Tuesday 14th July      Wednesday 15th July      Thursday 16th July      Friday 17th July     

Friday 17th July, 09:00 - 10:30 Room: N-132

Methodological Aspects of the Left-right Self-placement Scale
Convenor	Ms Cornelia Zuell (GESIS )
Coordinator 1	Dr Evi Scholz (GESIS)

Session Details

The left-right dimension is a core element of political science research, serving as an instrument that supports citizens’ orientations in a complex political world by condensing political contents. Analyses of the left-right dimension are usually based on responses to a left-right self-placement scale, and sometimes to open-ended questions about the meaning of left and right. Left-right self-placement on a uni-dimensional scale is a standard type of survey question that measures respondents’ ideological orientations in a minimalist way. Aspects that are taken into account are the context and the way the scale is traditionally calibrated or methodological aspects, e.g., offering a midpoint or a “can’t choose” option, or cross-cultural differences in understanding the scale by the respondents.

The proposed session aims at bringing together national and comparative researchers interested in
- the design of the left-right scale (for example, scale range, scale labels, use of midpoint, use of permitted “don’t know”, mode effects);
- methodological aspects of response behavior to the left-right scale or to open-ended questions;
- the use of probe question, for example, about the meaning of left and right or about reasons for selecting the midpoint;
- response bias and social desirability;
- cross-national equivalence (left-right; liberal-conservative) and comparability;
- other methodological aspects related to the left-right self-placement.

Paper Details

1. 10-point vs. 11-point? Effects of Left-right Scale Design in a Cross-national Perspective
Mrs Cornelia Zuell (GESIS)
Dr Evi Scholz (GESIS)

Left-right self-placement is a standard question to measure respondents’ ideological orientation. Although the left-right scale is a standard instrument, the scale design is not standardized. One design aspect is the offer of a midpoint. This paper is about design effects on central left-right scale placement in a cross-national context. How do respondents answer without true scale midpoint? We tested the effects of a 10-point vs. an 11-point scale in a cross-national online survey. Preliminary results seem to show that respondents find a virtual center in the 10-point scale and that non-

2. Stability and meaning of the left-right scale in the Czech Republic
Dr Lukas Linek (Institute of Sociology, Czech Academy of Sciences)

This paper analyses merits and limits of usage of the left-right scale in the Czech Republic. This is done by the analysis of three surveys which used experimental and quasi-experimental design for this purpose. First, left-right self-placement question was asked in two different formats in one survey, once with 11 and once with 5 points. Second, left-right self-placement question was asked in the panel survey. In the third survey, left-right self-placement question was followed by an open-ended question concerning its meaning.

3. Stability without validity? The left-right scale in Iceland from 1983-2013.
Dr Hulda Thorisdottir (University of Iceland)

In this talk I present research examining how Icelandic voters have used the left-right scale over time. The Icelandic National Election Studies (ICENES) from 1987–2013 will be used to assess changes over time and how they relate to demographics and external political events. Results will be compared to the European Values Study, which unlike the ICNES, uses a scale without a midpoint. The relationship over time between the scale and political attitudes will be reported. Findings demonstrate overall, the left-right scale seems to have high reliability in the Icelandic context but its validity is not as evident.

4. Is the left-right alignment of parties outdated? The German case
Professor Andranik Tangian (Institute of Economic and Social Research in the Hans-Boeckler-Foundation)

The advocates of western democracy promote the viewpoint that the class division is becoming outdated. We disprove it with the German political space of 28 parties who participated in 2013 federal election. We consider the official party positions on 38 topical issues shortly before the election and locate the parties in this 38-dimensional political space. The PCA reveals that the parties constitute a thin ellipsoid, whose first two diameters, explaining 83.4% of the total variance, result in the party order which exactly corresponds to the left – right alignment. Thus, the left – right alignment of parties is not outdated