Design Effects in Web Surveys: Comparing Trained and Fresh Respondents

Research output: Working paperDiscussion paperOther research output

373 Downloads (Pure)

Abstract

In this paper we investigate whether there are differences in design effects between trained and fresh respondents. In three experiments, we varied the number of items on a screen, the choice of response categories, and the layout of a five point rating scale. We find that trained respondents are more sensitive to satisficing and select the first acceptable response option more often than fresh respondents. Fresh respondents show stronger effects with regard to verbal and nonverbal cues than trained respondents, suggesting that fresh respondents find it more difficult to answer questions and pay more attention to the details of the response scale in interpreting the question.
Original languageEnglish
Place of PublicationTilburg
PublisherEconometrics
Number of pages31
Volume2008-51
Publication statusPublished - 2008

Publication series

NameCentER Discussion Paper
Volume2008-51

Fingerprint

Web survey
Satisficing
Rating scales
Experiment
Layout

Keywords

  • professional respondents
  • questionnaire design
  • items per screen
  • response categories
  • layout

Cite this

Toepoel, V., Das, J. W. M., & van Soest, A. H. O. (2008). Design Effects in Web Surveys: Comparing Trained and Fresh Respondents. (CentER Discussion Paper; Vol. 2008-51). Tilburg: Econometrics.
Toepoel, V. ; Das, J.W.M. ; van Soest, A.H.O. / Design Effects in Web Surveys : Comparing Trained and Fresh Respondents. Tilburg : Econometrics, 2008. (CentER Discussion Paper).
@techreport{4c8b40aa755748449474bf010586ac2e,
title = "Design Effects in Web Surveys: Comparing Trained and Fresh Respondents",
abstract = "In this paper we investigate whether there are differences in design effects between trained and fresh respondents. In three experiments, we varied the number of items on a screen, the choice of response categories, and the layout of a five point rating scale. We find that trained respondents are more sensitive to satisficing and select the first acceptable response option more often than fresh respondents. Fresh respondents show stronger effects with regard to verbal and nonverbal cues than trained respondents, suggesting that fresh respondents find it more difficult to answer questions and pay more attention to the details of the response scale in interpreting the question.",
keywords = "professional respondents, questionnaire design, items per screen, response categories, layout",
author = "V. Toepoel and J.W.M. Das and {van Soest}, A.H.O.",
note = "Subsequently published in Public Opinion Quarterly, 2008 (rt) Pagination: 31",
year = "2008",
language = "English",
volume = "2008-51",
series = "CentER Discussion Paper",
publisher = "Econometrics",
type = "WorkingPaper",
institution = "Econometrics",

}

Toepoel, V, Das, JWM & van Soest, AHO 2008 'Design Effects in Web Surveys: Comparing Trained and Fresh Respondents' CentER Discussion Paper, vol. 2008-51, Econometrics, Tilburg.

Design Effects in Web Surveys : Comparing Trained and Fresh Respondents. / Toepoel, V.; Das, J.W.M.; van Soest, A.H.O.

Tilburg : Econometrics, 2008. (CentER Discussion Paper; Vol. 2008-51).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Design Effects in Web Surveys

T2 - Comparing Trained and Fresh Respondents

AU - Toepoel, V.

AU - Das, J.W.M.

AU - van Soest, A.H.O.

N1 - Subsequently published in Public Opinion Quarterly, 2008 (rt) Pagination: 31

PY - 2008

Y1 - 2008

N2 - In this paper we investigate whether there are differences in design effects between trained and fresh respondents. In three experiments, we varied the number of items on a screen, the choice of response categories, and the layout of a five point rating scale. We find that trained respondents are more sensitive to satisficing and select the first acceptable response option more often than fresh respondents. Fresh respondents show stronger effects with regard to verbal and nonverbal cues than trained respondents, suggesting that fresh respondents find it more difficult to answer questions and pay more attention to the details of the response scale in interpreting the question.

AB - In this paper we investigate whether there are differences in design effects between trained and fresh respondents. In three experiments, we varied the number of items on a screen, the choice of response categories, and the layout of a five point rating scale. We find that trained respondents are more sensitive to satisficing and select the first acceptable response option more often than fresh respondents. Fresh respondents show stronger effects with regard to verbal and nonverbal cues than trained respondents, suggesting that fresh respondents find it more difficult to answer questions and pay more attention to the details of the response scale in interpreting the question.

KW - professional respondents

KW - questionnaire design

KW - items per screen

KW - response categories

KW - layout

M3 - Discussion paper

VL - 2008-51

T3 - CentER Discussion Paper

BT - Design Effects in Web Surveys

PB - Econometrics

CY - Tilburg

ER -

Toepoel V, Das JWM, van Soest AHO. Design Effects in Web Surveys: Comparing Trained and Fresh Respondents. Tilburg: Econometrics. 2008. (CentER Discussion Paper).