Academic journal article International Journal of Business and Information

Questionnaire Pretesting Methods: A Comparison of Cognitive Interviewing and Respondent Debriefing Vis-À-Vis the Study of the Adoption of Decision Support Systems by Knowledge Workers

Academic journal article International Journal of Business and Information

Questionnaire Pretesting Methods: A Comparison of Cognitive Interviewing and Respondent Debriefing Vis-À-Vis the Study of the Adoption of Decision Support Systems by Knowledge Workers

Article excerpt

1.INTRODUCTION

As a vehicle for data collection, the questionnaire is a vital component in achieving higher quality in social science research. The best of sampling schemes and estimation strategies would not yield accurate data if the answers provided by the respondents are not meaningful (DeMaio et al.,1998). A frequent difficulty with questionnaire design is that respondents often misinterpret the questions - a difficulty that has been recognized in the literature (Hilton, 2015). In recent years, however, there has been increased emphasis on building quality into questionnaire design through pretesting. A pretest aims to ensure quality by subjecting the questionnaire to some type of evaluations to assess its ability to collect the desired data (Hughes, 2004). A pretest can improve questionnaire response rate and is therefore a valuable method to evaluate whether a new measure is performed in the field as planned (Hilton, 2015). Whether constructing a new scale or revising an existing scale, through pretesting, researchers may confirm whether the scale uses clear and appropriate language, has no obvious errors or omissions, and has adequate psychometric properties. Questionnaire pretesting, therefore, is a crucial element of a good study design (Greco & Walop, 1987). Although a pretest may not guarantee success for the main study, it does improve the likelihood of its success (Drennan, 2003; Hilton, 2015).

Although the value of pretesting has been recognized as critical to the valid measurement of phenomena by survey methodology, only a few studies have been published on pretesting that report on the specific pretesting protocols with a significant level of detail vis-a-vis the methods undertaken (e.g., DeMaio, et al. 1998; Hughes, 2004; Hilton, 2015). Nevertheless, there is no consensus on best practices (Presser et al. 2004; Beatty & Willis, 2007). According to Collins (2003), an evaluation of methods used in pretesting is often lacking (e.g., Subar et al., in a study of improving food frequency questionnaires).

As the popularity and variety of pretesting methods increases, the question remains whether different methods yield consistent and/or complementary results. On one hand, the three most commonly used front-end methods of pretesting are: (1) a forms appraisal system, which is a detailed, formal system of coding questions in questionnaires in order to detect problems; (2) cognitive interviewing, which has traditionally been used to identify problems in a questionnaire at a relatively early stage of its development; and (3) expert reviews, which are frequently used as a method of evaluating draft questionnaires. People who have theoretical questionnaire knowledge or practical experience are asked to review draft questions with an eye to identifying questionnaire problems (DeMaio & Landreth, 2003; Hughes, 2004).

On the other hand, two final-stage questionnaire evaluations methods have been used during a field test or actual production survey (Hughes, 2004). These methods are: (1) behavior coding, which is a system of coding the interactions between an interviewer and a respondent; and (2) respondent debriefing, which involves incorporating follow-up questions in a standardized interview.

This paper compares two critical pretesting methods - cognitive interviewing and respondent debriefing - with respect to evaluating a series of questions for the study of executive information systems (EIS) adoption factors by knowledge workers. Although other pretesting methods are recognized in the literature, the methods we discuss were chosen because together they provide information from all three potential sources of measurement errors: the questionnaire, the respondent, and the interviewer (DeMaio et al., 1998). Pretesting is one of the key considerations of survey questionnaire construction processes, in addition to specifying the research aims and objectives, establishing the data collection methods, and designing the questionnaire. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.