Web surveys have become the dominant mode of survey data collection. They offer advantages in terms of cost and time over other modes and they are the new normal in survey research for many populations and in many fields of study. Nevertheless, web surveys, like other survey modes, are affected by the consequences of satisficing behavior, which is attributed, among other things, to low motivation among respondents. We assume that the motivation of respondents can be increased by using a respondent-friendly approach in the design of the questionnaire ( Dillman, 2000), which was achieved by implementing a chatbot-like questionnaire interface. In a randomized field-experiment conducted among university students we employed a between-subject design comparing a chatbot-like interface and a traditional web survey design administering the same questionnaire. We assessed respondent evaluation, as well as data quality indicators like response time, non-differentiation, item missing rates, and the length of answers to narrative open-ended questions. Results indicate that respondents perceived the chatbot-like design as more original and entertaining than the traditional web survey design. By contrast, participants rated the chatbot-like interface as more difficult to navigate. The analyses of response time and character count of answers to open-ended questions showed no significant differences between the two designs. The proportion of respondents with at least one item-missing was marginally lower in the chatbot-like design, while the degree of differentiation of one of the multi-item scales was higher in the web survey design. Comparaison exp & eacute;rimentale entre un questionnaire de type chatbot et un questionnaire web traditionnel. Les enqu & ecirc;tes en ligne sont devenues le mode pr & eacute;dominant de collecte de donn & eacute;es. Elles offrent des avantages de co & ucirc;t et de temps par rapport aux autres m & eacute;thodes et constituent d & eacute;sormais la norme dans la recherche par sondage pour de nombreuses populations et domaines d'& eacute;tude. N & eacute;anmoins, & agrave; l'instar des autres modes d'enqu & ecirc;te, les questionnaires web subissent les cons & eacute;quences des comportements de satisficing attribu & eacute;s, entre autres, & agrave; une faible motivation des r & eacute;pondants. Nous postulons que la motivation des r & eacute;pondants peut & ecirc;tre accrue en adoptant une approche plus conviviale (respondent-friendly) dans la conception du questionnaire ( Dillman, 2000), ce qui a & eacute;t & eacute; mis en oe uvre ici via une interface de type chatbot. Dans le cadre d'une exp & eacute;rience de terrain al & eacute;atoire men & eacute;e aupr & egrave;s d'& eacute;tudiants universitaires, nous avons employ & eacute; un protocole inter-sujets (between-subject design) comparant l'interface chatbot & agrave; un design web traditionnel pour un m & ecirc;me questionnaire. L'& eacute;tude & eacute;value l'exp & eacute;rience des r & eacute;pondants ainsi que plusieurs indicateurs de qualit & eacute; des donn & eacute;es : le temps de r & eacute;ponse, la non-diff & eacute;renciation des r & eacute;ponses (sch & eacute;mas de r & eacute;ponse uniformes), le taux de donn & eacute;es manquantes par item, et la longueur des r & eacute;ponses aux questions ouvertes. Les r & eacute;sultats indiquent que les r & eacute;pondants per & ccedil;oivent le design chatbot comme plus original et divertissant que le format traditionnel. En revanche, ils ont jug & eacute; l'interface chatbot plus difficile d'utilisation. Les analyses du temps de r & eacute;ponse et du nombre de caract & egrave;res pour les questions ouvertes ne r & eacute;v & egrave;lent aucune diff & eacute;rence significative entre les deux mod & egrave;les. La proportion de r & eacute;pondants ayant au moins une donn & eacute;e manquante & eacute;tait l & eacute;g & egrave;rement inf & eacute;rieure avec le chatbot, tandis que le degr & eacute; de diff & eacute;renciation sur l'une des & eacute;chelles multi-items & eacute;tait plus & eacute;lev & eacute; dans le format web traditionnel.

In open-ended numeric survey questions, certain numbers, typically round numbers, are disproportionately mentioned, creating unsubstantial “heaps” in the response distribution. These distortions often arise from satisficing behavior or imprecise memory, making it difficult to correct the resulting measurement error through post-hoc weighting. With the intention to reduce the incidence of heaping during data collection, we conducted an experimental study embedded in a web survey of the general internet population in Germany. Respondents answered two list-style numeric open-ended questions and were randomly assigned to either an experimental group, where round-number responses triggered immediate interactive feedback, or a control group getting no feedback. Results show that the feedback in the form of instructing and appreciating text-bubbles significantly reduced both the prevalence of round answers and the degree of roundness of the responses. This indicates that interactive feedback can effectively mitigate heaping in numeric open-ended questions.

Measurement error refers to differences between values reported by the respondents and their true values, with deviations in web surveys potentially stemming from the respondents themselves, from the survey instrument, or from the interaction between both respondent- and instrument-related factors. This chapter begins with a review of the literature on assessing and preventing measurement error in web surveys. The authors embedded several randomized between-subjects field experiments in three web surveys to examine the use of interactive feedback enabling direct interaction with the respondents and immediate reaction to their response behavior. Field experiments embedded in large ongoing surveys are typically preferred over lab experiments. Future research needs to always consider the potential impact of nonresponse bias on the external validity and generalizability of main effects demonstrated in field-experimental studies concerning visual design features in web surveys.

Check-all-that-apply questions are one of the most commonly used question formats in self-administered surveys. They are especially valuable because they allow respondents to select several responses from a list of alternatives that they consider applicable. In this study, we assessed the effectiveness of different types of instructions requesting a specific number of responses to a check-all-that-apply question in a web survey. We compared “static” instructions that are always visible together with the question stem, “dynamic” instructions that instantly appear once respondents start answering the question, and “combined” instructions taking advantage of both static and dynamic instructions. Findings showed that in view of respondent compliance with the instruction, the combination of a static and dynamic instruction is most effective. However, findings also revealed that the specific number of responses requested in the instruction has to be taken into account as a decisive factor influencing the response selection process and ultimately data quality.

Clarification features are used in Web surveys to improve the quality of responses. It is generally advised to place clarification features after the question stem. However, based on initial findings of an eye-tracking study (Kunz & Fuchs, 2012), we expected that the optimal position depends on the respective stage of the question-answer process as it is referred to by the clarification feature. In three Web surveys, the use and positioning of clarification features were tested in open-ended questions, with three different positions being experimentally varied: before the question stem, after the question stem, and after the answer box. Results indicated that contrary to expectations, the optimal position of clarification features did not differ depending on the respective stage of the question-answer process. Clarification features were principally most effective when they were positioned after the question stem, whereas clarification features placed before the question stem were least effective in improving the quality of responses.

This paper is concerned with the optimal wording of the recruitment question for a mobile phone panel survey in Germany. In order to learn more about the effects of different recruitment questions on the size and composition of the panel, we experimented with four different recruitment question versions. We analyzed the effectiveness of each question version with regard to three indicators: (1) What is the proportion of respondents who agree to take part in the panel? (2) Will the respondents who agreed to become panel members actually participate in the panel? (3) To what extent are differential nonresponse biases induced into the panel, since each question version may have differential effects on the composition of the recruited sample? Findings are discussed in light of an adaptive field work design: We propose a tailored request for participation for each individual respondent based on sociodemographic and other relevant variables.

In light of the growing importance of web-based data in the social and behavioral sciences, WEBDATANET was established in 2011 as a COST Action (IS 1004) to create a multidisciplinary network of web-based data collection experts: (web) survey methodologists, psychologists, sociologists, linguists, economists, Internet scientists, media and public opinion researchers. The aim was to accumulate and synthesize knowledge regarding methodological issues of web-based data collection (surveys, experiments, tests, non-reactive data, and mobile Internet research), and foster its scientific usage in a broader community.

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.

With the growing mobile-only population landline telephone surveys are increasingly complemented by mobile phone interviews using a dual frame approach. Typically it is assumed that a mobile phone is a personal device solely used by one individual. Even though several articles dealt with the eventuality that several persons may be reached when calling a mobile phone number, respondent selection procedures are currently not implemented. This paper provides further insight into this phenomenon. Using data from a 2010/11 survey conducted in the German cell phone population mobile phone sharing was examined indicating noteworthy prevalence rates. The sharing population also differed to the non-sharing population with respect to sociodemographic variables. Results are discussed in light of potential consequences for field work.

The continuously growing mobile-only population raises concerns regarding the representativeness of traditional landline telephone surveys. At this time, the mobile-only population differs significantly from general population, which leads to coverage bias when using fixed-line samples only for telephone surveys. However, in many European countries the mobile-only population is not the only source of coverage bias in telephone surveys. In addition, we have to consider coverage biases caused by considerable proportions of citizens without any telephone service. Since these two groups differ from the general population with respect to differential socio-demographic categories, in our view, the negative effects of mobile-only coverage error in traditional landline telephone surveys might in fact compensate—in part—for coverage bias caused by the no-phone population. To test this hypothesis of compensating coverage biases we calculated relative coverage biases caused by the mobile-only population and relative coverage biases caused by the no-phone population in 30 European countries for two socio-demographic variables in two points in time. Results are presented for four groups of countries that differ with respect to no-phone and mobile-only rates. Results suggest that—in general—mobile-only biases and no-phone biases do not compensate to a great extent, and thus the alarming mobile-only biases cannot be neglected when using telephone surveys in the estimation of population parameters. Nevertheless, there are several countries where the bias caused by the mobile-only population is far bigger than the joint bias caused by the mobile-only population and the no-phone population. This finding suggests that biases caused by the recent mobile-only population would be even more severe if the no-phone population did not exist.

Using cell phones in surveys poses new challenges to survey researchers. Amongst others, high proportions of nonworking numbers in randomly generated cell phone samples are reflected in increasing survey costs and extended fieldwork periods. In addition, ambiguous voicemail and operator messages which do not clearly indicate the status of a number negatively affect the proportion of numbers of unknown eligibility as well as the reliability of response rates. The simulated experiment reported in this article aimed at increasing the proportion of working cell phone numbers in the sample by screening out technically invalid or temporarily inactive numbers prior to fieldwork. Two methods were tested: number validation and text messaging. Findings from the application of different screening rules varying in strictness show that pre-call validation increases working number rates whereby survey costs and numbers of unknown eligibility can be reduced. Strict screening seems to be more efficient, however, at the expense of higher screening error.

Exploring Animated Faces Scales in Web Surveys: Drawbacks and Prospects Web surveys have been mimicking paper questionnaires with respect to their layout and appearance for a long time.Even though the rapid development of the internet and internet data collection methods offers various graphic and multimedia design features, very little is known about the influence of animated web survey questions on the question answering process.In a web survey among university journal readers, we conducted an experiment exploring the effects of implementing animated faces in scale questions.By varying the visual appearance of faces in a scale question, we enhance their influence on the question answering process.

79 Authors: Stephanie Steinmetz, Lars Kaczmirek, Pablo de Pedraza, Ulf-Dietrich Reips, Kea Tijdens, Katja Lozar Manfreda, Lilly Rowland, Francis Serrano, Marko Vidakovic, Carl Vogel, Ana Belchior, Jernej Berzelak, Silvia Biffignandi, Andreas Birgegard, Ernest Cachia, Mario Callegaro, Patrick J Camilleri, Gian Marco Campagnolo, Marta Cantijoch, Naoufel Cheikhrouhou, Daniela Constantin, Reuven Dar, Sophie David, Edith de Leeuw, Guy Doron, Enrique Fernandez-Macias, Niels Ole Finnemann, Muriel Foulonneau, Nicoletta Fornara, Marek Fuchs, Frederik Funke,

The field experiment (n = 24,999) reported in this paper was designed to decrease survey costs and the interviewers’ workload by scre ening out technically invalid numbers and numbers of unknown eligibility prior to fieldwork of a telephone survey in the cell phone frame. In addition, effects of pr e-call validation on data quality were examined. Two methods were tested: number validation and text messaging. Furthermore, several screening conditions of differ ent strictness were examined. Results indicate that both number validation and te xt messaging are effective methods to increase the percentage of working numbers in th e field. Contact and interview rates can also be increased. A reduction of the proportio n of numbers of unknown eligibility due to pre-call validation results in an increase o f response rates. Altogether, high percentages of screened out numbers achieve considerable cost savings. However, precall validation of cell phone numbers comes at the risk of screening out valid cell phone numbers which potentially causes biases.

Die Hauptfragestellungen in der Governance-Forschung (vgl. hierzu und zum Folgenden Babyesiza/Kehm 2009) richten sich auf Entscheidungsstrukturen, -prozesse und -gegenstände. Gefragt wird zum Beispiel danach, wie Leitungs- und Verwaltungsstrukturen aussehen (in einer Hochschule, einer Schule, in einem Krankenhaus oder in der öffentlichen Verwaltung). Sind interne und externe Stakeholder in die Entscheidungsprozesse involviert? Wie variiert dies je nach Entscheidungsgegenständen? Welche staatlichen und privaten Ebenen greifen in welcher Weise in die Entscheidungsprozesse ein? Aus den Antworten auf solche und ähnliche Fragen wurde einerseits das Konzept der Multi-level- oder Mehrebenen-Governance abgeleitet, andererseits – und ganz in einem normativen Sinne gemeint – das Konzept der ,good governance’. Dieses Konzept ist bisher noch recht vage geblieben. Es wurde von den Experten der Weltbank geprägt und bezieht sich auf Forderungen nach Effizienz und Rechtsstaatlichkeit in Schwellen- und Entwicklungsländern sowie in so genannten ,failing states’. Es gibt aber keine einheitliche und breit akzeptierte Definition von ,good governance’. Mit dem Begriff verbinden sich Vorstellungen von Transparenz, Effizienz, Partizipation, Verantwortlichkeit, Rechtsstaatlichkeit und Gerechtigkeit.

The family is a key factor in the occurrence of school violence. In this paper, we will transcend the traditional explanatory models according to which domestic violence and hegemonic masculinity increase the prevalence of school violence while parental support for their children might decrease the likelihood of them turning violent in school. In our view, in addition to the direct effects of the domestic conditions on school violence, we also need to consider the micro-social properties of the class as causal factors contributing to its occurrence. In this paper, we aim to demonstrate that there are not only children who suffer from unfavourable domestic conditions who turn violent in schools. Instead we assume, that the average level of domestic violence, of parental support and of hegemonic masculinity in a given class also contributes to the frequency of violence committed by those students who themselves are not subject of such disadvantaged conditions.