Does the Repeat Tourist Spend More Than the First Time? A Contribution to the Literature

There isn't a consensus about the impact of a first time or a repeat tourist on local economy since that the findings of studies comparing the amount of money spent by the two categories of tourists are inconclusive. We interviewed 500 people who went to any tourist destination more than once through a virtual platform in Brazil. This article uses Principal Component Analysis to analyze this issue. The results identified four types of repeat tourists, who tend to spend less than first-time tourists, but who may influence people to visit the destination due to their high degree of satisfaction. A recommendation for future research would be to use the Confirmatory Factor Analysis (CFA) to confirm the measurement model after conducting Exploratory Factor Analysis (EFA).


Introduction
The principal objective of economic activity is the profit maximization. For this, it is essential to obtain the most significant volume of revenue possible from the sale of products and services. In the case of tourism, this revenue is associated with the behavior of the tourist as a consumer, which can vary if the tourist is repeated or first-time, among other factors.
For example, Freytag (2010) have found that the number of times that tourists climbed the Eiffel Tower or visited the Louvre fell more than 60% after the third visit to Paris, while the drop was 50% for the Versailles Garden. Indeed, this behavior influences the revenue of a tourist destination. Furthermore, empirical evidence shows that repeated tourists are more price sensitive and more interested in looking for lower prices that first-time tourists (Li et al., 2008). A recent study made by IFT Tourism Research Centre (ITRC (2017), in Macao, showed that the average spending of first-time visitors is 14% higher than repeat visitors. Saladié, Clavé & Gutiérrez (2016) have identified that friends or relatives accommodated 27.7% of the repeat tourists while 58.5% of first-time tourists chose to stay in a hotel. It surely has a positive impact on the local economy.
But this result is not a consensus. For example, the study made by Mohamad & Ghani (2014) in Malaysia suggested that first-time visiting tourists are price-sensitive compared to repeat tourists. However, many studies show that the more the tourist repeats a tourist destination, higher the likelihood of the greater the possibility of it returning (Kozak & Rimmington, 2000;Badarneh & Mat Son, 2011). However, the most relevant studies seek to understand the determinants of tourist spending (Brida & Scuderi, 2013;Maroccu, 2015) but only some of them has investigated the differences between "first-time" tourists' and "repeat" tourists from the economic point of view (Petrick, 2004;Robinson & Gammon, 2004;Alegre & Cladera, 2010;Tjørve & Flognfeldt, 2015). Therefore, the objective of this article is to identify the attitudes, behaviors, and profiles of the repeat tourists to evaluate their effects on the local economy as a result of the found parameters. More specifically, this article seeks to test four hypothesis which, if confirmed, could reveal that the repeat tourist's impacts are more favorable to the local economy than first-time TAVARES, J. M.; IVANOV, S.; COBANOGLU, C. Does the repeat tourist spend more than the first time? A contribution to the literature. Revista Hospitalidade. São Paulo, volume 16, n.03, p. 118-133, 2019. Doi: https://doi.org/10.21714/2179-9164.2019 120 ISSN 1807-975X tourists. The importance of this objective is mentioned for Giraldi (2016) according to which "repeat motivations and the consequent effects have not been thoroughly investigated." This article desire to reduce the gap existing in the literature about this issue identifying which type of tourist contributes more to the local economy of the tourist destinationfirst-time tourists or repeat tourists. According to Wahid et al., (2016), the results show that the impact on local economies is inconclusive as well as for Pereda (2013, p.1) according to which "despite the efforts, it remains unclear why people undertake repeat visits and what kind characteristics hold repeat visitors." Besides it, "the study of tourist expenditure is an important thing in the formulation analysis of tourism marketing, strategies, and policies" (Sudanti et al., 2018, p.1).
Finally, another contribution is that most of the empirical studies on repeat tourist (and until about travel motivations) have not focused on Latin American countries, such as Brazil, for example.
As previously mentioned, these studies have investigated if there is a difference in terms behavioral incurred between a tourist visiting a destination for the first time and the one who has already been to the destination. However, this article opts for the approach that uses this behavior to make inferences about a possible distinction between the expenses of each one of them (firsttime and repeated tourists) and, consequently, and consequently the impact of these expenditures on the local economy.
These studies show that there is a significant difference between the actions of a first time and repeated tourists, mainly concerning visits to tourist attractions, use of public or private transportation, conventional or alternative routes, consumption of souvenirs, etc. According to Wahid et al., (2016, p.622), in a survey conducted in Malaysia, "there are significant differences and similarities between the first time and repeat tourist's in their demographic characteristics and travel behavior." Robinson & Gammon (2004) suggest that firsttime tourists prefer to escape from an environment and repeated tourist wishes to avoid for an environment. Research shows that "the first-time visitor to Tasmania generally spends more than the repeat visitor: in 2014, first-time visitors to Tasmania visited more attractions and participated in more activities than repeat visitors" (Tourism Tasmania, 2015).
In principle, the first-time tourist behavior would be closer to the "conventional tourists" image mentioned for McCabe (2005, p.98), namely as "being a tourist means that you only go sightseeing, without experiencing the people or the flair of the place/country you are visiting (...) a tourist feels extremely insecure when they venture away from the postcard sellers and the "touristic guides".
In principle, a tourist may revisit a tourist destination because of the level of satisfaction obtained in previous stays and by the opportunity to see or enjoy attractive not previously visited.
Moreover, the tourist may spend less due to prior knowledge of the destination or even bring friends and relatives who are unaware of the tourist destination (Wang, 2004;Tjørve & Flognfeldt, 2015). Besides it, Lee et al., (2009) have verified that "first-time visitors associated more strongly with relationships on food value and souvenir value than did repeat visitors." According to Chen & Xiao (2013), in a survey with repeated tourists, the most important motivation item to return to the destination was "take my family", and the most cited second was "to visit the natural attractions that had not visited before, "in a total of 20 possible motivations available to the respondent. Motivation such as "shopping" and "visit natural, cultural and historical attractions already visited before," weren´t even among the ten most cited. Petrick (2004b, p. 463), however, it is only a hypothesis that repeated tourists are the most desired visitors in a destination. However, the results obtained in the literature are controversial, as the following discussion.
For example, Tsitsiloni et al., (2012), have found that repeated tourists are staying more days and seem to spend more than the first time while Lau & McKercher (2004) have stated that higher the percentage of tourists who already have visited the destination, more would be achieved average expenditure. According to Visiting Britain (2016), an International Passenger Survey 2015 showed that first-time tourist and repeat tourist stay 8.6 and 6.4 nights, respectively, while the spends per visit were £ 623 e £ 578, respectively. Repeated tourists are more price-sensitive and more likely to seek lower prices than first-time tourists (Petrick, 2004). Craggs & Schofield (2009) find consistent evidence that infrequent tourists spend more than repeated tourists. According to Wang (2004, p.108), "repeated tourists visit fewer attractions" while Oppermann (1997), in a study conducted in New Zealand, reports that first-time tourists tend to spend more money than the repeat tourists. According to Tjørve & Flognfeldt, (2015, p.3), "an increasing proportion of repeat visitors have been proposed as an indicator of maturity (and therefore as a symptom of decline), in sea, sand, and sun (3S) destinations, together with low-status visitors, higher degree of seasonality, a decline in length of stay, and ultimately a decline in overall visitations". Chen & Gursoy (2000) statements that first-time tourists often use more external sources of information to familiarize yourself with your destination and buy fewer products already acquired previously, such as handicrafts, souvenirs, postcards, etc.
The intention of repeating the visit is negatively related to age and previous visits positively associated with spending. Gitelson & Crompton (1984) reports that repeated tourists are more likely to be older people looking to relax or visit and friends and relatives. In general form, repeated tourist tends to be a "tourist independent" (Sheng & Ping, 2006;Jia, 2008;Butler & Hannam, 2012): they use fewer guide services, find restaurants less disclosure, avoid travel agencies and use more public transport (BIN et al., 2009). In according to The Economist (2013), this recent growth in independent tourist is indeed a trend rather than an anomaly. Lau & McKercher (2004)  H2 -More likely of the repeated tourist to be an independent tourist -use fewer services guides, more public transport and find fewer disclosure restaurants -(based on Sheng & Ping, 2006;Jia, 2008;Bin et al., 2009;Butler & Hannam, 2012) H3 -Repeat tourists are more likely to take less money to the destination because they want to visit fewer paid attractions which have been visited previously (based on Oppermann, 1997;Tourism Tasmania, 2016); H4 -Repeat tourists regard highly the possibility of buying fewer products associated with the destination.

Methodology
Initially, the questionnaire was available on an online platform, which came to the respondents' knowledge through e-mail and mainly, in social networks. Although it has reach throughout Brazil, certainly most respondents would be resident in the State of Minas Gerais This questionnaire has six questions: the first three questions are related to the general identification of respondents; the fourth question concerns the type of tourism practiced when the destination is repeated; the fifth question seeks to investigate the reasons which made him repeat the tourist destination; the sixty question concerns the possibility for repeat tourist to do actions or to present certain behaviors, such as taking less money to the tourist destination, visiting fewer free tourist attractions, using more public transport, among others. The answers are based on a Likert scale from 1 to 5, where 1 means "no chance," and 5 means "total possibility." This article uses the Principal Component Analysis (PCA) as the main methodology - Wang (2004) has used this method to a similar purpose. In the PCA, a set of variables is transformed into another set and called the principal component. According to Mingoti (2005), principal component analysis's major objective is to "explain the variance and the covariance of a random vector, composed of a p-variable random, through the linear combination of the original variables, which are called principal components." Then, in this article, the main purpose of applying principal component analysis was to constructbased on the principal component responsible for most of the data variabilitydifferent profiles of tourists concerning their actions in a tourist destination, naming them according to these actions. In the multivariate statistical literature, there are three methods commonly used to choose the number of key components to be considered. In the method of Kaiser and scree plot, the other method is the percentage of the total variability explained. In it, a limit is set, and it appears the number of eigenvalues needed to achieve it. This study chose this method: this limit was close to 70%, because, according to Alves & Souza (2007, p. 5) "if the number of major components is too small can be an exaggerated reduction of dimensionality and much information can be lost." On the other hand, the objective of the technique of factorial analysis is "to replace the initial set of determinant characteristics, for others of a smaller number, but that keep significant original explanation of the problem, in order to raise the latent dimensions in the original variables of the phenomenon, aiming to give a more comprehensible interpretation according to common directions" (Maxwell, 2015, p.33).
It is necessary to highlight the difference between the PCA versus Factorial Analysis Exploratory (EFA). Both techniques seek to produce linear combinations of variables that capture as much as possible the variance of observed variables. In the PCA all variance is used. In EFA only the shared variance (Dancey & Reidy, 2004). Tabachnick & Fidell (2007, p.608) argued that "if you are interested in a theoretical solution uncontaminated by error variability, the factorial analysis should be your choice. If you only want an empirical summary of the data set, the principal component analysis is a better choice". For Garson (2009), the PCA is preferred for data reduction purposes while factorial analysis is generally preferred when the purpose of the search is to detect the structure of the data or the causal modeling. According to Hair et al., (2006), in most cases, both the PCA and the EFA reach the same results if the number of variables exceeds 30 or if the communalities exceed 0.60 for most of the variables.
Besides, factor loadings indicate when a factor explains a variable. The factor loadings may range from -1 to 1. Some variables may have high factor loadings on multiple factors.
Following Hair et al., (2006), this study has adopted 0.40 as the acceptable limit of the contribution of the variable in the creation of the factor to avoid the problem of the indetermination of the relationship between variables and factors. Finally, this study used rotation TAVARES, J. M.; IVANOV, S.; COBANOGLU, C. Does the repeat tourist spend more than the first time? A contribution to the literature. Revista Hospitalidade. São Paulo, volume 16, n.03, p. 118-133, 2019. Doi: https://doi.org/10.21714/2179-9164.2019 125 ISSN 1807-975X by the Varimax Criterion, which tries to minimize the number of variables strongly related to each factor. The collected material was tabulated and analyzed using the statistical package MATLAB for Windows. The questionnaires were tabulated, processed and analyzed the collected data using SPSS (Statistical Package for Social Sciences) version 11.5 for Windows.

Results
The sample included 500 respondents (55.4% are men and 44.6% female), and 64.2% of them were born between 1992 and 1998. Concerning family income, 37.6% and 35.60% receive up to U$ 1.210.00 and between U$ 1.211.00 to U$ 2.420.00, respectively. 70.3% of respondents prefer the sun tourism & mar as favorite, followed by cultural tourism/history (11.0%).
Concerning the reasons which have returned to the tourist destination, 43.2% said it the "affordable" is the main motivation, followed by the "quality of the 1st visit" (35.6%).
The variables "to user fewer guide services," "seek more involvement with the local community" and "use more public transport" were the items most likely to occur in return on the destination. On the other hand, the variables "less free to visit tourist attractions," "spend less time on destination" and "take less money to target" were items less likely to occur (Table 1). The results show that the first four factors account for 61.7% of the variability of data (Table 2). The loadings of the variables associated to the four most significant factors (of the largest variances explained) are presented below (Tables 3 and 4).  The first factor underlying the set of attributes associated with respondents' behaviour is significantly represented namely by four original variables: "buy fewer products (handicrafts, souvenirs, etc.)"; "Find restaurants less disclosed"; "Stay less time on tourist destination" and "visit less paid tourist attractions" However all the variable have loadings above than 0.40. In the same way, as in the first main component, this factor represents the tourist who seeks to minimize his travel expenses and does so, mainly, buying fewer products and getting less time in the tourist destination. Therefore, this tourist could receive label "tourist repeated independent and thrifty" The second underlying factor is representing predominantly and positively by the variables "seek more involvement with the local community" and "use more public transport".
On the other hand, the variables "visit less free tourist attractions" and "stay less time on tourist destination" are significant but in the inverse direction. Therefore, this tourist could receive the designation of "social, independent and economically engaged tourist".
The variables "Take less money to the destination" (positive signal) and "visit less paid tourist attractions" (negative signal) represent, significantly, the third factor. In this way, this tourist could receive the designation of "thrifty repeated tourist.". TAVARES, J. M.; IVANOV, S.; COBANOGLU, C. Does the repeat tourist spend more than the first time? A contribution to the literature. Revista Hospitalidade. São Paulo, volume 16, n.03, p. 118-133, 2019. Doi: https://doi.org/10.21714/2179-9164.2019 127 ISSN 1807-975X Finally, the fourth and last factor considered is significantly represented by two variables, namely, "seek more involvement with the local community" and "visit less free tourist attractions." Therefore, this tourist could receive the designation of "social and economically engaged tourist".
The KMO value of the sample is 0.695, i.e., higher than the critical threshold of 0.60 (Hair et al., 2006). Likewise, the Bartlett Test of Sphericity (BTS) -as suggested by Field (2005) -test is statistically significant because of p <0.000 (Table 5) and therefore it satisfies p <0.05. In both cases, the tests suggest that the data are suitable for factor analysis. Regarding the correlation pattern between the variables, the correlation matrix should display most of the coefficients with a value above 0.30. Table 6 shows, in a summarized way, the results of the EFA. Note that the most common characteristics of repeated tourists are "thrifty" and "independent."  Table 7 shows the results of the hypothesis tested based on EFA analysis.  Result H1 -A considerable possibility of spending less due to a prior knowledge of the destination (a thrifty tourist) Confirmed H2 -The opportunity to see or enjoy attractions that were not previously visited was an important motivation to return to the destination Rejected H3 -It is more likely that the returning tourist would be an independent touristusing less guided services, more public transport, and finding less publicised restaurants Confirmed H4 -Returning tourists are more likely to take less money to the destination, because they want to visit fewer paid attractions which they have visited previously Confirmed H5 -Returning tourists highly regard the possibility of buying fewer products that are associated with the destination (handicrafts, souvenirs, postcards, and so on.) Confirmed Source: Authors, 2018. SPSS

Conclusions
The revenue from a tourist destination is associated with the behavior of the tourist as a consumer, which may vary if the tourist is repeated or first-time, among countless other factors.
This study has found that repeated tourists adopt certain behaviors to return to the destination that may contribute to spending less than first time tourist.
The five hypotheses presented the following results. H.1 has been confirmed in this article (repeated tourist tends to spend less in reason to prior knowledge of the destination and, therefore, he would spend your money more efficiently)as obtained by Li et al., (2008) and Tjørve & Flognfeldt (2015). The H.2 has been confirmed. It means that the repeat tourist tends to be "more independent" (use fewer services guides, more public transport and find fewer disclosure restaurants). This behavior certainly implies decreasing expenses during your trip, and it follows the results obtained by Bin et al., (2009);Butler & Hannam (2012). H.3 also has been confirmed. The results showed that tourists repeat more inclined to take less money to the destination because they want to visit a smaller number of private attractive (and, therefore, to pay for tickets) but previously visited. About it, Fallon & Schofield (2004) have found that this kind of tourist prefers more social activities such as dining, and visiting friends and relatives, while the first-time tourist prefer visiting main-iconic attractions (in most of them the tourist pays for a ticket). Finally, the H.5 has been confirmed. It means that the tourists repeat consider as ISSN 1807-975X high the possibility of buying a smaller number of products associated with the destination (handicrafts, souvenirs, postcards, etc.) as found by Petrick (2004) and Lee, Lee, & Yoon (2009).
The key features of this repeated tourist are those of being predominantly "independent and thrifty," which certainly negatively impacts the local economy of a destination repeatedly visited by the tourist. The results for both the 1st component and the 1st factor -the parameters of greater weight in the analysis -are the same, reinforcing this conclusion. However, it should be emphasized that for the second and third factors, the results indicate an independent but economically engaged repeated tourist profile, that is, behaviors that contribute to the local economy.
Therefore, it seems there is a relation between the repeated tourist and impact on the local economy. If it prevails this sort of tourist, it needs to encourage them to go back to paid tourist attractions, preferably, as well as to consume local products, and to hire private transport services (transfer, city tour, tours of the region, etc.). If it prevails the first-time tourists, it is essential to invest in receivership since the first visit quality is entirely responsible for the tourist return induction. The more critical issue of this article can be verified in Wahid et al., (2016) according to the which "tourism managers need to make a clear distinction between the marketing strategies for first-time and repeat visitors." A recommendation for future research would be to use the Confirmatory Factor Analysis (CFA) to confirm the measurement model after conducting Exploratory Factor Analysis (EFA).