whether children committed significantly more errors of one type as
> compared to the other two types, i.e. test the scale *error 1 > error 2 >
> error 3 (">" is "more than").*  Which statistical test is most appropriate?
>
> Furthermore, I want to know whether one particular type of error is more
> common in one experimental condition than in the other, i.e. test
> whether *error
> 1 in condition 1 is more common than error 1 in condition 2*.
>
> Thnx a lot
>
> Julia Edeleva
>
> *Compare different types of errors in chidren's performance. Statistical
> Test? - ResearchGate*. Available from:
> https://www.researchgate.net/post/Compare_different_types_of_errors_in_chidrens_performance_Statistical_Test
> [accessed Jul 10, 2016].
>
>         [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> 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.



More information about the R-help mailing list