Academic Writing

All questions are based on an expanded version of the cross-national Democracy dataset ( i have the dataset ) For each question, include in your answer: ➢ A diagram of the hypothesised relationship ➢ The statistical procedure that you used and why ➢ The relevant null hypothesis ➢ The relevant results from your statistical analysis presented in nice tables or graphs (please don't just copy the Stata output!) ➢ The significance and – if significant – the form and strength of the relationship(s) in question ➢ A diagram of the relationship(s) given your results ➢ A conclusion of what your results mean (in terms of the real variables), and whether and why they confirm your expectations question example : Q1 Research hypothesis: Proportional representation electoral systems tend to make it easier for women to get into parliament Variables: ‘Women2008’, ‘Elecpr’ question answer example Sample question: Research hypothesis: A person’s astrological sign has no influence over their views on euthanasia. Variables: ‘zodiac’, ‘letdie1’ Sample answer: In this case, the research hypothesis is equivalent to the null hypothesis: that there is no relationship between the independent variable (astrological sign) and the dependent variable (attitudes towards euthanasia). As such, the hypothesis is consistent with the views of those who are sceptical about the capacity of astrological variables to explain any aspects of people’s attitudes or behaviour. The absence of an arrow in the diagram below reflects the hypothesis that the two variables are not statistically associated. To test that hypothesis, I use a crosstabulation and chi-square test. This is the appropriate significance test when dealing with two categorical variables. The table below shows the crosstab and the chi-square test result.