An Overarching Investigation into the Classification & Derivation of Emotional Display Rules in Tourism

Published: 7 February 2025| Version 2 | DOI: 10.17632/h77zwyh6bv.2
Contributors:
Sabal Sahoo,

Description

This research endeavour aims to deliver a panoptic understanding of the categorisation of emotional display rules applied in tourism. Towards this objective, the investigators applied a mixed research method incorporating qualitative methodology and participant observations. Data was collected in the form of 18 semi-structured interviews and 5 field observations on nature-based guides in different parts of Australia. 376 codes representing emotional display rules were extracted from the transcripts and were analysed through domain analysis and semantic analysis. In the obscure phases of its reconstruction, field observation data and its theoretical implications played a key role and helped the researcher in building constructs between the emergent themes and resultant theory. The findings of this data analysis suggest that the boundaries between sources of display rules i.e. societal, occupational and organisational norms are vague. Also, there are no clear demarcations between implicitness and explicitness of display rules propagated within guiding sector. These result in guides exercising considerable amount of freedom in deriving their emotional display rules. However, in subsequent application to the service context, guides experience mutual exclusivity in their implementation role i.e. regarding them as either positive or negative display norms and executing them as integrating, differentiating or masking displays of emotions. Therefore, the subsequent fit and utilisation of emotional display rules in the frontline interaction becomes largely streamlined. The emotion-laden content of interviews are disguised emotional display rules used by guides in their frontline interactions. These exemplar quotes serve as codes in the attached data typology analysis, which are classified simultaneously on the basis of several classification criteria for emotional display rules. Obscurity and challenges might be observed in classifying them on the basis of their sources and propagation medium, showing overlaps between them repeatedly across the 376 parts of data reconstruction. However, the process of assigning them a type on the basis of their direction of implementation and fit to the purpose in context would be considerably easier as their subcategories are mutually exclusive. These repetitive patterns help the investigators in cumulatively gathering a grasp of the relationships between different types of display rules and different criteria themselves across a variety of guiding contexts. Their visualisation would facilitate them in building a broad classification structure of emotional display rules incorporating all the criteria and simultaneously informing the nuances of deriving them from a variety of sources and applying them in specific social interactions. These data sets were obtained over different timeframes to minimize biases towards recent incidents of environmental and health-related crisis and their impacts on the society and economy.

Files

Steps to reproduce

As part of a hybrid method of data collection, the 18 semi-structured interviews and 5 field observations were conducted across 7 states in Australia. The sampling of participants for the 18 semi-structured interviews was mixed - 60 per cent random and 40 per cent purposive. The questionnaire included open-ended enquiries about nature-based guides' frontline interactions with visitors, job roles and their service environment set-up. The transcripts for each of those interactions were obtained using Otter.ai, anonymized and then carefully disintegrated in search of emotion-laden exemplar quotes on QSR NVivo version 1.3. These verbal threads rich with affect-based content served as codes representing emotional display rules. Each of these 376 codes were then simultaneously rendered classification on the basis 4 criteria in the attached emotional display rules type analysis excel workbook. The participant observation was discreetly carried out on 5 guides, each in different National Park within Australia, where the principal investigator posed as a customer and made observations on their frontline interactions with visitors. The observation notes were centred around their body language, their emotional expressions and content of verbal communication in response to the events in the service environment and responses from the visitors. Those notes were separately reflected upon by the researcher after the tours, for alignment with emotional labour literature as well as new theoretical implications and significance in the excel workbook attached. In alignment with the majorly contextual nature of emotional display rules and resultant process of emotional labour, it is highly recommended to perform this exemplar approach of manual coding. The use of domain analysis and semantic analysis for this customized deconstruction and reconstruction of data is largely dependent and embedded in the context of the corresponding interaction thread. Therefore, the investigators suggest that the mixed-methodology in this piece of social interpretivism predominantly leans towards qualitative methods and elaborate field observations.

Institutions

Griffith University

Categories

Tourism, Environmental Recreation, Emotional Labor, Tour Guide Interpretation

Licence