The quality of responses to grid questions as used in Web questionnaires (compared with paper questionnaires)

Online Data Collection to Evaluate a Theoretical Cognitive Model of Tinnitus

“…Online survey software makes it easy to download data directly into computer packages and greatly reduces the potential for data entry mistakes (Alessi & Martin, 2010). Survey respondents are less likely to submit incomplete questionnaires when submitted online rather than when they are required to send paper-based responses via the mail (De Rada & Domínguez, 2015). However, online data collections may also have a number of disadvantages.…”

Constructing Behaviour Profiles for Answer Behaviour Across Surveys.

“…This tendency may differ intra-individually when survey mode is switched (Aichholzer, 2013) and more generally with more extreme answers in interviewer-administered than self-administered surveys (De Leeuw, 1992). Studies reported relatively more extreme positive responding in telephone mode (Ye et al, 2011) and in postal surveys versus internet surveys (Díaz de Rada & Domínguez, 2015). An increase in extreme responding may refer to stable response behaviour.…”

“…For attitudinal statements, respondents may refrain from expressing a judgment by filling out the middle or neutral response of the answering options (see for instance Kalton, Roberts, & Holt, 1980;Krosnick & Fabrigar, 1997;Tarnai & Dillman, 1992). Or they may choose the most extreme option that expresses their judgment (see for instance Aichholzer, 2013;Díaz de Rada & Domínguez, 2015;Ye, Fulton, & Tourangeau, 2011). In both cases, the answer may refer to their true opinion, but may also be the result of satisficing.…”

“…This may lead respondents to differentiate to a smaller extent between the questions in their answers (Krosnick & Alwin, 1989). Straightlining, or non-differentiation, refers to giving the same answers to a series of questions arranged in a grid format (Díaz de Rada & Domínguez, 2015;Schonlau & Toepoel, 2015). Straightlining seems more common towards the end than the beginning of a questionnaire (Krosnick, 1991).…”

“…Acquiescence is defined as the tendency to answer affirmatively, regardless of the content of the question (Billiet & McClendon, 2000;De Leeuw, 1992;Heerwegh & Loosveldt, 2011;McClendon, 1991). Acquiescence may be modedependent (Díaz de Rada & Domínguez, 2015) and potentially lead to measurement error. Saris et al (2010) found relatively lower data quality for agree-disagree rating scales, referring to the tendency to acquiesce in case of agree-disagree items (O'Muircheartaigh et al, 2000;Schaeffer & Presser, 2003).…”