Riassunto analitico
Obesity is a global issue that does not hint to stop and it is leading to death almost 2,8 milions people per year, according to World Health Organization statistics. The problem once associated to high income countries is now turning as pandemic across the globe, affecting also low-income economies. Many researchers have conducted analysis of the topic focusing on certain medical and/or socioeconomic variables. Furthermore, they paid particular attention on single states or limited region, which cannot take into account the global reach of the phenomenon and all the variables involved in it. In this scenario, I tried to give my contribution expanding the numerous researches conducted in the previous decades by outlining a model that could explain the phenomenon through socioeconomic factors that could be useful worldwide. Moreover, I considered the differences derived by gender belongings in the model, in order to see which factors are the most determinant for each category. The work is based on a panel data of 188 countries in the last 25 years (1995 - 2019), analysed through a multivariate ordinary least square regression, carried out by adding each variable sequentially, starting from the base model. The data were available through secondary sources databases and represent mostly sociodemographic and socioeconomic factors, such as labour market participation, GDP per capita value and its growth, urbanization level, religion, food production and consumption of meat, finally, the introduction of gender gap index as link between the male and female sphere. The comprehensive model was then enlarged by geographical dummy variables, that tested the geographic influence on the phenomenon, but can also cluster the customs of a society. Furthermore, two more models were outlined with gender-specific variables and then compared to each other, in order to spot the differences. The effects of independent variables on the dependent variable obesity rate (total, female and male) were analysed by pooling a step-wise OLS regression. Fixed-time and country effects have been used in order to control for the robustness of results. The results of the total model show how the variables with major effects on the model are meat consumption, Livestock prod., GDP per capita and some religion belongings for both sexes. Geographically speaking, the regions MENA (Middle East and North Africa) and Oceania look to have relevance in the topic. Specifically, female obesity is greatly influenced by meat consumption, food availability, level of income, labour market participation, Muslim and Christian religion belonging, and most important by MENA location. Besides, men obesity looks to be influenced by meat consumption, religiosity belonging and urban residency. The study shows how dietary income of meat affects primarily obesity, but the major role of urbanization, customs and tradition of the world regions, income and gender parity have found large confirmation. Therefore, public institutions must be able to harmonise the efforts in order to deliver policies that embrace different aspects regarding the person, in order to limit the spread of the phenomenon.
|
Abstract
Obesity is a global issue that does not hint to stop and it is leading to death almost 2,8 milions people per year, according to World Health Organization statistics. The problem once associated to high income countries is now turning as pandemic across the globe, affecting also low-income economies. Many researchers have conducted analysis of the topic focusing on certain medical and/or socioeconomic variables. Furthermore, they paid particular attention on single states or limited region, which cannot take into account the global reach of the phenomenon and all the variables involved in it.
In this scenario, I tried to give my contribution expanding the numerous researches conducted in the previous decades by outlining a model that could explain the phenomenon through socioeconomic factors that could be useful worldwide. Moreover, I considered the differences derived by gender belongings in the model, in order to see which factors are the most determinant for each category.
The work is based on a panel data of 188 countries in the last 25 years (1995 - 2019), analysed through a multivariate ordinary least square regression, carried out by adding each variable sequentially, starting from the base model. The data were available through secondary sources databases and represent mostly sociodemographic and socioeconomic factors, such as labour market participation, GDP per capita value and its growth, urbanization level, religion, food production and consumption of meat, finally, the introduction of gender gap index as link between the male and female sphere. The comprehensive model was then enlarged by geographical dummy variables, that tested the geographic influence on the phenomenon, but can also cluster the customs of a society. Furthermore, two more models were outlined with gender-specific variables and then compared to each other, in order to spot the differences.
The effects of independent variables on the dependent variable obesity rate (total, female and male) were analysed by pooling a step-wise OLS regression. Fixed-time and country effects have been used in order to control for the robustness of results.
The results of the total model show how the variables with major effects on the model are meat consumption, Livestock prod., GDP per capita and some religion belongings for both sexes. Geographically speaking, the regions MENA (Middle East and North Africa) and Oceania look to have relevance in the topic. Specifically, female obesity is greatly influenced by meat consumption, food availability, level of income, labour market participation, Muslim and Christian religion belonging, and most important by MENA location. Besides, men obesity looks to be influenced by meat consumption, religiosity belonging and urban residency.
The study shows how dietary income of meat affects primarily obesity, but the major role of urbanization, customs and tradition of the world regions, income and gender parity have found large confirmation. Therefore, public institutions must be able to harmonise the efforts in order to deliver policies that embrace different aspects regarding the person, in order to limit the spread of the phenomenon.
|