Most preferred photo zone types
An analysis of respondents’ most preferred PHZ types found that natural-type PHZs were generally preferred more than artificial-type PHZs (
Table 5). Among the natural-type PHZs, the garden type was the most preferred, with a frequency of 25. 4%, followed by the natural phenomenon type (24.6%), grotto type (10.4%), and mountain scenery type (8.4%). In contrast, the waterfall scenery type and rock scenery type were relatively less preferred, with frequencies of 2.6% and 1.3%, respectively. Among the artificial-type PHZs, the character type had the highest preference, with a frequency of 8.1%, followed by the moon-shaped type (4.7%), heart-shaped type (4.5%), and frame type (3.2%). In contrast, the preference frequencies for the storytelling type, animal-shaped type, wing-shaped type, letter type, historical symbol type, and fruit-shaped type were relatively low, at 1.6%, 1.6%, 1.5%, 1.0%, 0.8%, and 0.3%, respectively.
An analysis of the most preferred PHZ types by age group is presented in
Table 5. Among respondents in their teens, the character type was the most preferred, with a frequency of 18 people, followed by the natural phenomenon type (12), grotto type (10), and moon-shaped type (10). Among those in their 20s and 30s, the natural phenomenon type was the most preferred, with a frequency of 59 people, followed by the garden type (49), character type (30), grotto type (28), and moon-shaped type (13). Among respondents in their 40s and 50s, the garden type was the most preferred, with a frequency of 72 people, followed by the natural phenomenon type (48), mountain scenery type (30), and grotto type (20). Overall, the respondents in their 40s and 50s tended to have a higher preference frequency for the natural types; among the artificial types, the heart-shaped type (14) and frame type (10) were relatively more preferred. Regarding the age of visitors to garden tourism,
Benfield (2013) reported that the median age of visitors to the 2011 “Garden Walk Buffalo” urban garden tour was 55.5 years old.
Lee et al. (2019) found that the age group in their 40s and 50s represented a relatively high proportion of visitors to local garden festivals. Similarly,
Kim and Huh (2016) reported that the highest age proportion of visitors to arboretums was in their 40s. Based on previous studies, it was found that those in their 40s and 50s prefer garden landscapes and actively participate in garden tourism. The PHZ type most preferred by this group was also the garden type, which features garden landscapes, showing consistent results with the highest preference frequency. Among those in their 60s, both the natural phenomenon and garden types had the highest preference, with a frequency of 33 people each, followed by the mountain scenery type (15). Overall, this group had a high preference for natural types, but a low preference for artificial types, with fewer than 10 people preferring each of the artificial subtypes.
An analysis of the preferred age groups based on PHZ types with a total preference frequency of 20 people or more showed that among the natural-type PHZs, the natural phenomenon type was generally preferred across all age groups, while the garden type was preferred by all age groups except those in their 10s. The grotto type had a high preference frequency among most age groups except for those in their 60s, and the mountain scenery type tended to be preferred by those in their 40s and older. Among the artificial-type PHZs, the character type and moon-shaped type showed a high preference frequency among younger age groups, including those in their 10s, 20s, and 30s, while the frame type and heart-shaped type tended to have a high preference frequency among the respondents in their 40s and 50s (
Table 5).
Preference analysis by photo zone type
An analysis of respondents’ preferences for 16 types of PHZs revealed that, overall, the average preference for natural types of PHZs was higher than that for artificial types of PHZs (
Table 6). This finding aligns with the finding that the preference frequency for natural types of PHZs exceeded that for artificial types of PHZs (
Table 5). The natural types of PHZs exhibited above-average preference levels (ranging from 3.17 to 3.98) across all subtypes. Among them, the garden type received the highest preference, with a mean of 3.98, followed by the natural phenomenon type (3.94), waterfall scenery type (3.67), grotto type (3.46), mountain scenery type (3.45), and rock scenery type (3.17).
The seven artificial types of PHZs showed above-average preference levels (ranging from 3.01 to 3.49). Among these, the moon-shaped type had the highest preference, with a mean of 3.49, followed by the historical symbol type (3.17), frame type (3.12), heart-shaped type (3.09), storytelling type (3.08), character type (3.07), and animal-shaped type (3.01;
Table 6). The historical symbol type showed a low frequency of respondents who preferred it the most, with 5 people (0.8%), but its average preference was relatively high among the subtypes of artificial-type PHZs. The character type had the highest frequency of respondents who preferred it the most among the artificial types, with 50 people (8.1%), but its average preference was relatively low. The letter type, wing-shaped type, and fruit-shaped type had below-average preferences, with means of 2.94, 2.95, and 2.55, respectively. The percentages of respondents who did not prefer these types were also high, at 33.3%, 34.3%, and 48.4%, respectively. This result aligns with the finding that the frequencies of respondents who most preferred these types were relatively low, at 6 (1.0%), 9 (1.5%), and 2 (0.3%), respectively (
Table 5). The letter type and fruit-shaped type are actively used to emphasize a unique local landscape and create a lasting local image for tourists (
Lee and An, 2023;
Park, 2019), and the wing-shaped type is increasingly used as a photo zone for social media posts. However, since the preference for these types was rated as below average, it seems that design improvements are needed.
To determine the difference in preferences by respondent age group for 16 types of PHZs, a one-way ANOVA was performed. The results showed that preferences for all PHZ types, except for grotto type, natural phenomenon type, and letter type, had a significant difference in mean by age group at the 1% significance level (
Table 7). Preferences for grotto type and natural phenomenon type did not show a significant difference in mean by age group. Preference for the letter type showed a significant difference in mean by age group, at the 5% significance level. Teenagers’ preference for this type was above average with a mean of 3.16, while it was below average for other age groups.
Among the natural-type PHZs, the preference for mountain scenery and garden types was relatively high in the 40s–50s and 60s and older age groups. The average preference for the rock scenery type was 2.77 in the 20s–30s age group, which was below average and significantly lower than that of other age groups. The preference for the waterfall scenery type was also relatively low in the 20s–30s age group. Among the artificial-type PHZs, the preference for character and storytelling types was significantly higher in the 10s, while the preference was below average in the 60s and older age group. Preference for the frame type and wing-shaped type was above average in the 10s and 40s–50s. The average preference for the frame type was 2.99 in the 20s–30s and 2.89 in the 60s and older age group, while the average preference for the wing-shaped type was 2.75 and 2.71, respectively, indicating a below-average preference. Preference for the historical symbol type was generally above average, but significantly higher in the 10s. Preference for the heart-shaped type was below average in the 20s and 30s with a mean of 2.71, while preferences were all above average in the other age groups. The preference for the moon-shaped type was generally above average, with a particularly higher preference in the 10s compared to other age groups. The preference for the fruit-shaped type and animal-shaped type was significantly higher in the 10s, with means of 3.19 and 3.71, respectively, while preferences in the other age groups were all below average.
When summarizing the respondents’ preferences by PHZ type, the preferences for all natural types of PHZs, except for the rocky scenery type, were above average for all age groups; and among artificial type PHZs, the preferences for the historical symbol type and moon-shaped type were above average for all age groups. This suggests that these types of PHZs could appeal to tourists of different age groups in garden tourism. Moreover, among the artificial-type PHZs, the character type and storytelling type were highly preferred by the 40s-50s age group, which was the main visitor group for garden tourism (
Benfield, 2013;
Kim and Huh, 2016;
Lee et al., 2019), as well as by the 20s–30s age group, which was reported to be more familiar with online searches for tourism activities (
Zhang et al., 2019). All age groups except the 60s and older showed an above-average preference for these types, indicating that they have a strong potential for introduction to garden tourism. Preferences for the letter type, fruit-shaped type, and animal-shaped type were low overall (
Table 6), but were above average in the 10s age group, significantly higher than for other age groups. Therefore, it is suggested that these types of PHZs could enhance tourism attractiveness and user satisfaction when introduced to themed gardens targeting children and adolescents.
Image analysis by photo zone type
To analyze the images recognized in the natural types of PHZs, a multiple regression analysis was performed, with the preference for the six natural types of PHZs as the dependent variable and the images identified in each PHZ type as the independent variables. The results are shown in
Table 8. All six regression models derived were statistically significant at the 1% significance level. Additionally, the tolerance of the input variables was greater than 0. 1, and the variance inflation factor (VIF) was less than 10, indicating that multicollinearity was not a problem (
Lee and Lim, 2018;
Park et al., 2022;
Huh et al., 2022). The image variables that predicted the preference for the grotto type were “beautiful,” “natural,” “interesting,” “mysterious,” and “impressive.” The R
2 value of the regression model was 0.591, indicating that these five independent variables explained 59.1% of the preference. Among the variables included in the model, “natural” had the highest explanatory power for the preference for grotto-type PHZs, with a beta value of 0.282, followed by “beautiful,” “interesting,” “mysterious,” and “impressive.” The image variables that predicted the preference for the mountain scenery type were “impressive,” “natural,” “touching,” “interesting,” and “mysterious.” The R
2 value of the regression model was 0.517, indicating that these five independent variables explained 51.7% of the preference. Among the variables included in the model, “natural” had the highest explanatory power for the preference for the mountain-scenery-type PHZs, with a beta value of 0.314, followed by “impressive,” “touching,” “interesting,” “mysterious,” and “mysterious.”
The image variables that predicted the preference for the natural phenomenon type were “beautiful,” “impressive,” “natural,” “interesting,” “touching,” and “mysterious,” and the R2 value of the regression model was 0. 573, indicating that these six independent variables explained 57.3% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for natural-phenomenon-type PHZs with a beta value of 0.264, followed by “natural,” “interesting,” “impressive,” “touching,” and “mysterious.” The image variables that predicted the preference for the rock scenery type were “beautiful,” “touching,” “natural,” “interesting,” “impressive,” “fancy,” and “friendly,” and the R2 value of the regression model was 0.606, indicating that these seven independent variables explained 60.6% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for the rock-scenery-type PHZs with a beta value of 0.248, followed by “natural,” “touching,” “interesting,” “impressive,” “fancy,” and “friendly.”
The image variables that predicted preference for the waterfall scenery type were “beautiful,” “interesting,” “distinctive,” “natural,” and “friendly,” and the R2 value of the regression model was 0.554, indicating that these five independent variables explained 55.4% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for preference for waterfall-scenery-type PHZs with a beta value of 0.433, followed by “interesting,” “distinctive,” “natural,” and “friendly.” The image variables that predicted the preference for the garden type were “beautiful,” “impressive,” “touching,” “friendly,” “interesting,” and “fancy,” and the R2 value of the regression model was 0.626, indicating that these six independent variables explained 62.6% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for garden-type PHZs with a beta value of 0.449, followed by “impressive,” “touching,” “interesting,” “friendly,” and “fancy.”
In the six regression models, the image variables that had high explanatory power for the preference for natural-type PHZs were “beautiful,” “natural,” “impressive,” and “interesting.” Notably, the variables “beautiful” and “natural” had a significant impact on the preference. The variable “impressive” had high explanatory power for the preference for mountain scenery type and garden type, while the variable “interesting” had high explanatory power for the preference for waterfall scenery type. Therefore, when planning a natural-type PHZ, it may be beneficial to suggest framing for photo shoots by exploring and selecting a location with a landscape that offers beauty and naturalness as a background (
Lee, 2010); or to consider designing a landscape that creates a strong impression or interest through natural materials, such as a garden or waterscape, and using it as a backdrop for photo shoots.
To analyze the images recognized in the artificial types of PHZs, a multiple regression analysis was performed with the preference for 10 artificial types of PHZs as the dependent variable and the images identified in each type as the independent variable. The results a re shown in
Table 9. All 10 regression models derived were statistically significant at the 1% significance level. Moreover, the tolerance of the input variables was greater than 0.1, and the VIF was less than 10, indicating that multicollinearity was not a problem (
Lee and Lim, 2018;
Park et al., 2022;
Huh et al., 2022). The image variables that predicted preference for the character type were “impressive,” “beautiful,” “interesting,” “natural,” “friendly,” and “mysterious,” and the R
2 value of the regression model was 0. 625, indicating that these six independent variables explained 62.5% of the preference. Among the variables included in the model, “impressive” had the highest explanatory power for the preference for the character-type PHZs with a beta value of 0.219, followed by “interesting,” “natural,” “beautiful,” “friendly,” and “mysterious.” The image variables that predicted the preference for the frame type were “beautiful,” “impressive,” “natural,” “interesting,” “simple,” and “touching,” and the R
2 value of the regression model was 0.610, indicating that these six independent variables explained 61.0% of the preference. Among the variables included in the model, “natural” had the highest explanatory power for the preference for frame-type PHZs with a beta value of 0.246, followed by “interesting,” “impressive,” “beautiful,” “simple,” and “touching.”
The image variables that predicted letter type preference were “beautiful,” “natural,” “simple,” “friendly,” and “interesting,” and the R2 value of the regression model was 0.521, indicating that these five independent variables explained 52.1% of the preference. Among the variables included in the model, “natural” had the highest explanatory power for the preference for letter-type PHZs with a beta value of 0.263, followed by “beautiful,” “simple,” “friendly,” and “interesting.” The image variables that predicted the preference for the historical symbol type were “natural,” “distinctive,” “friendly,” “symbolic,” “interesting,” “mysterious,” and “touching,” and the R2 value of the regression model was 0.578, indicating that these seven independent variables explained 57. 8% of the preference. Among the variables included in the model, “natural” had the highest explanatory power for the preference for the historical-symbol-type PHZs with a beta value of 0.294, followed by “interesting,” “friendly,” “symbolic,” “mysterious,” “distinctive,” and “touching.”
The image variables that predicted the storytelling type preference were “interesting,” “natural,” “touching,” “fancy,” “impressive,” and “symbolic,” and the R2 value of the regression model was 0.545, indicating that these six independent variables explained 54.5% of the preference. Among the variables included in the model, “interesting” had the highest explanatory power for the preference for storytelling-type PHZs with a beta value of 0.236, followed by “natural,” “fancy,” “touching,” “impressive,” and “symbolic.” The image variables that predicted the preference for the heart-shaped type were “beautiful,” “interesting,” “natural,” “impressive,” “touching,” “symbolic,” and “mysterious.” The R2 value of the regression model was 0.707, indicating that these seven independent variables explained 70.7% of the preference. Among the variables included in the model, “interesting” had the highest explanatory power for the preference for the heart-shaped-type PHZs with a beta value of 0.249, followed by “beautiful,” “natural,” “impressive,” “mysterious,” “symbolic,” and “touching.”
The image variables that predicted the preference for the wing-shaped type were “beautiful,” “natural,” “impressive,” “fancy,” and “interesting,” and the R
2 value of the regression model was 0.624, indicating that these five independent variables explained 62.4% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for wing-shaped-type PHZs with a beta value of 0.408, followed by “natural,” “impressive,” “interesting,” and “fancy.” Since the regression coefficient for the variable “fancy” was negative (b = −0.115; beta = −0.110), the stronger the “fancy” image was perceived in wing-shaped PHZs, the lower the preference tended to be. As such, the perceived “fancy” image in wing-shaped PHZs has a negative explanatory power for preference, suggesting that the “fancy” image had an adverse effect on the wing-shaped type, which already had a relatively low average preference (
Table 6). Therefore, when designing a wing-shaped PHZ, it is suggested to consider incorporating a more natural image rather than a fancy one. The image variables that predicted the preference for the moon-shaped type were “beautiful,” “natural,” “mysterious,” “simple,” “interesting,” and “touching.” The R
2 value of the regression model was 0. 639, indicating that these six independent variables explained 63.9% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for the moon-shaped-type PHZs with a beta value of 0.437, followed by “mysterious,” “natural,” “interesting,” “simple,” and “touching.”
The image variables that predicted the preference for the fruit-shaped type were “beautiful,” “natural,” “interesting,” and “touching,” and the R2 value of the regression model was 0.602, indicating that these four independent variables explained 60.2% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for fruit-shaped-type PHZs with a beta value of 0.333, followed by “natural,” “interesting,” and “touching.” The image variables that predicted the preference for the animal-shaped type were “natural,” “interesting,” “beautiful,” and “friendly,” and the R2 value of the regression model was 0.614, indicating that these four independent variables explained 61.4% of the preference. Among the variables included in the model, “beautiful” had the highest explanatory power for the preference for animal-shaped-type PHZs with a beta value of 0.287, followed by “natural,” “interesting,” and “friendly.”
In the 10 regression models, the image variables that had a high explanatory power for the preference for artificial-type PHZs were “beautiful,” “natural,” “interesting,” and “impressive.” Notably, “natural” and “beautiful” images had a significant effect on the preference. These results showed the same or similar trends as those derived from the regression model for the preference for natural-type PHZs. The variable “impressive” had a high explanatory power for the preference for the character type, while the variable “interesting” had a high explanatory power for the preference for the character type, storytelling type, heart-shaped type, and animal-shaped type. Since these types are mainly composed of specially shaped sculptures, it was suggested that the interest provided by the sculptures had a positive effect on the preference. Therefore, when designing artificial-type PHZs, it seems desirable to encourage user preference by using materials that harmonize with the environment and create a space that evokes naturalness and beauty. As in the design of PHZs, the overall image was perceived as more important than detailed design elements, notably for sculpture-centered PHZs, it seems possible to create a space that captures visitors’ interest and enhances their desire to take photos by seamlessly integrating such zones into the surrounding landscape and atmosphere, and incorporating detailed elements such as the shape, color, and material of the sculptures (
Ji and Kim, 2022). Previous studies on the attributes of tourist destinations suggest that the creation of PHZs that capture visitors’ interest can be a key attribute in enhancing tourist satisfaction and serve as a strong point for tourism destinations (
Kim and Jun, 2021;
Kim et al., 2016).