Journal of global business and technology · 2009 · 33 citations · 0 references
Social SciencesInformation SystemsEconometricsWeb AnalyticsWeb TechnologiesQuestionnaire DesignSampling TechniquesSampling TheorySurvey ProcessesSurvey ResearchSurvey SamplingOnline SurveysSurvey DesignSurvey Research DesignSurvey-based MeasurementWeb Survey Design
ABSTRACT The paper examines the relationship between web-based surveys and more traditional survey methods. The paper examines issues related to: sample quality, response rates, sample panels, and the use of incentives in a web-based environment. The paper also discusses web-based survey design issues and factors to be considered for effective utilization of web-based surveys in litigation and research contexts and identifies effective implementation techniques, including use of devices to assure the quality of the resulting data. Managerial implications and directions for future research are also discussed. INTRODUCTION The role of the Internet as a vehicle for communications is no longer questioned. Within the US, estimates of Internet usage range from 69% (Pew Research 2007) to 75% (Fadner and Mandise 2004) and over 40% worldwide, with growth in European and Asian markets growing at double-digit rates (Roster 2004). In fact, as noted by Cate (2008) 76% of Chinese have broadband and spend 20% or more time in an average week. Moreover, while studies have shown lower coverage levels for Hispanic s and residents of rural areas in the US, these differences are diminishing with every passing year (Zang 2000). On the other hand, studies in the UK (Roster et al 2007) found that, while web access in Northern Europe is equal or higher then US penetration, accessibility in Southern Europe, Spain, Portugal, Italy, and Greece was largely concentrated in the urban areas. Additionally, as noted by Cambiar (2006) online research continues its headlong march to become one of the most dominant (if not the most dominant) data collection methodology worldwide. As noted by ESOMAR (2008) online research continued its rapid ascent and is now the fastest growing methodology, with an increase of 14% year over year. They also noted that the largest absolute increase in 2007-2008 took place in the UK, where research almost doubled. Evidence that the Internet has become the vehicle of choice for many survey research organizations is found throughout the literature. For example, Internet research now represents 10-30% of customized research for AC Neilson and almost 47% of Harris Interactive' s polling research (Einhart 2003). In fact, in 2003 Wilson and Laskly (2003) predicted that 30% of research worldwide would be conducted by 2010. Three years later (2006) estimates are that 87% of research companies use methodologies (Cambier 2006), with an estimate that over 35% of data collected in 2006 was online. (Bradley 2 00 6) [ 1I The problem, however, as noted by Wilson and Laskey (2003), is that there is very little research outside the US to determine the current usage of internet-based research and the experiences of those involved in providing such methodology. Research evaluating on-line survey methodologies that has begun to emerge, has involved studies comparing differences in response rates and data quality between mail and/or phone surveys and Internet or web-based methods. Studies have also examined the impact of incentives, types of contacts or invitations used in web-based studies, and question format on data quality and willingness to respond. The question is, what guidelines should a researcher consider when deciding to utilize a web-based survey? This paper sets out those factors which are most likely to be raised when evaluating research, either as part of an organization's corporate strategy or, in fact, by judges and attorneys being asked to consider research results as part of a litigation strategy (GeIb and GeIb 2006). The starting point for such an articulation of guidelines for internet survey research is a review of traditional survey research methods since internet surveys are generally compared with more traditional methodologies, particularly telephone and face -to -face (i.e., mall intercept) studies. TRADITIONAL RESEARCH METHODS Telephone The traditional telephone survey was, throughout most of the latter part of the Twentieth Century, the method of choice for survey research, including in support of litigation. …