The influence of theocratic rule and political turmoil on tourists’ length of stay

Abstract

The average tourist’s length of stay (LOS) is in a global decline. This downward trend underscores the need to study the factors that affect this variable in order to enable more effective management and marketing. This paper moves beyond the literature by investigating the determinants of international tourists’ LOS in a Muslim destination amid political crisis. LOS was evaluated using a survival analysis approach with data from 726 international tourists in Tabriz, Iran to ascertain the significant factors influencing trip length. The results reveal that the determinant factors are as follows: socio-demographic profiles, trip characteristics, and destination attributes. In addition, political turmoil and religious regulations are pivotal factors in LOS. The empirical findings provide valuable theoretical contributions to researchers and actionable guidance to tourism managers and marketers.

Keywords

Destination image Islamic destination length of stay political turmoil survival analysis socio-demographic profiles trip attributes Tabriz

Introduction

A tourist’s length of stay (LOS), also known as trip duration, is a critical factor in the financial success of the tourism industry. Evidence has shown that increases in LOS are associated with higher levels of tourism expenditure (Barros and Machado, 2010); this means that LOS has an effect on tourism-generated income (Barros and Machado, 2010; Barros et al., 2008; Martinez-Garcia and Raya, 2008). For instance, when the duration of stay at a hotel (or other types of accommodations) rises, the fixed costs of the hotel drop relative to revenue while employment rates increase, meaning that hotels are able to increase their profits (Barros and Machado, 2010; Jacobsen et al., 2018; Peypoch et al., 2012). Additionally, LOS affects tourists’ activities and behavior. Davies and Mangan (1992) demonstrated that prolonged stays not only increase the participation of tourists in various activities but also lead to greater feelings of satisfaction.

In recent years, the tourism industry has faced a global decline in LOS—travelers now prefer to travel more but stay for shorter periods of time (Adongo et al., 2017; Aguiló et al., 2017; Barros and Machado, 2010; Gokovali et al., 2007; Jacobsen et al., 2018; Salmasi et al., 2012). The increased prominence of business trips and the emergence of low-cost airlines have exacerbated the trend of short-term stays (Barros and Machado, 2010). These shifts in the industry have resulted in a considerable decline in tourism revenue (Alén et al., 2014); despite an increased number of incoming tourists, tourism-generated income has diminished, largely due to reductions in LOS (Ferrer-Rosell et al., 2014).

LOS is considered a crucial factor in tourism research and both destination (Barros et al., 2010; Gokovali et al., 2007; Martínez-García and Raya, 2008) and hospitality management (Barros and Machado, 2010; Peypoch et al., 2012). The significance of this factor in tourism research is merited since LOS is one of the most important variables in visitors’ decision-making processes (Decrop and Snelders, 2004). Given the importance of the duration of visits for the industry, it is crucial to determine the factors that influence LOS. A comprehensive understanding of its determinants would provide planners and managers with the proper tools to design effective marketing strategies and lure in visitors who show a greater predisposition to prolonged stays. The determinant factors of LOS are integral to efficient planning for the sustainable development of tourist destinations.

The significance of LOS has recently become a central focus in tourism economics research. The steady growth of publications on this topic began in 2008 (Alegre et al., 2011; Ferrer-Rosell et al., 2014; Jackman et al., 2020; Martinez-Garcia and Raya, 2008; Peypoch et al., 2012; Rodríguez et al., 2018). The literature on tourist LOS suggests that various factors, such as tourist profile, trip characteristics, and destination attributes, can be influential (Gössling et al., 2018; Rodríguez et al., 2018). Nonetheless, a deeper look at previous studies indicates a number of gaps in the extant research. Despite the great importance of this variable for destinations, less work has been conducted in less developed countries. As far as we know, up until now, no previous research has investigated tourists’ LOS in Islamic destinations. This disparity has prompted calls for more studies in tourist destinations throughout the Muslim world to analyze different antecedents of tourists’ behavior from the length-of-stay perspective.

According to Alén et al. (2014) and Rodríguez et al. (2018), different destinations often see different behavior; LOS appears to vary by geographical area and/or tourist segment (Aguilar and Díaz, 2019). Therefore, any information provided by studies conducted in non-Muslim destinations may lead to inappropriate or suboptimal decisions by tourism managers and marketers in Muslim destinations.

A tourist’s destination choice and level of engagement with the host community are often affected by cultural norms (Brown and Osman, 2017). Religion, as a significant component of culture, tends to influence local hospitality and legal guidelines; its many ramifications for various aspects of tourism have garnered interest from researchers. However, there is a dearth of research on how theocratic rule or religious regulations affect tourist behavior. It is perceptible that the laws and regulations in Islamic destinations can influence tourists’ LOS, which is an important dimension of tourist behavior—ignoring this relationship could bias parameter estimates in an unknown direction.

Moreover, despite the well-established fact that political turmoil and instability influence tourism demand and tourist behavior (Lanouar and Goaied, 2019), no previous studies have investigated their effect on LOS, one of the most fundamental components of tourism demand. There is evidence that sociopolitical unrest disrupts tourists’ decision-making process (Seddighi et al., 2001); thus, it is reasonable to consider the fact that it could influence their LOS.

This study incorporates variables that have previously been identified in research as predictors of tourist behavior. With the aforementioned research gap in mind, this study aims to breathe new air into the literature by exploring how political turmoil and theocratic rule in Muslim destinations affect tourists’ LOS.

This hypothesized relationship is examined in Tabriz, Iran. Our choice of Tabriz as a case study was motivated by Alén et al. (2014), who called for greater emphasis on diverse destinations. Additionally, it pairs well with our hypothesis. Tabriz was chosen as the ‘2018 Capital of Islamic Tourism’ by the Organization of Islamic Cooperation, an intergovernmental organization of 56 Islamic nations. However, the sanctions imposed on Iran by the US have seriously hindered the country; its oil-oriented economy has been harshly affected by sanctions against crude oil exports. As a result, public dissatisfaction has dramatically increased. This country has been swept by social upheaval and domestic instability on account of popular street protests in dozens of cities, including Tehran, Mashhad, and Tabriz (Takeyh and Maloney, 2011).

Tabriz is located in northeastern Iran, in the East-Azerbaijan Province. It is home to many heritage assets, some of which are thought to date back more than 2,500 years (Gannon et al., 2020). The city also has a long history of traditional manufacturing and cottage industries—most notably, its internally respected carpet and craft industry—which are often considered to be protected intangible heritage assets (Light et al., 2013). According to the Iranian Tourism organization’s report (2017), the average LOS of inbound tourists in Tabriz was decreased to 8.7 days in 2016 from 9.2 days in 2000. The short LOS is certainly a negative sign regarding the tourism-generated income (Barros and Machado, 2010), which considerers a main concern for local government, destination and hospitality management. This Muslim destination, given its international appeal and simultaneous political unrest, provides an excellent opportunity to explore the effects of Islamic regulations and political unrest on tourist behavior, especially in terms of LOS.

This study, to the best of our knowledge, is the first to evaluate the effect of Islamic regulations and political unrest on tourists’ LOS in Muslim destinations. It provides new insights to marketers and managers that can help them formulate operational strategies aimed at increasing LOS.

The remainder of this article is organized as follows. The second section reviews previous studies on tourists’ LOS. The third section details our theoretical lens and research method. The fourth section presents the experimental results of our research. Finally, the last section reviews our key findings, discusses their policy implications, examines the limitations of this study, and comments on the potential for future research.

Literature review

Although we can trace studies on the general concept of LOS to the 1970s (Archer and Shea, 1975; Mak and Moncur, 1979; Mak et al., 1977), there was little research on tourism-related LOS before 2006 (Rodríguez et al., 2018). The number of publications in this area has steadily increased since 2008, and scholars have identified various factors that influence LOS in a destination.

Factors influencing LOS have received extensive attention from academics (Aguilar and Díaz, 2019). Previous studies have been heterogeneous in scope (i.e., different geographic areas and tourist segments) and in their variables and methodology. For example, regarding the different geographical regions, we can find articles evaluating LOS in various destinations, such as Latin America (Barros et al., 2008), the Azores (Menezes et al., 2008), Madeira (Barros and Machado, 2010; Machado, 2010), Madagascar (Peypoch et al., 2012), Brazil (De Oliveira Santos et al., 2015), Santiago de Compostela (Rodríguez et al., 2018), Spain (Aguilar and Díaz, 2019) or Barbados, North America (Jackman et al., 2020).

In respect to the tourist segment researchers focus on low-cost tourism (Martínez-García and Raya, 2008), golf tourism (Barros et al., 2010), inbound tourism (De Oliveira Santos et al., 2015; Ferrer-Rosell et al., 2014), cultural tourism (Brida et al., 2013), senior tourism (Alén et al., 2014; Fleischer and Pizam, 2002), student tourism (Thrane, 2016), leisure tourism (Jackman et al., 2020), or international tourist (Aguilar and Díaz, 2019).

In addition, we can find various methodological approaches within extant literature, for example, Tobit model (Fleischer and Pizam, 2002), OLS regression model (Thrane, 2016); Binomial logit model (Alegre and Pou, 2006), Ordered logit model (Ferrer-Rosell et al., 2014; Yang et al., 2011), Count quantile regression (Salmasi et al., 2012), Negative binomial model (Alén et al., 2014), a truncated OLS regression, framed in a Heckman selection model (Rodríguez et al., 2018), or Survival analysis (Aguilar and Díaz, 2019; Barros et al., 2010; Gokovali et al., 2007; Wang et al., 2012).

Finally, in a review of factors that directly or indirectly impact LOS, Rodríguez et al., (2018) indicated that, in terms of the choice of variables included in the analysis, the variation is substantial and we can classify these independent variables in three main categories: Tourist profile (age, gender, education, and income), trip characteristics (travel purpose, travel cost, means of transport, repeat visitation, physical distance), and destination attributes variables (quality of the services of the destination, type of accommodation, climate, price).

In Table 1 we present a summary of some recent studies from 2006 to 2020. The table provides a brief description of these papers in terms of region of focus, methodological approach, and principal variables.

Table 1.

Studies on tourists’ LOS.

Reference	Geographic area/tourist segment	Methodology	Conclusions: LOS determinants
Alegre and Pou (2006)	Balearic Islands (Spain)/British and German sun and sand tourists	Logit model	Labor status, nationality, accommodation type, number of yearly trips, and repeat visitation rate .
Gokovali et al. (2007)	Bodrum (Turkey)/No specific segment	Duration models or Survival analysis	Nationality, education, income, previous experience, party size, familiarity and daily spending.
Barros et al. (2008)	Latin America/Portuguese tourists (charter flights)	Duration models or Survival analysis	Economic budget, age, destination attributes (culture, climate and security), destination image.
Martinez-Garcia and Raya (2008)	Catalonia (Spain)/Low-cost travelers	Duration models or Survival analysis	Nationality, age, education, type of accommodation, the season of trip and the geographical area of destination.
Menezes et al. (2008)	Azores (Portugal)/sun-and-sand tourism	Duration models or Survival analysis	Nationality (geographical distance), gender, age, education, type of flight, previous experience, geographical region of destination (or destination ascendancy), destination image, and number of islands visited.
Barros and Machado (2010)	Madeira Island (Portugal)/Sun and sea tourism	Duration models or Survival analysis	Age, nationality, gender, education, hotel quality, and expenditure.
Barros et al. (2010)	Algarve (Portugal)/golf tourism	Duration models or Survival analysis	Age, education level, nationality, trip main reason, type of hotel, and destination characteristics (climate, event and hospitality).
Menezes and Moniz (2011)	Azores (Portugal)/no specific segment	Duration models or Survival analysis	Repeat visitors, trip main reason, type of flights.
Peypoch et al. (2012)	Madagascar (Africa)/no specific segment	Duration models or Survival analysis	Gender, age, income, characteristics of destination (desirable foods, destination sea and nature, weather, security, life style and appearance characteristics of destination)
Thrane (2012)	No specific destination/Scandinavian students	OLS and duration models	Internet booking, trip season, daily expenditure, and planned trip.
Wang et al. (2012)	Dalian (China)/no specific segment	Duration models or Survival analysis	Age, education level, nationality, repeat visitors, and expenditure per tourist per day.
Brida et al. (2013)	Bolzano (Italy)/Cultural tourism	Count data (zero truncated negative binomial model)	Nationality, age, income, travel costs, and destination weather.
Alén et al. (2014)	Spain/senior tourism	Count data (zero truncated negative binomial model)	Age, VFR motivation, climate, activities (such as shopping and sports activities), type of accommodation, and traveling alone.
Ferrer-Rosell et al. (2014)	Spain/no specific segment	Ordered logit model	Age, nationality, education, income, trip characteristics (season of the trip, climate, travel accompany), type of accommodation, and type of destination.
De Oliveira-Santos et al. (2015)	Brazil/no specific segment	Duration models or Survival analysis	Gender, age, education level, nationality, multi-destination trip, motivations, type of accommodation, trips by air, party size, previous visits, travel season, expenditure, and type of destination.
Thrane (2016)	No specific destination/Norwegian student tourists	Survival analysis, OLS, and zero truncated negative binomial models	Daily travel costs, booking time, tourism and trip motives, trip month, and gender.
Jacobsen et al. (2018)	Norway/ no specific segment	Binary logistic regression	Time for desired activities, number of vacation days, overall trip budget, accommodation.
Rodrígueza et al. (2018)	Santiago de Compostela (Spain)/ no specific segment	Heckman selection model and separate probit and zero-truncated OLS regression	Age, travel main motivation.
Soler et al. (2018)	Malaga (Spain)/no specific segment	Negative binomial model	Domestic tourists’ LOS determinants: trip main reason, marital status, type of accommodation, mode of transport, travel party. International tourists’ LOS determinants: Satisfaction, marital status, employment status, destination loyalty programs, type of accommodation, mode of transport.
Aguilar and Díaz (2019)	Spain/no specific segment	Duration models or Survival analysis	Trip main motivation, destination, age, travel season, type of accommodation, time and budgetary constraints, travel distance, Party size and composition.
Jackman et al. (2020)	Barbados (North America)/Leisure tourism	Poisson model (count data model)	geographic distance, cultural distance, climatic distance, and economic distance.

The models analyzing LOS determinants are based on the consumer-behavior theory (Rodríguez et al., 2018). According to this theory, consumers (in this case, tourists) must estimate the type of vacation and the LOS they both prefer and can afford when they plan a vacation (Brida et al., 2013; Rodríguez et al., 2018). In the estimated models, the demand function of LOS is primarily determined by personal characteristics (Martínez-García and Raya, 2008) alongside the vacation type and destination (Aguilar and Díaz, 2019; Rodríguez et al., 2018). From an empirical perspective, there is no theoretical guidance for selecting the specific factors for each case. Consequently, the covariates used to explain LOS in this study are based upon the literature specified above and were categorized into three groups: personal factors, such as age (e.g. Barros and Machado, 2010; De Oliveira Santos et al., 2015; Salmasi et al., 2012; Yang et al., 2011), gender (e.g. Martinez-Garcia & Raya, 2088; Peypoch et al., 2012), educational level (e.g. Barros and Machado, 2010; Barros et al., 2010; De Oliveira Santos et al., 2015), and income (e.g. Barros and Machado, 2010; Ferrer-Rosell et al., 2014; Gokovali et al., 2007); trip attributes, such as purpose of travel (e.g. Adongo et al., 2017; Correia et al., 2017; De Oliveira Santos et al., 2015), frequency of travel (e.g. Alegre and Pou, 2006; Alegre et al., 2011; Gokovali et al., 2007; Thrane and Farstad 2012), geographic distance (e.g. Jackman et al., 2020; Yang et al., 2011), and size of travel party (e.g. Alegre and Pou, 2006; Correia et al., 2017; Gokovali et al., 2007; Salmasi et al., 2012); destination attributes, such as daily spending in destination (e.g. Alegre and Pou 2006; Alegre et al., 2011; Thrane, 2016), type of accommodation (e.g. Aguilar and Díaz, 2019; Salmasi et al., 2012; Soler et al., 2018), number of attractions (e.g. Adongo et al., 2017; Ferrer-Rosell et al., 2014; Rodríguez et al., 2018), destination image (e.g. Jacobsen et al., 2018; Machado, 2010; Menezes and Moniz, 2013).

Based on the preceding discussion, the following hypotheses are proposed:

H₁: Tourists’ socio-demographic characteristics determine LOS.

H₂; Trip attributes determine LOS.

H₃: Destination attributes determine LOS.

A priori, two additional factors could affect tourist behavior in our selected case study: Islamic regulations and political unrest. Both of these have been recognized in the literature as predictors of tourist behavior. Therefore, our study moves beyond the extant literature and can be considered an advancement in the literature on LOS and Islamic destinations.

As mentioned in the introduction, religion is a staple component of culture. Previous research indicates that the cultural norms of a destination significantly influence tourist behaviors, as travel is uniquely linked to social context (Battour et al., 2011). This relationship has gained a degree of momentum in the tourism literature. In this stream of research, several scholars have attempted to thoroughly investigate the link between Islam, the world’s second-largest religion, and tourism (e.g. Brown and Osman, 2017; Hino, 2011). However, some aspects of this relationship have been relatively neglected (Battour et al., 2011; Zamani-Farahani and Henderson, 2010); various aspects of Islamic tourism have yet to receive the same level of attention. Some authors focus on the definition of Islamic tourism (e.g. Henderson, 2009; Zamani-Farahani and Henderson, 2010), Muslim tourists’ needs (e.g. Henderson, 2009), and tourists’ motivations (e.g. Weidenfeld, 2006). However, there is a clear shortage of work on the impact of a destination’s religion on tourism demand. Undoubtedly, the rules in some Islamic destination (e.g. dietary laws, strict dress codes, hotel restrictions for unmarried couples), which impose severe restrictions on tourists’ activities, can influence various aspects of their behavior and travel decisions, such as their LOS. Consequently, we propose the following hypothesis:

H₄: Islamic regulations and laws determine LOS.

Several scholars point out that tourism is a highly vulnerable industry that is susceptible to exogenous factors (Seddighi et al., 2001; Sonmez, 1998; Sonmez and Graefe, 1998). One major exogenous factor that can have lingering effects on the industry is political instability—it can jeopardize a destination’s entire tourism sector. Recently, political instability and social unrest have engulfed many countries, developing countries in particular; this has reignited scholars’ interest in the impact of political upheaval on tourists’ decision-making process (Farmaki et al., 2019). This stream of research has uncovered many examples of places where political instability led to a decline in the tourism industry or a complete disappearance from the tourism map (e.g. Butler and Suntikul, 2017; Hall, 2010; Sonmez, 1998). Such instability can disrupt tourists’ decision-making process and deter them from traveling to specific destinations (Seddighi et al., 2001).

The argument that political unrest always has negative effects on tourism demand, however, is not universally valid. For instance, despite high levels of corruption and terrorism in Uganda, the number of tourists in the country has risen annually by about 17% between 1999 and 2009 (Yap and Saha, 2013). Iran is a good example of this kind of destination; despite its conflict-ridden regional context, this country became a booming tourist destination (Khodadadi, 2016). One potential reason for this dissimilar trend may be that some political problems simply cannot negatively impact tourism demand, especially in countries with historical and natural heritage sites (Yap and Saha, 2013). Thus, while political turmoil may not drive tourists away completely, it may alter tourist behavior. This study aims to understand the effect that political unrest has on tourists’ LOS.

H₅: Destination’s political situation determines LOS.

Methodology

The pilot study questionnaire was completed by 40 tourists visiting Tabriz. The validity of the questionnaire was tested and confirmed. Moreover, the reliability coefficient was confirmed by its Cronbach alpha (0.725) after a pilot test, meaning it is both reliable and valid. It is worth noting that the Persian questionnaire was translated into English, Turkish, and Arabic, as all respondents could speak and understand at least one of these three languages.

Taking cue from previous researches (Adongo et al., 2017; Alén et al., 2014; Barros et al., 2010) the questionnaire was used to collect data. The data was collected during the summer of 2017 using a structured self-administered questionnaire that was hand-delivered to 1,023 international travelers by a researcher. Only one individual from each travel party was invited to take part because travel party members typically have similar opinions to each other. Finally, 78.98% of questionnaires (808) were completed and returned. This rate of return, according to Dillman (1978), was acceptable. After initial cleaning and data screening, 726 responses were used for analysis and entered into SPSS (82 questionnaires were omitted, due to incomplete answers).

It is important to note that, since the survey was carried out in city’s departure terminals, all interviewees reported their LOS, so censored data was not included in the analysis (i.e., censoring occurs when we have some information about individual survival time, but we do not know the survival time exactly). Moreover, due to the sensitive religion and politics investigation involved in the questionnaire items, and following recommendations from Podsakoff et al. (2003), respondents were assured that their involvement would be voluntary and anonymous, and there are no correct or wrong answers, so that they would express their personal views as honestly despite the potential bias brought about by the interviewer-participant interaction.

The demographic makeup of the participants was as follows: The average LOS was around 8.6 days and the composition of the sample is not balanced in terms of gender (58.4% men), tourists with a university level of education represent 61.5% of the sample; the largest share of the trips take place for medical purposes (31.8%), followed by VFR (29.1%) and leisure (27.6%) purpose and the majority of tourists stayed in paid accommodation for the duration of their visit (58%).

It is worth mentioning that the characteristics of the sample in our study are similar to those of the national profile, according to recent data from the Tabriz tourism organization report (2017).

Econometric methodology

The time-varying LOS variable (Ferrer-Rosell et al., 2014) has some characteristics that are difficult to deal with using conventional statistical methods, and extensive discussion about which may be the best (see the section 3.3 below for a more detailed explanation).

Duration, defined as the time elapsed until a certain event occurs, was analyzed using survival models (Allison, 2010; Gokovali et al., 2007). Survival analysis is a branch of statistics used to analyzing the predicted time until certain events or actions occur. For the purpose of this study, the event of interest is the tourist’s departure from Tabriz.

Two concepts are key in survival analysis: the survival function and the hazard function. The survival function is the probability of observing a survival time greater than or equal to some stated value (Kiefer, 1988). The hazard function is the rate at which the spell will be completed by duration t, conditional upon the spell lasting until t (Kiefer, 1988: 651). Put formally, the approach may be described as follows: let spell length be represented by a random variable T (in this study, LOS measured in days), and t for its realizations.

The cumulative distribution function of T, F(t), is the probability that a random variable will have a survival time less than some stated value t; it can be formulated as:

F (t) = P_{r} (T \leq t) = \int_{0}^{t} f (s) d s .

In this study, F(t) denotes the probability that an international tourist leaves Tabriz before time t, whereas the survival function of T, S(t) indicates the probability that an international tourist stays in Tabriz longer than the specified time t. Thus, it can be written as:

S (t) = 1 - F (t) = P_{r} (T \geq t) = \int_{0}^{\infty} f (s) d s .

The hazard function or hazard rate, h(t), is another fundamental concept in survival analysis. h(t) refers to the probability that the subject changes its current status at time t, given that it has been staying in that status up to time t (Allison, 2010; Hosmer and Lemeshow, 1999; Kiefer, 1988).

In this context, the hazard function gives the probability that an international tourist leaves Tabriz at time t, given that they have been staying from time 0 until time t. In other words, the hazard rate answers the following question: ‘Given that the spell has lasted until time t, what is the probability that it will end in the next short interval of time Δt?’ (Menezes et al., 2008), where indicates a small interval of time. More formally:

h (t) = lim_{Δ \to 0} Pr (t \leq T \leq t + Δ | T \geq t) / Δ

It is also worth mentioning that the hazard function is also known as the conditional failure rate; the conditional probability is the likelihood that a tourist’s LOS time, T, will lie in the time interval between t and t + Δ, given that the survival time is greater than or equal to t.

The hazard rate can be rewritten as $h (t) = \frac{f (t)}{[1 - F (t)]} = f (t) / S (t)$ , and the cumulative hazard function formulated as

H (t) = \int_{0}^{t} h (u) d u = \int_{0}^{t} \frac{f (u)}{S (u)} d u = \int_{0}^{t} \frac{d [1 - S (u)]}{S (u)} = - log [S (t)] .

Therefore, we can develop the survival function; $S (t) = e^{- H (t)} = exp [- \int_{0}^{t} h (u) d u]$

Finally, by introducing this relationship into hazard rate equation, we have

S (t) = e^{- H (t)} = exp [- \int_{0}^{t} h (u) d u] .

The advantages of survival analysis

The particular nature of LOS (Ferrer-Rosell et al., 2014) has some characteristics that are difficult to deal with using conventional statistical methods. One of the main challenges is that LOS has a non-negative nature—time can be any number equal to or greater than zero. This characteristic of time-varying variables was introduced by Kiefer (1988) as an ‘inherent aging process’ and echoed in subsequent works (Gokovali et al., 2007; Greene, 2000). Kiefer defined it as ‘the dependent variable under consideration which should be assigned positive values.’ This feature of LOS forces the use of a model in which the systematic component must yield fitted values that are strictly positive. Linear regression approaches, such as ordinary least square (OLS) and its derivations, can yield negative fitted values, particularly for subjects with short survival times. In summary, linear regression approaches are unable to deal with this feature of the LOS variable.

Additionally, the LOS variable is not normally distributed, meaning that OLS, which is based on the assumption of normality, is inadequate for a LOS study because the normality of errors is dramatically changed by the presence of extreme values and skewness. There are some remedial actions that can be taken to transform non-normal data into normal data, but no methods are reliably sufficient. One of the most common of these methods is to employ a transformation (i.e., change the distribution by applying a mathematical operation to each observation/data value). However, this normalization process has been criticized by several scholars, as the transformed LOS has little meaning for decision-making processes (Greene, 2000). The issue can instead be handled by conducting a survival analysis, which offers a variety of outcomes.

Furthermore, we are not interested in estimating binomial or multinomial distributions because LOS is continuous and may lie anywhere between 1 and 31 nights. Several scholars have tried to tackle these features of time-varying variables through various methodological approaches. For example, binary logit models (logistic regression analysis with a dichotomous function) have been used by several researchers to analyze the relationship between an independent variable and a time-varying variable. Alegre and Pou (2006), based on the McFadden’s (1974) discrete choice random utility model, employed binary logit to model the LOS of tourists in the Balearic Islands, Spain. They considered LOS a binary variable and coded it as 0 if shorter than 7 days and 1 if 7 days or longer. Of course, this process ignores exact information on trip length, meaning the research loses precision and relevance. When the LOS is distributed evenly, however, researchers may have no obvious cut-off point, leading to the arbitrary partition of the LOS.

As already mentioned, additional concerns arise when using survival analysis from left or right data censoring issues. However, censoring does not occur at all in our data, as the survey was carried out in airport departure with all interviewees directly reporting their LOS. In short, when LOS is used as a dependent variable, certain aspects can cause problems for data analysis using traditional statistical models—survival analysis is a solution to those problems.

Results

Two frequently used parametric models for adjusting survival functions for the explanatory variable effects on survival probability are the accelerated failure time (AFT) model and the proportional hazard (PH) ratio model. Based on the Cox’x semiparametric model estimation, according to the Schoenfeld residuals test, we notice the PH assumption holds, therefore, we estimate the most common PH models; exponential, Weibull, and Gompertz distributions. According to Akaike information criteria (AIC), Weibull distribution is the best-fitting model (see Table 2). Therefore, the exponential and Gompertz models were eliminated by this criterion and the results of the exponential and Gompertz specifications are not reported here. Only the results of the Weibull and Cox specifications are reported and discussed, as the Cox PH assumption was not rejected and the Weibull parametric estimation gave the smallest AIC value.

Table 2.

Akaike information of PH parametric models.

	Exponential	Weibull	Gompertz
Log-likelihood values	−153.87	−148.05	−159.31
AIC values	352.06	325.30	367.8

The results of both Cox’s and Weibull’s regressions are shown in Table 3. These regressions generate almost identical results. Therefore, we will center our comments around the variables that are significant in both models and around the values obtained in the Weibull model.

Table 3.

Estimation results of Cox’s and Weibull’s regressions.

Variable	Weibull regression (P-value)	Cox regression (P-value)
Age (Age)²	+0.023 (0.00) −0.000226 (0.00)	+0.013 (0.003) −0.00126 (0.015)
Gender Male* Female	0.00 −0.093 (0.00)	0.00 −0.058 (0.00)
Education level Primary* High educated	0.00 −0.078 (0.60)	0.00 -0.009(0.00)
Economic status Income (Income)²	−0.00165 (0.73) 0.0066 (0.67)	−0.0135(0.51) −0.001(0.48)
Travel purpose** Leisure* VFR Business Medical Other	0.00 0.11(0.00) −0.36(0.00) 0.083(0.003) 0.022(0.00)	0.00 0.29(0.00) 0.008(0.00) 0.046(0.00) 0.033(0.00)
Frequency of traveling (frequency of traveling)²	−0. 10(0.00) 0.004(0.00)	−0.23(0.00) 0.012(0.00)
First visit No* Yes	0.00 −0.22(0.001)	0.00 −0.29(0.00)
Distance less than 300km* 300–800 km 800–1300 km More than 1300 km	0.00 +0.011(0.00) +0.18(0.001) +0.27(0.00)	0.00 0.006(0.00) 0.013(0.00) 0.17(0.00)
Group size group size (group size)²	−0.019(0.87) −0.13(0.63)	0.001(0.72) −0.01(0.76)
Accommodation free dwelling* paid accommodation high price hotels low price hotels/app	0.00 −0.37(0.00) −0.21(0.00)	0.00 −0.201(0.001) −0.09(0.00)
daily spending (daily spending)²	−0.21(0.00) +0.01(0.72)	−0.08(0.00) −0.03(0.62)
Number of visited attractions	+0.32(0.00)	0.09(0.00)
Overall image of Tabriz Positive* Negative Destination image** cultural destination* medical destination touristic activities	0.00 −0.47(0.00) 0.00 +0.201(0.00) −0.01(0.00)	0.00 −0.29(0.00) 0.00 0.1419(0.00) −0.13(0.00)
Other dummy variables** Islamic regulations Political turmoil	−0.43(0.00) −0.39(0.00)	−0.27(0.00) −0.19(0.00)

* indicates the reference category in a group of dummy variables.

It should be noted that quantitative variables and their squared forms are indicated in Table 3 to show quadratic relationships between independent variables and LOS. For nominal variables, after setting one as the reference, the others are coded into dummy variables. The first line of each multinomial variable shows the p-values for the Wald test of the hypothesis that all coefficients are fixed to be zero.

In line with our a priori expectations and the previous literature (Aguilar and Díaz, 2019; Alen et al., 2014; Barros and Machado, 2010; De Oliveira Santos et al., 2015; Gokovali et al., 2007; Martínez-García and Raya, 2008; Yang et al., 2011), the econometric analysis indicates that LOS is predicted by demographic variables.

The results suggest that age has a non-linear effect on tourists’ LOS. Based on the existing coefficient, tourists’ LOS in Tabriz follows a significant concave function of their age. LOS increases alongside age until age 51(−b/2a = 0.023/ (2 * 0.000226) = 50.88), at which point LOS decreases as age rises. This relationship may stem from the higher purchasing power of older tourists; this is in line with the macroeconomic income-consumption theory, which asserts that wealth, and, in turn, expenditure, increase alongside age and then it has a descending trend which is approved by the results of regression for this theory about tourism (Fleischer and Pizam, 2002; Salmasi et al., 2012). According to Fleischer and Pizam (2002), the decrease in income after retirement and the deterioration of health after age 65+ likely causes a decline in LOS.

Our results indicate that male tourists generally spent more time in Tabriz—females stayed, on average, 8.9% (1 − exp (−0.093)) shorter than males.¹ This is in line with the results of several previous studies (Barros and Machado, 2010; Menezes and Moniz, 2013; Peypoch et al., 2012; Rodríguez et al., 2018). One reason for this outcome may be the strict Islamic dress code, which mandates a hijab for female tourists, and several other challenges and restrictions.

The income variable was expected, based on primary assumptions and the extant literature, to significantly influence tourists’ LOS. In general, according to basic economic theory, tourism is a normal product with a positive income elasticity of demand; in other words, higher income leads to lengthier stays (Barros and Machado, 2010; Ferrer-Rosell et al., 2014; Gokovali et al., 2007; Salmasi et al., 2012). This argument that income always has positive effects on LOS, however, is not universally valid. Another strand of research suggests that a higher income actually results in shorter stays (Mak and Moncur, 1979). However, in this study, both the squared and non-squared forms of income were found to be non-significant. The results show Tabriz tourism as a normal necessity commodity. This result is in line with those of De Oliveira Santos et al. (2015) and Rodríguez et al. (2018). One can point to travel purpose to justify this result, meaning that, as most participants traveled for non-leisure purposes (e.g. VFR and medical), it constitutes a necessary commodity.

The relationship between level of education and LOS was not found to be statistically significant, which is in line with Gokovali et al. (2007).

The results indicate that there is a significant relationship between trip attributes and tourists’ LOS. In terms of motive, travel purpose was found to be a significant determinant of tourists’ LOS (p < 0.01). The results indicate that the trips taken to Tabriz for VFR lasted 11.6% (1 − T (purpose=VFR) / T (purpose= leisure) = 1 − exp (0.11) = −0.116), longer than leisure trips. Type of accommodation used may effect on longer duration of such trips. Those traveling for VFR reason often spend at least a night at a family member’s or friend’s house, which would save them money and, in turn, allow for a longer trip. These results are in line with previous findings (e.g. Alén et al., 2014; De Oliveira Santos et al., 2015; Wang et al., 2012), though they contrast with Hellstrom (2006) and Menezes et al. (2008). Additionally, the results indicate that medical-purpose trips tended to have medium LOS while business tourists stayed for the least amount of time—roughly 37% less than VFR tourists.

The second trip attribute variable, frequency of traveling to Tabriz, was found to have a negative effect on tourists’ LOS and increase the destination-leaving risk by about 9.9%. Based on the cardinal utility theory, we conclude that visiting Tabriz is a composite commodity; frequent use reduces its utility and, thus, leads to a decrease in demand. This is in line with the findings of Paul and Rimmawi (1992) and De Oliveira Santos et al. (2015). However, this outcome is in contrast with the findings of several researchers (Alegre and Pou, 2006; Alegre et al., 2011; Barros and Machado, 2010; Gokovali et al., 2007; Jacobsen et al., 2018; Thrane and Farstad 2012; Yang et al., 2011) who claimed that repeat visitors tend to stay longer. Note that, based on our results, first-time visitors stay for 19.7% shorter periods than do repeat visitors. Moreover, previous scholars, aside from De Oliveira Santos et al. (2015), have not considered the differences between first-time visitors and frequent visitors.

With respect to the distance variable, our results indicate a robust positive relationship between physical distance—meaning the distance between Tabriz and the tourists’ place of origin—and LOS (p < 0.01). Tourists from over 1000 km away, on average, stay 31% longer than the rest. This outcome is in line with most previous studies. According to (De Oliveira Santos et al., 2015; Jackman et al., 2020; Menezes et al., 2008; Paul and Rimmawi,1992; Rodríguez et al., 2018; Wang et al., 2012; Yang et al., 2011), tourists who live far away from their destination tend to stay longer to make up for the higher overall travel cost and time; in other words, they spread the journey’s fixed costs over a longer period). Moreover, this result is in accordance with the first law of geography that ‘everything is related to everything else, but near things are more related than distant things’ (Tobler, 1970). In the context of tourism, this law generally refers to the negative impact of distance (McKercher et al., 2008)—the association between two locations becomes weaker as the distance between them grows (Jackman et al., 2020).

Based on the results, we have confirmed that cost has a negative effect on LOS. This is in line with the findings of earlier works (Alegre and Pou, 2006; Alegre et al., 2011; Hellstrom, 2006; Peypoch et al., 2012; Thrane and Farstad, 2012). The depreciation of Iran’s national currency is associated with an increase in the purchasing power of international tourists. Therefore, this depreciation has made it easier for Iran to attract international tourists and increase their LOS.

Moreover, the results indicate that an increase in the number of attractions visited leads to a statistically significant increase in LOS. This finding is in line with several previous empirical studies (Adongo et al., 2017; Ferrer-Rosell et al., 2014; Rodríguez et al., 2018). As Botti et al. (2008) claim, tourists formulate their expectations by ranking a destination’s attractions and potential activities before making decisions. For each attraction and activity, they evaluate the time they wish to spend on it.

Shorter LOS in Tabriz is associated with negative images held by tourists. Those with a negative image of Tabriz are expected to stay for 37.5%, (1 − T (Positive image)/T (negative image) = 1 − exp (−0.47) = 0.375), less time than those with a positive image. This is likely because tourism products are intangible and potential tourists, or at least first-time visitors, cannot directly experience them; their decisions and behaviors are based on perception rather than objective reality (Jacobsen et al., 2018; Machado, 2010; Menezes and Moniz, 2013). Additionally, tourists who choose Tabriz for its affordable and high-quality healthcare services generally stay the longest—22% longer than those who visit for historical reasons. Tourists who choose Tabriz for its touristic activities tend to experience the shortest stays.

Results indicate that Islamic regulations and political turmoil have a negative and significant effect on tourists’ LOS (p < 0.01). A negative sign means that as the value of a variable increases, the survival of tourists’ duration decreases. More precisely, tourists who consider ‘Islamic regulations’ as limitations and ‘political turmoil’ as an important risk tend to take shorter journeys. Evidently, these variables can hinder long stays in Tabriz. The results indicate that female tourists do not stay as long as male tourists, likely due to religious restrictions and feelings of insecurity in conservative Islamic destinations (Brown and Osman, 2017). Considering the positive effects of tourism on economic development, authorities should take substantive steps to reduce religious restrictions—at least for tourists—and educate locals on this matter so that tourism can thrive in Islamic destinations.

Policy implications and limitation

The results obtained from this analysis allow us to accept the majority of the proposed hypotheses. The general conclusion is that middle-aged men who traveled a long distance generally stay the longest in Tabriz. Moreover, LOS in Tabriz is positively associated with medical and VFR motivations, positive destination images, free/cheap dwellings, and tourist attraction visits. Additionally, this study has confirmed that travelers who consider Islamic regulations and political turmoil as significant restrictions generally stayed for less time than those who regard them as unimportant.

These findings show that policymakers and those who are invested in the tourism industry must develop efficient marketing strategies to increase LOS and attract the market segments that are more likely to stay for long periods of time. They must consider the impact of influential factors on international tourists’ LOS; while it is not possible to control all of the explanatory variables of LOS in a destination, it is possible to concentrate on several of the identified factors (Peypoch et al., 2012).

This study revealed that a primary goal of government should be to maintain political stability, as political turbulence tends to reduce LOS. Additionally, instability fosters a negative image of the country in tourists’ minds (Butler and Suntikul, 2017; Seddighi et al., 2001). Authorities should conduct recovery marketing, and advertising efforts should be integrated with crisis-management activities. As suggested by Sonmez and Graefe (1998), airlines should conduct promotional campaigns and generous incentive schemes, such as 2-for-1 ticket offers, free companion tickets, and free car rentals. Furthermore, hotels can advertise significant price cuts for long-term guests. Authorities in tumultuous areas should issue regular reports on the level of security and assist with travel plans in order to reassure potential tourists (Issa, 2006). The more sensitive potential tourists are to political instability, the more aggressive the marketing and promotional strategies of a tourist destination should (Seddighi et al., 2001). Furthermore, governments need to maintain stability and security and portray themselves through the media in a way that is comforting to tourists.

Importantly, the results indicate that Islamic regulations have a negative effect on LOS, particularly for female tourists; such religious restrictions can be serious barriers to attract new tourists (Zamani- Farahani and Henderson, 2010). Although cultural changes need long-term planning, considering the positive effects of tourism on economic development, authorities must swiftly take substantive steps to reduce religious restrictions, at least for tourists. one of the programs in the field of culture is through educational programs. Local and national media presentations can support appropriate interactions with tourists. So, they could take advantage of tourism opportunities in Islamic destinations, like Tabriz.

Medical tourism appears to be a good opportunity for Tabriz’s tourism industry. Medical trips tend to involve long stays, so special attention should be paid to the needs of these tourists. This outcome is particularly valuable for authorities and marketers in Tabriz, as they can shift their priorities toward medical and health tourism to increase LOS and enhance the economic benefits of tourism.

In consideration of extant studies that concluded cost benefit is one of the initial factors which encourages medical tourists to travel abroad for treatment (Momeni et al., 2018). The depreciation of Iran’s currency and its resultant tourism price competitiveness constitutes an opportunity for the Iranian government to establish a promotional strategy that depicts Tabriz as a destination for high-quality but affordable medical tourism. Therefore, the low prices for surgical operations is a major driver of medical tourism in the city, and the government should take advantage of that. Importantly, the maintenance and enhancement of medical programs should be a main priority of the Tabriz public sector (Momeni et al., 2018). Moreover, another marketing strategy is the establishment of agents of companies for medical tourism in cities of neighboring countries (Momeni et al., 2018) and lowering waiting times (Yu and Ko, 2012). For example, in Thailand, medical tourism facilities have contracts with airlines to offer reduced ticket rates designed for foreign patients requiring follow-up medical trip and lure in medical tourists (Buzinde and Yarnal, 2012).

We also found that the number of tourist attractions has a positive impact on LOS and a negative impact on the risk of leaving the destination. In other words, the more attractions a tourist visits, the longer they stay. According to Botti et al. (2008), LOS is one of the questions that is generally decided during the planning phase, when tourists are deciding what they want to do. Therefore, it is necessary for DMOs to create strategies to introduce and promote natural, cultural, and historical attractions. Additionally, they could increase tourists’ LOS by developing new attractions and activities, such as man-made attractions, festivals or events (Alén et al., 2014; Ferrer-Rosell et al., 2014; Rodríguez et al., 2018), that would increase the amount of time necessary for tourists to see everything they want to see. Importantly, Yap and Saha (2013) have found that while terrorism and political unrest are decisive factors in tourism demand, the negative effect is significantly smaller in countries with historical and natural attractions. Tabriz fits into this category, as it is an attractive tourist destination that is home to both tangible and intangible heritage assets (Gannon et al., 2020). The city must reevaluate its existing policies and design suitable strategies to introduce, promote, and advertise its historical and cultural attractiveness.

In summary, uncovering the determinants of LOS is critical to the design of marketing policies that promote longer stays and, in turn, result in higher revenue. From a marketing perspective, these findings can be used to develop successful marketing strategies that target the tourist groups that are most likely to stay for long periods of time. They also allow for destinations to cater to the needs and desires of these high-LOS tourists. As already stated, the results of this study indicate that theocratic rule and political instability seriously hinder tourists’ LOS lengthier stays. However, since this is a novel finding, more research is necessary to confirm it. One interesting direction for future research would be to develop a detailed analysis of the effects of these elements on tourists’ LOS. It would be beneficial to focus on other Islamic and politically unstable destinations to determine whether the effects of the analyzed variables on the LOS vary by destination. Another limitation of this study is the use of the following statement ‘Islamic laws and regulations/ political situation negatively affect my trip’s LOS’ in the questionnaire. The reason is that this statement could be suggestive and leading due to the sensitive context of this study (i.e., religion and political). Therefore, it is imperative for future studies to take the sensitivity of the subject into consideration when using such statements. Importantly, since our study only considered one destination, we could not claim any causal relationship between LOS and destination characteristics. In order to establish causality, an experimental design that considers multiple destinations would be optimal. Nevertheless, it is possible to partially investigate the relationship mentioned above by surveying multiple destinations with different attributes (e.g. different number of attractions, different destination’s image) and comparing the LOS among these destinations.

Finally, future work could also directly expand on this study. For example, one could evaluate whether a tourist’s religion influences their LOS in Islamic destinations.

Supplemental material

Supplemental Material, questionnaire - The influence of theocratic rule and political turmoil on tourists’ length of stay

Supplemental Material, questionnaire for The influence of theocratic rule and political turmoil on tourists’ length of stay by Fahimeh Hateftabar and Jean Michel Chapuis in Journal of Vacation Marketing

Footnotes

Acknowledgements

We would like to express our very great appreciation to Amid Miraeinejad for his valuable and constructive suggestions during the development of this research work. Acknowledgments are also extended to the anonymous reviewer for his/her constructive comments on the earlier draft of this paper.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Fahimeh Hateftabar

Supplemental material

Supplemental material for this article is available online.

Note

References

Adongo

Badu-Baydan

Boakye

(2017) The tourism experience-led length of stay hypothesis. Journal of Outdoor Recreation and Tourism 18: 65–74.

Aguilar

Díaza

(2019) Length of stay of international tourists in Spain: a parametric survival analysis. Annals of Tourism Research 79: 102768.

Aguiló

Rosselló

Vila

(2017) Length of stay and daily tourist expenditure: a joint analysis. Tourism Management Perspectives 21: 10–17.

Alegre

Pou

(2006) The length of stay in the demand for tourism. Tourism Management 27: 1343–1355.

Alegre

Mateo

Pou

(2011) A latent class approach to tourists’ length of stay. Tourism Management 32(3): 555–563.

Alén

Nicolau

Losada

, et al. (2014) Determinant factors of senior tourists’ length of stay. Annals of Tourism Research 49: 19–32.

Allison

DPD

(2010) Survival Analysis using SAS: A Practical Guide (2nd edn). Ernakulam: SAS publication.

Archer

Shea

(1975) Length of stay problems in tourist research. Journal of Travel Research 13(3): 8–10.

Barros

Machado

(2010) The length of stay in tourism. Annals of Tourism Research 1(1): 1–15.

10.

Barros

Correia

Crouch

(2008) Determinants of the length of stay in Latin American tourism destinations. Tourism Analysis 13(4): 329–340.

11.

Battour

MMK

Ismail

Battor

(2011) The impact of destination attributes on Muslim tourist’s choice. International Journal of Tourism Research 13(6): 527–540.

12.

Botti

Peypoch

Solonandrasana

(2008) Time and tourism attraction. Tourism Management 29(3): 594–596.

13.

Brida

Meleddu

Pulina

(2013) Factors influencing length of stay of cultural tourism. Tourism Economics 19(6): 1273–1292.

14.

Brown

Osman

(2017) The female tourist experience in Egypt as an Islamic destination. Annals of Tourism Research 63: 12–22.

15.

Butler

Suntikul

(2017) Tourism and Political Change (2nd edn). Oxford: Goodfellow.

16.

Buzinde

Yarnal

(2012) Therapeutic landscapes and postcolonial theory: a theoretical approach to medical tourism. Social Science & Medicine 74(5): 783–787.

17.

Correia

Serra

Artal-Tur

(2017) Steady tourists’ relationship with a mature destination: the case of Portugal. Tourism Economics 23(4): 803–815.

18.

Davies

Mangan

(1992) Family expenditure on hotels and holidays. Annals of Tourism Research 19(4): 691–699.

19.

De Oliveira Santos

Ramos

Rey-Maquiera

(2015) Length of stay at multiple destinations of tourism trips in Brazil. Journal of Travel Research 54: 788–800.

20.

Decrop

Snelders

(2004) Planning the summer vacation: an adaptable process. Annals of Tourism Research 31(4): 1008–1030.

21.

Farmaki

Khalilzadeh

Altinay

(2019) Travel motivation and demotivation within politically unstable nations. Tourism Management Perspectives 29: 118–130.

22.

Ferrer-Rosell

Martínez-García

Coenders

(2014) Package and no-frills air carriers as moderators of length of stay. Tourism Management 42: 114–122.

23.

Fleischer

Pizam

(2002) Tourism constraints among Israeli Seniors. Annals of Tourism Research 29: 106–123.

24.

Gannon

Rasoolimanesh

Taheri

(2020) Assessing the mediating role of residents’ perceptions toward tourism development. Journal of Travel Research 1–23.

25.

Gokovali

Bahar

Kozak

(2007) Determinants of length of stay: a practical use of survival analysis. Tourism Management 28(3): 736–746.

26.

Gössling

Scott

Michael

(2018) Global trends in length of stay: implications for destination management and climate change. Journal of Sustainable Tourism 26(2): 2087–2101.

27.

Greene

(2000) Econometric Analysis. India: Pearson Education.

28.

Hall

(2010) Crisis events in tourism: subjects of crisis in tourism. Current Issues in Tourism 13(5): 401–417.

29.

Hellstrom

(2006) A bivariate Count data model for household tourism demand. Journal of Applied Econometrics 21(2): 213–226.

30.

Henderson

(2009) Islamic tourism reviewed. Tourism Recreation Research 32(2): 207–212.

31.

Hino

(2011) The impact of Islam on food shopping and consumption patterns of Muslim households. In: Sandikci

Rice

(eds) Hand. Islamic Market. London: Edward Elgar.

32.

Hosmer

Lemeshow

(1999) Applied Survival Analysis. New York: Wiley.

33.

Issa

(2006) Impacts of political instability on tourism planning and development: the case of Lebanon. Tourism Economics 12(3): 361–381.

34.

Jackman

Lorde

Naitram

, et al. (2020) Distance matters: the impact of physical and relative distance on pleasure tourists’ length of stay in Barbados. Annals of Tourism Research 80: 102794.

35.

Jacobsen

SJK

Gossling

Dybedal

, et al. (2018) Exploring length of stay: international tourism in south-western Norway. Journal of Hospitality and Tourism Management 35: 29–35.

36.

Khodadadi

(2016) A new dawn? The Iran nuclear deal and the future of the Iranian tourism industry. Tourism Management Perspectives 18: 6–9.

37.

Kiefer

(1988) Economic duration and hazard functions. Journal of Economic Literature 26: 646–667.

38.

Lanouar

Goaied

(2019) Tourism, terrorism and political violence in Tunisia: evidence from Markovswitching models. Tourism Management 70: 404–418.

39.

Light

Rezaei

Dana

(2013) Ethnic minority entrepreneurs in the international carpet trade: an empirical study. International Journal of Entrepreneurship and Small Business 18(2): 125–153.

40.

Machado

(2010) Does destination image influence the length of stay in a tourism destination. Tourism Economics 16(2): 443–456.

41.

Mak

Moncur

JET

(1979) The choice of journey destinations and length of stay: a micro analysis. Review of Regional Studies 10(2): 38–48.

42.

Mak

Moncur

Yonamine

(1977) Determinants of visitor expenditures and visitor lengths of stay: a cross-section analysis of US visitors to Hawaii. Journal of Travel Research 15(3): 5–8.

43.

Martínez-García

Raya

(2008) Length of stay for low-cost tourism. Tourism Management 29(1): 1064–1075.

44.

McFadden

(1974) Conditional logit analysis of qualitative choice behavior. In Zarembka

(ed.) Frontiers in Econometrics. New York: Academic Press. pp. 105–142.

45.

McKercher

Chan

Lam

(2008) The impact of distance on international tourist movements. Journal of Travel Research 47: 208–224.

46.

Menezes

Moniz

(2011) Determinants of length of stay. A parametric survival analysis. Tourism Analysis 16: 509–524.

47.

Menezes

Moniz

(2013) The determinants of length of stay of tourists in the Azores: a count model approach. Revista tourismo & Desenvo 15: 25–40.

48.

Menezes

Moniz

Vieira

(2008) The determinants of length of stay in the Azores. Tourism Economics 14: 205–222.

49.

Momeni

Janati

Imani

, et al. (2018) Barriers to the development of medical tourism in East Azerbaijan province, Iran: a qualitative study. Tourism Management 69: 307–316.

50.

Paul

Rimmawi

(1992) Tourism in Saudi Arabia: asir national park. Annals of Tourism Research 19(3): 501–515.

51.

Peypoch

Randriamboarison

Rasoamananjara

, et al. (2012) The length of stay of tourists in Madagascar. Tourism Management 33(5): 1230–1235.

52.

Podsakoff

MacKenzie

Lee

(2003) Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology 88(5): 879–903.

53.

Rodríguez

Martínez-Roget

González-Murias

(2018) Length of stay: evidence from Santiago de Compostela. Annals of Tourism Research 68: 9–19.

54.

Salmasi

Celidoni

Procidano

(2012) Length of stay: price and income semi elasticities at different destinations in Italy. International Journal of Tourism Research 14: 515–530.

55.

Seddighi

Nuttall

Theocharous

(2001) Does cultural background of tourists influence the destination choice? an empirical study with special reference to political instability. Tourism Management 22(2): 181–191.

56.

Soler

Gemar

Correia

(2018) Length of stay for tourists' inland trips. Journal of Destination Marketing & Management 10: 49–60.

57.

Sonmez

(1998) Tourism, terrorism, and political instability. Annals of Tourism Research 25(2): 416–456.

58.

Sonmez

Graefe

(1998) International vacation decisions and terrorism risk. Annals of Tourism Research 25(1): 112–144.

59.

Takeyh

Maloney

(2011) The self-limiting success of Iran sanctions. International Affairs 87(6): 1297–1312.

60.

Thrane

(2016) Students’ summer tourism: determinants of length of stay (LOS). Tourism Management 54: 178–184.

61.

Thrane

Farstad

(2012) Tourists’ length of stay: the case of international summer visitors to Norway. Tourism Economics 18(5): 1069–1082.

62.

Tobler

(1970) A computer movie simulating urban growth in the Detroit region. Economic Geography 46: 234–240.

63.

Wang

Little

DelHomme-Little

(2012) Factors contributing to tourists’ length of stay in Dalian northeastern china—a survival model analysis. Tourism Management Perspectives 4: 67–72.

64.

Weidenfeld

(2006) Religious needs in the hospitality industry. Tourism and Hospitality Research 6(2): 143–159.

65.

Yang

Wong

Zhang

(2011) Determinants of length of stay for domestic tourists: case study of Yixing. Asia Pacific Journal of Tourism Research 16(6): 619–633.

66.

Yap

Saha

(2013) Do political instability, terrorism, and corruption have deterring effects on tourism development even in the presence of Unesco heritage? A cross-country panel Estimate. Tourism Analysis 18(5): 587–599.

67.

(2012) A cross-cultural study of perceptions of medical tourism among Chinese, Japanese and Korean tourists in Korea. Tourism Management 33(1): 80–88.

68.

Zamani-Farahani

Henderson

(2010) Islamic tourism and managing tourism development in Islamic societies: the cases of Iran and Saudi Arabia. International journal of tourism research 12(1): 79–89.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB