Statistics and Research Design: Conceptual
This course met the Research program component of the IT curriculum. I thought the instructor, Sky Huck, did a great job of introducing the course in a way that made it meaningful no matter what your intentions for it. Some of the students were going on to do doctoral work, and would need the training in how statistics are used in performing research, while other students would not be doing a thesis or dissertation, and would focus instead on how to critically read research findings.
Without getting too far afield into specifics (null and alternative hypotheses, rigor, internal/external validity, error types, etc), I’ll simply say that I learned a world of useful information in this course. Specifically, I learned how to interpret data: rather than lean on the researchers’ published conclusions in determining whether or not the data support a hypothesis, I can now look at the methodology itself to see whether or not the findings support the statement. I learned what to look out for – single versus two-tailed hypotheses, sample size, etc. With a knowledge of how inferential statistics work, it’s a lot easier to critically examine a study, and ask “well why was this omitted,” or “why are you using a lower confidence interval here, when a higher one would have caused you to fail to reject your null hypothesis, and then using a higher interval later when the data more strongly support your hypothesis?” In short, Sky’s class did a great job of being a survey course in statistics. Participants in the course were well on their way to being able to stand on their own two feet when it came to interpreting publications referencing information gained from inferential statistics.
There weren’t any papers to write in this class. We completed several projects, and took quizzes and exams. Here are the rules describing one of those projects, as well as a reporting form used to report the results of the exercise, to give you an idea of what sort of work we were doing.