Your Congressperson Is So Dumb, He Speaks at a Ninth-Grade Level
May 21, 2012 2:02 PM   Subscribe

The Sunlight Foundation, an organization dedicated to transparency in government, has analyzed 17 years of Congressional speech (per the Congressional Record) and found that the Flesch-Kincaid grade level of speech on the floor of the House and Senate has dropped by nearly a full year in the last decade.

While Republicans spoke at a slightly higher grade level than Democrats in the late 1990s, they plunged half a grade between 2006 and 2007, while the Democrats rose after retaking both houses of Congress. Democrats dropped almost as much in 2010, but both parties bounced up the next year for the first time since 2005.

The highest Flesch-Kincaid score of any member of Congress was Rep. Dan Lungren (R-California) at a whopping 20.5, crushing his nearest rival by nearly five years. In the Senate, highest honors went to Sen. Daniel Akaka (D-Hawaii) at 15.1. The lowest in Congress was Rep. Connie Mack at 6.7, with Sen. Rand Paul (R-Kentucky) at the bottom of the upper house at 8.0.
posted by Etrigan (38 comments total) 14 users marked this as a favorite
 
Hah! Idiocracy, here we come!
posted by Mental Wimp at 2:04 PM on May 21, 2012 [2 favorites]


I think we need a more sophisticated algorithm than the Flesch-Kincaid if we want to say something meaningful. Longer sentences and more-syllabic words do not equal smart or educated. (They might equal edumacated, though.)
posted by aught at 2:07 PM on May 21, 2012 [13 favorites]


I say that as someone who would be happy to see Rand Paul officially pegged as a dumbass.
posted by aught at 2:08 PM on May 21, 2012


Those scatterplots immediately made me think of the greatest science paper ever written.
posted by solotoro at 2:09 PM on May 21, 2012 [25 favorites]


But isn't a lower Flesch-Kincaid score considered a good thing? Why do we want legislative writing to be only readable to experts? Hemingway's score I believe was 8.0 too...
posted by Doleful Creature at 2:09 PM on May 21, 2012 [2 favorites]


Now do Metafilter!
posted by cmoj at 2:10 PM on May 21, 2012 [6 favorites]


Okay, this is my last word on the matter, I swear. When I heard this on NPR this morning, they mentioned that the average reading level of the general public was not far off this new Congressional level of discourse, and a part of me wondered whether that might not be a bad thing (to communicate effectively, I am thinking).
posted by aught at 2:11 PM on May 21, 2012 [2 favorites]


aught: "they mentioned that the average reading level of the general public was not far off this new Congressional level of discourse, and a part of me wondered whether that might not be a bad thing"

And the realities of an increasingly multilingual constituency, and lawmakers who are increasingly aware of their multilingual constituencies. I was disappointed that the article didn't really touch on that.
posted by roll truck roll at 2:13 PM on May 21, 2012


Martin Luther King’s “I Have a Dream” speech is at a 9.4 grade level.
I would rather listen to King speak for hours than some bore who knows a lot of long words. His score is still below the average score today. However, it is still an interesting study, and maybe worth comparing with speech grades in other English-speaking legislatures.
posted by Jehan at 2:16 PM on May 21, 2012


When asked about the Sunlight Foundation's conclusions Congressman Bruce Banner replied, "Puny analysis. Hulk smash."
posted by BitterOldPunk at 2:18 PM on May 21, 2012 [2 favorites]


You lie!
posted by miyabo at 2:19 PM on May 21, 2012 [3 favorites]


I would rather listen to King speak for hours than some bore who knows a lot of long words.

Sure, but the possibilities don't exist only at the extremes. The Gettysburg Address scores at a 13th-grade level.
posted by shakespeherian at 2:21 PM on May 21, 2012 [1 favorite]


We do not know if this is good or bad. There are many reasons that people may be using smaller words and shorter sentences in Congress. Some are good reasons. Some are bad reasons.

It is okay to use simpler words and sentences. This is good for many people. They can understand things better. Some people do not understand English well. They may be from other countries. But they are now Americans, so they have to understand people in Congress are saying.

But I think this is bad. I think that people are saying simpler things in Congress because they are saying stupid things, and because they are only saying things that people like. People in Congress should say things that are true. But true things are hard to understand sometimes. So they say things that people like. They do not say things that are true.

This is bad for America. The world is hard sometimes to understand. And saying simple things make people like you. True things make people sad. It is like this on MetaFilter too.

So I understand why people in Congress say simple things that are not true. But it is bad.
posted by alasdair at 2:23 PM on May 21, 2012 [21 favorites]


When asked about the Sunlight Foundation's conclusions Congressman Bruce Banner replied, "Puny analysis. Hulk smash."

Well, aint' he unglamourays!
posted by JHarris at 2:24 PM on May 21, 2012


Here is all you need to know about Flesch-Kincaid scores:

"But lo, what light from yonder window breaks" gets rated as Grade 2.

"Everybody dance!" gets rated as Grade 14.

It is a really profoundly stupid way to measure reading level. This study tells us jack shit about the level of discourse in Congress, or even about the grade level that congressmen are speaking or writing to, because the metric they're using is just fundamentally broken.
posted by nebulawindphone at 2:31 PM on May 21, 2012 [14 favorites]


Those scatterplots immediately made me think of the greatest science paper ever written.

That was just brilliant!
posted by Mental Wimp at 2:31 PM on May 21, 2012


Most congresspeople who speak at a ninth-grade level are actually overperforming, as they only think at a second -grade level.
posted by Benny Andajetz at 2:32 PM on May 21, 2012


Now do Metafilter!

Plastic.com had this feature, or something like it. I remember my posts were always at the level of Sports Illustrated or Time. Now and then they achieved Atlantic Monthly status.
posted by Calzephyr at 2:33 PM on May 21, 2012


It is a really profoundly stupid way to measure reading level.

I'm not disagreeing with this, but I think using a single sentence to get a measurement is probably prone to inaccuracies of how the metric is supposed to function.
posted by shakespeherian at 2:33 PM on May 21, 2012


Metafilter: Now do Metafilter!
posted by GenjiandProust at 2:42 PM on May 21, 2012 [3 favorites]


I'm not disagreeing with this, but I think using a single sentence to get a measurement is probably prone to inaccuracies of how the metric is supposed to function.

Yeah, no, that's true. You're actually not supposed to use it on passages that short at all.
posted by nebulawindphone at 2:49 PM on May 21, 2012


Shorter, pithier phrases make for better soundbites in the media; more quotable/tweetable/accessible. Oftentimes there's much more you can do stylistically (that is, in the delivery) of a shorter, simpler phrase and it will then be more reproducible and memorable (ie., catchy).
posted by iamkimiam at 2:50 PM on May 21, 2012 [1 favorite]


Now do Metafilter!

MS Word and its ilk have this sort of functionality built in. I usually use it to rewrite my software documentation based on the "foot level" I need to convey. I use the formula:

(Fk)/log(Fl) = Target Grade

Where Fk is the Flesch-Kincaid score, Fl is the "foot level", and log is the base 10 logarithm.

Example:

We have a document deliverable to someone in management who expresses constant amazement at "This new iPad thingee" about the innermost working of our software. Can you take the design document and make read at the 10,000 foot level?

Fk(design document) = 15.5
Fl = 10,000

(15.5)/1og(10,000) = 3.875

Open up the document and re factor entire paragraphs into sentences.

(I know this is all totally bunk, just felt like making a joke about explaining complicated things to people who have no express interest in understanding how it really works.)
posted by The Power Nap at 2:53 PM on May 21, 2012


It would be more interesting, I think, to try to correlate F-K scores with electability. You could measure the deviation in individual races from the generic D/R ballot in any given election cycle and then see if there was a positive correlation between that deviation and the difference in F-K scores for major speeches by opposing candidates.

No matter what the outcome, you'd learn a lot more about the persuasive efficacy of political rhetoric than anything this headline-grabbing study can tell you.
posted by R. Schlock at 3:03 PM on May 21, 2012


Also, lots of stuff that's difficult to compare here...writing is very stylistically different than speech. Speech genres are going to also vary by design (conversations, statements, debates, speeches, presentations...all very different). Audience matters too...people accommodate their speech based on who they're talking to and what the goal is. And what is the data going into this? Are the disfluencies, contractions, colloquialisms and such transcribed? Has this been done consistently?

That's all not to mention dialect and accent differences and the many ways to interpret the results.
posted by iamkimiam at 3:03 PM on May 21, 2012


Those scatterplots immediately made me think of the greatest science paper ever written.

Heh. I actually did something like Figure 1 in a physics lab. The professor asked me how I came up with the curve, to which I replied with 10 seconds of awkward silence, because "I used the MS Word drawing tool" seemed like an even worse response.

Having read dozens of papers since then, I've learned that you can get away with this as long as you say the curve is there to "guide the eye."
posted by dirigibleman at 3:11 PM on May 21, 2012


According to read-able.com this FPP and accompanying comments are currently getting a 6 on the F-K scale.
posted by Doleful Creature at 3:20 PM on May 21, 2012


I know from working on consent forms and having to make them readable at the 8th-grade level that these scoring systems are poor measures of comprehensibility, especially when you're trying to convey something complex. You get dinged for "long" sentences, which means that compound sentences and dependent clauses cost you, yet complexity almost demands use of them. I've not closely examined the research behind these simplistic measures, but I suspect it isn't all that robust.
posted by Mental Wimp at 5:03 PM on May 21, 2012


According to Read-able.com, the Gettysburg Address scores 6.9 and Richard Feynman's Nobel Lecture scores 10.4.

I guess we can conclude that the message of the Gettysburg Address is rather easier to understand than the gist of quantum electrodynamics, and that Lincoln and Feynman were rather good at making themselves understood.
posted by philipy at 6:15 PM on May 21, 2012


Do they have an equivalent of Fleisch-Kincaid in Chinese, where basically every word is two syllables?
posted by miyabo at 9:01 PM on May 21, 2012


You just have to look at the trend in the Inaugural Address of successive presidents.

I mean, when George Washington became president, his Inaugural Address began AMONG the vicissitudes incident to life no event could have filled me with greater anxieties than that of which the notification was transmitted by your order, and received on the 14th day of the present month. On the one hand, I was summoned by my country, whose voice I can never hear but with veneration and love, from a retreat which I had chosen with the fondest predilection, and, in my flattering hopes, with an immutable decision, as the asylum of my declining years—a retreat which was rendered every day more necessary as well as more dear to me by the addition of habit to inclination, and of frequent interruptions in my health to the gradual waste committed on it by time.

When George W. Bush became president, his Inaugural Address began, if memory serves, President Clinton, distinguished guests and my fellow citizens, the peaceful transfer of authority is rare in history, yet common in our country. But I win. The rest of you can suck it. Now watch this drive.
posted by twoleftfeet at 9:51 PM on May 21, 2012


"Everybody dance!" gets rated as Grade 14.

Dammit! My high school only went to Grade 13. I knew I missed out on something awesome!
posted by srboisvert at 2:14 AM on May 22, 2012


There's a difference between being stupid and addressing people you think are stupid, and it's an important difference.
posted by Legomancer at 5:30 AM on May 22, 2012 [1 favorite]


You can actually get arbitrarily high Flesch-Kincaid scores if you're willing to make up arbitrarily long words. A text that just says "supercallifragilisticexpialidocious" (horribly abusing the system again to make a point; bear with me) has a grade level of 138 and a reading ease score of -894.
posted by nebulawindphone at 10:16 AM on May 22, 2012


Cannot resist the opportunity to cite my favorite Winston Churchill quote: "Short words are best, and the old words, when short, are best of all."

(I figured I'd better double check that it was really Churchill before posting it Metafilter, because Churchill tends to attract quote attributions. And the only website I can find that actually cites a source for it ruins it by adding the two syllable words "broadly speaking" to the beginning!)
posted by OnceUponATime at 4:13 AM on May 23, 2012 [1 favorite]


"Short words are best, and the old words, when short, are best of all."

Or, never use a big word when a diminutive word will do.
posted by Mental Wimp at 11:39 AM on May 23, 2012


Further explanation of why this whole story is nonsense.
posted by iamkimiam at 11:46 PM on May 27, 2012 [2 favorites]


As the link says:

In fact, for readability — the purpose the tests were developed for — a lower score is always better, assuming the same information is conveyed.

If anything here tells us anything about the depressing state of political debate it's the fact that people are so eager to take a rough measure readability where lower is always better and convince themselves it's actually a worthwhile metric of smartness where higher is better.
posted by philipy at 9:57 AM on May 28, 2012


« Older The Man Who Thought He Was King   |   We gone be up here all day. We have a long wait. Newer »


This thread has been archived and is closed to new comments