¹ Laver, Michael, Benoit, Kenneth and Garry, John, ‘Extracting Policy Positions from Political Texts Using Words as Data’, American Political Science Review, 97 (2003), 311–332CrossRefGoogle Scholar; Monroe, Burt L. and Maeda, Ko, ‘Talk’s Cheap: Text-Based Estimation of Rhetorical Ideal-Points’ (presented at the 21st Annual Summer Meeting, Society for Political Methodology, Stanford University, 2004)Google Scholar; Hopkins, Daniel and King, Gary, ‘A Method of Automated Nonparametric Content Analysis for Social Science’, American Journal of Political Science, forthcomingGoogle Scholar; Diermeier, Daniel et al. , ‘Language and Ideology in Congress’ (presented at the annual meeting of the Midwest Political Science Association, 2007)Google Scholar; Slapin, Jonathan B. and Proksch, Sven-Oliver, ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’, American Journal of Political Science, 52 (2008), 705–722CrossRefGoogle Scholar.

² Slapin, and Proksch, , ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’.Google Scholar

³ Attina, Fulvio, ‘The Voting Behaviour of the European Parliament Members and the Problem of the Europarties’, European Journal of Political Research, 18 (1990), 557–579CrossRefGoogle Scholar; Brzinski, Joanne Bay, ‘Political Group Cohesion in the European Parliament, 1989–1994’, in Carolyn Rhodes and Sonia Mazey, eds, The State of the European Union (London: Longman, 1995), pp. 64–83Google Scholar; Raunio, Tapio, The European Perspective: Transnational Party Groups in the 1989–1994 European Parliament (Sudbury, Mass.: Dartmouth/Ashgate, 1997)Google Scholar; Kreppel, Amie and Tsebelis, George, ‘Coalition Formation in the European Parliament’, Comparative Political Studies, 32 (1999), 933–966CrossRefGoogle Scholar; Hix, Simon, ‘Legislative Behaviour and Party Competition in the European Parliament: An Application of Nominate to the EU’, Journal of Common Market Studies, 39 (2001), 663–688CrossRefGoogle Scholar; Noury, Abdul, ‘Ideology, Nationality and Euro-Parliamentarians’, European Union Politics, 3 (2002), 33–58CrossRefGoogle Scholar; Noury, Abdul and Roland, Gerard, ‘More Power to the European Parliament?’, Economic Policy, 17 (2002):35, 280–310CrossRefGoogle Scholar; McElroy, Gail, ‘Committee Representation in the European Parliament’, European Union Politics, 7 (2006), 5–29CrossRefGoogle Scholar; Han, Jeong-Hun, ‘Analysing Roll Calls of the European Parliament: A Bayesian Application’, European Union Politics, 8 (2007), 479–507CrossRefGoogle Scholar.

⁴ Hix, Simon, Noury, Abdul and Roland, Gerard, ‘Dimensions of Politics in the European Parliament’, American Journal of Political Science, 50 (2006), 494–511CrossRefGoogle Scholar; Simon Hix, Abdul Noury and Gerard Roland, Democratic Politics in the European Parliament (Cambridge: Cambridge University Press, 2007)Google Scholar.

⁵ Carruba, Clifford J., et al. , ‘Off the Record: Unrecorded Legislative Votes, Selection Bias and Roll-Call Vote Analysis’, British Journal of Political Science, 36 (2006), 691–704.CrossRefGoogle Scholar

⁶ Carruba, et al. , ‘Off the Record’, p. 692.Google Scholar

⁷ In their sample (5th EP, 1999–2000) co-decision votes were significantly under-sampled: only 0.77 per cent of co-decision votes were by roll call, see Carruba, et al. , ‘Off the Record’.Google Scholar

⁸ Carruba, et al. , ‘Off the Record’, p. 702.Google Scholar

⁹ Hix, , Noury, and Roland, , Democratic Politics in the European Parliament, p. 166.Google Scholar

¹⁰ McElroy, Gail, ‘Legislative Politics as Normal? Voting Behaviour and Beyond in the European Parliament’, European Union Politics, 8 (2007), 433–448, p. 437.CrossRefGoogle Scholar

¹¹ There are other methodological approaches to studying positions of national parties in the European Union, but they do not focus specifically on parliamentary behaviour. These approaches include expert surveys (Hooghe, Liesbet and Marks, Gary, ‘Chapel Hill 2002 Expert Survey on Party Positioning on European Integration’, http://www.unc.edu/ (2002)Google Scholar; Marks, Gary, Hooghe, Liesbet, Nelson, Moira and Edwards, Erica, ‘Party Competition and European Integration in the East and West – Different Structure, Same Causality’, Comparative Political Studies, 39 (2006), 155–175CrossRefGoogle Scholar; Benoit, Kenneth and Laver, Michael, Party Policy in Modern Democracies (London: Routledge, 2006)Google Scholar; Benoit, Kenneth and McElroy, Gail, ‘Party Groups and Policy Positions in the European Parliament’, Party Politics, 13 (2007), 5–28Google Scholar; Steenbergen, Marco R. and Marks, Gary, ‘Evaluating Expert Judgments’, European Journal of Political Research, 46 (2007), 347–366CrossRefGoogle Scholar); and there are also MEP surveys (Farrell, David et al. , ‘EPRG 2000 and 2006 MEP Surveys Dataset’, http://www.lse.ac.uk/collections/EPRG/ (2006))Google Scholar, mass survey research (Hix, Simon and Lord, Christopher, Political Parties in the European Union (Basingstoke, Hants.: Macmillan, 1997)CrossRefGoogle Scholar ), interest group ratings, and European election manifestos (Gabel, Matthew J. and Hix, Simon, ‘Defining the EU Political Space: An Empirical Study of the European Elections Manifestos, 1979–1999’, Comparative Political Studies, 35 (2002), 934–964CrossRefGoogle Scholar). However, none of these approaches actually studies the revealed preferences of the MEPs themselves. In addition, the alternative approaches have some methodological problems. For instance, McElroy points out that elite surveys suffer from sample response issues, preference measures on the basis of constituency characteristics are difficult given the weak electoral connection in the European Parliament, and interest group ratings tend to have selective samples, thus potentially exaggerating extreme positions (McElroy, , ‘Legislative Politics as Normal?’, p. 437Google Scholar).

¹² Corbett, Richard, Jacobs, Francis and Shackleton, Michael, The European Parliament, 5th edn (London: John Harper, 2003), p. 145.Google Scholar

¹³ Corbett, , Jacobs, and Shackleton, , The European Parliament, 5th edn.Google Scholar

¹⁴ See Rule 149 of the EP Rules of Procedure.

¹⁵ Judge, David and Earnshaw, David, The European Parliament (Basingstoke, Hants.: Palgrave Macmillan, 2003), p. 239.Google Scholar

¹⁶ Corbett, , Jacobs, and Shackleton, , The European Parliament.Google Scholar

¹⁷ To automate this task, we wrote a computer script which automatically extracted the agenda item and the number of speeches from the information available on the EP website.

¹⁸ Kreppel, and Tsebelis, , ‘Coalition Formation in the European Parliament’Google Scholar; Marks, Gary, Wilson, Carole and Ray, Leonard, ‘National Political Parties and European Integration’, American Journal of Political Science, 46 (2001), 585–594CrossRefGoogle Scholar; Gabel, and Hix, , ‘Defining the EU Political Space’Google Scholar; Aspinwall, Mark, ‘Preferring Europe: Ideology and National Preferences on European Integration’, European Union Politics, 3 (2002), 81–111CrossRefGoogle Scholar; Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’; Hix, Noury and Roland, Democratic Politics in the European ParliamentGoogle Scholar.

¹⁹ Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’; Hix, Noury and Roland, Democratic Politics in the European Parliament.Google Scholar

²⁰ Benoit, Kenneth, Laver, Michael and Mikhaylov, Slava, ‘Treating Words as Data with Error: Uncertainty in Text Statements of Policy Positions’, American Journal of Political Science, 53 (2009), 495–513.CrossRefGoogle Scholar

²¹ We thank an anonymous referee for pointing this out to us.

²² Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’.Google Scholar

²³ Budge, Ian, Klingemann, Hans-Dieter, Volkens, Andrea, Bara, Judith and Tanenbaum, Eric, Mapping Policy Preferences: Estimates for Parties, Electors, and Governments 1945–1998 (Oxford: Oxford University Press, 2001).Google Scholar

²⁴ Benoit, and McElroy, , ‘Party Groups and Policy Positions in the European Parliament’, p. 22.Google Scholar

²⁵ Slapin, and Proksch, , ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’. Wordfish is implemented in R and available at www.wordfish.org.Google Scholar

²⁶ Laver et al., ‘Extracting Policy Positions from Political Texts Using Words as Data’; Proksch, Sven-Oliver and Slapin, Jonathan B., ‘Institutions and Coalition Formation: The German Election of 2005’, West European Politics, 29 (2006), 540–559CrossRefGoogle Scholar; Slapin, and Proksch, , ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’Google Scholar; Hug, Simon and Schulz, Tobias, ‘Left–Right Positions of Political Parties in Switzerland’, Party Politics, 13 (2007), 305–330CrossRefGoogle Scholar.

²⁷ Laver, Michael and Benoit, Kenneth, ‘Locating TDs in Policy Spaces: Wordscoring Dail Speeches’, Irish Political Studies, 17 (2002), 59–73CrossRefGoogle Scholar; Laver et al., ‘Extracting Policy Positions from Political Texts Using Words as Data’; Monroe and Maeda, ‘Talk’s Cheap: Text-Based Estimation of Rhetorical Ideal-Points’; Giannetti, Daniela and Laver, Michael, ‘Policy Positions and Jobs in the Government’, European Journal of Political Research, 44 (2005), 91–120CrossRefGoogle Scholar; Diermeier et al., ‘Language and Ideology in Congress’.

²⁸ Laver, Michael, Benoit, Kenneth and Sauger, Nicolas, ‘Policy Competition in the 2002 French Legislative and Presidential Elections’, European Journal of Political Research, 45 (2006), 667–697.CrossRefGoogle Scholar

²⁹ Benoit, Kenneth, et al. , ‘Measuring National Delegate Positions at the Convention on the Future of Europe Using Computerized Word Scoring’, European Union Politics, 6 (2005), 291–313.CrossRefGoogle Scholar

³⁰ McGuire, Kevin T. and Vanberg, Georg, ‘Mapping the Policies of the U.S. Supreme Court: Data, Opinions, and Constitutional Law’ (prepared for delivery at the Annual Meeting of the American Political Science Association, Washington, D.C., 2005).Google Scholar

³¹ Slapin, and Proksch, , ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’.Google Scholar

³² We have applied this model to compare election manifestos from German parties between 1990 and 2005. We found that the technique is able to recover party positions estimated by other techniques (e.g. expert surveys and hand-coding of manifestos). Furthermore, the positions reflect important changes in the party system, in particular a rightward movement of the major social-democratic party, the SPD, in the 1990s. We could produce estimates over time by making the assumption that word weights are time-invariant (see Slapin and Proksch, ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’).

³³ Laver, et al. , ‘Extracting Policy Positions from Political Texts Using Words as Data’Google Scholar. While the technique has mostly been used to study political manifestos, it has been applied to legislative speeches as well (Laver, and Benoit, , ‘Locating TDs in Policy Spaces: Wordscoring Dail Speeches’Google Scholar; Laver et al. , ‘Extracting Policy Positions from Political Texts Using Words as Data’Google Scholar; Giannetti, and Laver, , ‘Policy Positions and Jobs in the Government’Google Scholar). Laver and Benoit use speeches from a confidence debate in the Irish Dáil in October 1991 over the future of the incumbent coalition government. They postulate a ‘pro- versus anti-government’ dimension and use the speech of the prime minister and of the opposition leaders as reference texts. The resulting placement of political parties on a scale of government versus opposition ‘is readily recognisable by any observer of Irish politics’ (Laver, et al. , ‘Extracting Policy Positions from Political Texts Using Words as Data’, p. 327Google Scholar).

³⁴ We did validate the Wordfish algorithm presented here with the Wordscores technique. To do so, we anchored the Wordfish dimension in Wordscores by using the speeches from the most extreme parties identified by Wordfish as reference texts. We estimated the Wordscores positions using a slightly updated version of the algorithm (Martin, Lanny W. and Vanberg, Georg, ‘A Robust Transformation Procedure for Interpreting Political Texts’, Political Analysis, 16 (2008), 93–100CrossRefGoogle Scholar). As expected, the results correlate very highly across all languages between the two techniques (correlation of 0.91 or higher).

³⁵ This number excludes new member state MEPs joining in 2004 for only a few weeks before the next election, but includes the presidents and vice-presidents of the EP who deliver mostly procedural speeches.

³⁶ Laver, et al. , ‘Extracting Policy Positions from Political Texts Using Words as Data’, p. 327.Google Scholar

³⁷ The inferences will only be valid for this total set of speeches and do not necessarily apply for subsets of speeches (e.g. specific policy areas).

³⁸ Hix, and Lord, , Political Parties in the European UnionGoogle Scholar; Raunio, The European Perspective; Kreppel and Tsebelis, ‘Coalition Formation in the European Parliament’; Kreppel, Amie, The European Parliament and Supranational Party System (Cambridge: Cambridge University Press, 2002Google Scholar); Hix, Simon, ‘Parliamentary Behavior with Two Principals: Preferences, Parties, and Voting in the European Parliament’, American Journal of Political Science, 46 (2002), 688–698CrossRefGoogle Scholar; Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’; Hix, Noury and Roland, Democratic Politics in the European ParliamentGoogle Scholar.

³⁹ Benoit, and McElroy, , ‘Party Groups and Policy Positions in the European Parliament’.Google Scholar

⁴⁰ Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’Google Scholar. There are no independent measures of ideology available at the individual level with the exception of the EPRG survey of MEPs themselves, which suffers from low response rates (Farrell et al., ‘EPRG 2000 and 2006 MEP Surveys Dataset’). If the researchers wish to compare roll-call positions with expert survey positions or CMP data, they must aggregate up to the level of national party.

⁴¹ We thank one of the anonymous referees for pointing this out.

⁴² We exclude new member state MEPs as they were only represented in the 5th European Parliament by nominated members for a few weeks between the date of enlargement (1 May 2004) and the elections to the 6th European Parliament (June 2004).

⁴³ We used Perl scripts to automate this task. The speech archive of the European Parliament is available at http://www.europarl.europa.eu/activities/archives/cre/search.do?language=EN, last consulted in April 2008.

⁴⁴ Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’.Google Scholar

⁴⁵ We use Will Lowe’s jfreq program, available at http://www.williamlowe.net/software/.

⁴⁶ The English Wordfish results using words mentioned by at least ten parties correlate with results using words mentioned by at least thirty parties at 0.99.

⁴⁷ The EU has fewer official languages (twenty-three) than member states (twenty-seven). German is spoken in Germany and Austria, English in the United Kingdom and Ireland, Greek in Greece and Cyprus, and Belgium and the Netherlands share common languages with their neighbouring countries.

⁴⁸ Corbett, , Jacobs, and Shackleton, , The European Parliament, p. 34Google Scholar; Judge, and Earnshaw, , The European Parliament, p. 163Google Scholar.

⁴⁹ These obligatory tasks result in considerable costs in the EU. In 2003, prior to the enlargement, EU institutions spent a combined 549 million euros on translation, and following enlargement to twenty-five members in 2004, the expense rose to an estimated 807 million euros per year, or approximately 1.78 euros per EU citizen (see European Commission Memo 05/10, January 2005, http://europa.eu/rapid/pressReleasesAction.do?reference=MEMO/05/10). In 2005, after enlargement by ten new member states, the EP had over one million pages of parliamentary documents translated. In addition, the EP provided interpretation services totalling 85,340 work days (see European Parliament Budget 2005, http://www.europarl.europa.eu/pdf/budget/rapportpublic2005_en.pdf).

⁵⁰ There are several reasons to believe that translation may affect the output of computer-based content analysis. The German language has a particular feature that allows the compounding of words to create new ones. For example, the phrase ‘workers’ rights’ is described by two words in English, three in French (‘droits des travailleurs’), but only one in German (‘Beschäftigtenrechte’). Moreover, translation itself possibly adds error to the data, which could lead to different results across language. Translation theorists have suggested that one can view translation as a series of choices that can be modelled as a decision tree (Levý, Jiří, ‘Translation as a Decision Process’, in To Honor Roman Jakobson II (The Hague: Mouton, 1967), 1171–1182Google Scholar). Each language presents the translator with a set of possible choices about which particular translation to choose. A stylistic choice a translator makes at one node may affect how he or she translates the rest of the text. This means that additional error may enter into the data both because different languages offer different choice sets and translators will make different decisions within those choice sets. Thus, we might get different results because some languages use different words and grammatical structures to express exactly the same content and because translators might follow different strategies in translation.

⁵¹ Appendix A shows the national party estimates using the English translations. The estimation is based on 4,859 unique words in English, 6,248 unique words in French and 7,369 unique words in German.

⁵² In contrast, speeches from a party whose native language is not English, German or French, are translated into all three languages.

⁵³ The estimation is based on 4,765 unique words.

⁵⁴ We can also calculate the average standard deviation of national parties based on the results from the individual level analysis. For those national parties with more than one MEP (n = 71), the average standard deviation of positions is 0.68, which is about two-thirds of the overall standard deviation of the positions (fixed at 1). If we include national parties with one MEP (n = 103), the mean standard deviation of the positions across national parties drops to 0.47. It would be interesting to explore the reasons for the variation of individual-level positions in future research.

⁵⁵ Hooghe, and Marks, , ‘Chapel Hill 2002 Expert Survey on Party Positioning on European Integration’; Marks et al., ‘Party Competition and European Integration in the East and West’; Steenbergen and Marks, ‘Evaluating Expert Judgments’.Google Scholar

⁵⁶ In addition to missing several small parties, the UNC data does not include parties from Luxembourg.

⁵⁷ Benoit, and Laver, , Party Policy in Modern Democracies.Google Scholar

⁵⁸ Benoit, and Laver, , Party Policy in Modern Democracies, p. 229.Google Scholar

⁵⁹ Benoit, and Laver, , Party Policy in Modern Democracies, p. 131Google Scholar. The scales used for these questions range between 1 and 20. The Benoit/Laver survey includes other measures of EU support; however, they all correlate highly and produce the same result.

⁶⁰ Although most of the missing estimates are for smaller parties, positions for parties from Ireland and France are missing entirely from the survey on these questions.

⁶¹ Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’.Google Scholar

⁶² Hix, , Noury, and Roland, , ‘Dimensions of Politics in the European Parliament’; Hix, Noury and Roland, Democratic Politics in the European Parliament.Google Scholar

⁶³ Cook, J. R. and Stefanski, L. A., ‘Simulation-Extrapolation Estimation in Parametric Measurement Error Models’, Journal of the American Statistical Association, 89 (1994), 1314–1328.CrossRefGoogle Scholar

⁶⁴ This method corrects for measurement error of the independent variables only. The dependent variable, the positions estimated from word counts in speeches, is also measured with error. Wordfish allows researchers to estimate the fundamental uncertainty surrounding the positions via a parametric bootstrap. We have shown elsewhere through simulations that the confidence intervals of the estimated positions in Wordfish significantly decrease as the number of unique words used in the analysis increases (Slapin and Proksch, ‘A Scaling Model for Estimating Time-Series Party Positions from Texts’). Because we use several thousand unique words to estimate the positions, the confidence intervals of those estimates are rather small (see Appendix B). Moreover, measurement error in the dependent variable will not cause the kind of attenuation bias in the regression coefficients that we worry about. (Poole, Keith T., ‘Measuring Bias and Uncertainty in Ideal Point Estimates via the Parametric Bootstrap’, Political Analysis, 12 (2004), 105–127Google Scholar). Alternatively, one could apply Bayesian statistical analysis to estimate positions and their uncertainty (Han, ‘Analysing Roll Calls of the European Parliament’).

⁶⁵ Benoit, , Laver, and Mikhaylov, , ‘Treating Words as Data with Error: Uncertainty in Text Statements of Policy Position’, American Journal of Political Science, 53 (2009), 495–513.CrossRefGoogle Scholar

⁶⁶ To estimate the SIMEX model as implemented in R, we use as the measurement error the mean standard deviation of responses across all parties.

⁶⁷ It is possible to generate uncertainty estimates for Nominate using a parametric bootstrap (Lewis, Jeffrey B. and Poole, Keith T., ‘Measuring Bias and Uncertainty in Ideal Point Estimates via the Parametric Bootstrap’, Political Analysis, 12 (2004), 105–127CrossRefGoogle Scholar). Alternatively, one could apply Bayesian statistical analysis to estimate positions and their uncertainty (Han, ‘Analysing Roll Calls of the European Parliament’).

⁶⁸ Average net contributions per capita for 1999–2003 are operating budgetary balances taken from the 2005 EU Commission report on the allocation of EU expenditures per member state divided by population, p. 138 (http://ec.europa.eu/budget/documents/revenue_expenditure_en.htm). We include those years of the 5th European Parliament for which the budget lists the balances for EU-15 member states only.

⁶⁹ Steenbergen, Marco and Jones, Bradford S., ‘Modeling Multilevel Data Structures’, American Journal of Political Science, 46 (2002), 218–237, p. 233.CrossRefGoogle Scholar

⁷⁰ Gelman, Andrew and Hill, Jennifer, Data Analysis Using Regression and Multilevel/Hierarchical Models (Cambridge: Cambridge University Press, 2007).Google Scholar

⁷¹ GDP per capita is significant in the models using the UNC survey data, which excludes Luxembourg. Luxembourg is an outlier on GDP per capita, so excluding it from the analysis alters the results.

⁷² To preserve space we only report the predicted values for the country-level variables that attain statistical significance in the hierarchical model found in Appendix B.