Case Study as a Research Method in Hospitality and Tourism Research: A Systematic Literature Review (1974–2020)

Case Study as a Research Design

Case study method is used in a wide range of disciplines and fields in the social sciences to investigate several different, social, and political phenomena related to individuals, groups, and organizations (Yin, 2003). It can be designed through both inductive and deductive logics (Dooley, 2002), and they can be qualitative or quantitative (Eisenhardt, 1991). While inductive case studies aim to build theories, deductive case studies test theories (Eisenhardt & Graebner, 2007). The case study method is conducive to investigate complex and new phenomena in both building and testing theories (Eisenhardt, 1991; Tasci et al., 2019). In their purest form, case studies are limited by time and space, where researchers collect detailed information through a wide variety of data collection techniques for a certain period of time (Creswell, 2014).

Case studies can include a single case or multiple cases (Gustafsson, 2017). Yin (2014) further proposed four types of case study designs: single-case holistic designs, single-case embedded designs, multiple-case holistic designs, and multiple-case embedded designs. The single holistic design indicates the existence of only one case, while a single embedded design also has subcases which signal embedded units in the case of interest. The same situation applies to multiple holistic and multiple embedded designs, except that there is more than one case to be investigated (Hoorani et al., 2019). Both designs offer several advantages and disadvantages. The single case study is an appropriate design for critical, unusual, common, revelatory, or longitudinal cases (Yin, 2018). The arguments obtained from multiple case studies are usually treated as more compelling, and the overall research is thus accepted as being more robust (Yin, 2014).

According to Yin (2014), there are six types of data sources in case studies: documentation, archival records, interviews, direct observations, participant observation, and physical artifacts. Observation can be grouped into five categories: covert versus overt, participant versus nonparticipant, systematic versus unsystematic, observations of natural versus artificial situations, and self-observation versus the observation of others (Flick, 2009, p. 222). Observation is often used by researchers to collect data in cases when there is no attainable data from other methods (Noor, 2008). Interviews can be applied face-to-face, over the phone (Fontana & Frey, 2005), or even online. Documents and document analysis can be complementary to other sources, or they may even be the sole source (Flick, 2009). Notes, case reports, contracts, drafts, diaries, statistics, annual reports, letters, or expert opinions are regarded as some of the documents from which researchers or scholars can benefit (Wolff, 2004). Photography and films can be crucial parts of visual sociology as they deeply provide detailed recordings of facts while ensuring a more comprehensive and holistic perspective regarding lifestyles and circumstances (Flick, 2009). Overall, photographs, art objects, videotapes, website main pages, e-mails, text messages, social media text, or any forms of sound generate the forms of qualitative audio and visual materials (Creswell, 2014).

The Legitimacy of Case Study Method

The legitimacy of research (Miles & Huberman, 1994), including case study method, is described with different terms such as reliable, valid, dependable, reasonable, confirmable, credible, useful, compelling, significant, and empowering. Qualitative approaches in general are criticized for their lack of rigor and credibility, and thus they are usually blamed for missing the tenets of good science (Decrop, 1999). Similarly, along with the reliability (dependability) matter, the issue of transferability (external validity) is one of the major problems that may discourage journals’ editors and reviewers to embrace qualitative case studies as they may not be convinced that these studies add a great deal of theoretical or methodological contribution to the current knowledge (Nunkoo et al., 2013) and thus researchers and scholars are discouraged.

There are four criteria to judge the legitimacy of case studies: construct validity, internal validity, external validity, and reliability (Yin, 2009). Construct validity refers to triangulation, based on multiple sources of data used in research, while internal validity reflects pattern matching to compare findings with research objectives (Adeyinka-Ojo et al., 2014; Xiao & Smith, 2006). According to Stake (2005), case studies receive credibility by triangulating the descriptions and interpretations by benefiting from multiple sources of data. Triangulation can provide rigor in a case study while increasing validity (Hoorani et al., 2019).

In response to the issues presented above, there are several different methods to provide reliability of qualitative research, including case studies. Several quantitative techniques with statistical applications have been suggested to increase the trustworthiness of case study research such as Krippendorff Kappa or Cohen’s Cappa Analysis (Forman & Damschroder, 2007; Hayes & Krippendorff, 2007; Landis & Koch, 1977), using alpha (α) as a reliability co-efficient to measure agreement among observers or coders (Krippendorff, 2011), or employing data triangulation to make research more rigorous and credible to avoid measurement bias (Creswell, 1998; Creswell & Poth, 2018). Besides, researchers are suggested to apply several steps to establish legitimacy of case studies such as increasing sample size of the unit of analysis, employing more multiple case studies approach from comparative analysis perspective to gain deeper insights about the phenomenon under investigation and benefiting from data triangulation that involves different and independent sources to provide wider acceptance of qualitative tourism researches (Decrop, 1999).

Data Collection

Data were obtained from 10 top-ranked hospitality and tourism journals that are classified by Scimago Journal and Country Rank developed by Elsevier using the Scopus database. These journals publish the bulk of the research on case studies. The basic reason behind the choice of the Scimago Journal and Country Rank as a data source is twofold: (a) it is regarded as a size-independent indicator and (b) it classifies and ranks journals according to their average prestige for each article and can be used for journal comparisons in the scientific assessment process (Vuignier, 2016). The unit of analysis consists of the aforementioned journal articles published between 1974 and 2020. Sampling of articles was based on three main approaches (Veal, 2011, p. 295):

The major hospitality and tourism journals are thus the Annals of Tourism Research, Current Issues in Tourism, Journal of Hospitality Marketing & Management, Journal of Hospitality and Tourism Research, Journal of Sustainable Tourism, Journal of Travel Research, Journal of Travel & Tourism Marketing, International Journal of Contemporary Hospitality Management, International Journal of Hospitality Management, and Tourism Management. During the course of data collection, a keyword search approach was performed from predefined key terms that were selected as “the case of,” a case of,” “a case study, “the . . . case,” “a/an . . . case study,” “case study,” “the case . . . of,” “a . . . case of,” “the case for,” “a case for,” “case of . . .,” “a . . . case study of,” “the case study,” “the . . . case of,” “a case from . . .,” “a case study from,” “cases studies from,” “case study with,” “a case study on . . .,” “a . . . case study from,” and “a case on the . . .” by adapting the research approach of Tasci et al. (2019). As such, a keyword search was applied to the title of studies. Many inclusion and exclusion criteria were considered during the course of the database search (see Table 1). Book reviews, reports, research in brief, commentary papers, research letters, short communications, viewpoints, and discussion papers were excluded from review. Only full hospitality and tourism journal peer-reviewed articles in English were included in the analysis process.

OPEN IN VIEWER

Table 1.

Stages Employed for Systematic Literature Review (Adapted from Bichler, 2019).

Stages	Applications for systematic literature review	Author(s)
1	Only peer-reviewed and full articles in hospitality and tourism (1974–2020)	Chang & Katrichis (2016), Bichler (2019)
2	SSCI listed journals by Scimago Journal and Country Rank	Papamitsiou & Economides (2014)
3	Keywords, for example, “the case of,” a case of,” “a case study,” “the . . . case.”	Tasci et al. (2019)
4	Applying exclusion and inclusion criteria	Müller et al. (2020)
5	Analyzing data, for example, thematic analysis	Isaac & Eid (2018); Prayag & Ryan (2011)
6	Screening, refinement, and eligibility process of data	Pahlevan-Sharif et al. (2019)
7	Creating a catalog that presents all instruments and their key figures	Müller et al. (2020)

Data Analysis

The study adopted a systematic literature review that has more recently received greater attention of scholars and reviewers in the field of tourism (Khoo-Lattimore et al., 2019). The thematic analysis was conducted to examine the data at several distinct stages (see Table 1).

The thematic analysis of keywords in article titles revealed categories of articles at the initial stage. Articles were classified independently by two authors and separated into four groups and themes within each group after the coding process was finalized. Next, the classifications were compared (e.g., Qualitative, Quantitative, Conceptual, and Mixed-Methods) and a high level of agreement was provided by two authors on the themes that emerged. Any inconsistency was resolved by discussion and a few modifications to accommodate inconsistencies. A disagreement about six articles resulted in their reclassification into a new category (i.e., papers such as Research Notes, Research Letters, Short Communications, Viewpoints, and Discussion Papers labeled as “Short Papers”).

Leximancer was used for data analysis. This software differs from other analysis programs as it uses quantitative methods to implement qualitative analysis (Tseng et al., 2015). Along with the frequency of the concepts that mostly occur, it also creates thematic clustering to map complex themes derived from the data (Cheng & Edwards, 2019).

The primary aim of employing a systematic literature review is usually to provide information to the researcher mapping and evaluating the existing intellectual field while building a research question to expand the current knowledge beyond the field (Tranfield et al., 2003). It is often used by researchers to elicit all empirical evidence that complies with the predefined inclusion criteria in response to answering a specific research question or hypothesis (Snyder, 2019).

Thematic Analysis of Articles

This section includes the key themes of the articles identified using Leximancer software that resulted from a qualitative, thematic analysis. It also aims to identify the related issues in the articles covered in this systematic review. The review of the articles was categorized into six main themes after systematic review process finalized. These major categories are as follows: (a) articles by method, (b) articles by year, (c) articles by method and journals, (d) articles by subjects and contexts, and (e) data collection method and statistical techniques used.

Articles by Method

Figure 1 shows that the majority of studies (52%) employed a quantitative method (e.g., Axelsen & Swan, 2010; Brida et al., 2015; Zátori et al., 2019), followed by a qualitative method (e.g., Matarrita-Cascante et al., 2010; Ottenbacher & Harrington, 2013; Smith et al., 2016), the mixed-method approach including both quantitative and qualitative methods (e.g., Biran et al., 2014; Page et al., 2017), and conceptual approach (e.g., Callahan, 1998; Mycoo, 2006). This implies the dominance of the positivist paradigm for the majority of journals. Another reason can be attributed to the attitudes of journals’ reviewers and editors who are prone to reject conceptual and/or qualitative case studies as they believe they cannot make a significant theoretical or methodological contribution to the literature (Nunkoo et al., 2013).

OPEN IN VIEWER

Figure 1.

Percentages of articles by method.

Articles by Year

Figure 2 shows the number of published articles had a sharp increase between 2004 and 2008. Results also show a slight decrease in the number of articles published on the case studies between 2008 and 2010, which can be evaluated as a temporal or coincidental situation or it might be a periodic process where this method may have reached academic saturation.

OPEN IN VIEWER

Figure 2.

Publication development of case studies in hospitality research.

Articles by Method and Journals

Figure 3 shows that Tourism Management (TM) published the most case study articles (25.8%) followed by Journal of Travel and Tourism Marketing (JTTM) (12.5%). Taking into consideration articles by method and journal, JoST and ATR differ from others as they published qualitative case studies more than quantitative ones.

OPEN IN VIEWER

Figure 3.

Distribution of articles by method and journal.

Figure 4 reflects that a large number of case study articles were employed in Asia (n = 236), mostly published by TM, followed by Europe (n = 221). The least number of published articles that adopted a case study approach was in the Polar region.

OPEN IN VIEWER

Figure 4.

Distribution of articles by location and journal.

Articles by Subjects and Contexts

The major topic areas of case studies have consisted of a wide range of fields, such as destination marketing and management, economic development of destinations, dark tourism, structure and analysis of service and hotel industry, crisis management, motivation and experiences, cultural and heritage tourism, sustainable tourism, stakeholder management, tourism governance, tourism planning and policy, innovation, sharing economy and interrelated fields like peer-to-peer accommodation, Airbnb, satisfaction, service quality, authenticity, and tourism education (Table 3).

OPEN IN VIEWER

Table 3.

Word Frequency Used by Research Context.

Words	Counts
Tourism	44,669
Tourists	11,412
Local	9,791
Travel	7,213
Hotel	6,962
Service	6,895
Visitors	6,680
Quality	6,556
Economic	6,526
Satisfaction	6,461
Experience	6,338
Value	5,730
Image	5,162
Marketing	4,973
Residents	4,485

Data Collection Method and Statistical Techniques Used

Most of the qualitative case studies relied on the data triangulation approach, which refers to multiple data sources, and thus tourism researchers and scholars have tended to use observation, documentation, and archival records or published materials as secondary data to ensure the trustworthiness of the research. It was also reported in the analysis that a number of more recent qualitative studies have used photographs and movies as visual sources of data (e.g., Im & Chon, 2008; Önder et al., 2016; Song & Kim, 2016). On the other hand, the number of qualitative studies that employed participant observation was limited, while the use of the interview as preliminary data was prominent.

Some of the qualitative research benefited from published materials and archival records along with notes of field observations as the primary empirical evidence, while some studies used secondary sources, such as printed and online documents, as primary data sources. Online websites of travel, tourism, and hospitality organizations and travel blogs used by tourists were also used by many scholars in their research as a source of data (Jeong, 2004; Law & Ngai, 2005; Marine-Roig, 2014; Xiang et al., 2010). While tourists were mostly the unit of analysis in quantitative studies, in qualitative studies, the main tourism stakeholders, consisting of both public and private sector representatives (e.g., local government authorities, tourism organizations, travel agencies, owners of travel agencies/tour operators, etc.), were the samples in qualitative case studies.

Results show that in both quantitative and qualitative case studies, the majority of research predominantly tended to use systematic literature as a review type since 2010. Also, a wide variety of quantitative methods and several different statistical methods and analysis techniques were employed including analysis of variance (ANOVA) (Oviedo-García et al., 2016), exploratory factor analysis (EFA) (Isaac & Çakmak, 2014), factor analysis (Fuchs & Reichel, 2006), data envelopment analysis (DEA) (Wöber & Fesenmaier, 2004), descriptive statistics (Mohsin, 2007), paired sample test (Suen & Law, 2001), generalized least squares (GLS) (Salazar & Zhang, 2013), multiple regression analysis (Sinclair-Maragh & Gursoy, 2015), linear regression analysis (George, 2010), and shift share analysis (SSA) (Zuo & Huang, 2020). In studies using mixed method, more than one statistical technique was used at the same time (e.g., Biran et al., 2014; Tang-Taye & Standing, 2016).

For qualitative case studies, use of content analysis as a data treatment technique was most prevalent (e.g., Paget et al., 2010; Peters & Kallmuenzer, 2018) which is followed by thematic analysis (e.g., Marschall, 2017) and discourse analysis, respectively (see Burns & Cowlishaw, 2014; Cui et al., 2020). More interestingly, an increasing number of innovative data analysis techniques were recently used in quantitative case studies such as affective mapping and emotinality of texts (Martini & Buda, 2019), along with computer-assisted data analysis software (see Chan & Hawkins, 2010; Lacey et al., 2012; Li & Ryan, 2018; Scott et al., 2008).