Abstract
The opportunity to experience nature-based activities at a destination with climate variations is a major driver of visitation for tourists. Despite significant research into seasonality and nature-based activity preferences, academic researchers are not profiling activity-oriented tourists into segments based on temporal factors such as seasons. To address this research gap, an expert panel was first asked to classify activities collected in a large secondary Norwegian tourist questionnaire into seasons. Next, 8,962 potential nature-based tourists were segmented based on summer, winter, and year-round activity preferences. When seasonality was taken into account, four clusters were identified. A combined model where seasonality was not addressed yielded fewer segments, and differing variables indicating that segmentation researchers may benefit from considering a fifth segmentation factor, namely temporal, in future. Theoretical and practical implications from this research are outlined and opportunities for future research are provided.
Introduction
The opportunity for a tourist to experience enjoyment, relaxation, and/or improvement in well-being through participating in activities that are outside of their usual lifestyle and surroundings represents a key driver of potential destination visitation (e.g., Bello and Etzel 1985; Meric and Hunt 1998; Milman 1998). Tourists will be motivated to satisfy intrinsic desires that can only be fulfilled through specific destination attributes such as nature-based activities that need to be experienced at a certain destination (Dann 1981). These activities are often affected by temporal factors such as seasonality (Gartner 1986). Most worldwide destinations are characterized by systematic fluctuations throughout the year caused by the different seasons (Baum and Hagen 1999; Hinch and Jackson 2000; Koc and Altinay 2007). Seasonality alters a destination’s climate through factors such as the variability in temperature, humidity, and snow depth (Becken 2013; Smith 1990). These variations in seasons provide certain destinations with tourism potential based on the attraction of favorable nature-based activities available at specific times of the year (Baum and Hagen 1999; Lundtorp 2001).
Countries such as Canada and Japan benefit greatly from tourism during the winter months because of the snow-oriented nature-based activities that include skiing and snowboarding, whereas Spain and the Caribbean are popular tourist destinations through the summer months because of their favorable warm climate and excellent beaches that allows for surfing and swimming. In certain countries such as Norway and New Zealand, tourists can be attracted to experience both summer and winter nature-based seasonal activities. Tourist preferences for climate-dependent attractions and the nature-based activities they provide occurs during peak seasons (Jang 2004; Spencer and Holecek 2007), and destination marketers and tourism operators aim to maximize revenue in the limited seasonal times to survive during the low “off-season” periods.
Despite tourists being motivated to participate in nature-based activities unique to a destination with seasonal variations, these activity preferences may vary between tourists based on their personal characteristics. In the past, market segmentation has been employed to enable destination marketers to better understand the differences among tourists who may frequent their destination (Perdue 1996). A plethora of studies have been conducted to provide an understanding of the types of tourists that are expected to visit a seasonal destination during a particular time period (or periods) and the activities that they wish to experience during this season (e.g., Andriotis, Agiomirgianakis, and Mihiotis 2007; Bonn, Furr, and Uysal 1992; Tangeland and Aas 2011). However, to date academics have not identified a fifth segmentation base, namely, temporal factors. This article examines temporal factors using a seasonal tourism destination to understand whether consideration of temporal factors may be warranted. Specifically, it is largely unknown whether tourism segments differ based on seasonality as a segmentation base. This research gap provides the impetus for this study. As tourists will visit a destination at a specific season to fulfill a specific motivation such as the need to participate in nature-based activities, it can be argued that tourists to a climate-variant destination may need to be segmented uniquely based on the temporal factor of season.
Literature Review
Activity Preferences
It is imperative to identify the activities that a tourist may wish to experience at a chosen destination, as it will determine the likelihood of the tourist traveling to the location to satisfy a specific need (Dann 1981). Here, it can be identified how intrinsic motivations such as a desire to escape, rest and relaxation, social interaction, and fitness can be fulfilled through specific attributes such as activities at a destination (Andreu et al. 2005; Uysal and Jurowski 1994). Some tourists may prefer shopping, city sightseeing, and the nightlife, whereas others may enjoy visiting wilderness areas, experiencing cycling, bushwalking, or going on a safari (Lang, O’Leary, and Morrison 1994). Motivation is described in terms of initiating actions and guides behavior, while preferences are portrayed as being linked more closely to the actual choice (Goodall 1991). Therefore, knowledge of potential activity preferences prior to potential travel will benefit destination marketers by identifying what is perceived as favorable and desirable among potential tourists (Tkaczynski and Prebensen 2012).
Seasonality and Nature-Based Tourism
Seasonality has long been regarded as one of the most unique and difficult facets of the tourism industry (Higham and Hinch 2002; Jang 2004). An understanding of this phenomenon is essential for predicting tourism demand (Baum and Hagen 1999; Koenig and Bischoff 2003; Lee and Jang 2013) and ensuring the efficient operation of tourism facilities and infrastructure (Butler 1994; Koc and Altinay 2007). Since the seminal work conducted by BarOn (1975), there has been a plethora of academic research into conceptualizing tourism seasonality (e.g., Baum and Hagen 1999; Butler 1994; Moore 1989). A frequently applied definition was proposed by Hylleberg (1992), who defined seasonality as the “systematic, although not necessarily regular, intrayear movement caused by changes in the weather, the calendar, and timing of decisions, directly or indirectly through the production and consumption decisions made by the agents of the economy. These decisions are influenced by the endowments, the expectations and the preferences of the agents, and the production techniques available in the economy” (p. 4). BarOn (1975) initially claimed that seasonality is caused by natural (e.g., climate, snow-depth) and institutional (e.g., religious or school vacations) factors. More recently, Butler (1994) added the dimensions of (1) social pressure or fashion, (2) sporting seasons, and (3) inertia or tradition. These additional three causes are usually combined with BarOn’s (1975) conceptualization (Frechtling 1996; Hinch and Jackson 2000; Lundtorp 2001) to define tourism seasonality.
While efforts of destination marketers seeking to reduce seasonality effects have been documented, Flognfeldt (2001) argued that tourism organizations must learn “how to live with strong seasonality” and “fit different types of tourism production into the seasonal patterns of other production activities, including an adjustment of some public services” (p. 110). Consequently, several authors have proposed strategies to minimize seasonality such as the development of product and market diversification (Getz and Nilsson 2004; Higham and Hinch 2002) which could include targeting new market segments at different time periods (Baum and Hagen 1999). Therefore, destinations that are greatly influenced by climatic conditions such as ski resorts can develop products and services that would appeal to different groups of tourists (e.g., business tourism) in the off-peak season such as conferences or retreats (Baum and Hagen 1999; Pegg, Patterson, and Gariddo 2012).
Seasonality is often inextricably linked to nature-based tourism because of the focus on outdoor, climate-oriented activities evident in both phenomena (e.g., Blamey 2001). Nature-based tourism is utilized to describe tourists’ preferences in terms of pursuing nature-based activities during their vacation (Laarman and Durst 1987). It is defined by Honey (2002) as “the travel to unspoiled places to experience and enjoy nature” (p. 1) and is depicted as a multifaceted construct including numerous activities and motivations related specifically to nature (Kuenzi and McNeely 2008; Tangeland and Aas 2011). For example, a nature-based tourist may be motivated to experience the untouched nature (Andreu et al. 2005; Park and Yoon 2009) through cycling, snowboarding, or surfing, which are dependent on the natural environment (Tangeland and Aas 2011). While the majority of these activities such as skiing (winter) and surfing (summer) may need to be experienced in one specific season to maximize the benefit of the nature conditions (Lundtorp 2001), there are certain nature-based activities that can be experienced that are not dependent on certain seasonal climates. For example, a tourist could experience a leisurely boat cruise or undertake a short bush walk in the summer, spring, fall, or winter periods. Despite the strong drawing power of highly seasonal activities for a destination’s tourism, these year-round (nonseasonal) activities can ensure that tourism organizations can attract tourists on a yearly basis to maximize revenue potential and to cater for the off-peak periods. Consequently, both seasonal and year-round nature-based activities can be frequently advertised in destination marketing material.
Despite their noted interest in nature, tourists will often vary in their interests in terms of how to use and what to do in the natural environment when vacationing (Haukeland, Grue, and Veisten 2010; Mehmetoglu 2005). For example, destinations that have a subtropical warm climate during their peak season may attract some tourists that would prefer to participate in water-based activities such as swimming and kayaking, whereas others may choose to hike or participate in beach sports. Tourists may also choose to travel alone (Mehmetoglu, Dann, and Larsen 2001) or in groups (Prebensen 2005) to experience nature-based activities in seasons that have a favorable climate. It has also been noted that while nature-based tourists share a mutual interest in outdoor-oriented activities, their activity level may vary (Andriotis, Agiomirgianakis, and Mihiotis. 2007; Mehmetoglu 2007). Active nature-based tourists may travel to a destination to experience seasonal activities such as skiing or extreme sports that require physical exertion (Jacobsen, Denstadli, and Rideng 2008; Mehmetoglu 2007). Conversely, passive nature-based tourists may prefer other activities that are not as strenuous such as taking a cruise or sun bathing (Mehmetoglu 2007; Tkaczynski and Prebensen 2012).
Temporal Factors in Market Segmentation
Largely because of the different nature-based activities that may be available during the different seasons of a destination, effective strategies need to be employed to ensure that destination marketers attract the types of tourists that are profitable and are most likely to frequent the destination. The literature suggests that the tourism market is heterogeneous and that different activities will appeal to different types of tourists (McKercher et al. 2002). Consequently, destination marketers will pay greater attention to tourists’ needs, wants, and preferences by supplying a greater variety of facilities, packages and services for tourists (Kotler, Bowens, and Makens 2010; Morrison 2009).
To cater for these differences while still upholding the key attraction(s), market segmentation is a technique that can be employed where a heterogeneous market (e.g., nature-based tourists) can be viewed as a number of smaller, homogeneous markets that can be distinguished by tourists’ (1) demographics, (2) geographics, (3) psychographics, and/or (4) behavior (Tkaczynski, Rundle-Thiele, and Beaumont 2009; Kotler, Bowens, and Makens 2010). For segmentation to be managerially useful, each segment needs to be accessible, measurable, and substantial (Sirakaya, Uysal, and Yoshioka 2003). Therefore, the segments that fulfil each of these criteria and are deemed as attractive and profitable to the destination marketers can then be targeted through specific marketing campaigns.
While market segmentation is most effective when a tourist viewpoint is taken (Uysal, Harrill, and Woo 2011), temporal factors such as seasonality may offer a fifth useful classifying criterion to profile tourists. Although certain locations close to the equator may have continual warm weather through the four seasons, the majority of countries will experience different seasons that will attract tourists to participate in different nature-based activities. Therefore, empirical examination of a fifth segmentation factor, and its consequent impact on the segments derived is needed to determine if segments that are empirically derived vary based on season (e.g., Bonn, Furr, and Uysal 1992; Calantone and Johar 1984).
To explore the relationship between seasonality and market segmentation, a review of current studies was conducted (see Table 1). This table was designed based on incorporating the five segmentation factors (demographic, geographic, psychographic, behavior, and seasonality). Each of these studies utilized tourist data and focused upon a specific season and researched activities that tourists have or would have experienced at a chosen destination. To date, studies have not empirically examined the segments derived when temporal factors such as seasonality are taken into account. Importantly, the majority (14 studies) employed the four segmentation bases as proposed by Kotler, Bowens, and Makens (2010). It can be concluded that most studies were conducted in highly seasonal destinations such as North America (12 studies) and Norway (3 studies). Furthermore, it was identified that researchers largely utilized one of two approaches to segment tourists based on seasonality and activity segmentation. The first approach focused upon a specific season (10 studies), identifying the relevant activities at the chosen destination(s) (e.g., Jacobsen, Denstadli, and Rideng 2008; Kozak and Rimmington 2000). In the majority of articles (7 studies), the peak season was targeted, with summer (6 studies) dominant. The second approach utilized activities relevant to a destination (e.g., beach activities at a summer resort) and compared the differences in participation and/or relevance of tourists across multiple seasons (e.g., Jang 2004; Spencer and Holecek 2007). All 25 studies utilized segmentation bases and variables as recommended by Kotler, Bowens, and Makens (2010). For example, Palacio and McCool (1997) identified four segments of summer-oriented nature-based tourists that differed based on their age, gender and group size (demographic), frequency and length of stay (behavioral). Conversely, Spotts and Mahoney (1993) segmented the fall tourism market and identified significant differences in tourists’ age, travel party composition, (demographic), the trip expenditure, information sources utilized, length of stay, trip planning, and past experience (behavioral).
Seasonality and Segmentation Studies.
Secondary data.
Conceptual Framework
The aim of this article is to examine whether the incorporation of a fifth factor (temporal) in segmentation studies can better capture heterogeneity in the tourism market. Despite considerable research (see Table 1) into segmentation and seasonal activity preferences (e.g., Bonn, Furr, and Uysal 1992; Canavan 2013; Spotts and Mahoney 1993), to date tourism research has failed to identify whether segments vary if temporal factors are included in segmentation analysis. While some studies have aimed to compare and contrast ratings of tourists in the different peak periods (e.g., Bonn, Furr, and Uysal 1992; Jang 2004), these studies tend to focus on activities that may be essentially relevant for only one particular season. For example, identifying the importance of off-peak seasonal activity such as sunbathing and jet skiing in winter may limit the profiles derived for destination marketing purposes by artificially inflating activities that are not possible at certain time points. Differences in activity participation and/or importance are likely to be evident, as tourists’ intrinsic motivations may only be able to be fulfilled at a specific destination during one particular season. This research aims to identify if the number of segments derived varies when temporal factors are incorporated into segmentation analysis. The first research hypothesis follows:
Hypothesis 1: The number of segments will increase when temporal factors are incorporated in segmentation analysis.
A second hypothesis has been proposed that aims to determine the relevance of year-round (nonseasonal) activities in the segmentation of tourists at a destination influenced by climatic conditions. As destination marketers aim to diversify their product offerings to cater for potential tourism markets in off-peak seasons, identifying the significance of year-round activities for yield management and financial survival is required. This research will address this issue and a second hypothesis is listed as follows:
Hypothesis 2: There are significant differences in tourist characteristics to a seasonal destination based on their year-round nature-based activity preferences.
A third and final research hypothesis has also been presented that aims to determine if utilizing all four segmentation bases as proposed by Kotler, Bowens, and Makens (2010) can provide a more holistic view of tourists. Recall that for segmentation to be purposeful, it needs to be accessible, sustainable, and substantial (Sirakaya, Uysal, and Yoshioka 2003). This research, therefore, will aim to further contribute to the literature (see Table 1) by identifying if tourists with nature-based activity preferences can be profiled based on personal characteristics of relevance to academics and practitioners. The third hypothesis follows:
Hypothesis 3: There are significant differences in nature-based activity–oriented tourists at a seasonal destination based on their profiling characteristics.
Destination
Norway is a highly seasonal country with a variety of summer and winter activities that aims to attract tourists during the different seasons (Innovation Norway 2011). The climate in Norway is relatively cold, situated on the same latitude as Alaska and Siberia. Norway has steep mountains and deep fjords indicating vast potentials for a variety of nature-based activities. This is further reflected in the promotional materials (Visit Norway 2012). Wintertime (October–April) includes temperatures from +15 to −40 degrees Celsius. Conversely, in the summer season (June–August), the temperature varies between 0 and 30 degrees Celsius (Bjørbæk 1998). There are nature-based activities that can be experienced throughout the whole year such as walking short and long trips in nature and taking a cruise along the coastal Fjords (Visit Norway 2012).
In 2010, there were more than 28 million tourist nights in Norway, which represented a 2% growth from 2009 (Statistics Norway 2011). Domestic tourism increased by 1% (72% of total). International tourism rose by 5% to approximately eight million commercial guest nights, with key source markets such as Sweden (942,070 an increase of 6%) and France (318,072 an increase of 7%) experiencing growth from the previous year. In 2010, 53% of all commercial guest nights were in the summer months between May and August (Statistics Norway 2011). The lowest period was in the fall and winter months of October to December, which represented approximately 200,000 commercial guest nights per month.
Methodology
The present study utilized data provided by Innovation Norway, a company mainly funded by the Norwegian state that is responsible for managing Norwegian trade and tourism (Innovation Norway 2011). In 2009, Innovation Norway examined residents who had an interest in taking a nature-based vacation in Norway. The focus of this study was on three international markets (Sweden, Britain, and France) and the domestic market (Norway). These markets were focused on by Innovation Norway as they were argued to have the highest growth potential for Norway (Statistics Norway 2011).
The data collection conducted by Innovation Norway was based on a sequential procedure of three phases. First, 1,000 phone interviews were conducted in each of the four potential markets to reveal possible interest in Norwegian nature-based vacations. The telephone interviews revealed nature-based target groups in the respective countries representing between 30% and 50% of the interviewed residents. For example, nearly half (47.2%) of French residents were within the nature-based target group. Respondents in this phase functioned as a basis for the convenience sample performed for the online questionnaire, which represented the second stage. Specifically, this questionnaire was sent to patrons from these key markets that visited the Innovation Norway website and acknowledged an interest in nature-based vacations. Third, within the online questionnaire, an additional procedure was carried out to select tourists with a previous history who would potentially visit Norway within the next three years for a nature-based vacation. An online questionnaire was used as the primary data collection method because of its ability to reach a large number of individuals in a timely and cost-effective manner (e.g., Van Selm and Jankowski 2006). In total, 8962 questionnaires were returned from the major tourism markets. A total of 2,010 French, 2,891 British, 2,034 Swedish, and 2,027 Norwegian residents completed the questionnaire.
The questionnaire was based on interviews with representatives from the tourism industry, previous studies on Norwegian tourism, and international statistics. The questions included age, education, employment, income, marital status (demographic), country of origin (geographic), activities (psychographics), expenditure, frequency [of travel to Norway], length of stay, mode of transportation, and trip planning (behavioral). Age, activities, frequency, and length of stay were treated as continuous variables with the remaining eight variables designed as categorical. The 18 nature-based activities were measured in a binary “yes” or “no” format. Prior to analysis, income and expenditure were recoded to allow comparisons to be made across respondents from the four countries. Based on a review of income from leading statistical agencies in each of the countries (e.g., Statistics Sweden for Sweden), an average income level was identified. Respondents were classed as either “high” or “low” based on whether they were above or below this average income level through a detailed analysis in IBM SPSS version 20.0 employing frequency distributions. Using a similar approach, expenditure levels were coded into three levels, low, mid, and high.
Expert Review to Classify Nature-Based Activities
As a second phase, expert judges were used to classify activities into seasons for the 18 nature-based activities. Scale development recommendations proposed by Bearden and Netemeyer (1999) and Hardesty and Bearden (2004) formed the basis for the expert review. An email was sent to experts from the Norwegian Tourism Industry Group asking members to classify each of the activities into “summer,” “winter,” or “year-round” categories. A reminder email was not sent to respondents who did not initially complete the requested information. Eleven experts (28.9%) completed the task and responses were tallied (coded as 1 if included, 0 if excluded). In a few instances, experts classed an activity as both summer and year-round to indicate that this variable was essentially a summer activity but could also be experienced in the winter periods. An activity with a majority score in one category was assigned to one of the three activity categories. Items were deleted that did not have a majority rating (at least 60%) for at least one of the three constructs measured (Hardesty and Bearden 2004). In total, four summer, four winter, and six year-round activities were identified by the experts. For example, “sun and bathing” and “going on a cruise (fjords)” received 100% scores for summer and year-round activities respectively. Of note, the four winter activities also received 100% scores from all respondents, indicating experts were unanimous in defining what represents a Norwegian nature-based winter activity.
TwoStep cluster analysis using the log-likelihood measure in IBM SPSS version 20.0 was employed to segment nature-based tourists to Norway. Each of the three seasonal models (summer, winter, and year-round) was derived. To validate each model, three measures were required. First, when employing the Bayesian information criterion (BIC) for statistical inference, the silhouette measure of cohesion and separation was required to be at or above the required level of 0.0. This would ensure that the within-cluster distance and the between-cluster distance was valid among the different variables (Norusis 2011). Secondly, chi-square and t-tests were performed on the categorical and the continuous variables respectively to indicate significant differences among clusters. Third, the input (predictor) importance was measured to determine the importance of variables in a cluster. A variable with a rating between 0.8 and 1.0 was highly important to the cluster formation, whereas if the rating was below 0.0 and 0.2, this variable was less important (Norusis 2011).
The three models were then compared and contrasted once the solutions were finalized. To be accepted, these three validation measures needed to be upheld and similarities across all three models such as number, size, and characteristics of the three cluster models were required (e.g., Tabachnick and Fidell 2012). To further validate the model, all 16 nature-based activities were collectively entered into TwoStep cluster analysis in addition to the demographic, geographic, and behavioral variables using the same outlined procedure. The aim of this process was to determine if similar segments can be identified with nature-based activities combined as opposed to being separately run based on the seasons.
Results
Initially two clusters were identified in both the summer and the winter models. In both instances, the silhouette measure of cohesion and separation was above the required level of 0.0, which warranted further analysis (Norusis 2011). In both models, the variable age was insignificant (p > 0.05) and had the lowest importance for both clusters (p < 0.0). Therefore, age was removed from both models. After rerunning the models, four valid clusters were identified and all variables had input (predictor) importance models (see Table 2). It was determined that for the year-round activity model, four of the variables (expenditure, length of stay, and two nature-based activities) were insignificant (p > 0.05). In evaluating the model, it was identified that the average age of respondents in the four clusters was similar, being mid- to late 40s. As firstly segmentation is most effective when differences are identified among segments (Dolnicar 2008; Loker and Perdue 1992) and secondly age was found to be an insignificant variable for the summer and winter models, this variable was removed. In rerunning the models, the nature-based activities “fishing” and “Visit North Cape” were not eliminated. In the third model, the remaining four year-round activities and the other profiling variables were validated and four clusters were identified (see Table 3). The results, therefore, supported all three hypotheses.
Cluster Input (Predictor Importance).
Final Cluster Model.
Note: TAFE = technical and further education.
In examining the constructs of each of the three clusters, it was identified that income, origin, and transportation were the most important variables for the cluster formation for the summer, winter, and year-round activity solutions. Trip planning and education were also extremely pertinent. Despite it being acknowledged that the unique seasonal activities were relevant for the formation of the four clusters in each of the models, it was determined that they were not the most important predictors of the clusters. All nature-based activities apart from “cruise (fjords)” for the year-round model were below the 0.2 level.
Table 3 was presented to compare and contrast the three models. The distribution of respondents among the four clusters was relatively equal for the three models, with clusters one and four representing approximately a quarter of the sample. The size of cluster two and three varied slightly for the summer, winter, and year-round models. Experiencing the “midnight sun” (summer), “northern lights” (winter), and “fjords” (year-round) were the most popular nature-based activities among the clusters. In all three models, being employed full-time and married/cohabit/couple was dominant. Minimal differences were identified in the respondents’ origin, transportation, trip planning, and frequency for the three models. Despite not being the dominant country of origin, the number of Swedish respondents varied for the second and third clusters in the three models. The third cluster of winter activity–oriented residents was also less sure of their form of transportation. While the winter and year-round respondents had similar trip planning preferences, the summer activity model concluded that organize the trip myself and travel alone was the most popular travel option. The frequency of travel for the second cluster of all three models was also varied.
The combined model (with all nature-based activities) initially produced six clusters, with an average silhouette measure of 0.0 which warranted further analysis (Norusis 2011). Origin and nature-based activities were invalid in discriminating the clusters and were removed from further analysis. TwoStep cluster analysis was rerun with the same average silhouette measure. Three clusters were produced and length of stay and another seasonal activity, “sun and bathing,” were invalid and insignificant and were therefore extracted. Three clusters were validated and the results are listed in Tables 2 and 4.
Combined Cluster Model.
Note: TAFE = technical and further education.
In comparing the cluster input importance (Table 2), the combined model shared similarities with the seasonal model in that income was highly significant (1.0) and the majority of nature-based activities were not (<0.2). However, age [not in the seasonal solutions], employment, and marital status were not identified as critically important in the summer, winter, and year-long solutions. Conversely, transportation, trip planning, education, and origin [not in the combined solution] had a lesser level of importance in this final solution.
Some aspects of the combined model were comparable to the individual season models. The activity preference percentages were similar across the models. For example, “experiencing the midnight sun” was between approximately 60% and 75% for the summer and combined models. Furthermore, the most popular summer, winter, and year-round activities were also identified in the combined model. The educational levels were similarly ranked at the higher levels and the majority of respondents were married/cohabit/couple and would prefer to organize the trip themselves and travel alone. Major contrasts in the models were also identified. The first critical difference was that the number and size of the clusters was completely different between the seasonal and the combined models. Next, the variables used to derive the combined cluster solution varied (origin and length of stay were omitted in the combined model, noting that origin had a high importance in classifying in seasonal models). Furthermore, the income level was further toward the lower level for the combined model, and the frequency (because of its low importance) was similar across the three clusters whereas it was a greater distinguishing variable in the seasonal models.
Discussion and Conclusions
Preferred activities represent a driver of tourism to many seasonal destinations. Certain destinations will experience different seasons that will attract different tourists largely because of the “various regular and recurring temporal changes in natural phenomena usually associated with climate variability, and as a result, influence demand for specific activities” (Andriotis, Agiomirgianakis, and Mihiotis 2007, p. 60). This research has contributed to the literature by segmenting potential tourists to Norway based on seasonality and their nature-based activity preferences. Three hypotheses were proposed and supported, which lends support for tourists to seasonal destinations to be segmented based on their seasonal and year-round activity preferences in addition to their personal characteristics.
Theoretical Contributions
The first theoretical contribution of this research is that it is the first known study to segment tourists into different clusters based on their preferred nature-based activities for summer, winter, and year-round seasons. Previous segmentation studies in the tourism and marketing literatures (e.g., Andriotis, Agiomirgianakis, and Mihiotis 2007; Jang et al. 2005; Palacio and McCool 1997) have employed between one and four segmentation bases to derive segments for seasonal destinations, namely (1) demographic, (2) geographic, (3) psychographic, and (4) behavior. This study provides a unique contribution to tourism literature suggesting that consideration of a fifth segmentation factor may be warranted. Specifically, the results of this study suggest that temporal factors such as seasonality may provide destination marketers with considerable improvement in the descriptive capability of segments derived when clear summer and winter seasons are evident.
This research confirmed the literature that respondents have varying profiling characteristics that differentiated themselves based on specific preferred activities (e.g., Andriotis, Agiomirgianakis, and Mihiotis 2007; Figini and Vici 2012; Masiero and Nicolau 2012; Spotts and Mahoney 1993). However, despite the focus on activities within the tourism literature as the driver of destination visitation (Jacobsen, Denstadli, and Rideng 2008; Tangeland and Aas 2011), these variables were among the least significant in differentiating the segments in all models. Conversely, demographic and behavioral characteristics were most important in defining the seasonal segments. Consequently, segmenting potential activity-based tourists in addition to variables such as income (Calantone and Johar 1984; Jang et al. 2005) and length of stay (Spencer and Holecek 2007; Thrane ad Farstad 2011) makes it possible for destination marketers to anticipate better development trends and supply well-diversified products that could satisfy certain requirements of tourists (Andriotis, Agiomirgianakis, and Mihiotis 2007).
A key finding from this research is that the most popular activity for each of the three solutions, “experiencing the midnight sun (summer),” “experiencing the northern lights (winter),” and “experiencing fjords (year-round)” represent sightseeing, which is consistently identified as an important activity for destination visitation (e.g., Lang, O’Leary, and Morrison 1994; Mehmetoglu 2007). Therefore, while nature-based activities (e.g., bicycling) may be prominent with certain segments, the key sightseeing activities that are promoted by Visit Norway are desired by potential nature-based tourists. However, contradictory to the majority of literature (Andriotis, Agiomirgianakis, and Mihiotis 2007; Calantone and Johar 1984; Thrane and Farstad 2011), age, which is cited as the most frequently employed destination market segment variable (Tkaczynski, Rundle-Thiele, and Beaumont 2009), was not a major discriminator of the four clusters in the seasonal models. However, other variables such as origin and length of stay, which are frequently acknowledged as relevant for segmentation purposes in the literature (Bonn, Furr, and Uysal 1992; Masiero and Nicolau 2012), were important in this study.
The final contribution of this research is that it exemplifies the importance of year-round activities in the formation of segments. Despite studies being conducted to target tourists in off-peak periods (Figini and Vici 2012; Kozak and Rimmington 2000) or to compare tourists in different seasons (Jang et al. 2005; Spotts and Mahoney 1993), this study contributed to the literature by identifying four significant activities that were year-round at a destination that has extreme temporal conditions. While this research acknowledges that different activities appeal to different types of tourists (Kardes 1999; Madrigal and Kahle 1994), activities such as experiencing fjords and the coast and culture are popular to all groups of segments. Therefore, to minimize seasonality issues, it is essential that product and market diversification for seasonal destinations are considered (Getz and Nilsson 2004; Higham and Hinch 2002) to target these tourists to uphold survival throughout the year (Baum and Hagen 1999).
Practical Implications
This research produced many implications that will contribute to practical knowledge of the segmentation of nature-based activity–oriented tourists to a destination that experiences seasonal variability. This study’s major practical implication is that it identified the types of tourists who will likely experience a nature-based activity–oriented vacation in Norway during different seasons. This study determined that Norway is attractive to tourists in both the summer and the winter seasons. For example, some of the seasonal variables such as “experiencing the midnight sun (summer),” “experiencing the northern lights (winter),” and “experiencing fjords (year-round)” had at least a 50% rating by all respondents in the three seasonal models as a potential nature-based activity to experience. However, it is noticeable that the first cluster rates these activities higher than the three other clusters, with a large difference noted with the two largely domestic clusters (clusters one and three). In that at least half of all respondents would choose to experience these offerings, these activities should still be in the promotional material, as they seem to be important and likely to attract tourists.
A second practical implication is that largely because of their high predictor importance, several of the variables were extremely important in differentiating the four clusters. For example, income level (high vs. low), origin (Norway, Sweden, United Kingdom, and France), and mode of transportation (car, plane, don’t know) were crucial and need to be considered when planning for segmentation purposes. As it is acknowledged that the size of the clusters is extremely similar across all three models, the Norwegian destination marketers need to determine which cluster(s) they wish to target based on its likely potential. The first cluster does have the highest education and expected expenditure level in addition to a high level of income and length of stay. This segment was almost universally from France. Consequently, particular promotional materials advertising the most popular activity in the summer and winter months could be utilized. Conversely, the second and fourth clusters come largely from Scandinavia (Norway and Sweden). While their level of expenditure is lower than the other two clusters, they travel far more frequently and spend a similar amount of time on a Norwegian nature-based vacation. Furthermore, the cost of promoting to these tourists would likely be less expensive because of their familiarity of Norway. The fourth cluster may be particularly attractive because of the higher full-time employment, income, and expenditure levels. Consequently, professionals within Scandinavia can be targeted with promotional material.
Limitations and Opportunities for Future Research
Despite producing many theoretical and practical contributions, this study is not without its limitations. The first limitation of this research is that it focuses on activity preferences. This, therefore, does not reflect actual visitation. However, research suggests that people’s perceptions and attitudes in addition to their past experience can predict future behavior (e.g., Ajzen and Fishbein 2010; Ajzen 1991). To validate the research findings, it is recommended that a follow-up questionnaire be used to identify if the sample within this study did experience a nature-based vacation. Future research can also be conducted to gain insights into the predictive capability of the segments derived.
A second limitation of this research is that it utilized a cross-sectional design in which data were collected in the winter month of February. An opportunity for future research is to conduct research over two specific peak seasons in Norway (summer and winter) to ensure that an accurate representation of the sample is obtained. Here convenience sampling can be used to identify if the size of the seasons (e.g., winter) is larger than the other peak period. Future research should also be conducted longitudinally to see if the segments described in this study are better able to predict the types of tourists traveling to the destination. A final limitation of this research arises from the data source. The current study utilized a large secondary source which indicated that the research instrument was initially designed for purposes other than this study. This suggests that questions and response rates were unable to be controlled. Future research in seasonal segmentation should ensure that the research design is primary to alleviate these issues and provide a more representative sample.
This article has focused upon nature-based activity preferences as a driver for visitation to Norway. It has been acknowledged in previous research (e.g., Jacobsen, Denstadli, and Rideng 2008; Prebensen 2005) that these unique temporal conditions will attract potential tourists to this country. While special-interest tourists (e.g., ski tourism enthusiasts) may require specific activities (e.g., skiing) that can only be experienced during finite time periods, other tourists to Norway may be constrained by nonclimatic seasonal factors such as institutional (e.g., school vacations) or social pressure that are outlined within the literature (e.g., BarOn 1975; Butler 1994). Consequently, a final recommendation for this research is to identify if the choice to visit the country or the season to visit is the primary determinant for visiting a climate-oriented destination such as Norway. Here, it could be identified that the nature-based activity–oriented respondents in this study that outlined a specific seasonal temporal nature-based activity (e.g., experiencing the midnight sun) did in fact visit Norway in the winter months.
A final recommendation of this research is to further examine the level of participation in each of these activities among potential tourists. Despite the study respondents’ acknowledging that they may participate in activities such as skiing, it is unknown if this would likely relate to their total holiday experience or only a minimal time period. As it has been acknowledged that Norway attracts many people for both summer (Prebensen 2005) and winter (Jacobsen, Denstadli, and Rideng 2008), level of participation may be of interest.
Footnotes
Acknowledgements
The authors thank Innovation Norway for access to the data and for positive feedback. The research in the present study is part of the research program “Service Innovation and Tourist Experiences in the High North: The Co-Creation of Values for Consumers, Firms and the Tourism Industry,” financed by the Norwegian Research Association, project No: 195306/140.
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research in the present study is part of the research program “Service Innovation and Tourist Experiences in the High North: The Co-Creation of Values for Consumers, Firms and the Tourism Industry,” financed by the Norwegian Research Association, project No: 195306/140.