Use of the pathfinder network scaling to measure online customer reviews: A theme park study

• Author: Feifei Xu,Xiaowei Ji,Qun Ren
• Date: 01 September 2019
• DOI: http://doi.org/10.1002/jsc.2288
• Published date: 01 September 2019

RESEARCH ARTICLE

Use of the pathfinder network scaling to measure online

customer reviews: A theme park study

Qun Ren

1

| Feifei Xu

2

| Xiaowei Ji

3

1

School of Business Law and Communications,

Solent University, Southampton, Hampshire,

United Kingdom

2

Department of Communications, Journalism

and Marketing, International Business School,

University of Lincoln, Lincoln, United Kingdom

3

Department of Tourism and Marketing,

Department of Mathematics, Uppsala

University, Uppsala, Sweden

Correspondence

Feifei Xu, International Business School,

University of Lincoln, Lincoln LN5 7AT, United

Kingdom.

Email: fxu@lincoln.ac.uk

Abstract

We use pathfinder network scaling (PENETS) approach to measure and evaluate

theme park visitors' online reviews. PFNETS as an effective tool of big data analytics

can be used to identify unobserved meaningful interrelationships between concepts.

Although there are many research analyzing online reviews, this study is the first

attempt to use an analytical approach of PFNETS to explore online reviews in theme

park visitor experiences. The article collects 14,142 effective reviews of the world's

first Disneyland in California from TripAdvisor. Using parallel and similarity compari-

son in pathfinder scaling, four individually but fully connected networks were gener-

ated to reveal different visitors' experiences in different segments. The findings

indicate the dissimilarity of concept relatedness between different segments and rev-

ealed the knowledge gap of marketing to different segments in theme parks.

1|INTRODUCTION

As the increasing growth and wide use of social network sites (herein-

after, SNS) has become a global phenomenon, visitors' online reviews

(or digital texts) on the SNS has a ubiquitous influence on a potential

visitor's understanding to a theme park's service quality, relationship

with customers as well as making his/her purchase decision (Arenas-

Gaitan, Rondan-Cataluña, & Ramírez-Correa, 2013; Wang & Chang,

2013). It is by no means clear, more and more organizations in the

hospitality industry have regarded online reviews as the key source of

knowledge (Roblek, Pejic Bach, Meško, & Bertoncelj, 2013). Conse-

quently, there is a growing research interest on analyzing online

reviews to understand customers' needs, to build brand loyalty, and to

encourage customer engagement (Crosby, Evans, & Cowles, 1990;

Hennig-Thurau & Klee, 1997; Sánchez-Franco, Muñoz-Expósito, &

Villarejo-Ramos, 2017).

Tourism has been one of the areas that online reviews have been

explored extensively. For example, Xiang, Schwartz, Gerdes Jr, and

Uysal (2015) used reviews from TripAdsivor to study hotel staying

experiences. Geetha, Singha, and Sinha (2017) used reviews to

explore customers sentiment with hotels. However, previous research

on theme park experiences have mainly been relying on traditional

surveys and interviews (Chiappa, Ladu, Meleddu, & Pulina, 2013;

Geissler & Ruck, 2011; Pikkemaat & Schuckert, 2007), therefore, inev-

itably have their limitations from traditional research methods. There-

fore, it would be helpful to evaluate and gain insight from large

amount of unstructured online reviews from consumers (Amith et al.,

2017). Recognizing this research gap, the article hopes to contribute

to the studies on customer experiences in theme parks by analyzing

online reviews.

In the field of marketing, over the years, researchers have become

increasingly aware of the significance of understanding customer seg-

ments (Rau, Gao, & Ding, 2008). Segmentation plays an important role

in marketing as different groups of customers will have different

needs, motivation and behavior, therefore, shaping different experi-

ences (Xu, Morgan, & Moital, 2011). How to identify these experi-

ences that can be shared by a certain group becomes critical to

marketing. With the advent of Web 2.0, using a large set of user gen-

erated content (UGC) makes it possible in identifying unstructured

hidden patterns when exploring customer experiences.

Among many methods of big data analytical tools, pathfinder scal-

ing is very effective. It is used to set a fully connected network during

the data analysis, which indicates the importance of these terms and

JEL classification codes: M10, M20, M30.

DOI: 10.1002/jsc.2288

Strategic Change. 2019;28:333–344. wileyonlinelibrary.com/journal/jsc © 2019 John Wiley & Sons, Ltd. 333