|
|
Re: OT: Summarization of Snit's 'Statistical Analysis' argument - Re:
Visualizing where to draw the standard deviation line
Posted:
Jun 19, 2012 12:12 PM
|
|
On Jun 19, 12:02 am, Snit <use...@gallopinginsanity.com> wrote:
> On 6/18/12 9:15 PM, in article
> 708534c3-a8f3-462b-924c-c6a41f686...@n16g2000vbn.googlegroups.com, "Onion
>
>
>
>
>
>
>
>
>
> Knight" <onionknight...@gmail.com> wrote:
> > On Jun 18, 10:45 pm, Steve Carroll <fretwiz...@gmail.com> wrote:
> >> On Jun 18, 3:11 pm,Snit<use...@gallopinginsanity.com> wrote:
>
> >>> What I did not do, of course, was show that this increase was based on the
> >>> *causative* factors I spoke of
>
> >> Bullsh*t... the facts show that you even called it an "overall
> >> claim" (which clearly places it in the realm of causation) that UI
> >> improvements **will** (this one word highlights your lie here) bring
> >> new users:
>
> >> "Now I have theories and reasons for this - and, frankly - it would
> >> not really be contrary to my overall claims that focus on the UI, if
> >> successful, will bring new users." -Snit
>
> >> That's clearly a 'cause and effect' statement if there ever was one.
> >> That you have the nerve to say you're speaking outside the realm of
> >> causation is absurd.
>
> >>> - I was very clear the data merely correlated
> >>> with the prediction I made based on those factors. You got confused between
> >>> causation and correlation... as did cc.
>
> >> You're confused and you're lying... and no one is falling for it.
>
> > You have no clue what the fuck you are talking about. Even your own
> > quotes show Snit making a prediction based on causes he thinks will
> > lead to a change. Then he showed that the change correlated with the
> > prediction. In 2011. He admitted it did not in 2012.
>
> > You are too fucking stupid to understand causation and correlation.
>
> As noted elsewhere: Steve confuses the concepts of showing a correlation,
> which I did
Wrong. But it matters not... your argument was buried with the
following... when you told a poster named kaba that you had no
statistical reason for creating your trend line, you said the reason
was *not* to have the line appropriately fit the data and it was *not*
for it to be useful with respect to that data in some way... kaba
wrote:
"I gather there are two questions here:
1) Is the trendline approriately fitted to that data?
2) Is the trendline useful in some way? " - kaba
Logical questions by kaba... and, of course, they highlight the 2 most
important reasons for even bothering to create a trend line at all.
You then dishonestly changed your claimed reason for creating your
trend line to this ridiculous reason:
"Not quite: the question was merely if the process of creating the
trend line was correct - did it follow the process of creating a
linear trend line that is supported by the build in "linear trend
line" properties of the program." - Snit
It's not hard at all to see what you tried to do as you abandoned all
statistical goals for your trend line.
|
|
