BMS, English Abstracts 2003-2013 (80-119)
N. 80, October 2003
LES ENQUETES DE VICTIMATION ET LA CONNAISSANCE DE LA DELINQUANCE
Philippe Robert, Marie-Lys Pottier, Renée Zauberman
Victimisation Surveys and Knowledge of Delinquency: This article is based on an 11 April 2002 presentation at the Survey Methods Seminar of INED and the French Society of Statistics at INED (salle Alfred Sauvy, 133, boulevard Davout, 75020 Paris). After presenting the historical background of victimisation surveys in the framework of quantitative data used in the analysis of crime, this article presents the contributions and limits of such surveys. It provides the basic structure for such questionnaires and various sample frameworks. In closing, it gives results based on a survey in the Paris region in 1998-2000, and compares them with related police statistics. Measuring Delinquency, Criminal Statistics, Victimisation Surveys.
THE ALLOW-FORBID ASYMMETRY IN QUESTION WORDING – A NEW LOOK AT AN OLD PROBLEM1
The famous “allow”-“forbid” example of question-wording effect, originally published in 1941 by Daniel Rugg, is tested in a local survey in Germany, based on a quota sample and on face-to-face interviews (N=620). Not only the original question wording enters into the split-ballot experiment, but also an extended-version questionnaire where the response alternatives are explicitly stated (“allow or not allow”, “allow or forbid”). The original effect described in the literature turns out be primarily the result of formulating the question in a one-sided manner without explicit alternatives. The specific wording itself has a minor effect. Respondents with lower education are primary affected by the one-sided formulation; respondents with higher education seem to be more frequently influenced by the wording itself. Allow, Forbid, Questionnaire Construction, Question Wording.
THE NEW DUTCH ICS 2003-2008 RESEARCH PROGRAM
Interuniversity Center for Social Science Theory and Methodology (ICS)
The Interuniversity Center for Social Science Theory and Methodology (ICS), which is a graduate school and research center, was founded in 1989 and includes the University of Groningen, Utrecht University and the University of Nijmegen. Last year, along with re-accreditation by the Royal Dutch Academy of Arts and Sciences (KNAW), a number of important changes took place at the ICS which are developed in detail in its recently-published Annual Report from which this article was developed. The ICS research program now includes a Social Networks, Solidarity and Inequality branch which is presented here along with the four other branches and the full list of ICS Ph.D students and their thesis topics. Netherlands, Social Science Research, Sociological Methodology.
N. 81, January 2004
EFFECTS OF SURVEY EXPERIENCE ON RESPONDENTS’ ATTITUDES TOWARDS SURVEYS
Volker Stocké, Bettina Langfeldt
In the following article we analyze whether or not and to what extent respondents’ evaluations of past interview experiences affect their generalized attitudes towards surveys. In particular, our study compares the relative significance of three orthogonal evaluation dimensions: the burden caused by interview participation, the “entertainment value” of an interview and the irritation due to confusing question wording. As a first result, the mode of administration of the last interview proved to be a significant predictor of the evaluation of surveys in general. The second and most important finding is that the burden experienced during the last interview, but not the other evaluation dimensions, has a significant effect on the generalized evaluation of surveys. In addition, the association between the subjective burden and survey attitudes is found to be conditioned by the cognitive accessibility of these evaluations, as measured by response latencies. Thus, cognitively more accessible and more salient instances of burdensome interview experiences are especially relevant for the respondents’ attitudes, and therefore influence most likely cooperation in future surveys. Attitude Accessibility, Attitudes Towards Surveys, Respondents’ Burden, Respondents’ Cooperation, Response Latency, Survey Experience, Survey Sponsor.
RENOUVELLEMENT DU TERRAIN PAR LA PHOTOGRAPHIE: LA COOPERATION D’UNE ETHNOLOGUE ET D’UN PHOTOGRAPHE
Béatrice Maurines, Angel Sanhueza
Renewing Field Work through Photography, Cooperation between an Ethnologist and a Photographer: This article is dedicated to the discovery of sensitive worlds rich in import for the analysis of social representations. At its origin is a joint research project of an ethnologist and a photographer in industrial anthropology. The article describes the development of the research project where pictures were used not only as a means of illustration but with the intention of obtaining a better understanding of this field work. Indeed, certain sectors are barely explorable with “classic” qualitative and quantitative means (interviews, observations, questionnaires). In our perspective, the enthnologist’s work should facilitate the passage “from the visible to the readable”, but it should be able to address the invisible — invisible for the researcher and often even for the informant — and the non-expressible without mediation. Photography, Methodological Use of Photography, Industrial Anthropology, Ethnologist-Photographer Cooperation.
CORRESPONDENCE & CO-WORD ANALYSIS OF TEN YEARS OF BMS ARTICLES (1993-2003)
Karl M. van Meter, Philippe Cibois, Mathilde de Saint Léger
Using Trideux factorial correspondence analysis and Calliope co-occurrence of key word analysis, we apply them to the data base of key words characterizing each research article or ongoing research report published by the BMS from December 1993 to October 2003. We present the results of these analyses, followed by the complete list of tables of contents, the author index and the article-title index for the articles and reports analyzed. BMS, Correspondence Analysis, Trideux, Co-Word Analysis, Calliope, Scientometry.
ACADEMY COLLOQUIUM ON “CREATION AND RETURNS OF SOCIAL CAPITAL. THE STATE OF THE ART
Henk Flap, Beate Völker
On 30-31 October 2003, an international colloquium took place on “The Creation and Returns of Social Capital. The State of the Art”, organised by the Royal Netherlands’ Academy of the Arts and Sciences at the Trippenhuis (Kloveniersburgwal 29) in Amsterdam, The Netherlands. About fifty people took part in the conference. Below, you will find abstracts of the contributions, addresses of the persons who presented a paper, and a short summary of the main results of discussions during the conference. Social Capital, Amsterdam, Royal Netherlands’ Academy of the Arts and Sciences.
N. 82, April 2004
Méthode et stratégie d’analyse des questions ouvertes du Panel Electoral Français
M. Brugidou, N. Mandran, M. Moine, A.-C. Salomon
Method and Strategy of Analysis of Open Quesitons of the French Electoral Panel: This article presents a voting survey method which includes many open questions, the detailed strategies of analysis employed and also, based on analyzed examples, the methodololgical advances – and their limits – of such strageties which mix quantitative and qualitative approaches. During the spring 2002 elections, a survey apparatus was set up including a panel which would permit the measuring of French political opinions and behavior over three waves of the survey. The first part of this article addresses problems engendered by this survey apparatus and how they have been handled. This involves the creation of a typology of open questions, the place of open questions in a questionnaire, interviewer/interviewee interactions, etc. Replies to part of the questions were transcribed and analyzed using the Alceste program. The choice of these questions and the use of a text analysis approach are discussed in the second part of the article. To return to a more explanatory approach, it is possible to cross-tabulate the classes produced by Alceste and the closed quesitons of the questionnaire. This procedure encounters several difficult problems that are both technical and epistomological that we address in the third part of the article. Open Questions, Analysis of Textual Data, Discourse Analyse, Themes, Alceste.
Une méthode de traitement sociologique de données filmées
A Method for Treating Filmed Sociological Data: This article presents a method of analysis of filmed observations developed during research which also involved opened interviews and questionnaires. This method of coding images resulted in the construction of standardized and quantitifed data on the informal behaivor or behavior that is only slightly objectified in the subjects’ dscourse. The steps of this standardization and the difficulties encountered are described. A preliminary analysis of these data shows the value of their construction and their confrontation or comparison with other data collected during the interviews. Surveys with filmed observations; Analysis of filmed data; Method for codifying filmed observations.
ONLINE OR NOT ONLINE? A COMPARISON OF OFFLINE AND ONLINE SURVEYS CONDUCTED IN THE CONTEXT OF THE 2002 GERMAN FEDERAL ELECTION
The article compares the results of three surveys that were conducted in methodologically very different ways in the run-up to the last German federal election. The first survey is a representative sample of the German population, the second one is a representative online survey of Internet users, the third one is an unsolicited open online survey with self-selected participants. The comparison yields considerable differences among the three surveys concerning demographic (age, sex, education) as well as substantial variables (voting intentions, interest in politics). It is also shown that these differences continue to exist after weighting the samples by sex and age. Voting Survey, Survey by Internet, Survey on the Web.
N. 83, July 2004
Computer Assisted Pretesting of CATI Questionnaires (CAPTIQ)
Observational pretesting or standard pretesting of CATI-questionnaires is problematic because the recording of observed respondent behavior has either to be carried out during the interview itself or after completion of the interview by filling out observation forms. Recording during the interview often places a heavy additional burden on the interviewer above and beyond conducting the interview correctly. Recording after the interview introduces the problem of reliability. In this paper, we present a method for Computer-Assisted Pretesting of Telephone Interview Questionnaires (CAPTIQ) which allows respondents’ behavior to be coded during the interview without overburdening the interviewer. The interviewers are able to code without interrupting the flow of the interview. The pretest data collected by CAPTIQ for each question and each respondent may be seen as longitudinal data which can be represented by a graph called an IPD (Interview Process Graph). The IPG, much like an electrocardiogram, reveals any problem zones occurying during the interview. As a result, information is collected on problems concerning response scales and on the learning process initiated by the respondents going through item batteries. Comprehension difficulties related to question wording or other factors also manifest themselves as oscillations of the IPG. The paper describes the CAPTIQ method and presents an illustration of the IPG by evaluating a CATI questionnaire used for a nationwide survey of health and media use. Computer-Assisted Telephone Interviews, Pretesting Questionnaires, Respondent Behavior, Interview Process Graph.
L’enquête d’opinion dans l’étude des pratiques sportives de montagne
Opinion Surveys in the Study of Mountain Sports: In the framework of the theory of social conflicts developed by J. P. Pagès, we have done research on practice of mountain sports. We first develope the structure of opinions on sports in the mountains with the objective of understanding the interplay between these opinions in the management of that space of practice. Beyond the presentations of the explanatory variables for these differecnes, the article presents the theoretical framework and methodology used in this particular survey. As part of the constructivist perspective, the construction of opinion questionnaires requires the mastery of a particular methodological procedure that is also presented here. Mountains, Sports, Public Opinion, Social Conflicts, Constructivism, Structuralism..
REPRESENTATIVENESS PROBLEMS INHERENT IN ADDRESS-BASED SAMPLING AND A MODIFICATION OF THE LESLIE KISH GRID
The problem of drawing a person from a household often occurs at the final stage of an address-based sample survey design; e.g., in telephone surveys after the households are contacted. The Kish grid gives an algorithm for this random selection. When evaluating the representativeness of samples obtained by this design, researchers often refer to the undersampling of males and overrepresentation of elderly people, the phenomenon originating from the practical realisation of the interview; e.g., males are more difficult to find at home, and less willing to participate. In the paper, some theoretical evidence will be given that explains the representation problems in the case of address-based samples without considering these assumptions. We found that, contrary to the opinion held by some researchers, the grid is not capable of providing representativeness by gender and age. The misconception stems from the fact that when the Kish grid was developed in the 1950’s, both randomness and representativeness could be achieved using the method, due to the household structure of the USA. We show that today this does not hold for most countries. Finally, a modification of the Kish grid is suggested that is more appropriate for selecting a representative sample. Since the performance of the grid depends on the household structure within the target population, its modification varies country to country. In the paper, those countries are considered where the Kish grid is in use. The main lesson is that in cases when the implementation of both address-based samples and population register samples is feasible, it may be worth considering the above mentioned aspects and deciding to choose a population register sample. Address-based Sample, Kish Grid, Kish Tables, Population Register Sample.
ESDS QUALIDATA ONLINE “WHITHER COMMUNITY STUDIES?” CONFERENCE REPORT
This report was prepared by one of the organizers of this conference and the head of the Economic and Social Data Service (ESDS) Qualidata service. The conference was in honour of Colin Bell and brought together several of his former colleagues and students, and covered past, present and future topics of community studies. Colin Bell, Great Britain, Community Studies.
N. 84, October 2004
L’analyse longitudinale de réseaux sociaux totaux avec SIENA – Méthode, discussion et application
Ainhoa de Federico de la Rúa
Longitudinal Analysis of Total Social Networks with SIENA – Method, Discussion and Applications: This article explains the functional basis of the SIENA statistical model for the analysis of longitudinal total social networks. In the first part, we explain the specificity of statistical models for analyzing network structure and give a brief history of the major models proposed in the past which produced dynamic models such as SIENA. We then present the basis on which SIENA is constructed and we discuss the theoretical implications. Model specification follows with a presentation of the principal effects which can be studied with this model. We end with an example of application on a network of international friendships. Longitudinal analysis of social networks, SIENA.
Assessing the Human Development Index through the Structure of Welfare
Erik H. Cohen
In this study, we assess the validity of the United Nations’ Human Development Index (HDI), a multi-faceted indicator of national welfare. For this purpose, the HDI is introduced into a previously published typology of welfare indicators (Cohen, E. H. 2000). The HDI appears to be a well-balanced indicator of welfare, close to the semantic center of the structure. However, it is somewhat more strongly correlated with the education and population rates regions of the structure than with the production and media regions. Human Development Index (HDI), Welfare, Smallest Space Analysis (SSA), External variables.
De la portée des récits de vie dans l’analyse des processus globaux
The Impact of Life Histories in the Analysis of Global Processes: This article shows the far-reaching impact of research based on the collection of life histories. While the compiling and comparison of life histories, item by item, allows us to construct typologies, the analysis of internal coherence of each life history and the comparison of individual and collective logics permits us to truely treate the complexity of social phenomena and related certain seemingly contradictory processes. To establish the scientific validity of such a procedure, researchers have experimented with different approaches: discourse form analysis, identifying biographical indicators and events, cross verification of life histories with other materials, preceding a qualitative survey with a quantitative survey, crossing life histories between each other at the collective membership leve, contextualizing them by situating them within monographies. Life Histories, Comparing Life Histories, Biographical Events, Calenders, Qualitative Research.
N. 85, January 2005
DES LOGICIELS D’ANALYSE TEXTUELLE AU SERVICE DE L’IMAGINATION SOCIOLOGIQUE
Text-Analysis Programs Serving Sociological Innovation: Beginning with the idea that what text analysis programs do is closely associated with their manner of application to a corpus, the Methods Thematic Network of the French Sociological Association (AFS) organized the analysis of the same corpus by different programs. This comparative perspective allows us to show that these programs are not simple instruments for generating proofs, but, on the contrary, constitute applicable resources — among others — for testing interpretations, for enriching theories, and thus for encouraging sociological innovation. Textual Analysis, Computer Programs, Prospéro, Calliope, Trideux, Alceste.
Le logiciel Prospéro à l’epreuve d’un corpus de resumes sociologiques
The Program Prospéro Tried Out Against A Corpus of Sociological Abstracts: The objective of this article is to test the potential of the program, Prospéro, for analyzing the abstracts of the first congress of the French Sociological Association (AFS). Prospéro was conceived as a program for the analysis of corpora of texts characterized by a certain variability (particularly in their use of argumentation and time). We consider these abstracts as a “dossier”; that is as a collection of texts characterized by the unknowns and the obligations of authors proposing a presentation in a Thematic Network (TN) session. To begin with, we show how much recoding — done by the researcher in the case of Prospéro — is decisive in determining the level of analysis and of hypothesis testing. We then propose a model describing the expectations of a potential presenter and construct categories, collections and fictitious entities to test this model. The program, in this context, functions as an instrument for identifying singular properties of the corpus (for example, judging the use of a category or a collection to characterize a TN). But it also identifies atypical texts which can encourage the modification of the model. Finally, we study how presenters adapt to the requirements of a proximity between the abstract and the call for papers distributed a few months before by the TNs. This approach shows that only a few TNs bring together texts sharing similar characteristics. This work permits us to identify certain interesting aspects of French sociology and, in particular, the possibilities and the limits of Prospéro. The program functions as a research tool which can test a researcher’s interpretations, hypotheses and models. The program would be even more useful if corpora could be complemented by texts permitting a variation (particularly historic) which would benefit the analysis of sociological scientific production. Computer Programs, Prospéro, Analysis of Corpora, Coding, Interpretation, Proximity, Sociology.
CARTOGRAPHIE DU PREMIER CONGRES DE L’ASF AVEC LA METHODE DES MOTS ASSOCIES
Mathilde de Saint Leger, Karl M. van Meter
The 2004 congress of the French Sociological Association (AFS) included more than 1,000 presentations. Considered as representative of contemporary French sociology, the corpus of abstracts is analyzed with Colliope, a text-analysis computer program based on the associated-words method. A cartography of the entire corpus shows the rather varied themes which are interconnected but also weakly structured. The role of the term WOMAN as attractor, in the sense that it is the link between many different themes. Cartographies of sub-sets of abstracts, grouped together by thematic network (RT), reveal more structured results which remain strongly interconnected. It would be interesting to compare these results with those of other text-analysis approaches which should be done soon. Text Analysis, Calliope, Associated-Words Method, Contemporary French Sociology.
De quoi parlent les sociologues réunis en congrès ? Eléments de complémentarité entre une analyse lexicale ouverte et le cumul de variables fermées
The first congress of the AFS was an occasion to provide a synthesis of the subjects of conversation current in French sociology, and we will attempt to analyze this vocabulary and its frequencies. This article confronts this first analysis with a more quantified one, based on the distance from the means, to find how the two approaches can be combined. The lexical analysis, with the Alceste program, stresses the co-occurence frequencies when words appear together in an abstract. The most frequent co-occurences are found in the class “society-inequality”, and the class “theory-authors”. Calculating the percentage difference from the means (program Trideux) stresses the differences, and puts into comparable position similar differences. In this case, the words that differ the least from the means will be those used the most by sociologists and are found in the classes “method” and “public polity”. If the authors are introduced into the analysis with variables for sex, institutional origin, location, and are associated with the words they use, we can combine the two methods and still conserve the families of words and those words which are the most significant, while cross-tabulating them with accumulated individual characteristics. The most discriminating variable est thus sex, with a concentration of women in the family “society-inequality”, and a concentration of men in the family “theory-authors”. The class on which there is the most agreement between the two methods is not mentioned individually in either method. It is the class “work-company”. Vocabulary, Occurrence of Key Words, Alceste, Trideux, Complementarity of Methods, AFS Congress, France, Sociology.
N. 86, April 2005
Aux abonnés absents : liste rouge et téléphone portable dans les enquêtes en population générale sur les drogues
François Beck, Stéphane Legleye, Patrick Peretti-Watel
Missing Subscribers, The Red List and Mobile Telephones in General Population Surveys Concerning Drugs: Survey designers using Computer-Assisted Telephone Interviews (CATI) are confronted with four types of populations: Owners of land-line telephones which include those on the Red (unlisted number) List; Owners of only mobile telephones; and those who don’t own a telephone. Over the past few years, the proportion of land-line users has decreased as only mobile users increased. Even if the telephone remains an excellent means for contacting individuals residing in France (only 1.2% have no telephone), the increase in Red List numbers and only mobile users complicates methods developed since the early 1990; the choice of not contacting these persons should be made on a knowledgeable basis. Two types of research presented here allow us to extend our knowledge of these populations in relation to the use of psychoactive substances: a random survey of land-line users (n=13,685) allows us to characterize Red List members concerning declared drug use and associated opinions; a survey (n=2,009) concerning the social representation of drug users, including a sub-sample of only mobile users (n=201) permits us to obtain a summary description of the later and measure differences between them and line-liner users on these same questions. Methodology, Drugs, Telephone Surveys, Alcohol, Tobacco, Cannabis.
Faire et défaire des groupes : L’information chiffrée sur les « populations difficiles à atteindre »
Making and Unmaking Groups, Numerical Information on “Difficult-Access Populations”: After having underlined the divergent points of view between statistics and sociology on how to approach “difficult-access populations”, this article proposes to explore the different manners of defining a distance from a norm, the periphery in relation to a center, and to show how, according to the choice of criteria of inclusion adopted, the size and meaning of the statistical group thus constituted are extremely variable. Hidden Populations, Methodological Sensitivity, Surveys.
Qui construit les données du sociologue ? Les problèmes posés par l’analyse secondaire des fichiers des licences fédérales
Who Makes Sociology’s Data? Problems Posed by Secondary Analysis of National Sport Club Membership Files: The secondary analysis of national files of sport club membership is a possible research technique in investigating careers in sports. But before engaging in such work, one must know if the stored data can be used for scientific research. In this framework, we examined the yearly membership files of the Fédération Française de Voile. This critical analysis had two aspects. First, the information on members (wind-surfing or not, competitions or not) was compared with that provided by another high-quality source. This comparison reveals major differences. To interpret the meaning of these differences, we reconstructed the means by which the data was produced for the national files. This work permitted us to evaluate the pertinence of the two recorded variables for eventual longitudinal studies. The result is that the data could be used in longitudinal studies with only slight confidence. The data tends to describe the margins of interpretation left to local authorities by the federal guidelines, and not the real activity of members. Secondary Analysis, National Sports Leagues, Club Membership, Careers, Longitudinal Analysis.
Report on the Third International Workshop on Comparative Survey Design and Implementation
The main goals of the Comparative Survey Design and Implementation (CSDI) program are to provide a forum and platform for researchers concerned with methodological issues in cross-national and cross-cultural surveys; to foster co-operation on projects aimed at improving comparative research; and to disseminate knowledge and promote best and good practices. The Third International Workshop on CSDI was held in Madrid on 10-12 March 2005, during which plenary sessions, at which participants presented papers, were combined with break-out sessions in which work groups met to discuss work in progress, as well as a common set of questions described below. Cross-National and Cross-Cultural Surveys, Comparative Methodology.
N. 87, July 2005
UTILISATION DES CARTES D’AUTO-ORGANISATION DANS LA CLASSIFICATION, ILLUSTRATION PAR DES EXEMPLES
P. Rousset, C. Guinot, J. Vero
Self-organising maps used for classification, illustration with several examples: The aim of this paper is to illustrate, with several examples, the possible uses of self-organizing maps (som), especially in classification and data-set visualization purposes. This classification method, based on the unsupervised learning algorithm of Kohonen, is presented here in a practical manner, and applied in the following fields: characterisation of human skin, Polish national daily consumption of electricity, training supply and professional careers. The diversity of the examples, in these application fields and for various statistical purposes (typology, classification of curves), illustrate many self-organizing map characteristics as compared to more traditional methods. In particular, we should mention, firstly, that the clustering method and its representation system are complementary; secondly, that som are sensitive to small distances and, finally, that they can be pertinent when the data set is large. These properties make it possible to visualize cluster proximities, local effects (restricted to a part of the population) and integrate many variables into the analysis (for example, as in data mining). Apart from clustering, self-organizing maps can also be used as a visualization tool for the representation of data-set intrinsic structure. In this case, as with projections to principal planes of the factorial analysis, this data-set representation can be considered a graphical support for any analysis method. To illustrate its characteristics, we use it to represent two hierarchical classification results. In this instance, the particularity of this technique comes from the specificity of its own symbolic representation. It gives to som a freedom that allows it to be well adapted to complex structures such as non-linear ones. In this paper, self-organizing maps are presented first as a clustering method, and then as a tool for the visualization of dataset intrinsic structure. Self-Organizing Maps, Cluster Analysis, Visualization Tools, Large Data Sets.
Being, Having and Doing, Modes of Existence: Confirmation and Reduction of a New Scale Based on a Study among Israeli Female Teachers, Student-Teachers and Counselors
Erik H. Cohen, Rachel Sagee, Rivka Reichenberg
Following Fromm’s (1976) and Rand’s (1993) theory, a scale of 51 items on attitudes towards the Being, Doing and Having “Modes of Existence” has been constructed (Reichenberg 1996). Our objectives are two-fold: first, an examination, in parallel, of the validity of the scale with two methods: Factor Analysis and Facet Theory. And second, a shortening of the scale from 51 questions to a more compact one. The present article is based on empirical data gathered from 386 female teachers, students and counselors in college education departments in Israel in 1995 and 1997. Modes of Existence, Facet Theory, Factor Analysis.
La « socio-anthropologie » : champ, paradigme ou discipline ? Regards particuliers sur les entretiens de longue durée ou d’observation
“Social Anthropology” – Field, Paradigm or Discipline? Particular views about prolonged or observational interviews: This article attempts to describe the methodological aspects of what is called « socio-anthropological » research. It describes the process of “rapprochement” between observation and interviewing in certain research protocols in the context of contemporary societies, particularly in the case of in-depth interviews : interviews over a long period of time with the same person. Comparing the diachronic and synchronic approches of these studies with extended interviews, as presented in several recents books, we can say that the distinction betwen qualitative sociology and anthropology in modern research fields loses its relevance. The consequence of this sociological and anthropological overlap is the creation of a new field of human sciences, and perhaps a new paradigm or discipline. Interviews Over Long Periods of Time, Observational Interviews, Socio-Anthropology.
N. 88, October 2005
LA “MISE EN VARIABLES” DES TEXTES : MYTHE OU REALITE ?
Gaël de Peretti
Turning Texts into Variables, Mythe or Reality: The statistical analysis of responses to open questions is the subject of many controversies between supporters and critics, but also among supporters concerning the choice of methods of analysis and of data construction. With an example, we focus our attention on two controversies: the interest of using open questions; the use of techniques for transforming corpora of texts (normalization and lemmatization). The pertinence of open questions is established, but turning texts into variables remains a delicate task in textual analysis and requires reflexive considerations necessary particularly in the interpretation of results. Textual Analysis, Open Questions, Quantification, Normalization, Language, Lemmatization.
PLAIDOYER POUR LE BRICOLAGE ET L’ENRACINEMENT DES METHODES D’ENQUETE DANS LE TERRAIN : L’EXEMPLE D’UNE RECHERCHE SUR LE CHANGEMENT DANS LES SERVICES PUBLICS LOCAUX
Plea for Adapting and Embedding Survey Methods in the Field, Example of Research on Changes in Local Public Services: Beginning with a distinction between theory and methodological practice, this article attempts to show, on the basis of a case study, the advantages of two types of methodological operations: adapting and embedding survey methods in the field. The description and the justification of procedures developed by these methodological operations will lead me to the discussion of the conditions of validity and use of certain methods of scientific investigation : classic methods — observations, interviews, etc. — or more uncommon — action research-intervention, scenarios method. I will also show the triple advantage of methodological adaptation and field embeddedness: permitting the research to translate his or her subject and objectives into terms compatible with the type of intended intelligibility; to permit access to field work that can prove to be fruitful for both the research and the actors; construct a research format adapted to the actors’ experience and permitting the translation of the researcher’s abstract questioning into terms compatible with this experience. Methodological Practice, Methodological Procedures, Adaptions, Embeddedness, Scenarios Method.
STUDYING SURVEY INTERVIEWERS: A CALL FOR RESEARCH AND AN OVERVIEW
Karl M. van Meter
To encourage an ongoing research project on survey interviewers, the BMS publishes here an overview of the extensive research on the subject that has appeared in the BMS over the past twenty years. While requesting contributions — particularly from the vast survey methodology experience of colleagues in the Netherlands, Germany and the United States — we share preliminary results from a detailed French study which made a statistical analysis of interviewer performance and followed up with direct face-to-face interviews with interviewers whose “statistical profile” was far from “standard”. The interviews with these interviewers provide a surprising description of the survey business milieu. Interviewers, Surveys, Literature Review, Survey Business.
N. 89, January 2006
ABSENCE DE RIGUEUR DANS LES SONDAGES D’OPINION : LES LECONS D’UN RETOUR D’EXPERIENCE
Patrick Che, Jean-Pierre Pagès
Absence of rigor in opinion polls, Lessons learned from experience: The authors statictically analyze data collected during a 2003 CAPI (computer-assisted personal interview) public opinion poll of 2,520 persons. Interviewer data collection is examined for number of interviews, lenght of interviews, response modality use and time between interviews. This analysis identifies a number of interviewers whose data production raises certain questions. Far from placing responsibility only on interviewers, the authors critique certain practices in polling that encourage the production of low-quality data. Opinion Polls, Interviewer Behavior, Data Quality.
INDELICATESSE ET MANQUE DE RIGUEUR DANS LES SONDAGES : ENQUETE AUPRES DES PROFESSIONNELS DU SONDAGE
Fabienne Pagès, Marianne Tribel, Alexis Bonis-Charancle
Indelicate Behavior and Lack of Rigor in Polling, Investigation Among Polling Professionals: Following statistical controls concerning the French EDF electricity board’s « Énergie et nucléaire » opinion poll survey (see references 1, 2 and 3), EDF decided to complete the study with an investigation among polling professionals to clarify observed data anomalies and to begin to understand how to remedy these anomalies. The authors were guided by these anomalies in conducting their interviews with survey interviewers. The dark picture which emerges from this study may be more somber than the real situation since polling professionals were encouraged to identify insufficiencies in the polling process they had witnessed during their work (with different institutions), particularly over the past few years. Certain of these insufficiencies reflect an increasing lack of rigor on the part of polling firm clients and the polling firms themselves. Opinion Polls, Polling Industry, Interviewers, Data Quality, Clients of Polling Firms.
INTERVIEWER AND SURVEY RESEARCHER: MUTUAL DEPENDENCIES
Johannes van der Zouwen
The explanation of behavior requires the use of subjective concepts like opinions and attitudes. For the measurement of these non-directly observable concepts, strict standardization of the process of question answering is necessary. If interviewers have to be used to ensure a proper process of data collection, these interviewers may also hamper the standardization. Is the interviewer an indispensable help to researcher and respondent, or a difficult obstacle for obtaining unbiased and comparable answers? To answer this question, transcripts of survey interviews have been analyzed; especially the methods interviewers use to repair inadequate answers of respondents. The outcome of this analysis, and insights gained from cybernetics, make clear that it is ineffective to strongly standardize repair behavior of the interviewers. The analysis also points at the mutual relationship between interviewer and survey researcher: the survey researcher, when constructing the questionnaire and giving instructions about appropriate ‘repair’ behavior may also be a help or a hindrance for the interviewer. Standardized Interview, Personal Interview, “Repair” by Interviewers.
[BMS Editorial Committee]
Many members of the International Sociological Association’s RC33, Research Committee “Logic and Methodology”, contributed information and contacted other interested researchers to have them contribute to this “interviewer/data quality” thematic issue. Several contributions of different types from France, Switzerland, Sweden, Germany and the Netherlands are included below. Interviewers, Data Quality, Research Resources.
N. 90, April 2006
INTRODUCTION AUX METHODES D’APPARIEMENT OPTIMAL (OPTIMAL MATCHING ANALYSIS)
Laurent Lesnard, Thibaut de Saint Pol
Introduction to Optimal Matching Analysis: This paper provides an introduction to a new statistical technique for describing sequences: Optimal Matching Analysis (OMA). Borrowed from biology, this technique is based on rather simple principles that can be adapted to suit the theoretical requirements of different analyses. Since OMA is not based on any causal assumptions, it is particularly well-adapted for sociologists who retemporalize action by analyzing it as a process. Two applications of Optimal Matching to time-use data are proposed to demonstrate its flexibility and also its sociological interest. Optimal Matching Analysis, Sequences, Epistemology, Dining, Temporality of Evening Meal, Work Schedules.
QUALITY ISSUES IN INTERVIEW SURVEYS – SOME CONTRIBUTIONS
This article is the introduction to a thesis with the same title (done at the Department of Statistics of the University of Stockholm) which deals with some of the issues associated with the quality in interview surveys. The interviewer has many tasks, and some of them are extremely error-prone. The author describes some general themes and how these are usually handled. She also suggests some new approaches and ideas for further work, both when it comes to gaining cooperation and when it comes to doing a good job in administering the question-answer process. She also evaluates a specific procedure to reduce nonresponse errors in interview surveys in terms of costs and error reduction. Survey Data Quality, Survey Interviewers, Survey Cooperation, Survey Nonresponse.
COMPTE-RENDU DES JOURNEES INTERNATIONALES : L’ANALYSE SECONDAIRE EN RECHERCHE QUALITATIVE, UTOPIE OU PERSPECTIVES NOUVELLES ?
Magdalini Dargentas, Mathieu Brugidou, Dominique Le-Roux, Annie-Claude Salomon
Report on the International Symposium: Secondary Analysis in Qualitative Research. Utopia and Perspective?: Qualitative secondary analysis is a new field of research in France. A symposium on this issue was organized by the group CAPAS and held in Grenoble (MSH-Alpes) on the 3rd and 4th of November 2005. The following issues were discussed: state of the art in qualitative research with interviews in social and human sciences; Anglo-Saxon and French practices of secondary analysis; ethical, legal and methodological questions regarding archiving and secondary analysis of interviews; epistemological questions. Qualitative Research, Interviews, Secondary Analysis, Archiving, Ethical and Legal Issues, Tools and Software for Secondary Analysis, Methods.
PRICING ENVIRONMENTAL AMENITIES: AN ASSESSMENT OF THE CONTINGENT VALUATION METHOD
Patricia A. Gwartney, Anthony V. Silvaggio
The authors are preparing an essay that will introduce non-economists to contingent valuation (CV) – a survey method developed by economists to estimate prices for environmental assets. Here, they present the CV method with references in the scieitific literature describing its development and evolution. Contingent Valuation (CV), Environmental Economics, Survey Research.
N. 91, July 2006
SOCIAL SCIENCES AND THE IFSE CONFERENCES
International Federation of Science Editors (IFSE), associated with UNESCO, organized its twelfth conference in October 2004 in Merida, Yucatan, Mexico. A debate among the organizers resulted in an opening toward the social sciences which had not been represented up until then. The following three articles constituted the first IFSE social science session. IFSE, Scientific Editing, Social Sciences.
Editing a Normal Science Journal in Social Science
Linton C. Freeman
This paper displays some differences between “normal science” journals in fields like physics and chemistry and those in “non-normal science” fields in the social sciences. It shows that one journal, Social Networks, looks more like a normal science journal than a typical social science journal. I argue that the normal science properties of social network research stem from its use of both graphic images and mathematical models and from the availability of computers that permit the analysis of relatively complex data structures. Normal Science, Social Sciences, Social Network Analysis, Scientific Journals.
SCIENCE EDITING IN THE SOCIAL SCIENCES: METHODS BELONG TO ALL OF US
H. Russell Bernard
Social sciences should be included in every IFSE conference because social science is now, and has always been, real science, including both the production of basic knowledge and the translation of that knowledge into technologies that have real impact on people’s lives. The historical development of several social science methods, including probability theory and regression, provide examples of these developments that have always been both quantitative and qualitative. Social Science Methods, History of Science, Qualitative and Quantitative Methods.
AUTHORS AS “ARTISTS” OR “HEAVY WEIGHTS” IN SCIENTIFIC PUBLISHING: THE SOCIOLOGICAL ANALYSIS OF SCIENTIFIC LITERATURE AND THE BMS
Karl M. van Meter
There are only two types of major authors who survive in such a new and competitive field of scientific development as early sociological AIDS research. The “artists” survive by publishing in successive, more-or-less neighboring specialties, probably associated with newly available financial or institutional resources. They also appear to change co-authors comparatively often. On the other hand, “heavy weights” dominate a specific specialty closely associated with mainstream scientific developments which are therefore assured of financial and institutional backing. They appear to change co-authors relatively less frequently. These results, based on co-word analysis can be compared with the results of divers other methods of sociological analysis of scientific literature which tend to show that there are two types of world sociology, that there are multiple, not unique, scientific elites, and that there are different national sociologies which can be compared for content and structure. Scientific Literature, Scientific Publishing, Scientific Careers, Co-Word Analysis, Comparing National Sociologies.
N. 92, October 2006
An Inquiry into the Efficacy of a Complementary Training Session for Telephone Survey Interviewers
Claire Durand, Marie-Eve Gagnon, Christine Doucet, Eric Lacourse
This article presents an experiment in which a short cognitive training session was given at the private pollster conducting the 2004 Canadian Addiction Survey of 14,000 people. Overall, 79 interviewers worked on the project for a total of 2,436 interviewer-days, interviewers worked a maximum of 65 days, and the survey was in the field for 109 days. We devised a one-hour training session that focused on knowledge acquisition in two areas: sampling and selection within household, and reasons for refusals. The evolution of interviewers’ daily performance, analyzed using group-based trajectory analysis, allowed us to identify low (n=25) and high (n=42) performers two weeks before training. The training session was aimed at low performers and newly hired interviewers. Among these, all the interviewers working on training day (n=18) attended one of the three training sessions. The impact of the training was assessed in three areas: knowledge acquisition and attitudes (through a post-training questionnaire two weeks later), retention, and performance. The results show that trained interviewers a) felt that the training helped them understand sampling and the reasons for refusals b) were more likely than high performers to feel that their performance had improved since the beginning of the project, c) were likely to go on working on the project after training longer than the control group, and d) improved their performance by .04 per day after training, compared with .015 per day before training, almost reaching the performance of high performers. Survey, Interviewer, Training, Performance, Retention, Growth Model, Multilevel Analysis.
Integrating Exchange and Heuristic Theories of Survey Nonresponse1
John Goyder, Luc Boyer, Guil Martinelli
Interest in theorizing about survey nonresponse has accelerated over the past decade. Scholars such as Dillman, Groves and Couper have drawn from across the social sciences to extract conceptual schemes for why people decide to accept or decline requests for surveys. With the conceptual enrichment, however, has come complexity and some confusion. Beginning with analysis of current usage in articles on survey nonresponse in leading journals, the relationship between social exchange theory and the theory of psychological heuristics for compliance is examined using a two-dimensional conceptual scheme. Building on work by Groves and Couper, one dimension is defined as “amount of decision-making.” This helps locate the heuristic interpretation of response decisions. The second dimension, termed “strength of cultural factors,” clarifies the different varieties of social exchange theory that pertain to response behaviour. Viewed thus, it becomes clear that one source of confusion in theorizing about survey nonresponse is the heterogeneity of the various approaches which gather under the umbrella term “social exchange.” At the most general level, social exchange is the background landscape behind most theorizing about survey nonresponse, but once defined more specifically a fit into one or another quadrant within the two-dimensional conceptual scheme becomes possible. Theories of Survey Nonresponse, Surveys, Exchange Theory, Heuristic Theory.
Brève sociographie du SECOND Congrès de l’AFS – Bordeaux septembre 2006
Brief Socio-Graph of the Second AFS Congress in Bordeaux in September 2006: In January 2005, the BMS published three different articles using four different methods to analyze the abstracts of presentations at the First Congress of the Association Française de Sociologie (AFS). Here a preliminary analysis with two différent methods is done of the abstracts of presentations made at the Second Congress of the AFS. French Sociology, Text Analysis, Alceste, Trideux.
N. 93, January 2007
Quelles temporalités travaillent les entretiens biographiques rétrospectifs ?
What Temporalities Are at Work in Retrospective Biographical Interviews?: Open and indepth biographical interviews are material which can take into account the temporalities of individual trajectories. But that contribution answers to epistemological, theoretical and methodological conditions very different from those that govern the use of closed questionnaires and precoded chronological agendas. Discursive material, as such interviews, is highly specific. The interviewee is the principal source of information selection, the recounted episodes are presented from a subjective point of view, and the recounted events are interpreted. Thus the temporality of biographical accounts cannot be disassociated from the narrative process. On the other hand, the intrigue construction of one’s personal history presupposes the production of a temporal framework. The latter is at the same time a construction of coherence and a construction of perspective concerning the account of a trajectory. It is intelligible at the crossing points of chronological chains of events (the articulation of sequences coherently constructed) and temporal resolutions (exits from sequences that are part of the perspective). This grammar is tested against a corpus of accounts of entering the labor market made by youths with few academic qualifications. It is schematized as four temporal forms called destiny or vocation, repetition or fatality, opportunities or projects, rupture or mourning. Each of these four forms covers various trajectories from the point of view of social status and position which are worked over, compared, interpreted during the interview. Finally, by showing how narrative production necessarily implies the construction of a temporal form, this proposed procedure calls for a plurality of sociological or scientific categorizations of biographical time adapted to the characteristics of the empirical material analyzed. Biographical Interviews, Temporalities, Professional Trajectories, Chronotopes, Temporal Grammars.
La conceptualisation en sociologie : influences paradigmatiques et implications méthodologiques. L’exemple de la notion de risque dans le sport
B. Soulé, J. Corneloup
Conceptualization in Sociology: Paradigmatic Influences and Methodological Implications: Such a seemingly self-apparent notion as “risky sports” is widely used in social science literature. Through the in-depth analysis of this term, the purpose of this article is to examine a crucial step of the sociological research process: definition of concepts. After having underlined simplifications and paradoxes inherent in the common sense use of this term, we present an overview of its scientific uses and definitions. Some terminological difficulties soon appear, along with differing points of view and plural available conceptualizations. The critical, structural, subjectivist, culturalist, and systemic approaches can thus be differentiated. The choice made between these options entail methodological consequences that are also discussed. Two epistemological positions can eventually be adopted. On one hand, one can be in favour of the abandon of such a category, on account of its intrinsic shakiness, and because it is merely impossible to reach by consensus the one and only definition of “risky sports”. On the other hand, one can acknowledge the polythetic nature of this term. The latter will then be easily observed through the specific constructions of “risky sports”, as a scientific object, according to each research problematics. Conceptualization, Hazards, Risky Sports.
PRELIMINARY CO-WORD ANALYSIS OF THE 2006 CONGRESS OF THE ASSOCIATION FRANCAISE DE SOCIOLOGIE
Karl M. van Meter, Mathilde de Saint Leger
Following the publication in the BMS, and since then in a book, of our analysis of all the abstracts of the first congress of the AFS in 2004 with the Calliope method, we carry out a first analysis of the abstracts of all the presentations made at the second congress of the AFS in Bordeaux in September 2006. The results of the analyses of the two corpuses are compared, revealing a stable structure for both congresses and a complex interplay between the key words Women and Work. Structure of Sociology, Co-Word Analysis, Calliope, Association française de sociologie.
Researching Informal Education: A Preliminary Mapping
Erik H. Cohen
Factorial correspondence analysis and smallest space analysis are applied to an original data set on informal education. Despite its widespread use, long history and growing importance in the postmodern era, little theoretical or cumulative research exists on informal education. Multi-dimensional data analysis techniques are applied to a literature survey of research on informal education to classify and to organize the content issues and methods identified. A “structural state of the art” of recent research on informal education is presented, allowing a preliminary mapping of the field. Factorial Correspondence Analysis, Smallest Space Analysis, Informal Education.
N. 94, April 2007
Une autre manière de construire des milieux sociaux ? Pourquoi ? Comment ?
Monique Dalud-Vincent, Odile Rissoan, Rachel Gasparini
Another Way for Constructing Social Milieus? Why and How?: A secondary analysis of the INSEE “Education and Family” survey led us to rethinking the construction of social milieus (and sub-milieus) based on the French PCS social-professional categories and professions of a child’s six parents and grandparents in an attempt to better explain child/parents relations. The milieus are characterized by a particular homogeneity/heterogeneity (equivalence/non-equivalence of PCS’s). The sub-milieus are the decomposition of different milieus according to social origins (using the PCS’s). Social Milieus, Sub-Milieus, Homogeneity, PCS Social-Professional Categories and Professions, Parents/Child Relations.
Les “défaillances de la mémoire” dans les enquêtes de victimation
“Memory Defficiencies” in Victimization Surveys: Responses in retrospective surveys are subject to different biaises associated largely with problems of memory recall. In victimaziation surveys, it is necessary to provide exact dates for the events mentioned. Among errors of date by respondents, Wesley G. Skogan distinguished three forms of chronological displacement (telescoping). On the basis of two victimization surveys in France, it is possible to show evidence for two of these forms (backward telescoping towards the reference period and forward telescoping within the period of reference), but not for the third (forward telescoping toward the period of reference). However, a more detailed analysis of one of these surveys reveals the possible influence of the seasonal variation of delinquency on the survey responses. Retrospective Surveys, Victimization Surveys, Measures of Delinquency, Biais, Memory Recall.
Toutes choses égales par ailleurs – Comparer deux congrès de l’Association française de sociologie
Didier Torny, Patrick Trabal
All Things Otherwise Equal, Comparing Two Congresses of the French Sociological Association: This article analyzes the corpus of the 1,190 summaries of the French Sociologicial Association congress (Bordeaux, 2006) from a comparative point of view. It is focused on the multiple means and significances of a comparative study by mobilizing three approaches. The first aims at questioning the permanence of the summaries’ authors and of the congress thematic organization. The second mobilizes certain Prospéro software algorithms to identify the more remarkable lexical and semantic variations. Finally, we continue the study certain objects whose heuristic value has been shown in our preceding analyses. Questioning comparative work, we show, at the same time, how the software provides a cognitive economy and empowers the interpretative inquiry of social scientists. French Sociology, Comparative Method, Text Analysis, Prospéro.
Un Vert, ça va. Dix Verts, bonjour les débats !
One Green, That’s OK. Ten Greens, Hello the Debate! [a play on words associated with a French anti-alcohol slogan: “Un verre, ça va. Trois verres, bonjour les dégats”!, One glass (of wine), that’s OK. Three glasses, hello the damage!]. A textometric analysis is applied to the corpus of the political declarations favorable to the Pacte écologique of Nicolas Hulot by ten of French 2007 presidential candidates. Different utilizations of the notion of “generalized type” (Tgen) converge toward two forms of discourse which differentiates the speakers. Lexicometry, Generalized Types, 2007 French Presidential Elections, Ecology.
N. 95, July 2007
Interviewer and Respondent Survey Quality Effects in a CATI Panel
Especially in panel surveys, respondent attrition, respondent learning, and interviewer experience effects play a crucial role with respect to data quality. We examine three interview survey quality indicators in the same survey in a cross sectional as well as in a longitudinal way. In the cross sectional analysis we compare data quality in the mature original sample with that in a refreshment sample, surveyed in the same wave. Because in the same wave an interviewer survey was conducted, collecting attitudes on their socio demography, survey attitudes and burden measures, we are able to consider interviewer fixed effects as well. The longitudinal analysis gives more insight in the respondent learning effects with respect to the quality indicators considered by considering the very same respondents across waves. The Swiss Household Panel, a CATI survey representative of the Swiss residential population, forms an ideal modelling database: the interviewer – respondent assignment is random, both within and across waves. This design avoids possible confusion with other effects stemming from a non-random assignment of interviewers, e.g. area effects or effects from assigning the best interviewers to the hard cases. In order to separate interviewer, respondent and wave effects, we build cross-classified multilevel models. Survey Data Quality, Centralised CATI, Random Assignment, Longitudinal Effects, Cross-Classified Multilevel, Interviewer Effects, Respondent Effects.
MÉTHODES D’ANALYSE DU CHANGEMENT FONDÉES SUR LES TRAJECTOIRES DE DÉVELOPPEMENT INDIVIDUEL : MODÈLES DE RÉGRESSION MIXTES PARAMÉTRIQUES ET NON PARAMÉTRIQUES
Véronique Dupéré, Éric Lacourse, Frank Vitaro, Richard E. Tremblay
Longitudinal Methods Based on Individual Development Trajectories – Parametric and Non Parametric Mixed Models: Generalized linear mixed models encompass a variety of modern longitudinal analytic approaches based on individual developmental trajectories. These models overcome many important problems inherent to other traditional analysis of longitudinal data. They all rely on two basic levels: the lower one express, through a set of parameters, the individual pattern of change over time (within-individual change), whereas the upper level captures the variations between these parameters describing individual trajectories (between-individual differences in change). However, other characteristics distinguish different sorts of mixed models, such as their assumptions concerning the distribution of the trajectories within the population. This introductory article presents the basic linear mixed model assuming a normal distribution of the unobserved heterogeneity, and the nonparametric mixture model that relies on a discrete approximation of the unobserved heterogeneity. Before comparing these two models, the first section of the article gives a general description of the notion of individual developmental trajectories. Mixed Models, Multilevel Models, Developmental Trajectories, Distributional Assumptions, Continuous Latent Variable, Discrete Latent Variable.
Compte rendu succinct de l’analyse du Congrès de l’AISLF (Tours 2004)
Succinct Report on the Analysis of the AISLF Congress (Tour 2004): Following the line of work of the January 2005 issue of the BMS (n. 85, based on the analysis of the abstracts of the February 2004 Association Française de Sociologie Congress), we try to demonstrate the optimal use that can be made of text analysis software in human sciences. But in this case, we use the reports from the most recent meeting ot the Association Internationale des Sociologues de Langue Française (AISLF) in July 2004. We use the ALCESTE software in a first step to extract a maximum of “terminal”lexical classes (15 in this case). The second step uses the stages of ALCESTE’s hierarchically descending classification analysis to examine the possible origins of these successive partitions. French Sociology, AISLF, Text Analysis, ALCESTE, Correspondence Analysis.
Sensibilité du chaos aux conditions extérieures
Sensitivity of Chaos to External Conditions: Using the example provided by Guy Tchibozo in the BMS (n. 74, 2002), the author uses simulation methods to study the results of various initial conditions for the model, showing the the results can be chaotic or stable depending to specific parameters. Simulation in Sociology, Unstable and Chaotic Results, Stability, Model Parameters.
N. 96, October 2007
Questionnaire Pretesting Methods: Do Different Techniques and Different Organizations Produce Similar Results?
Jennifer Rothgeb, Gordon Willis
During the past 15 years there has been a significant increase in the use of novel forms of questionnaire pretesting. Various evaluation techniques have been evaluated and the strengths and weaknesses identified. Limited research has been conducted about the effectiveness of cognitive interviews in ameliorating questionnaire problems. However, this research has not allowed for separating effects of pretesting method from effects of the organization applying these methods. The objective of our research was to determine how various methods, applied across three survey research organizations (The Census Bureau, Westat, Inc., and Research Triangle Institute), were found to predict problems. For this research, multiple researchers within each organization applied three pretesting methods (informal expert review, a formal cognitive appraisal system, and cognitive interviewing) to three questionnaires, according to a Latin Square-based experimental design. A classification scheme was developed to code problems identified through all methods. The problems detected by each pretesting method and organization were quantitatively compared. We will present results about the degree of agreement between organizations and between methods. We will make conclusions about the utility and appropriateness of each method, and the degree to which the results are common and systematic, as opposed to haphazard and organization-specific. Pretesting Methods Evaluation, Cognitive Interviewing, Expert Review, Cognitive Appraisal System, Question Testing.
Elaboration d’un ludogramme pour les sports sous forme de duel
Construction a Sports Ludogram as a Form of Duel: Playing a sport is not anodine; it is the result of a symbolic representation of that sport. Studying the gestures of a player during a match, before or after wining a point, after an adversary’s double faut or after a final winning point, can reveal that social symbolism by other than traditional methods. Sports reveal intense manifestations of emotions and how is one to account for thedifferent modes of these emotional manifestations in relation to the stages of a match? Tennis, table tennis, field tennis, badminton and volleyball serve as our research terrain. We explain the development followed in constructing a methodologic tools to data collection called the ludogram. Ludogram, Sports, Observation, Methodology, Tennis, Terrain, Gestures.
Comparaison et influence de deux méthodes de recueil de données différentes sur les résultats globaux d’une enquête quantitative
Comparison and Influence of Two Data Collection Methods on the Results of a Quantitative Survey: This article proposes to analyze and compare two sets of data obtained from the same survey questionnaireits by two different methods. In the framework of our research on social workers which includes a quantitative aspect, we used the Internet as a data collection tools along with the more classic method of sending out a questionnaire by mail. This article proposes to consider the different aspects to comparison possible based on a concrete analysis to the two sets of data obtained from the two samples according to a specific variable associated with our research topic. Then, based on a comparative analysis, we examine the cumulative consequences of the mail questionnaire and the Internet questionnaire data and their implications for our research. We also show how the origin of the data can or cannot influence research results. Data Collection, Online Questionnaires, Comparisons, Samples, Questionnaire Modes.
N. 97, January 2008
La méthodologie des sondages électoraux de l’élection présidentielle française de 2007, chroniques d’un problème récurrent
Election Poll Methodology and the 2007 French Presidential Elections – Chronicle of a Recurring Problem: The 2002 presidential election polls will be remembered for a long time because their error in predictions had dramatic consequences. What influence did those elections have on French opinion poll research and was it able to revise its methods? Was it able to improve its voting estimates? Research has shown that French polls are now more able to distinguish between voting for the left and the right than in 2002. However, French polls have not improved their capability to estimate intentions to vote for major candidates. The discrepancies between means for twelve polls published during the last week before voting vary from 1.8 points for Royal to 3.2 points for Sarkozy and 3.8 points for Le Pen. These results show that the estimation error for Le Pen is the same in 2007 as in 2002, but in the opposite direction. As in 2002, samples were biased concerning level of education – strong under-representation of the less educated – as well as for voting profile – strong non-declaration of voting for Le Pen. One can thus conclude that the problem of estimating the extreme right vote remains in its entirety and that polling institute methods have not changed greatly. Cooperation between the institutes and researchers should permit further explanations and possible solutions. Voting Polls, Estimation Methods, Sampling, France, 2002 Elections, 2007 Elections.
CO-WORD TEXT ANALYSIS APPLIED TO POLITICAL SCIENCE: 2006 INTERNATIONAL POLITICAL & “PARAPOLITICAL” HEADLINES
Karl M. van Meter, Mathilde de Saint Léger
A system of text analysis by co-occurrence of key words, Calliope, is applied to the 2006 data set of daily headlines and titles of media reports on international political and “parapolitical” events produced by the journal, Intelligence. These political (publicly avowed use of political power) and parapolitical (unavowed use of policital power) developments are divided into three successive four-month periods, each producing a two-dimensional “strategic digram” with a centrality and a density axis. Graphs over time of “emerging” and “declining” key words are also produced. These results permit an original and intriguing analysis and interpretation of these 2006 events. Text Analysis, International Politics, Parapolitics, Calliope, Co-Word Analysis.
Analyser les pratiques discursives en sciences sociales : Journée d’études du CEDITEC à l’Université Paris XII, le 27 avril 2007
Johannes Angermüller, Laurent Jeanpierre, Caroline Ollivier-Yaniv
Analyzing discursive practices in social science – The CEDITEC symposium at the University Paris XII on 27 April 2007: This symposium was organized around the question of the convergence between discourse analysis and certain currents of qualitative social science. The symposium was divided into two parts: “Ethnography, Field Work, Actor”, what happens when social actors express themselves during surveys; and “Lexicometry, Corpus, Announcement”, concerning certain analytical, conceptual et technical tools generated by the sciences of language, and, in particular, text analysis software and its use in current sociology. Discourse Analysis, Text Analysis, Interviews, Sciences of Language, Sociology.
The International Conference on Methodology of Longitudinal Surveys (MOLS), 12-14 July 2006
On 12-14 July 2006, at the University of Essex, Peter Lynn, with the UK Longitudinal Studies Centre and the International Association of Survey Statisticians (IASS), organized the International Conference on Methodology of Longitudinal Surveys (MOLS 2006). The BMS has collected material from the Web sites of MOLS 2006 and the International Statistical Institute, of which the IASS is a section, to constitute this conference report. MOLS 2006, Longitudinal Surveys, Panels, Longitudinal Analysis.
N. 98, April 2008
Harmonisation of Demographic and Socio-Economic Variables in Cross-National Survey Research
Jürgen H. P. Hoffmeyer-Zlotnik
The aim of the present paper is to demonstrate how demographic and socio-economic variables in cross-national comparative survey research can be harmonised. After a short introduction discussing the difference between translation and harmonisation, the path from a national concept and structure to an internationally-applicable measurement instrument is traced using the education variable as an example. Cross-national Surveys, Harmonisation of Variables, Education Variables.
Etude des opinions partagées dans des situations relationnelles variées : méthodologie des multi-appartenances
The Study of Shared Opinions in Varied Relational Situations – The Multiple-Association Methodology: This article examines the manner in which actors manage to form an opinion concerning certain aspects of AIDS, supposing that opinions are the product of interaction and not isolated acts. Cognitive and normative references are involved, but in societies of cultural diversity, an actor can be associated with different reference systems. Thus, we have adopted a structural approach for developing a theoretical model and a research protocol to study the (trans)formation of actor opinions associated with different relational contexts. The communicational convergence model we have adopted has led us to a survey tool capable of collecting information on actor opinions and the relational processes that produced them by comparing an actor’s opinions formulated in two different situations. The data were analyzed at the individual level, at the dyade level and at the triad level, showing the influence of socio-metric structures on the versatility of opinions. Communicational Convergence, Multiple Associations, Cognitive Heterogeneity, Structural Relations, Opinion Versatility.
L’utilisation du différenciateur sémantique en sociologie pour appréhender des facteurs agissant sur le choix des pratiques sportives
Using the Semantic Differenciator in Sociology to Understand the Active Factors in the Choice of Playing Sports: Surveys using different questionnaires are quite common in sociology, but indirect questioning is less common. Osgood’s semantic differenciator is one of the lesser-used tools. It employs a set of questions with word oppositions that the respondent judges according to a given stimulus. Initially used in psychology and marketing, it can be used in sociology. We show its use and the different possibilities of presenting results, along with multiple suggestions concerning its utilisation. It is applied in a study of sociology of sports students’ and players’ representation of five sports. Semantic Differenciator, Methodology, Stimuli, Representation, Sports, Players, Questionnaires.
Jeunes, génération numérique et sondage en ligne – L’exemple de deux enquêtes conduites auprès de jeunes québécois
Jacques Hamel, Gabriel Doré, Christian Méthot
Youth, Digital Generation and Online Surveys – The Example of Two Surveys Among Young Quebecois: After a brief presentation of the technical aspects of online surveys, the authors evaluate the advantages and disadvantages by using two online surveys: one on professional and social integration of the digital generation in the “new economy”; the other on college and university student values concerning sociological studies, social work and medecin. The authors ask whether or not this survey method can encourage young people’s participation in sociological surveys because of their familiarity with the new information technologies. Should researchers be encouraged to use such online surveys in order to obtain youth participation in sociological research? Youth, Digital Generation, Online Surveys, Sampling Methodology, Sociological Surveys, Response Rates, Likert Scales.
N. 99, July 2008
Characteristics of Personality and Identity in Population Surveys: AN Approache for operationalisATION and USE to explain life satisfaction
Max Haller, Bernadette Müller
Characteristics of identity and personality are usually not included in sociological studies. Questions of this sort are considered as belonging to the domain of psychology. However, sociology can also develop research instruments to operationalise them in a short and efficient way. This article, based on a representative Austrian population survey, shows that (1) it is possible to grasp such characteristics with relatively simple instruments, and (2) characteristics of personality and identity can explain a considerable amount of the variance in life satisfaction. Specific scales were developed and used to capture the “Big Five” personality factors, as well as four aspects of the self and eleven aspects of social identity. Big Five, Personnality Traits, Representative Surveys, Life Satisfaction.
Une approche réticulaire de la dynamique temporaire dans les narrations biographiques
Carlos Lozares, Joan Miquel Verd
Reticulated Approach to Temporal Dynamics in Biographic Narratives: This article proposes a specific approach to the analysis of biographic narratives by means of using the social networks perspective. More specifically, and more than other analyses, network formalisation can be used to capture the episodic memory present in narratives. To do so, the notions of environment/episode and of narrative passage are used to connect those features focused on the interaction and the dynamic dimension of narratives. The proposed procedure is illustrated by analyzing a fragment of a narrative interview. Biographic Narratives, Social Network Analysis, Narrative Passages, Network Text Analysis, Network Discourse Analysis, Social-Personal Networks.
Political Protest and Power Distance: Towards a Typology of Political Participation
Erik H. Cohen, José Valencia
Analyses of data regarding political protest collected in the World Values Survey are conducted to create a typology of non-institutional political participation in various nations. In a Smallest Space Analysis (SSA) of items related to political protest, a clear structure is found in one dimension, moving from the least involved and least demanding form of activity to the most involved and most demanding. In a two-dimensional SSA, the item “attended a lawful demonstration” emerges as the inflexion point differentiating between types of participation. Multidimensional Partial Order Scalogram Analysis (MPOSAC) indicates how countries of low, medium and high power distance are located in a two-dimensional structure. This typology of nations is considered in relation to Hofstede’s concept of power distance. Political Protest, Power Distance, Multidimensional Partial Order Scalogram Analysis, Smallest Space Analysis.
CARME-N – Correspondence Analysis and Related Methods Network, CARME 2007
Jörg Blasius, Michael Greenacre, Patrick Groenen, Michel van de Velden
In May 1991, a first correspondence analyse conference was organized in Cologne. Since then, three other such conferences have been organized and the Correspondence Analysis and Related Methods Network (CARME-N) has been established. The fifth conference took place in June 2007 in Rotterdam. Correspondence Analysis, Conferences, CARME-N.
N. 100, October 2008
ANALYSES OF A QUARTER OF CENTURY OF PUBLISHING AT THE BMS
Karl M. van Meter
The BMS started publishing in October 1983 and, one hundred issues later, we look at previous attempts to analyze the content the journal’s research articles. With this issue, we analyze all the research articles and ongoing research reports published in issue 1 to issue 99 with four different methods: Calliope co-occurrence of key word analysis; a descending (Alceste) and an ascending hierarchical classification analysis; and Trideux correspondence analysis. The results of these four analyses – published in this issue of the BMS – are commented and compared between themselves, and compared with the previous analyses of BMS content. BMS, Calliope, Alceste, Trideux, Thematic Content Analysis, Analysis of Scientific Literature, Bibliometrics.
Comment ont évolué les thématiques des 99 premiers numeros de BMS ? Analyse àVEC UN logiciel de fouille de texte
Mathilde de Saint Leger
How have the themes of the first 99 issues of the BMS evolved, An analysis with a text-mining software: The titles, abstracts (in French) and key words (in French) of the 315 articles published in issues 1 to 99 of the BMS from 1983 to 2008 were divided into three periods: 1983-1992 (85 articles), 1993-2000 (119 articles) and 2001-2008 (111 articles). These three corpora were individually analyzed with Calliope, an analysis by co-occurrence of key words in units of text, which produces thematic clusters. It also constructs strategic diagrams, which show the general thematic structure of each period with its axes of density (ties within a thematic cluster) and centrality (external ties between clusters). These diagrams, and those of the more dominant clusters, are presented along with their interpretation. BMS, Text analysis, Calliope, Co-Word Analysis, Bibliometrics.
Un quart de siècle de méthodologie – L’émergence du longitudinal, des réseaux et de la statistique textuelle
A Quarter of a Century of Methodology – The Emergence of Longitudinal Analysis, Social Networks and Textual Statistics: Using the titles, abstracts and key words (in French) of the 318 articles in the BMS, issues 1 to 99, the author analyzes this corpus with hierarchically ascending and descending classification programs and with factoral correspondence analysis. With the results of these three methods, he describes in detail five major thematic classes: Survey Situation, Longitudinal Analysis, Social Network Analysis, Theoretical Debate and Textual Analysis. Their interrelationship and evolution over time is also described. BMS, Classification Analysis, Factorial Correspondence Analysis, Thematic Evolution, Bibliometrics.
FIEVRE EUROPEENNE D’EVALUATION DE LA RECHERCHE DANS LES SCIENCES SOCIALES
European Research Evaluation Fever in the Social Sciences: After hitting Germany recently, the European fever of evaluating research in the social sciences has hit France with the classification of research journals by the official AERES. This dossier of the BMS consists of three contirbutions concerning bibliometrics, evaluation and the recent AERES classification. We reprint the “Introduction” of the research report byYves Gingras (UQAM), “Research Evaluation Fever – Bad Use of False Indicators”. We then publish a report by Laurence Coutrot (CMH-CNRS), “On the Recent Use of Bibliometric Indicators as Evaluation Tools in Scientific Research”, concerning the recent CNRS meeting on the question of bibliometrics. This is followed by a report prepared by François Briatta (ANCMSP). The French Agency for the Evaluation of Research and Higher Education (AERES) recently published journal rankings in the human and social sciences. This document compares AERES rankings for francophone journals in sociology/demography and political science with two former rankings, by Philippe Jeannin and by the authors of a CNRS-sponsored “Periodicals study” respectively. We first observe that the threefold classification used in each ranking resorts to different criteria, in such a way that no methodological consensus seems to emerge from the comparison. The rankings themselves are starkly heterogeneous: in political science, only 11 journals appear in the two compared rankings, only eight of them with the same rank. In sociology/demography, 26 “ranking agreements” and 17 divergent rankings are observed within the range of comparable journals. Bibliometrics, Research Evaluation, France, CNRS, AERES.
La fièvre de l’évaluation de la recherche – Du mauvais usage de faux indicateurs
Sur l’usage récent des indicateurs bibliométriques comme outil d’évaluation de la recherche scientifique
COMPARAISON INTER-CLASSEMENTS DES REVUES EN SOCIOLOGIE-DEMOGRAPHIE ET EN SCIENCE POLITIQUE
N. 101, January 2009
Extraction de réseaux égocentrés dans un (très grand) réseau social
Christophe Prieur, Alina Stoica, Zbigniew Smoreda
Extracting Ego-Centered Networks From Very Large Social Networks: This article presents a method at the intersection of macro and micro approaches in social networks. In examining ego-centered network, it proposes an approach to this type of study for very large networks. The calculation of several local indicators makes it possible to create aggregate statistics for each of the nodes of the network, which can then be seen as a corpus of ego-centered networks. An application is made on a network of telephone conversations between two million people, and a comparison is made with a classic ego-centered network study, based on an ethnographic research project. Social Networks, Ego-Centered Networks, Large Social Networks, Density, Communication Networks, Macro versus Micro, Qualitative Approaches [a corrected machine translation in English of this article is available on request from firstname.lastname@example.org].
A Behavioural Approach to Ranking Academic Departments by Prestige: The Case of Sociology in English-Speaking Canada
Academic prestige has, due to the 21st century culture of declining survey co-operation, become harder to measure accurately using reputational surveys. Even if the high nonresponse on a survey of professors were ignorable in terms of bias, the credibility of such a survey would be challenged. Within sociology, citation index counts are not as useful an alternative as in more consensual and article-based disciplines such as economics. Sociologists publish in an immense variety of outlets, with much of the most important work appearing in books. Reported herein is a behavioural approach to academic prestige, based on each department’s profile of doctoral orgins of staff members (termed a broad by fuzzy approach) and on dominance within the hiring exchange matrix (giving a narrow but clearer reading). The case study is departments of sociology within English-speaking Canada. Academic Prestige, Reputational Surveys, Departments of Sociology, Doctoral Origins, Hirings, Canada.
L’opinion et ses PUBLICS : Une approche pragmatiste de l’opinion publique
The Public Opinion and Its Audiences – A Pragmatic Approach to Public Opinion: That is the title of a recently published book by Mathieu Brugidou, with a preface by Nonna Mayer (Presses de Sciences Po, 2008), in the series “Sociétés en mouvement”. With the presentation of this book by its author, the BMS opens a file and a debate on public opinion research which will soon be followed by other contributions. Public Opinion, Research Methods, Public Debates, Public Controversies.
Applications de la théorie des jeux à l’éducation : pour quels types et niveaux d’éducation, quels modèles, quels résultats ?
Christelle Garrouste, Massimo Loi
Applications of Game Theory in Education – What Types and At What Levels: This paper examines the use of game theory in educational sciences. It describes the evolution of game theory from the defining axioms of Von Neumann and Morgenstern in 1944 to the present. After the Introduction and this description, the third part presents the methodology used to select research articles compared in this study. The final fourth part presents the results of this literature review. The study reveals a similar interest on the part of economists and educators for the application of game theory to issues of education. However, the extremely small number of published articles can be interpreted as the sign of a problem of adaptability of this method in this research discipline. The paper concludes with a discussion of the nature of this inadaptability. Game Theory, Educational Sciences.
N. 102, April 2009
THE AFS AND THE BMS: ANALYZING CONTEMPORARY FRENCH SOCIOLOGY
Karl M. van Meter
The French Sociological Association (Association Française de Sociologie or AFS) was founded in 2002 and has held congresses in 2004, 2006 and this month in 2009. The AFS Methods Section (Réseau Thématique 20 – Méthodes) used four text analysis methods – Prospéro, Calliope, Trideux, Alceste – applied to the abstracts of presentations at the 2004 congress. This resulted in three research articles published in the BMS in February 2005 and served as the basis for a book, Analyses textuelles en sociologie (PUR), in 2006. Since then, the BMS has published similar analyses of the 2006 AFS abstracts, and in this issue publishes the first two analyses of the 2009 AFS abstracts. French Sociology, AFS, Text Analysis, Prospéro, Calliope, Trideux, Alceste.
Permanences et transformations d’une institution : une analyse sociologique de trois congrès de l’AFS
Didier Torny, Patrick Trabal
Permanence and Transformation of an Institution – A Sociological Analysis of Three AFS Congresses: This analysis of summaries of presentations at the first three congresses of the Association Française de Sociologie (AFS) shows the stability of the 40 thematic sections founded in 2004, but a high turnover among presenters, 78% of whom have spoken in only one congress (1,777 out of 2,278 authors). Moreover, there is an increasing feminin presence (48% in 2004, 54.5% in 2009). At the third congress (Paris, 2009), the introduction of a general theme, “violence and society”, influenced the thematic sections’ activity, going from full acceptance in the concerned section, to the total lack of being taken into account. These changes allow us to demonstrate empirically the transformations generated by the subject when compared to previous conferences. French Sociology, Text Analysis, Prospéro, Institutional Demands.
FRENCH SOCIOLOGY AS SEEN THROUGH THE CO-WORD ANALYSIS OF AFS CONGRESS ABSTRACTS: 2004, 2006 & 2009
Mathilde de Saint Léger, Karl M. van Meter
We apply the Calliope text analysis software to the set of all abstracts of presentations at the 2009 French Association of Sociology (AFS) congress. We describe the data set’s overall structure and contents, and examine in detail the “mainstream” clusters associated with “violence”, “travail” (work), “femme” (woman), “salaire” (salary), “domaine scolaire” (educational domain), and “santé” (health). These results are compared with similar results for the analysis of 2004 AFS congress and 2006 AFS congress abstracts. Despite the 2009 dominance of the congress theme of “violence and society”, we find once again the major structuring role of the two dominant components associated with “travail” and “femme”. French Sociology, AFS, Calliope, Co-Word Analysis, Scientometrics.
N. 103, July 2009
Response Latency as an Indicator of Optimizing in Online Questionnaires
Mario Callegaro, Yongwei Yang, Dennison S. Bhola, Don A. Dillman, Tzu-Yun Chin
In answering questions, respondents carry out four mental steps: comprehension, retrieval, judgment, and answer selection. Those who carefully employ all steps are called optimizers, while those who invest less effort and skip steps are called satisficers. In this study, we compared the response latencies of optimizers and satisficers as they answered questions on a Web-administered personality assessment. Using spline regression and controlling for possible confounding variables, we provided evidence that optimizers invest more time than satisficers when answering questions in the first section of a questionnaire. This supports the perspective that deeper cognitive processing requires greater effort and takes more time. We also showed that, as a test progresses, both groups spend less time until they attain a maximum responding velocity. Response Time, Satisficing, Impression Management, Web-Administered Questionnaire, Personality Assessment.
Le lexique mobilisé lors de trois congrès de sociologues : Permanence et sensibilité à la conjoncture.
The Vocabulary Mobilized over Three Sociology Conferences – Permanence and Sensitiveness to the Current Situation: The Association Française de Sociologie (AFS) held its third congress in Paris in April 2009 and its 1,051 abstracts of presentations have been analyzed by two different text analysis programs in the BMS, n. 102. Here, the author used two others programs – Alceste and Multiple Correspondence Analysis – to analyze those same abstracts and to compare the results with those of the 2004 and 2006 AFS congresses analyzed by the same programs. Association Française de Sociologie (AFS), Textual Analysis, Alceste, Multiple Correspondence Analysis.
Méthodes statistiques globales et locales d’analyse d’un tableau de contingence par les tailles d’effet et leurs intervalles de confiance
Brice Lefèvre, Stéphane Champely
Global and Local Statistical Methods of Analysis for Contingency Tables by Size Effects and Confidence Intervals: This paper presents global and local analyses of a contingency table along the lines of Wilkinson et al. (1999) defining good statistical practices as using effect sizes, confidence intervals and plots instead of tests of significance. On one hand, the global relationship between the two qualitative variables is measured by Cramér’s V. A confidence interval may be calculated using the decentred chi-squared distribution by inverting the corresponding significance test. A bootstrap confidence interval is also proposed. On the other hand, Cibois’s PEM (percentage of maximum deviation from independence) is a local effect size that can be computed for every cell of the contingency table giving its specific distance from the independence hypothesis. A corresponding confidence interval is calculated by the resampling bootstrap technique. An original plot summarizing these different aspects for the entire contingency table is proposed. Contingency Tables, Cramér’s V, Size Effects, Confidence Intervals, Bootstrap, PEM.
Percentage of maximum deviation from independence (PEM): COMMENT ON Lefèvre & Champely’S “analyse d’un tableau de contingence” ARTICLE
Concerning the article by Lefèvre and Champely “Analysis of a Contingency Table”, Philippe Cibois explains the concept of the PEM, which was raised by the authors. He shows that the use of Chi-square avoids the Bootstrap procedure to assess the significance of a PEM. He suggested that instead of Cramér’s V, used by the authors, to use a generalized PEM whose algorithm he gives for maximizing the PEM in the case of an entire table. Cramér’s V, Percentage of Maximum Deviation from Independence (PEM), Confidence Intervals, Bootstrap, Contingency Tables.
N. 104, October 2009
GERMAN & FRENCH CONTEMPORARY SOCIOLOGY COMPARED: TEXT ANALYSIS OF CONGRESS ABSTRACTS
Karl M. van Meter, Mathilde de Saint Léger
We review results of the Calliope co-word analysis of all paper abstracts for the 2004, 2006 and 2009 congresses of the Association Française de Sociologie (AFS) and compare these results with those of a similar analysis of the 2004 Deutsche Gesellschaft für Soziologie (DGS) congress. German Sociology, French Sociology, Co-word Analysis, Calliope.
PESSIMISME ET OPTIMISME DANS LA MESURE DE LA LIAISON D’UN TABLEAU DE CONTINGENCE : RÉPONSE À PHILIPPE CIBOIS
Stéphane Champely, Brice Lefèvre
Pessimism and Optimism in the Measure of Association for Contingency Tables – A Reply to Philippe Cibois: In his commentary, Philippe Cibois suggests using the generalised PEM (percentage of maximum deviation from independence) instead of Cramér’s classical V to study the strength of the relationship in a contingency table. The PKM (percentage of maximum chi-squared of Cibois, 1990) seems more directly comparable to Cramér’s V2. A new algorithm is proposed to optimise the calculaltion of this PKM (and of the generalised PEM). It leads to results very similar to those obtained using Cramér’s statistic. Contingency Tables, Cramér’s V, PEM, PKM
Journée d’Etude du 21 Août 2009 sur la méthodologie « Alceste » Arguments des interventions
The 21 August 2009 Workshop on Alceste Methodology – The Presentations’ Arguments: Already presented in the BMS (n. 102: 90-92; n. 103: 102-103), the Alceste Summer School in Carcassonne on 19-21 August included a day devoted to methodology during which ten presentations were made and discussed by the participants. Alceste, Text Analysis, Discourse Analysis, Carcassonne.
AAPOR 2009 CONFERENCE “Public Choices in Changing Times”
Nick Moon, John Goyder
The sixth-fourth annual AAPOR Conference took place on 14-17 May 2009 in Hollywood, Florida, near Miami. With more than 400 papers presented and many important awards, it was major survey research meeting. American Association for Public Opinion Research, AAPOR, Survey Research Methods.
N. 105, January 2010
Comment traduire sous forme de probabilités les résultats d’une modélisation logit?
How to Translate a Logit Model into Probabilities: It is common in sociology to try to translate the results of a logit model into probabilities or percentages; in other words, into the language of cross tabulations. The purpose of this paper is to present and discuss three ways of performing this operation. This is done using numerical data taken from the FQP 2003 survey by INSEE. We present and discuss successively the method “of deviation from a reference situation”, the “experimental deviation” method and finally the method of “pure deviation”. Logit Models, Logistic Regression, Probabilities.
Les « choix » du sociologue avec ALCESTE – De la forme du corpus aux résultats obtenus
The Sociologist’s Choices with Alceste – From the Form of a Corpus to the Results Obtained: In this article we pose the question of the relationship between the form of a corpus of semi-structured interviews in sociology and the results obtained by the Alceste software. We show that “fixing” the vernacular or “including” the sociologist’s comments has an influence on the software’s results, and thus on the sociological analysis produced. Alceste, Textual Analysis.
Trideux software integrates Jérôme Deauvieau’s “How to translate a logit model into probabilities”
In his article “How to Translate a Logit Model into Probabilities?”, Jérôme Deauvieau shows that presentations as probabilities or percentages can cause problems because the results depend upon the choice of modalities that serve as a reference category. Deauvieau offers two solutions to this problem: the adjustment under experimental deviation and the adjustment as pure deviation. Both methods are now implemented in the Trideux software and an example is proposed. Logit Models, Logistic Regression, Probabilities, Trideux.
Report on the European Survey Research Association Conference 2009 in Warsaw
Johannes van der Zouwen
After the conferences in Barcelona (2005) and Prague (2007), this conference saw the presentation of some 360 papers on a dozen themes. There were also three keynote speeches and two workshops on issues of survey research.
N. 106, April 2010
Do we necessarily need longitudinal data to infer causal relations?
Guillaume Wunsch, Federica Russo, Michel Mouchart
It is generally admitted that causes precede their effects in time. This usually justifies the preference for longitudinal studies over cross-sectional ones, because the former allow the modelling of the dynamic process generating the outcome, while the latter cannot. Supporters of the longitudinal view make two interrelated claims: (i) causal inference requires following the same individuals over time, and (ii) no causal inference can be drawn from cross-sectional data. In this paper, we challenge this view and offer counter-arguments to both claims. We also argue that the possibility of establishing causal relations does not so much depend upon whether we use longitudinal or cross-sectional data, but rather on whether or not the modelling strategy is structural.
Tels parents, tels enfants ? L’utilisation de données de seconde main dans l’étude des influences politiques parentales en Suisse.
Like Parents, Like Children? Using Secondary Data in the Study of Parental Political Influence in Switzerland: This text is devoted to the intergenerational transmission of left-right ideological orientation. Using data from the Swiss Household Panel (www.swisspanel.ch), collected both directly and through intermediaries, it raises the question of the validity of second-hand information, which is discussed on several points. The article also shows that there is a strong long-term coherence between the ideological orientation of parents and their children. It also highlights the fact that the link is not the same for boys as for girls, and it varies between generations. Finally, it shows that strong upward social mobility, compared to the original environment, facilitates emancipation from parental ideology, social inertia being greater for downward and horizontal mobility.
Analyse des réseaux sociaux : une « French touch » ? Retour sur l’école thématique CNRS « Réseaux sociaux : enjeux, méthodes, perspectives », Cargèse (Corse), 15-20 septembre 2008
Claire Bidart, Alain Degenne, Michel Grossetti, Claire Lemercier
Social Network Analysis – A French Touch? Looking Back on the CNRS Summer School ‘Social Networks – Issues, Methods and Outlooks’, Cargèse (Corsica), 15-20 September 2008: This summer school was intended to present social network analysis and its recent developments to a French-speaking public. It was also the opportunity to take stock of work in this area by French and, more generally, French-speaking researchers. The article covers the presentations that were made on this occasion and attempts to identify what may be the specific nature of this work, which seems to be associated with a particular attention to theoretical and epistemological reflexivity.
“An ethnographic seduction”: how qualitative research and Agent-based Models can Benefit Each Other.
Paola Tubaro, Antonio A. Casilli
We provide a general analytical framework for empirically informed agent-based simulations. We believe that this methodology provides present-day agent-based models with a sound and proper insight as to the behavior of social agents – an insight that statistical data often fall short of providing at least at a micro level and for hidden and sensitive populations. In the other direction, simulations can provide qualitative researchers in sociology, anthropology and other fields with valuable tools for: (a) testing the consistency and pushing the boundaries, of specific theoretical frameworks; (b) replicating and generalizing results; (c) providing a platform for cross-disciplinary validation of results.
N. 107, July 2010
Representativeness in online surveys through stratified samples
Jörg Blasius, Maurice Brandt
While nationwide face-to-face samples are considered representative for the entire population, online samples are regarded as biased, especially in terms of age, sex and education. To consider this bias, the data can be weighted to receive a representative sample. In case of online surveys, elderly women with a low level of education receive a very high weight and young men with high education a very low one. Instead of weighting the data, we draw a stratified sample from over 20,000 participants of an online panel. Thereby, elderly women with relatively low formal education had a very high likelihood of entering the sample, young men with high education a very low one. To obtain a stratified sample by sex × age × education without coverage error, the sample is restricted to an age of up to 49 years. To compare these data with representative face-to-face data, we used a set of questions from the German Social Survey 2002. We compare the results of the two studies on the basis of more than 1,100 cases, each of which are equivalent in terms of age, sex and education. It is shown that stratified online samples are not representative for the entire population.
Les « choix » du sociologue avec ALCESTE : Du paramétrage des Unités de Contexte aux résultats obtenus
The “Choice” of the Sociologist with Alceste – From Determining Units of Context to the Results Obtained: In this article, we pose the question of the relationship between cutting up a corpus of semi-structured interviews in sociology by determing the units of content and the results obtained when the Alceste software analyzes the units of context. We show that the way the text is cut up has an influence on the results and thus the analysis produced.
Measuring and comparing levels of education; Methodological problems in the classification of educational levels in the European Social Surveys and the French Labor Force Surveys
Commonly used in social sciences to understand representations, behaviors or major social changes, the education level of individuals, its classification and its possible measures are too rarely the subject of research methodology. We show here, based on a comparison of different versions of the European Social Survey with the Labor Force Surveys carried out by INSEE in France, the importance of procedures of reporting, classification, aggregation and harmonization of education in comparable categories, from the national to the cross-national level (ISCED). The most common measures focus on the comparison of the levels of education. Taking into account the orientation of knowledge, including the difference between general, vocational and technical education thus significantly improves the measurement quality. This article argues for a development of sociological studies on these issues.
Pour une sociologie de l’activité au travail des dirigeants de grandes entreprises
For a Sociology of Work Activity of Managers of Large Businesses: This paper proposes a methodology for an integrated analysis of top executives practices: the process of professional socialization; how s/he handles her/his business and pursues financial goals; the process of personal enrichment during her/his career; and the way s/he legitimizes the social position s/he occupies. This research strategy combines biographical sources, grey literature, ethnographic observations inside companies and secondary analysis of accounting data and public relations data. We conclude on the irony of such a research procedure caught between the ideal of a scientific analysis of practices and the social imperatives of the staging of power.
N. 108, October 2010
NOTE SUR LES METHODES QUANTITATIVES EN SOCIOLOGIE DES VALEURS
Note on Quantitative Methods in the Sociology of Values: The sociology of values, once confined to a mere extension of “moral philosophy”, has been profoundly renewed by the application of empirical methods of research. Developed in the United States in the 1970s by large-scale statistical surveys, it has also resulted in France, over the past twenty years, in regular investigations. However, these surveys suffer from a number of problems: the lack of thought concerning methods, the lack of a definition of the concept of values, the restriction of conscious awareness, authors not taking into account consideration of pertinence, and the absence of contextualization. We will discuss in conclusion the effects of these methodological problems in terms of results, and we present some ideas for a more rigorous empirical study of values.
Moniteur sportif et sociologue : récit d’une observation participante clandestine
Bastien Soulé, Cédric Richet
Sports Instructor and Sociologist – A Covert Participant Observation: This paper describes the change of status and viewpoint inherent in the shift from a sports instructor job to the implementation of a participant observation. The latter was undertaken to grasp the stakes and difficulties of a sports facility delegated management. Beside the results of this investigation, we also intend to relate a peculiar process, including the combination of multiple and varying roles, as well as a dynamic relation to the fieldwork and its participants. As the observation remained concealed, this role duplicity had many consequences: beyond personal and psychological ones, the effects are also professional, relational, methodological, heuristic, and ethical.
The Mayors of BourGES FROM 1474 TO 1723
This article shows how genealogical research, by reconstructing the interwoven fabric of families (parentage, siblings and re-alliances), can stimulate the diachronic analysis of networks. It covers the patrician families who ruled Bourges though mayors and aldermen from 1474 for 250 years.
Effects (and defects) of the telephone survey in polling research: Are we abusing of the telephone survey?
Vidal Díaz de Rada
The objective of this paper is to evaluate the adaptation of telephone surveys to electoral forecasting to see if they show substantial improvements when compared to face-to-face surveys. To do so, two similar samples were used, one surveyed with face-to-face interviews and the other using telephone interviews. The objective is to compare the results in answering three questions concerning: 1) the important differences in results of the two methods which could notably affect their extrapolation; 2) telephone surveys’ higher evaluation of the intention to vote and more sincere responses (closer to the electoral results); 3) the important differences in the evaluation of the economic and political situation, definition of the most important social problems, and nationalist and regionalist sentiments.
N. 109, January 2011
Matrix assisted structural hypothesis construction
Erik H Cohen, Charles Tresser
This paper outlines a novel process for designing a structural hypothesis. Relevant factors in a study are identified and micro-hypotheses are enunciated in a matrix format. Levels of similarity are assigned to each pair of factors. Multi-dimensional scaling techniques are used to graphically portray the matrix. The example of “Quality of Life” (QoL) is used to illustrate the proposed matrix approach to generation of a structural hypothesis. A Smallest Space Analysis (SSA) of the hypothetical Quality of Life matrix shows a structure in which four polar regions surround a central region representing “general” QoL. This clear and logical structure of the hypothesized correlation matrix is a strong first confirmation that the approach allows for a detailed, systematic and theoretically based mode of reasoning and didactics. Additionally, the structure found here corresponds to a previous SSA analysis based on empirical Quality of Life data.
Analyse textuelle et analyse de RÉSEAUX: Exemple du traitement d’une base de données bibliographiques à l’aide des logiciels Alceste et Pajek
Monique Dalud-Vincent, Romuald Normand
In this paper, we demonstrate how textual analysis and network analysis are complementary in displaying various groups from a bibliographic data bank, then in building the corresponding networks of co-publication. The Alceste software and Pajek, used on the same “set” of data including 5,000 references, allowed us to objectify the existence of various epistemic communities, as well as the links within and between communities related to the issue of “measurement” in education.
To Participate or Not to Participate: Decision Processes Related to Survey Non-response
This paper proposes a different methodological approach to understanding the determinants of survey participation. Data was collected by means of a Web panel survey conducted with a population of university students (N = 330). First, we focused on theory-driven measures of consequences of participation: costs of participating, survey situation, generalized attitudes about surveys, and past survey-taking behaviours using a paper-and-pencil questionnaire. A subset of this sample was subsequently invited, twice, to participate in a Web panel survey. This comparison allows the theory-driven prediction of unit non-response. We found that time resources and costs (fear of data misuse) are the primary determinants of participation in the first wave; and salience of topic and costs (burdensome survey experience in the first wave) are the major predictors of participation in the second wave. These results support the theory of rational action to explain unit non-response and match well with the existing empirical evidence.
DATA COLLECTION IN A PROBABILITY-BASED INTERNET PANEL: HOW THE LISS PANEL WAS BUILD AND HOW IT CAN BE USED
In this paper, we introduce a methodology to set up a panel that combines the scientific standards for a longitudinal panel with the advantages of Internet interviewing as a method of data collection. A panel that is intended for scientific research demands a probability sample, covering the whole population of interest and thus including people without Internet access and people who do not actively volunteer to answer questions. We will show how such a panel can be built and maintained, and how researchers can use it for their own, cost free data collection.
N. 110, April 2011
Réseaux personnels, réseaux sociaux
Ainhoa de Federico de la Rúa, Catherine Comet
Guest editors’ presentation of this thematic issue
Studying Relational Chains from Narrative Material
Michel Grossetti, Jean-François Barthe, Nathalie Chauvac
Relational chains are a poorly understood approach to social networks, whose methods of analysis are usually classified according to the personal networks / complete networks dichotomy. Rather than mapping out networks statically, the study of relational chains reconstructs the concrete mobilisations of social relations in processes for accessing resources or networking with people. The method we present reconstructs chains from a narrative material established on the basis of crossed interviews. This method is especially well suited to capturing phenomena of embeddedness of economic activities in social networks. We introduce it in this article based on two surveys, one on science-industry relations and the other on enterprise creation.
Dynamique des réseaux sociaux et trajectoire d’entreprises informelles à Bobo-Dioulasso (Burkina Faso)
Jean-Philippe Berrou, Claire Gondard-Delcroix
Dynamics of Social Networks and Trajectories of Informal Businesses in Bobo-Dioulasso (Burkina Faso): This paper analyzes the evolution / transformation of social relations and networks for access to resources during the career of micro entrepreneurs in an informal urban economy in Sub-Saharan Africa. From a methodological point of view, the analysis of life histories from micro and small entrepreneurs in Bobo-Dioulasso (Burkina Faso) was done by combining different forms of qualitative and quantitative analysis (Grossetti and Barthe, 2008). This novel method can identify the characteristics of the joint evolution of the business activity and the entrepreneur’s social network, and then provide a comprehensive analysis of the processes underlying these dynamics. It appears that the gradual professionalization and institutionalization of the social network of the entrepreneur are forms of securing access to resources promoting the development of the economic activity.
Anatomy of a Fraud: Trust and Social Networks
How do social networks support trust in the case of fraud? The Santa case is a “strategic case” given its scope and its duration. This case of fraud has marked specificities, insofar as it concerns primarily expatriate military. Two officers canvassed these members of the military to propose so-called high-return investments to them. I am more particularly interested in the leverage used by these salesmen; this is, the mutual relations they used as referrals, and in the nature of the linkages between the levers and the targets according to target’s status (civilians, non-commissioned officers, or officers). The analyses are based on a questionnaire-based survey of 117 victims of the fraud.
Introduction du générateur de liens sociaux par contextes (GLSC) dans une approche mixte : Etude sur l’hétérogénéité dans les liens de collaboration des chercheurs en biotechnologie et en sciences de la vie
Claude Julie Bourque
Introduction to the Social Context Name Generator (GLSC) in a Mixed Methods Approach – Study of the Heterogeneity of Researcher Cooperation Links in Biotechnology and Life Sciences: The field of biotechnology research is characterized there by social dynamics more heterogeneous than the others? To answer this question, we built a tool to produce comparable empirical data on collaboration networks in three practice settings. The generator of social ties by contexts (GLSC) was administered to 735 researchers in Quebec life sciences, natural sciences and engineering in the fall of 2009. The results indicate that for most researchers, whatever their discipline, the morphology of social bonds is significantly different in the context of financing. Moreover, the results refute the hypothesis that collaborative practices are significantly more heterogeneous in research in biotechnology and life sciences.
Two different methods for measuring personal networks compared
This article compares two different tools for analyzing personal networks: the FNG (free name generator) by which respondents freely list the names of people with whom they more often interact, and the GNG (guided name generator) by which respondents list the names of people with whom they more often interact, following a standardized 20-item list of types of exchange. The two techniques are tested together in an integrated questionnaire, the data of 37 personal networks analyzed, a typological-factorial sampling design is discussed, and a standard procedure is suggested and tested.
N. 111, July 2011
Effects of prepaid monetary incentives on mail survey response rates and on self-reporting about delinquency – Empirical findings
Rolf Becker, Guido Mehlkop
We carried out a methodological experiment to investigate whether mail survey response rates can be influenced by monetary incentives. Prepaid monetary incentives are expected to elicit higher response rates, as predicted by social exchange and subjectively expected utility theory. This is true especially in a survey on the respondents’ delinquent behavior, since prepaid incentives are likely to strengthen the respondents’ trust towards the researcher. For surveys in which money is promised once the questionnaire is returned, we expect the response rates to be lower. The empirical findings clearly support our assumptions, even if the social mechanisms of reciprocity which underlie response behavior cannot be observed directly. Our analysis also shows that monetary incentives evoke neither desirability, nor sponsorship effects which would cause overreporting of delinquency. Finally, our results show no systematic variation regarding socio-structural characteristics and correlates of delinquent behavior.
N. 111, July 2011
Measuring Race and Hispanic Origin: Cognitive Test Findings Searching for “Truth”
Jennifer Hunter Childs, Rodney Terry, Nathan Jurgenson
This research describes an attempt by USCensus Bureau staff to create a “gold standard” assessment of the “truth” of self-identified race in order to evaluate the performance of an experimental panel of race and Hispanic origin questions. This gold standard is achieved by asking about race in three ways: (1) an open-ended question that allows the respondent to self-identify with any races or Hispanic origins; (2) a series of yes/no questions aimed at measuring identification with the US government’s race and Hispanic origin categories;2 and (3) a summary measure which attempts to gather the respondent’s usual or typical report to race and Hispanic origin questions. We argue that while no single measure taken alone captures the truth of race, all the three measures, taken together, do provide a robust portrait of self-identified race and Hispanic origin for nearly all respondents in our sample.
Inferring Causality through Counterfactuals in Observational Studies – Some epistemological issues
Federica Russo, Guillaume Wunsch, Michel Mouchart
This paper contributes to the debate on the virtues and vices of counterfactuals as a basis for causal inference. The goal is to put the counterfactual approach in an epistemological perspective. We discuss a number of issues, ranging from its non-observable basis to the parallelisms drawn between the counterfactual approach in statistics and in philosophy. We argue that the question is not to oppose or to endorse the counterfactual approach as a matter of principle, but to decide what modelling framework is best to adopt depending on the research context.
Rapport de thèse de doctorat en Science politique (Paris Dauphine) de Philippe Blanchard : Les médias et l’agenda de l’électronucléaire en France. 1970-2000
Report on the Political Science Doctoral Thesis (Paris Dauphine) of Philippe Blanchard: The Media and the Agenda of Nuclear Power in France, 1970-2000: This book review focuses on the thesis of Philippe Blanchard, The Media and the Agenda of Nuclear Power in France 1970-2000. In this thesis, the agenda has been studied in two complementary ways: one based on modelling, the other on sociology. Their combined use helps to understand the construction of the nuclear issue as a long-term public controversy and policy. This note highlights the important methodological issues addressed by this thesis.
Pourquoi traduire sous forme de probabilités les résultats d’une modélisation logit ? Réaction à J. Deauvieau (BMS, 2010)
Première réponse à Marion Seltz
Why Translate into Probabilities the Results of Logit Models?: Jerome Deauvieau suggested in the BMS (2010, 105: 5-23) different ways of translating the results of a logit model into the language of two-way tables. This note shows the limits and drawbacks of such operations, which are lose the notion of “all things equal otherwise”, or the significance of the result.
N. 112, October 2011
XML Content Warehousing: Improving Sociological Studies of Mailing Lists and Web Data
Benjamin Nguyen, Antoine Vion, François-Xavier Dudouet, Dario Colazzo, Ioana Manolescu, Pierre Senellart
In this paper, we present the guidelines for an XML-based approach for the sociological study of Web data such as the analysis of mailing lists or databases available online. The use of an XML warehouse is a flexible solution for storing and processing this kind of data. We propose an implemented solution and show possible applications with our case study of profiles of experts involved in W3C standard-setting activity. We illustrate the sociological use of semi-structured databases by presenting our XML Schema for mailing-list warehousing. An XML Schema allows many adjunctions or crossings of data sources, without modifying existing data sets, while allowing possible structural evolution. We also show that the existence of hidden data implies increased complexity for traditional SQL users. XML content warehousing allows altogether exhaustive warehousing and recursive queries through contents, with far less dependence on the initial storage. We finally present the possibility of exporting the data stored in the warehouse to commonly-used advanced software devoted to sociological analysis.
EST-IL POSSIBLE ET SOUHAITABLE DE TRADUIRE SOUS FORME DE PROBABILITES UN COEFFICIENT LOGIT ? Réponse aux remarques formulées par Marion Selz à propos de mon article paru dans le BMS en 2010
Is It Possible and Desirable to Translate a Logit Coefficient into Probabilities – Reply to Comments by Marion Selz on My BMS 2010 Article: Marion Selz responded to my BMS article on the translation of a logit coefficient into probabilities. I propose here to respond to her objections by showing that the translation is not a problem when one thoroughly understands what takes place. We first show the relationship between linear models and logistic models, and then the relationship between the translation of a logit coefficient and standardization in demography, and finally the sociological interest of such a translation in the general framework of logit modeling.
L’expérience de la mobilité sociale : plaidoyer pour une approche par le discours
The Experience of Social Mobility – A Plea for a Discourse Approach: This article discusses the theoretical and methodological debates concerning the experience and consequences of upward social mobility between generations. The methods employed by researchers working on this subject are diverse, and the results they arrive at are sometimes contradictory. This article proposes, firstly, to give an overview of the different traditions of studying the experience of social mobility in order to identify the conditions for potential “cumulative” knowledge. In a second step, we argue that it is mainly through a study of the narratives of mobility that one can, simultaneously and comprehensively, grasp the ambivalence and multiplicity of effects that produce upward social mobility.
Matrix assisted structural hypothesis construction: Further explorations
Erik H Cohen, Charles Tresser
In January 2011 (issue 109), this journal published our article on a new process for developing a structural hypothesis, using a matrix approach and multi-dimensional data analysis techniques (Cohen and Tresser, 2011). The present article continues to develop the matrix assisted hypothesis construction method (matrix method for short) and offers directions for research utilizing this approach.
Italian Sociologists’ approach to qualitative interviews
It is common to come across interviews, both transcribed and recorded, that were conducted according to remarkably different strategies that seem to contrast with the recommendations and indications of a large part of the literature on this subject. To understand the reasons behind these different strategies, I will present the outcomes of a study carried out among Italian sociologists who base their research on qualitative interviews. The aim of my study is to examine the attitude of a group of researchers and interviewers towards the interviewees in the field of qualitative research.
N. 113, January 2012
UNDERSTANDING SEQUENCING IN SOCIAL NETWORK COMMUNICATIONS
Guang Ying Mo, Barry Wellman
Sequencing is an indispensable decision-making process during information flow. This paper proposes the conceptualization of sequencing to understand how and why information senders prioritize some network members when they communicate with others. We examine the usefulness of this conceptualization with the data collected from GRAND, a scholarly network. The concept of sequencing enables researchers to explore the decision-making process that occurs prior to information flow and link individuals’ behavior to the social context at a higher level.
SUPPLEMENTAL APPLICATION OF CONFIGURAL FREQUENCY ANALYSIS TO FREE SORTING DATA
This paper analyses the application of Con?gural Frequency Analysis (CFA) to free sorting data. It is intended to start a discussion about this simple supplementary analysis tool for free sorting data. Common features of sorting data and their relevance for CFA are discussed, in its approach to using both descriptive and con?rmatory data analysis. Sample size requirements of card sorting are also revisited for descriptive analysis. The sample required to some extent depends on the speci?ed domain. The average number of different con?gurations encountered depends on the number of sortings collected, the number of items included in the domain and the qualities of the domain. A simple exact significance test is proposed for hypothesis testing of configural frequencies in free fuzzy sorting data.
LE CHOIX DE LA LANGUE DANS LE LOGICIEL ALCESTE : EXEMPLE DU TRAITEMENT D’UNE BASE DE DONNEES BIBLIOGRAPHIQUES
Monique Dalud-Vincent, Romuald Normand
“The Choice of Language for the Alceste Software: Example of the Analysis of a Bibliographic Data Base” Using Alceste requires a priori choosing a language. This choice is crucial since the analysis is based on the use of a dictionary of the language used, but it is problematic when the corpus is a bibliography of references in English, French and German. In this paper, we show, for a largely bilingual (English and French) bibliography, with 5,000 references, the variations in the software results when you change the settings.
Transfer of variables between different data sets, or Taking “previous research” seriously
Given two methodologically similar surveys, a question not asked in one survey could be seen as a special case of the missing data problem. Hence, the transfer of data across data sets (“statistical matching” or “data fusion”) could be achieved applying the procedures for Bayesian multiple imputation of missing values. To tackle the problem of conditional independence, which this approach creates, a simulated data set could serve as the “third data set” that conveys information about the relationship between variables not commonly observed. This paper presents a model for transferring data between different data sets based on multiple imputation (MI) approach. The results show that statistical matching based on MI principles can be a useful research tool. The entire enterprise is interpreted in the sense of taking the “previous research” into account seriously.
NONRESPONSE VERSUS MEASUREMENT ERROR: ARE RELUCTANT RESPONDENTS WORTH PURSUING?
Joop J. Hox, Edith D. de Leeuw, Ann Chang
To increase response rates, survey researchers intensify their efforts to bring sampled persons into the respondent pool. The question is whether or not “reluctant” survey respondents provide answers of lower quality than “eager” respondents. We define eager respondents as persons who respond to the first round of a mail survey, and reluctant respondents as persons who respond in later rounds. We used a multitrait-multimethod (MTMM) design, which allows statistical separation of substantive or trait variance, method variance, and error variance. The results show that the measurement structure does not differ between eager and reluctant respondents. There was also no systematic difference in the reliability and validity estimates for both groups. Data Quality, Measurement Error, Multitrait-MultiMethod (MTMM), Nonresponse Bias, Reluctant Respondents, Total Survey Error
N. 114, April 2012
THE INTERPRETATION OF STATISTICS IN SOCIOLOGY
The interpretation of statistics in sociology, in particular in data analysis, faces the problem of realistic interpretations or not of the ideal types that the researcher can identify. With an investigation of school willingness of children from immigrant families, we show how one can hesitate in choosing a realistic interpretation or a constructivist interpretation of the results. We broaden the discussion by a fictitious dialogue between supporters of each camp. Data Analysis, Realism, Constructivism, Spontaneous Philosophy of Scientists
COUNTING AND MEASURING ONLINE: THE QUALITY OF INTERNET SURVEYS
Edith D. de Leeuw**
We still cannot look inside people’s heads. To measure emotions, opinions, evaluations, associations, and intentions, we have to ask people about these phenomena. Our data collection has changed over time, and many different methods are available. At present, Internet surveys are widely used in market research and are becoming an important data collection tool in universities. The main advantages of Internet surveys are the speed and cost reduction in comparison to more traditional data collection methods, the positive effect of self-administration, such as more privacy, use of more complex questionnaire routing, and in general a greater interactivity. Disadvantages are problems of Internet coverage of the general population, especially under-coverage of certain subgroups such as the elderly and the lower educated, and non-response. This contribution provides a review of existing studies comparing online surveys with more traditional data collection methods. Web Surveys, Coverage, Non-response, Data Quality, Survey Design
LES ELECTIONS PRESIDENTIELLES DE 2002 : LES SONDAGES S’ETAIENT-ILS TROMPES ?
French Presidential Elections of 2002 – Were the Polls Wrong? This note intends to show that on the eve of the first round of the 2002 presidential elections in France, polls published in the media had properly informed the public about the evolution of public opinion (slight decrease for Jospin, increase of Le Pen), and about indecision on the part of the electorate (which decides the day before or on election day). It is therefore likely that the published results were accurate and it is the predictions made on the basis of these results that were clearly erroneous. Public Opinion Polls, Voting Intentions, April 2002, France
TYPOLOGIES DE PARCOURS ET DYNAMIQUE LONGITUDINALE
Patrick Rousset , Jean-François Giret, Yvette Grelet
Life Course Typologies et Longitudinal Dynamics: In this paper, we propose a method for developing typologies of individual itineraries that takes into account the dynamics of longitudinal timelines. It applies when these timelines are developed from calendars. The originality of this method lies both in the method of calculating the distance between trajectories, and the classification procedure based on Kohonen self-organizing maps. The main property of the metric is to take into account the proximity between different states and its evolution over time. We propose an application in the analysis of career paths using Céreq data for monthly monitoring over seven years for youths who left the education system in 1998. Our results highlight the importance of temporal dynamics in the construction of timelines of entry into the workforce. Finally, we examine the role and contributions of this method by comparing it to the most commonly used methods for constructing timeline typologies. Classification, Typologies, Life Course, Entering the Workforce
METHODOLOGICAL RESEARCH ON “SENSITIVE” TOPICS: A DECADE REVIEW
Yeon-Ok Lee, Raymond M. Lee
Drawing on van Meter’s (2000) article, examining the methodological literature relating to research on “sensitive” topics, we analyse the corresponding journal literature in the decade following. We present data on authorship patterns, identify thematic continuities and discontinuities, and draw attention to gaps in the existing literature. Sensitive Topics, Methodological Trends, Methods Literature
N. 115, July 2012
Social Media Censorship in Times of Political Unrest – A Social Simulation Experiment with the UK Riots
Antonio A. Casilli, Paola Tubaro
Following the 2011 wave of political unrest, extending from the Arab Spring to the UK riots, the formation of a large consensus around Internet censorship is underway. The present paper adopts a social simulation approach to show that the decision to “regulate”, filter or censor social media in situations of unrest changes the pattern of civil protest and ultimately results in higher levels of violence. Building on Epstein’s (2002) agent-based model, several alternative scenarios are generated. The systemic optimum, represented by complete absence of censorship, not only corresponds to lower levels of violence over time, but allows for significant periods of social peace after each outburst. UK Civil Violence, Internet Censorship, Agent-based Modelling, Social Simulation, Social Media, Flash Mobs
Mesure statistique, mesure politique – Le cas des personnes sans-abri
Statistical Measure, Policy Measure – The Case of Rough Sleepers: Since the 1980s, in the US and Europe, various statistical methods have been developed to investigate people without housing, including those who sleep in places “not meant for habitation”, the “rough sleepers”. Street counts can be done locally by local authorities or by organizations serving homeless people to assess policies toward the rough sleepers or seeking to attract the attention of the general public and authorities. Among national statistical agencies, national street counts of rough sleepers are carried out and highlighted by a civic type of argument. A third type of surveys involves a vision of the rough sleepers as occupying a situation which may be only transitory, and involves public action to improve the situation of homeless people, but also to prevent it. By drawing on the work of Alain Desrosières, we will attempt to reconsider these statistical tools, these ways of thinking about society, and these policies of acting on society, at least for the fraction of interest here. Homeless, Unsheltered Homeless, Rough Sleepers, Statistical Survey, Hard to Reach Populations, Weight Sharing Method
Calendar Interviewing and the Use of Landmark Events – Implications for Cross-cultural Surveys
Tina Glasner, Wander van der Vaart, Robert F. Belli
This paper discusses potential methodological issues in the design and implementation of calendar recall aids such as the Life History Calendar for cross-cultural surveys. More specifically, it aims to provide insights into how the use of landmark events in calendar interviewing may be influenced by cross-cultural variability. As an example, we compare the landmark events reported by Dutch and American respondents in two studies in which calendar recall aids were used. The study discusses differences that were found between the two countries in the numbers and types of reported landmark events, as well as in the temporal distribution of those events. The outcomes suggest that it is important for researchers to examine how landmark events in calendar instruments translate in diverse cultural contexts. Retrospective questions, Life history calendar, Cross-cultural surveys, Autobiographical memory
Traiter des « masses » de données prosopographiques par la numérisation d’annuaires – Espoirs et vertiges
Sylvain Laurens, Francis Marchan
Treating “Masses” of Prosopographical Data by Scanning Directories – Hopes and Disorientation: This note aims to provide an update on the progress made in optical character recognition (OCR) and the contribution of these techniques to the creation of prosopographical data bases in social sciences. With the example of a European investigation of European business associations, it highlights the progress made possible by OCR with the analysis of several biographical directories identifying groups of business interests. Based on this example, it is hypothesized that the development of digital technologies allow the creation of corpuses of data much larger than in the past in the framework of quantitative inquiries conducted by smaller teams. However, this article also highlights the fact that this extension of corpuses – made possible by scanning – raises new problems of method, starting with the increased time devoted to the standardization of digital data. Data entry, Optical character recognition (OCR), Prosopography, Recoding data, EU lobbies
Ségrégation et fragmentation socio-spatiale – L’épreuve de la mesure
Segregation and Socio-spatial Fragmentation – The Ordeal of Measurement: This note is a reflection on the meaning and scope of segregation and socio-spatial fragmentation indicators and, beyond, on the role of statistics in sociological reasoning. It shows that if these indicators can be used to test theories of segregation or fragmentation, they must nevertheless themselves be put to the test of sociological reasoning. The choice of statistical indicators is actually not a purely technical exercise but is strongly linked to the meaning we give to the phenomenon. Their construction and interpretation always depend on how it was defined previously, how its causes have been pictured, and its consequences have been anticipated. In this sense, statistical indicators can only be a step in a particular argument. Segmentation, fragmentation, statistical indicators, sociological reasoning, measurement, falsifiability
Human or Machine Coding of Open-ended Questions
It is argued in studies in which open-ended questions are used that the choice is not between human or machine coding. It is the position the investigator takes with regard to the coding process that is relevant: from whose perspective is the coding performed? This choice determines whether human or machine coding can be performed. The results one obtains when these approaches are used can be very different. Text analysis, open-ended questions, coding, instrumental view, representational view.
N. 116, October 2012
Harpoon or Bait? A Comparison of Various Metrics in Fishing for Sequence Patterns
Nicolas Robette, Xavier Bry
The use of sequence analysis in the social sciences has significantly increased during the last decade or two. Sequence analysis explores and describes trajectories and “fishes for patterns” (Abbott, 2000). Many dissimilarity metrics exist in various domains (bioinformatics, data mining, etc.); therefore a crucial and pervasive issue in papers using sequence analysis is robustness. To what extent do the various techniques lead to consistent and converging results? What kinds of patterns are more easily fished out by each of the metrics? Here we propose a systematic comparison of about ten metrics that have been used in the social science literature, based on the examination of dissimilarity matrices computed from a simulated sequence data set including various patterns that sociologists can try to identify. This should help scholars in picking the method best suited to their data design and inquiry objectives. Sequence Analysis, Optimal Matching, Geometric Data Analysis, Comparison, Simulation
Une approche « hétéro-statistique » et graphique des masses de données d’enquête – le logiciel PointG
Stéphane Champely, Brice Lefèvre, Julie Thomas, Sylvain Ferez
A “Hetero-statistical” and Graphical Approach for Massive Survey Data – The PointG Software. In sociology, one is often confronted with the problem of graphic presentations based on processing massive quantities of data from questionnaire surveys. Moreover, potential users can be of low statistical expertise, thus requiring relatively easy to use software such as PointG with drop-down menus. It is aimed at facing the difficulties due to the mass of collected quantitative data, both at the strategic, tactical and operational levels of statistical processing. It offers conventional ways of analysis going from univariate to multivariate data, but also automates treatments on groups of variables and calculates the size of overall and local effects. Apart from the graphic aspects, its distinctive “hetero-statistic” approach permits the user not to worry about the nature of the variables in bivariate analyses and during regression and factor analysis. PointG was developed within the powerful distribution free R programming environment. Sociology, Survey Questionnaire Data, Statistics, R Software, Graphics.
Impact of the Mode of Data Collection on the Quality of Answers to Survey Questions Depending on Respondent Characteristics
The Internet is used more and more to conduct surveys. However, moving from traditional modes of data collection to the Internet may threaten the comparability of the data if the mode has an impact on the way respondents answer. In previous research, Revilla and Saris (2012) find similar average quality (defined as the product of reliability and validity) for several survey questions when asked in a face-to-face interview and when asked online. But does this mean that the mode of data collection does not have an impact on the quality? Or may it be that for some respondents the quality is higher for Web surveys whereas for others it is lower, such that on an average the quality for the complete sample is similar? Comparing the quality for different groups of respondents in a face-to-face and in a Web survey, no significant impact of the background characteristics, the mode and the interaction between them on the quality is found. Web and face-to-face Surveys, Survey Answer Quality, MTMM Approach, Data Collection Mode.
Paul Felix Lazarsfeld His Methodological Inspirations and Networking Activities in the Field of Social Research, Prague, 25 – 27 September 2011
Eva Balazova, Jan Marsalek
This short text informs about the international conference dedicated to the life and work of Paul F. Lazarsfeld which was held in Prague, September 2011. It organizes the conference presentations into three blocks: 1) Critical analyses of social research methods that Lazarsfeld propagated; 2) Commemorations of his personality and intellectual style; 3) Studies of Lazarsfeld’s posterity and of his presence in current social research. P. F. Lazarsfeld, Methods of Social Research, Inspirations.
Who is Responsible for the Disappearance of Social Classes?
Who made social classes disappear? This was the issue raised at a conference held on 12 October 2011 at the ENS Ulm by the CNRS “Economics & Sociology” research group. The conference was entitled: “Social Classes: Have They Been Dissolved by Socio-economists in Networks, Generations and the Income Hierarchy?”. But basically, it was about asking who had made social classes disappear from the traditional means of interpreting the social world in sociology and other social sciences. Social Classes, Social Networks, Evolution of French Social, Classifications.
Comment décrire les liens entre mobilité « objective » et mobilité « subjective » ? Retour sur la proposition de D. Merllié
How to Describe Links between “Objective” Mobility and “Subjective” Mobility? Back to D. Merllié’s Proposal: We propose, like D. Merllié, to study the links between “objective” mobility and “subjective” mobility without constructing an a priori hierarchy of all socio-professional groups or categories. But to avoid the risks of over-interpretation of mean scores, we construct the cross-tabulation between these two mobility variables to analyze it with well known methods such as factorial correspondence analysis, calculations of percentages of optimistic pessimistic optimistic/pessimistic opinions, and even graphic representations. Objective Social Mobility, Subjective Social Mobility?
N. 117, January 2013
Extended Field Efforts to Reduce the Risk of Non-response Bias: Have the Effects Changed over Time? Can Weighting Achieve the Same Effects?
Julia Hall, Victoria Brown, Gerry Nicolaas, Peter Lynn
We examine the effects on survey estimates of extended interviewer efforts to gain survey response, including refusal conversion attempts and attempts to make contact with difficult-to-contact sample members. Previous research on this topic has identified that extended efforts do appear to affect estimates, and in ways that seem consistent with bias reduction. We extend the previous research in three ways. First, we provide the first study of changes over time in the effects of extended efforts on estimates. We study change in the UK over a ten-year period. Second, we use a more precise measure of the difficulty of contact and third, we assess the effects of extended efforts conditional on weight adjustments for non-response estimates as well as on unweighted sample statistics. Surveys, Interviewers, Non-contact, Non-response, Refusal.
Response Rate and Nonresponse Bias – Impact of the Number of Contact Attempts on Data Quality in the European Social Survey
Marek Fuchs, Dayana Bossert, Sabrina Stukowski
Increasing respondent contact problems and decreasing respondent willingness to cooperate have contributed to declining response rates in general population surveys, which has raised concerns of survey accuracy. To counteract nonresponse, several methods have been employed, including incentives, advanced letters, alternative survey modes for reluctant respondents, and increased field efforts to contact potential respondents. In particular, the number of contact attempts has been increased for many surveys. Even though more contact attempts increase survey costs, they are a reliable means for increasing response rates. However, the assumption that high response rates foster data quality and smaller nonresponse bias has been challenged. In this paper, we used contact data from the European Social Survey for Norway, Finland and Slovenia to see whether or not additional contact attempts resulting in a higher response rate can potentially reduce nonresponse bias. Response Rate, Nonresponse Bias, Contact Attempts, European Social Survey.
Les médias et l’opinion – Eléments théoriques et méthodologiques pour une analyse du débat sur l’identité nationale
Emmanuel Marty, Pascal Marchand, Pierre Ratinaud
The Media and Public Opinion – Theoretical and Methodological Analysis of the Debate on National Identity: This article deals with the “great debate on national identity” initiated by the Ministry of Immigration, Integration, National Identity and Mutually-Supportive Development in the winter of 2009-2010. Using the theory of framing (Bateson, 1972; Entman, 1993; Chong and Druckman, 2007), it seeks to identify the relationship between permeability and public debate as it is shown on the Web and the media coverage of the latter. To do this, a crossed lexicometric analysis of two corpora was done: on one hand, the 18,240 contributions posted on the dedicated ministerial site; on the other, the 1,436 press articles published from 25 October to 2 December 2009. A descending hierarchical classification was done with the Iramuteq software and identified, by their lexicon, places of permeability between personal and media frameworks in the apprehension of the debate, but also the specific universes of discourse to one or other corpora. We present them here in a detailed manner while proposing explanatory approaches of their dynamics. National Identity, Framing Theory, Lexicometry, Media Coverage, Opinion.
Bref rappel de trois problèmes méthodologiques de l’histoire de vie en sociologie
Brief Review of Three Methodological Problems of Life Histories in Sociology: This article reviews the popularity of life histories in sociology in the early 1970s, especially in light of the work done at that time by Daniel Bertaux. Forty years later, what can be said of it? After a brief historical overview, we consider: 1. The issue of representativeness; 2. The status conferred on common sense; and 3. The rigor of the analysis developed on the basis of a life history in the framework of grounded theory. These three topics are discussed so as to show possible solutions to the methodological problems they raise. A simple example of the analysis of a life history illustrates this point. Life Histories, Representativeness, Common Sense, Qualitative Analysis, Sociology.
Meta-analysis for Sociology – A Measure-driven Approach
David J. Roelfs, Eran Shor, Louise Falzon, Karina W. Davidson, Joseph E. Schwartz
Meta-analytic methods are becoming increasingly important in sociological research. In this article we present an approach for meta-analysis which is especially helpful for sociologists. Conventional approaches to meta-analysis often prioritize “concept-driven” literature searches. However, in disciplines with high theoretical diversity, such as sociology, this search approach might constrain the researcher’s ability to fully exploit the entire body of relevant work. We explicate a “measure-driven” approach, in which iterative searches and new computerized search techniques are used to increase the range of publications found (and thus the range of possible analyses) and to traverse time and disciplinary boundaries. We demonstrate this measure-driven search approach with two meta-analytic projects, examining the effects of various social variables on all-cause mortality. Meta-analysis, Sociology, Measure-driven, Alternative Search, Mortality.
Sample Size Requirements for Stable Clustering of Free Partition Sorting Data
Joachim Harloff, Adrian Stringer, Jennifer Perry
Convergence rates of cluster analysis results are studied for six sets of partition sorting data. Hierarchical cluster analysis, fuzzy cluster analysis, k-medoids clustering and four variants of the consensus clustering method are covered. The six data sets are kinship terms, colour plates, drug data, consultancy Web site items, social groups and prenatal tests. Convergence rates depend on the number of clusters extracted, on the clustering method and on the domain. For domains of 25 items or more, cluster analysis results identical to the final result were most consistently obtained using the third method of Gordon and Vichi (2001). Example evaluations are given for the most stable numbers of clusters for five sets of partition sorting using the method of Gordon and Vichi. The interpretations resemble previously published interpretations of results obtained from non-partitioning statistical methods.
N. 118, April 2013
Discrete Graphical Models in Social Mobility Research – A Comparative Analysis of American, Czechoslovakian and Hungarian Mobility before the Collapse of State Socialism
Renáta Németh, Tamás Rudas
Variants of path models have been widely used for the analysis of the social status attainment process. The methods presented here differ from earlier approaches in several ways. Social status is considered a categorical variable and path models are developed starting from graphical models, using the marginal log-linear approach. Overall model fit may be tested by standard techniques. Under these models, the status attainment process is completely characterized by a set of parameters that measure the strengths of the relevant effects. This is in sharp contrast with estimating and interpreting ad hoc parameters, without paying attention to overall model fit and to other effects influencing the process. The method is applied to the social status attainment process in the USA, Hungary and Czechoslovakia at the end of the last century, and shows that policies in the latter socialist countries to prevent status inheritance had little success. Graphical models, Marginal models, Marginal log-linear parameters, Intergenerational mobility.
L’analyse des réseaux sociaux – Un survol à travers quelques jalons
The Social Network Analysis – An Overview via Certain Milestones: Social network analysis has multidisciplinary origins (anthropology, social psychology, sociology, mathematics). It brings with it to sociology a point of view based on the study of social circles, on one hand, and relations between persons, on the other. Social cohesion, social roles, mediation, diffusion of innovations and ideas, small world, social capital, etc. are among the topics to which social network analysis has provided interesting insights. This text is based on the keynote presentation at the 15-20 September 2008 CNRS Summer School on social networks in Cargèse, Corsica. Social Networks, History, Instrumentation, Sociology.
Ecole thématique CNRS « Etudier les réseaux sociaux », Porquerolles,10-14 septembre 2012
Michel Grossetti, Claire Lemercier, Claire Bidart, Michel Bertrand
CNRS Thematic School “Studying Social Networks”, Porquerolles, 10-14 September 2012: Following the success of the first CNRS Summer School on social networks in Cargèse, Corsica, 15-20 September 2008 (see the preceding article by Alain Degenne), the organizers and the CNRS decided that a second thematic school should be organized on the Mediterranean island of Porquerolles, this time with more of a concentration on social sciences other than sociology, and specifically on history. This is a report on this rather successful second thematic school on social networks. Social Networks Analysis, France, CNRS Summer Schools, Porquerolles.
Considering Dialect in Survey Research
Isabelle Renschler, Brian Kleiner
This paper describes ways in which dialect variation may have important effects on the implementation and results of surveys in particular linguistic settings. Specifically, such differences of speech across populations within countries may significantly interfere with survey cooperation and with normal processes of understanding and responding to survey questions. The Swiss context is presented as a case study of how complications in survey implementation due to language and dialect variation may jeopardise data quality. Greater awareness among researchers of the potential impact of dialect variation on survey participation and responding is needed. Also, there is a need for more research internationally on the nature and scope of potential problems due to dialect, and on how to develop targeted strategies and remedies. Survey interviewing, Sight translation, Measurement equivalence, Interviewer adaptation, Dialect?
Justifier et préciser l’interprétation des données statistiques
Justify and Clarify The Interpretation of Statistical Data : Sociological criticism of statistical data is expressed in semantic terms. The problem of classification has been widely recognized in France, but that concerning interpretation, related to contextualism, remains relatively disregarded. Three kinds of solutions are presented. To recognize the contribution of these interpretative strategies also leads to seeing that it is important, but often difficult, to specify the contents of these sociological results. Statistics, Contextualism, Categorization, Data significance, Sociological interpretation, Semantics.
Empirical Data and Theory Construction: An Example of Application in Social Science Research
The classical approach in sociology, as much as in other scientific fields, consists in the use of hypotheses and hypothesis testing processes to determine what, if any, effects can be attributed to particular factors in the case being studied. Other scholars however propose another procedure that eliminates the presence of previous hypotheses and privileges an interpretive – as opposed to a hypothetic-deductive – approach to data analysis. However, in the latter approach crucial questions appear for the researcher: how does one approach the data and data analyses to ensure the credibility of findings? Blumer’s suggestion of “sensitizing concepts” seems to be appropriate in order to reach not only reliable interpretations of data but also to retain the possibility of “building theory”. This process could be supported by computer-assisted research. Methodology, Qualitative Approach, Sensitizing concepts, Computer-assisted analysis, Grounded Theory.
N. 119, July 2013
Karl M. van Meter
Alain Desroisères: This colleague, friend and internationally known specialist on the history and development of statistics, especially in the social sciences, died on 15 February 2013 before we could publish his article which follows in English. This is a short description of his life and work. Alain Desroisères, History of Statistics, Development of Statistics.
The History of Statistics as a Genre: Styles of Writing and Social Uses
To what extent does the history of statistics belong to the literary genre of “social constructions”? Does this genre include the development of national statistics? This brief historical summary of selected steps in the social uses of history and, more recently, of the history of science, can be helpful for examining the different ways in which the history of statistics has been written and used. It includes five sections: The First Period of National Statistical Histories (1800-1920); The Difficulties of Statistical Internationalism; The History of the Mathematization of Statistics; The New Wave of National-Statistics History (since 1975); The Bielefeld Group and the “Probabilistic Revolution”. History of Statistics, National Statistics, Social Uses of History, Internationalization of Statistics, Bielefeld Group.
An Order on Cross-Tabulations and Degrees of Association
It is often argued that an order exists in a cross-tabulation when the table’s margins have such a structure. We can free ourselves from this point of view and clearly define an order on the table itself. As Louis Guttman noted previously in the case of scalogram analysis, one must often move rows and columns about to be able to create a scale. In this case, it is the order of the table’s structure which induces an order on the margins and not the reverse. However, Goodman and Kruskal, when they proposed the gamma index that defines the strength of an association in the ordered case, only use the margins’ order, and they have since then been followed by most researchers. One should return to the original intuition of Guttman and show that at least an approximate order is almost always present in a table. The ordered cross-tabulation generated by ordered questions is only one case among many others and conversely a table with a strong order structure induces an order on question modalities. With real examples, we show that the criteria are available to define an order on a table, that there are formalized methods to reveal the associated structure, that there are also different indices to measure the degree of association, and finally that there are tests to assess the level of significance. Cross-tabulations, Order Structures, Gamma Index, PEM Index.
La passation de questionnaire – Chronique d’un solliciteur de l’espace public
Filling Out Questionnaires – Chronicle of a Public Space Solicitor: In a divers public space increasingly exposed to solicitations of all kinds, a questionnaire survey can be problematic. If the culture of the solicitation is integrated by the vast majority, mechanisms of refusal are equally integrated in the culture. The aim of this paper is to profile the solicitation work of surveyors through the ethnographic description of a questionnaire interviewer’s work. It relies in particular on the need to understand the interaction framework that unites the respondents and interviewers to determine the efficiency with which the questionnaire is administered. The ability to cope with this framework – the set of tangible and intangible elements that structure and limit, but also make it possible the execution of the interview – is one of the skills of interviewers in action. Quantitative Studies, Questionnaire, Filling Out Questionnaires, Solicitation Work, Framework for Interaction.
Forms and Modes of Apprehending Interdisciplinarity – A Socio-Computer Analysis of Sports Sciences
Cécile Collinet, Philippe Terral, Patrick Trabal, Matthieu Delalandre
This article considers the interdisciplinary character of sports sciences, which holds a double challenge. On one hand, demonstrating how the analysis of a corpus of scientific texts contributes – without exhausting the magnitude of the scientific work – to a better knowledge of a field of interdisciplinary research: sports sciences. On the other hand, on a methodological level, tools and computer strategies have been developed in order to analyze textual data. This article reveals how interdisciplinarity explicitly appears in scientific papers, not only as a discussion topic, but also as an organizational watchword in sports sciences. Besides, the analysis of this corpus helps to finely characterize active forms of interdisciplinary work organized around hybrid disciplines or specific research subjects. Interdisciplinarity, Multidisciplinarity, Discipline, Sports Sciences, Socio-Computer.
XI Conference of the International Society for Quality-of-Life Studies (ISQOLS)
“Discovering New Frontiers in Quality of Life Research” (1-4 November 2012, Venice, Italy)
This report on the recent eleventh conference of the International Society for Quality of Life Studies (ISQOLS), which took place in Venice, traces the development of methods and movements involved in measuring quality of life, including the new BES project, established at the Italian National Institute of Statistics, and the initiative “How is Life”, led by OECD. Wellbeing, Quality of Life, Social Indicators, International Society for Quality-of-Life Studies (ISQOLS), Venice Conference.