Social Science Computer Review

Volume 29, No. 2

Summer 2011



Special issue on Web Survey Methods


Measuring political knowledge in telephone and web surveys:  A cross-national comparison / Zan Strabac & Toril Aalberg


The fast pace of technology changes makes conducting high-quality web surveys increasingly easy, though performance of web surveys should be continuously monitored. In this article a comparison is made of the results of telephone and web surveys of items measuring international news knowledge. We compare web surveys of general populations conducted in the USA and Norway in January 2009 with their telephone replications conducted in the same month. Results show rather small differences between web and telephone surveys, particularly in Norway. We discuss the results and make recommendations for use of web surveys and for future methodological research.


Words, numbers and visual heuristics in web surveys: Is there a hierarchy of importance?   / Vera Toepoel &  Don A. Dillman


Respondents follow simple heuristics in interpreting the visual features of questions. We carried out two experiments in two panels to investigate how the effect of visual heuristics affects the answers to survey questions. In the first experiment we varied the distance between scale points in a five point scale to investigate whether respondents use the conceptual or visual midpoint of a scale. In the second experiment we used different endpoint labels of a five point scale, by adding different shadings of color and numbers that differed both in sign and value (2 to -2), to study whether options that are similar of appearance are considered conceptually closer than when they are dissimilar in appearance. We predicted that there is a hierarchy of features that respondents attend to, with verbal labels taking precedence over numerical labels, and numerical labels taking precedence over visual cues. Our results confirmed our hypothesis: the effect of spacing of response options and different endpoints was only apparent in polar point scales and not in fully labeled scales. In addition, our study on two panels, with one consisting of extremely trained respondents and the other of relatively fresh respondents, shows that trained respondents are affected by the distance between response options while relatively new respondents are not. To reduce the effect of visual cues, taking into account the robustness of results, we suggest it is better to use fully labeled five-point scales in survey questions.


Response rates and data quality in web and mail surveys administered to PhD holders / Maite Barrios, Anna Villarroya, Ángel Borrego, & Candela Ollé


Despite the extensive use of Web surveys today, there are certain methodological factors related to participant cooperation and data quality which remain unclear and require further study. Here, we compare responses to a survey administered in two formats – electronic, or by post – in terms of overall response rate and the quality of the data collected. Web and mail questionnaires were sent to a sample of 572 PhD holders, asking them about aspects related to their academic career and personal and family data in order to investigate the factors that determine scientific productivity. The Web questionnaire elicited a significantly higher response rate than the mail questionnaire. Response rates did not differ between males and females; however, topic salience had an effect on the response rate. Finally, data quality was higher in Web surveys than in the mail surveys, with fewer overall errors, fewer missing items and longer responses in open-ended questions.


Sliders for the smart: Type of rating scale on the web interacts with educational level / Frederik Funke, Ulf-Dietrich Reips, & Randall K. Thomas


Slider scales and radio buttons scales were experimentally evaluated in horizontal and vertical orientation. Slider scales lead to statistically significantly higher break-off rates (odds ratio = 6.9) and substantially higher response times. Problems with slider scales were especially prevalent in participants with less than average education, suggesting the slider scale format is more challenging in terms of previous knowledge needed or cognitive load. An alternative explanation, technology-dependent sampling (Buchanan & Reips, 2001), cannot fully account for the present results. The authors clearly advise against the use of Java-based slider scales and advocate low-tech solutions for the design of Web-based data collection. Orientation on screen had no observable effect on data quality or usability of rating scales. Implications of item format for Web-based surveys are discussed.




Reports and Communications


Is public trust in government associated with trust in e-government? / Simon Horsburgh, Shaun Goldfinch, & Robin Gauld 


The term e-government describes the use of information and communications technology, particularly the internet, for the delivery of public services. As governments invest in e-government, there is only limited knowledge of the extent of public trust in the new electronic modes of delivery; we also know little about whether there is any relationship between trust in government and trust in e-government. This article reports on research designed to probe this issue. Drawing on survey data from Australia and New Zealand, a series of hypotheses are tested pertaining to relationships between public trust in government and in e-government, in the use of information and communications technology and trust in e-government, and support for e-government investment and development. Trust in government was found not to be correlated with trust in facets of e-government service provision, but was associated with support for e-government investment. More intensive internet users were more likely to trust e-government services.


Political participation of teenagers in the information era: The case of the 2008 Korean Candlelight Protests / Yun Seongyi & Chang Woo-young



This study examines the environments and the characteristics of political participation of teenage political participation.  The 2008 Candlelight Protests of Korea is used as the case of this study.  Traditionally, teenagers in Korea have known to disengage from political affairs because they are usually worn by intense academic challenge and competition.  In that sense, participation of teenagers at such a large scale shown in the 2008 Candlelight Protests of Korea can only be explained as huge shift in political preference of teenagers and conventional social norms. In that sense, the case of 2008 Candlelight Protests of Korea has shown great implications for the political potential of new media, which is capable of revolutionizing the political socialization patterns of youth. Survey results demonstrated that the internet had become an important tool from which the teenagers collected political information and channels which they utilized to organize and mobilize.  Numbers also showed that the degree of the youth’s socio-political interests were higher than the adult’s.  Other notable fact found was that female students showed more aggressive involvement than male students and this could be explained by the difference in the internet usage pattern between male and female students.  In using media, adolescent girls displayed more relationship and objective oriented behaviors than the boys.


VODYS:  An agent-based model for exploring campaign dynamics / Girish J. Gulati, Charles R. Hadlock, & Juliet F. Gainsborough


The literature on campaigns has considered a number of factors that affect whether and how someone votes, including demographics, campaign strategies, and social milieu. Understanding the dynamics of campaigns, however, is complicated by the fact that researchers cannot observe much of what happens during an election cycle. Typically, studies rely on voter recollections of conversations, contacts, and media exposure. In addition, because data are collected at discrete points in time, most models of voter turnout cannot capture the dynamic nature of an individual’s interactions during a campaign cycle. Agent-based models offer a way to overcome these data limitations by allowing us to model the dynamics of voter turnout over the course of many weeks as individuals move back forth between home and work environments, interacting with neighbors and colleagues. In this paper, we present an agent-based model of campaign dynamics, VODYS, and conduct three simulations to demonstrate the utility of agent-based models for exploring the effects of contact and context on political behavior.