Who tends to answer open-ended questions in an e-service survey? The contribution of closed-ended answers. Issue 12 (2nd December 2017)

• Record Type: Journal Article
• Authors: Zhou, Ronggang; Wang, Xiaorui; Zhang, Leyuan; Guo, Haiyan
• Is Part Of: Behaviour & information technology. Volume 36:Issue 12(2017)
• Page Start: 1274
• Page End: 1284
• Publication Date: 2017-12-02
• Subjects: Open-ended questions -- closed-ended questions -- web survey -- e-service -- user-centred design; Electronic data processing -- Periodicals; Human engineering -- Periodicals; Information technology -- Periodicals; 303.4833
• DOI: 10.1080/0144929X.2017.1381165
• Languages: English
• ISSNs: 0144-929X

Abstract

ABSTRACT: This study presents a web survey investigating the effects of gender, age, prior usage behaviours, and closed-ended answers on response behaviour for open-ended questions in a user satisfaction or experience evaluation. Two types of open-ended questions were measured: general comment-specific questions designed to collect comments and explanation-specific questions designed to determine the reason for a higher or lower score for the closed-ended questions. Using an online structured questionnaire on an e-service telecom website, 13, 346 valid responses (73.1% male; all aged 19 or above) were analysed. More than 75% of the respondents did not answer any of the open-ended questions. Although respondents tended to answer comment-specific questions (19.9%) more often than explanation-specific questions (11.9%), personal characteristics emerged as significant predictors of participants' response behaviour in both types of open-ended questions. Males, younger participants, and those who had more e-service usage behaviours answered more often than other participants. Regarding the relationship between answers in closed-ended and open-ended questions, respondents' scaled scores were significant predictors of responses to open-ended questions, particularly comment-specific questions. This study suggests that a score of four on a five-point scale may indicate an interesting answer, with negative responses (5.5%) more likely than positive responses (3.9%).