Posted in 2019
view() provides a generic interface to the GUI function
View() in base R and RStudio. It makes it possible to extend it to
data objects of the classes “
descriptions”, and “
A method for “
data.set” objects allows to transfer these
objects more easiliy into the “tidyverse”, i.e. facilitates the use of
functions from these package ecosystem on data sets imported or created with
as_haven() function translates “
data.set” objects into
“tibbles” with that extra information that the “haven” package adds to
“tibbles” imported with the help of that package. This should allow to view
and post-process data imported with memisc more or less the same way as if
the data were imported with “haven”.
- 16 April 2019
Using data from the European Social Survey (ESS) I examine the impact of social class on electoral turnout, the support for left and social democratic parties and for right-wing populist parties. I find that the impact of class on voting for leftist and social democratic parties seems to have disappeared in some countries. Yet there is clear evidence that class has an impact on electoral turnout and on support for right-wing populist parties. The class-pattern of support for left/social-democratic parties tends to be weak in countries where electoral turnout has a particular strong class pattern. Further, changes in turnout affect left and social democratic parties more than other parties.
A new release 0.99.17.1 of my package memisc has been published on CRAN. The new release has the following improvements:
mtable() results now include a legend for significance symbols.
codeplan() function creates a data frame that describes the
structure of an
"data.set" object. Such
“codeplans” can be used to copy the “item structure” or code plan from
one object to another.
- 01 February 2019
A small number of top-level units in multilevel analysis is a problem that has worried comparativists for quite some time. In a paper that is forthcoming in the British Journal of Political Science (Elff, Schaeffer, Heisig, and Shikano 2020 (forthcoming)) my co-authors and I show that this problem can be satisfactorily addressed by only moderate modifications of common techniques of statistical inference: using restricted maximum likelihood (REML) instead of (ML) estimators (Patterson, and Thompson 1971) and by avoiding the assumption of asymptotic normality for the sampling distribution of coefficient estimates and assume a t-distribution instead. (Satterthwaite 1941; Kenward, and Roger 1997)