The Preference of Domestic Tourist Towards the Tourism Destination During Pandemic Period in Bali

The COVID-19 Pandemic that occurred around the world had an impact on the decrease in people's income which has implications for changing behavior in choosing tourist destinations. This study aims to: 1) analyze the relative importance attribute from Indonesian tourists who came to Bali during the COVID - 19 Pandemic 2.) Get the most optimal combination of tourist destination products choice based on the preferences of Indonesian tourists during the COVID – 19 Pandemic. The research was conducted at Benoa Harbour, I Gusti Ngurah Rai Airport and Kuta Beach from January to April 2021 through the dissemination of questionnaires to 160 Indonesian tourists by accidental sampling. Data analysis by con-join method and analyzed with SPSS 2.0 program. The results showed that the relative importance attribute sorted from the highest value to the lowest value is cleanliness, price, attractiveness, service, recommendations, security, and promotion. Tourists' preference is the willingness of hygiene facilities, the price of > Rp.50.000, flexible artificial tour booking, friends/family, low case rates and discounts on airline tickets.


INTRODUCTION Background
Tourism is one of the industrial sectors that has been affected byCOVID-19 pandemic. The paralysis that occurred in the tourism sector was caused by restrictions on activities, some destinations were closed, and some events such as concerts had to be cancelled, resulting in a decrease of tourist visits. The tourism industry itself is the main sector for the regional economy of Bali Province and many people depend on the tourism sector so that the tourism sector is a very important sector in supporting the welfare of the community. The drastic decline in tourist arrivals had a huge impact on the Balinese economy and during the pandemic, many people lost their jobs and were forced to close their businesses. One solution to maintain people's income and economy is to rebuild the tourism sector by increasing tourist visits.
According to Horner and Swarbrooke (2007) there are two main things that are considered by tourism consumers in choosing a destination, they are: 1) factors that determine whether consumers can take vacations or not and 2) factors that determine the type of trip and experience if the traveler decides to travel. Besides finding out the motivational factors of tourists, it is further necessary to know the changes in tourist behavior caused by changes in habits during the COVID-19 Pandemic.
According to Kotler (2008), consumer behavior is a process and a person's activities related to the quest, selection, purchase, usage, and evaluation of products or services to meet consumer wants and needs so that in the process there is a form of preference from these consumers. Consumer behavior can be influenced by the circumstances and situations of the surrounding community such as the place where consumers were born and raised. This means that consumers with different origins or environmental layers of society will have different assessments, needs, opinions, attitudes, and tastes, so that decision making in the buying stage will be influenced by several of these factors.
One way to revive Bali tourism is to make efforts to attract tourists to visit Bali by understanding changes in tourist behavior and preferences during the visit. Through information about the wishes, needs and priorities of tourists, it can be used as a basis for consideration in creating tourism products that are appropriate and needed by tourists currently.
Based on the existing problems and urgency, the research objectives are 1) Obtaining the most optimal combination of tourist destination products based on the preferences of domestic tourists during COVID-19 Pandemic 2) Analyzing the level of interest of domestic tourists who come to Bali during COVID-19 Pandemic.

Consumer behavior
The definition and understanding of consumer behavior is quite difficult and very complex. This is because there are quite a number of factors that can influence and these factors tend to interact with each other. Consumer behavior is a number of real actions of individuals or consumers that can be influenced by internal and external factors and then direct them to assess, choose, obtain and use the goods and services they want. Consumer behavior (customer behavior) can be defined as the activities of an individual who is directly involved in the process of obtaining and using a product or service, including a decision-making process.

Conjoint Analysis
Referring to the theory of Hair, et al. (2010) stated that conjoint analysis is a multivariate technique that has been specifically developed to be able to understand a person's preferences regarding a type of object such as a product, service, or idea. In the conjoint analysis there are 9 stages, they are: 1) determine the purpose of conjoint analysis, 2) determine the attributes and level of research attributes, 3) design the conjoint model, 4) determine the presentation model, 5) design stimuli, 6) determine the estimation measurement technique, 7) evaluate the conjoint model, 8) perform interpretations results and 9) measuring the accuracy of predictions.

Traveler's Preferences
According to Frank (2011) preference is the process of ranking all things that can be consumed with the aim of obtaining a preference for a product or service. According to Schiffman and Kanuk (2009), preference is defined as the way a person sees the world and its surroundings. Tourist preferences appear at the alternative evaluation stage in the purchasing decision process where at that time tourists are faced with a variety of different product and service choices so that at this time tourists will rank priorities according to their wishes and needs. This study will focus on the preferences of domestic tourists in choosing tourist destinations in Bali. Having knowledge of these market patterns will produce an analysis that can be used as a strategy in increasing domestic tourist visits to Bali.

Domestic Tourist
A domestic tourist is someone who travels in the territory of a country, in the area of Indonesia with a trip duration of less than 6 months and does not aim to earn income at the place visited and is not a routine trip (school or work), to visit tourist objects, commercial accommodation, and/or stay in commercial accommodation, and/or travel distance greater than or equal to 100 (one hundred) kilometers round trip (Report of Nusantara Tourist Statistics, 2018).

Tourist Characteristic
According to Seaton dan Bennet (in Suwena and Widyatmaja, 2010) the matter concerning tourist can be categorized into two types, they are based on: 1.) trip descriptor the characteristic of travel such as recreation, visiting relative, business journey, and other trips 2) tourist descriptor, it is tourist characteristic based on socio demographic, geographical characteristic, and psychogeography characteristic.

Travel Destinations
According to the Tourism Law No. 10. Year 2009 (article 1 paragraph 6) states that tourism destination areas, hereinafter referred to as tourism destinations, are geographical areas located within one or more administrative areas of public facilities, tourism facilities, accessibility, and communities that are interrelated and complement the realization of tourism.

Destination Selection Factors
According to Middleton & Fyall (2009) there are several factors from tourist destinations that can affect tourists, namely attractiveness, availability of facilities, accessibility, image of tourist destinations, prices of tourist destinations, and promotions. These factors are adjusted to the wishes and needs of individual tourists in choosing the products/services they want. These factors will produce tourist preferences that are driven by their motivation. In this study the use of several attributes that are adapted to current conditions and needs, namely: (1) price, (2) tourist attraction, (3) recommendations, (4) security, (5) cleanliness (6) service and (7) promotion by taking into account changes in tourist behavior as a benchmark in making statements.

METHODS
The time of research was carried out from January to April 2021 in three strategic locations, the location was I Gusti Ngurah Rai Airport, Gilimanuk Harbor and Kuta Beach. The choice of location was based on considerations regarding the exit and entry of tourists, namely I Gusti Ngurah Rai Airport and Gilimanuk Harbor, while the consideration of choosing Kuta Beach was based on strategic locations and tourism favorites in Bali.
The process of determining the sample in this research is based on accidental sampling technique, which is a process of determining the sample which is carried out without careful planning (Kusmayadi, 2004). Tourists who were selected as research respondents had criteria, namely tourists who came to Bali during COVID-19 Pandemic. Referring to the theory of Hair et.al. (2010) that the best sample size has 100 samples or larger and according to general rules, the minimum number of respondents is fiove times more than the total number of question items that have been determined, and the sample size is acceptable if the sample uses a ratio of 5:1 to 10:1. In this study there were 16 total question items, so the sample size needed in this study was a minimum of 16 x 10 = 160 samples. e-ISSN 2407-392X. p-ISSN 2541-0857 This study uses a quantitative descriptive approach which is supported by a qualitative descriptive analysis. The analytical method used is conjoint analysis technique. In the conjoint analysis there are several steps that must be done (Hair et. al., 2010).

Determining Attribute and Attribute
Level of Research The Attribute Determination and Attribute Level of Research refer to the theory and concept of research. Airplane ticket discount Source: Primary data (processed data), 2021.

Designing the Conjoint Model
In this research, the presentation method applied was the traditional conjoint model, where the respondents assessed all attributes at the same time and give a value with a predetermined value scale. The method has criteria, namely the maximum number of attributes is 8 attributes, can be done manually or computerized and can produce individual data analysis.

Determining the Presentation Model
The presentation model used in this research is a full profile presentation model. Through the use of this model, it is possible to reduce the number of stimuli by using the fractional factorial design method, which is a special class of fractional design, called an orthogonal array so that it is possible to estimate all the main effects.

Designing Stimuli
The design of the combination of stimuli consists of a group of attributes and attribute levels that will be evaluated by the respondent. The process of designing the combination of stimuli will use the SPSS 2.0 program with the Orthogonal Design method so that the number of combinations of stimuli in this study is 16 combinations as follows. Source: Primary data (processed data), 2021.

Determine the Estimation Technique
In this research, the estimation technique applied is a rating technique with a scale of 1 to 5 with the provisions of a value of 1 = least preferred and 5 = most preferred.

Measuring the Accuracy of Predictions
According to Gudono, (2015) the measurement of predictive accuracy can be seen through the correlation between the actual value and the predicted value with Kendall Tau, and the Pearson correlation. As for the assessment provisions, if the significance value is < 0.05 then each attribute has a strong significance relationship and if the correlation value is > 0.5 then there is a close relationship between the actual and predicted values.

Measuring the Accuracy of Predictions
According to Gudono, (2015) the measurement of prediction accuracy can be seen through the correlation between the actual value and the predicted value with Kendall Tau, and the Pearson correlation. The assessment provisions are if the significance value is <0.05 then each attribute has a strong significance relationship and if the correlation value is >0.5 then there is a close relationship between the actual and predicted values.

RESULT AND DISCUSSION
The main results of the conjoint analysis in the form of attribute utility values and attribute levels, the value of the importance of attributes either individually or as a whole (aggregate) then to determine the level of relationship between the actual value and the predicted value, the measurement of the level of Kendall Tau correlation or Pearson's correlation can be used. As for the discussion of the analysis will be carried out as a whole (aggregate). The following is a complete interpretation of the results.

Attribute Level Utility Analysis
The utility value explains the respondent's choice of attribute levels in tourist destinations. The provisions of the utility value can be stated as follows. If a positive utility value states that the respondent chooses the level of an attribute, if a negative value states that the respondent does not choose or reject the level of an attribute, and if the utility value at an attribute level has a value of zero, it means that the attribute level is neutral with each other. or the attribute level does not affect the respondent. Source: Primary Data (processed data), 2021.
Based on Table 3, the results of the attribute utility value analysis of 160 respondents who have evaluated the questionnaire show:

Price Attribute Level Utility.
The results of the analysis of the utility value of the price attribute show that the price attribute level above Rp. 50,000 gets a value of 151, the price attribute level of Rp. 10,000 -50,000 gets a value of 020 and the level of the price attribute less than Rp. 10,000 gets a value of -170 then this shows that in general the response-en prefers destination entry prices above Rp. 50,000 then likes destination entry prices with a price range of Rp. 10,000 -Rp. 50,000 and do not like the entry price for destinations below Rp. 10,000. Based on the results of the tourist questionnaire, it was argued that a tourist destination with an entry price above Rp. 50,000 would be more promising for comfort, security and the experience that would be obtained such as Garuda Wisnu Kencana. If it is associated with the characteristic profile of tourists who came during the COVID-19 Pandemic to Bali, it can be concluded that the selection of a price attribute level above Rp. 50,000 as the preferred level can be influenced by the characteristics of tourists who have an income above 7.2 million per month.

Attractiveness Attribute Level Utility Value
The results of the analysis of the utility level of the attraction attribute show that the attribute level of artificial tourist attraction gets a value of 022, the attribute level of natural tourist attraction gets a value of -004 and the attribute level of cultural tourist attraction gets a value of -018, so this shows that in general tourists prefer attractiveness. artificial tourism such as shopping centers, zoos, museums and so on, while for natural and cultural tourist attractions are not accepted or disliked by tourists. Based on the results of the questionnaire, some tourists who choose artificial tourist attractions are based on joint decisions with their families and are more interested in artificial tourism that is providing discounted prices than natural tourist attractions that have been visited.

Recommended Attribute Level Utility
The results of the utility analysis of the recommendation attribute level show that the recommendation attribute level from friends or family gets a value of 082 and the recommendation attribute level from social media gets a value of -082, this shows that tourists prefer recommendations given by their closest friends and family and do not really like the recommendations from social media. Based on the results of the questionnaire, tourists think that assessments from closest friends and family are more appropriate and real when compared to assessments from social media.

Security Attribute Level Utility
The results of the security attribute level utility analysis show that the security attribute level of destinations in the lowcase zone gets a value of 007 and the security attribute level of CHSE-certified destinations gets a value of -007, this shows that tourists prefer the security attribute level of destinations that are in the low-case zone and refuse or reject. disliked the level of security attribute of CHSE certified destinations. Based on the questionnaire, tourists think that Bali is a fairly safe tourist destination when compared to the location of origin of tourists and some tourists do not pay much attention to destinations that have been certified by CHSE.

Cleanliness Attribute Level Utility
The results of the utility level analysis of the cleanliness attribute show that the level of the cleanliness attribute of the destination with the availability of clean facilities gets a value of 317 and the level of the cleanliness attribute of the destination that is maintained gets a value of -317, so this shows that tourists prefer destinations that have the availability of clean facilities and do not really like the cleanliness of destinations that are clean. Based on the tourist opinion questionnaire, the selection of destinations that provide clean facilities such as hand sanitizers, sinks and disinfectants is more important during this COVID-19 pandemic.

Service Attribute Level Utility
The results of the service attribute level utility analysis show that the service attribute level in the form of ordering flexibility gets a value of 142 and the service attribute level in the form of accident insurance availability gets a value of -142, this shows that tourists prefer the service attribute level in the form of booking flexibility and do not really like the service attribute level in the form of Availability of accident insurance. Based on the questionnaire, tourists argue that in uncertain times such as today, the flexibility of ordering in the form of changes and cancellations of orders is very important to ensure that ordered products purchased by tourists can be refunded or rescheduled.

Promotion Attribute Level Utility
The results of the utility level analysis of promotional attributes show that the attribute level in the form of discounted airline tickets gets a value of 028 and the attribute level in the form of discount accommodation gets a value of -028, so this shows that tourists prefer the attribute level in the form of discount airline tickets and do not really like the attribute level. accommodation discount. Based on the results of the questionnaire, tourists think that they need a discount on airplane tickets because the expenditure is quite high for tourists. In addition, if you refer to the characteristics of the research respondents who are dominated by tourists from DKI Jakarta and surrounding areas, the selection of discounted flights is quite appropriate considering the distance from the location of origin of the tourists.
Based on the description above, it can be concluded that the most optimal combination of tourist destination attributes consists of destination entry prices above Rp. 50,000, tourist destinations with artificial tourist attractions such as museums, shopping centers, sports venues and others, have been directed. recommended by close friends or family, the location of the destination is in a low-case area of COVID-19, getting services in the form of easy bookings, cancellations and changes to tourist activities, and getting discounted flight tickets.

Attribute Importance Level Analysis
Importance values analysis is used to see the results of respondents' evaluation of the importance value of a particular attribute, where the higher the value of the importance level of the attribute, the higher the influence of these attributes in influencing the choice of tourist destinations. Source: Primary data (processed data), 2021.
Based on Table 4, the results obtained that the analysis of the level of importance of the attributes from the highest value to the lowest value are as follows: (1) the cleanliness attribute with the highest importance value, which is 21,543, (2) the price attribute with an importance value, which is 18,873. , (3) attractiveness attribute with an importance value of 17,608, (4) service attribute with an importance value of 13,575, (5) recommendation attribute with an importance value of 9,871, (6) security attribute with an importance value of 9,609, and (7) promotion attribute with an importance value of 8,922. Based on this description, it can be concluded that there are three attributes that are highly preferred and considered by domestic tourists in choosing tourist destinations during the COVID-19 Pandemic, namely cleanliness attributes, price attributes and attractiveness attributes.
The attribute of cleanliness is the most important attribute of the choice of tourists in choosing a tourist destination during the COVID-19 Pandemic. This is due to the increasing public attention to the cleanliness aspect to protect themselves from exposure to the corona virus. According to Suprihatin, (2020) tourist behavior during the COVID-19 pandemic is motivated by the motivation to get satisfaction from tourism products that pay attention to aspects of protection and comfort in the form of cleanliness, health, and safety as the main motivation in purchasing decision attitudes. . To be able to meet the needs of safe travel, the cleanliness of tourist destinations is a benchmark that is highly considered by tourists. According to Glimpses of Recovery Journal, (2020) cleanliness is a driving force for material demand, differentiating, and supporting factors for the image of a product so that the cleanliness aspect can improve the image of the destination for the better to gain a sense of trust from tourists towards the destination.
The price attribute is the second most important attribute for tourists in choosing a tourist destination. Price itself is one of factors that influence tourists in choosing a tourist destination where there are considerations regarding the type of destination, the quality of the destination, and the suitability of the price with the destination to be visited. In the COVID-19 Pandemic situation, which has caused a decline in people's incomes and even a global economic downturn, many tourists consider the price aspect as a form of efficiency. This is supported by Yuni's statement (2020) that financial considerations and travel costs are a major concern for almost all tourists.
The attractiveness attribute is the third attribute that is an important consideration for tourists. According to Middleton & Fyall (2009) without an attraction it will be very difficult to make tourists go on vacation to a tourist destination, but in this study the attribute of attraction is in the third position in the level of tourist interest. This is due to the priority of tourist behavior which is adjusted to the needs during the COVID-19 Pandemic.

Correlation Value
The main purpose of calculating the accuracy of predictions is to get a significant and correlation value between the estimated results and the actual results.

Value
Sig.
Based on Table 5, the results of the Pearson's correlation value are 0.744 with a significant value of 0.000 and the results of the Kendall's Tau correlation value are 0.56 with a significance value of 0.001. The results of the correlation values from the two analyzes show a value of > 0.5 and a significance value of < 0.05, so it is concluded that there is a close relationship between the results of the study (actual value) and the expected preference (prediction value).

Conclusion
According to the results of the research conducted on 160 respondents, the results of tourist preferences for destination selection during the COVID-19 Pandemic in Bali, can be concluded as follows: 1. Based on the characteristics of tourists who came to Bali during COVID-19 pandemic, it was found that tourists from DKI Jakarta, were male, aged 26-55 years, with the last education of bachelor degree, worked as a private employee, are married, and have an income above Rp. 7,200,000 per month are the characteristics of tourists who are the most respondents in this study. 2. Based on the travel patterns of tourists who came to Bali during COVID-19 Pandemic, it was found that tourists with a visit period during school holidays, with a visit length of 4-7 days, had visited 2-5 times, planning a trip personally, using air transportation, choosing five-star accommodation, having vacation with family, and the main purpose of travel is for vacation are the characteristics of tourists based on travel patterns who are the most respondents in this research. 3. The most optimal combination of attributes based on tourist preferences is destination entry prices above IDR 50,000, tourist destinations with artificial tourist attractions such as museums, shopping centers, sports venues and others, have been recommended by close friends or family, the location of the destination is in an area with low cases of COVID-19, getting services in the form of ease of ordering, canceling and changing tourism activities, as well as getting discounted flight tickets. 4. The level of importance of the attributes that are most preferred by tourists from the highest to the lowest are (1) cleanliness attributes, (2) price