When it comes to research, I tend to use the following ...
30% of time to work out a hypothesis.
60% of time to work out how to test the hypothesis.
10% of time to test the hypothesis.
... I'd be curious on others experience.
30% of time to work out a hypothesis.
60% of time to work out how to test the hypothesis.
10% of time to test the hypothesis.
... I'd be curious on others experience.
i.e. about twice as much effort goes into working out how to test the hypothesis compared to developing the hypothesis and the actual test is usually a lot less than working out how to test.
Of course, it's a rule of thumb thing.
Of course, it's a rule of thumb thing.
X : The point.
Me : None, just a personal observation. The hard bit for me is not the hypothesis nor even running the test (including analysis) ... the really hard bit is always working out how I am going to test something. That's the thing I find diificult. Curious about others.
Me : None, just a personal observation. The hard bit for me is not the hypothesis nor even running the test (including analysis) ... the really hard bit is always working out how I am going to test something. That's the thing I find diificult. Curious about others.
I mention it because I see a lot of things written about hypothesis, testing and analysis ... but maybe I've just not been looking .... there seems to be a lack of things written about how to work out how to test something and for me, that's the hardest problem of all.
i.e. I have some things that I've not been able to work out how to test over decades of thinking about it.
I've got one at the moment which is related to the culture map and how to test it ... after many years. I still don't know how to effectively test it.
An interesting model - sure. Helps in discussions - yep. Some observations to support it - fine.
But how to test it?
An interesting model - sure. Helps in discussions - yep. Some observations to support it - fine.
But how to test it?
Roughly speaking, this is how I think of research ... I'm curious to others thoughts ...
X : Why does correlation lead to new research?
Me : You can always find correlations in any data if you go looking for them. That doesn't mean they are anything but random occurrence. So, you have to consider them as untested observations and start an entire new research project.
Me : You can always find correlations in any data if you go looking for them. That doesn't mean they are anything but random occurrence. So, you have to consider them as untested observations and start an entire new research project.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
