1 Description of the variables included in the index, together with their relative importance within each of the pillars of Rural Development. Full understanding of the main determinants of economic and social growth of rural areas remains one of the main policy issues (Bryden 2002), and given the multiple dimensions of rural development, there is a great interest among policy makers to learn more about the extent and trends in overall welfare in rural regions (Michalek and Zarnekow 2012a).
CAN RURAL DEVELOPMENT BE MEASURED? DESIGN AND
I NTRODUCTION
This is all the more remarkable when we consider that until recently most of the world's population lived in rural areas and depended on agriculture. A full understanding of the main determinants of economic and social growth of rural areas remains one of the most important policy issues (Bryden, 2011), and given the many dimensions of rural development, there is a huge interest among policy makers to learn more about the size and trends in overall well-being in rural areas (Michalek and Zarnekow, 2012a).
T HEORETICAL F RAMEWORK
The Kageyama Rural Development Index has four different dimensions or sub-indices: demography (Population Index), economy (Economic Development Index), social welfare (Social Welfare Index) and environment (Environment Index) - so that to classify the territorial units according to their Rural development. The Kageyama index ranges from 0 to 1, which means that the closer it is to 1, the higher the level of rural development of the territory in question.
M ATERIALS AND M ETHODOLOGY
- Construction of the Rural Development index RDI Abreu
- Data collection
Demographic dependency index - the ratio of the elderly (age 65 and over) plus the youth (under 15) to the working-age population (age 15-64). Literacy (ratio, values from 0 to 1) – education being one of the most indisputable variables of development, the literacy rate is its own. The index proposed here uses a geometric mean within each dimension (population, social, economic and environmental) in order to sum the variables, instead of the arithmetic mean, avoiding a substitution effect.
The nature of the values used to calculate the index (standardized data) provides index values between 0 and 1.
R ESULTS AND D ISCUSSION
Furthermore, the economic dependence of rural areas on agricultural activities decreases when other sources of economic activity are present, such as the case of the largest municipalities and Campo Maior. It is also possible to analyze these results more in depth from various perspectives, looking for any correspondence between the characteristics of the different areas and its RDI values. Regarding the Population dimension, it can be highlighted that the values increased in most of the studied municipalities in the analyzed period, although the two largest cities (Portalegre and Elvas) showed significant decreases.
However, it can be observed that in two of the analyzed municipalities (Castelo de Vide and Gavião) the value of this component has decreased sharply.
C ONCLUSIONS
Rural areas and agriculture are no longer synonymous, and the positive externalities generated by the multifunctionality of agriculture are unanimously recognized, supporting other economic activities and promoting the development of the region (with rural tourism as a good example). The fact that rural areas are composed of a wide variety of small territorial units with their own specificities leads to difficulties in finding appropriate measures for rural development, which should always make it possible to take into account the particularities of the different areas , which is investigated. .
R EFERENCES
Implementation of a national program to strengthen family agriculture in the state of Rio Grande do Sul. Using the Rural Development Index in the Analysis of Rural Regions in Poland and Slovakia.
THE ASSESSMENT OF RURAL DEVELOPMENT
I NTRODUCTION
These policies pursue objectives and support initiatives that often overlap and create gaps due to the lack of a clear identification of the distribution of institutional responsibilities (Ferrão 2014). A first approximation to this RD index would come from the use of various indices designed to measure overall development, such as the Gross Domestic Product (GDP) (World Bank 1997), the UN Human Development Index (UNDP 2016b) or Multidimensional Poverty. Index (Oxford Poverty and Human Development Initiative 2010a). Examples of the use of both indices can be found in the work of (Haag 2009), who applied the Kageyama index to measure the effectiveness of public policies in Brazil, and in that of (I. Abreu et al. 2019), who applied their index to measure the RD of 15 Portuguese municipalities.
It should not be overlooked that one of the main political issues is understanding the main drivers of economic and social growth in rural areas (Bryden 2002), especially after the growth of community movements fighting against the abandonment and depopulation of rural areas (Alonso 2020; Ramírez 2020 ).
M ATERIAL AND M ETHODOLOGY
- The Delphi Method
- Development of the study
- Consensus
Finally, Section 4 describes the main conclusions of the study and shows some recommendations for stakeholders and policy makers. This first phase enabled the selection of 88 indicators included in the first draft of the questionnaire. A preliminary version of the first round questionnaire was sent to 3 experts (not included in the final sample) who performed a final review.
The formula used to calculate this indicator was the difference between the standard deviation of the ratings from the first and second rounds, as used in other Delphi studies (Alcon et al.
R ESULTS AND D ISCUSSION
- Population
- Social welfare
- Economy
- Environment
- Selected set of indicators
In this sense (Spellerberg et al. 2007) points out that population potential and infrastructural factors are drivers of differences between rural areas. 7 The demographic dependency index is the ratio between the elderly (aged 65 and over) and the young (under 15) and the working-age population (aged 15-64). It should be noted that several selected indicators are related to natural resources and environmental quality.
This is in line with what has been indicated by other authors (Sánchez-Zamora et al. 2014b) about the influence of the environment and natural resources on the dynamics and development of rural areas.
C ONCLUSIONS
R EFERENCES
However, the process needs to be improved by ensuring that the required data are available and accessible, and that feedback is provided after the selected indicators are applied and problems are identified, for example due to lack of up-to-date data or the difficulty in accessing data. Towards strategies to adapt to pressures on fresh produce security due to climate change. Towards a common set of criteria and indicators to identify forest restoration priorities: A panel based expert.
The development of indicators of regional sustainability and the role of academia in this process: the Portuguese practice.
DESIGN AND VALIDATION OF AN INDEX TO MEASURE
I NTRODUCTION
In a world that is increasingly developed and urbanized and where population movements from the rural areas to the cities have been taking place for decades, the rural environment finds itself in an increasingly unfavorable situation with regard to the urban areas (I. Abreu et al. . 2019). This is reflected in an increasingly pronounced imbalance between the level of development of rural and non-rural areas. Like any other policy, RD must be evaluated to determine the effectiveness of the measures implemented, to analyze new initiatives or to decide which areas need more attention (I. Abreu et al. 2019).
Next, different aggregation methods accepted by the scientific community are used to build the RD index: simple aggregation of indicators by geometric/arithmetic mean, as used on the HDI (Conceição 2019) or on the RD index proposed by (I. Abreu et al. 2019); weighted arithmetic mean of the indicators using the results of a panel of RD experts through a Delphi approach (I. Abreu and Mesias 2020);.
M ATERIALS AND M ETHODOLOGY
- Data collection
- Aggregation methods
- RDI Abreu
- RDI PCA
- RDI Delphi
- Assessment by experts
For some indicators obtained from INE and due to the type of data provided, some methodological adjustments had to be made, as detailed below. We use the normalized values of the indicators and first calculate the arithmetic mean of the indicators belonging to each dimension. In this study, the Delphi RDI was calculated as a weighted arithmetic mean using weights previously generated for each of the indicators by a panel of Delphi experts (I. Abreu and Mesias 2020).
The results of the three RDIs were then assessed by a panel of 25 Portuguese rural development experts, who had also previously participated in the preliminary selection of the indicators (Abreu and Mesias, 2020).
R ESULTS AND D ISCUSSION
- Assessment by experts
The experts were contacted by email and sent a summary of the results of the three RDIs for the nine Portuguese NUTS III regions analyzed. From the information presented in Figure 3.3 we can conclude that there are no major differences in the range and distribution of the three calculated indices. They were also asked to indicate which of the outcomes had been most important to their responses.
The result of the experts' assessment agrees with the more usual trend in composite index construction (Greco et al. 2019; OECD 2008).
C ONCLUSIONS
Analyzing the reasons given by the experts in support of their decisions, the main points of disagreement with the RDI PCA stem from the relative position of the Coimbra and Açores regions. On the other hand, Açores is the Portuguese region with one of the worst health service indicators and the highest income inequality (Diogo 2019), so experts do not accept its highest position on the RDI PCA. The results showed the statistical similarity of two of the proposed indices (RDI Abreu and RDI Delphi) compared to the RDI PCA index.
They believed that RDI Abreu more accurately reflects different levels of rural development, while experts disagreed with some of the results generated by RDI PCA.
C REDIT A UTHORSHIP C ONTRIBUTION S TATEMENT
It also makes it difficult to identify areas where the use of public funds would be more efficient, resulting in less effective rural development policies. In addition, the various approaches used to construct these indices often suffer from a lack of stakeholder involvement, both in the selection of indicators to be included in the study and in the various ways of combining them to create a unit of measurement. . The approach used in this paper aims to bridge these gaps by using a series of indicators previously created by a group of rural development experts, which became the input data for the creation of several rural development indices using different aggregation methodologies (arithmetic/geometric mean with and without weighting -RDI Abreu and RDI Delphi) and clustering using principal component analysis -RDI PCA).
Finally, a panel of rural development experts reviewed the scores generated by the three indices for a set of 9 NUTS III Portuguese regions.
R EFERENCES
On the methodological framework for composite indices: A review of the issues of weighting, aggregation and robustness. Since none of the existing indices solves the problem of applicability to small territories (e.g. at the municipal level) or takes into account the special conditions of rural areas, it was proposed to work on an index specially designed for rural areas. These questions have been addressed by this work through the use of the qualitative Delphi methodology, based on the opinion of a panel of experts with different roles in rural development, which allowed to define a set of demographic, environmental, economic and social welfare indicators that could used to assess rural development and related policies.
Based on the consensus of the experts, 25 indicators were selected, covering the 4 dimensions on which the rural development analysis is structured.
