scholarly journals Analiza regresji z programem R – przykład użycia w badaniach politologicznych

2020 ◽  
Vol 2020 (55) ◽  
pp. 396-405
Author(s):  
Mateusz Wajzer ◽  
Monika Cugier-Syguła

The aim of the article is to present the basic functionalities of the R program for the creation of regression models describing political phenomena. A database of voter turnout during the 2014 U.S. Congress elections categorised according to voters’ age was used for the analyses. The statistical procedures (linear and second-degree polynomial models) applied were discussed in detail, with paths to their respective commands being provided. The article is addressed primarily to postgraduate students in political science and related disciplines, as well as to researchers who have never used the R program before.

FLORESTA ◽  
2013 ◽  
Vol 43 (3) ◽  
pp. 439 ◽  
Author(s):  
Saulo Jorge Téo ◽  
Alan Marcon ◽  
Tiago Ehlers ◽  
Júlio César Bianchi ◽  
Adriano Peloso ◽  
...  

Os objetivos deste trabalho foram ajustar e selecionar modelos de afilamento para estimar diâmetros a várias alturas e volume total, com e sem casca, de árvores de Pinus elliottii Engelm., na região de Caçador, SC, para estudar a forma dos fustes das árvores de diferentes idades. Ao todo, 78 árvores, com idades de 10, 18 e 27 anos, foram abatidas, cubadas rigorosamente e tiveram os seus volumes obtidos por meio do método de Smalian. Os modelos de afilamento testados foram os Polinômios de 2º e de 5º Graus e o Polinômio de Potências Fracionárias de Hradetzky. A função de afilamento que apresentou melhor desempenho para estimar o diâmetro com e sem casca ao longo do fuste das árvores foram baseadas no Polinômio de Potências Fracionárias de Hradetzky, para as idades de 10, 18 e 27 anos. O Polinômio de 2º Grau propiciou as melhores estimativas de volume total com casca para as idades de 10 e 27 anos e de volume total sem casca para as idades de 18 e 27 anos. Já o Polinômio de Potências Fracionárias apresentou melhores estimativas de volume com casca para 18 anos e de volume sem casca para árvores de 10 anos. Os fustes das árvores de Pinus elliottii apresentam melhor forma nas idades mais avançadas. AbstractTaper models for Pinus elliottii, at different ages, in the region of Caçador - SC. The research aims to fit and select taper models in order to estimate diameters at different heights and total volumes, inside and outside bark, of Pinus elliottii Engelm. trees, in the region of Caçador - SC, Brazil, as well as to analyze  behavior of stem form for trees at different ages. Altogether, 78 trees at 10, 18 and 27 years of age, had been cut and their volumes obtained by Smalian method. The taper models tested were the 2nd and 5th Degree Polynomial and the Hradetzky Polynomial of Fractioned Potencies. The best performance taper function for estimate diameters inside and outside bark throughout the stem were based on the Hradetzky Polynomial of Fractioned Potencies, for 10, 18 and 27 years of age. The 2nd Degree Polynomial provided the best estimates of outside bark volume for 10 and 27 years of age, and of inside bark volume for 18 and 27 years of age, although the Hradetzky Polynomial presented the best estimates of outside bark volume for 18 years of age and of inside bark volume for 10 years old trees. The stems of the Pinus elliottii trees present better form for the oldest tress.Keywords: Polynomial models; profile models; stem form.


2017 ◽  
Vol 7 (2) ◽  
pp. 371-393 ◽  
Author(s):  
Jairo Nicolau ◽  
Lilian Oliveira

Abstract The article analyses the production of Brazilian political science, as published in six periodicals (Dados, Revista Brasileira de Ciencias Sociais, Novos Estudos Cebrap, Lua Nova,Opiniao Publica and Brazilian Journal of Political Science) over a period of five decades (1966-2015). The text emphasizes two aspects: first, the distribution of articles by thematic areas of political science; and second, the type of method used, with an emphasis on the variety of statistical procedures used by the authors. The article observes the growing volume of articles published by Brazilian political scientists, particularly from the second half of the 1990s. One of the conclusions of the paper is that although statistics are used - and in recent years more advanced techniques have spread - Brazilian political science is far from being considered an eminently quantitative discipline.


2017 ◽  
Author(s):  
Luke Keele ◽  
Randolph T. Stevenson

Social scientists use the concept of interactions to study effect dependency. Such analyses can be conducted using standard regression models. However, an interaction analysis may represent either a causal interaction or effect modification. Under causal interaction, the analyst is interested in whether two treatments have differing effects when both are administered. Under effect modification, the analysts investigates whether the effect of a single treatment varies across levels of a baseline covariate. Importantly, the identification assumptions for these two types of analysis are very different. In this paper, we clarify the difference between these two types of interaction analysis. We demonstrate that this distinction is mostly ignored in the political science literature. We conclude with a review of several applications.


2013 ◽  
Vol 107 (2) ◽  
pp. 362-374 ◽  
Author(s):  
JAMES H. FOWLER ◽  
CHRISTOPHER T. DAWES

The American Political Science Review recently published a critique of an article we published in the Journal of Politics in 2008. In that article we showed that variants of the genes 5HTT and MAOA were significantly associated with voter turnout in a sample of 2,300 subjects from the National Longitudinal Study of Adolescent Health. Here, we address the critique first by conducting a replication study using an independent sample of 9,300 subjects. This study replicates the gene-environment interaction of the 5HTT gene variant with church attendance, but not the association with MAOA. We then focus on the general argument of the critique, showing that many of its characterizations of the literature in genetics and in political science are misleading or incorrect. We conclude by illustrating the ways in which genopolitics has already made a lasting contribution to the field of political science and by offering guidelines for future studies in genopolitics that are based on state-of-the-art recommendations from the field of behavior genetics.


2017 ◽  
Vol 62 (2) ◽  
pp. 29-60
Author(s):  
Mircea Comşa

Abstract Turnout decline in former communist countries has attracted a great deal of scholarly attention. In this paper, I re-test some of the previous hypotheses on new data and I propose a new hypothesis that considers the impact of external migration. Using multivariate regression models on a dataset of 272 presidential and parliamentary elections held in 30 post-communist countries between 1989 and 2012, I have found strong support for the “migration hypothesis”: other things being equal, an increase of migration rate by 1 percentage point reduces voter turnout by around 0.4 percentage points. Most of the previous hypotheses related to causes of turnout decline are supported too.


Sign in / Sign up

Export Citation Format

Share Document