The Priorities of People's Dissatisfaction
Recent Gallup pollingshows a couple of notable points: 1) A clear majority of Americans (63 percent) say they are dissatisfied with the state of the nation as it pertains to government surveillance of U.S. citizens and 2) It ranks low on the list of priorities they'd like to see Congress and the president address in the next year. Just 42 percent say it is an extremely or very important priority placing it lower than more than a dozen other matters.
Only the state of the nation regarding the economy, poverty and homelessness are held in lower regard by the public than surveillance of Americans. And only abortion, race relations and acceptance of gays and lesbians are seen as lower priorities for Congress and Obama to deal with.
http://www.washingtonpost.com/blogs/the-fix/wp/2014/01/17/how-the-public-is-unhappy-with-surveillance-but-unmoved-in-one-chart/?tid=up_next
Tags
Who is online
70 visitors

I have long suspected that the state of public understanding of complex math concepts was extremely low . Here we have clearly demonstrated that a large poll taking organization , Gallup , is included in that ignorance . Apparently they have fired any data analysts who would be capable of understanding even something as basic as correlation .
What else could explain the bass-ackwards organization of that chart ?
Petey
What exactly is your complaint abut the chart
It is a little busy but the four quadrants which highlight the combination of the levels of importance and dissatisfaction is pretty easy to understand
All that being said, it is the data that is important to me not how it might be presented in the chart
Appreciate your perspective on the Gallup polling organization
When a chart is organized to display correlation it is not just the 4 quadrants that provide useful info , it is the entire chart . For example , what conclusion is suggested by the position of the "Economy" topic ?
I disagree
I have seen this type of display in any number of charts for sales, marketing and business risks over the years it is quite valuablewhen determining a course of action
As the "economy: topic it is quite clear that it is very important to those polled and that they are very dissatisfied with how it is being handled.
It is not the "type of display" I am objecting to . It is the bass-ackwards organization of the variable on the vertical axis . If it were reversed to show [in effect ] high satisfaction then the chart would be much easier to interpret .
All points on the main diagonal would be interpreted as being dealt with appropriately to their public perception .
But all point far off the main diagonal would demonstrate problems of one sort or another . Then only 2 quadrants need to be further examined .
That must have been very traumatic for you . Have you considered suing your school district ?
Petey
Okay
Potato potatoe I guess
I didn't find it confusing but I guess it could be clearer
But as I said it is the data that is important not the preentaion
Always appreciate little humor
Thanks BF
It is very unclear . There needs to be a configuration on the graph that reveals well correlated data points . Can you find such a configuration ?
If you connect all the dots, a ghostly image will appear.
Yours might be ghostly . Mine is merely conceptual ...
I suppose my view would cover all of the above, my dissatisfaction lies with what wehave allowed ourselves to become as a people, society and most of all what we have allowed our government (or lack there of it) to morph into.
We should all be ashamed, and yet still we stand by and wring our hands.
Well, yes and no. Because IMO there are basically two things you can use this chart for:
1. To discover the general "trend"-- to get a quick idea as to correlation between the two variables graphed here (degree of satisfaction and degree of priority.) Whether or not the two variable are correlated at all (either negatively or positively) and if there is any correlation, is it a high degree of correlation-- or not?
2. Aside from the general trend (degree of correlation overall between the two variables using all cases listed), you can get 2 pieces of information for any specific case (the economy, for example-- according to this data, there is a high degree of dissatisfaction and it is considered a high priority).
The advantage of this sort of chart is that you can quickly & easily get a rough idea of these things.
However, for more accuracy (for example, to determine the exact correlation coefficient between satisfaction & priority overall) you would have to resort to using the appropriate mathematical formula.More accurate but more time consuming...
Simple. Several conclusions actually come immediately to mind (taking more time I'm sure others could be found):
1. Probably the most important: In terms of the two variables at play here (satisfaction & priority), "Economy" is considered to be a very high priority and also a very high degree of dissatisfaction.
2. Of all the issues graphed here, the highest priority is given to "economy".(Clinton was right...)
3. The issue(s) with the most dissatisfaction are "economy" and "poverty/homelessness"
4. People have the same amount of dissatisfaction with "poverty/homelessness" as they do with "economy". (Perhaps not surprising as "poverty/homelessness"issues might be viewed as a subset "Economy")
5. This is a bit fuzzier, but the issues of which people have views (re: these two variables) that are most different from their views re: "Economy" are "Race Relations" & "Acceptance of Gays/lesbians".
I think you're missing my point . The issue is not finding a correlation coefficient . It is having an easy to interpret graph . You can explain a point like Economy without that but how about a point like "Govt Surveillance" ?
Question: In which quadrant of the graph does your point fall? How important would most people polled think your point is? How high a priority would they give it. And, what correlation does your point have with any other point?
Jus' kiddin'!
But seriously:
When looking at a graph like this, I want to get all the information out of it that I can. When you have several topics (Economy, Race Relations, etc) and then the graph shows, how they stack up re: 2 variables (in this case "Importance/Priority" and Degree of Satisfaction,) it seems to me that one thing that you might want to look at when there are two variables is the degree, if of correlation between the two variables. However, that is not the only thing I would want to look at.
Of course people have different interests, but one of the things that interests me greatly are relationships between things, especially between variables. The correlations between the two. (And then if there's a relationship of cause and effect or not, etc. ) Part of the reason that I am so interested in correlations is that this is extremely important in two of my main interests-- studying the stock market, and Astrology).
There are potentially several things one could learn from this graph-- different people may be interested in different things.
I don't see a problem with that. You would poll people on that just as you would with the other issues-- than place a dot for it on the graph. The position of the dot would be determined by how important people think that is, and how satisfied/dissatisfied they are with it. Just like all the other topics.
This is why I advocate simplifying the graph by a variable transform . You would be left with a graphical representation of correlation on the main diagonal ... simple . And that is where "Govt Surveillance" would end up ,
right near the main diagonal . As things are now you would have to look at the other diagonal which shows negative correlation .
Looking at the current graph, I see that there is a band (or "channel") emcompising most points that goes from lower left to upper right. Indicating a correlation.
Its not a perfect straight line (which would indicate perfect correlation). And its not a thin band which would indicate a higher correlation. So that means to me that the correlation isn't extremely high, but it does exist.
Three points are outside this channel: Military, Terrorism, & Gov't Surveillance. This indicates that the correlation isn't very high, but OTOH since there are only these three outside the band, correlation isn't extremely low either.
The chart doesn't actually show two lines deliniating the upper & lower limits of the band, but you can see it if you look (most points in a somewhat wide band from lower left to upper right).
If you cover up the three points mentioned, the band becomes more obvious.
Indicating that a correlation between "Importance/Priority" and "Dissatisfaction does indeed exist.
Petey
Like I said I did not find it confusing, you did and maybe others did as well and that is unfortunate
But as I said get past the fact that you do not like the format, it is the data presented that is important even if you have to did and think about it a bit to understand it
The data is showing us what is important to the people and what the people are dissatisfied with and I think that is important
Krish
Thanks a lot for the perspective
It seems that I am not alone in being able to see what the data means and to feel that it is important
A P
Never thought of that what image did you find?
The ghost of Joe McCarthy or of NSA Directors past?
luther
Thanks a lot for the perspective and I agree 100% that "we" have allowed these things to happen.
Petey
Can you put together a chart, based on the data presented here, in the format you advocate and share it with us?
Perhaps then it would be easier for me to understand your fixation on the format rather than the data.
Okay. When I first glanced at the chart I saw it's meaning clearly. Then when I noticed others deciding that there were problems in the display, I looked at it a little more in detail.
From what I see the chart is nothing more than a standard x-y Cartesian coordinate layout which would place the (-x,-y) or lowest importance, lowest dissatisfaction in the third quadrant. The inverse of that would be located in the first quadrant which it is.
As far as the location of the topics on the chart, they look to be in a reasonable order. Thus I would say that Robert has found a chart that depicts things in by importance and satisfaction level rather well.
FLYNAVY1
Appreciate the feedback
I stumbled on the article and the chart, so I can take little credit for picking out a good format, but I too found it fairly easy to understand and isolate the issues by importance to the people and dissatisfaction level with the issue
I also found that the issues which seem to be the most critical to the nation were in the top right quadrant (high importance/high dissatisfaction) - poverty, taxes, economy (jobs), healthcare, education and social security/medicare.
It would seem that those are the issues that out elected leaders should be focused on
BadFish
I am glad you enjoyed
I tend to like the simple, directpresentation of information as well
Now that we are all clear on this.... back to the show!
Is that what you want to display ? That means the important issues are NOT being addressed .
From my POV I would find the display more meaningful if it displayed the inverse of the dissatisfaction scale . If that showed a lack of correlation [it does] then it demonstrates a general failure of the political system to deal with important issues .
I don't have software that outputs graphs in jpeg or gif format . That seems to be the only way to get it to display on NT . Do you know another way to make a graph display here ?
You can see what the individual trees mean but not the entire forest .
Petey
Actually the trees and the forest are visible (using your metaphor) in this chart even though the format seems to have thrown you a little
No worries and like I said if you create a better formatted chart with the data from this one, please share it with us and explain why it is so much better
Have a good day
Fine . So what does the visible forest tell you ?
Petey
What it seems have told everyone else, but you, the level of importance and dissatisfaction the people place on the issues of the day
I have not seen a single comment made that goes towards the general level and its correlation . But by all means point out where that conclusion has been drawn ...
Petey
No one else has whined and complained about the format of the chart
Others have even discussed the priorities and what they mean
There is no value in continuing as you have given up on the data because you do not like the way it is portrayed.
I accept that I bid you good evening
In a previous comment in this discussion, I have said that looking at the graph, I definitely see a correlation (and, in addition, that it wasn't a perfect correlation, not an extremely high correlation coefficient-- but strong enough that there is a significant correlation).
Of course our conversation re: the merits (or lack thereof) of the graph has had the effect of derailing the more relevant conversation which is about the actual contents of the chart-- what it all means, what is the significance of the data presented (as opposed to the way in which it was presented).
"Forest through the trees"..that sort of thing.
Krishna
I thank you for that point and for your perspective on this thread
(Bolding added by RIO
odd, no mention that according to a Gallup Poll conducted January 14 through 16, Obamas approval rating has slipped to 39 percent. A majority of those polled said they are dissatisfied with the nations president. The poll included voters of all races.
JoAnne
Perhaps because it has little or nothing to do with the topic of the thread