Why Are We Expanding the Federal Role in Early-Childhood Education?

Attempts to replicate successful early-childhood education programs have not worked. We can’t afford that kind of failure on a national scale.

Attorney General Eric Holder (L) and Education Secretary Arne Duncan talk to preschoolers Dylan Hunt (2ndR) and Khalil Robinson (R) while they pretend to play doctors in their class at J.O. Wilson Elementry School. On March 21, 2014 in Washington, DC. President Obama has proposed expanding government-funded early childhood education.
National Journal
David B. Muhlhausen
Add to Briefcase
David B. Muhlhausen
April 24, 2014, 2:04 p.m.

Over the past 50 years, there has been a tre­mend­ous amount of re­search in­to the plight of the na­tion’s poor chil­dren. Com­pared with chil­dren raised in more for­tu­nate cir­cum­stances, we know that many dis­ad­vant­aged chil­dren have so­cial and aca­dem­ic skills de­fi­cits when they form­ally enter school.

Com­ment­ing on the cre­ation of Head Start in 1965 — a Great So­ci­ety preschool pro­gram in­ten­ded to help dis­ad­vant­aged chil­dren catch up to chil­dren liv­ing in more for­tu­nate cir­cum­stances — Pres­id­ent Lyn­don John­son as­ser­ted, “I be­lieve this re­sponse re­flects a real­ist­ic and a whole­some awaken­ing in Amer­ica. It shows that we are re­cog­niz­ing that poverty per­petu­ates it­self.” Ever since, the fed­er­al gov­ern­ment has been act­ively de­voted to help­ing the na­tion’s poor chil­dren catch up — spend­ing more than $202.5 bil­lion on Head Start.

Early-child­hood edu­ca­tions pro­grams, such as Head Start, are auto­mat­ic­ally as­sumed by ad­voc­ates to level the play­ing field by help­ing dis­ad­vant­aged chil­dren ar­rive at school without learn­ing de­fi­cits. From time to time, an early-child­hood edu­ca­tion pro­gram will ap­pear to work. When a par­tic­u­lar in­nov­at­ive early-child­hood edu­ca­tion pro­gram seems to pro­duce com­pel­ling evid­ence of suc­cess, poli­cy­makers and ad­voc­ates of gov­ern­ment so­cial pro­grams around the coun­try ap­pro­pri­ately take no­tice.

Such is the case with the High/Scope Perry Preschool and the Car­o­lina Abe­ce­dari­an Pro­jects — two small-scale, highly in­tens­ive early-edu­ca­tion pro­grams that served minor­ity chil­dren. Based on the ex­per­i­ences of a 58 preschool­ers and 65 chil­dren not gran­ted ac­cess to preschool, Uni­versity of Chica­go eco­nom­ist James Heck­man and his team of re­search­ers es­tim­ate that the Perry pro­gram pro­duced $7 to $12 in long-term so­ci­et­al be­ne­fits for every dol­lar in­ves­ted. The ma­jor be­ne­fit of the pro­gram is de­rived from re­duced crime. The Abe­ce­dari­an Pro­ject, and its study of 111 chil­dren, was found to have a long-term ef­fect on in­creased edu­ca­tion­al at­tain­ment with the treat­ment and con­trol groups av­er­aging 13.46 and 12.31 years of edu­ca­tion, re­spect­ively. However, the Abe­ce­dari­an Pro­ject had no meas­ur­able im­pact on in­come or crim­in­al con­vic­tions.

Based on these clearly lim­ited eval­u­ations of these pro­grams, Pres­id­ent Obama called for a large ex­pan­sion in fed­er­al fund­ing of the early-child­hood edu­ca­tion pro­grams. In his fisc­al 2015 budget pro­pos­al, Obama states: “Re­search shows that one of the best in­vest­ments we can make in a child’s life is high-qual­ity early edu­ca­tion. This year, we will in­vest in new part­ner­ships with states and com­munit­ies across the coun­try to ex­pand ac­cess to high-qual­ity early edu­ca­tion, and I am again call­ing on the Con­gress to make high-qual­ity preschool avail­able to every 4-year-old child.”

The pres­id­ent’s pro­pos­al is well-mean­ing. However, the pres­id­ent’s reas­on­ing is based upon the “single-in­stance fal­lacy.” This fal­lacy oc­curs when a per­son be­lieves that a small-scale so­cial pro­gram that ap­pears to work in one in­stance will yield the same res­ults when rep­lic­ated else­where. Com­pound­ing the ef­fects of this fal­lacy, we of­ten do not truly know why an ap­par­ently ef­fect­ive pro­gram worked in the first place. So how can we rep­lic­ate it?

There are good reas­ons to ques­tion the as­sump­tion that the fed­er­al gov­ern­ment can rep­lic­ate the be­ne­fi­cial out­comes pur­por­ted to have been caused by the Perry and Abe­ce­dari­an Pro­jects. Ig­nor­ing the fact that these stud­ies are not based upon well-im­ple­men­ted ran­dom-as­sign­ment stud­ies, the eval­u­ations of these small-scale pro­grams are out­dated. And des­pite all the hoopla, the res­ults have nev­er been rep­lic­ated. In more than 50 years, not a single ex­per­i­ment­al eval­u­ation of the Perry ap­proach ap­plied in an­oth­er set­ting or on a lar­ger-scale has pro­duced the same res­ults. The same holds true for the Abe­ce­dari­an pro­gram, which began in 1972.

Simply put,there is no evid­ence that these pro­grams can pro­duce the same res­ults today. If we really knew how these pro­grams ac­tu­ally pro­duced suc­cess, would not these res­ults have been rep­lic­ated else­where?

In ad­di­tion, the fed­er­al gov­ern­ment has a poor track re­cord of rep­lic­at­ing suc­cess­ful pro­grams on a na­tion­al scale. This point is al­most nev­er raised by ad­voc­ates of ex­pand­ing the fed­er­al gov­ern­ment’s in­volve­ment in early-child­hood edu­ca­tion pro­grams. Just con­sider what we really know about Head Start and Early Head Start.

Des­pite Head Start’s long life, the pro­gram nev­er un­der­went a thor­ough, sci­en­tific­ally rig­or­ous eval­u­ation of its ef­fect­ive­ness un­til Con­gress man­dated an eval­u­ation in 1998. Ad­voc­ates of Head Start as­ser­ted that such an eval­u­ation was un­ne­ces­sary be­cause we already knew that Head Start worked be­cause of the Perry Preschool Pro­ject.

After dec­ades of claim­ing the Perry Preschool Study in­cluded all the in­form­a­tion the coun­try needed, the Health and Hu­man Ser­vices De­part­ment com­mis­sioned a na­tion­al rep­res­ent­at­ive sample of Head Start cen­ters and ran­domly as­signed al­most 5,000 chil­dren to study and con­trol groups. The short-term res­ults meas­ur­ing the im­pact of the pro­gram at the end of Head Start were pub­lished in 2005. Re­search­ers pub­lished in­form­a­tion about the same group of chil­dren after they com­pleted kinder­garten and first grade in 2010. They fol­lowed this up with a 2012 study eval­u­at­ing the stu­dents after third grade.

The res­ults have been dis­ap­point­ing. The Head Start eval­u­ation showed that al­most all of the be­ne­fits of par­ti­cip­at­ing in Head Start dis­ap­pear by kinder­garten. Alarm­ingly, Head Start ac­tu­ally had a harm­ful ef­fect on 3-year-old par­ti­cipants once they entered kinder­garten, with teach­ers re­por­ted that non­par­ti­cip­at­ing chil­dren were more pre­pared in math skills than the chil­dren who at­ten­ded Head Start.

Early Head Start has not proven much bet­ter. Cre­ated dur­ing the 1990s, Early Head Start is a fed­er­ally fun­ded, com­munity-based pro­gram that serves low-in­come fam­il­ies with preg­nant wo­men, in­fants, and tod­dlers up to age 3. The pro­gram was in­spired by the find­ings of the Abe­ce­dari­an Pro­ject. A pair of lar­ger-scale and bet­ter-qual­ity stud­ies made pub­lic in 2005 and 2010 ex­amined the lives of chil­dren who at­ten­ded 17 Early Head Start sites se­lec­ted by the HHS. Re­search­ers eval­u­ated a group of 3,001 study and con­trol fam­il­ies when par­ti­cip­at­ing chil­dren reached age 3 and again in fifth grade.

It’s true that by the time par­ti­cipants reached age 3, Early Head Start had a few mod­est be­ne­fi­cial im­pacts on child cog­nit­ive and lan­guage de­vel­op­ment and child-so­cial-emo­tion­al growth. However, the mod­est short-term ef­fects of Early Head Start dis­ap­peared by the fifth grade.

All fed­er­al gov­ern­ment pro­grams are not fail­ures. NASA’s Apollo pro­gram was won­der­fully suc­cess­ful. Send­ing Amer­ic­ans safely to and from the moon was a sin­gu­lar achieve­ment. For that mat­ter, does any­one doubt that the ini­tial cre­ation of the in­ter­state high­way sys­tem was a suc­cess?

However, fed­er­al so­cial pro­grams in­ten­ded to im­prove hu­man be­ha­vi­or, like Head Start and Early Head Start, have not pro­duced sim­il­ar suc­cesses. The real­ity is that many Amer­ic­ans have an in­flated sense of what the fed­er­al gov­ern­ment can achieve when it comes to so­cial en­gin­eer­ing. In many ways, so­cial en­gin­eer­ing is a much more elu­sive hu­man en­deavor than build­ing rock­ets and pav­ing high­ways.

Fur­ther, the de­bate over ex­pand­ing the fed­er­al role in early-child­hood edu­ca­tion is miss­ing a much more sig­ni­fic­ant prob­lem faced by young chil­dren. As polit­ic­al sci­ent­ist Charles Mur­ray aptly poin­ted out earli­er this year, “Amer­ica has far too many chil­dren born to men and wo­men who do not provide safe, warm, and nur­tur­ing en­vir­on­ments for their off­spring — not be­cause there’s no money to be found for food, cloth­ing, and shel­ter, but be­cause they are not com­mit­ted to ful­filling the ob­lig­a­tions that child-bear­ing brings with it.”

The like­li­hood of the fed­er­al gov­ern­ment cre­at­ing a so­cial pro­gram to solve the di­lemma iden­ti­fied by Mur­ray is about as likely as the fed­er­al gov­ern­ment suc­cess­fully scal­ing up early-child­hood edu­ca­tion pro­grams. Yet, ad­voc­ates of in­creased fed­er­al spend­ing on early-child­hood edu­ca­tion pro­grams ig­nore the fed­er­al gov­ern­ment’s poor track re­cord in rep­lic­at­ing small-scale pro­grams ori­gin­ally thought to be suc­cess­ful.

And here is the prob­lem. With no sci­entif­ic cer­tainty, ad­voc­ates can­not an­swer the fol­low­ing ques­tion: Will in­creased fed­er­al spend­ing on early-child­hood edu­ca­tion pro­grams im­prove chil­dren’s fu­tures? In­stead, the de­cision to fa­vor a fed­er­al ex­pan­sion of preschool learn­ing op­por­tun­it­ies is most of­ten based on the an­swer to a less sci­en­tific­ally rig­or­ous ques­tion: Will pro­pos­ing in­creased fed­er­al spend­ing on early-child­hood pro­grams make ad­voc­ates feel that they are mak­ing a dif­fer­ence in the lives of chil­dren?

The an­swer to the lat­ter, sim­pler ques­tion is al­most cer­tainly yes. Un­for­tu­nately, this faulty de­cision mak­ing pro­cess of­ten res­ults in fed­er­al boon­doggles like Head Start and its sib­ling Early Head Start.

For the sake of tax­pay­ers, let’s not cre­ate more of them.

Dav­id B. Muhl­hausen, Ph.D., is re­search fel­low in em­pir­ic­al policy ana­lys­is in the Cen­ter for Data Ana­lys­is at the Her­it­age Found­a­tion and the au­thor of Do Fed­er­al So­cial Pro­grams Work?

What We're Following See More »
FCC Tightens Internet Privacy Standards
11 hours ago

Along party lines, the Federal Communications Commission on Thursday voted to tighten privacy standards for Internet service providers. "The regulations will require providers to receive explicit customer consent before using an individual’s web browsing or app usage history for marketing purposes. The broadband industry fought to keep that obligation out of the rules."

Obama Commutes Another 98 Sentences
11 hours ago

President Obama commuted the sentences of another 98 drug offenders on Thursday. Most of the convicts were charged with conspiracy to distribute drugs or possession with intent to distribute. Many of the sentences were commuted to expire next year, but some will run longer. Others are required to enroll in residential drug treatment as a condition of their release.

Clinton Up 9 in USA Today Poll; Up 3 According to Fox
16 hours ago

A new USA Today/Suffolk University poll finds Clinton leads Trump by 9 points nationwide, 47% to 38%. A Fox News national poll has Clinton up just three points, 44% to 41% over Trump.

Cruz: Eight Justices Could Be an Ongoing Situation
18 hours ago

Sen. Ted Cruz (R-TX) said that "there was “precedent” for a Supreme Court with fewer than nine justices—appearing to suggest that the blockade on nominee Merrick Garland could last past the election." Speaking to reporters in Colorado, Cruz said: "I would note, just recently, that Justice Breyer observed that the vacancy is not impacting the ability of the court to do its job. That’s a debate that we are going to have.”

DNC Sues RNC Over Trump’s Rigged Vote Comments
21 hours ago

The Democratic National Committee sued the Republican National Committee in U.S. District Court in New Jersey for aiding GOP nominee Donald Trump as he argues that the presidential election is "rigged." The DNC claims "that Trump's argument is designed to suppress the vote in minority communities."


Welcome to National Journal!

You are currently accessing National Journal from IP access. Please login to access this feature. If you have any questions, please contact your Dedicated Advisor.