Questionnaire Redux: Pro or Con - Results

[Frau_Blucher]

Okay, the results...

A bit broader sampling - 40 altogether including Mr. Average. The Average in fact hasn't changed - 3.4 as before. The Standard Deviation has however changed - it's gotten tighter from a 0.70 before to a 0.58. It could mean that the population is possibly slightly less diverse in its views than before or simply that cognizance of the poll's dynamics has possibly led to an either witting or unwitting moderation of answers. Regardless, here are the raw averages for all the questions - closer to 1 tends to mean more conservative while closer to 5 tends to mean more liberal (I excluded the Mac and Hart questions for the averages):

<img src="http://www.pbase.com/myirwin/image/94687568/original.jpg">

A SDev of 0.58 means that everyone from "Ti" to "Bt" are within a typical distribution of a population that would represent about 68% of that population if it were statistically "normal". Folks that are 2 SDevs out like "Ph" and beyond or "Ky" and beyond would typically represent about a 27% minority of the population. Folks beyond 2 SDevs like "Hu" and beyond and "Mk" and beyond are a somewhat extreme 4% of a "normal" distribution and are clearly not to be trusted members of proper society. [;)]

There is only a single person out past 3 SDevs (<1.6) who would be termed, um, [i]fairly unique[/i] - <0.2% of the population! But that's just averages, and averages don't come close to illuminating the individuality of peoples' choices. For that we need my favorite...CLUSTERS! [8D]

<img src="http://www.pbase.com/myirwin/image/94687590/original.jpg">

What you're seeing is a correlation clustering both of peoples' answers to the questions (the top clusters) as well as how the answers to the questions themselves tended to cluster together (the side clusters). More Red tends to mean more liberal and more Green means more conservative. Be sure to note that the closer you and another individual are connected at a node or [i]branch[/i] of a cluster tree, the more closely you are related to that individual's choices. Your proximity to a person laterally does not necessarily relate to your similarity as they may be connected to you far further up the tree and you may just have been drawn closer to someone else on this 2-dimensional representation of a 3-dimensional data set. For that same reason, there's also not a simple liberal or conservative trend as you scan from left to right either.

Having said that, it's clear there are 2 major clusters, A and B. To me, they sure seem to scream out [i]Liberal[/i] and [i]Conservative[/i]...in general of course! The A node seems to further have a possibly more liberal left A1 node (ahem) and a perhaps more conservative (libertarian??) right A2 node, but that's really splitting hairs. I don't see any such obvious correlation between the B1 and B2 nodes on the right side - both are very green but just with different patterning. In fact, two members, "Xx" and "Ex" are fairly red in spots, but something about their patterning puts them in this B node - perhaps a strong polarity in their answers with a few such answers just shifting them into this node.

On the questions side (including the Mac and Hart ones), there are 4 major nodes, Q1 through Q4, meaning that the questions within those nodes tended to be answered similarly. You can see that the Mac and Hart are interestingly in their own node, Q3, possibly meaning it's a good validation of the process that they didn't really fall into any political or life-view pattern. Cluster Q2 looks to be the more liberally answered questions while Q4 looks to be the more conservatively answered. Q1 looks fairly well mixed. Questions 4 & 5 about Affirmative Action were the most closely related at 0.92 correlation and reflecting broadly negative or ambivalent sentiments - not really a great guage since they're very similar questions. The closest supposedly "unrelated" questions were 19 and 38 - George Bush and the Iraq War. How about that?! No surprise that they're strongly and largely negative.

Some interesting individual points. One really could consider "Ti" to be a third cluster all to himself - there's a certain moderation or even a "dis-patterning" to his choices that might either make him a) quite an agreeable person or b) confoundingly obtuse! Elsewhere, the Average (asterisked) is most closely related to our dedicated mod Marie3 with a quite high 0.86 correlation (1.0 = identical). The other members in that sub-cluster - myself, KYYX, and Swampthing - are similar by a correlation of 0.58. That's quite different than what our averages might convey, but that might hint that some similarity reflects differing shades of similar issues rather than opposing positions on those issues. I think that could be the most interesting subtext to this that averages do not illuminate at all. It could also mean that one could possibly chose to wriggle absolute numbers and averages around but not necessarily shift their clustering, ie. general underlying beliefs...if of course we're being generally earnest about our answers. [;)]

Ya see anything else??

Edit: Reposting the chart with a different color map that seems to make the distinctions clearer. Blacks are 3s - down the middle:

<img src="http://www.pbase.com/myirwin/image/94692757/original.jpg">

[i_like_lectric_motors]

There I go again ... being shockingly average in my views. [:(]

Thanks for the awesome charts, Blir.




<a href="http://smg.photobucket.com/albums/v604/motors/?action=view¤t=andysahomo.jpg" target="_blank"><img src="http://img.photobucket.com/albums/v604/motors/andysahomo.jpg" border="0" alt="Photobucket"></a>

[Spit]

Blir , The charts and all the various clusters look great .The addition of the second chart with the added black clears up my understanding of how this all played out , except for one thing.What band ended up being #1 for the year 1982 ?

[Frau_Blucher]

[quote][i]Originally posted by Spit[/i]
<br>Blir , The charts and all the various clusters look great .The addition of the second chart with the added black clears up my understanding of how this all played out , except for one thing.What band ended up being #1 for the year 1982 ?
[/quote]
I'm pretty sure that was Selfish Cunt with "Fuck the Poor".

[club80]

Hey...is that Glok all the way on the right (and I do mean, "the right")?



We didn't sell a lot of records, but somehow we left an impression.
-Johnny Ramone-

[i_like_lectric_motors]

[quote][i]Originally posted by club80[/i]
<br>Hey...is that Glok all the way on the right (and I do mean, "the right")?



We didn't sell a lot of records, but somehow we left an impression.
-Johnny Ramone-
[/quote]

On this chart, right = liberal. [:D]

<a href="http://smg.photobucket.com/albums/v604/motors/?action=view¤t=andysahomo.jpg" target="_blank"><img src="http://img.photobucket.com/albums/v604/motors/andysahomo.jpg" border="0" alt="Photobucket"></a>

[Spit]

[quote][i]Originally posted by Blir[/i]
<br>[quote][i]Originally posted by Spit[/i]
<br>Blir , The charts and all the various clusters look great .The addition of the second chart with the added black clears up my understanding of how this all played out , except for one thing.What band ended up being #1 for the year 1982 ?
[/quote]
I'm pretty sure that was Selfish Cunt with "Fuck the Poor".
[/quote]
Now how did I miss that.If I'm not mistaken,that song was from their first album "Shellfish Cunt".
Joking aside , nice job on the charts.I don't no how the hell you figure this shit out . [8D]

[club80]

[quote][i]Originally posted by i_like_lectric_motors[/i]
<br>[quote][i]Originally posted by club80[/i]
<br>Hey...is that Glok all the way on the right (and I do mean, "the right")?



We didn't sell a lot of records, but somehow we left an impression.
-Johnny Ramone-
[/quote]

On this chart, right = liberal. [:D]

<a href="http://smg.photobucket.com/albums/v604/motors/?action=view¤t=andysahomo.jpg" target="_blank"><img src="http://img.photobucket.com/albums/v604/motors/andysahomo.jpg" border="0" alt="Photobucket"></a>
[/quote]


Oops! I should really read the opening paragraphs more often.



We didn't sell a lot of records, but somehow we left an impression.
-Johnny Ramone-

[glokstadt]

[quote][i]Originally posted by club80[/i]
<br>[quote][i]Originally posted by i_like_lectric_motors[/i]
<br>[quote][i]Originally posted by club80[/i]
<br>Hey...is that Glok all the way on the right (and I do mean, "the right")?



We didn't sell a lot of records, but somehow we left an impression.
-Johnny Ramone-
[/quote]

On this chart, right = liberal. [:D]

<a href="http://smg.photobucket.com/albums/v604/motors/?action=view¤t=andysahomo.jpg" target="_blank"><img src="http://img.photobucket.com/albums/v604/motors/andysahomo.jpg" border="0" alt="Photobucket"></a>
[/quote]


Oops! I should really read the opening paragraphs more often.



We didn't sell a lot of records, but somehow we left an impression.
-Johnny Ramone-
[/quote]

Sure looks that way, who did I beat out? MikeP? Mk = ?

At least I'm on the polar end of Kimiii... [:p]


<a href="http://photobucket.com" target="_blank"><img src="http://i97.photobucket.com/albums/l237/vespa_boy_2006/Smallerglokstadt.jpg" border="0" alt="Photo Sharing and Video Hosting at Photobucket"></a>

Visit my scooter club's website for Club Impact: www.clubimpact.us

[KYYX4ever]

Thanks for the interesting charts, Blir. [:)]
I don't know how you figure all this stuff out.