Understanding sense of safety and trustworthiness of tourism information among migrant visitors

Purpose – The migration-tourism discourse has gained attention in global tourism and mobility among scholars because of the vast contributions of migrant visitors to various sectors of the host country’s economy, including the tourism and hospitality industry. However, few studies appear to have been undertaken on the subjectmatter, particularlywithin the developing country tourism context. The purpose of this researchwas to assess and understand migrant visitors’ sense of safety among five different nationalities (United Kingdom, USA, Germany, China and India) and their perceptions on trustworthiness of tourism information. Design/methodology/approach – Anchored on the national cultural dimension of Hofstede model, this study sampled 306 migrant visitors, with the use of one-way analysis of variance (ANOVA) and the Games– Howell post-hoc test to examine mean differences of nationalities, with regard to trust of tourism information and sense of safety. Finally, the multiple regression analysis was conducted to establish the significant effects of national culture characteristics and trustworthiness of tourism information on sense of safety of migrant visitors. Findings – The study finds that there were no significant differences in the sense of safety by the five nationalities. However, national culture characteristics (power distance and uncertainty avoidance) and trustworthiness of tourism information were significant predictors of sense of safety of migrant visitors. Research limitations/implications –This research adopted only two cultural dimensions of the Hofstede’s model, so future studies within the migration-tourism literature should consider adopting other cultural dimensions of the model. Practical implications –The research provides, first, insight into sense of safety, trustworthiness of tourism information and national culture characteristics which are relevant for destination marketers, the ministry of tourism and local tourism officials to promote safety tourism. Originality/value – The study expands the application of the cultural dimension of Hofstede model within the migration-tourism literature and establishes that national cultural characteristics and trustworthiness of tourism information are significant predictors of sense of safety of migrant visitors.


Introduction
Safety and security are important concerns for most travellers to foreign tourist destinations and have dominated research designs over the past few decades (Spencer and Tarlow, 2021;Wang et al., 2019) and could have an impact on travel decisions. Doubtless, over the years, the growth in international tourism and the events of 11th September 2001 and terrorist attacks on the USA, among others, have raised concerns about the safety and well-being of travellers and so on (Ghaderi et al., 2017;United Nations, 2015). Recent reports from UN World Tourism Organization (UNWTO, 2020) revealed a 22% decline in arrivals in the first three months of the year 2020, following the lockdown in many countries, translating into a loss of 67m international arrivals and about US$80bn in receipts (i.e. exports from tourism). Furthermore, the International Association of Tourists Safety (IATSS) reports on a lot of uncertainties about safety and security in tourism destinations (IATSS, 2020).
In this respect, reliability and trustworthiness of information become very essential among international and migrant visitors, especially given that tourism is information-based, complex and dynamic in nature (Trogisch and Fletcher, 2020;Zarezadeh et al., 2018). The purpose of this study therefore is to understand tourists' safety perspectives among different migrant visitors in Ghana and their perceptions on tourism information. Migrant tourists or visitors in Ghana comprises skilled, semi-skilled and unskilled individual who have different nationalities with diverse cultural characteristics, are identified in this study. Certainly, international tourists and migrant visitors are likely to place much emphasis on public safety and trustworthiness of information. However, extant studies on tourism reveal the lack of consensus as regards trustworthiness of travel information, and this theme continues to be topical and debatable among academics and practitioners within the tourism sector (Berhanu and Raj, 2020;Munar and Jacobsen, 2014). Furthermore, there have been calls for further studies for insights into migrant visitors' concerns, regarding political security, public safety, disaster prevention and getting access to reliable information for visitors (Preko, 2021). Significantly, Zou and Meng (2019) found differences in five dimensions of tourists' sense of safety, based on their actual experience and safety expectation among tourists, but was limited to only Chinese nationals. To a large extent, the tourism literature has focussed on certain nationalities and cultures (e.g. Batat and Prentovic, 2014;Farahani and Mohamed, 2011;Wang et al., 2019). There is therefore a gap in the tourism literature due to the limited studies on "sense of safety" from the perspectives of migrant visitors in developing country destinations, hence the direction of this study. In this study, "sense of safety" of migrant tourists is operationalized as tourists' level of satisfaction and confidence in safety measures within the host tourism environment that would lead to positive attitude towards the tourism service propositions.
Notably, the literature attests that tourist attitude towards safety and processing of information could differ, possibly due to differences in levels of personal development, personal experience, national culture characteristics and so on (Canavan, 2016;Farahani and Mohamed, 2011;Zeng and Gerritsen, 2014). Interestingly, migrant visitors usually maintained their values and cultural characteristics consistent with national cultures of their home country (Marcher et al., 2020;Okano and Naoi, 2020). Profoundly, there is scant literature on migrants' cultural backgrounds; migrants' perceptions on sense of safety; and the reliance and trust of information in developing economies.
The specific objective, therefore, of this study is to: (1) to explore differences in "sense of safety" among migrant tourists in Ghana; (2) assess tourists' perspectives on trustworthiness of tourism information (TTI); and (3) to understand the nexus of sense of safety, information reliability and national cultural characteristics (NCCs) of migrant tourists in Ghana. The following research questions (RQ) will therefore guide the study:

RQ1.
Are there any significant differences in "sense of safety" among migrant tourists of different nationalities?
RQ2. What are migrant tourists' perspectives on trustworthiness of tourism information?
RQ3. Are there significant relationships among "sense of safety," information reliability and NCCs?
This direction of study will expand the scope of tourism research, enhance understanding of tourists' perception of safety of destinations and thereby create host-guest relationships within the tourism market. After a review of the literature devoted to questions of sense of safety variables and reliability of tourism information and cultural characteristics, the methodology is described in greater detail. The results are then presented and discussed, and implications for future research, policy and other tourism stakeholders are submitted.

Literature review
Theoretical background This study is underpinned by Hofstede's model (Hofstede, 1980) on national culture, which established national cultures akin to different countries, based on a survey on employees of IBM, a US multinational, from 40 countries. The Hofstede framework (Hofstede, 1980) and subsequent adaptations (Hofstede, 1983;Hofstede et al., 2010) classified five dimensions as: power distance, individualism/collectivism, masculinity/femininity and uncertainty avoidance for national culture values and long/short-term orientation. Power distance highlights the extent to which the less powerful members of organizations within a country expect and accept how power is distributed. Second, collectivism versus individualism highlights whether members of a society are expected to "care for themselves" or to first "seek the welfare of the society". Femininity versus masculinity explores to ascertain whether prevailing values of a society are "masculine" (e.g. assertive and competitive) or if gender roles do overlap. Uncertainty avoidance finds whether members of a culture feel threatened by uncertain or unknown situations and how such situations are avoided. The last dimension -"long-term orientation versus short-term orientation", combines the values of persistence, thrift, ordering relationships by status. Longterm orientation focusses on future; while short-term focusses on the present or past and considers that more important than the future. As such, national cultures with short-term orientation value tradition hold on to current social hierarchy and fulfil all social obligations. Critics of the Hofstede framework, such as McSweeney (2002) argued that surveys are unsuitable ways for examining cultural differences of nations and that results of employees from one company could not be generalized to an entire nation's scores. Subsequently, Hofstede (2002) contended that the survey measured the differences between nations and not absolute numbers. Crotts (2004) also argues that perception-based assessment of cultural differences is subjective and ambiguous construct, and this position might not allow one to make inferences on tourist behaviours based entirely on national cultures. Despite the critiques, this study was guided by the Hofstede's framework model because, first, it is the most widely developed dimensions of national culture that allows differentiation among national cultures and found to be usefulness in the tourism literature, for assessing behavioural and destination intentions and so on (e.g. Filimonau and Perez, 2018;Li, 2012;Li and Lu, 2016). Second, earlier studies underpinned by Hofstede's model have shown strong influence of cultural values on key operational areas, in various disciplines such as marketing, consumer behaviour, consumer preference, brand choice and so on (Farahani and Mohamed, 2011;Koc, 2020;Reisinger and Crotts, 2010). To the best of our knowledge, no study has applied the Hofstede model to explore and understand trustworthiness of information and perceptions of safety akin to different migrant tourists in developing destination contexts. In this study, uncertainty avoidance, and power distance, were applied to investigate sense of safety and trustworthiness of tourism information in Understanding sense of safety of migrant visitors Ghana, from the perspectives of migrant visitors from Germany, India, USA, China and the United Kingdom. Specifically, the study was guided by insights on "uncertainty avoidance" and "power distance" cultural dimensions in the Hofstede framework. The reason being that "uncertainty avoidance" plays an important role in understanding travelling risks, and more important than other cultural dimensions in predicting cross-cultural behaviour (Alcantara-Pilar et al., 2017;Crotts, 2004). The "power distance" dimension has been strongly associated with adherence to safety rules and regulations in the tourism sector (Farahani and Mohamed, 2011), and therefore would be relevant for achieving the specific objective of this study. Furthermore, we maintain that sense of safety and trust of tourism information about a destination could be assessed and understood first on the basis of the impact of the culture of the country of origin of the migrant visitors in a destination country.

National cultures and tourists' behaviour
Remarkably, the literature concurs that national and cultural diversities ought to be preserved, protected and enjoyed, as the world has become a global village (Almuhrzi and Al-Azri, 2019). Notably, regional and national cultures have become a major content provider in tourism and have been of interest to regional trading schemes (Anastasiadou, 2011;Crotts, 2004;Gravari-Barbas et al., 2019;Richard, 2019). Further, the literature acknowledges that tourism services have become increasingly aligned with national/regional cultures, tourists' subcultures and organizational culture of tourism enterprises in host destinations (Crotts, 2004;Koc, 2020;Richard, 2019). Hence, this study focusses on migrants from the European Union (EU), the Association of Southeast Asian Nations (ASEAN) and North American Free Trade Agreement (NAFTA) because they represent well-established regional trading blocs which churn out large volumes of tourists globally (Anastasiadou, 2011;Anastasiadou and de Sausmarez, 2006;Wang et al., 2019). The literature attests that the behaviour and attitudes of nationals from these blocs could differ due to their peculiar national subcultures, sociocultural characteristics, experiences, which affect the level of client satisfaction, loyalty, and so on, with regard to service/product offering (Filimonau and Perez, 2018;Li and Lu, 2016;Tsiotsou, 2019;Zeng and Gerritsen, 2014). Peculiar cultures and subcultures could also influence their appreciation of public safety and information processing and how tourists behave in the host country. Significantly,  have argued that the role of national cultures in global tourism was under-researched, especially in the context of emerging tourist markets, creating a gap in literature which this study attempts to address. In this regard, our study focussed on migrant tourists, from the four trading blocs because they dominate the tourism market in Ghana.
The migration-tourism discourse has gained attention in global tourism and mobility because of the enormous contributions of migrant tourism to various sectors of the host country's economy (Marcher et al., 2020;Okano and Naoi, 2020;Preko, 2021). According to the United Nations report (UN, 2019), international migrants who searched for leisure, improved standard of living and so on numbered about 272m in 2018, representing 3.5% of the world population. Largely, their continuous stay in the host destination would depend on their safety and the extent to which tourism information is readily accessible and reliable.

Sense of safety, trustworthiness of tourism information
Tourist safety is a pertinent global issue affecting travellers and destinations (Trogisch and Fletcher, 2020;Zou and Meng, 2019). In recent times, there is substantial evidence regarding a decline in global tourism visitations due to the lack of safety and security in tourist destination (Ghaderi et al., 2017;UNWTO, 2020). A recent study on "sense of safety" revealed a strong tendency of low safety perceptions affecting German tourists' behaviour in Turkey and in Egypt destinations (Isaac and Velden, 2018) and among Dutch tourists in Sri Lankan markets (Isaac and Van der Bedem, 2020). A similar work amongst Chinese nationals in unfamiliar environments revealed concerns about safety and issues on reliability of information in the tourist destinations (Zou and Meng, 2019;Zou and Mawby, 2020). Most of such studies, however, are within different destination contexts, leaving a gap for further studies from the perspectives of migrant visitors in a developing country, such as Ghana.
Trust and reliability of products and services are important factors that cannot be overemphasized or disregarded in consumer market operations. Undisputed, trustworthiness of information is one of the fundamental elements that visitors take into consideration when planning to visit certain destinations (Berhanu and Raj, 2020;Hu and Jiang, 2014;Kang et al., 2019). Basically, travellers sought information on sightseeing cites, gastronomy, accommodation, traffic, shopping, amusement and so on, before making travelling decisions (Hu and Jiang, 2014). Therefore, provision of reliable information on these services and facilities is likely to create some form of comfort and calmness among foreign tourists who visited the destinations. Undoubtedly, the literature acknowledges that credibility and trust of tourism information are important variables that determine patronage intentions of international tourists (de la Hoz-Correa and Muñoz-Leiva, 2018;Ponnapureddy et al., 2017;Zarezadeh et al., 2018). Specifically, the study of Ponnapureddy, et al. (2017) on German tourists in Portugal found that general trust, hotel services and perceived usefulness of the brochure information were positively and significantly related to booking intentions. Other authors (e.g. Chiappa et al., 2015;Mieli and Zillinger, 2020;Zeng and Gerritsen, 2014) found that information content generated by direct consumers was more trusted than those from official tourism websites, tour operators, travel agents and mass media advertising. Furthermore, Berhanu and Raj (2020) established that visitors' perception with regard to trustworthiness of social media travel information sources was largely influenced by demographic characteristic (age, gender, education) of tourists. In Preko's (2021) study, however, gender had no relationship with migrant tourists' risk perceptions in the host country, and these findings give room for further insights into the influence of personal characteristics on trustworthiness of information, among different nationals and so on. This study is unique because it is the first, to the best of our ability, that investigates in much larger context, national cultural characteristics, information reliability and tourists' sense of safety, underpinned by Hofstede's model, within a developing country context.

Study respondents
This study employed quantitative methodology that focussed on gathering data from the first five nationalities (diaspora) identified as the large number of international migrant workers and was one of the inclusion criteria for sampling. The nationalities are: Chinese (70,000), Indians (10,000), Americans (3,000), British (2,200) and Germans (1,300) residing and working in Ghana (Ghana Immigration Statistics, 2021). In addition, the participants should be 18 years or above; should speak, write and understand English language; and should have visited at least any official tourism destination in Ghana. This study concentrated on TTI, NCC and sense of safety of foreign migrant tourists in Ghana because of the following reasons. First, international tourism is relevant to Ghana, because of the multi-dollar potential foreign tourists present to the country. A report showed that Ghana will annually earn US$8.3bn from estimated 4.3m international visitors year per year by 2027. Migration-tourism literature (Forsyth et al., 2012;Nghiêm-Ph u, 2016) evidenced that migrant tourists spent more on products and services than domestic tourists on their visits to tourism sites. Second, Ghana has signed and ratified the International Convention on Protection of the Rights of All Migrants Workers and Members of Their Families on the 7th September 2000. This protocol spelt out the level of safety to be provided to all international migrants and their families in the host countries. Third, Ghana is ranked the second safest country (1.796 GPI Score) among the Understanding sense of safety of migrant visitors ten countries listed on the African continent (Traveling Lifestyle, 2020). This ranking will enhance international migrants travelling to Ghana for work and tourism purposes. Interestingly, there are 217,556 foreign migrant workers living in Ghana, out of which the summation of the five nationalities provided above, the total population estimated stand at 86,500 migrant workers (UNPDES, 2019). Using Krejcie and Morgan (1970) sample size determination table suggested 382 participants for the current study. Data for this research were collected over 5 months in two major towns (Accra and Tema) with the highest number of migrant workers. The questionnaires were scripted from a paper-based to digital format supported by the opine software installed on tablets and smartphones, to enable complete adherence to all COVID-19 safety protocols. Multiple sampling approaches were deployed in the data gathering, so all the migrants were not positioned at one destination. Initially, the convenience sampling was employed to gather data, and then the migrants referred the enumerators to other relevant migrants, as in snowball sampling. After the snowball sampling was exhausted, the enumerators adopted the intercept approach to sample participants at public places such as churches, mosques, shopping centres, sport clubs, golf playing grounds, workplaces or organizations with high number of migrant workers. Specifically, this took place in Tema suburb, where majority of foreign companies are situated, given that the first national shipping port is located in the area. Earlier, permissions were granted for data collections in these intercepts. Again, the ethical issues, confidentiality and anonymity were adhered to accordingly. In this study, the researchers responded to the issues of non-response rate and attrition rate in adding 15% (51) to the estimated sample. This resulted into 433 samples. Approximately, 450 questionnaires were administered and 345 questionnaires were received, and 306 were useable, which constitute 68% response rate. In this study, the researchers enhanced the response rate by applying the following steps. The researchers notified the participants in advance and made follow-ups; the questionnaire was designed in simple English language and mode of delivery was the tablet data collection. These steps are recommended to improve response rate within the hospitality research context (see: Ali et al., 2020). In all, the topic of safety was relevant to the tourists, hence their readiness to participate in the study.

Survey instrument
The study developed a survey measurement based on previous literature. To measure "sense of safety", sixteen (16) items were adapted (Boakye, 2010;Pizam and Smith, 2000;Poku, 2016;Zou and Meng, 2019), six (6) items for TTI (Chiappa et al., 2015;Berhanu and Raj, 2020) and eight (8) items for NCC of migrant visitors were developed based on the Hofstede national culture which was guided by earlier study of (Farahani and Mohamed, 2011). Two tourism professors assessed content validity of the survey instrument. In all, the participants were presented with a total of thirty (30) items anchored on the five-point Likert scale from strongly disagree as "1" to strongly agree "5" while five demographics of participants were also considered in the questionnaire. A pilot test with 30 migrant tourists was conducted prior to administering the final instrument, to ensure mitigation of potential errors of clarity, relevancy, reliability and suitability of survey instrument. The final instrument was revised accordingly to the Ghanaian context and reversing coding was applied to two questions of NCC that were negatively formulated prior to the final analysis. Furthermore, the common method bias was conducted to check distortion of significant interaction effects in the data. Using the Harman single-factor method with the help of a principal component analysis all the 30 items simultaneously loaded, produced an acceptable total variance of 27% (<50%) (Podsakoff et al., 2003), signifying that common method bias is not an issue. Following the recommendation of earlier studies (Greenland, 2014;Rothman and Greenland, 1998), a sensitivity analysis was conducted to screen for the ultimate confounding variables, when utilizing cross-sectional data. According to Berhanu and Raj (2020), these specific potential confounders, such as gender, age, educational, status are associated with TTI and NCC. Thus, this study adjusted for these specific demographics by calculating the crude and ultimate regression coefficients.
Furthermore, this research used varied statistical methods to analyse data. First, the exploratory factor analysis was deployed to identify the underlying factor structure of the main constructs (Hair et al., 2010). Second, the main constructs of "sense of safety", TTI and NCC were validated using the confirmatory factor analyses (CFAs). Third, the analysis of variance (ANOVA) was utilized to find out whether the mean difference is observed utilizing nationalities as independent variable. Finally, the multiple regression analysis was performed to confirm the significant effects of NCC and TTI on "sense of safety" of migrant tourists.

Findings
The findings showed that 64.2% of the migrants were male and 82.2% were married. Further, majority of migrant visitors were aged 40 years and above (78.3%). Correspondingly, 77.4% of the visitors were graduates. The highest number of migrant visitors were from Asia (55.5%). To ensure fairness in assessing which tourism destinations the migrants have visited in Ghana, this study provided multiple response questions, which allow the participants to select more than one destination (Table 1). The results showed that the forts/ castles, beaches/rivers/waterfalls and botanical gardens were the three most visited destinations in context. Descriptive statistics of average showed sense of safety (3.70), trustworthiness (3.67), power distance (4.57) and uncertainty avoidance (4.08) (see Table 3). On average, this meant that first tourists agreed that they were satisfied and confident with the safety measures and trusted the tourism information presented within the host tourism environment. Second, majority of the migrant visitors agreed that they preferred visiting safety destination, follow the safety rules and regulations (power distance) and avoided visiting tourism destinations that were unsafe (uncertainty avoidance). In all, the findings on power distance and uncertainty avoidance signified the extent to which migrant visitors' distance and avoid visiting destinations sites that are unfamiliar or unsafe.
Principal axis factorial approach with varimax rotation method was employed to identify the underlying factor structure of the adapted constructs. Of the 30 items loaded, four factors were identified where factor I explained the 35.8% of variance, with 16 items related tourism environment and safety information, so it was labelled "sense of safety" (multidimensional factor). Factor II explained 23.25% of variance, with eight items related to tourists' culture information, so it was labelled "NCC" (multidimensional factor). Factor III explained 14.76% variance, with six items related to tourism information, so it was labelled "trustworthiness of tourism information" (unidimensional factor). The assumption in exploratory factor analysis were met (Kaiser-Meyer-Olkin (KMO 5 0.787; Bartlett's test of sphericity x 2 ¼ 2; 276:34; Sig: < 0:001Þ (Hair et al., 2010) signifying the adequacy in conducting a factor analysis.  Table 2). The CFA was used to assess the validity of the measures of the main constructs. The initial measurement model, comprising of one 16-item factor (sense of safety), one 8-item factor (NCC) and one 6-item factor (TTI) produced acceptable fit indices ðx 2 ¼ 304:27; df ¼ 56; CFI ¼ 0:95; GFI ¼ 0:94; NFI ¼ 0:95 and RMSEA ¼ 0:06Þ with a significant regression weights at 0.01. Since all the regression weights were significant at 0.001 level, the model was not modified. In Table 3, the AVE scores for all the constructs were higher than 0.50, and the CR values also were higher than 0.80 cutoff point establishing convergent valid. Discriminant validity was also tested to determine the latent instruments distinction from others. For discriminant validity to be established within the data set, the square of average variance extracted (AVEs) should be greater than the correlations between the instruments in a model (Fornell and Larcker, 1981). This can be seen in Table 3. Foregoing, the confounding analysis revealed significant cruel regression coefficient of NCC on sense of safety β ¼ 0:193; t ¼ 5:07; p < 0:05 and TTI on "sense of safety" β ¼ 0:771; t ¼ 20:16; p < 0:05. In the first stage, gender, age and education were regressed on NCC and the probability values obtained for gender ðp ¼ 0:36Þ and education ðp ¼ 0:85Þ were above the threshold p > 0:25 (Greenland, 2014) and were removed from the analysis. In second stage, the percentage change in cruel coefficient of NCC ðβ ¼ 0:481Þ and coefficient of age ðβ ¼ 0:689Þ of the participants revealed À30% which has met the acceptable threshold of ± > 10% (Greenland, 2014). This implies that age is an ultimate confounder. Next, the same procedures were conducted for TTI and the demographic (age, gender and education). Again, age was identified as an ultimate confounder with a percentage change in cruel coefficient of TTI ðβ ¼ 0:843Þ and age beta ðβ ¼ 0:321Þ produced 163%, which met the accepted cutoff point ± > 10%. In all, the age of the migrant visitors was found to be an ultimate confounder and was retained in the main regression analysis. Table 4 shows the results of the one-way ANOVA and the Games-Howell post-hoc test was conducted using "sense of safety" and TTI as the dependent variables and nationalities as the fixed factor. The Games-Howell post-hoc test was conducted to examine whether differences exist in (1) the way the five nationalities assess their "sense of safety" tourism and (2) the migrant visitors' assessment of TTI. Games-Howell test was chosen as it is considered the most powerful and accurate in cases where the size of the clusters is not equal (Field, 2013). The findings of the Games-Howell test showed that no significant differences existed between the five nationalities in their "sense of safety" and TTI. In all, it implies that "sense of safety" and TTI mean the same for migrant visitors of all the selected nationalities.
The multiple regression analysis was conducted to examine the effects of NCC and TTI including the confounder (age) on "sense of safety" (see Table 5). Results showed that TTI ðβ ¼ 0:625; p < 0:05Þ, NNC ðβ ¼ 0:104; p < 0:05Þ, age ðβ ¼ 0:095; p < 0:05Þ were found to be significant predictors of "sense of safety". Collectively, the three predictors explained 64.9% of the variance associated with sense of safety of migrant visitors. It is important to note that the Durbin-Watson test revealed 2.13 which was within the recommended threshold of 1.5-2.50 (Hair et al., 1998), implying that the assumption of the independence of errors was not violated. Finally, the correlation coefficients shown in Table 3 are not above 0.80, revealing that multicollinearity was not an issue (Hair et al., 2010).

Discussion of findings
Understanding risks perceived by foreign travellers is very important as those perceptions could determine travellers' behaviour (Bjork and Kauppinen-R€ ais€ anen, 2011). It is therefore essential for destination managers, marketers and researchers to appreciate tourists' perceptions on safety and trustworthiness of information among tourists of different nationalities. Hence, this study was heavily grounded on previous works of Berhanu and Raj (2020)

IHR
Mohamed (2011) that attested to the importance of understanding the impact of national culture orientation in tourism development. To this end, our study sought to assess differences in "sense of safety" among migrant tourists in Ghana; assess tourists' perspectives on trustworthiness/ reliability of tourism information; and understand the nexus of sense of safety, information reliability and national cultural characteristics, among migrant tourists from the USA, India, China and United Kingdom, from three trading blocs (EU, ASEAN, NAFTA). These trading blocs have shown significant contributions to the global tourism industry (Anastasiadou, 2011;Anastasiadou and de Sausmarez, 2006;Wang et al., 2019). The concept of "safety and security" has become an important phenomenon within the tourism sector, leading to advocacy from institutions such as UN calling for collective,   Table 3. Inter-factor correlation matrix Table 5. Regression analysis of migrants' sense of safety tourism Understanding sense of safety of migrant visitors multilateral frameworks to cover the environment, health and economic threats (Preko, 2021;United Nations, 2015). Particularly, during this health-crisis (COVID-19 pandemic), the issue of sense of safety information is relevant to tourists. For example, WHO (2020) has indicated that understanding the mitigation efforts to curb the COVID-19 pandemic ultimately falls on individual countries and destination sites to provide the appropriate public health, safety and health systems capacities, particularly at the destination's points of entry. Hence, this study assessed "sense of safety" among migrant tourists in the context of a developing country destination. Notably, this study revealed no significant difference in "sense of safety" among the migrant tourists of the four different national culture characteristics. Contrary, Zou and Meng (2019), in a study on Chinese domestic tourists in Chinese tourist sites, found significant differences in "sense of safety" of the visitors. Meanwhile, Isaac and Velden (2018) confirmed low safety perceptions among German tourists, and this affected their decisions to travel to Turkish and Egyptian tourism destinations. The inconsistencies of findings of these studies seem to support Crotts' (2004) argument that a perception-based study driven by national culture is subjective and ambiguous. Hence, "sense of safety" could be regarded as a basic requirement for tourists and could depend largely on individual expectations and type of destination under consideration and not entirely due to national cultures characteristics. In this regard, destination managers ought to focus on developing standardized safety measures that would be appreciated by tourists of different cultural persuasions. According to Choi et al. (2019), trustworthiness is a key factor for decision-making and enhances the successful running of businesses in the travel and hospitality sectors. Also, some research studies have shown that national culture of tourists is likely to influence their information search behaviour (e.g. Gursoy and Umbreit, 2004). In this study, the findings show no significant differences with regard to trustworthiness of information among the migrant visitors. This finding is inconsistent with Mieli and Zillinger's (2020) study that affirmed differences in information search and behaviour among tourists of different nationalities (Europe, Canada, USA, Brazil); albeit, Mieli and Zillinger (2020) did not specifically look at trust concerns among the nationals. In a similar cross-cultural study on processing of information among Spanish and British tourists, Alc antara-Pilar et al. (2017) found differences in perceived usefulness and tourists' attitude towards website information. Once tourists find tourism information to be useful, they are likely to have a positive feeling towards the destination under consideration. These findings mean that tourism information needs are generic and a fundamental requirement among tourists, irrespective of nationality, and ought to be given due attention in planning and promoting tourism markets, in order to attract visitors to tourism destinations.
Furthermore, this study revealed that trustworthiness of information had effect on "sense of safety" of the migrant visitors. In a similar study among international travellers to the Finnish tourism destinations, Bj€ ork and Kauppinen-R€ ais€ anens (2011) found that risk perceptions on information sources affected the perceived safety image of a whole country. This meant that international travellers would glean information from all available sources to ascertain the level of risk of the host destination before making travelling decisions. Again, this study finds that NCCs of migrant visitors influence migrant visitors' "sense of safety". This finding is consistent with the work of Batra (2008) which explored foreign tourists' perceptions on safety concerns of leisure travellers to Bangkok. The findings showed that personal factors, such as nationality, had an influence on how visitors perceived safety and security of host destinations. It is interesting to note that while our study found no significant differences in the level of perceived safety among the four nationals (USA, UK, India, China), Batra (2008) confirmed that European tourists expressed a lower sense of safety, compared with visitors from other regions (i.e. Asia, America, Africa). The differences could be attributed to confidence in the safety measures taken and the tourism site under consideration, since a specified destination could be perceived as being safe based on one particular pedestal, such as "food safety" in Singapore (Tarulevicz, 2018). Despite the differences among the nationals, the influence of national characteristics on "sense of safety" supports our proposition that irrespective of the nationality, safety concerns ought to be given due attention, to develop the tourism markets across the globe.
Theoretical implication. This study concurs with other authors that consumer attitudes and behaviour could differ based on differences in their NCCs and so on (Canavan, 2016;Farahani and Mohamed, 2011;Filimonau and Perez, 2018;Prentovic and Batat, 2014;Zeng and Gerritsen, 2014). However, most of these studies have limited focus, in terms of the number of nations under consideration. This study makes significant contribution to the body of knowledge in the tourism literature. First, the items in the literature used to measure "sense of safety" have considered mainly, safety concerns, safety information and facilities and equipment, the environment and regional culture (Preko, 2021;Zou and Meng, 2019). This study makes a significant contribution by expanding the multidimensionality and confirming items on "sense of safety" (including public safety measurements). This improves the measurement rigor for "sense of safety" and adequately defines the complex nature of safety and security, within the developing destination context. Second, this provides further insights, in a larger context, on information reliability and tourists' "sense of safety", across five different national cultures of migrant visitors, underpinned by Hofstede's model, within a developing tourism and hospitality context. Specifically, it focusses on application of "power distance" and "uncertainty avoidance", which is least explored in tourism cross-culture literature (Farahani and Mohamed, 2011).
Remarkably, this study extends the work of Canavan (2016), beyond the perspectives from the visitors of the same cultural background in a developed country context, as in the study on Chinese domestic tourists' which confirmed differences in their "sense of safety". While some studies have explored and assessed safety concerns from the perspectives of foreign visitors from just one nation (see: Bj€ ork and Kauppinen-R€ ais€ anens, 2011; Isaac and Velden, 2018; Isaac and Van der Beden, 2020), this study looked at visitors from five nationals in a single study. Interestingly, this study encompasses the perspectives of nationals from three different trading blocs, thus bringing new data from these blocs to the body of cross-cultural studies, and shows the significance of reliable information and safety for the promotion of tourism markets across the globe.
Managerial implications. Sense of safety could be regarded as basic requirement and expectation for all tourists irrespective of their nationalities and tourism destination. First, destination managers ought to focus on developing standardized safety measures that would be appreciated by tourists of different nationalities, bearing in mind the basic protective measures against COVID-19 pandemic suggested by World Health Organization (WHO, 2020). This includes hand hygiene, physical distancing, avoiding touching eyes, nose, mouth, among others. Scholars have widely acknowledged that safety tourism is regarded as the bedrock for tourism development (Chauhan, 2007;Neumayer, 2004;Preko, 2021). Invariably, a safe destination will serve as a marketing tool in promoting a destination to potential and current consumers of tourism services and goods. Second, the finding on the significant influence of TTI on sense of safety suggests to managers that credible and timely tourism information will aid tourists' travel decision-making and patronage intentions (de la Hoz-Correa and Muñoz-Leiva, 2018;Zarezadeh et al., 2018). This research supports these viewpoints and further recommends that managers should continue providing reliable travel information to tourists. Third, though the main aim of this research was not to identify the most frequent visited destinations in Ghana by the migrant visitors, the multiple response analysis suggested an interesting result which will inform managers, practitioners, the Ministry of Tourism Arts and Culture, Ghana Tourism Authorities and other tourism Understanding sense of safety of migrant visitors agencies to pay attention in developing awareness programs that will promote less visited destinations in context. In all, the findings of this study cannot be overlooked because significantly, this study responded to safety tourism which is key in the industry, though there are some limitations. First, this research findings were limited to the five countries from two continents out of the seven continents; therefore, no generalization could be made beyond this scope. Second, this study did not aim at examining duration of stay and frequency of visits of migrant tourists on "sense of safety" of tourism destinations in context, which could have provided more insight into the phenomenon under consideration. Third, this study specifically did not concentrate on sense of safety of a particular tourism destination site. Besides, this research used only two of cultural characteristics (power distance and uncertainty avoidance) of the Hofstede's model in determining the national cultures of respondents.
The above-stated limitations of this study provide opportunities for future studies. First, future studies could consider including other countries from other continents to advance the understanding of "sense of safety" tourism within or in similar geographical contexts. Second, it is recommended that future study examine specific destination sites with focus on on-site data collection after tourists have completed their tours. Since this study focussed on two of cultural characteristics (power distance and uncertainty avoidance) of the Hofstede's model, future studies within the tourism culture literature should consider adopting other national culture dimensions, such individualism, masculinity, of the Hofstede's framework and so on.