Notes on the memrise paper #1

Okay, just rethinking, restarting, getting going again  :-)

I spent a few minutes this afternoon trying to work on the data from the Memrise pilot study. After looking at the numbers and playing around with a little bit, what I can see is unless I'm doing something completely wrong, which I could check and probably should, it looks like the differences aren't statistically significant between the control group and experimental group. So when thinking about reformulating is the three hypotheses:

• Hypothesis 1: A class that uses memrise as a vocabulary acquisition tool, will show improvement in their vocabulary knowledge.
• Hypothesis 2: A class that uses memrise will improve more than a class that is left to their own devices.
• Hypothesis 3:  Students that use memorized on a regular basis will improve to a greater extent than those who do not use that tool.
• Hypothesis 4: Students that use the memorized tools extensively will improve more than those who indicate substantial effort in traditional methods

 I think working through these hypotheses will get me started on the paper. But let me outline the what the data shows at this point:

• Hypothesis 1: Yes, statistically significant improvement
• Hypothesis 2:  no statistical significance, although greater improvement in the raw numbers
• Hypothesis 3:   need to figure out how to analyze that  [although step one is to decide how to differentiate the groups – I was thinking three. And then what statistical analysis would be used to see if there are differences between – I was thinking ANOVA]
• Hypothesis four: need to figure out how to analyze that, but steps in the process are same as above.

 Okay, that's a good start for right now. I should cook dinner :-).