1,098
edits
Changes
no edit summary
Five social and economic variables were selected to develop a synthetic vulnerability index for Barcelona and Madrid: non Spanish population, illiterate or uneducated population, average income per capita, unemployed and social welfare beneficiaries. Data were sourced from the National Statistics Institute (experimental statistics) and the statistics portals from Madrid and Barcelona.
The economic variables were chosen to take into account citizens with lower income, unemployed and those in in-work poverty. Immigrants and illiterate or uneducated people are the groups potentially most exposed to poverty, as is confirmed in the VIII Foessa Report: ''Exclusion and social development in Spain 2019 (Exclusión y desarrollo social en España 2019)'', and the Oxfam Intermón publication (2020) ''A fair reconstruction is possible and necessary. This is not the time for austerity; let us choose dignity (Una reconstrucción justa es posible y necesaria. No es momento para la austeridad, elijamos dignidad)''. A long-lasting lack of income leads to depending on social welfare benefits. These variables have been included and assessed in important studies on vulnerability, such as the ''Atlas of Urban Vulnerability (Atlas de la Vulnerabilidad Urbana) and the Urban Analysis of Vulnerable Neighbourhoods in Spain (Análisis urbanístico de Barrios Vulnerables en España)'', published by the Ministry of Development (Ministerio de Fomento, 2012, 2015).
{| cellspacing="0" cellpadding="0" style="margin-top: 0; margin-right: 0px; margin-bottom: 0em; margin-left: 0px; float:right; width:350px"
| style="vertical-align: top; border-style: solid; border-width: 1px; border-color: #ffffff; background-color:#ffffff" |
|}
|}
Once the information was compiled and processed, a synthetic vulnerability index was developed using the principal component analysis technique. Variables were displayed in rows and principal components in columns. Ideally, each principal component relates positively to a few variables (with correlation coefficients close to +1 or -1) and poorly to the other variables (with coefficients close to 0). In this case, of the two principal components obtained, the first was considered the most representative one and it was therefore selected. Lastly, the results were transferred to a geographical information system and sorted using the natural breaks classification method.
