Interpretive media that attract park visitors and enhance their experiences: A comparison of modern and traditional tools using GPS tracking and GIS technology

Tourism Management Perspectives 7 (2013) 59–72

Contents lists available at SciVerse ScienceDirect

Tourism Management Perspectives journal homepage: www.elsevier.com/locate/tmp

Interpretive media that attract park visitors and enhance their experiences: A comparison of modern and traditional tools using GPS tracking and GIS technology Isabelle D. Wolf a,⁎, Heidi K. Stricker b, Gerald Hagenloh a a b

NSW National Parks and Wildlife Service, Office of Environment and Heritage, Department of Premier and Cabinet, Hurstville, NSW 2220, Australia School of Biological, Earth and Environmental Sciences, University of New South Wales, Sydney, NSW 2052, Australia

a r t i c l e

i n f o

Article history: Received 28 March 2013 Accepted 3 April 2013 Keywords: Interpretive media Audio tour Information pamphlet Signage GPS tracking Visitor satisfaction Learning

a b s t r a c t Advances in technology have expanded the range of media available for interpretation in parks however their effectiveness in nature-based settings has not been well established. This study compared the performance of modern technical media, namely a GPS-triggered multi-media tour and an MP3-player audio tour, with traditional media: text-rich versus image-rich pamphlets and signs. Performance was evaluated by a questionnaire-based survey along with GPS tracking of visitors who used the different media along a scenic walking track in an Australian national park. The GPS tracking proved to be an efficient and versatile tool to ascertain three performance measures for interpretive media, specifically, the attracting, holding and distracting powers of interpretive media. The latter is defined first in this study as their power to encourage people to visit attractions off the main path. The GPS navigation tour performed well compared to traditional media in achieving an intermediate attracting power, the highest distracting power and the highest holding power. Compared to the audio tour, it was rated more highly for the overall experience with the medium and for facilitating fun. Further, visitors were more willing to provide word-of mouth recommendation for the GPS navigation tour. Both modern media achieved the highest satisfaction ratings for discovery and learning and were most efficient at facilitating factual learning. Traditional media were more conducive to socialising and more relaxing and consistent with a nature-based experience. Signage outperformed pamphlets by achieving stronger attracting and holding powers, higher overall satisfaction with the medium and greater word-of-mouth recommendation. The minor differences between image-rich and text-rich media were that the former received a higher satisfaction score for facilitating a fun experience but it achieves a lower degree of factual learning. Whilst our study demonstrated that modern technical media can be effective tools for park interpretation, traditional media continue to play an important role in nature-based experiences without the intrusion of technology. © 2013 Elsevier Ltd. All rights reserved.

1. Introduction Effective interpretation is an important tool for park managers in attracting visitors and enhancing their experiences in national parks (Light, 1995; Moscardo, Woods, & Saltzer, 2004; Tubb, 2003). Park interpretation aims to educate and increase visitor appreciation and awareness of the cultural and natural heritage aspects of a specific site (Hughes & Morrison-Saunders, 2002a; Hwang, Lee, & Chen, 2005), and fosters place attachment and identification with a park (e.g., Ballantyne, Packer, & Beckmann, 1998; Brody, Tomkiewicz, & Graves, 2002). Successful park interpretation draws visitors away from competing leisure attractions and encourages return visits to sites where people experienced high levels of satisfaction (e.g., Bramwell & Lane, 1993). The latter may result in tourists recommending the tour to others which is an important means to attract new visitors (Munro, ⁎ Corresponding author. Tel.: +61 4 0330 3550; fax: +61 2 9585 6601. E-mail address: [email protected] (I.D. Wolf). 2211-9736/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tmp.2013.04.002

Morrison-Saunders, & Hughes, 2008). Additionally interpretation is an appropriate tool for guiding visitors in space and time enabling them to select places or features within the park that may be particularly attractive or meaningful to them (Bramwell & Lane, 1993). Given that there is a range of interpretive media available and that interpretation programs are costly to implement and maintain, it is imperative that research is undertaken to evaluate and identify the most effective media for a specific recreation setting (Light, 1995; Munro et al., 2008). Numerous studies have assessed interpretive media in indoor settings such as galleries, museums and visitor centres (e.g., Alt & Shaw, 1984; Moscardo & Pearce, 1986; Peart, 1984; Screven, 1976; Washburne & Wagar, 1972). In contrast, far fewer studies have evaluated interpretive media for outdoor settings such as national parks (e.g., Brody et al., 2002; Light, 1995; Ruchter, Real, & Düpmeier, 2005). The results of studies from indoor settings are not always applicable outdoors where interpretive media need to be particularly effective at competing with a multitude of random sensory stimuli from the environment and visitors' need to pay attention as they travel along

60

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

recreational tracks and explore sites (Hughes & Morrison-Saunders, 2002a; Rademaker, 2008). In addition to this, people's motivation to learn and therefore their incentive to engage with interpretive media may be greater in museums and other educational indoor settings (Light, 1995; Novey & Hall, 2007). We focus on the evaluation of self-guided, traditional (pamphlets, signage) vs. modern (audio MP3 tours, GPS-triggered multi-media tour) media for a walking trail in a well-frequented national park in the Sydney metropolitan area. GPS-triggered multi-media tours can be delivered via smart-phone applications or on GPS navigation devices as points of interests (POIS). Both technologies have become very popular to retrieve location-based information whilst travelling and compared to others, radio-frequency identification systems (Ferrer, Dew, & Apte, 2010) for example, they are comparatively easy to provide as the user supplies the equipment rather than park management. This is a particular advantage for large park systems where many parks need to be serviced with appropriate interpretive information. The central question in our paper is how modern technical media compare to traditional media as tools for park interpretation. We hypothesised that people prefer traditional media over modern media as the technicality of the latter may distract from the outdoor experience in the natural environment, which is a major motivation to visit national parks. However, modern media were expected to enhance short-term factual learning and to have a higher holding power as they likely increase visitor's attention span. To evaluate the effectiveness of the different media types we used a questionnaire-based survey in combination with the GPS tracking of visitors, which is a comparatively new and insightful technique that generates high-precision data on visitor movements (Shoval & Isaacson, 2007; Wolf, Hagenloh, & Croft, 2012). To our knowledge this is the first study that compares the performance of interpretive media based on an in-depth evaluation of spatio-temporal visitor data. We provide insights into the effectiveness of GPS tracking and GIS for such an evaluation process and how well this approach complements more traditional techniques. Our study exemplifies how the implementation of GPS tracking and GIS technology is crucial to generate reliable and novel performance measures that assist park management in decision-making on the types of interpretive experiences that best meet their requirements. 2. Literature review 2.1. Range of interpretive media When implementing interpretive programs along recreational tracks, there is a range of self-guided interpretive media available from which park managers can choose. Museum studies have shown that people actively engage with interpretive media and that the way material is communicated to them influences their level of absorption (McManus, 1988, 1989). People may briefly read parts of interpretive text in order to confirm the ‘predicted’ meanings of the object which is being interpreted (McManus, 1989), and to decide whether to further engage with the object and the interpretive medium itself. Print media such as pamphlets and trail-side interpretive signage that alert and inform visitors of park attractions are traditionally used. Pamphlets have proved to be effective tools for conveying information to park visitors (Andereck, 2005; Brody et al., 2002; Moscardo et al., 2004). They are useful as they can be read at any time during the walk and retained for later reference (Moscardo et al., 2004). Pamphlets may also be perceived as less visually polluting or distracting than signage which is permanently installed in the environment. In contrast, interpretive signage, another common means for visitor interpretation, provides access to information at a specific location of interest (Moscardo et al., 2004). Signage is known to enhance self-guided trails and allows visitors to move at their own

time and pace (Moscardo et al., 2004). In comparison with modern technical media, print media pose no technical difficulty that may distract visitors from enjoying their tour (Light, 1995). However, a noted drawback of signage may be its inadequacy of interpreting fauna wherever wildlife sightings occur randomly in space (Moscardo et al., 2004). There are also other aspects of traditional media that may reduce the quality of the visitor experience. For example users of pamphlets may need to rely on a map to find the location of interest, whereas no such additional effort is required for other interpretive media (Ruchter et al., 2005) such as GPS-triggered multi-media tours. Even with interpretive signage, visitors have a chance to engage with panels that they have discovered serendipitously by following a marked track. Given that visitors can opt to read pamphlets at the sites which they interpret or elsewhere, may entail that people do not read the content at all or miss out on a possibly more engaging and memorable form of on-site interpretation. An important choice in print media relates to the presentation of information because quantity and layout of text and the number and size of images significantly influence their effectiveness (Bitgood, 2000; Cole, Hammond, & McCool, 1997; Hughes & Morrison-Saunders, 2002a; Moscardo, 1999b). Such design features should be optimised to maximise attention span in visitors (Bitgood, 2000; Cole et al., 1997; Ham, 1992). According to Bitgood (2000) “. . . capturing visitor attention is the first step in the interpretive process. Visitors must first pay attention to a label before it has any chance of delivering an interpretive message”. High-intensity use of signage on trails may result in information overload and visual pollution as evidenced by previous studies (Bramwell & Lane, 1993; Cole et al., 1997; Hughes & MorrisonSaunders, 2002a, 2005). In the pursuit of enhanced visitor satisfaction and short term retention of new facts, there is a fine balance between the number and size of panels and any resulting visual impacts (Hughes & Morrison-Saunders, 2002a,b, 2005). Recent advances in park interpretation include the use of modern technical media that deliver auditory and visual (text, imagery) tour content via conventional MP3 players or multi-media tools such as GPS navigation devices and smartphones. These tours may be very attractive to visitors as they are novel, multi-sensory and audio-visual and require the least amount of effort for consuming interpretive information (Light, 1995; Moscardo, 1996). There is evidence that people are more responsive to dynamic multi-sensory and interactive experiences, which enhance visitor learning and satisfaction levels (Light, 1995; Moscardo, 1996; Novey & Hall, 2007; Rademaker, 2008). Modern media that include an audio component can be particularly effective at raising the attention span of visitors leading to higher levels of engagement than traditional media (Light, 1995; Novey & Hall, 2007; Peart, 1984). For example Davidson, Lee, and George (1991) found that for a museum exhibit 72% of visitors listened to the recorded labels as opposed to 34% who read them. Given the diverse range of visitors and ages that utilise national parks, park management needs to decide on the level of control and technicality of the device such as a manual or automatic trigger of audio tour content (Moscardo, 1996; Rademaker, 2008). An automatic trigger assists in locating tour stops (Ruchter et al., 2005), and may encourage people to explore sights which they may have otherwise missed. However, it may also diminish the joy of discovery and inhibit visitors' choice of path (Rademaker, 2008; Ruchter et al., 2005). In addition, some people may react negatively to the loss of autonomy to decide whether or not to trigger the tour and therefore experience lower levels of satisfaction (Rademaker, 2008). Although the novelty of modern media may be very attractive to some visitors, others may consider them in conflict with the naturalness of the visitor experience in a national park. Whilst GPS navigation devices may be considered less visually and physically intrusive than a sign, the audio components may conflict with visitors who seek a quiet walk away from urban noise. Simply dealing with the

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

technical device can frustrate some people, especially those who consider a nature-based experience as an escape from their technologically dominated lives (Buckley, 2000; Rademaker, 2008). 2.2. Evaluation of the effectiveness of interpretive media Selection of appropriate media for visitor interpretation in outdoor environments is challenging and requires a robust process of evaluation (Hughes & Morrison-Saunders, 2002a). In this study we used four dependent variables as performance measures; namely, visitor satisfaction, word-of-mouth recommendation, short-term factual learning and guidance of visitors through space and time. The former three are well-established performance indicators of interpretive media (e.g., Moscardo, 1999a; Prentice, Guerin, & McGugan, 1998) that are commonly assessed through pre- and post-experience questionnaires (Brody et al., 2002). The level of influence which interpretive media exert over the spatio-temporal behaviour of visitors is typically measured in the form of their attracting power (percentage of visitors that stop at a point of interest) and holding power (duration of stay) (Bitgood, 2000; Screven, 1976) since both measures were found to be responsive to differences in media types (Davidson et al., 1991; Peart, 1984). The variation in the response may be a direct measure of the amount of interest and attention visitors afford towards tour stops in response to different media (Novey & Hall, 2007). The holding power is not only a performance measure but also informs about the optimum length of time for tour stops (Hughes & Morrison-Saunders, 2002a). For the purpose of this study, an additional term has been defined as ‘distracting power’. This measurement describes the number of detours upon which people embark given the power of a medium to distract them from the main track. To capture in-depth data on the attracting, holding and distracting powers of interpretive media we tracked visitors by GPS and used GIS for analysis. The use of GPS tracking is a relatively new visitor monitoring method (McKercher & Lau, 2008; Shoval & Isaacson, 2006, 2007) compared to more traditional means such as, direct observations (Hartmann, 1988), itinerary mapping (Connell & Page, 2008) or self-administered travel diaries (Fennell, 1996). GPS tracking is particularly useful for gathering data where low sample sizes of visitor groups are dispersed over an area where spatial extent impedes direct observation (Orellana, Bregt, Ligtenberg, & Wachowicz, 2012; Shoval & Isaacson, 2007). Using this technique, consistent data can be gathered from several visitors along the entire trail of a park vs. sporadic and patchy observations of a random sub-sample of visitors by direct observation (Shoval & Isaacson, 2007). In addition, GPS tracking provides a more accurate, economical and unobtrusive way to acquire ample data on spatio-temporal behaviour of tourists (Wolf et al., 2012; Xia, Arrowsmith, Jackson, & Cartwright, 2008). Evaluation of interpretive media needs to be undertaken in light of visitor characteristics as they significantly influence the overall effectiveness of a medium (Light, 1995; Prentice et al., 1998). These factors include; motivations for visit, whether the visitor was alone, or as part of a group, the length of time the visitor spends at the site, age, gender and social class (Ballantyne et al., 1998; Light, 1995; Moscardo, 1999a; Prentice et al., 1998; Wearing & Schweinsberg, 2008). Frequency of visitation is another, particularly important factor that influences visitor response to media and their satisfaction with the interpretive experience (Ballantyne et al., 1998; Hughes & Morrison-Saunders, 2002b; Wearing & Schweinsberg, 2008). For example, previous research has found that the intentions of first time visitors differ to those of repeat visitors, in that the former were more likely to be interested in exploring a recreation site than repeat visitors who are more motivated by relaxation and relationship building (Ballantyne et al., 1998; Gitelson & Crompton, 1984; Hughes & Morrison-Saunders, 2002b; Lau & McKercher, 2006; Li, Cheng, Kim, & Petrick, 2008). The tendency for first time visitors to be more

61

exploratory, whilst repeat visitors limit their activities and visit fewer attractions is also reflected in larger scale studies such as those of Oppermann (1999) and McKercher, Shoval, Ng, and Birenboim (2012) who respectively compared first time and repeat visitors to New Zealand and Hong Kong. Such differences in the motivation to visit may have an important influence on the effectiveness of interpretive media including the amount of learning achieved (Light, 1995). The above factors demonstrate that there are several aspects of the provision of interpretive media for visitors to natural areas that park managers must consider to ensure that their interpretive programs are effective. In essence, park managers need to provide a flow of information that guides people strategically through the park, similar to the manner in which the travel industry organises tourist flows by the intensive use of signs and images (Lash & Urry, 1996). 3. Methods 3.1. Study area and tour stops We conducted our research along the Bradleys Head walking trail in Sydney Harbour National Park. The park is renowned for its scenic views of Sydney Harbour, European and Aboriginal heritage sites and offers ample opportunity to view unique Australian flora and fauna. Bradleys Head is a 0.4 km 2 peninsula located on the north shore of the Sydney Harbour National Park, which conserves much of the Sydney harbour foreshore and islands. Its close proximity to the city centre and easy accessibility make Bradleys Head a desirable location for both local residents and visitors to Sydney. Estimates of visitor numbers to Sydney Harbour National Park were close to 1 million in 2010 (NSW National Parks and Wildlife Service. State of the Parks survey 2010) (Fig. 1). The Bradleys Head walk (‘BH walk’) leads visitors to iconic and internationally significant Aboriginal and European heritage sites and provides insights into Australian naval history. Visitors have the opportunity to experience unique Australian flora typical of the Hawkesbury sandstone heaths and woodlands with Red Bloodwoods (Corymbia gummifera) and Sydney Peppermint (Eucalyptus piperita). Bradleys Head is also home to a rich avifauna with species such as the Rainbow Lorikeet (Trichoglossus haematodus) and supports an abundant reptile community including the Eastern Water Dragon (Physignathus lesueurii). A tour was developed consisting of eight stops at heritage and natural attractions distributed over 1.8 km of the unpaved BH walk along the coastline providing scenic views across the harbour to the iconic Sydney Opera House. Three (‘detour stops’) of the eight stops were accessible via staircases leading to a beach, a café or a historic cannon battery. To discern the influence of the different media types which attracted visitors to tour stops from the attracting power of other visitor assets such as signage, benches and viewing platforms, tour stops were located at least 20 m away from these already existing facilities. 3.2. Interpretive media The tour script was developed by professional interpretation specialists (Acoustiguide, Australia) in consultation with NSW National Parks and Wildlife Service staff. The draft tour script was tested on-site with staff and visitors and revised accordingly. We used the revised script for the development of six types of interpretive media: namely traditional media, including text-rich vs. image-rich pamphlets and signs, and modern media: a GPS-triggered multi-media tour (‘navi tour’) and an MP3-player audio tour. The pamphlets were printed in colour, on both sides of an A3-paper, which was folded in half across the middle to create a standard brochure format. The first page of the brochure displayed a topographic map of the tour and numbered stops. Text-rich pamphlets

62

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

Fig. 1. Location of the study area, Bradleys Head, within the Sydney Harbour National Park in New South Wales, Australia.

consisted of the 2100-word tour script, without any imagery aside from the tour map. The text on the image-rich pamphlets was reduced to approximately 50% of the original script and presented in the form of bullet points. In spite of the reduction in text, all key information was retained as best as possible by eliminating contextual information and redundant text. Twenty images added visual support to the text. The same text and imagery were used in the development of text-rich and image-rich information signs. The signs were in A3-format boards and positioned in clearly visible locations along the trailside. The sign boards were removed from the posts on days when the other interpretive media were assessed. At the staircases leading to the detour stops, signage was placed to encourage visitors to take the detour on days when signage was tested. Whilst such signage was not set up when pamphlets were tested, the same persuasive messages were used directly on the pamphlets. At each tour stop, clearly visible numbered tags of 20 × 20 cm provided easy reference to the stops described in the interpretive media. The audio tracks for the audio tour were produced by itours (Australia) in consultation with NSW National Parks and Wildlife Service staff using the Acoustiguide tour script. In a similar fashion to the printed media tours, the audio tour consisted of eight stops, which were referenced by numbered tags and three detour stop announcements. At each tour stop, audio tracks were played at an average length of 1.9 min. Visitors triggered the tracks manually on an MP3

player and listened to them with headphones. Although the MP3 player had no display to indicate the track number, each track commenced with an announcement of the stop number. To develop the navi tour, we used an online program by Geovative Solutions (USA). Tours were created by uploading the geographic coordinates of the individual stops, which were then linked with MP3s, imagery and text. We used the MP3s from the audio tour, some of the imagery provided in the image-rich print media and a summary of the tour stops' content in text format. The tour was loaded onto a Garmin (USA) nüvi 765 vehicle navigator, which offers a pedestrian mode for explorations on foot. Audio tracks commenced automatically as visitors approached the tour stops. Touch-screen buttons allowed visitors to access imagery and text. Users of both the audio tour and the navi tour received pamphlets providing instructions on the use of the media, including the tour map. 3.3. Visitor sampling The visitor sampling took place on weekends from August to November and during the school holidays in 2011. Randomly selected visitors were intercepted at the beginning of the BH walk, approximately 150 m before the first tour stop. Those willing to participate (78% of those approached) were asked to embark on the walk either without a tour medium (=control) or whilst using one of the text-rich or image-rich pamphlets, an audio or a navi tour. When

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

testing audio tours, we offered MP3 players to the other members of the travel party to create a realistic scenario. In contrast, the navi tour contained built-in speakers which broadcast the audio track to the entire travel party. On days when signage was tested, no other media were handed out. We aimed to sample at least 40 participants in each media type with one randomly selected person per travel party as the focus of the research. We restricted our sample to those visitors who intended to complete the BH walk and excluded people who planned to spend the day at the café or in one of the beaches. We also excluded visitors whose primary motivation was to exercise. In addition to the instructions on the use of the technical devices provided in the supporting pamphlets, visitors equipped with MP3 players or navigators were given aural instructions. A paired pre- and post-experience survey (in Appendix) was conducted by interviews. The pre-experience survey captured data on socio-demographics, frequency of previous visitation to the BH Walk, travel motivations, and general preferences for tour media. The post-experience survey focussed on word-of-mouth recommendation (WOM), factual learning and visitor satisfaction with various aspects (tour medium itself, having fun, relaxing, socialising, experiencing nature, discovery of sights, learning) of the walk which accrue from using a specific tour medium. Questionnaires administered to control visitors were the same as those provided to media users with the exception that control users were asked for their satisfaction with aspects of the walk irrespective of a tour medium. To investigate learning, six multiple-choice questions were posed about information covered at different tour stops. Four response choices were provided per question along with an ‘Unsure/Don't know’ category, which was scored as incorrect. In addition, visitors were asked to indicate prior knowledge of the correct response to any of the six questions. Visitors with pamphlets were instructed not to refer to the text for responses. To determine the spatio-temporal behaviour of visitors, participants were equipped with a GPS receiver (Garmin, USA, Forerunner 305 wristband GPS), which has been used successfully in other visitor tracking studies (Edwards, Griffin, Hayllar, Dickson, & Schweinsberg, 2009). The GPS units were programmed to record time-stamped geographical position data in one-second intervals of the route (track points) that each participant had taken and their velocity. Visitors were instructed not to tamper with the devices and to return the units to the researcher who administered the second part of the survey at the end of the walk. 3.4. Analysis 3.4.1. Survey data The responses from the visitor survey were analysed with IBM SPSS Statistics for Windows 19.0 (IBM SPSS, 2012). Point-scale data were treated as ordinal (Rhodeghier, 1996; Scheff, Saucier, & Cain, 2002) so that the mean WOM and satisfaction ratings could be compared with a two-way factor ANOVA including media type and frequency of visitation, followed by a Tukey post hoc test. A two-way factor ANOVA was also conducted to analyse the influence of media type and frequency of visitation on factual learning (number of responses learnt) and to identify differences in knowledge prior to embarking on the tour (number of responses known prior). The interaction between the two factors was excluded from all final ANOVA models due to non-significance. In the models on WOM and the satisfaction ratings, the factor ‘frequency of visitation’ was excluded for the same reason. We had initially included a range of explanatory factors (age, group type, group size, gender) in all models however these factors were excluded as they were found to be non-significant. The lack of influence of these visitor characteristics may be partially explained by the fact that the content of the interpretive tour had

63

been designed (see Section 4.2) with a broad park visitor audience in mind and hence was equally appealing to different sub-groups of visitors. We observed no significant bias in the sub-groups of the six types of interpretive media with respect to the explanatory factors included in our analysis.

3.4.2. Spatio-temporal visitor data The visitor tracking data were spatially analysed with ArcMap 9.3 (ESRI, 2008) to yield the attracting, distracting and holding powers of the different media types. These variables are the result of the collective movement of visitors (Orellana et al., 2012) whose properties can be aggregated over a defined space (tour stop zones, see next section) and time (duration of stay within these zones). A concentric buffer zone of 10 m around each tour stop was defined alongside other visitor asset categories as the tour stop zone. We selected a tight buffer of 10 m to effectively discriminate between users and non-users of the interpretive media and assets. The lower limits of this threshold were determined by the accuracy of our GPS which was estimated at less than ±10 m. By visual inspection of the tracks prior to analysis, we discarded waypoints that appeared to be obvious outliers. Visitors spending at least 1 min in these zones were considered to use the respective facilities. This parameterisation of the spatial and temporal dimensions of movement suspension/stopping (Orellana et al., 2012) and consequent aggregation of visitor movement data in summary statistics (Shoval, 2010), was based on the following criteria: (1) direct observations of visitors around tour stops and visitor assets, (2) average walking time to pass from one end of the buffer zone to the other (approx. 30 s), (3) an estimate of the additional minimum time needed to engage with at least a part of the tour stop information (another approx. 30 s) based on field testing and literature (see for instance, Cole et al., 1997). For reference, listening to the full length of an average audio track required approx. 1.9 min. The spatio-temporal thresholds were validated through direct observations of visitors (n = 218), which confirmed that at least 93% and 89% of people who spent 1 min or longer within a 10 m zone around tour stops and specific visitor assets (such as benches and picnic tables) respectively were in fact using these. We then performed a spatial join between the buffered tour stops (and other assets) with the visitor tracks to calculate the number of track points of each visitor within a buffer. Given the one-second recording interval, the number of track points equalled the duration of time spent in each buffer zone (holding power). The attracting power was defined as the percentage of visitors who spent at least 1 min within a buffer zone. The absence or presence of track points determined whether people had taken a detour to access a detour stop. We captured this information in the number of detours taken (distracting power). For the statistical analysis of these data, we used a univariate two-way factor PERMANOVA+ (Anderson, Gorley, & Clarke, 2008), with 999 permutations and type III sum of squares based on Euclidean distance in PRIMER v6 (Clarke & Gorley, 2006). We opted for a PERMANOVA+ rather than an ANOVA as it releases the data from the assumptions inherent to parametric tests, which were not fulfilled. The main and interaction effect of media type and frequency of visitation as fixed factors were included in the PERMANOVA+ model. The interaction was excluded in the final model due to non-significance. To account for repeated measurements of the same participants at the different tour stops ‘participant’ was included as a random factor nested into the fixed factors. Subsequent pairwise comparisons were employed to identify differences between the levels within each factor such as ‘media type’ and ‘frequency of visitation’. For all analyses of both the survey data and the spatio-temporal visitor movement data, the α-level of the P-value was set to 0.05. Unless otherwise stated, average values are presented ±1 SE. All maps were produced in ArcMap 9.3 (ESRI, 2008).

64

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

4. Results 4.1. Characteristics of the sample We surveyed and GPS-tracked 306 park visitors > 18 years of age along the 1.8 km BH walk; from the beginning of the BH walk near the Sydney ferry terminal to the BH historic site. This included 42 control visitors (without interpretive media), 43 audio tour and 54 GPS navigation tour (‘navi tour’) users, 43 users of text-rich and 41 users of image-rich information pamphlets, 42 users of text-rich and 41 users of image-rich signage. Participation rate was at 76%. Twelve of the recorded GPS tracks, equally spread across treatments, were excluded from the GIS analysis as they were incomplete. Participants were sampled throughout the day with 56.2% having commenced their walk before 12 pm. The majority (63.4%) of participants were tracked during cloudy weather conditions, with temperatures averaging between 15 °C and 30 °C (vs. b 15 °C) in 63.1% of cases. 4.2. Visitor profile The survey revealed that the majority of participants were from Australia, particularly from within the Sydney metropolitan area (88.9%). Most of them travelled as adult couples (42.8%), friends/ relatives (27.8%) or single travellers (27.8%) (vs. parents with 1–2 children: 1.3%; organised groups: 0.3%). The majority of people (62.8%) were older than 44 years (≤ 24: 3.3%, ≤ 34: 14.1%, ≤ 44: 19.9%, ≤54: 27.8%, ≤64: 26.8% and 64 and plus: 8.2%). Many people had undertaken the BH walk before, either frequently (=at least once per month: 29.1%) or occasionally (40.5%). Despite differences in the frequency of visitation, visitors shared most of their motivations (multiple responses) for undertaking the BH walk: Particularly important were the enjoyment of nature/peace/quiet (58.5%), landscape/views/natural landmarks (54.9%) and viewing flora/fauna (31%). Pleasure/leisure (61.4%) and rest/relaxation (40.8%) were also among the primary motivations. Touring and sightseeing (28.6%) along with education/learning (18.6%) were mainly important for first-time and some occasional visitors. Whilst our study excluded the many frequent visitors who use the BH walk for exercising, the physical aspect of walking/hiking (79.7%) was still a very important incentive to visit BH. Further motivations included socialising/family experience (24.5%), photography/art (17.3%) and heritage (5.9%). 4.3. General preferences for interpretive media Prior to embarking on the BH walk, visitors were asked to rate their preference for a variety of interpretive media. People showed the greatest preference for signage (average 5-point Likert scale rating: 4.3 ± 0.06) and a moderate preference for pamphlets (3.1 ± 0.09). Modern media (multi-media) that announced stops automatically allowed users to decide whether or not to trigger an audio tour component and access additional text or imagery via a touch-screen, scored highly at 3.7 ± 0.08. Modern media (audio tour) without stop announcements and access to further information were far less popular, independent of whether the audio component needed to be loaded manually (1.8 ± 0.07) or automatically (1.8 ± 0.06). People commented that they found it convenient to be alerted to the proximity of a sight but still wished to decide for themselves whether or not to access audio or other interpretive tour content. Additional tour content was viewed as a ‘nice bonus’, and people suggested that it could encompass information not delivered by the audio tour, along with a synopsis of the most important facts presented at each stop or even a complete transcript. Nearly a third (33.7%) of the visitors found it important to very important that a tour could be downloaded from the internet to their own devices. These visitors clearly preferred smartphones (89%) for delivering downloaded tours. A small number favoured MP3 Players

(8.3%), GPS navigation systems with the capability to present tours as points of interest (2.6%) and tablet PCs (0.5%). People commented however that loading the tour onto a GPS navi with a pedestrian mode could be more practical than a smartphone if the tour was for a scenic drive or a combined drive and hike. Whilst visitors favoured bullet points (84.1%) over continuous blocks of text on signage and pamphlets, 68.2% of visitors still preferred complete sentences over a list of catchwords/key facts that summarised the tour information. When asked for the optimum ratio of various tour components, people opted for media that consisted of 70% of imagery with text, 20% stand-alone text and 10% stand-alone imagery. Control and media users showed similar preferences for the number of tour stops of the BH walk and their average duration: Control visitors, who were asked to indicate their general preferences for the BH walk, found 4.1 (±0.5) stops ideal with up to 2 detours leading to optional tour stops. The preferred maximum time needed to engage with the tour medium (independent of the media type) at each stop was 2.0 (± 0.6) min. The majority of media users who were asked to indicate to which extent the BH tour should be modified, opted to maintain a number of 8 stops (76% vs. 9.8%/14.2% voting for an increase/decrease) and their average duration (80.4% vs. 9.6%/ 10% voting for increase/decrease) of 1.9 min. Users of text-rich media along with occasional and frequent visitors were somewhat overrepresented in the groups voting for fewer and shorter stops. Visitors commented that stops should be longest wherever sights were stationary and hence viewing success guaranteed, compared to stops where sights such as fauna may or not be present at the time of people's visit. 4.4. Visitor satisfaction and word-of-mouth recommendation Visitors were most satisfied using signage along the BH Walk, together with navi tours and followed by image-rich pamphlets (Table 1). Text-rich pamphlets and audio tours were considered the least satisfying of the six media types. Word-of-mouth-recommendation showed a similar pattern (Table 1). Controls experienced lower levels of fun along the BH walk than users of image-rich print media and navi tours (trend), similar levels as users of text-rich print media however higher levels than audio tour users (trend). Traditional media users rated their experience as similarly relaxing, social and close to nature as that of control users whereas levels of relaxation and socialising were much lower with modern media (Table 1). In contrast, users of modern media were significantly more satisfied with their learning experience and the discovery of new sights (Table 1) compared to controls or traditional media users with the exception of text-rich signage, which achieved a similar level of satisfaction with respect to the learning experience. The frequency of visitation had no significant effect on any of these performance measures. Navi users considered the automatic onset of the audio recordings upon approach to the tour stops as a convenient feature. Nonetheless, 17% of them commented that they would have preferred an automatic announcement of stop names and the option to trigger the audio recordings manually. This is consistent with visitors' preference for different media types as stated prior to embarking on the BH walk. Opinions about using a technical device such as the GPS navi or MP3 player in a national park setting were divided. Fourteen percent of visitors found the actual handling of the technical device a nuisance, whilst 11% were reluctant to use such a device on a park walk as, in visiting nature, they sought to escape from their technologically dominated lives. Sixteen percent of people found listening to an audio recording, whether delivered by MP3 player or GPS navi, a distraction from enjoying the natural soundscape and socialising with members of their travel party. This issue became particularly prominent for audio tours that required headphones, which

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

65

Table 1 Mean word-of-mouth (WOM) recommendation of the tour to other visitors and mean satisfaction (±1 SE) ratings on a 5-point Likert scale (with 1 = very unlikely/low; 5 = very likely/high) of the overall satisfaction with the medium and with six different components of the tour experience depending on the media type.

Control PamphImage PamphText SignImage SignText Audio Navi F(df) P-value

WOM

Medium

Fun

Relax

Social

Nature

Discovery

Learn

N/A 3.64 (±0.2)b 3.36 (±0.2)b 4.34 (±0.2)a 4.23 (±0.2)a 3.22 (±0.2)b 4.21 (±0.2)a F(5,129) = 4.8 b0.001

N/A 3.77 (±0.2)b 3.43 (±0.2)b,c 4.17 (±0.2)a 4.1 (±0.2)a 2.98 (±0.2)c 3.94 (±0.2)a,b F(5,252) = 7.5 b0.001

3.84 (±0.1)b,c 4.39 (±0.1)a 3.74 (±0.1)b,c 4.2 (±0.1)a 3.83 (±0.1)b,c 3.4 (±0.1)c 4.02 (±0.1)a,b F(6,289) = 5.2 b0.001

4.62 (±0.2)a 4.31 (±0.1)a,b 4.22 (±0.2)b 4.23 (±0.2)b 4.03 (±0.2)b 3.55 (±0.2)b 3.5 (±0.2)b F(6,274) = 7.5 b0.001

4.34 (±0.2)a 4.3 (±0.2)a 4.08 (±0.2)a 4.13 (±0.2)a 3.96 (±0.2)a 3.3 (±0.2)b 3.47 (±0.2)b F(6,203) = 4.5 b0.001

4.41 (±0.2)a 4.47 (±0.1)a 4.28 (±0.1)a 4.49 (±0.1)a 4.36 (±0.1)a 3.26 (±0.1)c 3.66 (±0.1)b F(6,265) = 8.2 b0.001

3.41 (±0.2)b 3.74 (±0.2)b 3.66 (±0.2)b 3.64 (±0.2)b 3.68 (±0.2)b 4.11 (±0.2)a 4.1 (±0.2)a F(6,266) = 2.5 0.026

3.08 (±0.2)c 3.54 (±0.2)b 3.72 (±0.2)b 3.55 (±0.2)b 4.03 (±0.2)a,b 4.19 (±0.2)a 4.07 (±0.1)a F(6,277) = 5.7 b0.001

The same (superscripted) letters indicate similar grouping by post hoc tests following significant differences in the ANOVAS. Bold letters indicate top ratings for each perfomance measure.

was the major complaint (78%) among users that gave the lowest satisfaction ratings for these tours. On windy days, however, navi tour users commented that headphones would be essential to listen to the recordings clearly. In contrast, other visitors (21%) commented favourably on using modern media for park interpretation and considered these tours the only sensible solution for future park interpretation, particularly if tours could be loaded onto personal devices (such as smartphones). These tours were thought to offer the greatest freedom and most relevant information to visitors, particularly if they could be customised (e.g., choice of different lengths of audio recordings, tour content, imagery) and content was updated regularly. Twelve percent of people also suggested the provision of several shorter recordings about a variety of topics at each tour stop rather than one long recording. Other suggestions included a theme-based form of interpretation where people could choose recordings/text/imagery that would suit their particular interests (e.g., educational vs. entertaining tours; tours focussing on flora/fauna vs. anthropogenic heritage; adult vs. children tours). 4.5. Short-term factual learning There were no significant differences in prior knowledge on the questions asked in relation to tour content between users of different media types (mean number of questions known prior to going on the tour: 1.1). Frequent and occasional visitors however had a somewhat greater prior knowledge than first-time visitors (Table 2). The media type (F(6,297) = 8.2, P b 0.001) and frequency of visitation (Table 2) had a significant effect on the number of responses learnt. Media users learned on average more responses (= 2.0) than control visitors (=0.6). This gain was greatest for users of media with audio components (audio = 2.6; navi = 2.3), followed by users of text-rich media (pamphlets = 2.2; signage = 1.8). Visitors using image-rich media attained the lowest number of learnt responses (signage = 1.5; pamphlets = 1.3) in spite of containing the same factual content as the other media. First-time visitors learnt significantly more factual material than occasional and frequent

visitors, which was slightly greater than the difference (0.45) in knowledge prior (see above in Section 4.5) to taking the tour (Table 2). Hence, first-time visitors learnt more than occasional and frequent visitors however they were starting from a lower baseline of knowledge. 4.6. Spatio-temporal behaviour Visitors spent an average of 30 min on the BH walk (controls: 25.1 ± 0.9; media users: 35 ± 1.4; see Table 2 for differences between first-time, occasional and frequent visitors). At tour stops media users stopped on average 3.2 times for 0.5 min or longer and 2.2 times for 1 min or longer. Visitors who stopped for 1 min or longer spent, on average, 2.2 min within the 10 m radius of the tour stop zones. Further, media users took on average 0.4 (±0.02) detours leading to tour stops. Tour stops located directly along the BH walk strongly attracted media users. An average of 27% (±3.1) of people stopped there for 1 min or longer (Fig. 2.1). In contrast, this percentage was much lower (4.7% ± 0.8) at detour stops and various other visitor assets, including benches, and viewing platforms (Fig. 2.1; pseudo-F(2,136) = 35.1, P(perm) b 0.001). The average time spent at either tour stops (2.3 ± 0.2 min) or detour stops (2.1 ± 0.2 min) was similar and significantly higher compared to other visitor assets (1.5 ± 0.2 min) (Fig. 2.2; pseudo-F(2,136) = 4.5, P(perm) = 0.001). Media exerted a significant attracting power to tour stops with virtually no control visitors stopping there for 1 min or longer. This is most likely because people stopped to engage with the media at the tour stops, which is underpinned by our field observations (see Section 3.4.2). Significant differences were also found between different media types (Fig. 3.1; pseudo-F(6,159) = 16.0, P(perm) = 0.001). Signage (Fig. 3.1 and 3.4) attracted the highest percentage of visitors to stop, followed by audio (Fig. 3.1) and navi tours (Fig. 3.1). In contrast, pamphlets (Fig. 3.1 and 3.3) demonstrated the weakest attracting power. The provision of either text- or image rich media had no significant impact on the attracting power of tour stops (Fig. 3.1).

Table 2 The mean (±1 SE) spatio-temporal behaviour and short-term factual learning of (1) first-time, (2) occasional (a few times/year, rarely), or (3) frequent (at least once per month) visitors who used one of six types of interpretive media along the Bradleys Head Walk.

Time (min) spent on walk No. of stops (≥0.5 min) No. of stops (≥1 min) % of stoppers (attracting power) Time (min) spent at stops (≥1 min) (holding power) No. of detours (distracting power) No. of questions known prior to walk No. of correct responses to questions unknown prior to walk

First-time visitors

Occasional visitors

Regular visitors

31.6 ± 1.2 4.6 ± 0.6 3.2 ± 0.4 31.9 ± 3.1 2.6 ± 0.13 0.5 ± 0.03 0.8 2.1

28.0 ± 1.4 2.8 ± 0.4 2.1 ± 0.2 11.3 ± 3.1 2.3 ± 0.17 0.4 ± 0.03 1.2 1.5

28.9 ± 0.9 2.2 ± 0.4 1.8 ± 0.2 12.1 ± 1.9 1.9 ± 0.07 0.3 ± 0.03 1.3 1.5

Statistical results

Pseudo-F(2,159) = 39.0, P(perm) b 0.001 Pseudo-F(2,136) = 10.5, P(perm) = 0.001 Pseudo-F(2,873) = 7.5, P(perm) = 0.01 F(2,297) = 6.1, P = 0.03

66

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

(1) Percentage of visitors

(2) Their average stop time (min)

Fig. 2. The (1) percentage of visitors (‘attracting power’) and (2) their average stop time (min; ‘holding power’) who stopped for 1 min or longer at one of eight tour stops or various other types of visitor assets along the Bradleys Head Walk. Visitors used one of six types of interpretive media; audio tours, GPS-triggered multimedia tour (Navi), text-rich pamphlets or signage, image-rich pamphlets or signage.

The holding power of tour stops also depended on the media type (Fig. 4.1; pseudo-F(5,136) = 5.9, P(perm) = 0.001) and was greatest for modern media (audio: Fig. 4.1; navi tours: Fig. 4.1 and 4.3). In contrast, the holding power of traditional media (Fig. 4.1 and 4.2) was similar and on average 32% lower than that of modern media. Finally, the distracting power of tour stops was dependent upon the media type (pseudo-F(6,873) = 5.1, P(perm) = 0.01): The number of detours leading to tour stops that navi tour users (0.6 ± 0.04) took was between 33% and 50% higher compared to the rest of the media users (audio, text-rich pamphlets, text-rich and image-rich signage: 0.4 ± 0.04; image-rich pamphlets: 0.3 ± 0.04) and the control (0.3 ± 0.04). The frequency of prior visitation of visitors significantly influenced the attracting, holding and distracting powers of tour stops (Table 2): The percentage of first-time visitors stopping at tour stops for 1 min or longer was approximately three times higher than that of occasional or frequent visitors. First-time visitors who used interpretive media and spent 1 min or longer at tour stops spent on average 13% more time than occasional visitors and 36% more time than frequent visitors. First-time and occasional visitors took, on average, significantly more detours than frequent visitors. 5. Discussion Our findings demonstrate the complexity of choices that confront park managers when they need to evaluate the performance of interpretive media in accordance with the objectives of their interpretive program. These may be to maximise learning, promote longer stays at tour stops or a high overall visitor satisfaction. As a result, performance

needs to be judged using a range of criteria. In our study, these encompassed the effect of interpretive media on the spatio-temporal behaviour of visitors, their satisfaction, word-of-mouth recommendation and short term factual learning. 5.1. Influence of visitor interpretation on the spatio-temporal behaviour of visitors Our research revealed that the media type significantly influenced the spatio-temporal behaviour of visitors along a walking track in Sydney Harbour National Park and hence their attracting, holding and distracting powers. We successfully captured these effects through GPS tracking of visitors and subsequent GIS analysis. The ability of the medium to guide people in space is an important aspect for park managers who seek to encourage people to stop at the more interesting and informative sites (Light, 1995). This was certainly achieved by our media. Visitors who used one of six interpretive media stopped more frequently along the trail, and for a longer period, including at sites without specific landmarks. They also took more detours than visitors who travelled without interpretation apart from what was already in place. The use of interpretation creates a more meaningful experience than if no media are used and is effective at guiding visitors to attractions (Jacobson, 1988; Rademaker, 2008). The attracting and holding powers of interpretive media and their tour stops were also greater than for facilities such as benches, platforms and toilets. Generally, a varying degree of usage of facilities and attractions is typical for parks (e.g., Orellana et al., 2012) and reflects their individual popularity. Interestingly in our study, controls

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

67

(1)

(2) Without interpretive media (=control)

(3) Image-rich pamphlets

(4) Image-rich signage

Fig. 3. The percentage of (1) visitors who stopped (‘attracting power’) for 1 min or longer at one of eight tour stops along the Bradleys Head Walk. Visitors travelled without interpretive material (=control group), or used an audio tour, GPS-triggered multimedia tour (Navi), text-rich pamphlet (Pamph-TR), image-rich pamphlet (Pamph-IR), text-rich signage (Sign-TR) or image-rich signage (Sign-TR). The (2) control and media with the (3) weakest and (4) strongest attracting power are displayed separately. Signage was not set up on days where control visitors or other media were sampled.

68

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

(1)

(2) Text-rich Pamphlets

(3) Navi Tours

Fig. 4. The average stop time (min; ‘holding power’) of (1) visitors who stopped for 1 min or longer at one of eight tour stops along the Bradleys Head Walk. Visitors used an audio tour, GPS-triggered multimedia tour (Navi), text-rich pamphlet (Pamph-TR), image-rich pamphlet (Pamph-IR), text-rich signage (Sign-TR) or image-rich signage (Sign-TR). The media with the (2) weakest and (3) strongest holding power are displayed separately. The control group (without interpretive media) is not shown because very few control visitors stopped at the tour stops (Fig. 3). Signage was not set up on days where control visitors or other media were sampled.

did not compensate for bypassing tour stops by stopping more at other facilities. Hence visitors without a tour medium spent less time along the trail, thereby missing out on a more extensive engagement with the site. Similarly, in a national park without an interpretive program, Orellana and Wachowicz (2011) found that visitors missed out on spending time at attraction points. In their case, however, visitors may have compensated as they used almost double as many facilities as attractions despite the latter accounting for more than twice the number. If park managers want to encourage people to stop at specific sites, they need to use interpretive media with a strong attracting power to prevent visitors from flowing through the park or missing out on natural and heritage attractions in favour of facilities (Orellana & Wachowicz, 2011; Orellana et al., 2012). In our study, signage held a greater attracting power than both modern media, all of which clearly outperformed pamphlets. Approximately a third of visitors stopped at tour stops with signage which is consistent with other studies (e.g., Light, 1995; Peart, 1984). Similarly, Light (1995) found that signage was more popular and received greater attention than for instance audio tours and exhibitions. Media such as signage which pose minimal difficulty in the access, uptake and absorption of information, satisfy peoples' desire for easy learning and having new experiences. Thus they encourage people to visit and stop at attractions,

even at those without specific landmarks, which corresponds with other studies (Moscardo, 1998; Moscardo et al., 2004; Sandifer, 2003). Using the attracting power of media as an indicator for the degree of engagement with the medium is likely most accurate for signage as people need to read it where it is located. This is somewhat more difficult to infer for the modern media as participants may have continued walking whilst for instance listening to the audio component. Similarly, albeit people tend to read pamphlets when stationary, they may do so away from the corresponding tour stop. Nonetheless, in all cases a greater attracting power means that more people stop and potentially engage with a specific site. Hence, the interpretation is utilised where it is most relevant and may consequently facilitate the engagement with that site. Navi tours were the most powerful at attracting visitors to take detours to tour stops off the main track. The greater distracting power of navi tours may be attributed to the automatic trigger of the detour announcements along with the voice-over, which was audible to all members of the travel party. According to Bitgood (2000), visitors readily respond to media that require minimal effort. In our case, although text-rich and image-rich pamphlets provided sufficient information to enable visitors to decide to embark on a detour, the passive form of communication had probably less persuasive power (Cole et al., 1997; Hughes & Morrison-Saunders, 2005).

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

Both audio and navi tours achieved the greatest holding power likely because they were most effective at maintaining people's attention. Similarly to Novey and Hall (2007), the audio component of both tours was prone to mitigate attention fatigue and hence performed better than text media (Light, 1995; Peart, 1984). Interactive media incorporating audio components have been linked to very long attention spans of more than 5 min (Tubb, 2003). Various studies have found that visitor attention span plays a significant role in the effectiveness of interpretive media (Bitgood, 2000; Ham, 1992; Novey & Hall, 2007). The ecological and historical sounds and recordings used in our modern media enhanced the audio experience and likely provoked an emotional response that achieved prolonged attention times (Ogden, Lundburg, & Maple, 1993). In addition, we found that people are more mindful and attentive when they have control over an interpretive experience as evidenced by our results from our interactive navi tour. This finding is similar to that of Moscardo's (1996). The multi-sensory nature of this tour, with extra information accessible through the touch screen, likely also increased attention span particularly compared to pamphlets (Sandifer, 2003). To a lesser extent the increased holding power of modern media may have also been attributed to time spent handling the technical devices (Ruchter et al., 2005) albeit pamphlet users may also spend some time when they extract pamphlets from their pockets or bags. There are few studies that have investigated holding time in a nature-based setting. However one such study by Cole et al. (1997) found that holding time based on reading signs was 0.25 min. Our average holding time of 2–2.5 min is significantly higher than that and more comparable with holding times reported from indoor settings such as museums or caves (Bitgood, Pierce, Nichols, & Donald, 1987). Although not all of the holding time may be attributed to attention generated by the medium, we expect that a considerable part of it was. Adding for instance an audio component to tours is known to increase their holding power (Bitgood et al., 1987; Davidson et al., 1991). Moreover, we defined tour stops as a dwell time of at least 1 min. Hence we excluded shorter stops, which may explain our reasonably long average holding times. Finally, the holding time encompassed all stages of the engagement with the tour stops. This includes walking through the zone, retrieval of tour stop information and post-media engagement with the site. However, even the latest stage can indicate the performance of a medium. As expected, the influence of interpretive media on the spatiotemporal behaviour of first time visitors was greater than for repeat visitors, with the former more easily attracted, distracted and held than the latter. They were also found to stop significantly longer at tour stops than occasional and frequent visitors, both of which may reflect their exploratory nature in a new environment and their higher capacity to absorb information. Our findings are supported by other studies (Ballantyne et al., 1998; Hughes & Morrison-Saunders, 2002b). There was, however, no interactive effect between frequency of visitation and media type which may suggest that the interpretive experience itself, is more important than the delivery of it. For example, repeat visitors may use all types of media to the same degree due to the fact that they may have already acquired knowledge about the area and may not perceive even novel technical media to be more appealing. 5.2. Influence of visitor interpretation on visitor satisfaction and word-ofmouth recommendation Visitor satisfaction with various aspects of the tour was considerably influenced by the type of medium provided. Visitors were most satisfied using signage along the BH walk, together with navi tours and followed by image-rich pamphlets. Text-rich pamphlets and audio tours were considered the least satisfying of the six media types. Word-of-mouth-recommendation showed a similar pattern as the overall satisfaction with the media. When visitors are satisfied with their tour, they are more likely to share this positive experience

69

with others, resulting in word-of mouth-recommendations (Baker & Crompton, 2000). This inexpensive form of promotion also has a high credibility (Andereck, 2005) and is therefore very potent. The high levels of satisfaction with signage may partly be attributed to their number and spacing on the walk, which were considered to be ideal by most visitors in our study. Finding the right balance between providing information on signage and information overload and visual pollution is critical (Bramwell & Lane, 1993; Hughes & MorrisonSaunders, 2005). As the number of signs increases, visitors typically pay less attention to individual signs (Bitgood, 2000). Conversely, Hughes and Morrison-Saunders (2002b), found that visitors on a walk in Western Australia, were dissatisfied with the sparse signage that was designed for minimal visual impact. In our study we appeared to have found the proper balance. In addition, signage is likely attractive to visitors as accessing information requires little effort and visitors have complete control to decide whether or not to read the information. Satisfaction levels with print media may also be influenced by their design such as the relative proportions of text and imagery (Cole et al., 1997; Moscardo et al., 2004). In our study image-rich and text-rich media were similarly satisfying to visitors except that the former was considered to be more fun. Our survey found that visitors preferred a ratio of 70% of imagery with text, 20% stand-alone text and 10% stand-alone imagery. Such a combination of visually appealing elements with factual information appears very attractive to visitors and was better realised in our image-rich media. Although bullet points were preferred over blocks of text, complete sentences were selected over catchwords which confirm the design principles proposed by Moscardo, Ballantyne, and Hughes (2003). Their recommendations foresee short and easy-to-process sentences with pertinent information that attracts visitor attention. Traditional media achieved higher satisfaction ratings for visitor relaxation, socialising and experiencing nature compared to modern media. This is consistent with Ruchter et al. (2005) who reported that people found using paper guides more relaxing than mobile nature guides. This resulted from people's frustration with an unfamiliar technical device. In our study additionally, some visitors considered the audio component of the tour disruptive to interactions with their travel party or experiencing nature. This issue was particularly prominent with the audio tour as the MP3 players required people to use headphones therefore isolating the individual from their group and their surroundings (McManus, 1987). The comparatively low satisfaction ratings achieved by audio tours for socialising is contrary to Novey and Hall (2007) who found that audio users were able to maintain social interaction. However in their case a handheld wand was used, which does not restrict visitors in their hearing of other group members' conversation as headphones may do, and which also has a novelty factor. This may also explain why our navi tours achieved a higher overall satisfaction compared to audio tours, and ultimately in the much higher inclination of people to recommend the tour to others. Whilst the satisfaction ratings of the modern media were similar across most components, navi tours were considered to be more fun than audio tours. Modern media performed better in relation to satisfaction with discovery and learning than traditional media. Whilst Rademaker (2008) noted that modern media may dampen people's sense of discovery, this was not the case in our study. In fact modern media animated people to take more detours than the other media. These findings should encourage park managers to consider the use of modern media as a means to disperse visitation to the more covert attractions. 5.3. Influence of visitor interpretation on short-term factual learning Short-term factual learning was most effective with modern media, likely due to their greater holding power. Visitors who spend more time engaging with a medium and the surroundings, with which the medium interprets, should absorb more information.

70

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

Similarly, Novey and Hall's (2007) study demonstrated that audio tours enhanced the short-term retention of facts. The ability of modern media to impart knowledge was reflected in the higher satisfaction ratings for learning. The amount of text in media was also found to influence the degree to which visitors retained factual information. Although text-rich media contained the same factual content as image-rich media, short-term factual learning was increased. Therefore, the reduction of contextual information and seemingly redundant text in image-rich media was counter-productive to learning. Visitors may have falsely perceived that image-rich media contained less factual information, which could have decreased their capacity or motivation to retain the existing information. As a result, where learning is a high priority, interpretation should include an adequate amount of text on media to be able to achieve this objective. As expected, first time visitors learnt significantly more than occasional and frequent visitors, which may be partly a result of their spatio-temporal behaviour; namely their increased attraction to tour stops and their higher average duration of stay along the walk. This was consistent with the findings in van Marwijk's (2009) study where first time visitors also demonstrated a significantly greater knowledge gain than occasional and frequent visitors. These findings likely reflect the emphasis that first time visitors place on learning, information gathering and becoming familiar with their surroundings (Hughes & Morrison-Saunders, 2005; Lau & McKercher, 2006). 5.4. Methodological implications We employed GPS tracking and GIS technology to calculate three performance measures for our interpretive media that summarise complex geospatial data in a way that is easily comprehensible. This approach allows us to ascertain and map differences not only in the performance of media (Figs. 3–4) but also between individual tour stops and other facilities (Fig. 2). It can also accommodate a range of external factors such as visitor characteristics (e.g., frequency of visitation, Table 2). Hence, we consider it a very effective and versatile tool for communication with park management. Whilst the attracting and holding powers have been used in previous studies, these are typically determined by direct observations (Hartmann, 1988; Keul & Kühberger, 1997) or a survey approach (Cole & Daniel, 2003; Young, 1999). Using visitor movement data accrued from GPS tracking to calculate these measures offers several significant advantages: (1) data are not tempered by observer bias (Arnberger, Haider, & Brandenburg, 2005) or recall which can flaw survey results (Cessford & Muhar, 2003; Cole & Daniel, 2003); (2) data are collected simultaneously at many sites by relatively little labour (1 researcher hands out tracking equipment); (3) even small differences in the performance of media can be detected, which are difficult to capture with traditional visitor monitoring techniques. This is especially relevant in our context because, as we have discussed above, the relatively short holding times of interpretive media require a small temporal and site-specific scale for assessment. Conversely, a major drawback currently is the high manual effort required to prepare the data for analysis and the cumbersome analysis itself which has been deplored by others (Orellana et al., 2012; Shoval & Isaacson, 2010). However this issue is likely to be alleviated as the methodology becomes more prevalent in people monitoring and hence the need increases to develop software solutions that automate at least part of the data processing. Apart from that we did not experience other potential disadvantages such as tampering with the GPS units or failure to be returned. 6. Conclusions Our research revealed that the choice of media is fundamental for effective interpretation as it influences visitor satisfaction, word-of-mouth

recommendation, short-term factual learning and all three attracting, distracting and holding powers. This study further demonstrated that modern technical media, specifically GPS-triggered multi-media tours, performed well in most of these measures. The use of rapidly advancing technology such as navigation devices and smartphones is becoming widespread in society. Therefore, modern media may be more readily received by a wider audience than for instance pamphlets. In addition, modern technical media appeal to more senses than traditional media thus holding visitor attention for longer periods. Another advantage of using modern media is that they are not subject to vandalism such as graffiti on signage which cause substantial costs to parks particularly in metropolitan areas. As a result of visitor feedback, we can recommend numerous measures to improve the performance of modern technical media even further: (1) offering tours for download to a personal device with which people are comfortable and which enables them to choose whether to use headphones; (2) automatically triggered tour stops and detour announcements with the option to trigger the audio component manually; (3) the provision of several short recordings over one long recording; (4) multi-media content that combines audio, text and imagery; (5) a thematic form of interpretation where people can choose recordings/text/imagery that would suit their particular interests. The ability to personalise products has become very prevalent in society and visitor interpretation can greatly benefit from giving people choices. Modern media could realise this in the form of multi-lingual audio recordings and thematic information on relevant topics of interest such as flora and fauna, anthropogenic heritage and art, or tours for different age-classes (Ruchter et al., 2005). Information that is relevant to visitors is likely to motivate them to focus on the content and engage with the medium (Moscardo, 1996). For those who prefer traditional media, signage is an appropriate option. Signage has a long history as a traditional form of park interpretive media and our study demonstrated its high performance. Signs should be presented with an appropriate combination of text and imagery which enhances learning and fun. This combination appeals to both those who are receptive of text for learning and those who are more visual in their senses. The range of interpretive media in a park setting needs to accommodate the wide-ranging ages, interests and learning styles of its visitor community (Beer, 1987; Light, 1995). These insights were gained by a mixed-method approach of questionnaire-based surveys and analysis of fine-scale spatial data of visitor movements. This combined approach enabled a comprehensive assessment of traditional and novel performance criteria for interpretive media that account for the range of objectives that interpretive programs pursue. Whilst the latter is particularly effective for detecting differences in the engagement with different interpretive media, the former may ascertain the influence of other factors such as visitor characteristics on the interpretive experience of visitors in parks. Acknowledgements The research was funded by the Tourism and Partnerships Branch of the NSW National Parks and Wildlife Service, Office of Environment and Heritage, Department of Premier and Cabinet. The authors greatly thank two anonymous reviewers for helpful comments on an earlier version of this manuscript and B. Weiler from Southern Cross University for an insightful discussion. J. Briere is to be thanked for his assistance in the design of the interpretive media, data collection and analysis. The authors also thank I. Miranda for her help with media design and all volunteers from Macquarie University, University of Technology Sydney and the University of New South Wales for their help in collecting data. We further gratefully acknowledge C. Togher

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

from the Office of Environment and Heritage for his advice on GIS issues. We thank P. Hay from the Office of Environment and Heritage for granting us permission to conduct our research in Sydney Harbour National Park. We further acknowledge L. Malouf for providing the tour content, some technical equipment and constructive discussion. Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.tmp.2013.04.002. References Alt, M. B., & Shaw, K. M. (1984). Characteristics of ideal museum exhibits. British Journal of Psychology, 75, 25–36. Andereck, K. L. (2005). Evaluation of a tourist brochure. Journal of Travel & Tourism Marketing, 18, 1–13. Anderson, M. J., Gorley, R. N., & Clarke, K. R. (2008). PERMANOVA+ for Primer: Guide to software and statistical methods. Plymouth: Primer-E. Arnberger, A., Haider, W., & Brandenburg, C. (2005). Evaluating visitor-monitoring techniques: A comparison of counting and video observation data. Environmental Management, 36, 317–327. Baker, D. A., & Crompton, J. L. (2000). Quality, satisfaction and behavioral intentions. Annals of Tourism Research, 27, 785–804. Ballantyne, R., Packer, J., & Beckmann, E. (1998). Targeted Interpretation: Exploring relationships among visitors' motivations, activities, attitudes, information needs and preferences. The Journal of Tourism Studies, 9, 14–24. Beer, V. (1987). Great expectations: Do museums know what visitors are doing? Curator, 30, 206–215. Bitgood, S. (2000). The role of attention in designing effective interpretive labels. Journal of Interpretation Research, 5, 31–45. Bitgood, S., Pierce, M., Nichols, G., & Donald, P. (1987). Formative evaluation of a cave exhibit. Curator, 30, 31–39. Bramwell, B., & Lane, B. (1993). Interpretation and sustainable tourism: The potential and the pitfalls. Journal of Sustainable Tourism, 1, 71–80. Brody, M., Tomkiewicz, W., & Graves, J. (2002). Park visitors' understandings, values and beliefs related to their experience at Midway Geyser Basin, Yellowstone National Park, USA. International Journal of Science Education, 24, 1119–1141. Buckley, R. (2000). Neat trends: Current issues in nature, eco- and adventure tourism. International Journal of Tourism Research, 2, 437–444. Cessford, G., & Muhar, A. (2003). Monitoring options for visitor numbers in national parks and natural areas. Journal for Nature Conservation, 11, 240–250. Clarke, K. R., & Gorley, R. N. (2006). Primer v6: User manual/tutorial. Plymouth: Primer-E. Cole, D. N., & Daniel, T. C. (2003). The science of visitor management in parks and protected areas: From verbal reports to simulation models. Journal for Nature Conservation, 11, 269–277. Cole, D. N., Hammond, T. P., & McCool, S. F. (1997). Information quantity and communication effectiveness: Low-impact messages on wilderness trailside bulletin boards. Leisure Sciences, 19, 59–72. Connell, J., & Page, S. J. (2008). Exploring the spatial patterns of car-based tourist travel in Loch Lomond and Trossachs National Park, Scotland. Tourism Management, 29, 561–580. Davidson, B., Lee, C., & George, H. E. (1991). Increased exhibit accessibility through multisensory interaction. Curator, 34, 273–290. Edwards, D., Griffin, T., Hayllar, B., Dickson, T., & Schweinsberg, S. (2009). Understanding tourist ‘experiences’ and ‘behaviour’ in cities. Gold Coast: Sustainable Tourism Cooperative Research Centre. ESRI (2008). ArcMap 9.3. Redlands: Environmental Systems Research Institute. Fennell, D. (1996). A tourist space-time budget in the Shetland Islands. Annals of Tourism Research, 23, 811–829. Ferrer, G., Dew, N., & Apte, U. (2010). When is RFID right for your service? International Journal of Production Economics, 124, 414–425. Gitelson, J. R., & Crompton, L. J. (1984). Insights into the repeat vacation phenomenon. Annals of Tourism Research, 11, 199–217. Ham, S. H. (1992). Environmental interpretation: A practical guide for people with big ideas and small budgets. Toronto: Fulcrum Pub. Hartmann, R. (1988). Combining field methods in tourism research. Annals of Tourism Research, 15, 88–105. Hughes, M., & Morrison-Saunders, A. (2002a). Impact of trail-side interpretive signs on visitor knowledge. Journal of Ecotourism, 1, 122–132. Hughes, M., & Morrison-Saunders, A. (2002b). Repeat and first time visitation in an experience specific context: The Valley of the Giants Tree Top Walk. Journal of Tourism Studies, 13, 20–25. Hughes, M., & Morrison-Saunders, A. (2005). Influence of on-site interpretation intensity on visitors to natural areas. Journal of Ecotourism, 4, 161–177. Hwang, S. N., Lee, C., & Chen, H. J. (2005). The relationship among tourists' involvement, place attachment and interpretation satisfaction in Taiwan's national parks. Tourism Management, 26, 143–156. Jacobson, S. K. (1988). Media effectiveness in a Malaysian park system. The Journal of Environmental Education, 19, 22–27. Keul, A., & Kühberger, A. (1997). Tracking the Salzburg tourist. Annals of Tourism Research, 24, 1008–1012.

71

Lash, S., & Urry, J. (1996). Economies of signs & space. London, UK: SAGE Publications. Lau, G., & McKercher, B. (2006). Understanding tourist movement patterns in a destination: A GIS approach. Tourism and Hospitality Research, 7, 39. Li, X., Cheng, C. -K., Kim, H., & Petrick, F. J. (2008). A systematic comparison of first-time versus repeat visitors via a two phase online survey. Tourism Management, 29, 278–293. Light, D. (1995). Visitors' use of interpretive media at heritage sites. Leisure Studies, 14, 132–149. McKercher, B., & Lau, G. (2008). Movement patterns of tourists within a destination. Tourism Geographies, 10, 355–374. McKercher, B., Shoval, N., Ng, E., & Birenboim, A. (2012). First and repeat visitor behaviour: GPS tracking and GIS analysis in Hong Kong. Tourism Geographies, 14, 147–161. McManus, P. M. (1987). It's the company you keep…: The social determination of learning-related behaviour in a science museum. Museum Management and Curatorship, 6, 263–270. McManus, P. M. (1988). Good companions: More on the social determination of learning-related behaviour in a science museum. The International Journal of Museum Management and Curatorship, 7, 37–44. McManus, P. M. (1989). Oh, yes, they do: How museum visitors read labels and interact with exhibit texts. Curator: The Museum Journal, 32, 174–189. Moscardo, G. (1996). Mindful visitors: Heritage and tourism. Annals of Tourism Research, 23, 376–396. Moscardo, G. (1998). Interpretation and sustainable tourism: Functions, examples and principles. Journal of Tourism Studies, 9, 2–13. Moscardo, G. (1999a). Communicating with two million tourists: A formative evaluation of an interpretive brochure. Journal of Interpretation Research, 4, 21–37. Moscardo, G. (1999b). Making visitors mindful: Principles for creating quality sustainable visitor experiences through effective communication. Champaign: Sagamor Publishing. Moscardo, G., Ballantyne, R., & Hughes, K. (2003). Interpretive signs-talking to visitors through text. In T. Griffin, & R. Harris (Eds.), Current research, future strategies: Bridging uncertainty.. Sydney: University of Technology Sydney. Moscardo, G., & Pearce, P. L. (1986). Visitor centres and environmental interpretation: An exploration of the relationships among visitor enjoyment, understanding and mindfulness. Journal of Environmental Psychology, 6, 89–108. Moscardo, G., Woods, B., & Saltzer, R. (2004). The role of interpretation in wildlife tourism. In K. Higginbottom (Ed.), Wildlife tourism: Impacts, management and planning (pp. 231–251). Altona: Common Ground Publishing. Munro, J. K., Morrison-Saunders, A., & Hughes, M. (2008). Environmental interpretation evaluation in natural areas. Journal of Ecotourism, 7, 1–14. Novey, L. T., & Hall, T. E. (2007). The effect of audio tours on learning and social interaction: An evaluation at Carlsbad Caverns National Park. Science Education, 91, 260–277. Ogden, J. J., Lundburg, D. G., & Maple, T. L. (1993). The effects of ecologically-relevant sounds on zoo visitors. Curator, 36, 147–156. Oppermann, M. (1999). First-time and repeat visitors to New Zealand. Tourism Management, 18, 177–181. Orellana, D., Bregt, A., Ligtenberg, A., & Wachowicz, M. (2012). Exploring visitor movement patterns in natural recreational areas. Tourism Management, 33, 672–682. Orellana, D., & Wachowicz, M. (2011). Exploring patterns of movement suspension in pedestrian mobility. Geographical Analysis, 43, 241–260. Peart, B. (1984). Impact of exhibit type on knowledge gain, attitudes, and behaviour. Curator, 27, 220–237. Prentice, R., Guerin, S., & McGugan, S. (1998). Visitor learning at a heritage attraction: A case study of Discovery as a media product. Tourism Management, 19, 5–23. Rademaker, L. G. (2008). Interpretive technology in parks: A study of visitor experiences with portable multimedia devices. Missoula: University of Montana. Rhodeghier, M. (1996). A practical guide to survey research using SPSS. Surveys with confidence. Chicago: SPSS Inc. Ruchter, M., Real, P., & Düpmeier, C. (2005). Comparing a mobile nature guide and a paper guidebook in the field. 4th workshop “HCI in mobile guides” at mobile HCI 05, September 19–22, 2005, Salzburg, Austria (pp. 5). Sandifer, C. (2003). Technological novelty and open-endedness: Two characteristics of interactive exhibits that contribute to the holding of visitor attention in a science museum. Journal of Research in Science Teaching, 40, 121–137. Scheff, S. W., Saucier, D. A., & Cain, M. E. (2002). A statistical method for analyzing rating scale data: The BBB locomotor score. Journal of Neurotrauma, 19, 1251–1260. Screven, C. G. (1976). Exhibit evaluation: A goal-referenced approach. Curator, 19, 271–290. Shoval, N. (2010). Monitoring and managing visitors flows in destinations using aggregative GPS data. In U. Gretzel, R. Law, & M. Fuchs (Eds.), Information and communication technologies in tourism (pp. 171–183). Vienna: Springer. Shoval, N., & Isaacson, M. (2006). Application of tracking technologies to the study of pedestrian spatial behavior. The Professional Geographer, 58, 172–183. Shoval, N., & Isaacson, M. (2007). Tracking tourists in the digital age. Annals of Tourism Research, 34, 141–159. Shoval, N., & Isaacson, M. (2010). Tourist mobility and advanced tracking technologies. New York: Routledge. Tubb, K. N. (2003). An evaluation of the effectiveness of interpretation within Dartmoor National Park in reaching the goals of sustainable tourism development. Journal of Sustainable Tourism, 11, 476–498. van Marwijk, R. B. M. (2009). These routes are made for walking: Understanding the transactions between nature, recreational behaviour and environmental meanings in Dwingelderveld National Park, the Netherlands. : Wageningen University (PhD Thesis).

72

I.D. Wolf et al. / Tourism Management Perspectives 7 (2013) 59–72

Washburne, R. F., & Wagar, A. J. (1972). Evaluating visitor response to exhibit content. Curator, XV, 248–256. Wearing, S., & Schweinsberg, S. (2008). Best practice interpretation research for sustainable tourism. Australasian parks and leisure, Vol. 11. (pp. 39). Australia: Parks and Leisure. Wolf, I. D., Hagenloh, G., & Croft, D. B. (2012). Visitor monitoring along roads and hiking trails: How to determine usage levels in tourist sites. Tourism Management, 33, 16–28. Xia, J., Arrowsmith, C., Jackson, M., & Cartwright, W. (2008). The wayfinding process relationships between decision-making and landmark utility. Tourism Management, 29, 445–457. Young, M. (1999). Cognitive maps of nature-based tourists. Annals of Tourism Research, 26, 817–839.

Dr. Isabelle Wolf is a Research and Analysis Officer at the Office of Environment and Heritage, Australia. Isabelle is responsible for all aspects of park visitor research and monitoring, including visitor experience development and management. She is leading numerous GIS-related visitor monitoring projects and New South Wales statewide visitor research. In this role she capitalises on her expertise in both the social and ecological sciences. As an ecologist her speciality is the interface between people and their environment with recent work on animal behaviour and flora and fauna communities in fragmented and disturbed habitats. Isabelle has completed a PhD at the University of New South Wales and published in both social and environmental science journals.

Heidi Stricker is an Ecologist (University of New South Wales) and environmental consultant currently pursuing research on the ecological role of turtles in the Murray River at the University of Western Sydney. She also has more than 8 years of experience working in the hospitality and tourism sector, including international resorts. Her strong interests in promoting sustainability and conservation through education have led her to undertake park visitor research with the NSW Office of Environment and Heritage.

Gerald Hagenloh is an architect with a long-standing interest in landscape planning, including that of recreationbased settings. He has worked on a range of consulting projects in this field. His speciality is progressive technical and mixed-method approaches in people monitoring and travel route development.