This book deals with one of the most important scientific developments of recent years, namely the exponential growth of data science. More than a savvy term that rings of robotics, artificial intelligence and other terms that for long were regarded as part of science-fiction, data science has started to become structurally embedded in scientific research. Data, meaning personal data as well as information in the form of digital files, has become available at such a large scale that it can lead to an expansion of knowledge through smart combinations and use of data facilitated by new technologies. This book examines the legal implications of this development. Do data-driven technologies require regulation, and vice versa, how does data science advance legal scholarship? Defining the relatively new field of data science requires a working definition of the term. By data science we mean the use of data (including data processing) for scientific research. The availability of massive amounts of data as well the relatively cheap availability of storage and processing power has provided scientists with new tools that allow research projects that until recently were extremely cumbersome if not downright impossible. These factors are also often described with the term ‘big data’, which is characterized by three Vs: volume, velocity and variety.The term data science is nonetheless broader, because it can also refer to the use of data sets that are large but still limited—and therefore, unlike big data, of a manageable size for processing.
Browse by title
Edited by Vanessa Mak, Eric Tjong Tjin Tai and Anna Berlee
Edited by Hans-W. Micklitz, Anne-Lise Sibony and Fabrizio Esposito
Multi Actor Multi Criteria Analysis
Edited by Cathy Macharis and Gino Baudry
The Enactive Approach
Antonino Mario Oliveri
Using the general categories of the total survey error (TSE) paradigm, this chapter discusses issues related to the construction and administration of structured questionnaires in face-to-face interviewing. The main sources of non-sampling error are discussed, which emerge at this stage in the surveying process, as well as some solutions which can be adopted to control or limit these errors. Examples taken from tourism research are presented.
Robert E. Manning
Norms are a theoretical construct that has been widely used in sociology and the social sciences more broadly. Norm theory and related empirical methods have been applied in a range of park, recreation, and tourism contexts. This chapter reviews and illustrates the resulting body of scientific and professional literature. In particular, this body of work has measured the personal and social norms of recreation visitors and other stakeholders through survey research, and illustrates the ways in which resulting data can help inform 1) standards of quality for the ecological and experiential conditions of park, recreation, and tourism areas, and 2) the associated carrying capacity of such areas. A variety of research issues are addressed, including question and response formats, norm prevalence, norm salience, evaluative dimensions of norms, crystallization of norms, norm congruence, statistical measures of norms, stability of norms, effect of existing conditions on norms, and the validity of norms.
Faizan Ali, Woody G. Kim and Cihan Cobanoglu
Theory building in business research requires analytical accuracy and sophistication in research methods (Sarstedt et al., 2014). Nonetheless, the significance of newer analytical methods depends on the researcher’s willingness to learn, adopt, and apply them within the research process (Zahra and Sharma, 2004). A review of the literature shows that traditionally empirical studies in hospitality research used only basic statistical techniques. For instance, Line and Runyan (2012) reviewed the hospitality marketing research published in four top hospitality journals from 2008 to 2010. They stated that among 274 articles published, 103 (37.5 per cent) used some type of descriptive and multivariate analysis (for example, descriptive statistics, analysis of variance, regression, or factor analysis). However, Line and Runyan (2012) also stated that more recently an increase in the usage of advanced statistical tools including structural equation modelling (SEM) can be observed. These findings are confirmed by Yoo et al. (2011) in their assessment of empirical articles published in four top hospitality journals during 2000 and 2009. They revealed that out of a total of 570 empirical studies, 254 (44.5 per cent) of the articles used descriptive analytical methods such as descriptive statistics, t-tests, and cross-tabulation. Not only hospitality, but also other important academic fields such as marketing (Babin et al., 2008), family business research (Sarstedt et al., 2014), operations management (Peng and Lai, 2012), and tourism (Nunkoo et al., 2013) have observed a recent rise in the usage of sophisticated and rigorous quantitative methodologies. Amongst these methodologies, SEM is the most commonly applied method across a variety of academic disciplines such as strategic management, marketing, and psychology over the last few years (Astrachan et al., 2014; Chin et al., 2008; Hair et al., 2011). Lei and Wu (2007) stated that SEM characterizes an advanced version of general linear modelling procedures and is applied to examine whether ‘a hypothesized model is consistent with the data collected to reflect [the] theory’ (ibid., p. 34). In simple terms, SEM is a multivariate analytical tool that is used to test and estimate causal and/or hypothetical relationships among the variables concurrently (Astrachan et al., 2014). Its ability to allow statistical inspection of the relationships among theory-based variables and simultaneously employing confirmatory factor analysis (CFA) and linear regression models has contributed to its widespread application (Hair et al., 2014a). However, this argument holds true for covariance-based SEM (CB-SEM) and not for the partial least squares-based SEM (PLSSEM). CB-SEM is the most extensively applied approach of SEM and therefore many scholars refer to it as SEM (Astrachan et al., 2014). However, Hair et al. (2014b) refer to this argument as naive because PLS-SEM is an advantageous and increasingly applied method to assess structural equation models in different disciplines, including marketing, information systems, strategic management, tourism, and so on (Hair et al., 2012b; Hair et al., 2012a; Ringle et al., 2012). Yet, its use in hospitality research remains at an early stage of development (Ali et al., 2018) where its application is much lower as compared to its application in other disciplines including marketing, management information systems (MIS) and strategic management (see Figure 29.1).