Quality of Life Determinants in Spain’s Smart Rural Areas During the Pandemic: A Better Alternative to Urban Living

Abstract

In the context of widespread rural depopulation, recent global developments have sparked shifts in lifestyle preferences, underscoring the often-overlooked benefits of rural living. These changes have prompted many people to consider relocating to rural areas for improved quality of life and more tranquility. This study explores the differences in quality of life between rural and urban environments, analyzing key factors that impact individuals’ perceived well-being within a broader societal context. Focusing on Spain, our research aims to identify the diverse elements influencing social happiness. Using data from a national survey on quality of life for 908 participants, we compared the perspectives of rural and urban residents during the pandemic, revealing a clear attraction effect toward rural areas. Our findings reveal a significant trend toward relocating to rural areas, but only those with good connectivity (smart rural areas). To effectively address rural depopulation and promote these regions, it is crucial to invest in comprehensive connectivity enhancements, including technological advancements, improved infrastructure, and wider availability of services. Methodologically, a comparative analysis approach was applied, using an analysis of variance (ANOVA) to identify differences and regression techniques to determine the influence of various factors on habitat happiness, understood as quality of life in social terms.

1. Introduction

In the early 19th century, less than two percent of the world’s population lived in cities. By 1900, that figure had increased to five percent, further rising to thirty percent by 1950. At the beginning of the 21st century, in 2007, more than half of the world’s population resided in urban environments. According to the United Nations [1], there are more than 512 cities with a population of more than one million, 45 of which exceed five million inhabitants, and 31 exceed ten million inhabitants (megacities); furthermore, it is predicted that two out of three people will live in urban areas by 2050. In other words, we see a trend of continuing emergence and growth of megacities, with urban spaces set to be home to 75 percent of the world’s population. However, none of the existing megacities score highly on quality-of-life criteria.

This ongoing increase in urbanization raises the question of whether it can be associated with greater well-being as well as with higher salaries and more employment opportunities. In this respect, it is essential to differentiate between what is considered rural and urban; however, there is no consensus in the literature, which may be an obstacle to validating results. Rural areas vary in characteristics such as population density and size, economic factors and distance from urban centers [2,3]. Wagenfeld [4] suggests that isolation and low population density are the most common defining features of rural areas, but he also argues that any such narrow definition ignores other important variables. The definition of urban is also unclear, as cities are heterogeneous, constantly changing realities [5]. Some researchers argue that rural and urban should be treated as points on a continuum rather than dichotomous categories [6], although most consider population size to be an important factor [7]. In this vein, Gross-Manos and Shimoni [8] define three types of areas for Israel—center, urban periphery and rural periphery—using the peripherality index developed by the Israel Central Bureau of Statistics. Similarly, González-Carrasco et al. [9] define an additional category for Spain, semi-urban, which they apply to medium-sized settlements of 2000 to 10,000 inhabitants. Sørensen [10] distinguishes three types of areas according to population level: rural areas, which are towns or villages with fewer than 5001 inhabitants; urban areas, defined as towns between 5001 and 100,000 inhabitants; and urban areas, defined as cities with more than 100,000 inhabitants. Thus, in light of this literature review and the fact that Eurostat guidelines [11] consider an urban cluster to have a minimum population of 5000, for this study, we define rural areas as those with fewer than 5000 inhabitants and urban areas as those with more than 5000 inhabitants, with the latter group including the urban areas and cities as defined by Sørensen [10].

Having quantitatively defined these areas, we turn to people’s choice of residence and, therefore, the possibility of mobility, given the trend for moving to urban spaces. This requires a comprehensive analysis of all studies on residential mobility, counter-urbanization, and naturbanization, which extends beyond the scope of this work. Thus, it is necessary to address, among other aspects, how rural areas can benefit from the dispersion of work from urban centers, as well as from housing prices [12], cultural clashes [13], new residential preferences [14], environmental concerns [15] and the attractiveness of natural protected areas (NPAs) as part of the urban deconcentration process [16,17]. Therefore, we are going to focus on the fact that choosing between an urban or rural area to live in offers advantages and disadvantages in terms of residents’ quality of life.

The choice between living in an urban or rural area offers advantages and disadvantages in terms of residents’ quality of life. Some studies highlight the positive aspects of rural living, including less pollution, closer connections and relationships with family and neighbors, and lower stress levels [10,18,19]. On the other hand, among the disadvantages, research remarks on lower income levels, lower employment rates, and more limited public healthcare, education, welfare, and other services [4,20,21]. At the other end of the scale, many analysts agree that urban areas offer better material conditions, higher employment rates and income levels, and more educational opportunities [22,23]. Drawbacks include pollution, congestion, feelings of alienation, fewer social relationships [6,24], increased risk of developing mental health problems [24,25] and higher stress levels [26]. It follows that a key aspect in the choice of one area or another depends on the perceived quality of life and the determinants of this quality of life. In this regard, [27] found that rates of unhappiness with quality of life increase significantly when the size of a city exceeds hundreds of thousands of people.

Based on the above, this study examines three key hypotheses:

Hypothesis 1 (H1).

There are significant differences in the levels of quality of life between rural and urban residents.

Hypothesis 2 (H2).

The determinants of quality of life in both rural and urban environments have evolved due to socio-economic and structural changes.

Hypothesis 3 (H3).

The COVID-19 pandemic had a differential impact on quality-of-life perceptions depending on place of residence. The lockdown context facilitated migration to rural areas, attracted by better living conditions during the pandemic.

Specifically, we explore how, during the early waves of the pandemic (2020–2021), many urban residents in Spain relocated to rural areas despite regional lockdown restrictions. In many cases, this was facilitated by the possession of a second home in a rural setting and/or the increased flexibility of remote work. As suggested [28,29], these net migration losses in cities and gains in rural areas in Spain were particularly pronounced after the removal of strict lockdown measures. At the end of these periods, it appears that these changes have been temporary and have not significantly altered the existing structures of the national migration system. This shift suggests that rural areas are perceived as safer, and policies facilitating tele-commuting could represent a potential solution to the demographic decline in these areas—provided that they are accompanied by structural transformations [28,29]. Consequently, the objective of this study is twofold: first, to analyze the main factors influencing the quality of life in both rural and urban settings, and second, to determine which types of rural areas are most attractive for this type of migratory movement, as they are likely to be the primary beneficiaries of population redistribution.

In this context, a key issue to consider is the social structure that affects the quality of life in rural areas, such as the rural-urban income gap, labor market dynamics, limited structural and technological change in rural areas, the welfare state, and gender bias in the rural world, together with the limited effectiveness of the policies in place. However, during the pandemic, residential shifts were driven not only by the search for safety but also by changes in labor dynamics, such as the ability to work remotely.

To conduct this study, we rely on a survey on the quality of life perceived by Spanish residents of rural and urban areas during the pandemic, which provided a total of 908 valid responses. Based on this survey, we can identify the key determinants of quality of life for each of the area types and consequently recommend actions and policies to encourage mobility toward rural areas, thereby providing potential solutions to the issue of rural depopulation affecting all developed countries. Thus, in Section 2, we review the literature that addresses differences in the quality of life perceived by citizens in rural and urban areas, as well as the aspects that shape these perceptions. Section 3 presents the database we use to empirically analyze the situation in Spain, while Section 4 outlines the main results. Finally, Section 5 provides the main conclusions and recommendations for enhancing life in rural areas.

2. A Review of the Literature on Quality of Life: Rural vs. Urban

The concept of quality of life has been the subject of numerous studies, but the literature has not reached a consensus on a definition or how to measure it. The main reason for this is its multidimensional and interdisciplinary nature, encompassing fields such as economics, sociology, politics, medicine, and demography [29]. This complexity results in a diversity of perspectives on the concept of quality of life. In many cases, it is used to assess the overall well-being of individuals or societies, while in others, it is linked to more abstract notions such as freedom, human rights, and happiness.

Quality of life is commonly described as having two interrelated dimensions: the objective dimension (including physical, social, and economic aspects) and the subjective dimension (psychological well-being and satisfaction). Consequently, measuring quality of life is not a straightforward task, as there are no widely accepted objective indicators that allow for direct comparisons between countries, regions, or cities.

In an effort to capture both dimensions, various indices have been developed by authors such as Marans and Cooper [30] and Turkoglu [31]. Similarly, recognizing the mul-tidimensional nature of quality of life, El Din et al. [32] distinguish seven key dimensions of urban quality of life: environmental, physical (infrastructure), mobility (accessibility), social (citizen integration and participation), psychological (perceptions and feelings), economic (employment and business opportunities), and political (urban governance and legislation).

Researchers have proposed multiple methodologies to measure quality of life at different territorial levels (national, regional, and urban), leading to diverse findings. However, there is no consensus on a universally applicable index or standardized dimensions. Most studies adopt a subjective rather than an objective approach and collect data through surveys—as we do in this study. Consequently, the approach we take is multidimensional, encompassing aspects of well-being, satisfaction, and happiness, enabling a comprehensive and subjective measurement of quality of life.

Different research disciplines in economics examine the differences between urban and rural quality of life. For example, in urban economics, Tassinari et al. [33] report that life satisfaction in cities is slightly higher than in rural areas, although this difference decreases with the level of development and disappears in high-income countries. These differences are caused by gaps in health problems, physical infrastructure, expectations about future economic conditions, and housing affordability. In addition, the authors find that life satisfaction varies according to the size of the city, with large cities defined as those with between one and five million inhabitants. Moreover, regional studies on quality of life play a significant role in rural development, particularly in understanding rural-urban interactions and interrelations [34], advancing theoretical debates and interdisciplinary approaches to rural-urban differentiation [35], and informing strategies to bridge existing disparities [36].

Depopulation can be seen as a factor that shapes the quality of life of rural residents, potentially driving migration to urban areas. At the same time, rural communities are increasingly eager to connect, engage in modern cultural life, and become more efficient or “smart”. In this sense, the rural paradigm is changing, as well as with its patterns [37]. In this regard, a key aspect to analyze when comparing rural and urban areas is the perceived quality of life in the different area types and the factors that influence this perception. Furthermore, in line with our objective of assessing the effects of the pandemic on urban-to-rural residential shifts in Spain, the literature review includes a specific analysis of studies focused on the pandemic’s impact on quality of life.

Therefore, bearing in mind that quality of life is a multidimensional concept encompassing socio-economic, cultural, and environmental aspects unique to each territory (H1), it is essential to identify the components of quality of life and the factors that influence it in urban and rural contexts. Tools for assessing the components of well-being at the territorial level could help to better define development policies; however, the effectiveness of those policies is still conditioned by the capacity to involve key actors from the affected communities in decisions about policy in terms of representation, knowledge and power [38]. Thus, rural and regional policy must evolve to address the complexities of the current reality, partially redefining territorial or place-based approaches and the New Rural Paradigm in line with the principles of “Rural Cohesion Policy” [37].

Different studies on well-being reveal that an important aspect of choosing a rural location is the quality and accessibility of services, such as health care, education, transport, shops, banks and cinemas [39]. In this regard, Casini et al. [40] and Peel et al. [41] conclude that there is dissatisfaction with well-being in rural and marginal environments. In other words, well-being is closely linked to the development of a given area. Development, in this sense, involves identifying and making use of a territory’s own resources and potential, both economic and non-economic. Therefore, the decision to leave a particular place is not driven by economic factors alone. Therefore, evaluations of well-being should not focus exclusively on the specific material features of the analyzed territories since the same availability of resources can be assessed differently depending on how individuals perceive the value of these resources in relation to the contexts in which they live. Addressing the challenges of quality of life (objective and subjective dimensions) in rural areas requires analyzing residents’ perceptions and the factors influencing them, which could contribute to mitigating depopulation.

As discussed above, the level of urbanization is increasing, but it would be useful to know whether this phenomenon is driven by an increase in employment opportunities and wages in cities, which, in turn, leads to higher well-being. Conversely, what is known as the urban paradox [42,43] may arise; that is, if incomes rise and technology advances even further, while transport and digital infrastructure also improve, rural areas become more accessible and diversified, leading to a reduction in urban-rural living condition gaps to the point where average happiness levels in rural areas, villages and small towns approach and even exceed those of large cities. The digital transformation of rural regions depends on targeted policies, strategic infrastructure investments, and ongoing education initiatives [44].

In short, our quality of life is conditioned by our living conditions [6]; that is, the context of a rural or urban environment affects a person’s subjective determinants of quality of life well-being (H2). Following Diener [45] and Diener et al. [46], this subjective well-being can be measured by emotional reactions to events, moods and judgments about fulfillment, specific life domains and general life satisfaction. Ultimately, as Easterlin [47] suggests, urbanization leads to economic growth, but such prosperity is not necessarily accompanied by an increase in happiness, especially in more developed countries.

Furthermore, the literature explores urban-rural disparities in well-being by territory, distinguishing between developed and developing countries and regions. That is, various studies reveal a strong link between differences in quality of life and the level of development in each country. In less developed countries, people reported being happier in urban areas, whereas in more developed economies, rural areas were equal to or even surpassed urban areas in life satisfaction [6,22,48,49]; for example, in the EU [10,50], in Scotland [51,52], in Australia [53] and in the United States [48,54]. Other studies have also shown that residents of the major cities of developed countries have lower subjective well-being than those living elsewhere; for example, in Stockholm [55], Dublin [56], New York [27,57], London [58], Sydney [43], Helsinki [59], Bucharest [50] and Toronto [60]. Along the same lines, studies have been carried out for specific regions. For example, in the case of Spanish regions [61], it has been shown that social happiness decreases as per capita income increases and that prosperity (social well-being, as an assessment of an individual’s quality of life in his or her social context, in terms of [62] can be achieved without depending exclusively on material growth. In contrast, there are fewer studies that examine the possible reasons for this situation. While various socio-economic, environmental and psychological factors are often considered, they do not fully explain it [10,27,43,63]. Moreover, there have been calls for eudaimonic well-being to be accounted for in studies [64].

Finally, we examine the contributions of various scholars on the impact of the pandemic on quality of life (H3). In this regard, consideration should be given to different schools of thought on how to manage the health crisis, such as interventionists (pro-state control) vs. liberals (pro-civil society action) [63]. Moving on from the pandemic, authors such as Qian et al. [65] present an evaluation framework for global smart cities with the aim of promoting sustainable development in the post-COVID era. Conversely, Morales et al. [66] pinpoint the pandemic as the turning point for a return to the countryside, with technology being the driving force of the population, remote work flowing from the big cities, and the green infrastructure of the rural environment serving as a lure for connected human resources. However, the geographer Sexto [67] warns of an opportunistic mercantilism of the idealized space, which he calls the “virtual village”. In that space, the post-COVID society is headed up by affluent families of urban origin, fulfilling their green desires and developing in touch with nature. In his analysis, the author makes a veiled criticism of the spatial interests of the real estate market. Focusing on Italy, Savini et al. [68] find that the highest values of infection to COVID-19 are associated with areas with commercial centers, near the most populated cities, and in industrial areas. Accordingly, this factor may prompt a return to rural areas, although perhaps only temporarily.

Based on citizens’ objective assessments, our analysis involves measuring the factors that influence their quality of life (mainly living conditions, working conditions, residential community, mobility, accessibility, safety, health and education) while observing the differences between the rural and urban worlds. The results can inform policies that seek to prevent or reverse rural depopulation. Thus, our aim is to gain a better understanding of the influence of some relevant determinants of the quality of life of Spanish citizens, as individuals generally tend to actively select their place of residence, considering job opportunities and the availability of public goods and services [69]. People’s choice of where to live is associated with their social and economic prospects [70] and thus with their pursuit of social quality of life [71]. The success or failure of municipalities to provide opportunities for residents to achieve a good quality of life determines their ability not only to attract new residents but also to encourage current residents to stay. Thus, to help prevent rural depopulation, our study aims to offer an understanding of the degree of influence of each of the well-being factors.

3. Materials and Methods

This study relies on a database made up of a total of 908 responses obtained from a questionnaire administered via email and social networks, targeted at citizens from all over Spain. The survey was anonymous in order to encourage objective responses. A random sampling technique was used, although it should be acknowledged that the sample pertains to a population with above-average digital capabilities, considering the data collection process. The survey had a margin of error of 3.25% with a confidence level of 95%. This refers to the statistical bias that must be considered when extrapolating information from a sample to a population. However, we also considered the significance criteria of the analyzed groups (rural and urban) to ensure consistent results. Despite the rigorous approach taken, potential sample bias may arise in terms of gender, with a higher representation of female respondents. However, this bias is not directly related to the objective of the analysis. A multidimensional approach was used to assess quality of life, in line with UN happiness surveys [72].

The questionnaire was first tested in a pilot exercise in 2020. Then, the information was collected between 15 February and 15 April 2021. The fact that the data collection period corresponds to a time when the COVID-19 pandemic was still active should be borne in mind, as it may have influenced some of the results obtained.

The questionnaire is made up of three blocks. The first captures socio-demographic variables, including the classification of respondents’ place of residence, allowing us to differentiate between rural and urban areas (information is also obtained on gender, marital status, age, education, employment status, economic sector, income, and region of residence). The second block corresponds to variables related to the citizens’ assessment of the quality of life: respondents are asked to make one direct assessment of their quality of life and another assessment using 22 variables related to key aspects of quality of life, formulated as questions about the living environment (family situation, trust, environment, commercial accessibility and public transport, green areas, culture and sport, foreign population, health, education, housing prices and safety), and about respondents’ working and educational environments (economic situation, employment status, working environment, training, access to the internet). The third block includes two questions related to the impact of COVID-19: one assessing its direct effect on quality of life and another identifying the area of life most affected by the pandemic. The questions in blocks 2 and 3 use a 10-point Likert-type scale (ranging from 1, “not at all satisfied”, to 10, “very satisfied”), except for the item on the area of life most affected by COVID-19, which offers the following options: personal and family life; work and the economy; education and training; residence; and “I have not been affected”.

In order to differentiate between people living in rural and urban areas, we used a question about the size of the municipality, with municipalities of up to 5000 inhabitants considered part of the rural world—a total of 167 (18.4%) of the responses in the sample—while the rest are urban, that is, those living in municipalities with more than 5000 inhabitants—in this case, 741 (81.6%) of the total number of responses. These figures are representative if we consider the population data for Spain (

Table 1. Distribution of the Spanish population by municipality size (number of inhabitants).

Municipality Size	2000	2010	2020	2021
Total (Spain)	40,499,791	47,021,031	47,450,795	47,385,107
Fewer than 101 inhabitants	57,677	62,186	79,850	78,172
From 101 to 500	726,261	679,957	650,018	651,192
From 501 to 1000	818,273	756,402	718,689	721,010
From 1001 to 2000	1,450,856	1,313,221	1,227,235	1,240,089
From 2001 to 5000	3,134,288	3,178,057	3,014,825	2,996,629
Rural	6,187,355	5,989,823	5,690,617	5,687,092
From 5001 to 10,000	3,515,907	3,933,865	3,844,677	3,902,355
From 10,001 to 20,000	4,622,754	5,020,865	4,827,217	4,894,070
From 20,001 to 50,000	5,609,225	7,435,693	7,858,627	7,837,738
From 50,001 to 100,000	4,241,165	5,915,160	6,189,980	6,178,689
From 100,001 to 500,000	9,368,333	11,076,146	11,288,818	11,216,339
More than 500,000	6,955,052	7,649,479	7,750,859	7,668,824
Urban	34,312,436	41,031,208	41,760,178	41,698,015

Source: National Statistics Institute (2002) [73].

), whose rural population figures in this century range from 16 to 12%. Furthermore, we can observe that the depopulation trend stabilized in 2021, coinciding with the pandemic. In this analysis, we aim to measure the factors that influenced the slowdown in the trend of rural depopulation.

For the sake of transparency regarding the data collection process,

Table 2. Sample structure.

		Area
Variable	Categories	Rural	Urban	Total
Civil Status	Single	37.1%	24.8%	27.1%
	Married or in a stable relationship	56.3%	67.1%	65.1%
	Divorced	4.8%	7.3%	6.8%
	Widowed	1.8%	0.8%	1.0%
Gender	Male	34.1%	39.8%	38.8%
	Female	65.9%	60.2%	61.2%
Age	18 and 19 years old	3.6%	1.6%	2.0%
	20–29 years old	24.0%	21.9%	22.2%
	30–39 years old	15.0%	11.3%	12.0%
	40–49 years old	26.9%	31.6%	30.7%
	50–59 years old	18.6%	24.4%	23.3%
	60–69 years old	9.6%	7.7%	8.0%
	Over 70 years old	2.4%	1.5%	1.7%
Economic sector	Business	3.6%	6.1%	5.6%
	Construction	2.4%	2.0%	2.1%
	Education	24.6%	26.0%	25.8%
	Unemployed or inactive	31.7%	19.6%	21.8%
	Energy, water	1.8%	1.8%	1.8%
	Finance and insurance	4.8%	7.0%	6.6%
	Manufacture	1.2%	1.1%	1.1%
	Other services (communication, information…)	13.8%	23.2%	21.5%
	Agriculture, animal husbandry, fishing (Primary)	3.6%	1.3%	1.8%
	Healthcare	7.8%	7.6%	7.6%
	Transport and storage	3.0%	1.8%	2.0%
	Tourism, catering, hospitality	1.8%	2.6%	2.4%

Source: Own Elaboration.

shows the classifications used and their percentages in the sample. Other demographic factors of the survey worth noting are that 38.8% of the respondents were men and that the most predominant age group is between 40 and 49 years old, representing 30.73% of the sample. The social and productive structures in urban and rural areas are quite similar in the study (Table 2). However, some differences in the productive sector (unemployed or inactive) reflect the reality of the population.

The next step is to measure residents’ quality of life using a multidimensional approach (residence, family and work) based on the variables reported in the survey. Regarding the informational variables, our analysis focuses on the Residential Satisfaction (RS) variable along with the final social Happiness (HP) variable. As we explain later, these two variables allow us to define a proxy variable that we label Habitat Happiness (HH). This variable (HH) approximates a measure of the quality of life in the citizen’s place of residence. The RS variable is an initial approach to citizens’ perception of quality of life, for which respondents are asked at the beginning of the survey about their level of satisfaction with their life in their place of residence. On the other hand, the HP variable is captured by a question included at the end of the questionnaire, asking respondents to indicate their degree of happiness after having answered all the other questions (by dimensions). It thus focuses on social happiness, considering all the variables or dimensions evaluated within the residential, family, and work spheres.

In addition, we analyze the influence on HH of factors considered relevant in the quality-of-life literature: family situation and home environment (FAM); neighborhood trust (NTR); polluting factors—pollution, noise, waste (PNW); public transport, mobility, commercial accessibility (TMC); sports and cultural facilities (S and C); individual and home safety (SAF); economic capacity and incomes (ECC); and happiness at work (HaW). We also analyze COVID-19 as an external factor that is assumed to have a negative influence on quality of life and which was prevalent at the time the information was obtained.

Based on the nature and the availability of the information, we opted for an Analysis of Variance (ANOVA) and a regression analysis. The ANOVA allows for the evaluation of differences in average values with respect to a single factor: in our case, rural vs. urban areas. Specifically, the ANOVA tests the null hypothesis that the group averages are equal. It is, therefore, the statistical tool used to test H1, which postulates the existence of significant differences in quality of life in rural and urban settings. The main limitations of ANOVA are related to the assumptions about the data, specifically normality, homoscedasticity, and independence. Normality is not a particularly important issue when handling large datasets, as in our case. Homoscedasticity is tested using the Levene statistic; if the assumption of homogeneity of variance is accepted, the F-statistic is used to perform the ANOVA. However, if the assumption is rejected, alternative methods, such as the Welch test, should be applied. The Levene statistic is defined as:

$$W={\displaystyle \frac{\left(N-k\right)}{\left(k-1\right)}}{\displaystyle \frac{\sum _{i=1}^k{N}_i{\left(Z_i.-Z..\right)}^2}{\sum _{i=1}^k\sum _{j=1}^{N_i}{\left(Z_{ij}-Z_i.\right)}^2}}$$

(1)

where k is the number of different groups, N is the total number of cases, N_i is the number of cases in the i-th group, and Z_ij is defined as:

$$Z_{ij}=\left|Y_{ij}-{\overline{Y}}_i.\right|$$

(2)

with Y_ij being the value of the measured variable for the j-th case from the i-th group and ${\overline{Y}}_i.$ being the average of the i-th group. Moreover, using Z_ij we can calculate Z_i. as the group average and Z.. as the overall average.

Finally, the two subsamples are, by definition, independent of one another. Moreover, ANOVA only indicates a difference between groups, not which specific groups are different—a multiple comparison test would need to be used to distinguish specific groups.

To test H2, we use regression analysis to examine the influence of various factors on the perceived quality of life of citizens in rural and urban areas, with the aim of identifying the significant factors in each case. This analysis will help us determine the aspects that require special attention to improve citizens’ perception of quality of life in rural areas. Since the variables are measured on a similar scale, linear correlation provides an efficient tool for identifying potential determinants, provided that the endogeneity condition is well defined, as is the case with the HH variable, and that sufficient information is available on the inputs (up to 22 variables). The main issues associated with the use of linear correlation analysis for this type of sample are related to the homogeneity of the sample and its representativeness of the population, a factor that has been proven to be efficient.

The model to be estimated links the k explanatory factors (f) determining HH for i citizens. The mathematical relationship follows a linear formulation, as shown by López et al. [74] in their study of citizen happiness, according to the following equation where ɛ is a random variable with an expected value of zero, uncorrelated and with constant variance.

$${\displaystyle \frac{\left({RS}_i+{HP}_i\right)}{2}}={HH}_i=\alpha +{\sum }_{s=1}^k{\gamma }_s{f}_{is}+{\epsilon }_i$$

(3)

If we run the model [3] using ordinary least squares for both the rural and urban samples, we will be able to compare the factors that determine the explained variable (HH).

To conduct these analyses, we use the statistical software R version 4.2.3 (R Core Team, 2020) for the ANOVA and EViews 12 for the regression analysis. The main results are presented in the following section.

4. Results

In this section, we present the results of the analyses outlined above, aimed at examining the reasons explaining the residential quality of life during the pandemic period, in which the trend of rural depopulation was halted. The first objective of the analysis is to determine whether there are significant differences in the mean values of rural and urban citizens’ perceptions of quality of life assessed at the beginning and at the end of the survey and to investigate the different factors that are thought to play a key role in these perceptions.

The comparative analysis of rural and urban areas is based on the endogenous variable capturing respondents’ average satisfaction with life in their place of residence (RS) and their residential life satisfaction after assessing the main factors involved in its measurement (HP). The result is an objective, unobservable proxy variable—Habitat Happiness (HH)—derived from the univariate measure of RS combined with the multivariate HP measure, which, in turn, reflects citizens’ assessments of the quality of life in relation to factors within the home, family, and work environments.

On the other hand, the factors included in the model as significant exogenous variables—and therefore as explanatory factors (f) for HH—are, as mentioned above, family and home situation (FAM), neighborhood trust (NTR); polluting factors —pollution, noise, and waste—(PNW); public transport, mobility, and commercial accessibility (TMC); sports and cultural facilities (S and C); individual and home safety (SAF); economic capacity and income (ECC); and happiness at work (HaW); with COVID-19 as a negative external factor affecting quality of life.

For the initial analysis of key factors and differences, we conducted a one-factor ANOVA (rural-urban). As previously explained, we assessed homogeneity of variance using Levene’s test, which determines the appropriate statistic for the comparative analysis of mean values in citizens’ assessments of various variables across rural and urban areas. The results of Levene’s test are presented in

Table 3. Levene’s test for homogeneity of variance.

Variable	Levene Statistic	df2	df2	Sig.
FAM	0.120	1	906	0.729
NTR	6.440	1	906	0.011
PNW	16.898	1	906	0.000
TMC	24.121	1	906	0.000
S and C	5.323	1	906	0.021
SAF	3.255	1	906	0.072
ECC	1.303	1	906	0.254
HaW	0.977	1	906	0.323
COVID-19	2.711	1	906	0.100
RS	0.138	1	906	0.710
HP	0.268	1	906	0.605
HH	0.003	1	906	0.957

Source: Own Elaboration.

.

The results show homogeneity of variance for some variables, indicating the use of the F statistic. However, whenever variances cannot be considered homogeneous, that is, the significance levels are below 0.05, we will use Welch’s test (W) to compare means. The values of these statistics, together with their significance, are shown in

Table 4. One-factor analysis of variance.

Factor	df1	df2	Statistic	Sig.
FAM	1	906	0.207 (F)	0.649
NTR	1	226.430	3.247 (W)	0.073 *
PNW	1	296.907	80.591 (W)	0.000 **
TMC	1	211.964	52.596 (W)	0.000 **
S and C	1	229.641	13.724 (W)	0.000 **
SAF	1	906	3.054 (F)	0.081 *
ECC	1	906	0.032 (F)	0.858
HaW	1	906	1.069 (F)	0.301
COVID-19	1	906	1.805 (F)	0.179
RS	1	906	0.165 (F)	0.684
HP	1	906	1.019 (F)	0.313
HH	1	906	0.593(F)	0.441

Source: Own Elaboration. * are significant at the 10% level; ** are significant at the 5% level.

.

The results in Table 4 indicate that there are no significant differences between rural and urban areas for any of the quality of life variables; that is, the initial assessment, the final assessment, and the proxy variable defined for use as the endogenous variable in the regression model. However, the same cannot be said for the explanatory factors, as significant differences at the 5% level are observed in the mean assessment values between rural and urban areas for the variables PNW, TMC, and S and C, along with differences significant at 7.3% for NTR and 8.1% for SAF. For the remaining variables, differences are not statistically significant, although

Table 5. Mean values: rural vs urban (scale 1 to 10).

	FAM	NTR	PNW	TMC	S and C	SAF	ECC	HaW	COVID-19	RS	HP	HH
Rural	7.99	6.88	8.21	5.89	6.58	8.17	7.28	7.34	7.75	7.72	7.28	7.50
Urban	8.06	7.22	6.80	7.39	7.26	7.93	7.25	7.14	8.01	7.79	7.42	7.60
Total	8.05	7.16	7.06	7.11	7.14	7.97	7.26	7.18	7.96	7.77	7.39	7.58

Source: Own Elaboration.

presents the mean values recorded for all variables.

Table 5 shows that among the variables with significant differences in mean values, PNW and SAF show higher values in rural areas, whereas NTR, TMC, and S and C display higher values in urban areas. In line with previous findings [24,25], the results regarding the roles of rural and urban areas in determining quality of life—and, by extension, the reasons for residing in a given area—show that the rural model offers advantages in factors such as safety and sustainability, whereas the urban model provides better conditions related to public transportation, accessibility to services, sports activities, and cultural offerings.

In the case of COVID-19, although the difference is not significant, it is found to negatively influence the quality of life and the variables measuring the quality of life to a greater extent in urban areas. Moreover, the dynamic element of this analysis was carried out in 2020. Despite the importance of family (FAM) and economic/labor-related (ECC, HaW) issues as factors influencing quality of life, they are not sensitive to distinctions between rural and urban areas.

Following the ANOVA, we construct the linear model (Equation (3)) that measures the direct relationships between HH and each factor. The results show notable differences between the rural and urban settings.

Table 6. Urban and rural Models in HH.

Variable/Factor	Urban		Rural
	Coefficient	Elasticity	Coefficient	Elasticity
Independent	1.038096 (4.35) ***	0.136582	0.925092 (2.15) **	0.123296
FAM	0.374491 (17.67) ***	0.397298	0.332747 (9.07) ***	0.354258
NTR	0.051676 (2.47) **	0.049070
PNW	0.037817 (1.98) **	0.033821	0.098893 (2.02) **	0.108206
TMC	0.055296 (2.57) ***	0.053754	0.122976 (3.95) ***	0.096477
S and C	0.101279 (4.69) ***	0.096784	0.080900 (2.18) **	0.070957
SAF	0.056420 (2.06) **	0.058844	0.119895 (2.34) **	0.130517
ECC	0.058759 (2.52) **	0.056057
HaW	0.125334 (6.16) ***	0.117790	0.118850 (3.65) ***	0.116289
R2	0.5798		0.6399
n	741		167

Source: Own Elaboration. ** are significant at the 5% level and *** are significant at the 1% level.

presents the γ coefficients along with their significance and elasticity values for ease of comparison. The elasticity values provide the proportional weight of each factor in the positive relationship with HH. The key factors with the highest value for both scenarios are marked in bold. The goodness-of-fit is adequate, as shown by the coefficient of determination (R²). In this regard, the rural model has a higher R², indicating greater socialization among these residents; that is, the social factors analyzed contribute more to their social happiness, with individual factors accounting for a maximum of 36% of the variance (1 – R²) compared to 42% in the urban model, in line with the finding reported by López et al. [74].

The regression results highlight two key issues. First, quality of life is more dependent on social conditions analyzed in rural areas. Specifically, they represent almost 64% of the total, compared to 58% in urban areas (see R²). Second, different factors determine the quality of life in the two area types. Notably, economic conditions (ECC) are a crucial determinant of quality of life in urban areas, whereas sustainability-related factors (PNW, SAF) play a more significant role in rural settings. The higher values for these factors (elasticity in Table 6) imply that for residents of rural areas, both home and personal safety, as well as low pollution and environmental conditions, are key factors in their quality of life. Additionally, accessibility and communication (TMC) emerge as critical aspects in rural areas, which indicates that they are a particularly pressing need. While the average value for TMC is clearly lower in rural areas compared to urban ones, its weighting (elasticity) is significantly higher.

5. Discussion

To understand the differences in the determinants of social happiness and the way these factors interact in urban and rural settings, it is opportune to examine behavioral patterns during the critical period of the pandemic, as this was a time of significant population movement from cities to rural areas in Spain. The temporary slowdown in rural depopulation trends in Spain provides a relevant and opportune context for this analysis. For this purpose, we rely on the information provided by residents and on previous research, which allows us to discuss the results within this framework.

Urban residents clearly differ regarding their preferences in two factors. First, the economic capacity (ECC) variable is a key factor when choosing their place of residence, whereas this factor is not essential for rural residents’ habitat happiness (HH). Second, trust in neighbors and the broader community (neighborhood) is essential for residents of Spanish cities, perhaps due to the spatial proximity that defines these urban environments. In contrast, rural residents view their sense of safety as extending to their entire environment rather than just their immediate surroundings (community). In other words, the size of the city means residents must break it down into smaller spaces (neighborhoods) and rely on trust among those who live nearby.

For the remaining determinants of Spanish citizens’ quality of life, it is useful to note both the similarities and differences between the two scenarios. Family situation and job satisfaction have a similar influence, as evidenced by the average values. However, regarding the living condition gap in relation to the analyzed determinants, aside from safety, two key needs emerge for rural residents. The first is environmental policy: issues related to pollution, waste management, and noise control are assigned a significantly higher priority by rural residents than urban ones. The second is connectivity and access to services, which many urban residents have and which rural residents see as essential to achieving quality of life. In cities, the availability of sports and cultural facilities is clearly greater than both the demand for and availability of such facilities in rural areas.

Therefore, urban and rural residents are similar in terms of the aspects they deem most influential for quality of life, such as family and work, which is in line with the quality of life model for developed countries [6,22]. However, some differences emerge in the analysis, which may currently encourage residency in cities but could also attract people to rural areas that follow the “smart rural area” model. This is corroborated by preferences observed during the pandemic in Spain, when the depopulation process slowed as people sought safety and a sustainable way of life while still prioritizing efficient physical and digital access to services and work that enhance quality of life. These results are very much in line with studies such as that by Leviäkangas et al. [44], who examine policies and best practices to stimulate a shift toward smart and digitalized rural regions and communities.

Based on the factors analyzed, it can be concluded that a typical resident in a Spanish city achieves quality of life primarily through their family situation, followed by factors related to job quality and economic status, then cultural and leisure activities, and finally safety, trust in the immediate environment (neighborhood), mobility and lastly, environmental policies. An interesting finding of this study is that these factors represent key reasons for residing in an urban environment in Spain and thus point to specific policy areas for action.

For the corresponding profile of a typical rural resident, in addition to the family situation (which constitutes the core of emotional support, stability and relationships that promote general and social well-being), factors such as safety, environmental policies and environmental quality, job quality, mobility, and connectivity to services and markets are clearly important. Therefore, active policies aimed at addressing rural depopulation must focus on the labor market, connectivity to services and green spaces, and improving the provision of cultural and sports facilities. According to these findings, related fiscal policies—besides challenges concerning their equitable application—are also open to criticism regarding their effectiveness. Tax cuts are not a sufficient motivation to attract people away from the city to the ideal green space [53]. Intervention in rural areas requires policies targeted at connectivity, digitalization, infrastructure, and housing that may not be economically feasible on the public budget nor easily implementable across all territories. This finding is also in line with other recent studies in rural areas [44] that address the specific challenges faced by rural communities, such as small markets, low population density, and long distances, highlighting the need for digitalization to improve the quality of life and sustainability in these areas. Similarly, the study by Tešin et al. [75] suggests promoting sustainable development initiatives (in line with the Sustainable Development Goals (SDGs) while strengthening the sense of place-dependence) and improving social cohesion since the economic, social and environmental dimensions of sustainability are essential to the quality of life in rural areas.

Transforming rural areas into “smart rural areas” capable of attracting inhabitants requires sustained intervention over time, combining public housing policies with direct digital access to employment and service hubs, as well as the continuation of remote work, particularly in service-oriented sectors. In line with other research, we can confirm that in Spain, the choice to reside in a rural area aligns with a model that prioritizes sustainable development as a guiding social system (PNW variable), although it must also incorporate accessibility, communication, digitalization, and transport efficiency (TMC variable). In line with Casini et al. [40] and Muhtar et al. [76], we label this type of habitat as “smart”, which encompasses rural areas with good access to services and facilities that support social development and safety (SAF variable), and technological efficiency enabling access to employment, commerce, or services.

Finally, regarding the COVID-19 effect, it turned out to be a temporary factor limited to the early stages of the pandemic; as it spread across the country, it gradually lost its significance as a factor supporting population retention in rural areas. However, we can still learn from the experience of the pandemic to identify the factors that drive such change, even if only temporarily. Moreover, several issues of interest arise from the sample profile that goes beyond the scope of our current analysis. For example, according to the responses obtained in this survey, we clearly observe that young people were less satisfied with their lives during the pandemic and were also more vulnerable, while those with employment showed higher levels of social happiness. These issues could counterbalance each other in rural and urban settings, as they move in opposite directions; the urban population is younger than the rural population, but it also has higher employment rates than the population in rural areas. In this respect, this analysis aligns with authors such as Shek [77], who examined how the COVID-19 pandemic affected various aspects of quality of life, reflecting on 12 key issues that emerged during the pandemic and that are crucial to better prepare for future pandemics. Furthermore, we should take into account the findings of other studies, such as that by Qian et al. [65], in which they explored how smart cities play a crucial role in promoting post-pandemic sustainable development, as their insights may also be applicable to rural areas. A comprehensive approach that includes appropriate policies, investment in technology and infrastructure, and active community participation is therefore essential.

6. Conclusions

The first conclusion of this study is that the social happiness derived from the factors considered in terms of quality of life in a Spanish person’s place of residence does not depend on whether the environment is urban or rural. However, we can confirm that the factors driving habitat happiness do indeed differ between rural and urban areas. This distinction allows us to identify the specific aspects that influence residents’ social happiness with their surroundings and leverage these to implement policies aimed at slowing the trend of migration from rural to urban areas. In turn, this will help mitigate the depopulation of geographical spaces, which poses challenges for sustainable development.

In this regard, it should be noted that the key factors influencing rural residents’ quality of life are environment-related, notably aspects such as green policies, pollution and contamination, and access to services. Thus, in line with Brereton et al. [54], we believe that regional policies focused solely on income measures are inadequate and that a more appropriate approach would incorporate additional quality-of-life components while aiming to ensure greater equality in economic opportunity. Specifically, regional policy should prioritize improving people’s quality of life through the provision of social and environmental services and by recognizing the importance of rural amenities; particularly essential in this respect is robust rural connectivity to attract residents.

In terms of quality of life, measures to encourage mobility in rural areas do not appear to rely on fiscal or financial incentives such as subsidies but rather on connectivity in all its forms. This approach extends beyond technology to include infrastructure and services provided by both the welfare state and the private sector. Consequently, the depopulation of rural areas located far from major population centers and lacking strong attractions in terms of green, cultural, or sports amenities appears difficult to reverse. Human beings tend to be happier in societies with larger populations, where they can meet their relational and employment needs. Furthermore, the fact that our society is highly service-oriented means the urban environment is one that continues to offer the best quality-of-life factors, particularly if urban management trends are moving toward sustainability, waste management, energy efficiency, security and the expansion of green, cultural and sports spaces.

In this scenario, the smart rural area model is viable for spaces that, benefiting from their geographical location and digital specialization, can implement public intervention strategies to attract residents. A good example of a Spanish region that is moving in the right direction is the central region of Castilla-La Mancha, which is predominantly rural but is continuing to experience demographic growth in smart areas near large cities and infrastructure.

We thus examine the nature of the depopulation problem and potential responses. In this regard, the policies implemented should be focused on the local context but embedded in wider spheres; that is, they should be comprehensive policies with civil society playing a leading role. Thus, public policies should be structured around the diversification of production, the closing of gaps (gender and access to services), housing, digital infrastructures, the role of immigration, environmental protection, and the strengthening of social capital. In short, they should facilitate development by improving the perception of quality of life in those environments.

Regarding the pandemic, rural areas with low population density became attractive during the first wave of the pandemic in 2020 due to the sense of security they offered. Subsequently, as the virus spread further, the failure of remote work as a viable alternative to office jobs meant that the temporary advantage of rural over urban spaces faded away.

It is difficult to imagine future generations returning to the rural areas and turning their backs on the array of benefits that cities have generated in 21st-century society, including the post-pandemic context. People’s needs are not solely economic in nature but rather are based on a sustainable growth model, where work is conducted in large cities with energy efficiency, technology and service connectivity. In other words, people are calling for smart cities, which will manage the main social spaces and offer a high quality of life.

Finally, this research presents some limitations, which will be addressed in future studies. They include the need to work with a much larger sample of citizens and to consider the difference in sample sizes between urban and rural areas. Although justified by Spain’s population distribution, this difference could pose a risk of biased results. In this vein, the results could also be affected by the profile of the respondents. Another limitation is that, since this work is part of a longitudinal study, it is difficult to increase the sample size each year when asking about COVID-19. Therefore, in future research on ‘smart territories’, it would be interesting to omit issues such as the pandemic and include other indicators that can affect the quality of life, such as tourism or ICT, especially in rural areas.