Damian Rosas and Dr. Darren Hawkins, Political Science Department
Main Text
Political rhetoric in Latin America has reflected the trends that different countries in the region have experienced during their democratic history. There are several factors that may influence the contents and attitudes towards democracy of presidential speeches. The most influential factor is likely to be the economic situation of the country. In an effort to integrate the economic and political knowledge that I acquired as an International Relations major I developed my research project in an interaction of these two disciplines.
Under the direction of Darren Hawkins and Kirk Hawkins I participated in a research project that consisted in the creation of a collection of presidential speeches from Latin American countries. This database of digital version of the speeches allows for a better identification of trends and patterns in the speeches through comprehensive analysis and coding. This base of available data was a determinant factor to carry out this research project.
The main objective of my project was to determine if general macroeconomic conditions in a given country affect the way that the president speaks about democracy. This is relevant because according to scholars “attitudes toward democracy and a favorable international political environment…have made a decisive difference in whether competitive regimes survive or break down. If the main actors are committed to democracy and if the international political environment is favorable, democracy can survive” (Mainwaring and Hagopian 2005, 7). The correlation that I expected to find in the statistical analysis is that during times of economic hardship the presidents will tend to reduce the amount of positive references to democracy in their rhetoric. In order to determine this I selected for analysis a sample of presidential speeches from four Latin American countries -Argentina, Mexico, Chile and Peru-. I chose one president from each of these countries based on the fact that during their administrations the country experienced a period of economic crisis and another one of economic prosperity. The sample of speeches was obtained from the collection we gathered in the Political Science department under Professor Darren Hawkins’ direction.
I selected the following presidents from each of the four countries, -Eduardo Frei (1994-2000) from Chile, Ernesto Zedillo (1994-2000) from Mexico, Alan Garcia (1985-1990) from Peru and Carlos Menem (1989-1999) from Argentina. I considered the time these presidents were on office in order to assure that they had been exposed relatively to the same international situations as well as the fact that they were elected democratically. According to economic data, all four presidents experienced periods of economic growth and a period of economic contraction. Since the economic downturn could have happen at the beginning or at the end of the president’s term, I randomly selected a speech for each year of that term up to six years.
I expected to find statistical evidence that economic conditions do influence the way that elite leaders speak about democracy in a given country. More particularly, that negative economic conditions are reflected in presidential rhetoric by a reduction in the positive references to democracy made in speeches to the public. The speeches were analyzed and coded by different research assistants following a coding rubric that has been developed to measure positivity or negativity towards democracy. These research assistants are qualified students who speak Spanish and have experience with this type of documents.
The research assistants read the speeches and determined the meaning of the different phrases, and coded them according to the rubric. After the coding was finished, the total number of codes was added and divided by the total number of sentences in order to obtain a ratio that would be used in the regression model. After all the speeches were coded and the ratios were calculated, this generated numerical data was organized in a table for statistical analysis. Each speech was coupled with the value of GDP growth that the country of the president that delivered it experiences that year. Statistical tests were carried out to determine if there was any significance in the relation between economic conditions and the way the president speaks about democracy.
According to the regression results, with a 1% increase in GDP growth the president will increase the ratio of positive references to democracy by ½ %. Considering the maximum and minimum GDP growth, these results are of important magnitude because a president with a negative 11% GDP growth will make about 12% less positive references to democracy in his speeches than a president with a 12% GDP growth in his country. This relation was is statically significant. Control variables were included to give more relevance to the results and to account for external factors.
These results are relevant but as a start for further research. Difficulties arise during the coding process because it is time consuming and requires qualified analyzers. In order to include a larger sample of speeches with more countries, more years and more speeches per year, more human and financial resources would be necessary. However, Political Science professors are working with this type of data and small end studies like mine contribute to polish coding systems and to see some results of the application of these research methods.
I very much enjoyed working in this project together with Political Science faculty that offered advice and help whenever needed. I now have graduated but the knowledge I obtained had been an important asset in pursuing a graduate degree. I hope to eventually apply this research knowledge to real life situations in politics and diplomatic activities.