Modify the original observed frequency table considering only 3 Regions and 2 Income Groups

What are your conclusions, if you run a Contingency Analysis, using α=0.05 ?

The 3 countries from Middle East & North Africa and the 2 from South Asia were grouped together in the same region (Middle East & North Africa & South Asia) and the Contingency Analysis was run.

The p-value is

The expected number (if Ho is true) of Low Income countries in South Asia is 

Is GDP per capita region related ?

Only 5 regions were considered (the country from Latin America & Caribbean was excluded from the sample) and GDP per capita mean values and confidence intervals were computed and plotted in this graph:

If you run an ANOVA (and the assumptions are met) what is your expectation for the p-value?

The 20 countries, randomly selected for this study, were grouped by geographic regions, allowing for a more nuanced analysis of economic and social indicators across different parts of the world, with a particular focus on investigating the Infant mortality rate is Region dependent.

The country from Latin America & Caribbean was excluded from the sample, reducing the total number of countries analysed and an ANOVA was run. The p-value was found to be smaller than 0.05.

The test statistic (F) is

Run the ANOVA (described in Question 8), to investigate if Infant mortality rate is Income Group dependent.

The p-value is

If you run the test described in Question 8, how many degrees of freedom will you assign to the Residuals sum of squares ?

Focus on Infant mortality rate and Ho: μ_{Low income}=μ_{Middle income}=μ_{Upper income}.

Homoskedasticity is met! Can we perform the test, with α=0.01, with the provided sample (n=20)?

Again regarding this histogram (from Question 6) built with a sample size n=45:

With this sample, would you be able to run a test such that Ho: π≥0.7 vs H1: π<0.7, where π is the proportion of countries (in the World Bank & United Nations Development Programme database) with a Population density above 500 ?