P1aMjRcc

1The Case Against Education:
2A Professional Student Explains Why Our Education
3System is a Big Waste of Time and Money
4Bryan Caplan
5[alternate title: The Magic of Education:
6Why The Real World Pays You to Waste Your Time In the Ivory Tower]
7[cover idea: kids sleeping at their desks]
8[cover idea: underwater basket-weaving]
9© Bryan Caplan 2015
10Please do not circulate without permission. 
11Chapter 7: Nourishing Mother: Is Education Good for the Soul?
12As you can see, I have memorized this utterly useless piece of information long enough to
13pass a test question. I now intend to forget it forever. You’ve taught me nothing except how
14to cynically manipulate the system. Congratulations.
15-Calvin and Hobbes
16College graduates often proudly name-drop their alma mater. Few realize, however, that the
17phrase contains a worldview. In Latin, “alma mater” means “nourishing mother.” A rich metaphor.
18A nourishing mother doesn’t merely teach you practical skills or help you land a well-paid job.
19She nurtures your whole person. She teaches you right from wrong. She shows you the magic
20of life. A nourishing mother speaks from her heart and fills your soul with goodness.
21Most economists are dedicated fans of education, but they still roll their eyes at such lofty words.
22They push education because they think it has a high social return, not because they think it good
23for the soul… whatever that means. Once in a while, philosophical economists muse that
24education might be a “merit good” – a product with value above and beyond customers’
25willingness to pay.i
26 But such musings rarely sink in with their pragmatic peers.
27Economists’ philistine ways expose them to a sharp humanist critique. An old-school fan of liberal
28arts education could chide, “You base your support for education on its high social return. But
29Caplan convincingly shows that this ‘social return’ is poor at best. Those who live by the sword,
30die by the sword.” The humanist could then turn to me and add, “Stop smirking. You may have
31beaten your fellow economists at their own game, but that game isn’t worth playing. The humanist
32tradition continues to offer the most meaningful perspective: Education really is good for the soul.
33Laugh all you want, but ideas and culture matter more than dollars and cents.”
34In all sincerity, I’m not laughing. For all my iconoclasm, I love ideas and culture. “Impractical”
35ideas and “uncommercial” culture are my life. The journey from ignorance to enlightenment
36moves me. Consider Malcolm X’s spell-binding story about teaching himself to read in prison:
37I saw that the best thing I could do was get hold of a dictionary—to study, to learn some
38words...
39I spent two days just riffling uncertainly through the dictionary’s pages. I’d never realized
40so many words existed! I didn’t know which words I needed to learn. Finally, just to start
41some kind of action, I began copying…
42I woke up the next morning, thinking about those words – immensely proud to realize that
43not only had I written so much at one time, but I’d written words that I never knew were in
44the world…
45I was so fascinated that I went on – I copied the dictionary’s next page. And the same
46experience came when I studied that. With every succeeding page, I also learned of people
47and places and events from history. Actually the dictionary is like a miniature encyclopedia.
48Finally the dictionary’s A section had filled a whole tablet – and I went on into the B’s. That
49was the way I started copying what eventually became the entire dictionary…
50I suppose it was inevitable that as my word-base broadened, I could for the first time pick
51up a book and read and now begin to understand what the book was saying. Anyone who
52has read a great deal can imagine the new world that opened.ii
53Unlike most economists, moreover, I think the value of ideas and culture reflects more than my
54personal tastes. Philosophy and opera really are merit goods. Reading David Hume’s Enquiry
55Concerning Human Understanding is intrinsically better than reading E.L. James’ Fifty Shades of 
56Grey. Hearing Richard Wagner’s Tristan und Isolde is intrinsically better than hearing Toby
57Keith’s White Trash With Money. Humanist fans of education grasp profundities that my fellow
58economists carelessly dismiss.
59Old-school humanists nevertheless overstate their case. Education definitely can be good for the
60soul. But that hardly shows that actually-existing education achieves this noble end. In practice,
61education may turn out to be a neglectful or abusive mother rather than a nourishing one.
62Meritorious Education
63The Master said, In old days men studied for the sake of self-improvement; nowadays
64men study in order to impress other people.
65-Confuciusiii
66Practical education has no need to be uplifting. As long as students graduate with skills to apply
67outside the classroom, even academic drudgery serves a worthwhile purpose. When you have
68instrumental value, you don’t need intrinsic value. To be a worthwhile end-in-itself, though, the
69educational process must meet higher standards. To plausibly qualify as a merit good, education
70needs three ingredients.
71The first ingredient: Worthy content. Learning about great ideas and glorious culture uplifts the
72soul. Learning about half-baked ideas and so-so culture, not so much. While the liberal arts
73tradition wisely hails the value of grappling with error, this only holds for well-argued, thoughtful
74errors. 
75The second ingredient: Skillful pedagogy. Learning from enthusiastic teachers who have
76mastered their subjects uplifts the soul. Learning from uninspired teachers who parrot the
77textbook, not so much. Mediocre instruction is tolerable for practical training, but worthless for
78intellectual or artistic inspiration.
79The third ingredient: Eager students. Sharing great ideas and glorious culture with students who
80find them fascinating uplifts their souls. Force-feeding great ideas and glorious culture to students
81who couldn’t care less, not so much. Indeed, the charade degrades students, teachers, and the
82subjects themselves. Opera is divine, but herding country music fans into opera houses is not
83only futile, but cruel. Many educators assuage their consciences by insisting that youthful forcefeeding
84will in time blossom into mature fascination. Even if they’re right, the force-feeding is a
85regrettable pathway to the merit good of mature fascination, not a merit good in itself.
86How does actually-existing education measure up against these standards of merit? As long as
87you’ve had a vaguely typical education, you already know the answer. The content of education
88is mixed at best: Pockets of greatness, surrounded by insipid busywork. The pedagogy is poor:
89Most teachers are, to be frank, pretty damn boring. The students are worse: No matter how great
90their teachers, few yearn for the life of the mind. Private education is arguably slightly better, but
91it’s cut from the same cloth as public education. Harvard University’s Steven Pinker sadly reports
92that the best students in the world yawn at the best teachers in the world:
93A few weeks into every semester, I face a lecture hall that is half-empty, despite the fact
94that I am repeatedly voted a Harvard Yearbook Favorite Professor, that the lectures are
95not video-recorded, and that they are the only source of certain material that will be on the
96exam. I don’t take it personally; it’s common knowledge that Harvard students stay away
97from lectures in droves, burning a fifty-dollar bill from their parents’ wallets every time they
98do.iv
99When I judge our education hollow, it is isn’t just my opinion; it’s almost certainly your opinion,
100too. Honestly, how many educators do you find fascinating? Do you really think kids find them
101any more fascinating than you do? Even if you refuse to voice the unseemly answers, your
102behavior speaks unvarnished truth. However big a nerd you are, you don’t nourish your soul by
103watching YouTube videos of average teachers. No one does. The empirics on student boredom
104we’ve covered underscore the obvious: Stimulating education is the exception that proves the
105rule.
106v
107
108While I broadly accept the humanist critique of philistine economists, economists have a weighty
109comeback: Cost matters. Suppose opera really is good for the soul, and education genuinely
110promotes love of opera. These facts mean little until we know the per-capita cost of conversion.
111“Exposing a single person to opera is worth any expense” is childish bravado. Cost matters
112whenever you spend your own money. How could cost cease to matter when you spend taxpayer
113or donor money? Every dollar spent is a dollar that could have been repurposed.
114This economic caveat is now more compelling than ever. Ours is an age of science fiction. Almost
115everyone in rich countries – and about half the earth’s population – can access machines that
116answer virtually any question and teach virtually every subject.vi
117 The internet doesn’t merely
118satisfy our curiosities; it connects us to global communities that share our curiosities. These
119global communities are more than clubs of novices; they include many of the greatest teachers
120on the planet.vii
121 The internet provides not just stream-of-consciousness enlightenment, but
122outstanding formal coursework. Khan Academy is the only most famous example. To top it all
123off, this ceaseless intellectual feast is, with rare exceptions, free of charge. If education is a merit
124good, the internet is the Merit Machine.
125On reflection, this Merit Machine is swiftly making traditional humanist education policy obsolete.
126Once everyone can endlessly enrich their souls for free, government subsidies for enrichment
127forfeit their rationale. The objection, “But most people don’t use the internet for spiritual
128enrichment” is actually a damaging admission that eager students are few and far between.
129Subsidized education’s real goal isn’t to make ideas and culture accessible to anyone who’s
130interested, but to make them mandatory for everyone who isn’t interested.
131The rise of the internet also undercuts the Machiavellian line that intellectual force-feeding
132ultimately blossoms into sincere appreciation. Today’s adults are the product of over a decade
133of mandatory exposure to abstract ideas and high culture. If educational force-feeding worked
134reliably, most educated adults would adore these nerdy realms – and eagerly tap the internet to
135explore them. To understate, they rarely do. “Kim Kashdashian” gets about twenty times as
136many Google hits as “Richard Wagner” and about two hundred times as many as “David Hume.”
137Insisting that “the end justifies the means” is comical when progress toward the end is nearinvisible.
138A philistine could summarily dismiss this logic: “Of course adults rarely bother studying ideas and
139culture online. There’s no money in it.” But this chapter isn’t aimed at philistines. It’s aimed at
140anyone who holds that actually-existing education is good for the soul. Once you embrace this
141stance, the rise of the internet has two unsettling lessons. First: The humanist case for education
142subsidies is flimsy today because the internet makes enlightenment practically free. Second: The
143humanist case for education subsidies was flimsy all along because the internet shows that low
144consumption of ideas and culture stems from apathy, not poverty or inconvenience. Just look:
145When the price of enlightenment drops to zero, demand for enlightenment remains embarassingly
146low.
147The Soulful Fallback
148Education is a special, deeply political, almost sacred civic activity. It is not merely a
149technical enterprise – providing facts to the untutored. Inescapably, it is a moral and
150aesthetic enterprise – expressing to impressionable minds a set of convictions about how
151most nobly to live in the world.
152– William Bowen and Derek Bok, The Shape of the Riverviii
153There are intermediate positions between “philistine” and “humanist.” Restrained educational
154idealists could fault the humanist tradition for overemphasizing students’ inner life. Sure, students
155rarely feel their souls being enriched. But why not pragmatically equate “enriching the soul” with
156fostering desirable adult attitudes and behavior? From this perspective, “Education is good for
157the soul” counts as true as long as education appreciably shifts society in the right direction. It
158follows that education cannot be good for the soul unless it appreciably shifts society in some
159direction.
160This is a reasonable fallback position, and we’re ready to delve into it. Identifying the “right
161direction” for societal change is up to the reader. Ballparking education’s effect on society,
162however, is up to me.
163To preview: Education does seem to shift students’ values, though less than teachers and parents
164advertise. To isolate education’s influence on society, however, you must unpack how education
165sways students. Is the mechanism “leadership” – planting teachers’ ideas in students’ heads?
166Then education remolds society. Is the mechanism “peer effects” – sorting kids into distinct
167groups? Then education reshuffles society without remolding it.ix
168Suppose you funnel an extra kid into college. His peer group seismically shifts. Given human
169conformity, the freshman will likely try to blend in with his new peers. College youths are less
170religious, for example, so one would expect the student to veer in a secular direction. This does
171not imply, however, that college makes society less religious. The existence of college splits kids
172into two subcultures with opposing peer effects. If college kids are less religious than average
173kids, then non-college kids must be more religious than average kids. Members of each
174subculture adjust their behavior to locally fit in. Religious conformity pressure in the non-college
175pool offsets secular pressure in the college pool. Net effect on society’s religiosity: unclear, even
176if college demonstrably makes students less religious.
177Both leadership and peer effects are at work in the real world. But there are three hefty reasons
178to think that peer effects dominate leadership. First: On the dimensions where academic
179leadership seems most intense, education’s effect on attitudes and behavior is mild. Second:
180Although schools focus their preaching on a handful of issues, education sways attitudes and
181behavior on many dimensions. Indeed, as we shall soon see, educated opinion occasionally
182spurns the pedogogical consensus. Third: Individual and social effects rarely “add up.” Boosting
183an individual’s education usually changes that individual far more than boosting a society’s
184education changes that society.
185High Culture Falls on Deaf Ears
186To the question “of what use are the humanities?”, the only honest answer is none
187whatsoever. And it is an answer that brings honor to its subject. Justification, after all,
188confers value on an activity from a perspective outside its performance. An activity that
189cannot be justified is an activity that refuses to regard itself as instrumental to some larger
190good. The humanities are their own good.
191-Stanley Fish, “Will the Humanities Save Us?”x
192Educators hope to enrich the soul in a hundred different ways. But there’s one form of enrichment
193high school and college pursues more explicitly and energetically than any other: instilling
194appreciation for high culture. English classes push classic novels, plays, and poetry: William
195Shakespeare, Washington Irving, Edgar Allen Poe, Mark Twain, Edith Wharton, Sinclair Lewis,
196Robert Frost. Music classes push traditional music, especially classical music: Antonio Vivaldi,
197Ludwig van Beethoven, Wolfgang Amadeus Mozart, and above all else, John Philip Sousa. Art
198classes are more hands-on, but still try to raise the status of visual works in top museums. Even
199school’s iconoclasm is conservative: Academic curricula often cover Kurt Vonnegut, Arnold
200Schoenberg, or Jackson Pollock, but rarely George R.R. Martin, Lady Gaga, or Frank Miller.
201Though some schools promote high culture more energetically than others, academic curricula
202are plainly tilted against pop culture.
203How effectively has this tilt fostered high culture? Earlier in the book, I appealed to the truism that
204education can’t be responsible for more than 100% of what we know. The same principle allows
205us to set an upper bound on education’s cultural impact: Education can’t be responsible for more
206than 100% of the high culture our society consumes.
207How high is that upper bound? Let’s start with books. Consumer demand is shockingly low
208overall: American spend 0.2% of their income on all reading materials, barely more than $100 per
209family per year.
210xi
211 Americans used to spend more on reading, but never spent much: Back in 1990,
212well before the rise of the web, reading absorbed 0.5% of the family budget.
213xii
214 Today’s Americans
215spend about four times as much on tobacco and five times as much on alcohol as they do on
216reading.
217xiii
218 Within this small pond, high culture is no big fish. Here are three rankings of the bestselling
219English-language fiction of all time. Sales figures include school purchases and assigned
220texts, so they overstate sincere affection for the canon.
221Table 7.1: Best-Selling English-Language Fiction of All Time
222Source
223Rank Wikipedia Ranker.com How Stuff Works
2241 The Lord of the Rings
225(Tolkien)
226A Tale of Two Cities
227(Dickens)
228A Tale of Two Cities
229(Dickens)
2302 Harry Potter and the
231Philosopher’s Stone
232(Rowling)
233The Lord of the Rings
234(Tolkien)
235The Lord of the Rings
236(Tolkien)
2373 And Then There Were
238None (Christie) The Hobbit (Tolkien)
239Harry Potter and the
240Sorcerer’s Stone
241(Rowling)
2424 The Hobbit (Tolkien) And Then There Were
243None (Christie)
244And Then There Were
245None (Christie)
2465 She: A History of
247Adventure (Haggard)
248The Lion, the Witch,
249and the Wardrobe
250(Lewis)
251The Lion, the Witch,
252and the Wardrobe
253(Lewis)
2546 The Lion, the Witch,
255and the Wardrobe
256(Lewis)
257She: A History of
258Adventure (Haggard)
259The Da Vinci Code
260(Brown)
2617 The Da Vinci Code
262(Brown)
263The Da Vinci Code
264(Brown)
265Harry Potter and the
266Half-Blood Prince
267(Rowling)
2688 Harry Potter and the
269Half-Blood Prince
270(Rowling)
271The Catcher in the
272Rye (Salinger)
273Harry Potter and the
274Chamber of Secrets
275(Rowling)
2769 The Catcher in the Rye
277(Salinger)
278Anne of Green Gables
279(Montgomery)
280The Catcher in the
281Rye (Salinger)
28210 Harry Potter and the
283Chamber of Secrets
284(Rowling)
285Black Beauty (Sewell) Harry Potter and the
286Goblet of Fire
287(Rowling)
28811 Harry Potter and the
289Prisoner of Azkaban
290(Rowling)
291Charlotte’s Web
292(White)
293Harry Potter and the
294Order of the Phoenix
295(Rowling)
29612 Harry Potter and the
297Goblet of Fire
298(Rowling)
299The Tale of Peter
300Rabbit (Potter)
301Harry Potter and the
302Prisoner of Azkaban
303(Rowling)
30413 Harry Potter and the
305Order of the Phoenix
306(Rowling)
307Harry Potter and the
308Deathly Hallows
309(Rowling)
310Ben Hur (Wallace)
31114 Harry Potter and the
312Deathly Hallows
313(Rowling)
314Jonathan Livingston
315Seagull (Bach)
316Lolita (Nabokov)
31715 Lolita (Nabokov) Angels and Demons
318(Brown)
319Harry Potter and the
320Deathly Hallows
321(Rowling)
32216 Anne of Green Gables
323(Montgomery)
324Kane and Abel
325(Archer)
32617 Black Beauty (Sewell) To Kill A Mockingbird
327(Lee)
32818 The Eagles Has
329Landed (Higgins)
330Valley of the Dolls
331(Susann)
33219 Watership Down
333(Adams)
334Gone With the Wind
335(Mitchell)
33620 Charlotte’s Web
337(White)
338The Thorn Birds
339(McCullough)
340While sales figures are plainly flawed, all three lists paint similar pictures of the public’s long-run
341literary tastes. High culture is but a niche market. Dickens’ A Tale of Two Cities tops two of the
342three lists. The Catcher in the Rye, Ben Hur, To Kill a Mockingbird, Gone With the Wind, and
343Lolita all appear on at least one list. But fantasy – Tolkien, Rowling, Lewis – dominates. The
344point is not that fantasy lacks literary merit; by my lights, Lord of the Rings towers over Catcher in
345the Rye.
346xiv
347 The point is that the books high school and college classes hail for their supreme
348literary merit lose out to much less prestigious genres. By and large, literature teachers fail to
349“get through” to their captive audiences: They rarely spark love of reading, much less love of the
350genre they urge their students to admire.
351In music, pop culture’s victory over high culture is even more decisive. The Three Tenors in
352Concert is the best-selling classical album ever.
353xv
354 With twelve million copies sold, it does not even
355break into the top fifty albums of all time.xvi
356 Looking at overall sales, classical music is only 1.4%
357of the U.S. music market. Country is eight times as popular, and rock/pop over thirty times as
358popular.xvii
359 Classical does better globally, but still only commands a 5% share of the world’s music
360marketplace. Well, at least it beats jazz.xviii
361 The point, again, is not that classical music alone is
362aesthetically worthwhile. Bad Religion isn’t Bach, but it’s good. The point, rather, is that schools’ 
363aesthetic priorities have negligible cultural impact. Even if American schools are the root cause
364of all U.S. consumption of classical music, their combined efforts only boost its market share from
3650% to 1.4%.
366Why is high culture so marginalized? Humanists may be tempted to blame poor salesmanship:
367Students would learn to love Shakespeare and Brahms if they only had the right teachers. The
368straightforward story, though, is that high culture requires extra mental effort to appreciate – and
369most humans resent mental effort. Students are overwhelmingly bored by Shakespeare, and the
370rare fan of high culture would probably have come to love the Bard on his own. Students sample
371a little high culture when their grades depend on it. As soon as the exam ends, however, the vast
372majority of students rush back to their low-brow comfort zone.
373Anyone reading this book is probably a bird of a different feather. You may even remember the
374names of the teachers who opened your eyes to the finer things in life. I owe my love of classical
375music to Mr. Zainer (General Music, 7th grade), and my love of literature to Mrs. Ragus (Honors
376English, 11th grade). A quick look at the basic facts, however, shows that our experiences are
377abnormal. The vast majority of our classmates emerge from years of cultural force-feeding with
378their aesthetic palates unchanged.
379The Paper Tiger of Political Correctness
380Three years of school was nice,
381In rice fields, planting rice.
382You did good work you see,
383Re-educating me…
384But men, will always be men.
385They washed out my brain,
386I'm still who I am.
387Deep inside, I know what I know.
388Wherever I go, I speak Uncle Ho,
389And think Uncle Sam.
390- Miss Saigon
391American educators lean left. There’s no denying it. The party breakdown for K-12 public school
392teachers is lopsided: roughly 45% Democrat, 25% independent, and 30% Republican.
393xix
394 The
395breakdown for college faculty is starkly lopsided: A nationally-representative study of all
396professors – including professors in two-year colleges – finds 51% Democrats, 35%
397independents, and 14% Republicans.
398xx
399 A similar study of four-year college faculty reports 50%
400Democrats, 39% independents/other, and 11% Republicans.
401xxi
402 Left-wing dominance seems even
403stronger at more elite schools.
404xxii
405
406Colleges are least balanced in the most politically-charged subjects. There are about five
407Democrats per Republican in the humanities, and eight Democrats per Republican in the social
408sciences.xxiii
409 As recently as 2006, 5% of humanities professors and 18% of social scientists were
410self-described “Marxists.”xxiv
411 The ratio of self-styled “liberals” to “conservatives” is less extreme,
412presumably because – unlike party affilation – these are relative terms.
413xxv
414 When 18% of their
415colleagues are outright Marxists, many mainstream Democrats understandably feel “moderate”
416by comparison.
417None of this proves that teachers and professors use their classrooms to “enlighten” or
418“brainwash” their students. But the conditions for enlightenment/brainwashing are most
419auspicious. Educators’ distinctive worldview provides a compelling motive to mold students’
420minds. Captive student audiences provide a golden opportunity to mold students’ minds. Even 
421if teachers avoid blatant proselytizing, ideological neutrality requires Vulcan self-discipline. When
422you and your colleagues share an ideology, your common ideology soon sounds like common
423sense – and there’s nothing unprofessional about promoting common sense. Even if lectures
424and readings are only subtly slanted, won’t a subtle slant, maintained year after year, win students’
425hearts and minds in the end?
426No. In the data, the well-educated are only microscopically more liberal. The General Social
427Survey has people place themselves on a seven-step scale, where 1 is “extremely liberal,” 4 is
428“moderate,” and 7 is “extremely conservative.” An extra year of education seems to make people
429.014 steps more liberal.xxvi
430 Taken literally, this means that over seventy years of education are
431required to shift ideology a single step. Statistical corrections make education’s impact on
432ideology look stronger, but it stays weak.
433xxvii
434If the effect on ideology is slight, the effect on partisanship is slightly perverse. As education rises,
435people actually grow less Democratic. The General Social Survey’s respondents place
436themselves on a seven-step scale, where 0 is “strong Democrat,” 3 is “independent” and 6 is
437“strong Republican.” An extra year of education seems to make people .071 steps more
438Republican.xxviii
439 Statistical corrections makes this effect look weaker, but education still appears
440to mildly boost support for the party that teachers and professors disfavor.
441xxix
442The plot thickens when you analyze education’s effect on specific issues. Abundant research
443confirms that education raises support for civil liberties and tolerance, and reduces racism and
444sexism.
445xxx
446 These effects are only partly artifactual. Correcting for intelligence cuts education’s
447impact by about a third.xxxi
448 Correcting for intelligence, income, occupation, and family background
449slices education’s impact in half.
450xxxii
451 When all corrections are done, education still fosters a
452package of socially liberal views.
453At the same time, abundant research also confirms that education raises support for capitalism,
454free markets, and globalization.xxxiii
455 These effects, too, are partly artifactual. Correcting for
456intelligence cuts education’s impact by about 40%. Correcting for intelligence, income,
457demographics, party, and ideology halves it.
458xxxiv
459 But when all corrections are done, education still
460fosters a package of economically conservative views.xxxv
461If educators are as left-wing as they seem, why would education have such contradictory effects
462on students’ stances? The charitable story is that educators scrupulously keep their politics out
463of the classroom. The more plausible story, though, is that educators are unpersuasive. The
464Jesuits say, “Give me the child until he is seven and I’ll give you the man.” Society gives liberal
465educators the child until he’s fifteen, eighteen, twenty two, or older. But issue-by-issue, teachers
466are about as likely to repel their students as attract them. Educators could protest, “The problem
467isn’t that we’re unpersuasive, but that students are stubborn,” but students revise their opinions
468all the time. The longer they stay in school, the more they revise. They just don’t revise in a
469reliably liberal direction.
470Critics who highlight educators’ leftist leanings usually have an ideological ax to grind (or swing):
471“Look at how liberal our teachers are! Leaving education of the young in the hands of ‘politically
472correct’ ideologues endangers our democracy. School should be a vibrant marketplace of ideas,
473not a center for indoctrination.” I grant the political imbalance; as a college professor, I’ve
474repeatedly endured it. The point, though, is that political correctness is a paper tiger. Even
475extreme political imbalance leaves little lasting impression on students’ minds. Contrary to the
476indoctrination story, education doesn’t progressively dye students ever brighter shades of red.
477Since education raises social liberalism and economic conservativism, neither liberals nor
478conservatives should cheer education’s effect on our political culture. What about people who
479are both socially liberal and economically conservative? Should they admit that education really
480is “good for the soul” after all? It’s complicated. If teachers aren’t molding their students, the
481logical inference is that students are molding each other. But peer effects, to repeat, are doubleedged.
482When schools cluster socially liberal, economically conservative youths inside the Ivory
483Tower, they inadvertently but automatically cluster socially conservative, economically liberal
484youths outside the Ivory Tower. If education is good for the souls of the former, it’s bad for the
485souls of the latter. Net effect on the polity? Ambiguous.
486Getting Out the Vote
487Voting is a bipartisan tenet of our secular religion. “It doesn’t matter how you vote, but vote”:
488Educators say it loud, proud, early, and often. Staunchly partisan teachers plead with their
489students to vote, silently hoping they’ll “vote correctly.” Genuinely apolitical teachers are often
490equally insistent: Democracy is at stake. K-12 teachers probably preach the duty to vote more
491energetically than college professors, but the whole education system reads from the same
492prayerbook.
493Do their sermons work? It’s complicated. Voter turnout rises sharply with education. Substantial
494effects of education on turnout usually linger after statistically correcting for income,
495demographics, intelligence, and so on.xxxvi
496 Despite some thoughtful naysayers, the limited
497experimental data also shows that extra education boosts turnout.xxxvii
498The fly in the ointment is that education has sharply risen over the last century, but turnout has
499gently fallen. This could mean that offsetting factors masked education’s pro-voting effect.xxxviii
500
501But several prominent researchers defend a straightforward alternative: Turnout depends on 
502relative education.xxxix
503 People don’t vote because they’re educated, but because they have more
504education than others. This could, as usual, reflect peer effects: The longer you stay in school,
505the more politically active your social circle, and the more politically active you have to be to fit in.
506But it could also stem from sheer status: The higher your rank in the pecking order, the more you
507want to crow.
508Suppose you’re convinced that voting enriches the soul. As long as some version of relative
509education is what counts, education redistributes the enrichment rather than creating it. Schooling
510one more person makes him more likely to experience the wonder of democratic participation.
511But it also makes the rest of the citizenry less likely to partake in the wonder.
512The Modern Lifestyle
513Culture and politics aside, we also have stereotypes about education’s effect on lifestyle. In our
514minds, the well-educated favor the “modern” way of life; the less-educated, the “traditional” way
515of life. The modern are secular and bohemian, with few children; the traditional are religious and
516stodgy, with many children. Schools may not loudly favor modernity over tradition, but perhaps
517they subtly turn their students into Modern Men and Modern Women nonetheless. Education
518could be good for the soul because it frees us from the dead hand of the past.
519Or not. For all their appeal, stereotypes about education and modernity are unreliable. In the
520data, the well-educated are more modern in some ways, more traditional in others.
521Religion. Stereotypes say that the well-educated are less religious, but this is a half-truth. The
522well-educated are less religious theologically. As education goes up, faith in God and the literal
523truth of the Bible recede. Yet the well-educated are actually more religious sociologically. As 
524education goes up, so do church membership and church attendance. These are well-established
525patterns, at least in the United States.xl
526
527Statistical corrections make education’s theological effect look smaller but its sociological effect
528look bigger. The General Social Survey measures Americans’ confidence in God’s existence on
529a 1-6 scale (1= “I don’t believe in God,” 6= “I know God really exists and I have no doubts about
530it” ) and religious attendance on a 0-8 scale (0= “never,” 8=”more than once a week”). After
531correcting for income, intelligence, social status, demographics, and time, a year of education
532reduces faith in God by .04 steps, but increases religious attendance by .06 steps.xli
533 Global
534studies of religion typically conclude that intelligence reduces both forms of religiosity, but
535education’s effect on religion remains fuzzy. The only clear-cut evidence that education seriously
536undermines religion comes from former Communist countries, where the curricula – not to
537mention the governments themselves – were harshly atheistic.
538xlii
539How can education – including college – matter so little for religion? Sociologists Jeremy Uecker,
540Mark Regnerus, and Margaret Vaaler paint a plausibly cynical picture:
541[S]ome students have elected not to engage in the intellectual life around them. They are
542on campus to pursue an "applicable" degree, among other, more mundane pursuits, and
543not to wrestle with issues of morality or meaning. They instead stick to what they "need to
544know" -- that which will be on the exam. Such students are numerous, and as a result
545students' own religious faith (or lack of it) faces little challenge. Indeed, many university
546curricula are constructed to reward this type of intellectual disengagement… What is not
547contested, then, cannot be lost. Faith simply remains in the background of students' lives
548as a part of who they are, but not a part they talk about much with their peers or professors.
549[emphasis added]
550All this is unsurprising given how little youths know about their religions in the first place:
551[W]hile higher education opens up new worlds for students who apply themselves, it can,
552but does not often, create skepticism about old (religious) worlds, or at least not among
553most American young people, in part because students themselves do not perceive a great
554deal of competition between higher education and faith, and also because very many
555young Americans are so undersocialized in their religious faith (before college begins) that
556they would have difficulty recognizing faith-challenging material when it appears.xliii
557None of this shows that education is inherently religiously impotent. Perhaps heavy-handed
558parochial or atheistic education would spur widespread conversion or apostasy. Actually-existing
559education’s religious effects are however mixed and weak.
560Marriage and divorce. Since the 1940s, lifelong marriage retreated as education advanced.xliv
561
562Could the reason be that schools sap belief in traditional values in favor of bohemian
563permissiveness? Unlikely. Despite broad trends, the link between education, marriage, and
564divorce is tenuous.
565In the modern United States, getting married is slightly more common for male college grads, but
566slightly less common for female college grads. Being married is more common for all college
567grads, because both male and female college grads are now less likely to divorce. Yet on closer
568look, the patterns blur. Americans who start college but fail to finish are less likely to marry and
569more likely to divorce than Americans who avoid college altogether. American women with
570advanced degrees are less likely to be married than American women who stop after the B.A.xlv
571
572Looking back in time or around the world makes the story blurrier still. Until recent decades,
573educated American women tended to stay single.
574xlvi
575 Internationally, the well-educated are less
576likely to marry in some countries, but more likely to marry in others; the same holds for divorce.xlvii
577How much of education’s apparent effect is genuine? Evidence is thin, but researchers who try
578to statistically isolate education’s effects usually find that – at least in the modern United States –
579education truly encourages marriage and marital stability.
580xlviii
581 In the General Social Survey, the
582estimated effects are small but solid – at least for the last two decades. Correcting for
583demographics, intelligence, church attendance, and era, each year of education raises the chance
584of being married by 0.7 percentage-points, and reduces the chance of being divorced by 0.3
585percentage-points.xlix
586
587Overall, then, friends of traditional marriage have little to fear from education. While its effect is
588uneven, time in school now seems to make Americans less bohemian, not more. Why would this
589be so? As usual, peer effects are the go-to answer. Humans care far more about what their
590“social equals” are doing today than what their teachers said or insinuated years ago. Consistent
591with this story, social class fully accounts for education’s effects on marriage and divorce in
592modern America.
593l
594 Staying in school gets individuals into the elite club, which in turn helps them
595get married and avoid divorce. When everyone stays in school longer, however, the elite club
596jacks up its membership requirements, leaving the overall state of marriage unchanged.
597
598Fertility. Educated people have fewer kids. Few demographic laws are more strictly enforced.
599This law doesn’t just fit the modern United States.li
600 It fits nearly every country, whether “modern”
601or “developing.”
602lii
603 And it fits year after year, at least since 1900 or so.liii
604 Fertility gaps are big:
605Averaging over the world, low-education women outbreed high-education women by about onethird.liv
606 Nation-by-nation, disparities of a full child or more are common.lv
607 This demographic law
608even fits tolerably across countries and over time: High-education countries are less fertile than
609low-education countries, and countries’ fertility erodes as education advances.
610In principle, education could be a mask for income, intelligence, status, democratization, or
611modernization. When statistically challenged, though, education stays strong.
612lvi To illustrate, take
613the United States from 1972 to 2012. During this era, a year of education seems to cut Americans’
614fertility by .12 children. After statistically correcting for income, intelligence, demographics, and a
615time trend, a year of education still seems to prevent .10 births.
616lvii
617 While education cools fertility
618for both sexes, it cools women’s more. In married couples, wife’s education matters three or four
619times as much as husband’s education.lviii
620
621Does education sway child-bearing via leadership or peer effects? The leadership story is
622straightforward: Almost all schools – even schools with celibacy-only sex education – explicitly
623urge students to delay child-bearing. And most schools at least insinuate that high-powered
624careers are better than big families. Peer effects make sense, too: Look at the Baby Boom.
625Globally, the mix remains unclear.lix
626 At least in the modern United States, however, peer effects
627seem weak. Even though social class fully explains education’s effects on marriage and divorce,
628social class explains none of education’s effect on fertility. Dropouts who climb into the upper
629class still breed like dropouts; Ph.D.s who stumble into the lower class still breed like Ph.D.s.
630When schools prompt their students to have fewer kids, then, they’re plausibly prompting society
631to have fewer kids, too. Education leads society toward a less populous future. Out of all the
632educational consequences we’ve scrutinized, this is the most impressive. The key question: Are
633these consequences impressively good or impressively bad?
634If you’re convinced that every country on Earth is overpopulated, education’s anti-natal effect is a
635strong point in its favor. Anyone who accepts the dangers of low fertility, however, should tremble.
636Almost all developed countries are below replacement fertility. Germany, Japan, and Russia’s
637populations are already falling. Many other lands will join them in coming decades. Even worse,
638education doesn’t just sap overall population. It targets the educational elite, because the people
639most inclined to linger in school restrict their child-bearing the most. Whatever weight you put on
640human capital versus signaling, or nature versus nurture, this is demographically perverse. The
641flip side, happily, is that countries can cheaply arrest their demographic troubles. Indeed, the cost
642is less than nothing: Governments can apparently make babies with budget cuts.
643Broadening Horizons
644I’ve been griping about curricula since kindergarten. Whenever teachers gave “stupid”
645assignments, I voiced my malcontent to teachers and parents alike. Their standard response:
646Even the “stupidest” assignments serve the higher purpose of broadening horizons. The world is
647rich with possibilities, yet most students are poor in curiosity. Teachers have a sacred duty to
648make closed-minded youths sample this cornucopia. Academic “tasting menus” don’t just enrich
649students emotionally. They enrich students vocationally by esposing them to overlooked career
650paths.
651In hindsight, the theory of my teachers and parents was sound. Kids are closed-minded, and
652schools can help them by nudging them to try new things. Unfortunately, educators misapply this
653noble theory. Rhetoric aside, educators are at least as narrow-minded as kids. Most of the items
654on the academic tasting menu have the same stale flavor – unsurprising when you remember that
655teachers typically teach whatever they were taught. When schools decry “narrow-mindedness,”
656their real goal is to replace students’ narrowness with their own.
657Think about what passes for “broadening students’ horizons.” Teachers expose students to an
658ossified list of subjects: music, art, poetry, drama, foreign language, history, government, dance,
659sports. Some kids respond eagerly, especially to music and sports. Yet the greater their
660excitement, the greater their ultimate disappointment: Almost no one grows up to be a violinist, 
661painter, poet, actor, historian, politician, ballet dancer, or professional athlete. More importantly,
662all the kids who respond eagerly to none of these time-honored subjects must wait until college
663for the mandatory “broadening” to relent.
664What’s the alternative? For starters, give students numerous and diverse options. Instead of
665making students study yet another American poem, expose them to Japanese graphic novels.
666Rather than forcing kids to perform one more play, show them a few films from the 1980s. When
667you run out of ideas, assign a random Wikipedia article. (I just perused the entry on the Geelvink
668imperial pigeon). If you want to help kids discover what emotionally “clicks” for them, repeated
669trial and error beats academic tradition cold. Anyone who objects that reading Japanese comic
670books and watching old movies is “useless” should check their double standard. How are comic
671books and movies any more useless than poems and plays?
672All else equal, of course, exposing students to plausible careers is better than esposing them to
673mere hobbies. To live the adage “Do what you love, and you’ll never work a day in your life,”
674students have to know what lovable jobs are available. Give students numerous, diverse, yet
675realistic options. Start with the Bureau of Labor Statistics’ figures on “employment by major
676occupational group” and “occupations with the most job growth.”lx
677 Expose boys to nursing.
678Introduce strong math students to insurance. Tell upper middle class kids what plumbers and
679electricians do – and how much they earn. See how many students try Python programming if it
680fulfills their foreign language requirement. When you run out of ideas, have students check out
681an unfamiliar job from Bureau of Labor Statistics’ Occupational Outlook Handbook.
682lxi
683I doubt schools will try any of these reforms. But their curricular conservatism teaches us a sad
684truth about actually-existing education: “Broadening horizons” is a slogan educators use to trump
685students’ sensible doubts. If educators really wanted to broaden students’ horizons, curricula 
686would give students a tour of all the world has to offer – not a tour of what the educators were
687forced to learn when they were students.
688The Merit of Play
689Rather than give children 30 minutes to while away the time as they please, he said, it
690makes more sense to teach them a skill, like dancing or gymnastics.
691– New York Times on Atlanta schools superintendant Dr. Benjamin Canadalxii
692Education can be glorious. At its best, to quote Roman philosopher Lucretius, it is a “voyage in
693mind throughout infinity.”lxiii
694 But education is not the sole glorious experience. Since students
695only have so many hours each day, even the finest education risks crowding out competing
696experiences of greater worth.
697What could possibly outshine the wonders of education? It is tempting to focus on prestigious
698activities like writing a killer app or training for the Olympics. Students’ most relevant competing
699experience, though, is play – savoring the joys of youth. The more time and effort students devote
700to their studies, the less remains for care-free exploration of their world. Recall the classic backto-school
701essay, “What I Did During My Summer Vacation.” Some kids fritter away their free
702months watching reality t.v. in a lonely basement. But plenty of others bond with grandparents or
703cousins, pick up an interest in seashells, play Dungeons and Dragons with their best friends, or
704tour the country. If kids spent more time in school, some of this enrichment would clearly be lost.
705If kids spent less time in school, can there be any doubt that more of this enrichment would be
706gained?
707Psychologist Peter Gray could well be the world’s greatest spokesman for the merit of free play.
708Kids have more fun and learn vital lessons when adults give them their space. Gray:
709“Playing well and having fun are more important than winning” is a line often used by Little
710League coaches after a loss, rarely after a win. But with spectators watching, with a trophy
711on the line, and with so much attention paid to the score, one has to wonderful how many
712of the players believe that line, and how many secretly think that Vince Lombardi had it
713right. The view that “winning is the only thing” becomes even more prominent as one
714moves up to high school and then to college sports…
715In informal sports, playing well and having fun really are more important than winning.
716Everyone knows that; you don’t have to try to convince anyone with a lecture. And you
717can play regardless of your level of skill. The whole point of an informal game is to have
718fun and stretch your skills, sometimes in new and creative ways that would be disallowed
719or jeered in a formal game… If you are a better player than the others, these are ways to
720self-handicap, which make the game more interesting for everyone. In a formal game,
721where winning matters, you could never do such things; you would be accused of betraying
722your team.lxiv
723The lesson isn’t that all play and no school are best for kids. The lesson is that champions of
724academic soulcraft shouldn’t fixate on education. Instead, they should seek out what mix of
725school and play is best for the soul. Unfortunately, thanks to the high status of education and the
726low status of play, we tend to compare school at its best to play at its worst: Another hour of Angry
727Birds can’t compete with a guest lecture on literature from the teacher Robin Williams played in
728Dead Poets Society. The smart way to discover the best mix of school and play, though, is to
729compare school and play as they really are. Both fall very short of their promise, but it’s unclear
730which falls shorter.
731Still, there’s little reason to favor a dominant role for education over play – and in our society,
732education dominates children’s days. School and study time has been high and growing for 
733decades. According to leading studies of 6-12 year-olds’ schedules, weekly school and study
734time rose from about 31 hours in 1981 to 37 hours in 1997 and 2003. Playtime is small by
735comparison, about 10 hours a week. This counts computer games, but excludes t.v. time, which
736fell from over 18 hours in 1981 to 14 hours in 2003.lxv
737 Outdoor play has atrophied over the last
738generation: 70% of mothers say they played outside every day when they were kids, while only
73931% of their children do the same.lxvi
740 Only a small minority of elementary schools have abolished
741recess, but one major study found that 20% of school districts cut recess during the first five years
742of the No Child Left Behind Act. Virtually no district made recess longer.
743lxvii
744Longer school days do serve one socially useful function: They warehouse kids so both their
745parents can work. But more hours in school needn’t mean more hours of school. Schools could
746have used kids’ extra campus hours to expand recess. Indeed, if they wanted kids to keep some
747independence, they could have offered an array of fun yet frugal options in lieu of extra class time.
748My pet cause: Keep the school library open so studious and intellectually curious kids have a
749tranquil place for free reading. Until college, every school I ever attended had a well-stocked
750library that was almost never open to the student body. Free play takes many forms. Why not
751deploy a couple teachers’ aids to keep the peace, and turn the library into a bookworms’
752sanctuary?
753For college kids, you may recall, playtime is now longer than ever.lxviii
754 The college workload
755slimmed down as the K-12 workload bulked up. Most critics of modern education take this as a
756tragic fall in standards. Once you accept the merit of play, the rise of Leisure College, USA
757becomes a blessing in disguise. College gives students ample time for care-free exploration of
758their world – time they rarely had in childhood. Plenty of undergrads fritter away their opportunity
759in a drunken stupor. Yet others sample a wide range of fascinating options, acquiring passions
760that last a lifetime. My undergraduate years at UC Berkeley were my favorites precisely because 
761classes were so undemanding. Every day was packed with ample hours for play, and play I did.
762I read philosophy, listened to opera, wargamed with my friends, and argued politics with strangers
763past midnight. These experiences changed me forever. I owe my soul to lax academic standards.
764The Cynical Idealist
765Economists are a cynical bunch. Most are tone-deaf to the humanist thesis that education
766enriches the soul. They studiously measure education’s career benefits. They grant education
767has consumption benefits. Yet most balk at claims about education’s intrinsic worth. When
768humanists face my calculations of education’s selfish and social returns, they assume I’m being
769a typical cynical economist, oblivious to the transformative ideals so many educators hold dear.
770I am an economist and I am a cynic, but I’m not a typical cynical economist. I’m a cynical idealist.
771I embrace the ideal of transformative education. I believe wholeheartedly in the life of the mind.
772What I’m cynical about is people.
773I’m cynical about students. The vast majority are philistines. The best teachers in the world
774couldn’t inspire them with sincere and lasting love of ideas and culture. I’m cynical about teachers.
775The vast majority are uninspiring; they can’t even convince themselves to love ideas and culture,
776much less their students. I’m cynical about “deciders” – the school officials who control what
777students study. The vast majority think they’ve done their job as long as students obey. Deciders
778barely care if students inwardly transform for life as long as they outwardly submit until graduation.
779Anyone who searches their memory will find noble exceptions to these sad rules. I know plenty
780of eager students and passionate educators, and know of a few wise deciders. They’re the salt
781of the earth. Still, my forty years in the education industry – many at the “best schools in the 
782world” – leave no doubt that eager students, passionate educators, and wise deciders are
783hopelessly outnumbered. Meritorious education survives, but does not thrive.
784I don’t hate education. Rather I love education too much to accept our Orwellian substitute.
785What’s Orwellian about the status quo? Most fundamentally, the idea of compulsory
786enlightenment. Educators routinely defend compulsion on the ground that few students want to
787explore ideas and culture. They’re right, but forget a deeper truth: Eager students are part and
788parcel of intrinsically valuable education. Mandatory study of ideas and culture ruins the journey.
789Even if you bite the end-justifies-the-means bullet, compulsory enlightenment yields little
790enlightenment. For all their Orwellian self-congratulation, schools are unconvincing. Under
791auspicious conditions, they fail to make either high culture or liberal politics noticeably more
792popular. Regimentation may be a good way to mold external behavior, but it’s a bad way to win
793hearts and minds – and a terrible way to foster thoughtful commitment. As Stanford education
794professor David Labaree remarks, “Motivating volunteers to engage in human improvement is
795very difficult, as any psychotherapist can confirm, but motivating conscripts is quite another thing
796altogether. And it is conscripts that teachers face every day in the classroom.”lxix
797
798Even top students respond to incentivized soulcraft by gaming the system, not reforming their
799priorities. British universities base admissions on academic performance, not overall merit.
800American universities use a “balanced” approach. When British professor Greg Clark began
801teaching at Stanford University, his elite American students looked like better human beings than
802their British counterparts. He soon learned Americans’ superiority was fake:
803In my second year as an assistant professor at Stanford University, I was assigned the
804task of mentoring six freshmen. Each appeared on paper to have an incredible range of
805interests for an eighteen-year-old: chess club, debate club, history club, running team, 
806volunteering with homeless shelters. I soon discovered that these supposed interests were
807just an artifact of the U.S. college admission process, adopted to flesh out the application
808forms and discarded as soon as they have worked their magic.lxx
809Still, humanists should not despair. The savior of transformative education has arrived: the
810internet, the Merit Machine. Though online education isn’t poised to put brick and mortar schools
811out of business, it already beats traditional education in the quest for enlightenment. The internet
812enlightens the money-poor. Out-of-pocket cost is normally zero. The internet enlightens the timepoor.
813Commuting cost is normally zero, too. The internet enlightens the intellectually isolated.
814Search engines and ratings mark the most promising autodidactic paths. Cyberspace is full of
815junk, yet finding top-notch content is child’s play.
816Many idealists object that the internet only provides enlightenment for those who seek it. They’re
817right, but petulant to ask for more. Enlightenment is a state of mind, not a skill. State of mind,
818unlike skill, is easily faked. When schools require enlightenment, students predictably respond
819by feigning interest in ideas and culture. This in turn gives educators a false sense of
820accomplishment.
821When enlightenment is optional, in contrast, educators’ failure to transform their students is
822undeniable. Cynics may lose whatever hope they had left, but cynical idealists will wonder, “How
823can we do better?” The obvious responses are better pedagogy and better marketing. This isn’t
824wishful thinking: Online education, broadly defined, refines pedagogy and marketing every day.
825Educators’ less obvious response is broadening the audience. Most humans interested in
826abstract ideas and high culture are working adults. Instead of lamenting youthful apathy,
827passionate educators should redirect their energy to humans who are ready for enlightenment. 
828There is little money in blogging, podcasting, or uploading lectures to Youtube. But if, like me,
829you love education to the depths of your soul, such efforts are their own reward.
830i Cites
831ii X, Malcolm. 1964. The Autobiography of Malcolm X. NY: Ballantine, pp.174-6.
832iii Waley, Arthur, translator. 2005. The Analects of Confucius. NY: Routledge, p.187.
833iv Pinker, Steven. September 4, 2014. “The Trouble With Harvard: The Ivy League is Broken and Only
834Standardized Tests Can Fix It.” The New Republic.
835v Self-cite
836vi http://www.internetworldstats.com/stats.htm
837vii See generally Carey 2015a.
838viii Bowen and Bok 1998, p.xxii.
839ix Cites on peer effects; who acknowledges zero-sumness?
840x Fish, Stanley. January 6, 2008. “Will the Humanities Save Us?” New York Times.
841xi U.S. Bureau of Labor Statistics. 2013. “Consumer Expenditures in 2011.” BLS Reports #1042, pp.2-3.
842URL http://www.bls.gov/opub/reports/cex/consumer_expenditures2011.pdf
843 
844
845xii United States Census Bureau. 2011. “Average Annual Expenditures of All Consumer Units by
846Selected Major Types of Expenditure.” Statistical Abstract of the United States 2012, Table 684. URL
847http://www.census.gov/compendia/statab/2012/tables/12s0684.pdf
848xiii U.S. Bureau of Labor Statistics. 2013, ibid, p.3.
849xiv http://econlog.econlib.org/archives/2014/07/what_to_learn_f_1.html
850xv Angel, Amanda. September28, 2011. “Top Five Classical Record Holders.” WQXR. URL
851http://www.wqxr.org/#!/story/161365-top-five-classical-record-holders/
852xvi http://en.wikipedia.org/wiki/List_of_best-selling_albums
853xvii Nielsen. “2014 Nielsen Music U.S. Report,” p.2. URL
854http://www.nielsen.com/content/dam/corporate/us/en/public%20factsheets/Soundscan/nielsen-2014-yearend-music-report-us.pdf
855xviii Music and Copyright’s Blog. July 15, 2014. “Pop still the biggest music genre, but retail sales slide
8567.6% in 2013.” URL https://musicandcopyright.wordpress.com/2014/07/15/pop-still-the-biggest-musicgenre-but-retail-sales-slide-7-6-in-2013/
857xix Moe, Terry. Special Interest: Teachers Unions and America's Public Schools, pp.84-87. Unfortunately,
858to the best of my knowledge this is the only recent systematic data on K-12 teachers’ politics.
859xx Gross and Simmons. 2007. “The Social and Political Views of American Professors,” unpub. ms.,
860p.31-32.
861
862xxi Rothman, Stanley, S. Robert Lichter, and Neil Nevitte. 2005. “Politics and Professional Advancement
863Among College Faculty.” The Forum 3, pp.5-6.
864xxii Cardiff, Christopher, and Daniel Klein. 2005. “Faculty Partisan Affiliations in All Disciplines: A VoterRegistration
865Study.” Critical Review 17, p.243. Gross and Simmons similarly report that the ratio to selfreported
866liberals to conservatives is lowest at community colleges, and highest at four-year liberals arts
867and elite Ph.D.-granting colleges.
868xxiii Gross and Simmons 2007, p.33, Rothman et al. 2005, p.6. For a good survey of the literature, see
869Klein, Daniel, and Charlotta Stern. 2009. “By the Numbers: The Ideological Profile of Professors.” In
870Maranto, Robert, Richard Redding, and Fredrick Hess, eds. The Politically Correct University: Problems,
871Scope, and Reforms. Washington, DC: American Enterprise Institute, pp.15-37.
872xxiv Gross and Simmons 2007, p.41.
873xxv Zipp, John, and Rudy Fenwick. 2006. “Is the Academy a Liberal Hegemony? The Political
874Orientations and Educational Values of Professors.” Public Opinion Quarterly 70, pp.304-26; Gross and
875Simmons 2007. Rothman et al. is a notable exception; their liberal/conservative ratios are about as high
876as their Democrat/Republican ratios.
877xxvi Results from regressing GSS variable identifier POLVIEWS on a constant and years of education.
878xxvii The main problem with the simple approach: The well-educated are richer, and the rich are more
879conservative. As a result, income conceals some of education’s effect. If you regress POLVIEWS on a
880constant, years of education, and log family income, a year of education makes you .028 units more
881liberal – double the estimate from the simple approach. Further correcting for race, sex, age, and year, a
882year of education makes you .024 units more liberal.
883
884xxviii Results from regressing GSS variable identifier PARTYID on a constant and years of education,
885excluding respondents who support third parties.
886xxix If you regress PARTYID on a constant, years of education, log family income, race, sex, age, and
887year, a year of education makes you .029 units more Republican.
888xxx See e.g. Coenders, Marcel, and Peer Sheepers. 2003. “The Effect of Education on Nationalism and
889Ethnic Exclusionism: An International Comparison.” Political Psychology 24, pp.313-43; Weakliem,
890David. 2002. “The Effects of Education on Political Opinions: An International Study.” International
891Journal of Public Opinion 13, pp.141-57; Nie, Norman, Jane Junn, and Kenneth Stehlik-Barry. 1996.
892Education and Democratic Citizenship in America, Chicago: University of Chicago Press; Golebiowska,
893Ewa. 1995. “Individual Value Priorities, Education, and Political Tolerance.” Political Behavior 17, pp.23-
89448; and Case, Charles, and Andrew Greeley. 1990. “Attitudes Toward Racial Equality.” Humboldt
895Journal of Social Relations 16, pp.67-94.
896xxxi Nie et al. 1996; Bobo, Lawrence, and Frederick Licari. 1989. “Education and Political Tolerance:
897Testing the Effects of Cognitive Sophistication and Target Group Affect.” Public Opinion Quarterly 53,
898pp.285-308.
899xxxii Kingston, Paul, Ryan Hubbard, Brent Lapp, Paul Schroder, and Julia Wilson. 2003. “Why Education
900Matters.” Sociology of Education 76, pp.53-70.
901xxxiii Caplan, Bryan. 2007. The Myth of the Rational Voter: Why Democracies Choose Bad Policies.
902Princeton, NJ: Princeton University Press, pp.22-93; Weakliem 2002; Caplan, Bryan. 2001. “What
903Makes People Think Like Economists? Evidence on Economic Cognition from the Survey of Americans
904and Economists on the Economy." Journal of Law and Economics 44(2), pp.395-426. Althaus, Scott.
9052003. Collective Preferences in Democratic Politics. Cambridge: Cambridge University Press, pp.97-144
906similarly finds that, all else equal, better-informed people are more economically conservative.
907
908xxxiv Miller, Steven, and Bryan Caplan. 2010. “Intelligence Makes People Think Like Economists:
909Evidence from the General Social Survey.” Intelligence 38, pp.636-47, plus supplementary calculations
910from the authors.
911xxxv Measuring effects issue-by-issue neatly explains education’s puzzlingly small impact on ideology and
912party. Since education simultaneously increases social liberalism and economic conservatism, its effect
913on “liberalism” is ambiguous. And while their social liberalism makes the well-educated more Democratic,
914their economic conservatism makes them more Republican, leaving partisanship nearly untouched.
915xxxvi For overviews of the research and some basic results, see e.g. Burden, Barry. 2009. “The Dynamic
916Effect of Education on Voter Turnout.” Electoral Studies 28, pp.540-9; Nagler, Jonathan. 1991. “The
917Effect of Registration Laws and Education on U.S. Voter Turnout.” American Political Science Review 85,
918pp.1393-405; Powell, G. Bingham. 1986. “American Voter Turnout in Comparative Perspective.”
919American Political Science Review 80, pp.17-43.
920xxxvii The most notable naysayers: Kam, Cindy, and Carl Palmer. 2008. “Reconsidering the Effects of
921Education on Political Participation.” Journal of Politics 70, pp.612-31 finds that after fully accounting for
922“preadult experiences and influences in place during the senior year of high school,” higher education has
923no effect on turnout; Tenn, Steven. 2007. “The Effect of Education on Voter Turnout.” Political Analysis
92415, pp.446-64 reports that immediately after gaining an extra year of education, individuals are no more
925likely to vote than they were in the previous year. Sonderheimer, Rachel, and Donald Green. 2010.
926“Using Experiments to Estimate the Effects of Education on Voter Turnout.” American Journal of Political
927Science 54, pp.174-89 examines three sets of experimental evidence on education and turnout.
928xxxviii Burden 2009 reviews the contrast between micro- and macro-level evidence, and summarizes the
929top contending “offsetting factors.” 
930
931xxxix The leading defenses of the relative education theory are Tenn, Steven. 2005. “An Alternative
932Measure of Relative Education to Explain Voter Turnout.” Journal of Politics 67, pp.271-82; and Nie et al.
9331996.
934xl Glaeser, Edward, and Bruce Sacerdote. 2008. “Education and Religion.” Journal of Human Capital 2,
935pp.188-215; Iannaccone, Lawrence. 1998. “Introduction to the Economics of Religion.” Journal of
936Economic Literature 36, pp.1470-74.
937xli In the GSS, I regressed GOD and ATTEND on years of education, the log of real family income,
938cognitive ability (WORDSUM), social class (CLASS), age, age squared, race, gender, and year.
939xlii Zuckerman, Miron, Jordan Silberman, and Judith Hall. 2013. “The Relation Between Intelligence and
940Religiosity: A Meta-Analysis and Some Proposed Explanations.” Personality and Social Psychology
941Review 17, pp.325-354, after reviewing a large research literature, concludes that intelligence
942consistently reduces religiosity, especially religious belief. Education, in contrast, has no clear effect on
943religiosity after correcting for intelligence. Meisenberg, Gerhard, Heiner Rindermann, Hardik Patel, and
944Michael Woodley. 2012. “Is It Smart to Believe in God? The Relationship of Religiosity with Education
945and Intelligence.” Temas em Psicologia 20, pp.110-3, finds a large negative effect of intelligence – but no
946clear effect of education – on religiosity. McCleary, Rachel, and Robert Barro. 2006. “Religion and
947Political Economy in an International Panel.” Journal for the Scientific Study of Religion 45, pp.149-75,
948actually finds that, at the global level, education seems to increase both religious belief and behavior.
949Both studies find that former Communist countries are markedly less religious than one would expect
950given their other traits.
951xliii Uecker, Regnerus, and Vaaler 2007, pp.1683.
952xliv See Ono 2009 and Torr 2011.
953
954xlv See Schoen and Cheng 2006, pp.5-6; Isen and Stevenson 2010, pp.111-24, Lefgren and McIntyre
9552006, pp.793-5; and Torr 2011.
956xlvi See Goldstein and Kenny 2001, pp.512-6, and Torr 2011, pp.8-9.
957xlvii Härkönen and Dronkers 2006; Kalmijn 2013.
958xlviii Heaton and Blake 1999, Lefgren and McIntyre 2006, Martin 2004, Ono 2009, and Torr 2011 estimate
959education’s effect on diverse measures of marital status in the modern United States, variously controlling
960for demographics, time trends, age at marriage, marital duration, number of children, family background,
961and income.
962xlix In the GSS, I regressed dummy variables for currently married and currently divorced on age, age
963squared, year, race, sex, vocabulary scores, and church attendance, limiting the sample to 1994-2012.
964Running the same regression on GSS data for 1972-1992 reveals no effect of education on marriage or
965divorce.
966l After adding social class to the previous GSS regressions, education’s effects on marriage and divorce
967are approximately zero.
968li Moore and Ellis 2014; Isen and Stevenson 2010, pp.129-32.
969lii Balbo, Billari, and Mills 2013; Meisenberg 2008
970liii Skirbekk 2008
971liv Skirbeek, p.161.
972
973lv Lutz and KC 2011, p.590.
974lvi See generally Meisenberg 2008 and Skirrbek 2008.
975lvii The intial results come from regressing number of children (GSS variable identifier CHILDS) on years
976of education. The corrected results come from regressing number of child on years of education,
977vocabulary test score, log real family income, age, age squared, race, and sex. Controlling for spousal
978education cuts the effect of a year of respondents’ education from -.10 children to -.06 children.
979lviii Results from adding spousal education to the preceding regression, then separately analyzing male
980and female GSS responses.
981lix See especially Basu 2002.
982lx United States Bureau of Labor Statistics 2013a, 2013b.
983lxi United States Bureau of Labor Statistics 2014.
984lxii Johnson 1998
985lxiii Lucretius 1997, p.5
986lxiv Gray 2013, p.162-3.
987lxv Hofferth and Sandberg 2001, p.206; Hofferth 2009, pp.17-18.
988lxvi Clements 2004, p.72.
989
990lxvii Center on Educational Policy 2007, p.7.
991lxviii Babcock and Marks 2011, 2010.
992lxix Labaree, Someone Has to Fail, pp.137-8.
993lxx Clark 2014, p.280.