[
You’re being visually misled by the Coeur d’Alene Press online poll.
Look at the bar chart below. It was captured from the Press online at 7 p.m. Friday, September 30. For purposes of illustration, I have removed the candidates’ names and the percentages the Press represents each candidate has received. The issue is how the Press presented information in a misleading way, not how any particular candidate rates in a novelty poll.
Visually at first glance, what might you conclude about the public’s support for each candidate?
The bar chart is apparently supposed to visually depict the difference in numbers of online poll votes received by Candidate Blue versus Candidate Green. A quick glance at just the bars might cause someone to conclude that Blue has overwhelming community support over Green.
The Press chart included information that Blue received 60% and Green received 40%. What did each receive a percentage of? We don’t know, because the Press doesn’t reveal if those percentages apply to the total number of votes cast. Neither does the Press include the exact number of votes cast for each candidate, Blue and Green.
We might guess that if Blue’s apparent 230 votes represents 60% of the total as the Press said, then the total votes cast must be approximately 383. So if there were, in fact, 383 votes cast, then Green’s 40% must represent approximately 153 votes cast. Is that the visual image and impression you get from just looking at the chart above from the Press‘s website?
No, of course it isn’t, because the Couer d’Alene Press chart has manipulated the Y axis. It doesn’t start at zero. In fact, we don’t know what it starts at. All we know is that the height of Candidate Green’s bar represents some figure less than 160.
Shown below is what a more accurately prepared chart might look like if we assume that there were a total of 383 votes cast and that Blue received 60% (230) and Green received 40% (153) of them.
With a more accurate visual representation of the data, suddenly the difference between Blue and Green isn’t visually as dramatic.
good eye Bill but I always do the math.
Comment by Ancientemplar — October 1, 2011 @ 4:35 pm
Ancientemplar,
Thank you. I suspect most readers who even looked at it would have caught it. My concern is that the Press even allowed such a deceptive technique to appear on its website. When I first saw the poll, my eye was drawn first to the graphics, not to the Y-axis numbers. Step two was to look at the numbers, but I do wonder how many people might have not proceeded to step two and could have been influenced merely by the apparent visual disparity.
Comment by Bill — October 1, 2011 @ 4:59 pm
That is pretty pathetic. Then again the Presses polls are all pretty pathetic so I’ll call this ‘candy stupid juvenile’. This type of data is best depicted in a pie chart where the total pie area represents 100% of the votes.
Comment by Wallypog — October 2, 2011 @ 5:43 am
Wallypog,
Pathetic is the correct word, even for a novelty poll. I’m trying to recall if it was in the 4th or 5th grade when I learned about charts and graphs and how it was important to be careful not to use them to visually deceive people. For an alleged newspaper to use this technique is insulting to the paper’s readers.
Comment by Bill — October 2, 2011 @ 6:27 am
I rather liked the look of it but, alas when looking at the actual numbers, not so much. Very wonky chart/vote app they are using. Doubt they have control of the graph but would be easy to find out.
Comment by adamgraves — October 2, 2011 @ 6:21 pm
O yes and http://www.adam-graves.com.
Comment by adamgraves — October 2, 2011 @ 6:22 pm
adamgraves,
Thanks for posting. To the extent the Press feels any sense of duty to accurately and completely present information, it needs to use online software that will allow accurate and complete presentation. Given the Press’s history of sloppy, incomplete, skewed reporting of news, it is no surprise that it employed a deceptive graphic display technique on a novelty poll. It would be even more regrettable if someone actually drew any conclusions about the candidates’ viability from the visual display.
Comment by Bill — October 2, 2011 @ 6:47 pm
Bill, i’ve got the viability. If i can get the vote out is yet to be seen. To hit what looks to be 300+ on the press poll i think is telling dont you?
Comment by adamgraves — October 2, 2011 @ 6:55 pm
adamgraves,
The only poll results that count are the ones on November 8, but as we’ve learned from the 2009 election contest lawsuit, even election night numbers are not necessarily reliable unless the election was administered strictly according to law.
My concern with the Press’s poll is that the Press is visually misrepresenting the data. It is hard to imagine that in this day and age anyone that made it past the sixth grade would be influenced by such an obvious deception, but then, the Press consistently demonstrates its low opinion of its readers.
Comment by Bill — October 2, 2011 @ 7:09 pm
“….the Press consistently demonstrates its low opinion of its readers.”
Why shouldn’t they? After all, they ARE part of the good ole boy association. It is the same contempt and disdain the mayor, city council, tony berns and good ole boys have always had for the local electorate. It can never change unless those WITHOUT hidden agendas are elected. There is a small glimmer of hope now that we have a new county clerk.
Comment by concerned citizen — October 3, 2011 @ 7:32 am
Bill, thanks for pointing this discrepancy out. People should read the numbers but we all know the visual is quick and, for most folks, it’s just a glance and move on. It’s a lot like headlines, where titles can slant the facts of an article but most people don’t read it so they take away the wrong impression.
Comment by mary — October 3, 2011 @ 9:53 am
Mary,
If we are expected to believe that the Press wants to provide accurate and complete information, then we have to believe Thompson and Patrick and Buley could and should have seen the same thing I did. They would not carelessly misrepresent information to their readers — or so one would think and expect. That leads me to conclude the paper’s visual misrepresentation was intentional, not accidental.
Comment by Bill — October 3, 2011 @ 1:07 pm