Political Preferences and Individual Choice: A Latin American's Countries Perspective

Madalozzo, Regina Carla; Chebib, Afonso Mariutti

doi:10.1590/1678-987317256301

Acessibilidade / Reportar erro

Brasil

Revista de Sociologia e Política

Español English

Brasil

Español English

sumário « anterior atual seguinte »

Sumário

Artigo • Rev. Sociol. Polit. 25 (63) • 01 Sept 2017 • https://doi.org/10.1590/1678-987317256301 copy

Political Preferences and Individual Choice: A Latin American's Countries Perspective

Preferências políticas e escolhas individuais: Uma perspectiva de países Latino Americanos

Authorship SCIMAGO INSTITUTIONS RANKINGS

Abstract

Analyzing the “left” and “right” political positions of individuals is challenging because personal attributes may influence political decisions without directly causing them. This issue may be even more pronounced in Latin America, where young democracies encounter the challenge of stabilizing political choices over time. This study contributes to the literature by analyzing the influence of personal attributes on political choices, focusing on the early 2000s, when the “left” turn occurred. The present study relies on the World Values Survey's fifth wave (2005-2008) to fulfil this objective. This dataset is composed of data that have been collected globally, and the questions are related to diverse subjects associated with the quality of life of individuals. From the available sample, we included all of the Latin American countries that participated in this wave: Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Uruguay. In this study, the aim is to directly understand the impact of these individuals own attributes on their declared self-positioning about the political leaning. To this aim, an ordered logit model was used to analyse how each variable exerts influence on the political leaning of the respondents. Our results found that political cleavages depend on demographic factors, economic factors, and individual opinions in agreement with previous studies. Increased age, religious service attendance, and satisfaction with one's financial life increase the tendency of individuals for self-positioning to the right of the political spectrum. The possession of a university degree and residence in a large city increases the likelihood of individuals of self-identifying with a leftist political position. This study contributes to the literature by analysing the influence of personal attributes on political choices. Although this research represents an important step toward understanding political leanings in Latin American countries, a significant amount of future research remains. The definitions of “left” and “right” continue to be unclear as they relate to dimensions that include democracy, autocracy, and political reform. Understanding the ways individuals set up their choices would increase the responsibility of political parties and authorities for the hidden claim the population has about their deliverables. At the same time, studies like ours may enhance the awareness of the general impression over political party choices of candidates. Finally, even with so many confounding aspects in this antagonist position, the left and the right continue to be a simple way to characterize veiled assumptions. Therefore, it may be an ambiguous form of defining choices, but this seemingly binary choice is still very significant for voters in Latin America.

Keywords:
political preferences; Latin American voters; political leanings; individual attributes; World Values Survey

	Argentina	Brazil	Chile	Colombia	Mexico	Peru	Uruguay	Total
Left-Right	5.53 (1.89) ^I I The standard deviations are listed in parentheses.	5.43 (2.39)	5.38 (2.15)	6.75 (2.58)	6.18 (2.98)	5.65 (2.28)	5.23 (2.45)	5.92 (2.55)
Democracy	9.11 (1.68)	8.24 (2.15)	8.22 (2.07)	7.86 (2.44)	8.69 (2.28)	8.37 (2.26)	8.91 (1.76)	8.37 (2.23)
Agrees with Strong Leadership	1.14 (0.93)	1.73 (0.90)	1.09 (0.89)	1.15 (0.81)	-	1.43 (0.99)	1.26 (0.86)	1.30 (0.91)
Satisfaction with Financial Life	6.46 (2.40)	5.87 (2.63)	5.80 (2.35)	-	7.13 (2.66)	5.68 (2.46)	6.18 (2.24)	6.20 (2.54)
# Observations ^II II These values reflect the number of people who answered the question regarding political positioning between left and right. Varying numbers of people answered each question.	690	1,370	706	2,253	1,293	1,128	867	8,307

	Argentina	Brazil	Chile	Colombia	Mexico	Peru	Uruguay	Total
Demographics
Age	42.55 (17.59)^I I The standard deviations are listed in parentheses.	39.96 (15.68)	42.93 (16.98)	36.96 (13.97)	39.69 (15.72)	37.62 (14.90)	46.53 (18.65)	39.88 (16.02)
Female	0.53 (0.50)	0.58 (0.49)	0.55 (0.50)	0.50 (0.50)	0.51 (0.50)	0.51 (0.50)	0.56 (0.50)	0.53 (0.91)
Education Profile
Incomplete elementary	10.78	34.83	14.93	5.26	19.78	13.27	15.33	15.10
Elementary	22.95	13.73	8.12	30.12	16.70	10.60	23.95	19.72
Incomplete secondary 1	17.56	10.18	16.53	-	4.69	11.67	31.56	9.99
Secondary 1	22.26	25.59	30.06	-	18.75	22.87	13.93	15.88
Incomplete secondary 2	3.59	-	3.51	-	5.78	7.20	0.10	2.55
Secondary 2	6.99	-	8.82	45.32	12.14	13.27	0.30	18.16
Incomplete university	9.08	6.83	6.51	-	7.90	9.53	7.01	5.62
University	6.79	8.84	11.52	19.31	14.26	11.60	7.82	12.99
Social Class
Lower	9.11	25.92	13.36	15.50	1.57	11.60	17.05	13.75
Working	47.96	35.66	25.71	47.65	16.66	48.64	37.55	38.47
Lower Middle	31.10	35.73	46.15	11.88	18.43	26.48	34.83	25.49
Upper Middle	11.52	2.01	13.26	24.13	62.56	12.65	10.04	21.51
Upper	0.31	0.37	1.52	0.84	0.79	0.63	0.52	0.77
Profession/Occupation
Employer/Manager (1)	0.10	0.77	0.87	-	1.21	-	1.22	0.76
Employer/Manager (2)	3.29	5.86	1.24	-	12.45	-	7.33	5.81
Professional worker	7.19	7.88	10.92	-	17.53	-	10.51	10.29
Supervisory	4.49	1.39	3.85	-	3.63	-	2.93	3.08
Non-manual	18.36	6.83	6.45	-	9.31	-	11.74	10.25
Foreman	0.80	1.19	1.12	-	2.18	-	0.73	1.23
Skilled manual	18.66	15.69	20.10	-	11.00	-	16.14	16.32
Semi-Skilled manual	8.38	16.32	20.10	-	20.92	-	7.09	15.23
Unskilled manual	19.16	16.95	32.51	-	8.59	-	29.83	19.87
Farmer (owner)	0.80	1.12	0.00	-	0.85	-	5.62	1.21
Agricultural worker	1.20	7.32	1.86	-	11.97	-	3.42	5.47
Armed forces	1.50	2.23	0.99	-	0.36	-	3.42	1.61
Never had a job	16.07	16.46	0.00	-	0.00	-	0.00	8.87
Size of Town
Less than 2,000	7.98	0.00	0.00	-	16.16	-	8.80	6.93
2,000 to 5,000	5.99	0.67	0.00	-	9.23	-	9.00	5.01
5,000 to 10,000	0.00	6.00	0.00	-	3.08	-	6.00	3.27
10,000 to 20,000	5.99	10.00	0.00	-	5.38	-	11.10	6.68
20,000 to 50,000	5.99	14.00	0.00	-	6.15	-	16.40	8.74
50,000 to 100,000	3.99	12.67	16.10	-	8.46	-	5.10	9.47
100,000 to 500,000	7.98	26.67	83.90	-	21.54	-	0.00	27.30
More than 500,000	62.08	30.00	0.00	-	30.00	-	43.60	32.60
Frequency to the Religious Cult
More than once a week	5.69	22.41	7.24	10.83	11.98	10.49	9.58	11.77
Once a week	14.47	26.42	18.10	34.70	34.32	30.53	17.15	28.34
Once a month	15.87	17.06	15.51	18.91	17.97	22.13	15.81	18.19
Special days	10.98	3.61	8.58	7.61	10.69	12.78	8.46	8.73
Once a year	7.19	3.41	7.65	5.92	5.22	6.79	9.58	6.02
Less than once a year	14.87	21.47	13.96	10.73	8.44	9.48	12.03	12.59
Never	30.94	5.62	28.96	11.30	11.40	7.80	27.39	14.35
# Observations ^II II These values reflect the number of people who answered the question regarding political positioning between left and right. Varying numbers of people answered each question.	690	1,370	706	2,253	1,293	1,128	867	8,307

	Estimated Coefficients [Robust Standard Error]	Estimated Coefficients [Robust Standard Error]
Age	0.003^ indicates that the coefficient is significant at the 5 percent level, and [0.002]	0.008^* ***** indicates that the coefficient is significant at the 1 percent level. [0.002]
Female	0.007 [0.042]	0.047 [0.081]
Education Indicators:
Elementary	-0.128 [0.087]	0.052 [0.115]
Incomplete secondary 1	-0.355^ indicates that the coefficient is significant at the 5 percent level, and [0.098]	-0.116 [0.157]
Secondary 1	-0.465^* ***** indicates that the coefficient is significant at the 1 percent level. [0.085]	-0.119 [0.136]
Incomplete secondary 2	-0.590^* ***** indicates that the coefficient is significant at the 1 percent level. [0.133]	-0.031 [0.270]
Secondary 2	-0.491^* ***** indicates that the coefficient is significant at the 1 percent level. [0.095]	-0.012 [0.225]
Incomplete university	-0.609^* ***** indicates that the coefficient is significant at the 1 percent level. [0.106]	-0.293 [0.180]
University	-0.745^* ***** indicates that the coefficient is significant at the 1 percent level. [0.091]	-0.331^* * indicates that the coefficient is significant at the 10 percent level, [0.184]
Social Class Indicators:
Upper class	0.705^* ***** indicates that the coefficient is significant at the 1 percent level. [0.266]	0.323 [0.521]
Upper middle class	0.085 [0.087]	0.346^* * indicates that the coefficient is significant at the 10 percent level, [0.186]
Lower middle class	0.133^* * indicates that the coefficient is significant at the 10 percent level, [0.076]	0.107 [0.129]
Working class	0.040 [0.072]	0.046 [0.125]
Frequency to Religious Cult:
More than once a week	0.389^* ***** indicates that the coefficient is significant at the 1 percent level. [0.089]	0.585^* ***** indicates that the coefficient is significant at the 1 percent level. [0.147]
Once a week	0.390^* ***** indicates that the coefficient is significant at the 1 percent level. [0.073]	0.286^ indicates that the coefficient is significant at the 5 percent level, and [0.129]
Once a month	0.297^* ***** indicates that the coefficient is significant at the 1 percent level. [0.076]	0.322^ indicates that the coefficient is significant at the 5 percent level, and [0.134]
Special days	0.417^* ***** indicates that the coefficient is significant at the 1 percent level. [0.089]	0.430^* ***** indicates that the coefficient is significant at the 1 percent level. [0.167]
Once a year	0.407^* ***** indicates that the coefficient is significant at the 1 percent level. [0.103]	0.538^* ***** indicates that the coefficient is significant at the 1 percent level. [0.177]
Less Often	0.250^* ***** indicates that the coefficient is significant at the 1 percent level. [0.079]	0.245^ indicates that the coefficient is significant at the 5 percent level, and [0.130]
Democracy	0.022^ indicates that the coefficient is significant at the 5 percent level, and [0.011]	-0.042^ indicates that the coefficient is significant at the 5 percent level, and [0.021]
Size of Town:
2,000 to 5,000	-	-0,103 [0.294]
5,000 to 10,000	-	0.150 [0.339]
10,000 to 20,000	-	-0,296 [0.254]
20,000 to 50,000	-	-0,360 [0.247]
50,000 to 100,000	-	-0,319 [0.251]
100,000 to 500,000	-	-0,593^ indicates that the coefficient is significant at the 5 percent level, and [0.240]
More than 500,000	-	-0,565^ indicates that the coefficient is significant at the 5 percent level, and [0.223]
Profession/Occupation
Employer/Manager (1)	-	-0.205 [0.528]
Employer/Manager (2)	-	0.102 [0.197]
Professional worker	-	-0.086 [0.182]
Supervisory	-	-0.378 [0.250]
Non-manual	-	0.007 [0.147]
Foreman	-	-0.254 [0.387]
Skilled manual	-	-0.152 [0.131]
Semi-Skilled manual	-	0.005 [0.139]
Farmer (owner)	-	0.04 [0.354]
Agricultural worker	-	-0.032 [0.217]
Armed forces	-	0.166 [0.240]
Never had a job	-	0.008 [0.148]
Life Satisfaction	-	0.085^* ***** indicates that the coefficient is significant at the 1 percent level. [0.018]
Strong Leader	-	0.058^* ***** indicates that the coefficient is significant at the 1 percent level. [0.043]
Brazil	-0.196^ indicates that the coefficient is significant at the 5 percent level, and [0.076]	-0.206^* * indicates that the coefficient is significant at the 10 percent level, [0.114]
Chile	-0.099 [0.082]	-0.029 [0.147]
Colombia	0.921^* ***** indicates that the coefficient is significant at the 1 percent level. [0.079]	-
Mexico	0.539^* ***** indicates that the coefficient is significant at the 1 percent level. [0.090]	-
Peru	0.164^ indicates that the coefficient is significant at the 5 percent level, and [0.074]	-
Uruguay	-0.038 [0.110]	-0.429^ indicates that the coefficient is significant at the 5 percent level, and [0.212]
# Observations	7,518	2,455
Pseudo R²	0.0181	0.0141

	Estimated Coefficients [Robust Standard Error]	Estimated Coefficients [Robust Standard Error]
Age	0.003^ indicates that the coefficient is significant at the 5 percent level, and [0.002]	0.007^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.003]
Age – Uruguay	-	0.024^ indicates that the coefficient is significant at the 5 percent level, and [0.013]
Female	0.006 [0.042]	0.061 [0.081]
Education Indicators:
Elementary	-0.231^ indicates that the coefficient is significant at the 5 percent level, and [0.092]	0.011 [0.147]
Incomplete secondary 1	-0.108 [0.107]	-0.106 [0.159]
Inc. Sec. 1 – Mexico	-0.517^* * indicates that the coefficient is significant at the 10 percent level, [0.310]	-
Inc. Sec. 1 – Uruguay	-0.940^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.098]	-
Secondary 1	-0.306^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.089]	-0.203 [0.138]
Sec. 1 – Mexico	-0.497^ indicates that the coefficient is significant at the 5 percent level, and [0.205]	-
Sec. 1 – Uruguay	-	1.601^ indicates that the coefficient is significant at the 5 percent level, and [0.480]
Incomplete secondary 2	-0.268^* * indicates that the coefficient is significant at the 10 percent level, [0.141]	-0.106 [0.280]
Inc. Sec. 2 – Mexico	-0.827^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.304]	-
Secondary 2	-0.114 [0.124]	-0.011 [0.229]
Sec. 2 – Colombia	-0.687^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.153]	-
Sec. 2 – Mexico	-0.691^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.246]	-
Sec. 2 – Uruguay	0.959^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.202]	-
Incomplete university	-0.258^ indicates that the coefficient is significant at the 5 percent level, and [0.128]	-0.325^* * indicates that the coefficient is significant at the 10 percent level, [0.183]
Inc. University – Argentina	-0.392^ indicates that the coefficient is significant at the 5 percent level, and [0.174]	-
Inc. University – Mexico	-0.811^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.252]	-
Inc. University – Uruguay	-1.043^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.297]	-
University	-0.246^ indicates that the coefficient is significant at the 5 percent level, and [0.118]	-0.099 [0.221]
University – Brazil	-0.309^* * indicates that the coefficient is significant at the 10 percent level, [0.177]	-0.591^ indicates that the coefficient is significant at the 5 percent level, and [0.251]
University – Colombia	-0.908^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.155]	-
University – Mexico	-1.015^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.223]	-
University – Uruguay	-1.261^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.374]	-
Social Class Indicators:
Upper class	0.697^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.267]	0.587 [0.514]
Upper middle class	0.074 [0.088]	0.313^* * indicates that the coefficient is significant at the 10 percent level, [0.189]
Lower middle class	0.080 [0.077]	0.071 [0.132]
Working class	0.019 [0.073]	0.132 [0.129]
Working class – Chile	-	-0.754^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.221]
Frequency to Religious Cult:
More than once a week	0.169 [0.129]	0.667^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.147]
More than weekly – Argentina	0.540^ indicates that the coefficient is significant at the 5 percent level, and [0.255]	-
More than weekly – Brazil	0.462^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.158]	-
More than weekly – Colombia	0.330^* * indicates that the coefficient is significant at the 10 percent level, [0.176]	-
More than weekly – Mexico	-	-1.856^ indicates that the coefficient is significant at the 5 percent level, and [0.733]
More than weekly – Uruguay	-	-0.355^ indicates that the coefficient is significant at the 5 percent level, and [0.098]
Once a week	0.411^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.073]	0.309^ indicates that the coefficient is significant at the 5 percent level, and [0.128]
Once a month	0.235^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.083]	0.324^ indicates that the coefficient is significant at the 5 percent level, and [0.134]
Once a month – Argentina	0.285^ indicates that the coefficient is significant at the 5 percent level, and [0.139]	-
Once a month – Mexico	0.333^* * indicates that the coefficient is significant at the 10 percent level, [0.170]	-
Special days	0.335^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.092]	0.422^ indicates that the coefficient is significant at the 5 percent level, and [0.170]
Special days – Mexico	0.434^ indicates that the coefficient is significant at the 5 percent level, and [0.214]	-
Once a year	0.444^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.103]	0.537^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.178]
Less Often	0.279^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.080]	0.243^* * indicates that the coefficient is significant at the 10 percent level, [0.131]
Democracy	0.033^ indicates that the coefficient is significant at the 5 percent level, and [0.016]	-0.189^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.029]
Democracy – Argentina	-0.176^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.040]	-
Democracy – Brazil	-	0.244^ indicates that the coefficient is significant at the 5 percent level, and [0.041]
Democracy – Chile	-0.191^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.036]	-
Democracy – Colombia	0.053^ indicates that the coefficient is significant at the 5 percent level, and [0.026]	-
Democracy – Uruguay	-0.122^ indicates that the coefficient is significant at the 5 percent level, and [0.057]	-
Size of Town:
2,000 to 5,000	-	0.075 [0.299]
5,000 to 10,000	-	0.389 [0.347]
10,000 to 20,000	-	-0,012 [0.262]
20,000 to 50,000	-	-0,029 [0.253]
50,000 to 100,000	-	-0,034 [0.263]
100,000 to 500,000	-	-0,277 [0.253]
More than 500,000	-	-0,111 [0.261]
More than 500,000 – Argentina		-0.375^ indicates that the coefficient is significant at the 5 percent level, and [0.184]
Profession/Occupation
Employer/Manager (1)	-	-0.126 [0.502]
Employer/Manager (2)	-	0.100 [0.197]
Professional worker	-	-0.084 [0.199]
Professional – Chile		-0.809^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.281]
Supervisory	-	-0.399 [0.247]
Non-manual	-	0.145 [0.149]
Non-manual – Uruguay		-1.902^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.620]
Foreman	-	-0.262 [0.404]
Skilled manual	-	-0.128 [0.131]
Semi-Skilled manual	-	0.051 [0.139]
Farmer (owner)	-	0.051 [0.317]
Agricultural worker	-	0.038 [0.225]
Armed forces	-	0.121 [0.248]
Never had a job	-	0.015 [0.151]
Life Satisfaction	-	0.086^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.018]
Strong Leader	-	-0.033 [0.053]
Strong leader – Argentina		0.239^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.087]
Brazil	-1.761^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.370]	-2.189^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.400]
Chile	-0.118 [0.435]	0.236 [0.217]
Colombia	-0.645^* * indicates that the coefficient is significant at the 10 percent level, [0.383]	-
Mexico	-0.648 [0.395]	-
Peru	-1.461^* ***** indicates that the coefficient is significant at the 1 percent level. Although we included many variables in our models, the average VIF is low enough (3.85) to consider that multicollinearity does not threat our results. [0.373]	-
Uruguay	-0.055 [0.581]	-1.374^ indicates that the coefficient is significant at the 5 percent level, and [0.621]
# Observations	7,518	2,455
Pseudo R²	0.0229	0.0248

Universidade Federal do Paraná Rua General Carneiro, 460 - sala 904, 80060-150 Curitiba PR - Brasil, Tel./Fax: (55 41) 3360-5320 - Curitiba - PR - Brazil
E-mail: editoriarsp@gmail.com

Acompanhe os números deste periódico no seu leitor de RSS

[TFN11] ^I
The standard deviations are listed in parentheses.

[TFN12] ^II
These values reflect the number of people who answered the question regarding political positioning between left and right. Varying numbers of people answered each question.