Welcome to Martin Elff’s website!¶
My name is Martin Elff, I am a political scientist with research interests in political behaviour, political sociology, comparative politics, and the methodology of political and social sciences. Besides teaching, research, academic governance, I also wrote some software, mostly in R.
Since 2015 I am a professor of political sociology at Zeppelin University (Friedrichshafen, Germany).
You can also find me on
- ORCID: 0000-0001-9032-9739
- Github: http://github.com/melff
- Web of Science: J-1604-2019
- Twitter: @martin_elff
What you will find here
- my publications (including links to replication material, where available)
- scientific data that I contributed
- software packages I created
- subjects I have been teaching
- news about publications and software
- contact information
- my vita
Below you find a quick overview of these topics.
Elff, Martin, Jan Paul Heisig, Merlin Schaeffer, and Susumu Shikano. 2021. “Multilevel Analysis with Few Clusters: Improving Likelihood-based Methods to Provide Unbiased Estimates and Accurate Inference”. British Journal of Political Science 51(1): 412-426.
Elff, Martin. 2020. Data Management with R: A Guide for Social Scientists. London: SAGE Publications.
Elff, Martin and Sebstian Ziaja. 2018. “Method Factors in Democracy Indicators”. Politics and Governance 6(1): 92-104.
Elff, Martin. 2013. “A Dynamic State-Space Model of Coded Political Texts”. Political Analysis 21(2): 217-232.
Elff, Martin. 2009. “Social Divisions, Party Positions, and Electoral Behaviour”. Electoral Studies 28(2): 297-308.
There are a few software packages I wrote for helping me in my research, creating this website, or which I just wrote for fun and out of curiosity (however I have only little time for this).
On this website you find information about the following R packages in particular:
- memisc: Management of Survey Data and Presentation of Analysis Results
- mclogit: Multinomial Logit Models, with or without Random Effects or Overdispersion
- munfold: Metric Multidimensional Unfolding in R
- mpred: Generic Predictive Margins
- manifestos: Spatial Modelling of Party Manifestos (and other political texts)
- iimm: Improved Inference for Multilevel Models with Few Clusters
- EMfit: An Infrastructure for Latent Variable Model Fitting using EM Algorithms
My are research interests in the following areas:
Political behaviour, in particular electoral behaviour and its relation with social structure. More specifically:
- class voting
- religious voting
- long term change
- the influence of parties’ (changing) positions
- with a focus on Germany
- with a focus on (Western) Europe
- Reconstructing the political positions of parties
Quantitative methods of the social sciences
- Multilevel modelling
- Modelling of qualitative/categorical dependent variables
- Cross-level inference
Throughout my career I taught a couple of subjects, including
Political sociology and political behaviour
- Introduction to political sociology
- Attitude formation and public opinion
- Electoral behaviour
- German politics
- West European politics
- Parties and party systems
Socials science research methods
- Research design and methods (at various levels)
- Data analysis (at various levels)
- Data analysis and graphics with R
- Mathematical tools for social scientists
- 04 February 2023 - RKernel: Yet another R kernel for Jupyter
- 25 November 2020 - New Book: Data Management in R
- 10 August 2020 - A “normalizing” re-alignment of voting behaviour in East Germany?
- 13 May 2020 - Now available as open access article: “Multilevel Analysis with Few Clusters”
- 22 November 2019 - New memisc release 0.99.20.1 improves compatibility with RStudio and “tidyverse”
- 23 March 2019 - New memisc release 0.99.17.1 facilitates checking your data
For a more extensive list see the News page.
About this website
This site is built with Sphinx, a generator for documentation and static websites, using a heavily modified variant of the insipid theme. The website builds on flat files in restructured text format and in some instances on Jupyter notebooks (in particular the support material for my recent book).
Because it is built with Sphinx, there is a bit of Python programming involved in this website. In particular I wrote a few Python modules that extend Sphinx for the purposes of this website, e.g.
- attaching files to webpages
- creating bibliographies and linking publication details from BibTeX files
- improving the navigation of the site
- providing an infrastructer for blog posts (with much inspiration from `ABlog`__)
I will share these extensions as soon as I found the time to document them appropriately.