Minority voting and representation The impact of religion, migration background and gender on voter preferences for in- and out-group politicians in France, Germany and the Netherlands Sanne van Oosten
Sanne van Oosten Minority voting and representation The impact of religion, migration background and gender on voter preferences for in- and out-group politicians in France, Germany and the Netherlands
Author: Sanne van Oosten Cover Illustration: Naomi Meelker Cover Lay-out: Ilse Modder (www.ilsemodder.nl) Lay-out: Ilse Modder (www.ilsemodder.nl) Printing: Gildeprint (www.gildeprint.nl) ISBN: 978-94-6496-230-7 © 2024, Sanne van Oosten, The Netherlands. All rights are reserved. No part of this thesis may be reproduced, distributed, stored in a retrieval system, or transmitted in any form or by any means without citing the author.
Minority voting and representation The impact of religion, migration background and gender on voter preferences for in- and out-group politicians in France, Germany and the Netherlands ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. ir. P.P.C.C. Verbeek ten overstaan van een door het College voor Promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op vrijdag 11 oktober, 2024, 16.00 uur door Sanne van Oosten geboren te Berkeley, California, United States of America
Promotiecommissie Promotores: dr. L.M. Mügge Universiteit van Amsterdam dr. F.F. Vermeulen Universiteit van Amsterdam Co-promotores: dr. A. Hakhverdian Universiteit van Amsterdam dr. D.J. van der Pas Universiteit van Amsterdam Overige leden: prof. dr. L. Morales Sciences Po Paris dr. S. Azabar Radboud Universiteit prof. dr. S.L. de Lange Universiteit van Amsterdam prof. dr. T.W.G. van der Meer Universiteit van Amsterdam dr. G. Schumacher Universiteit van Amsterdam Faculteit der Maatschappij- en Gedragswetenschappen This research is supported through Liza Mügge’s NWO-VIDI grant (Grant Number 016. Vidi.175.355) and Daphne van der Pas’ NWO-VENI grant (Grant Number 451-17-025).
For my family
Table of content Summary (English) Samenvatting (Nederlands) Acknowledgements Publication overview Open science information Introduction Unraveling voter preferences for in-group politicians Chapter 1 Meta-analysis of race/ethnicity in candidate experiments Chapter 2 Favoring in-group politicians and the representation of Muslims Chapter 3 Voting for DENK and ethnic/religious identification Chapter 4 Stereotyping or projection in expectations of Muslim politicians Conclusion The barriers to descriptive and substantive representation of Muslims References Appendix: online 10 11 12 17 19 21 43 65 91 115 139 149
10 Summary Who prefers politicians with whom they share the same religion, migration background and/or gender and why? Many voters and political observers assume that minority voters tend to favor politicians from their own minority groups, but this phenomenon has rarely been investigated in the European context. In this dissertation, I employ a theoretical framework that integrates concepts from representation theory, social identity theory, and heuristics theory to study in-group voting patterns among both minority and majority citizens in France, Germany, and the Netherlands. I ran a meta-analysis, subsequently gathered original data by sampling 3,049 respondents, the majority of whom belong to minority groups and ran an array of conjoint experiments presenting respondents with politicians and asking them about their preferences. The analysis reveals that shared migration background and gender have minimal effects on voter choices, while shared religious identity significantly influences voter preferences. Notably, self-identifying Muslim voters display a clear preference for politicians who share their religious identity, with Muslims in the Netherlands showing the most significant preference for in-group politicians, followed by Germany and France. Focusing on the Dutch case, I investigate why Dutch Muslims vote for DENK, a Dutch political party predominantly led by Muslim politicians who claim to represent Muslims in the Netherlands. I find that political issues and experiences with discrimination do not drive voting for DENK as much as Muslim in-group favoritism does. Additionally, non-religious voters exhibit in-group preferences similar to those of Muslim voters. To understand why non-religious voters prefer their in-group, particularly over the Muslim out-group, I explore the dynamics of stereotyping and projection onto Muslim politicians. Both mechanisms play a role in shaping expectations of Muslim politicians among the mostly non-Muslim electorate, and I call for further research on heuristics to understand in- and out-group voting dynamics. These findings provide insights into the complexities of achieving diverse descriptive and substantive representation and underscore the urgent need to improve the representation of Muslims in European politics. Keywords Descriptive representation, gender, heuristics, migration background, Muslim politicians, projection, religion, social identity, stereotyping, substantive representation.
11 Samenvatting Wie heeft een voorkeur voor politici met wie ze dezelfde religie, migratieachtergrond en/ of geslacht delen, en waarom? Veel kiezers, opiniemakers en politieke commentatoren gaan ervan uit dat minderheidsgroepen de voorkeur hebben voor politici uit hun eigen minderheidsgroep, maar dit fenomeen is zelden onderzocht in de Europese context. In dit proefschrift maak ik gebruik van een theoretisch kader dat concepten integreert uit de theorie van vertegenwoordiging, de sociale identiteitstheorie en de heuristische theorie om stemmen voor de eigen groep (in-group) te bestuderen onder zowel minderheids- als meerderheidsgroepen in Frankrijk, Duitsland en Nederland. Ik heb een uitgebreide meta-analyse uitgevoerd, en vervolgens originele gegevens verzameld onder steekproeven van 3.049 respondenten, waarvan het merendeel tot minderheidsgroepen behoort. Ook heb ik een reeks conjoint-experimenten uitgevoerd waarbij respondenten politici kregen voorgelegd en hun voorkeur konden aangeven. De analyses onthullen dat een gedeelde migratieachtergrond en geslacht minimale invloed hebben op kiezerskeuzes, terwijl een gedeelde religieuze identiteit van aanzienlijke invloed is op kiezersvoorkeuren. Kiezers die zichzelf als moslim identificeren vertonen een duidelijke voorkeur voor politici die hun religieuze identiteit delen. Hierbij vertonen moslims in Nederland de grootste voorkeur voor politici van de eigen groep, gevolgd door Duitsland en Frankrijk. Vervolgens ga ik dieper in op de Nederlandse situatie en onderzoek ik waarom Nederlandse moslims op DENK stemmen, een politieke partij die voornamelijk wordt geleid door moslimpolitici die beweren moslims in Nederland te vertegenwoordigen. Ik ontdek dat beleidsstandpunten en ervaringen met discriminatie weinig invloed hebben op stemmen voor DENK, maar ik ontdek wel dat voorkeuren voor de eigen groep aanzienlijke invloed heeft op stemmen voor DENK onder moslims. Daarnaast vertonen niet-religieuze kiezers vergelijkbare voorkeuren voor de eigen groep als moslimkiezers. Om te begrijpen waarom niet-religieuze kiezers de voorkeur geven aan de eigen groep, met name boven islamitische politici, onderzoek ik de dynamieken stereotypering en projectie. Beide mechanismen spelen een rol bij het vormgeven van verwachtingen van islamitische politici onder het grotendeels niet-islamitische electoraat, en ik pleit voor verder onderzoek naar heuristieken om stemkeuze te begrijpen. Deze bevindingen bieden inzicht in de complexiteit van het bereiken van diversiteit in politieke vertegenwoordiging en benadrukken de dringende noodzaak om de vertegenwoordiging van moslims in de Europese politiek te verbeteren. Zoektermen Descriptieve vertegenwoordiging, gender, heuristieken, migratie-achtergrond, islamitische politici, projecteren, religie, sociale identiteit, stereotypering, substantieve vertegenwoordiging.
12 Acknowledgements Standing on the shoulders of giants is a fundamental aspect of academia. This dissertation’s reference section lists the previous research I’ve built upon, but just as crucial are the many individuals who shaped my ideas, cheered me on, and lent a helping hand. Completing a PhD with children, particularly through a pandemic, requires tremendous support. It’s impossible to do it alone. It takes a village to finish a PhD. Here’s to my village, my village of giants. A heartfelt thank you to my parents for raising me to think critically, questioning my youthful assumptions and encouraging me to doubt everything. Our conversations at the dinner table or on the couch have profoundly shaped how I see the world around me. My father, on the one hand, was the first to introduce me to positivist thinking. From a young age, I was enthralled by his stories of experiments with gravity and plant growth, and we had fun together rebuilding the solar system and exploring scientific concepts. My mother, on the other hand, introduced me to interpretivist thinking. She made me aware of how language shapes our thinking and actions, the metaphors we live by, emphasizing that while our perceptions might be constructed, they are very real in their consequences. She was the first to make me think critically about sexism, racism, and classism. Our reflections on the role of religion in society have also influenced my research, a theme in my thinking that has its reflection in this PhD. My PhD-supervisors made all of this possible, from beginning to end. Liza Mügge, thank you for dedicating so much time, effort, and funding to my dissertation. You went out of your way, always pushing me to dream big. One moment when I felt especially cherished during my PhD journey was when I received feedback on my 8-month paper. I remember reading it repeatedly, studying and pondering your handwritten notes. I physically carried that feedback with me for months, referring to it often and feeling incredibly fortunate to have a promotor who invested so much time and effort into helping me succeed and develop my ideas. Floris Vermeulen, thank you for bringing peace and order to my PhD when I needed it most and for guiding it to completion. You have the unique ability to know what I need and when I need it. Your feedback was always clear, concise, and in the right place and at the right time. Armèn Hakhverdian, we had some wonderful lunches and drinks. Thanks to you, I have learned to focus on what is essential, in academic writing, argumentation and just life in general. Daphne van der Pas, you have pushed me to be a better and more thorough researcher, never letting me get away with anything wrong or half-baked. Your attention to detail and commitment to academic rigor have been both challenging and inspiring, and I will recommend anyone to work with you. My work with Boris van der Ham also significantly influenced my thinking. The discussions we had while he was writing his book “De Koning Kun Je Niet Spelen” in 2014 shaped my understanding of representation. At the time, I didn’t realize that
13 the central tenet his book paralleled Pitkin’s symbolic representation and Saward’s emphasis on audience for a representative claim to be accepted. While working at WOMEN Inc., I came into contact with Chafina Ben Dahman, who helped me when I was in trouble and welcomed me as the first employee into her amazing Rose Stories. I often reflect on that time. Rose Stories’ mission was to represent underrepresented groups, such as Muslim women, in the media through children’s books, television, movies, and more. I will never forget the discussions I had with Chafina Ben Dahman, Nadia Zerouali, and Rachida el Moussaoui about representation. Without the ideas developed during those conversations, I might never have secured this PhD position in the first place. Additionally, Chafina introduced me to Astrid Sy, Houria el Moussaoui, Mimoun Oaïssa, Nadia Moussaid, Rahma el Mouden, Raja Felgata, Rhimou el Ahmadi, Saloua el Moussaoui, Soundos el Ahmadi, and Yousef Gnaoui. All of these amazing individuals have inspired the work that lies before you today, through their insights, experiences and the luck I had to have many long conversations with them about the representation of minorities. Before I began my PhD journey, there were more who warmly encouraged me and welcomed me into the program group. Wouter van der Brug brainstormed potential PhD positions with me, Maria Kranendonk discussed my research proposal, Natalie Welfens provided feedback on my application, and Ellis Aizenberg generously shared her experiences in a heartfelt conversation. Gijs Schumacher and Roderik Rekker not only encouraged me to pursue a PhD but also wrote glowing letters of recommendation. Marcel Maussen consistently believed in my potential, he protected me in the junior-lecturer jungle, he stressed me out about all the people who were passing me left and right, and he put in a good word when I needed it the most. He even entrusted me with the role of coordinator for the methodology thread, a position for which I initially felt underqualified, but which gave me just the confidence I needed to dare to apply for the PhD-position that I have now completed. In the early stages of my PhD journey, designing my survey experiment was my complete fixation, and I owe immense gratitude to those who provided invaluable feedback while I was “on tour” to gather feedback. Uğur Aytaç offered insights on Turkish names and taught me so much about Turkish politics; Ebe Ouatera provided feedback on French names. I also deeply appreciate the contributions of Laura Mulder, Gijs Schumacher, Bert Bakker, Mike Medeiros, Wouter van der Brug, Marcel Hanegraaff, Matthijs Rooduijn, and Maria Kranendonk. A special mention goes out to Catherine de Vries, who generously offered feedback over lunch, focusing on the expectations voters have of politicians, laying the basis for my first published article. Evelyn Ersanilli provided precise feedback on the mechanisms tested, while Sarah Kalaï, Aya Elyamany and Mo Akachar’s insightful conversations shaped the survey design and gave me confidence in its resonance with Muslim and Moroccan respondents. Additionally, I extend thanks to Uğur Aytaç, Manal Kartoubi, Mike Medeiros, Olga Sezneva, and Lana
14 Amin for their invaluable help in checking translations and names across French, German, Turkish, Arabic, and Russian. The first article I worked on in my PhD was my meta-analysis, and I am deeply thankful to all colleagues who provided invaluable feedback, particularly: Amanda Friesen, Amma Panin, Anne Louise Schotel, Armen Hakhverdian, Ebe Ouattara, Eline Severs, Joost van Spanje, Martin Rosema, Michael Jankowski, Orly Siow, Robin Devroe, Sabrina Mayer, Sönke Ehret, Tom Louwerse, and Wouter van der Brug. Each of these individuals played a crucial role in shaping this article in unique ways. I also wish to express particular appreciation to all individuals with whom I had email correspondence and who generously shared their databases and replication materials. Your contributions were invaluable to the completion of this research. There were also some people who really shaped my commitment to Open Science and supported me along the way. Amanda Friesen, Bert Bakker, Gijs Schumacher, Isabella Rebasso, Maaike Homan, and Matthijs Rooduijn were instrumental in fostering my dedication to Open Science principles. My PhD journey wasn’t all smooth sailing. Along the way, I faced challenges, and during those times of difficulty, several people played pivotal roles in helping me navigate through them. Anne Kervers, Anne Loeber, Anne Louise Schotel, Armen Hakhverdian, Brian Burgoon, Daphne van der Pas, Ebe Ouattara, Ellis Aizenberg, Enzo Rossi, Floris Vermeulen, Gijs Schumacher, Haylee Kelsal, Isabella Rebasso, Judith de Jong, Natalie Welfens, Nicky Pouw, Samera Ouchene, Sarah de Lange, Sander Kunst, Uğur Aytac, Valerie Schreur, Wouter van der Brug, Zahra Runderkamp—each of them provided invaluable support and guidance. I found some dear friends who made this experience truly enjoyable: Anne Kervers, Anne Louise Schotel, Ebe Ouattara, Haylee Kelsal, Isabella Rebasso, Judith de Jong, Kirsten Vegt, Larissa Böckmann, Maaike Homan, Natalie Welfens, Uğur Aytaç, Valerie Schreur, Wouter Schakel, and Zahra Runderkamp. Your companionship made this PhD fun and memorable. I especially thank Wouter, Anne Louise and Natalie for their exceptional feedback throughout. Ebe, for helping me to learn to code. Isabella, for shelter and friendship with the entire family. Zahra, thank you for the enjoyable outings and insider AISSR insights. Judith, our discussions on theoretical implications were incredibly fruitful. Wouter and Maaike, your inspiring ideas have sparked thoughts for future work. The suggestions and discussions we shared will stay with me for years to come. Since moving to Oxford, I’ve been fortunate to have the support of several remarkable people. Mariña Fernández-Reino stands out as the most supportive PI I’ve had the privilege to work with. Rob McNeil has promoted me like no other and provided invaluable media advice and insight. Madeleine Sumption consistently challenged my ideas, pushing me to become a more rigorous researcher. Matthew Porges encouraged me to expand my horizons and develop my ideas further. Will
15 Allen for giving me valuable tips to help me pursue the research I’ve always wanted to sink my teeth into. I am also thankful for the contributions of Carlos Vargas-Silva, Denis Kierans, Federica Genovese, José Ignacio Carrasco, Miriam Driessen, Peter Walsh, Ruta Nimkar, and Sarah Spellman. Each of them has enriched my experience here in Oxford in unique and meaningful ways. To all my current co-authors, from both the EqualStrength Consortium and beyond, thank you for showing me how enjoyable and beneficial collaboration can be. I am deeply grateful to Pelin Atay, Héctor Cebolla Boado, Daniel Capistrano, Mathew Creighton, Valentina Di Stasio, Mariña Fernández-Reino, Flavia Fossati, Abel Ghekiere, Matthew Hepplewhite, Johanna Hildebrandt, Ketevani Kapanadze, Ilkhom Khalimzoda, Zohreh Khoban, Jeremy Kuhnle, Bram Lancee, Vasilena Lachkovska, Isabel Lázaro, Anna-Lena Nadler, Ruta Nimkar, Wan Munira Wan Jaafar, Josef Montag, Gefjon Off, Shomaila Sadaf, Maïlys Samba, Ágota Scharle, Bori Simonovits, Stefanie Sprong, Stephanie Steinmetz, Ely Strömberg, Alvaro Suarez, Susanne Veit, Paolo Velasquez, Pieter-Paul Verhaeghe, Veronika Vass-Vigh, Lim Mengzhen, and Eva Zschirnt. Working with you has shown me how collaboration truly enhances research. Having children all throughout my PhD journey, I’ve been fortunate to receive support from many people. Both sets of grandparents filled in the gaps when my husband and I couldn’t. Their love, care, and willingness to help have been invaluable in balancing parenthood with academic pursuits. We are deeply grateful for their unwavering support, which has made a significant difference in our lives and allowed me to focus on completing my dissertation. Carla has been an exceptional childminder, providing outstanding care. Ykje, your unwavering dedication to our Urban Family to taking our daughters to swimming lessons made the experience so much more enjoyable for all of us, and their swimming diplomas are as much yours as theirs. Special thanks to Juf Lisa and Juf Misha for creating supportive environments for my daughter’s condition in their classrooms. Juf Fatima, Juf Saskia, Juf Maghtaria, and Juf Fatima, thank you for taking my daughter’s condition seriously and ensuring she could eat safely at school without worry. Lastly, Dr. Kraakman, for your enlightening talk at my daughter’s school about her condition, debunking myths and misconceptions. For my youngest child, thank you for always bringing excitement into my life. Your inventiveness and creativity never fail to surprise me. Your caring nature is incredibly touching, and I feel proud knowing that we’ve played a part in nurturing such a compassionate person. Embrace your quirky spirit as an artist; it’s exactly what I’ve always admired in you. Stay true to yourself and let your artistic passion flow freely through your veins. I cherish accompanying you through each of your fascinations and journeys. Your beautiful artwork has also enriched this dissertation, adding depth and beauty to my work. Keep embracing life as an artist, painting my world with joy and beauty. I appreciate how you continually challenge my perspectives and inspire me to see the world differently. Moving to a new country, adapting to a new school
16 with uniforms and rules, making new friends, and switching schools three times were challenging transitions for you. I witnessed your struggles, and I admire your resilience throughout it all. I hope these experiences, though difficult, will someday be remembered fondly as moments of growth and transformation. To my eldest child, thank you for being the sunshine of my life. Your infectious smile and happy-go-lucky attitude bring me immense joy every day. Your smile lights up my world, and your dance moves are contagious. Your ambition and determination are truly inspirational. Whether you choose to become a prime minister, astronaut, CEO, climate scientist, football player, or all of the above, you will undoubtedly make our world a better place. Your dedication to helping others is a source of great pride and inspiration. You always strive to do what’s best for others, and your caring nature shines through in everything you do. Moving to a new country was a big adjustment for you as well, and I am deeply grateful for your willingness to move countries and you have gone through it all with flair. Throughout my PhD, you have given me reason to pause and cherish the moments together. Without you, I might have kept working tirelessly without taking a moment to appreciate my biggest joy of life: you. Above all, there is one person to whom I owe everything: my husband. He was the first to believe in my ideas when no one else did. Every endeavor I undertake begins with him. From the ideas I pursue to the grants I apply for and the papers I write, he is my sounding board and confidant. I discuss everything with him first, and I reflect on my adventures with him when they are done. My husband supports me no matter what, scrutinizes my ideas with care, and encourages me to explore my research more thoroughly. He made tremendous sacrifices, especially by moving with me to a new country for my work, but also by supporting me through all of it. His commitment and dedication, driven by love, have been immense. While it wasn’t always easy for him, he made it happen. He is not only my best friend in life but also the best father to our children. I look forward to spending the rest of my life with him.
17 Publication overview Chapter 1: Published: Van Oosten S, Mügge L and Van der Pas D (2024) Race/Ethnicity in Candidate Experiments: a Meta-Analysis and the Case for Shared Identification. Acta Politica 58(1). Palgrave Macmillan UK. DOI: 10.1057/s41269-022-00279-y. Contributions by co-authors: - Van Oosten S: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, and Writing – review & editing - Mügge L: Supervision, funding acquisition, conceptualization, writing – review & editing - Van der Pas D: Supervision, conceptualization, data curation, validation, formal analysis, writing – review & editing Chapter 2: Under review at Journal of Politics, preprint available at: Van Oosten S (2023) Who favor in-group politicians? In-group voting in France, Germany and the Netherlands and the challenges to the descriptive and substantive representation of Muslims. OSF Preprints. DOI: https://doi. org/10.31219/osf.io/rkejd. Chapter 3: Published: Van Oosten S, Mügge L, Hakhverdian A and Van der Pas D (2024) What explains voting for DENK: Issues, discrimination or in-group favouritism? Representation, Journal of Representative Democracy. Taylor & Francis. DOI: https://doi.org/ 10.1080/00344893.2024.2387011. Contributions by co-authors: - Van Oosten S: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Project administration, Resources, Software, Validation, Visualization, Writing – original draft, and Writing – review & editing - Mügge L: Supervision, funding acquisition, conceptualization, writing – review & editing - Hakhverdian A: Supervision, conceptualization, investigation, formal analysis, writing – review & editing - Van der Pas D: Supervision, investigation, validation, formal analysis, writing – review & editing
18 Chapter 4: Published: Van Oosten S (2022) What shapes voter expectations of Muslim politicians’ views on homosexuality: stereotyping or projection? Electoral Studies 80 (December). Elsevier Ltd: 1–11. DOI: 10.1016/j.electstud.2022.102553.
19 Open science information Please find the appendix, dataset and replication materials in the following locations: Van Oosten S, Mügge L, Hakhverdian A, Van der Pas D and Vermeulen F (2024) French Ethnic Minority and Muslim Attitudes, Voting, Identity and Discrimination (EMMAVID) - EMMAVID Data France. Harvard Dataverse. DOI: https://doi. org/10.7910/DVN/ULQEAY Van Oosten S, Mügge L, Hakhverdian A, Van der Pas D and Vermeulen F (2024) German Ethnic Minority and Muslim Attitudes, Voting, Identity and Discrimination (EMMAVID) - EMMAVID Data Germany. Harvard Dataverse. DOI: https://doi. org/10.7910/DVN/GT4N9J Van Oosten S, Mügge L, Hakhverdian A, Van der Pas D and Vermeulen F (2024) Dutch Ethnic Minority and Muslim Attitudes, Voting, Identity and Discrimination (EMMAVID) - EMMAVID Data the Netherlands. Harvard Dataverse. DOI: https:// doi.org/10.7910/DVN/BGVJZQ All information in this dissertation can be verified through requests using the following information: 1. Ethics approval: if interested in confirming this information please put in a request at the Academic Institute for Social Science Research (AISSR) – Tel: +31(0)20-525 2262 - Email: aissr@uva.nl - ERB number 2018-AISSR-9546 – Review carried out by Alix Nieuwenhuis. 2. Pre-registration: the pre-registration can be viewed here: osf.io/jtdqw – Titel: An MP Who Looks Like Me - How do race, religion and gender influence preferences for political representatives amongst individual citizens in France, Germany and the Netherlands? 3. Sampling strategy: the research accountability can be obtained through contacting Kantar Public – Tel: +31(0)20-5225 444 - Email: info@ kantarpublic.com - with reference number: H6764 – contact person is Ingmar Doeven.
Introduction Unraveling voting preferences for in-group politicians in the midst of minority broadstancing
22 Introduction Many believe that minority voters tend to vote for in-group politicians, who, in turn, will represent their interests (Dancygier, 2017; Stephens-Dougan, 2020). Even when reality is much more complicated, this in-group voting assumption fundamentally shapes the dynamics underlying narratives on diversity in political representation (idem): this assumption perpetuates a view that paints minority politicians as being “beholden” to their group’s interests at the cost of the majority, stemming from a zerosum worldview often prevalent among members of the majority group (StephensDougan, 2020). This assumes minority voters simply vote for their in-group. I argue this portrays minorities as lacking rationality in forming their own democratic choices, thereby not taking their thoughtful participation in the democratic process seriously. Consequently, minority politicians, assumed to have gained office through the support of in-group voters, have their electoral accomplishments attributed to merely their group membership, overlooking their views and efforts. Reality is, however, much more complicated than the in-group voting assumption might suggest. In many political contexts, minority politicians are incentivized to broaden their electoral appeal by distancing themselves from their in-group, a phenomenon I call minority broadstancing (van Oosten, 2024). The most famous example of a minority broadstancer is probably Barack Obama (Stephens-Dougan, 2020). For instance, he famously called for Black people to “get off the couch” (2020: 67), distancing himself from Black stereotypes of laziness, and blaming racism for their problems (idem). These stereotypical remarks are electorally beneficial: white moderates might be persuaded Barack Obama is not a threat to the racial status quo, thus broadening his electoral appeal (idem). Broadstancing can range from the distancing from stereotypes in the case of Obama, to “suppressive representation:” when minority politicians adopt suppressive positions towards the cultural and/or religious rights and freedoms of their in-group (Aydemir and Vliegenthart, 2016). Nonetheless, the in-group voting assumption remains ubiquitous (StephensDougan, 2020). The consequences of the perpetuation of the in-group voting assumption undermines minority politicians’ legitimacy and disrupts the functioning of our pluralistic democracy by hindering equal representation. Meanwhile, it perpetuates and originates from a power hierarchy that places the majority at the center and the minorities at the periphery. Understanding the dynamic relationship between minority voters and their in-group politicians is crucial, because it exerts a firm grip on the representation of minorities. The examination of this relationship from the perspective of the minority politician has been well executed (Aydemir and Vliegenthart, 2016; Dancygier, 2017; Stephens-Dougan, 2020), and now I am completing the puzzle from the viewpoint of the minority voter. Hence, the central research question is: Who prefers politicians with whom they share the same religion, migration background
23 and/or gender and why? In this dissertation, I studied in-group voting by citizens of France, Germany and the Netherlands with a first- or second generation migration background in Turkey (all countries), North-Africa/Morocco (France, the Netherlands), Sub-Saharan Africa (France), the Former Soviet Union (Germany), Surinam (the Netherlands) as well as those without a migration background, and/or those who self-identify as Muslim, Christian or non-religious. I emphasize voting for individual politicians with a secondary focus on voting for political parties. When taking political parties out of the equation, the existing literature reveals a consensus that minority voting citizens tend to prefer voting for candidates from their own racial or ethnic background and gender (Carlson, 2015; Dolan, 2008; Kao and Benstead, 2021; Kaslovsky et al., 2021; Lemi and Brown, 2019; Sanbonmatsu, 2002). This field, however, lacks research on Muslims in Europe, where debates on minority politics center on Muslim politicians and Islam (Vermeulen, 2018). My incorporation of Muslim politicians and examination of religious self-identification within the study of in-group voting stands out as a distinctive approach in the field. The influence and significance of Muslim voters and politicians emerged as pivotal factors shaping the dynamics of in-group voting for minority and majority groups. Amid the broader exploration of minority in-group voting, this dissertation centers on analyzing the in-group voting behavior specifically of self-identifying European Muslims. While Muslim European citizens face underrepresentation in European national parliaments (Aktürk and Katliarou, 2021), their societal position continues to be a contentious topic in political discourse (Vermeulen, 2018) and they experience distinct anti-Muslim racism in broader society (Fernández-Reino et al., 2023). Meanwhile, Muslims express low feelings of representation, political trust and satisfaction with democracy (Grewal and Hamid, 2022; Maxwell, 2010) and boosting the presence of Muslims in politics increases trust in the democratic process, encourages active participation (Spierings and Vermeulen, 2023), increases voter turnout among minority groups (Poertner, 2022) and ultimately contributes to a more robust and better functioning democracy (Mishler and Rose, 2001; Akachar et al., 2017). Understanding when and why in-group voting can contribute to the political inclusion of Muslims is vital for comprehending the democratic process in our increasingly diverse societies. While existing European research indicates that Muslims often vote for Muslim politicians (Azabar et al., 2020), distinguishing whether this is driven by shared religious identity or aligned policy positions presents a challenge. In some contexts, sharing the same religion and sharing the same policy preference is likely to align. In other contexts, this is not the case at all. Muslim politicians can be incentivized to distance themselves from prevailing stereotypes about Muslims, possibly impeding in-group voting amongst Muslims (Dancygier, 2017; StephensDougan, 2020). An exploration of in-group voting needs to disentangle shared identity and shared policy in order to understand which factor is driving voting amongst
24 minority citizens of France, Germany and the Netherlands. In this dissertation, I have discovered that the connection between Muslim politicians and Muslim voters is strong, showing a positive relationship in terms of voting patterns within their group, particularly in contexts with examples of Muslim politicians who stand for the same policy positions as most Muslim citizens. Conversely, the relationship between non-religious voters and Muslim politicians is comparable in strength but notably negative, possibly driven by stereotypes of the policy positions that Muslim politicians might stand for. Within the post-9/11 European context, Muslims have become a highly salient group. This research sheds light on the unique role of Muslims in current European politics, while also revealing broader insights into the dynamics experienced by underrepresented and marginalized groups in the political arena. Understanding these dynamics is key to establishing equitable minority representation and fostering diversity in our increasingly multicultural societies. What is representation and why do we need it? Understanding the dynamics and implications of in-group voting offers insights into which mechanisms can empower underrepresented and marginalized groups, enabling their voices to be heard and their concerns addressed within the political sphere. Ingroup voting is, therefore, essential within the larger framework of understanding political representation. Several scholars have advocated for fair political representation, highlighting its moral and epistemic importance and underscoring its critical role in the functioning of our democracies (Mansbridge, 1999, 2003; Williams, 2000; Young, 2000). When minority citizens receive adequate representation, it enhances the legitimacy of the democratic system, affirming the principle that every voice and vote holds equal importance (idem). This inclusive approach to political involvement not only upholds justice and strengthens democracy but also fulfills the moral mandate of fostering a political landscape where every citizen’s voice is acknowledged and valued. This dissertation aims to examine the role in-group voting might play in the political representation of minority groups. To understand representation, I use the framework laid out in Hanna Pitkin’s canonical work, The Concept of Representation (1967), which conceptualizes representation using four dimensions. First, formalistic representation consists of the “authorization” and “accountability” that representation begins and usually ends with (Pitkin, 1967: 51, 57). Second, descriptive representation concerns the degree to which representatives are “sufficiently like” those they represent (1967: 81). Third, substantive representation concerns the “activity to speak for, act for, [and] look after the interests of their respective groups” (1967: 117). Fourth, symbolic representation refers to the “state of mind, the condition of satisfaction or belief,” and includes the question of whether “the representative [is] believed in” (1967: 106, 102). Pitkin describes symbolic representation as “the symbol’s power to evoke feelings or attitudes” (1967: 97). I understand symbolic
25 representation, as defined by Hanna Pitkin, as the dimension relating most to voting. Symbolic representation embodies the beliefs and attitudes inherent in the process leading up to voting, making it the dimension most akin to the voting process. Following a period of minimal attention, scholarly interest in representation reemerged in the 1990s. Notably, this resurgence emphasized a normative standpoint advocating for the descriptive representation of women and other historically marginalized groups. Phillips (1995) argues for the increased presence of women in politics, stating that it would enhance the visibility of “particular interests of women that would be otherwise overlooked” (Phillips, 1995: 62–63). Similarly, Williams (2000) argues that marginalized groups, particularly those who are “immutable” (Williams, 2000: 55), need to be represented in politics to ensure that their “voices are heard” (2000: 62, 78, 131, 138), particularly given the “legacy of state-sponsored discrimination” and continued “systemic biases” (2000, 199). Mansbridge (1999) suggests that descriptive representation can help establish the “legitimacy” (1999: 648) and prove the “ability to rule” (1999: 628) of marginalized groups. Young (2000) also emphasizes the importance of descriptive representation because it reduces bias against “embodied forms of expression” (2000: 56) and increases the visibility of “situated knowledges” (2000: 70). Overall, the normative arguments for descriptive representation range from increasing the visibility of marginalized interests, having their voices be heard through epistemic justice to counter historic (epistemic) injustices, increasing legitimacy and proving they are able to rule. In the 2000s, the focus shifted towards a constructivist approach, revisiting the conceptualization of representation rather than its normative aspects (Disch, 2015). Scholars started emphasizing the dynamic relationship between voters and politicians, particularly exploring identity dynamics (Saward, 2010: 16, 298). This emphasis on identity becomes crucial in understanding why citizens might prefer in-group politicians. In this dissertation, I employ the conceptual framework established by Hanna Pitkin, delving into the dimensions of descriptive, substantive, and symbolic representation. I aim to critically examine and assess the normative propositions put forth by scholars like Mansbridge, Young, and Williams, by shifting the focus towards the perspective of the individual citizen, much akin to the approach taken by Saward. To understand the representation of identities, I draw on the literature of Social Identity Theory (Tajfel, 1974; Tajfel and Turner, 1979; Turner, 1975; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Turner and Oakes, 1986). They state that all humans strive for a positive self-image, i.e. positive distinctiveness. Social Identity Theory proposes that individuals use different strategies to achieve positive distinctiveness, including individual mobility, social creativity, and social competition, depending on various factors such as group status. For low-status groups, the choice of strategy depends on factors such as the permeability of group boundaries and the perceived homogeneity/ heterogeneity of the out-group. If low-status individuals perceive boundaries as
26 permeable, they will strive for individual mobility to join the high-status group, leading to out-group favoritism. If group relations are seen as legitimate and stable, individuals will try to achieve positive distinctiveness through social creativity by redefining what it means to be part of their group, an indirect form of in-group favoritism. If group boundaries are perceived as impermeable and status differences as illegitimate and/or unstable, low-status groups are more likely to choose social competition, leading to direct and open in-group favoritism. For high-status groups, the same three strategies exist, but they always lead to in-group favoritism. If group boundaries are perceived as permeable, high-status groups expect low-status groups to exert individual mobility and join the high-status group. If not, high-status groups may argue that low-status groups are guilty of causing their own inferiority. If group boundaries are perceived as impermeable, legitimate, and stable, high-status group members may exhibit highminded benevolence while engaging in latent discrimination and covert repression. If a high-status group perceives group relations as unstable and threatening, they may resort to supremacist ideologizing, conflict, open hostility, and antagonism by directly promoting the out-group’s inferiority. In summary, low-status groups may prefer their in-group or out-group under certain conditions, while high-status groups tend to prefer their in-group under all conditions, leading to universal in-group favoritism (for an overview, see Haslam, 2001: 21–26). Another explanation for in-group voter preferences might lie in Heuristics Theory, which ascribes more logical reasoning to the process shaping voting preferences. This theory states that voting citizens use heuristics - shortcuts or cues, often based on politician identity - to infer policy positions of politicians, particularly from their in-group, by simplifying voting citizens’ decision-making process (Cutler, 2002) and offering solutions when voting citizens are faced with overly difficult decisions (Koch, 2000). McDermott (1998) studied this phenomenon using candidate experiments. She moved away from stereotypes being a ‘bias against’ race or gender and moved the field towards understanding heuristics as a more ‘content-based’ view of how race and gender cues serve as ‘shortcuts [in] low-information elections’ (McDermott, 1998: 896). Indeed, she says one should depart from the common practice of focusing on ‘potential bias or racism’ towards ‘neutral stereotyping’ influencing the voting behavior of both white and black voters (1998: 901). Voting citizens use candidate race and gender to inform voting behavior based on voting citizens’ political beliefs. Indeed, some people use gender as a heuristic when making difficult voting decisions (Holman et al., 2016; Koch, 2016; Lau and Redlawsk, 2001). By intertwining representation theory, social identity theory, and heuristics theory, this dissertation puts forward a holistic understanding of voter behavior of minority voters and perceptions of minority politicians. The combination of these theoretical perspectives reaches beyond traditional scholarly disciplines by investigating the relationship between representation, identity, and heuristics.
27 Empirical evidence on migration background, religion and gender In this dissertation I refer to the term minority when I discuss people who belong to groups that are minoritized by being both a numeric minority and/or part of a discriminated group1 encompassing French, German and Dutch citizens with a migration background, Muslims and women. The studies on Muslims in politics are very limited, as most empirical studies on representation focus mostly on women but also ethnic minorities, with a particular interest in exploring whether (and how) descriptive representation leads to substantive representation (Lowande et al., 2019; Saalfeld and Bischof, 2013; Sobolewska et al., 2018). Besides overlooking the plight of Muslims in politics, this also leaves the question of whether descriptive and/or substantive representation leads to symbolic representation largely unaddressed (Schwindt-Bayer and Mishler, 2005; Verge and Pastor, 2018) and the existing scant literature focusses almost exclusively on symbolic representation of women, not ethnic/racial and religious minorities. To answer who prefers politicians with whom they share the same religion, migration background and/or gender and why we need to build on studies on in-group voting spanning religion and migration background, as well as gender. In the following section, I first discuss how voters and politicians sharing the same migration background influences voting, then religion, and lastly, gender. Migration background In this dissertation, I use the term migration background to denote individuals who were born in a different country, or who have one or both parents born in a different country. In France, Germany and the Netherlands, this term is commonly used to delineate ethnoracial differences (Elrick and Schwartzman, 2015), while terms such as race, ethnorace or ethnicity are more often used in the context of the United States (US). In the US, there is much experimental evidence for in-group voting (Burge et al., 2020; Kaslovsky et al., 2021; Lerman and Sadin, 2016; Philpot and Walton, 2007). This applies to both Black US Americans preferring Black politicians over their white counterparts (Stout and Le, 2017; Tate, 2003) and Asian and Latinx Americans alike (Schildkraut, 2013). A majority of highly engaged black voters stated ‘race’ as the most important reason to vote for Barack Obama in the 2008 primaries (Sullivan and Johnson, 2008: 59) and in-group voting increased after a number of public figures made ‘racially insensitive comments’, pointing to the importance of racial solidarity (2008: 60) as a driver of voting choice in addition to using race as a heuristic for the candidates’ policy positions (2008: 61). Moreover, descriptive representation leads to increased perceptions of legitimacy of the government (Ostfeld and Mutz, 2021) as well as turnout rates of ‘co-ethnics’ (Geese, 2020; Miller and Chaturvedi, 2018; 1 FRA: European Union Agency for Fundamental Rights (2017) Second European Union Minorities and Discrimination Survey (EU-MIDIS II): Main Results. DOI: 10.2811/902610.
28 Poertner, 2022; Whitby, 2007). Outside the US there are many studies suggesting voters prefer co-ethnic politicians (Aguilar, Cunow, Desposato, et al., 2015; Bermeo and Bhatia, 2017; Besco, 2015; Bird et al., 2016; Carlson, 2015; Chauchard, 2016; Goodyear-Grant and Tolley, 2019; Teney et al., 2010). In Europe, few experiments on ethnic/racial minority citizen assessments of politicians have been conducted. The only one I know about shows that the respondents have a significant negative bias against politicians with Arab names (Dahl and Nyrup, 2021), though non-experimental research points towards a co-ethnicity effect amongst voters in the UK (Fisher et al., 2014). Why might sharing the same migration background increase in-group voter preferences? First, voting citizens could be motivated through a preference towards descriptive representation in and of itself, without that leading to substantive representation (Fisher et al., 2014: 889; Ostfeld and Mutz, 2021). Some voting citizens have been found to choose descriptive representation over substantive representation, even when the two are at odds (Herron and Sekhon, 2005: 173). Descriptive representation has been shown to lead to greater societal respect (Ruedin, 2009) and showing the public your group has “ability to rule” (Mansbridge, 1999: 628) or as a “survival strategy” when faced with a lack of societal acceptance (hooks 1995 as cited in Lemi & Brown, 2019: 272). Ethnic minority voting citizens might be targeted by political parties (Goerres et al., 2020), or mobilized through civic (Fennema and Tillie, 1999), online (Kizgin et al., 2019) or personal (Geese, 2020) networks. The assumption of ethnically-based patronage networks may motivate voters to vote for their ethnic in-group members and the parties they are running for (Bermeo and Bhatia, 2017: 15) irrespective of whether patronage ensues or not. Even in a context where ethnic minority voter mobilization is discouraged, researchers find ethnic affinity voter preferences are prevalent (Teney et al., 2010: 275, 279) and ethnic minority voters “vote as a bloc” (Bergh and Bjørklund, 2011: 313, 323). Second, apart from the preference for in-group politicians, sharing the same ethnicity might be viewed as a heuristic for sharing the same policy position. There is much less literature on this second mechanism (except Arnesen et al., 2019; Lerman and Sadin, 2016) and this dissertation aims to change that. Religion I understand religion as a group that an individual can identify with and which is associated with a set of beliefs, practices, and rituals and thus encompasses both organized faith systems and personal spiritual beliefs. There is much less literature on co-religious voting even though Islam has become “politicized” (Fleischmann et al., 2011), “mobiliz[ed]” electorally (Schmuck and Matthes, 2019: 739) and “racialized” (Bracke and Hernández Aguilar, 2020; Elver, 2012; Meer and Modood, 2012) already since before 9/11 (Swyngedouw and Ivaldi, 2001: 15). Even those with positive
29 attitudes towards immigrants are far more critical towards Muslims, pointing towards the importance in how voting citizens reject religious fundamentalism in the name of political liberalism (Helbling and Traunmüller, 2018). With these trends towards politicization, collective identity and identity-threat become quantifiably stronger, as decades of evidence suggests (Frey, 2022; Lajevardi and Oskooii, 2018; Oskooii, 2016; Simon and Klandermans, 2001: 327; Voas and Fleischmann, 2012). In turn, religion is an important aspect of a Muslim’s identity, with Muslims scoring extremely high on numerous religiosity indicators (Fleischmann and Phalet, 2012: 329). However, if I only include politician ethnicity/race, one does not know whether people are reading their preferred religion into ethnicity/race or whether they prefer shared-ethnicity as it is (Di Stasio et al., 2019). That is why I disentangle ethnicity/race and religion in order to understand exactly which one is the driver of results. The only European experiment measuring co-religion between politician and citizen finds clear co-religiosity effects (Arnesen et al., 2019: 54). However, the experiment does not make a distinction between Christian and Muslim respondents. Given the underrepresentation of Muslim respondents in most surveys I expect that this co-religiosity effect is actually either a co-Christianity effect or an anti-Muslim effect, or both. Furthermore, Muslim citizens of Antwerp, Belgium show a strong preference for Muslim candidates (Azabar et al., 2020) and researchers find a co-Muslim effect in India as well (Heath et al., 2015). What explains Muslim affinity voting? Many seemingly logical reasons have been discredited in the literature. First, it is unlikely to be caused by religious practice, because Muslims who practice their faith more actively do not necessarily vote for Muslim candidates more often (Azabar et al., 2020: 8). Second, I expect the use of heuristics by Muslim citizens to be limited. Whereas majorities might use heuristics to assess politicians, Muslim minorities are more likely to resist stereotypes (van Es, 2019), be aware of the diversity within the group of Muslims (Baysu and Phalet, 2017; Baysu and Swyngedouw, 2020; Ellethy, 2016; Fleischmann and Phalet, 2012; Phalet et al., 2010; Verkuyten and Yildiz, 2009) and, therefore, be less inclined to resort to heuristics compared to out-groups who might be more likely to see Muslims as a monolithic bloc. Third, experiences of descriptive representation that did not lead to substantive representation, might lead to feelings of betrayal (Akachar, 2018; Anderson, 1997), weakening the likelihood of heuristic use. Indeed, Muslims might experience descriptive representation as being suppressive as they see politicians with a Muslim background explicitly speak out against Muslims (Aydemir and Vliegenthart, 2016). Given that so many explanations of Muslim affinity voting have been debunked, there is still a need to understand why Muslims do prefer voting for their in-group. One explanation for Muslim affinity voting that still stands lies in feelings op exclusion amongst Muslim voters, leading to voting for fellow Muslim politicians (Azabar et al., 2020: 8). My dissertation therefore focusses on the relationship between voting for
www.gildeprint.nlRkJQdWJsaXNoZXIy MjY0ODMw