In this video I discuss Type 1 error control, or how to make decisions based on studies without fooling yourself in the long run, within a Neyman-Pearson approach to statistical inferences. This is video 3.1 in my MOOC on Coursera on "Improving Your Statistical Inferences"

The quite recent, but already classic paper illustrating the
problems of inflated Type 1 error rates is:
Simmons, J. P.,
Nelson, L. D., & Simonsohn, U. (2011). False-Positive Psychology:
Undisclosed Flexibility in Data Collection and Analysis Allows Presenting
Anything as Significant. Psychological Science, 22(11), 1359–1366.
http://doi.org/10.1177/0956797611417632
A good discussion of Type 1 error control is provided by:
Rutherford, A.
(2011). ANOVA and ANCOVA: a GLM approach (2nd ed). Hoboken, N.J: Wiley.
Sections 3.6-3.10
If you want to use sequential analyses and benefit from
their efficiency, without inflating type 1 error rates, I’ve written an
accessible introduction:
Lakens, D. (2014).
Performing high-powered studies efficiently with sequential analyses: Sequential
analyses. European Journal of Social Psychology, 44(7), 701–710.
http://doi.org/10.1002/ejsp.2023
A discussion of a way to directly control the false
discovery rate can be found in:
Benjamini, Y., &
Hochberg, Y. (1995). Controlling the false discovery rate: a practical and
powerful approach to multiple testing. Journal of the Royal Statistical
Society. Series B (Methodological), 289–300.