This research was conducted in May and June 2024 using a qualitative research method with a contractual content analysis approach. The statistical population included professors and experts in new technologies and tourism from Semnan Province who were thoroughly familiar with health tourism. This included professors from Semnan University, Semnan University of Medical Sciences, Shahrud University of Medical Sciences, Shahrud Azad University, Payam Noor University of Semnan, Payam Noor University of Garmsar, and Payam Noor University of Majn (with a background in Tourism Management and IT Engineering). Additionally, professors from Shahrud University of Technology and Damghan University (with degrees in Computer Engineering and IT) were also included, as well as experts familiar with health tourism and modern marketing science from all travel and tourism agencies in Semnan Province. The sampling methods employed were purposive and snowball sampling.
The main criteria for selecting the sample members included a comprehensive knowledge of the IT industry, expertise in modern marketing and health tourism, and a master's degree or higher. Sampling continued until theoretical saturation was reached (
23). The saturation point is when no new topics emerge, and researchers, after reviewing all scenarios, determine that sufficient data are available to complete the information (
24).
The data collection tool was in-depth and semi-structured interviews, which were recorded and coded after obtaining permission from the interviewees. Initially, key ethical issues such as goals, outcomes, satisfaction, identity, relationships, confidentiality, and protection were shared and communicated with the interviewees. The interview questions focused on marketing through the metaverse and health tourism. The interviews began with demographic questions about the interviewee and continued with semi-structured questions related to marketing through the metaverse and health tourism.
After a thorough review of the literature on "metaverse technology" and "health tourism," the interview questions were designed and developed in collaboration with experts, including professors and specialists in new technologies and tourism who were thoroughly familiar with health tourism.
The reliability of the questions was assessed using two methods. In the first method, the research results were shared with experts, and their approval was obtained. In the second method, to calculate test-retest reliability, three interviews were selected, and each was coded by the researcher at 15-day intervals. The reliability between the two coders for the conducted interviews was calculated using the following formula:
Reliability = (Number of agreed codes/between two coders total number of codes) × 100%
This resulted in a reliability score of 0.77. Since the reliability was above 60%, the coding trustworthiness was confirmed, and it can be concluded that the current interview analysis has suitable reliability. The semi-structured interview questions included:
(1) What are the advantages of using metaverse marketing in the health tourism industry in military and civilian hospitals?
(2) What are the benefits of employing metaverse technology specialists in health tourism marketing?
(3) What opportunities can metaverse marketing offer in developing health tourism services in military and civilian hospitals?
(4) What challenges can metaverse marketing pose in developing health tourism services in military and civilian hospitals?
(5) How can metaverse technology in the health tourism marketing industry reduce the high costs of advertising and related expenses?
(6) What factors influence metaverse marketing in health tourism in military and civilian hospitals?
(7) How can the benefits and hospital services within the country be introduced to foreign health tourists using metaverse technology?
(8) How can the advantages of using metaverse technology be communicated to travel and tourism companies to attract health service seekers?
All interviews were conducted in person at the workplace. Each interview lasted about 45 minutes. At the end of each interview, after thanking the interviewees, they were asked to elaborate on any remaining topics. In total, eleven interviews were conducted.
In addition to the interviews, published texts and articles on "metaverse technology," "health tourism," and "marketing" from various scientific databases were reviewed and analyzed. The data from the semi-structured interviews were analyzed using the content analysis method of Graneheim and Lundman (
25).
For data analysis, each interview was recorded, transcribed, and listened to by the researcher. The entire text was transcribed line by line. After transcribing the interviews and conducting multiple reviews and comparisons, open codes were extracted and categorized. Irrelevant information was removed, and the final categories and dimensions were refined. The criteria for categorizing topics were based on the main stakeholders in the metaverse marketing process in the health tourism sector.
The main categories were identified, weighted, and ranked using the Shannon entropy method in the quantitative phase. According to experts, the Shannon entropy method is considered more robust and valid for data analysis. In the Shannon entropy method, categories are first counted based on their frequency. Then, based on the frequency table, the following steps are performed:
(1) Normalizing the frequency matrix using the formula:
Where:
F is the category frequency, p is the normalized frequency matrix, i is the interviewee number, n is the number of categories, m is the number of interviewees, and, j is the category number., Calculating the information load of each category and placing it in the relevant column using the formula:
is the information load, p is the normalized frequency matrix, i is the interviewee number, m is the number of interviewees, j is the category number, and ln is the natural logarithm. Using the information load of the indicators, the importance coefficient of each indicator is calculated. The higher the information load of an indicator, the higher its importance coefficient :
Where
is the importance coefficient of the category, and
is the information load. It should be noted that in the calculation of
,
values that are zero are replaced with a minimal value of 0.00001 to avoid errors and infinite results in mathematical calculations.
is an index that indicates the importance coefficient of each category in each message based on the interview format. Moreover, based on the W vector, the categories resulting from the message are also ranked (
26).
This study is derived from a research project approved by the Research and Technology Deputy of Semnan University of Medical Sciences, with the ethics code
IR.SEMUMS.REC.1403.021 in 2024.