Quantitative Methods in Social Sciences (E774). December 4, 2009. Gender Differences A mong IHEID 1st Year Mdev Students Catherine Doe Adodoadji Nathália Estevam Fraga Aleksandra Žaronina Maurice Tschopp 4 December 2009. Introduction and Hypothesis.
December 4, 2009
GenderDifferencesAmongIHEID 1st YearMdev Students
4 December 2009
Hypothesis: There is no significant difference between male and female
MDev students’ personal profiles and opinions on global issues
Internal- the degree to which conclusions about causes of relations are likely to be true, in view of the measures used, the research setting, and the whole research design
External - to what extent one may safely generalize the (internally valid) causal inference (a) from the sample studied to the defined target population
Mostly qualitative and discrete
Encompassing both opinions and numeric data
Data collected is consistently
Units of measurement standardized
Augmenting level of complexity of techniques
Errors tests performed
Estimation and significance
Policy Paper 1:
In this PP we mainly used descriptive statistics instruments to discuss differences in opinions among female and male Mdev Students.
Method: Comparing proportions and distribution of variables with the help of tables and figures.
Results and conclusion: comparable patterns in the answer of male and female students:
Example: 74 % of female students believe that Global poverty is the most important global issue. 62 % of male students have the same opinion.
Method: using Z calculations and test of hypothesis to see if the differences in percentages were significant enough.
87.5 % of male students believe that their country has been negatively affected by world crisis. (77% have the same opinion). Test for male students
H0: π = 0.77 H1: π > 0.77
Z = π - π 0 = 0.875 – 0.77 = 1.35 Z (table) =1.65 (one tailat 95%)
SE 0.0826 Z< Z (table)
We cannot reject H0. Therefore we cannot say with a confidence interval of 95 % that there is a statistical differences on opinions of both male and female students on this issue.
Additional hypothesis: Students who spend a lot of time and internet also spend more time on phone and have better computer proficiency.
We also looked at our “old” hypothesis, that states that there is no significant difference among men and women on the variables tested.
Method: Using correlations and regressions to find similar patterns in male and female subsample and to test our second hypothesis.
Regression time internet //time on phone for femalestudents
R Squared = 0.4916
Beta = 0.41 p-value= 0.000
Regression time internet //time on phone for male students
Beta= 0.069; n = 12
What’s new about our approach:
What we learnt:
Shortcomings of the dataset and missing elements of research:
Suggestions: 2nd round of surveys at the end of the semester