Bettina Grün

Challenging “Factor–Cluster Segmentation”

The concept of market segmentation has been widely accepted and warmly embraced both by tourism industry and academia. In tourism research, this increased interest in segmentation studies has led to the emergence of a standard research approach. Most notably a concept referred to as “factor–cluster segmentation” has been broadly adopted. The aim of this article is to demonstrate that this approach is not generally the best procedure to identify homogeneous groups of individuals (market segments).

What affects public acceptance of recycled and desalinated water

This paper identifies factors that are associated with higher levels of public acceptance for recycled and desalinated water. For the first time, a wide range of hypothesized factors, both of socio-demographic and psychographic nature, are included simultaneously. The key results, based on a survey study of about 3000 respondents are that: (1) drivers of the stated likelihood of using desalinated water differ somewhat from drivers of the stated likelihood of using recycled water; (2) positive perceptions of, and knowledge about, the respective water source are key drivers for the stated likelihood of usage; and (3) awareness of water scarcity, as well as prior experience with using water from alternative sources, increases the stated likelihood of use. Practical recommendations for public policy makers, such as key messages to be communicated to the public, are derived.

Fitting finite mixtures of generalized linear regressions in R

R package flexmix provides flexible modelling of finite mixtures of regression models using the EM algorithm. Several new features of the software such as fixed and nested varying effects for mixtures of generalized linear models and multinomial regression for a priori probabilities given concomitant variables are introduced. The use of the software in addition to model selection is demonstrated on a logistic regression example.

Environmentally friendly behavior - can heterogeneity among individuals and contexts/ environments be harvested for improved sustainable management?

The study of behavior with environmental consequences (recycling, water conservation, etc.) has received significant attention from social scientists over the past few decades. However, few studies have closely examined the systematic heterogeneity of behavior with environmental consequences. This study tests two specific hypotheses about such heterogeneity: that individuals differ systematically in their patterns of behavior with environmental consequences and that behavioral patterns systematically differ between context/environments. Both hypotheses are investigated empirically in the home and vacation environment. Results support the assumption that systematic differences in behavioral patterns exist across individuals. With respect to context/environment dependence, some groups of individuals do not change their behavior much between contexts/environments. The majority, however, tend to engage in fewer proenvironmental behaviors in the vacation context. These findings have significant implications for environmentally sustainable management, both for local councils and tourism destinations.

Validly Measuring Destination Image in Survey Studies

Destination image is among the most frequently measured constructs in empirical survey research. Academic tourism researchers tend to use multi-category scales, often referring to them as “Likert scales,” while industry typically uses “pick-any” measures. But which leads to results that are more valid? Findings from a large-scale experimental study show that a “forced-choice full binary” format (where respondents have to tick “yes” and “no” for each destination-attribute combination) performs better than both current preferred formats in academic and applied studies.

Water conservation behavior in Australia.

Ensuring a nations long term water supply requires the use of both supply-sided approaches such as water augmentation through water recycling, and demand-sided approaches such as water conservation. Conservation behavior can only be increased if the key drivers of such behavior are understood. The aim of this study is to reveal the main drivers from a comprehensive pool of hypothesized factors. An empirical study was conducted with 3094 Australians. Data was analyzed using multivariate linear regression analysis and decision trees to determine which factors best predict self-reported water conservation behavior. Two key factors emerge: high level of pro-environmental behavior; and pro-actively seeking out information about water. A number of less influential factors are also revealed. Public communication strategy implications are derived.

Required Sample Sizes for Data-Driven Market Segmentation Analyses in Tourism

Data analysts in industry and academia make heavy use of market segmentation analysis to develop tourism knowledge and select commercially attractive target segments. Within academic research alone, approximately 5% of published articles use market segmentation. However, the validity of data-driven market segmentation analyses depends on having available a sample of adequate size. Moreover, no guidance exists for determining what an adequate sample size is. In the present simulation study using artificial data of known structure, the impact of the difficulty of the segmentation task on the required sample size is analyzed in dependence of the number of variables in the segmentation base. Under all simulated data circumstances, a sample size of 70 times the number of variables proves to be adequate. This finding is of substantial practical importance because it will provide guidance to data analysts in academia and industry who wish to conduct reliable and valid segmentation studies.

Quick, simple and reliable: forced binary survey questions

Consumers are increasingly saturated by market research, which leads to decreasing response rates and an increased danger of response bias. Market researchers thus face the challenge of recruiting respondents, increasing response rates and reducing respondent fatigue by making questionnaires as short and pleasant as possible. One way of achieving this is to replace traditionally used ordinal multi-category answer formats (such as Likert-type scales) with forced binary scales. This proposition is attractive only if it indeed shortens the survey time while not compromising the quality of managerial insights from the data. This study investigates these conditions. Results from a repeat-measurement design indicate that managerial interpretations do not differ substantially between the two answer formats, responses are equally reliable, and that the binary format is quicker and perceived as less complex.

Gingival Tissue Transcriptomes Identify Distinct Periodontitis Phenotypes

The currently recognized principal forms of periodontitis—chronic and aggressive—lack an unequivocal, pathobiology-based foundation. We explored whether gingival tissue transcriptomes can serve as the basis for an alternative classification of periodontitis. We used cross-sectional whole-genome gene expression data from 241 gingival tissue biopsies obtained from sites with periodontal pathology in 120 systemically healthy nonsmokers with periodontitis, with available data on clinical periodontal status, subgingival microbial profiles, and serum IgG antibodies to periodontal microbiota. Adjusted model-based clustering of transcriptomic data using finite mixtures generated two distinct clusters of patients that did not align with the current classification of chronic and aggressive periodontitis. Differential expression profiles primarily related to cell proliferation in cluster 1 and to lymphocyte activation and unfolded protein responses in cluster 2. Patients in the two clusters did not differ with respect to age but presented with distinct phenotypes (statistically significantly different whole-mouth clinical measures of extent/severity, subgingival microbial burden by several species, and selected serum antibody responses). Patients in cluster 2 showed more extensive/severe disease and were more often male. The findings suggest that distinct gene expression signatures in pathologic gingival tissues translate into phenotypic differences and can provide a basis for a novel classification.

Finite mixtures of generalized linear regression models

Generalized linear models have become a standard technique in the statistical modelling toolbox for investigating relationships between variables. The assumption of homogeneity of regression coefficients over all observations can be relaxed by incorporating generalized linear models into the finite mixture framework. The model class consisting of finite mixtures of generalized linear models is presented. Model identification is discussed given that difficulties might be encountered due to trivial and generic identifiability problems. These problems have already been observed for mixtures of distributions, but the extension to mixtures of regression models introduces additional identifiability problems. Details on model estimation are given and the application is illustrated on several examples.