The study was conducted in 24 tourist villages representing tourist villages located in Central Java, East Java, West Java, Special Region of Yogyakarta, and West Nusa Tenggara. The selection of these villages was based on several factors: the availability of data, they were suspected of being affected by the Covid-19 shock, and their presence was assessed as having impacted the economy and social the rural society. Table 1 provides an overview of the profiles of the 24 villages and their main tourist attractions.

This study uses secondary data in the form of documents of tourist village management. The data collection process uses the documentation observation method. To assess the resilience of these villages in facing the Covid-19 shock, data was collected from 2019 (before the Covid-19 pandemic) to 2022 (during and after the Covid-19 pandemic).

This study aims to assess the resilience of tourist villages over time, specifically before and after the Covid-19 pandemic. By incorporating a time dimension and utilizing a composite index, this approach provides a unique perspective on the interplay between normalization, aggregation, and temporal factors (Bacchini et al., 2020). While previous research by (Frigerio et al., 2018) has explored the concept of vulnerability in relation to resilience, this study does not include spatial correlation analysis. This study focuses on examining the temporal variations in resilience levels across 24 tourism villages, dividing the study into three distinct periods: pre-crisis (2019- 2020), crisis (2020-2021), and post-crisis (2021- 2022). This stage follows the concept of crisis management, which consists of three steps: pre- crisis, crisis, and post-crisis (Coombs, 2023)

To measure economic resilience of tourist village is used a synthetic indicator known as the Adjusted Mazziotta-Pareto Index (AMPI) was developed by (Mazziotta & Pareto, 2016). AMPI is a composite index that allows data comparison across units and time, resulting in compatibility with the spatiotemporal approach. The choice of the AMPI method also corresponds to the characteristics of a tourism system that is complex, dynamic, and associated with many variables (Baggio, 2020) (Lv et al., 2021) so that it cannot be measured by a single indicator ((Mazziotta & Pareto, 2013); (Mazziotta & Pareto, 2017); (Scaccabarozzi et al., 2022)).

The AMPI method started with normalizing data or indicators using the following formula:

where is the matrix of n rows containing unit analysis and m columns containing indicators, and and are the goalspots for indicator j. Such normalization is a refinement of the MPI method to appreciate absolute changes over time (Mazziotta & Pareto, 2104) The range of normalization is varied between 70 and 130. If we denote and as the mean and standard deviation of normalized value of unit , respectively, the generalized form of AMPI is given by the following equation:

Where represents the coefficient of variation of the unit i. The sign indicates whether the phenomenon to be measured is maximized (the higher, the better) or minimized (the lower, the better).

As stated earlier, the AMPI needs a “goal spot” to facilitate interpreting the results. A reference point of 100, which is the average of indicators in a given year, is used. The AMPI value higher or lower than this reference point indicates whether the unit being analyzed is progressing or regressing. In this case, it indicates whether the units are more resilient or vice versa. The procedure to set the goal spot is the following:

where and represents the maximum and the minimum of indicator across all periods and the reference value of indicator j (Mazziotta & Pareto, 2017).

Measuring resilience using the AMPI method involves several stages, starting with the selection of indicators for tourism village resilience. At this point, due to the lack of unanimous consensus on both definitions and measurement ((Stanícǩová & Melecký, 2018) (Martin, 2012); (Muštra et al., 2020)), the process of determining economic resilience indicators in this study adheres to specific principles. These principles must involve various dimensions and variables that are considered proxies for a phenomenon, allowing for appropriate actions to be taken (Proag, 2014). This concept of vulnerability signifies a measure of risk associated with the physical, social, and economic aspects and the implications arising from the system's ability to cope with the resulting event. Concepts of resilience can take two broad forms, namely: (1) the selection of non-neutral variables, each representing a different aspect of resilience in relation to shocks, depending on the territorial context (Compagnucci & Urso, 2021).

In this study, economic resilience tourist village indicators are compiled from two main components: capacity and performance dimension. Capacity relates to resources that are part of the tourist village system, which is measured through indicators: (1) capacity building (cbdg), (2) employees (emp): (3) Village Development Index (VDI). Meanwhile, performance is related to the results of the work of the tourist village during and after disturbances, which is measured through indicators: (1) tourist (trs) (2) income (inc), (3) cost (cst). Indicators of tourist village economic resilience are listed in Table 2.

In accordance with the AMPI stages, data on resilience indicators from 24 tourist villages is normalized first. Data normalization results are shown in Table 3,Table 4, Table 5.

Futhermore, a typology of tourist village resilience will be presented in the 2019-2022, 2020-202, and 2019-2021 periods to describe the resilience characteristics of each tourist village. Determining the resilience typology of tourist villages uses the resilience trend matrix developed by (Compagnucci & Urso, 2021), which classifies resilience typologies into eight types (Table 7). This resilience trend matrix helps investigate how the use of different indicators will affect resilience measures and helps explore whether certain indicators are more appropriate for assessing the resilience of tourist village.

The resilience trend metric will analyze and compare the conditions of each village based on sequentialn variations of the resilience index (∆) during the 2019-2022 period, calculated as the geometric value of the resilience index through the following equation:

[1]

[2]

[3]..[6]

In equation [1] we consider the ratio between number of capacity building activities (Σcbdg) at the end of the period (t+k) on the value the capacity building activities (t), we raise the result to the power of one divided by the period length (k) and we subtract one from the subsequent result. In equations [2], [3] to equation [6] according to the number of variables, we perform the same calculations using the variables number of tourists (Σ trs), total income (Σ inc), number of employees (Σ emp), and total costs (Σ cost) and the value of the Village Development Index (VDI)2.

Table 8 shows that one year after the pandemic, all tourist villages suffered from a lower performance, as indicated by decreases in their AMPI scores. The delta (Δ) score from 2019- 2020 showed changes in the resilience index of 24 tourist villages in 2019-2020. This period was the most critical period to hit the tourism village. All 24 villages studied experienced a decreased resilience index (negative delta AMPI). The impact, however, affected villages differently. Some experienced a slight reduction in their resilience score, while others were significantly affected.

During 2019-2020, the Pentingsari tourist village experienced the most significant plunge in the resilience index compared with other villages. The government’s travel ban policy resulted in a degradation dramatically in the number of tourist arrivals. Therefore, the tourist village, offering rural and agricultural cultural attractions, closed its services rather than bearing costs disproportionate to its income. However, several tourist villages remain open despite the number of visitors and their income decreasing sharply. They kept their activities to maintain their status of tourist village by engaging in other activities such as training their staff or maintaining facilities.

Compared with 2019-2021 and 2020-2021, they have depicted extreme differences in the tourist village resilience index. From 2020 to 2021 (one year after the Covid-19 pandemic), almost all villages, except Tambaksari village, showed a remarkable recovery evidenced by positive changes in their AMPI scores (delta AMPI positive). These conditions indicate that during this period, the tourist village has adapted to the shocks caused by Covid-19. The resilience index generally increases because tourist villages implemented health protocols in the tourism sector (Cleanliness Healthy Safety Environment or CHSE Protocol) and conducted training related to services during the new normal period, and several villages modified tourist destinations in digital formats by offering travel packages. Digital tourism was developed to target visitors who could not visit in person or were still afraid of catching Covid-19. In addition, the recovery was also supported by the government policy that gradually reopened tourism activities. As a result of this policy, the number of visitors gradually increased.

As seen from Table 8, during 2020-2021 (Δ2), the most significant increase in the resilience index occurred in Karangrejo village. Karangrejo village is a community-based tourist village offering rural and agricultural cultural attractions. With full support from the community, especially in providing lodging facilities and implementing the CHSE protocol to ensure visitors’ health during their tours, this village raised the number of visitors, followed by increases in other resilience indicators.

Table 9 compares AMPI resilience scores before, during, and after Covid-19 (i.e., from 2019 to 2022, using 2022 as a “goal spot”). As can be seen from Table 9, the overall AMPI scores showed significant climbs toward 2022, indicating a strong recovery trend from the shock. The most significant rise in AMPI scores occurred in 2021- 2022, when all villages experienced an increase in the index, reaching double digits. This condition illustrated that tourist villages have adapted to the Covid-19 shock and can be fully recovered. One of the villages that showed strong resilience, indicated by the highest positive value of delta (Δ), was Karangrejo. One of the reasons for the success of Karangrejo village was its ability to build partnerships with several parties, especially with State-Owned Enterprises, by forming the Village Economic Center (known locally as Balkondes). This is in accordance with research (Fafurida, 2017) Public and private partnerships can increase the economic growth of the tourism sector.

In addition, the Tourism Awareness Group (known as Pokdarwis) has played a pivotal role in strengthening resilience to the Covid-19 shock. The collaboration of the two institutions is influential in developing creativity and encouraging visitor arrival. Karangrejo also

pointed out a high level of community involvement in providing homestays and other supporting facilities that have been adapted to health protocols, which have increased the performance of this tourist village, both in terms of the number of visitors and income. This condition is evidence of the successful implementation of community- based tourism, which has successfully dealt with external shocks. The success of the Karangrejo tourist village has earned it an award from the Indonesian government as a sustainable tourist village.

The results of the AMPI analysis (Table 8-Table 9) were used to determine and analyze the typology of tourist village resilience. Futhermore, based on the trend resilience scheme (Table 7), tourist villages are grouped based on their resilience trend (Table 10). Table 10 shows that at the start of the pandemic (2019-2020), all villages showed a negative AMPI index. However, in 2020-2021, 12 villages (50%) could adjust so that their AMPI index values were positive. These conditions continue, so all villages showed positive resilience in 2021-2022.

Based on the trend of resistance variation during 2019-2020; 2020-2021; 2021-2022, as many as 50% of the total villages (12 tourist villages) are included in the turnaround category ( - - + ), meaning that at the beginning and during the pandemic they were not able to survive, but then recovered after the pandemic ended. Meanwhile, 12 other villages were shaken by the pandemic in the initial period (2019-2020) and soon recovered in the following period (2020-2021; 2021-2022), so they are classified as villages with a positive jolt typology ( - + + ). If seen per region, the typology trend of the tourist village typology is more diverse (Table 11). In Central Java, five villages (50% of the tourist villages analyzed lead to a turnaround typology, while five other villages are on a positive jolt typology. In East Java, all observed villages (2 villages) lead to a turnaround typology. In Yogyakarta (DIY), conditions are more diverse; one village (16.67%) leads to a systemic declining typology, three villages (50%) lead to a turnaround typology, and three villages (50%) lead to a positive jolt typology. Meanwhile, in West Java, there is one village (50%) towards a turnaround, and one village (50%) toward a positive jolt. In West Nusa Tenggara (NTB), one village (25%) leads to a turnaround typology, and three villages (75%) lead to a positive jolt.

An analysis per indicator is carried out to find out the indicators that determine the resilience of tourist villages. This analysis will find out how different indicators different effects the resilience of tourist villages. The analysis uses the data in Tables 3-6, and based on the resilience trend scheme in Table 7, the analysis results are shown in Table 12 and Table 13.

Table 12 shows that if the tourist village resilience typology is based on indicators of the number of visitors, then there are seventeen villages (70.8%) leading to the standard resilience typology ( + - + ), and seven villages (29.2%) leading to the resilience typology ( + + + ). If the resilience typology is based on total income indicators, there are three villages (16%) lead to the standard resilience typology ( + - + ), sixteen tourist villages (66.6%) lead to turnaround typology ( - - + ), three villages village (12.5%) leads to a typology of resistance ( + + + ), one tourist village (4.16%) leads to a typology of systemic decline ( - - - ), and one village (4.16%) leads to positive jolt typology ( - + + ). Furthermore, if resilience is based on cost indicators, there are eight tourist villages (33.3%) that lead to a turnaround typology ( - - + ), fifteen villages (62.5%) lead to a resistance typology ( ++ + ), one village (4.16%) leads to the standard resilience typology, and one village (4.16%) leads to the positive jolt typology ( - + + ). This analysis shows that in the dimensions of resilience performance, the indicator that has the greatest influence on the typology of resilience in tourist villages is the number of visitors. Meanwhile, cost is an indicator that has the least influence on the resilience of a tourist village.

Table 13 presents a typology of tourist village resilience based on capacity dimensions. If the resilience typology is based on the number of employee indicators, there are five tourism villages (20.8%) that lead to a turnaround typology ( - - + ), fourteen villages (58.3%) lead to a resistance typology ( + + + ), three villages ( 12.5%) leads to a standard resilience typology ( + - + ), two villages (0.08%) leads to a positive jolt typology ( - + + ). If resilience is based on the Development Village Index (DVI) indicator, there are fiveteen villages (62,5%) leading to a systemic declining typology ( - - - ). four villages (16,6%) leading to a counter-cyclical typology, (

- + - ), three tourist villages (12.5%) lead to a

typology of lagged shocks ( + + - ), two tourist villages (0.08%) lead to a typology of turnaround ( - - + ). Furthermore, if the resilience typology is based on capacity-building indicators, eleven villages (45.837%) lead to a turnaround typology ( - - + ), five tourist villages (16.6%) lead to a standard resilience typology ( + - + ), nine tourist villages (16.6%) leads to the resistance typology ( + + + ). This analysis shows that in the dimensions of resilience capacity, Development Village Index (DVI) indicator has the greatest influence on the typology of resilience in tourist villages.