Embedded Files
Intro Stats: Islamic Approach -- Part 2: Probability and Statistics -
Lecture 1: Random Samples, Concept 2: Populations Slide 3: Example: Panel Data on Infant Mortality
This is a data set on Infant Mortality from the WDI Data set. For each country, the data set traces changes in Infant Mortality over time. For any give YEAR, the data is a cross-section of countries. For any fixed country, the data is a time series. This type of a data set is called PANEL DATA. The number are deaths of infants within first year of birth per 1000 live births.
An Important goal of this course is for students to learn to do data analysis visually and intuitively. Learn by looking at and plotting the data. Formal methods are often misused and lead to confusion instead of understanding.
With respect to RANDOM SAMPLING, it is important to state that IT IS ALMOST NEVER REASONABLE TO DRAW A RANDOM SAMPLE OF COUNTRIES FROM THE WDI DATA SET. This is because we have the ENTIRE data set in front of us, and we can do all necessary analysis on this data set. The ENTIRE OBJECT of random sampling is that random sample resemble the population and this permits us to make calculations about the population on the basis of the random sample. When the population is in front of us, then random sampling is not needed. Furthermore, a small random sample MAY NOT be representative and could easily be misleading.
Page updated
Google Sites
Report abuse