6. Statistical Inference
"There are three kinds of lies: lies, damned lies, and statistics." - Mark Twain
Lesson Prerequisites
This lesson assumes that you have taken a basic statistics course and have some familiarity with hypothesis testing. If you have not taken statistics before, it is strongly recommended that you take the free Udacity courses in descriptive statistics and inferential statistics before attempting this lesson. This lesson also assumes that you have beginner Stata skills and can complete the exercises in the lessons on Intro to Stata and Stata Best Practices.
0. Intro to the lesson
Statistical inference is the basis for all quantitative tools in the IDinsight toolkit, including needs assessment, impact evaluation, process evaluation, and monitoring.
1. Why we do statistical inference
Statistical inference provides a key input in decision-making about programs and policies.
2. The Central Limit Theorem
An intuitive understanding of the Central Limit Theorem is critical for diagnosing statistical problems and correctly interpreting the results of statistical tests.
3. Testing a hypothesis
The hypothesis test is the most important tool in the statistical inference toolbox.
4. Steps to testing for equality of means (part 1)
The hypothesis test has five main steps. Statistical software will only do two of them for you. (steps 1 & 2)
5. Steps to testing for equality of means (part 2)
The hypothesis test has five main steps. Statistical software will only do two of them for you. (steps 3, 4, & 5)
6. One-sample hypothesis tests in Stata
Test a single sample mean against a null hypothesis using data from the EG DIB evaluation.
7. Two-sample hypothesis tests in Stata
Test the equality of two sample means using data from the EG DIB evaluation.
Banner photo: William Playfair's bar chart showing the price of wheat over time, an early example of a bar chart. Accessed from https://www.lindahall.org/william-playfair/