I would even go further. The biggest problem isn't just using p values and R^2. I think the biggest problem is that a lot of people didn't learn statistics properly.
Properly is a vague term. So what do I mean? Instead of obsessing with tons of techniques going back to the basic and actually learn how to do design studies, work with data, learn statistical reasoning and critical thinking.
I took quite a few courses in statistics because I liked it. But a lot other people – especially those that apply statistics – maybe take one or two courses in stats and then do research / studies. The results can be pretty terrible. In conclusion, more fundamentals and less icing on the cake.
While I agree with the premise that most people do not know enough statistics, there are several underlying problems to that.
When I was first encountered to statistical topics as an undergrad, the lecturer himself was not confident of the matter and taught a very bare subset that consisted of linear fitting, error propagation, means and standard deviation. Even back then I had the feeling that the equations provided were insufficient and contained a lot of things that were not motivated or explained.
Nowadays I can see why I was not introduced to statistics and probabilities back then as I was introduced to algebra, analysis and quantum mechanics. The field of statistics is complex, full of contradicting best-practices and analytically challenging. Let alone Bayes-vs.-Frequentist, etc. In a way I doubt that all researchers working with Poisson-distributed data once in their life as scientists can work through the details Poissonian statistics analytically.
Maybe an illustrative introduction would be more beneficial. I imagine people could perform statistical experiments before working with real data and experience first hand how misleading a small data subset can be for example, or how fundamentally data plots can change their face. Maybe then people would stop being overenthusiastic about their N=20 experiments.
Statistics are a tool that is quite easily abused. Look at the damage a regression can do in the hands of someone who only had a couple weeks to learn it at the tail end of a stats or econometrics class.
Stats are a field where it does pay to be a degree snob - it's best to have folks who studied it extensively, and then actually used it in the real world.
I completely agree on the package manager issue. I switched to OS X about 6 years ago. Before that I mainly used Arch Linux with wmii. My switch from Arch to OS X took about 6 months in which I slowly migrated. At first, it was pretty great but when I wanted to upgrade software or install some new one it was horrible. Back then I used macports and I can't remember all the problems. It was a downgrade from pacman (and apt). I gave up coding a bit because it wasn't so much fun anymore.
Lately, I started using a linux vps as my development environment and it's so much better. It's worth switching just for a good package management system.
Just a small tip which may ease the search for methods. The general term for "on the fly" learning is online learning [1].
The rest depends on your problem but there are often online variants of offline methods, e.g. when you work with Gaussian process regressions
I think these numbers are based on one blog post the company made[1].
They wrote:
"[...] defined that an app has to hold a position for at least 7 days to be considered as "ranked".
That was the case for 265,959 apps from the 18th of July 2012 to 25th of July 2012. To the remaining 410.023 apps off the ranks, we refer as app zombies, leading a life outside a prospering market."
and
"[...] in theory app zombies can still be downloaded, we concluded that an average zombie is getting zero to ten downloads a day, depending on the country."
I personally think that this is a neat idea but VoiceBunny isn't the ideal tool. You pay about $70 for 400 words which is about the length of a short-mid sized blog post. If you just cover 5 posts per day, you will need to invest about $10k per month in VoiceBunny. At this point, looking for your own voice actors is probably cheaper.
Hi! VoiceBunny co-founder here. Our marketplace is new and the median prices are coming down on a weekly basis as more talents compete for work. Also, you are free to set the prices you want to pay. For ideas like hearablog.com, we are also implementing a royalty-based system, where talents get paid as the recording makes money for our client.
Hmm, that scatterplot suggests a less straightforward relationship; the for/against dots look like they plausibly came from slightly different distributions, but not very much different ones.
Properly is a vague term. So what do I mean? Instead of obsessing with tons of techniques going back to the basic and actually learn how to do design studies, work with data, learn statistical reasoning and critical thinking.
I took quite a few courses in statistics because I liked it. But a lot other people – especially those that apply statistics – maybe take one or two courses in stats and then do research / studies. The results can be pretty terrible. In conclusion, more fundamentals and less icing on the cake.