2016
DOI: 10.1016/j.annals.2015.10.006
|View full text |Cite
|
Sign up to set email alerts
|

Guidelines for treating unobserved heterogeneity in tourism research: A comment on Marques and Reis (2015)

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
30
0

Year Published

2016
2016
2022
2022

Publication Types

Select...
5
4

Relationship

1
8

Authors

Journals

citations
Cited by 48 publications
(31 citation statements)
references
References 40 publications
1
30
0
Order By: Relevance
“…We start by applying the FIMIX-PLS algorithm to narrow the range of statistically wellfitting segments. The FIMIX-PLS algorithm was executed 10 times for g = 2-5 segments, using the Akaike Information Criterion (AIC), Modified AIC with Factor 3 (AIC3), Bayesian Information Criterion (BIC), Consistent AIC (CAIC), Hannan-Quinn Criterion (HQ), and the normed Entropy Statistic (EN) as indicators to identify the appropriate segmentation solution (Sharma & Kim, 2012;Sarstedt, Ringle, & Gudergan, 2016). According to , the appropriate number of segments depends on a joint evaluation of the CAIC and AIC 3 indicators.…”
Section: Unobserved Heterogeneity and Subgroup Analysismentioning
confidence: 99%
“…We start by applying the FIMIX-PLS algorithm to narrow the range of statistically wellfitting segments. The FIMIX-PLS algorithm was executed 10 times for g = 2-5 segments, using the Akaike Information Criterion (AIC), Modified AIC with Factor 3 (AIC3), Bayesian Information Criterion (BIC), Consistent AIC (CAIC), Hannan-Quinn Criterion (HQ), and the normed Entropy Statistic (EN) as indicators to identify the appropriate segmentation solution (Sharma & Kim, 2012;Sarstedt, Ringle, & Gudergan, 2016). According to , the appropriate number of segments depends on a joint evaluation of the CAIC and AIC 3 indicators.…”
Section: Unobserved Heterogeneity and Subgroup Analysismentioning
confidence: 99%
“…The described segments should be interpreted in terms of observable and practically meaningful variables. To do so, we have conducted a post-hoc analysis with 5 explanatory variables (degree of co-creation, gender, place of origin, type of museum and age) to explain the posterior probabilities membership (Hahn, Johnson, Herrmann, & Huber, 2002;Sarstedt, 2008;Sarstedt, Ringle, & Gudergan, 2016). The 4 items of co-creation were averaged as their reliability was very high (ρ = 0.81; AVE = 0.54) and all factor loadings gave high values in standardised terms and significantly different from zero.…”
Section: Model-based Clustering Procedures In the Context Of Simultanmentioning
confidence: 99%
“…Thus, further research could conduct an extended moderator analysis examining in greater detail different properties of attitude strength. In a similar vein, research could explore whether other sources of (un)observed heterogeneity characterize the strength of the quality–delight–loyalty and quality–satisfaction–loyalty systems (Sarstedt, Ringle, and Gudergan 2016; Ratzmann, Gudergan, and Bouncken, 2016). Fourth, further research could assess the role of nonlinear relationships.…”
Section: Implications and Conclusionmentioning
confidence: 99%