Co-auteur
  • OLIVIER Mathieu (4)
  • DUWEZ Emmanuelle (3)
  • DANCIU Alina (2)
  • NAM Malick (2)
  • Voir plus
Type de Document
  • Communication non publiée (8)
66
vues

0
téléchargements
Les recherches quantitatives sont souvent réalisées au prix d’importants investissements en termes de temps et de ressources. Pourtant, la qualité des données ainsi recueillies est parfois insuffisante et rend difficile les tests d’hypothèses ainsi que la formulation de conclusions. Une raison principale en est que les questions ne sont pas toujours comprises par les répondants conformément aux attentes des concepteurs de questionnaires. Afin de prévenir ce problème, il convient de pré-tester les instruments de collecte. Lors de cette séance, nous passerons rapidement en revue différentes méthodes de pré-tests dont l’efficacité et la facilité de mise en place sont très variables. Nous nous focaliserons en particulier sur la méthode des entretiens cognitifs qui présente l’avantage d’explorer le ressenti du répondant à une question posée et aux modalités de réponses proposées. Des exemples concrets de travail réalisé au Centre des Données Socio-Politiques (CDSP), dans le cadre des projets Enquête Longitudinale par Internet pour les Sciences Sociales (ELIPSS) et Enquête Sociale Européenne (ESS), seront présentés. L’exemple d’ESS permettra d’évoquer, en outre, les biais liés à la traduction de questions et les enjeux d’harmonisation linguistique dans le cadre d’enquêtes multiculturelles. Une discussion sera animée afin d’inciter les participants à appliquer des techniques de tests et de traduction de questions.

Publié en 2019-07 Nom de la conférence ESRA - European Survey Research Association
ITALIANO Patrick
POLLIEN Alexandre
16
vues

0
téléchargements
The European Social Survey (ESS) is an academically driven cross-national survey conducted every second year across Europe, in which Belgium, France and Switzerland all participate since the first round in 2002. Since ESS measures attitudes, beliefs and behaviours of populations in over thirty countries, comparative analysis between the different national contexts and over time is crucial and methodologically demanding. In order to ensure the validity of the comparison, a key methodological condition relates to translations. ESS uses pragmatically equivalent questionnaires, that is translations aimed at maintaining the intended measurement properties rather than literal translations. The source questionnaire is designed in English and translated following the TRAPD methodology (Translation, Review, Adjudication, Pretesting and Documentation). After each country has reviewed two own, independent translations, a harmonization meeting takes place between countries sharing the same language. For the French version of the questionnaire, Belgium, France and Switzerland are involved. The aim of this meeting is to agree on a common French questionnaire as similar as possible and understandable in the same way across the three countries. Through a deep and detailed review of the work during these harmonisation sessions, this presentation proposes to take a critical look at the issues in need of decision-making during the process toward a shared French version. Over the years, we have been confronted with multiple and varied situations, which have implied concerted decision-making regarding the harmonization of translation. Through a typology of the discrepancies encountered, and the associated decision-making procedures, choices toward an expected convergence or differences to be preserved, we intend to propose a descriptive analysis of our practice aimed at identifying sources of difficulties in the translation process, in order to help the upstream work of disambiguating the source questionnaire and to guide the translation instructions.

Publié en 2019-07 Nom de la conférence ESRA - European Survey Reasearch Association
22
vues

0
téléchargements
The European Social Survey (ESS) is an academically driven cross-national survey conducted every two years across Europe, to which France has participated since the first round in 2002. The ESS measures the attitudes, beliefs and behaviour patterns of various populations in more than thirty countries. One of its main goals aims to provide analysis elements on how Europe’s social, political and moral fabric is changing. In the ESS, data are collected via face-to-face interviews. In France, the fieldwork of the 7th round led to a survey replication on the panel ELIPSS (Longitudinal Internet Studies for Social Sciences). Every month the panel members answer a 30 minutes self-administered questionnaire developed by researchers and selected by a scientific and technical committee. In order to answer the surveys, they are provided with a touch-screen tablet and a 3G Internet subscription. In the ELIPSS panel, members are randomly selected by the French census bureau (INSEE). Traditionally, this is also the same for the ESS, but it was different for the 7th round. This time, the fieldwork institute in charge of data collection has proposed its own sampling methodology. One can find significant differences in methodologies between these two surveys, from the sampling procedure to the data collection collect method and analysis tools. In 2017, we focused on the results comparison between the two devices; this year, we intend to focus on the different ways to set up a survey. In particular, we want to aim at the importance of the gender within the interviewer-respondent relationship. The replication of a face-to-face survey into an auto-administered panel makes it possible to study the effects resulting from being surveyed by a person of the same or the opposite gender. Consequently, it asks the question : to what extent auto-administered surveys could provide an access to objective measurement ?

Publié en 2017-11 Nom de la conférence 2017-11-14, 4e journées annuelles du réseau MATE-SHS, Rouen
41
vues

0
téléchargements
Depuis 2005, le CDSP documente et met à disposition de la communauté scientifique des enquêtes et données quantitatives. Les métadonnées liées à une étude, à ses résultats et aux données de l'étude elles-mêmes sont renseignées avec le logiciel Nesstar. Cette communication traite de la documentation de séries d’enquêtes quantitatives répétées dans le temps, et, plus précisément, des métadonnées au niveau de la description de l’étude. Par leur caractère longitudinal, ces séries posent des questions spécifiques de documentation. Nous proposons de comparer les pratiques de documentation des enquêtes produites par le CDSP, pour lesquelles la documentation commence dès leur conception (Enquêtes annuelles ELIPSS, Pratiques numériques...) et des dispositifs d’enquêtes dans lesquels le caractère récurrent n’est intégré qu’après la fin de leur terrain (Agoramétrie, Image de la Science...). L’objectif de cette comparaison est double : d’une part, identifier les éléments qui semblent intéressants à conserver dès la conception d’une enquête en vue de sa documentation ; d’autre part et plus généralement, interroger les enjeux posés par la prise en compte du cycle de vie dans la documentation.

Publié en 2017-09 Nom de la conférence 7th Conference of the European Survey Research Association
26
vues

0
téléchargements
The European Social Survey (ESS) is an academically driven cross-national survey conducted every two years across Europe, in which France has participated since the first round in 2002. The ESS measures the attitudes, beliefs and behaviour patterns of various populations in more than thirty countries. One of its main goals aims to track stability and changes in the social structure of European societies and to provide analysis elements on how Europe’s social, political and moral fabric is changing. In the ESS, data are collected via face-to-face interviews. In France, the fieldwork of the 7th round led to a survey replication on the pilot of the Elipss panel (Longitudinal Internet Studies for Social Sciences). Elipss is a probability based online panel that is representative of the French population aged 18-75. Panel members are randomly selected by The French National Institute of Statistics and Economic Studies (INSEE) and equipped with a touch-screen tablet and a 3G Internet subscription. Every month they are asked to answer a 30 minutes self-administered questionnaire proposed by researchers and selected by a scientific and technical committee. The Center of Socio-Political Data of Sciences Po coordinates the ESS fieldwork for France and conducts the Elipss panel. Consecutively we developed an expertise on the methodology and the process of each mode. The face-to-face fieldwork was carried out from November 2014 to February 2015. To replicate it on the Elipss panel, we used a slot between December 2014 and January 2015. The administration of the first part of the ESS core questionnaire on the Elipss panel gave an opportunity to wonder how the different strategies for collecting data may impact the response behaviour. Knowing that differences are already emerging in the specificities of these two protocols, the questionnaire needed some adjustments that we must consider in such a comparison. Indeed, for its replication in a self-administered mode on a mobile device we had to adapt the design of some questions, and it could have impacted the answer situation. The difference in the structure of the sample should be taken into account to explain the observed differences. The length of the questionnaire, the format of the answer categories, the presence or absence of an interviewer, the survey experience of Elipss panel members could also account for differences in response behaviour. Focusing on type and design of questions, we will pay special attention to the social desirability effect often pointed out in face-to-face surveys. This paper will highlight the specificities of the two survey designs (face-to-face vs self-administered online questionnaire) in order to discuss the scope of such a comparison. Finally, we will compare the answers according to whether the data is collected by interviewers or self-administered on mobile device.

Publié en 2016-12 Nom de la conférence colloque des 10 ans du CDSP : « Des données et des services pour les sciences sociales »
37
vues

0
téléchargements
Depuis 2002, la France participe à l’European Social Survey (ESS), destinée à mesurer les comportements et les attitudes des citoyens de plusieurs pays européens sur un ensemble de thèmes socio-politiques. La réalisation du terrain français de la 7e édition de l’ESS entre novembre 2013 et mars 2014 a donné lieu à une administration parallèle d’une partie du questionnaire au panel ELIPSS. C’est l’occasion de s’interroger sur les effets des modes d’administration dans la réponse à ces deux enquêtes. L’objectif de cette communication est de mettre en lumière les spécificités de l’étude selon qu’elle ait eu lieu en face-à-face ou via l’Internet mobile mais également d’apporter des éléments de comparaison sur les réponses recueillies par chaque dispositif.

Publié en 2016-11 Nom de la conférence L’enquête sociale européenne (ESS) : recherche, analyse, méthodologies
35
vues

0
téléchargements
Il s’agira dans cet atelier, à partir d’un module thématique répété lors des éditions 3 et 6 de l’ESS de travailler sur la mesure du bien-être. En interrogeant, pour point de départ, l’utilité de mesurer le bien-être, la présentation aura pour objectif de construire des indicateurs de bien-être permettant une comparaison entre les pays et dans le temps. à travers les étapes de recodage des données, d’harmonisation des variables ainsi que de pondérations des résultats, l’ensemble du processus de traitement des données ESS sera présenté aux participants.

41
vues

0
téléchargements
The French Center of Socio-Political Data has presented its reflection on the process of shifting from DDI-C to DDI-L at EDDI14. This year, we will discuss the creation and storage of a DDI-L compliant XML record by capturing metadata of a nine-wave political study of the ELIPSS panel. Determining how best to recognise continuities between metadata collections within the same study, including question continuity and methodological continuities has been a primary challenge. To answer it, the starting point was the creation of a questions database. As seen at the 2014 DDI workshop in Dagstuhl, the minimum requirements that a metadata system should meet before being able to import/export DDI-L are uniqueness of items, versioning and granularity. To conceive such a database, we had to start by using simple tools. We first identified metadata in CSV files that include variable-level information. We then performed a semi-manual import from these files to the database using importing scripts. Once we removed automatically the redundancy, with a further stage of human control, we generated the structure of the DDI-L compliant XML file. Our paper will present this process and discuss its replication to other DDI-C documented studies.