ANCOVA for pretest-posttest designs

[drbenvincent]

Add ANCOVA as a new capability of the package

• Glossary: add post-test only and pretest-posttest variants to the nonequivalent group design section
  • Flesh out with more detail?
• Get an example up and running with new PrePostNEGD class. Use synthetic data.
• Plot the data in both forms next to each other, so it's easier to see how data from pretest-posttest NEGD can be analysed by ANCOVA or DID.
• Add similar sklearn example notebook
• OPTIONAL: Update the example with a real dataset from a published paper. Probably best to follow up with a novel example, or to replace the initial example
• Add more explanation to the example notebook.
• Add to the README + index.rst
• Add data as a package dataset
  • Add that to the tests
• Add to integration tests

[drbenvincent]

Possible studies:

• Cummins, S., Petticrew, M., Higgins, C., Findlay, A., & Sparks, L. (2005). Large scale food retailing as an intervention for diet and health: quasi-experimental evaluation of a natural experiment. Journal of Epidemiology & Community Health, 59(12), 1035-1040.
• Huang, E. Y., Lin, S. W., & Lin, S. C. (2011). A quasi-experiment approach to study the effect of e-mail management training. Computers in Human Behavior, 27(1), 522-531.