Good good good. You will be far more educated than most. I
didn't take statistics [so things like "R values" and stuff
don't make a darn bit of sense to me], but I learned it "on the
job" and as I was asked to change various business metrics when
compiling large amounts of data (I was processing diverse data
in the millions from a few dozen incompatible sources [and
making them compatible] via Excel 2000 and VBA using sheets and
codes that I wrote)...
...and I didn't like what it was doing to the results.
I could see what was happening. It's not that it wasn't accurate
per se... but it was a shell game - a "don't look at the Wizard
behind the curtain" thing.
Sometimes the only clue that you might be being led in a wrong
direction could be a SINGLE WORD in the report description....
if you even got that. --- == Ah! So i guessed right that it
was R values smile emoticon [I just looked it up - and yup,
that's what I was asked to do].
The thing is, it was 'mostly harmless'. it gave an air of
positivity to the sales force so they could do their jobs
enthusiastically, and it was for the higher ups to be smiling as
they walked down the halls.
It didn't really change things otherwise in a fundamental way
but it was the statistical equivalent of "spin" - all for
internal use; it wasn't jiggling accounting figures and stuff
like that. =- Sliding windows. You can do SO MUCH with sliding
windows of time and moving averages. == stock market is great
for that one. They zoom you into ups and down on 1 or 2 day
scales when the REALLY IMPORTANT information is hiding in the 9
month sliding windows.
Yeah and it is - it's sad and pathetic that it's funny and it
*is* funny - in a wry sort of "yeah that's just how it is, isn't
it?" sort of way.. Anything and everything can be (and is) a
tool for bullshit. tongue emoticon == One time it did.
I got called over to a "special project" one time, because I was
their Excel Guru. [Systems Analyst II - I didn't even KNOW what
that meant until a few months ago and looked it up... and i
haven't worked there in 14 years!]
I never got a college degree or special training.
Anyway, it was about drop samples. Sales people are supposed to
drop off samples to doctors. The doctors often give them away to
patients... many sell them them even though they're not supposed
to, but that wasn't our problem.
It's to encourage them to switch over, along with coupons for
unrelated things like baseball games and stuff.
Anyway, I had to gather ALL of the drops for every segment of
the country - every sales rep in every geographical area. This
was 2001. I had Excel to work with but it's my baby.
My job was to spot anomalies. it was not easy because the data
WAS A MESS. Wow, it was a mess. Huge mess. Lots of it came in
from scanned PAPERS... translating 1 to l and 0 to O - stuff
like that.
But I did it. Took a few months. Cleaned up the data. Matched it
up. I'm a perfectionist for accurate data, whatever they wanted
me to do with it.
So then I found the outliers. This guy here in Texas, his
numbers were -15. A reasonable range might be -3 - +3. Mind you
- I 'made up' "What's reasonable" based upon what "most people"
seemed to do (it was about 8000 people in around 2000 districts,
with 4 per district rotating in time).
All in all, there were a few dozen anomalies.
Each of them not only got FIRED but got ARRESTED for fraud. They
were selling them out of their garages or to chiropractors, or
just keeping them.
They were like Girl Scouts eating the cookies they were supposed
to sell, or Den Mothers stealing money out of the Girl Scout
cookie jar so the girls could get her a sandwhich.
Except... this is pharma. They got guys with black suits and
black glasses to contend with tongue emoticon == Yup. And every
phase has to be correct. Philosophy of data collection. Metrics.
Process of data collection. Accuracy of data. What's acceptable
to do with averaging? normalizing? median? How will you report
the data? What is the audience? What are their expectations?
What image do you want to present in the final reports? How will
it be received?
It's a very human endeavor because it goes from human to human
ultimately.
I love spotting BS in statistics, in Science, in Media, in
economics. I get a giddy feeling because it's like I can see
right through to the day they decided to fudge a little. Tweak.
Not entirely lying, but not entirely truthful.
And I can sit back in my chair here and go, "AH HA! CAUGHT YOU!,
you sneaky sneak tongue emoticon " == It's true. I just enjoy
searching for bias, whether it's statistics, or historical
accounts, or wherever. Bias doesn't make someone dishonest, as
bias is often unconscious. But it points to possible intent...
and the "game of determining intent" can be fun at times.
I stopped watching TV about 2.5 years ago (except Doctor Who..
and sometimes I'll sit with someone here at the house and see
"Desperate Housewives" or some such nonsense because it's what
they're watching - and then I get a kick out of interpreting
_that_ data. Brain's always going - it never shuts up).
I like going on the attack against MEMES. I hate when my
favorites are misquotes. Bad Einstein quote - even if I like it,
is still not something he said and so I try to track it down.
So many bad Tesla quotes that I had to compile all of his
writings in a text file just so search it for verification.
Strange things I do for fun. Dead can't speak for themselves so,
I try to help. Absurdism - i think it's the only way to cope
tongue emoticon I'm always happy for the same reason. The only
thing I don't like is when trusted people mislead with something
that matters to me. That's when I go on the attack.
I also accept my influence is limited but still that urge to
throtle the writer of a misleading headline... or a person who
writes something about history and just skips over their
ignorance of 500AD - 1500AD with a few glib sentences... is hard
to ignore tongue emoticon === Go for it. I support you. I
don't have - and never had - a problem with what i call
"reasonable religions". Be a good person. Don't be an asshole. I
think most major organized religions _generally_ have something
similar in that way.
Plus you get to sing or dance and join up with other sometimes
and it's free psychology that generally ok.
But the cults... well, that's something different. Their sales
tactics are barbaric in what they do to convince people.
I think that's why I don't like advertising, marketing,
politics, scientists or religious people entering into the
political realm... JOIN US IN THE FIGHT AGAINST... [some
misrepresented group].
Young people are especially suspectable too. Between the mid
teens to mid-late 20s, people (at least in the West, I don't
know if it's universal) seem to be easily brought into
ideologies of various kinds.
Causes. Fights. Cults, etc.
They want someone to have all the answers. There's always
someone ready to pop up and give it to them.
Once an ideology holds, you don't even know you're a part of it
yourself.
<br data-reactid=".cx.1:5:1:$replies9448879389000
