Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top
Published in:
Notes on Comparative and Causal Analyses Using Loglinear, Logit, Logistic, and Other Effect Coefficients Read chapter Read first chapter

2020 | OriginalPaper | Chapter

Interest in Science: Response Order Effects in an Adaptive Survey Design

Author : Uwe Engel

Published in: Grundlagen - Methoden - Anwendungen in den Sozialwissenschaften

Publisher: Springer Fachmedien Wiesbaden

Log in

Activate our intelligent search to find suitable subject content or patents.

search-config
loading …

Abstract

The chapter covers a survey experiment on two kinds of response effects. Response order is analysed against the background of the prominent primacy/recency hypothesis in survey methodology. Since this hypothesis refers to unordered scales, a modified version is suggested for the case of ordinal scales. The use of four vs. five response categories represents a second experimental factor. The probit regression analysis confirms both the “modified primacy-effect hypothesis” and the “missing-equivalence hypothesis” on the use of four vs. five response categories. The experiment is embedded in an adaptive survey design. The data come from the “Bremen City-of-Science Survey” which was conducted in mixed – web and telephone – mode in the spring of 2016. For this survey, a probability sample of residents aged 18 + was drawn from the population register of the city of Bremen.

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

inform now

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

inform now
previous chapter Antwortskalenrichtung und Umfragemodus
next chapter Endgerätespezifische und darstellungsabhängige Bearbeitungszeit- und Antwortverhaltensunterschiede in Webbefragungen
Footnotes
1
See Engel and Köster (2015b) for an analysis of the latter case, esp. of “framing effects”.
 
2
This assumption is perfectly in line with Krosnick and Alwin’s expectation that respondents “choose the first acceptable alternative(s) among the offered choices”, especially in case of “survey items involving a large number of response options.” (Tourangeau et al. 2000, p. 250).
 
3
Weisberg (2005, p. 107 f.) provides a short review of the debate on whether to include middle categories for scale questions.
 
4
See Engel and Can (2017) for a more detailed description of the study design.
 
5
This strategy follows closely the recommendation to form a fully grouped propensity score (Little and Rubin 2002, p. 48 f.).
 
6
See Engel et al. (2017) for more information on this subject.
 
8
Question wording (in German language): „Wie groß ist im Allgemeinen Ihr Interesse an wissenschaftlichen Themen?“(1) „sehr gering“(2) „eher gering“(3)“teils, teils“(4) „eher groß“(5) „sehr groß“.
 
9
Question wording (in German language): “Wie stark interessieren Sie sich für wissenschaftliche Themen?” (1) „überhaupt nicht interessiert“(2) „eher nicht interessiert“(3) „eher interessiert“(4) „stark interessiert“
 
10
The identification topic is dealt with on various occasions, too (e.g., Long 1997, p. 117 f., 122–124; Long and Freese 2006, pp. 184–187, esp. 187). There are two common alternative parameterizations to arrive at an identified set of parameters (Rabe-Hesketh and Skrondal 2008, p. 294). In Stata (used here), the identifying assumption for ordinal regression models is that the intercept is zero, whereas the values of all thresholds are estimated.
 
11
To obtain for all four groups the complete set of estimates (incl. estimates of statistical significance), each model was, independently of each other, estimated in two equivalent versions. First, by including explicit terms for the first three groups (as displayed in the left-hand part of Table 1), using the fourth as reference group. Secondly, by including terms for the last three groups, using the first as reference. This yields perfectly equivalent and with respect to quantities estimated in both versions explicitly identical estimates, since the parameter solution is invariant of the choice of the reference category when using effect coding.
 
12
As described in the previous note, two equivalent model versions are set up with respect to group membership coding, of which each of the parallel model version includes all four \( x_{2g} \) terms for scale direction coding. This yields perfectly equivalent/identical parameter estimates across the two model versions.
 
13
Rabe-Hesketh and Skrondal (2008, p. 289) provide a good illustration of a threshold model for a three-category ordinal probit model.
 
14
Difference of entries in a row, column D minus column C.
 
15
If conducting a two-tailed test. However, based on the expectation derived from previous evidence from the PPSM panel (as outlined above), a one-tailed test would be arguable too. Then, statistically significant estimates of scale-direction effects would be attestable for two conditions, namely the “full interview” and the “basic questions” condition (Table 5).
 
16
Kendall’s tau-b is 0.21 and 0.24 respectively in the unweighted full sample.
 
Literature
Bethlehem, J. 2009. Applied survey methods: A statistical perspective. Hoboken: Wiley.CrossRef
Bethlehem, J., F. Cobben, and B. Schouten. 2011. Handbook of nonresponse in household Surveys. Hoboken: Wiley.CrossRef
De Leeuw, E.D. 2008. Choosing the method of data collection. In International handbook of survey methodology, Eds. E.D. de Leeuw, J.J. Hox, and D.A. Dillman, 113–135. New York: Lawrence Erlbaum Associates.
Engel, U. 2015. Response behavior in an adaptive survey design for the setting-up stage of a probability-based access panel in Germany. In Improving survey methods. Lessons from recent research, Eds. U. Engel, B. Jann, P. Lynn, A. Scherpenzeel, and P. Sturgis, 207–222. New York: Routledge.
Engel, U., and B. Köster. 2015a. Response effects and cognitive involvement in answering survey questions. In Improving survey methods. Lessons from recent research, Eds. U. Engel, B. Jann, P. Lynn, A. Scherpenzeel, and P. Sturgis, 35–50. New York: Routledge.
Engel, U., and B. Köster. 2015b. Framing effects. In Survey measurements. Techniques, data quality and sources of error, Ed. U. Engel, 58–75. Frankfurt: Campus.
Engel, U., S. Bartsch, C. Schnabel, and H. Vehre. 2012. Wissenschaftliche Umfragen. Methoden und Fehlerquellen. Frankfurt: Campus.
Little, R.J.A., and D.B. Rubin. 2002. Statistical analysis with missing data, 2. edn. Hoboken: Wiley.CrossRef
Long, J.S. 1997. Regression models for categorical and limited dependent variables. Thousand Oaks: Sage.
Long, J.S., and J. Freese. 2006. Regression models for categorical dependent variables using stata, 2. edn. College Station: Stata Press.
Lynn, P. 2003. PEDAKSI: Methodology for collecting data about survey non-respondents. Quality & Quantity 37:239–261.CrossRef
Rabe-Hesketh, S., and A. Skrondal. 2008. Multilevel and longitudinal modeling using stata, 2. edn. College Station: Stata Press.
Tourangeau, R., L.J. Rips, and K. Rasinski. 2000. The psychology of survey response. Cambridge: University Press.CrossRef
Weisberg, H.F. 2005. The total survey error approach. A guide to the new science of survey research. Chicago: University of Chicago Press.CrossRef
Metadata
Title
Interest in Science: Response Order Effects in an Adaptive Survey Design
Author
Uwe Engel
Copyright Year
2020
Book
Grundlagen - Methoden - Anwendungen in den Sozialwissenschaften

Print ISBN: 978-3-658-15628-2

Electronic ISBN: 978-3-658-15629-9

Copyright Year: 2020

DOI
https://doi.org/10.1007/978-3-658-15629-9_13

Premium Partner

    Image Credits