Reminding Parents about their Children’s Education

Re­search shows that bet­ter com­mu­ni­ca­tion be­tween schools and par­ents tends to sub­stan­tial­ly im­prove learn­ing out­comes in poor and rich coun­tries alike. Why is that? Is it be­cause par­ents are un­in­formed about stu­dent ef­fort, and in­for­ma­tion pro­vid­ed by the school solves that prob­lem? Or is it be­cause par­ents have lim­it­ed at­ten­tion, and mes­sages from the school in­duce them to ac­quire in­for­ma­tion about stu­dent ef­fort in­de­pen­dent­ly, solv­ing the prob­lem by them­selves?

CCWD supports the Sustainable Development Goals

The Sus­tain­able De­vel­op­ment Goals of this Pro­ject

In 2012, Vitória da Con­quista – a mu­nic­i­pal­i­ty in a poor Brazil­ian State – spent over USD 700,000 on mi­crochips em­bed­ded in pub­lic school stu­dents’ uni­forms. They hoped to de­crease tru­an­cy by in­form­ing par­ents in real-time when their chil­dren missed class­es. This pol­i­cy was in­spired by the suc­cess of in­for­ma­tion­al in­ter­ven­tions in af­fect­ing many fun­da­men­tal eco­nom­ic de­ci­sions, in­clud­ing those linked to im­proved ed­u­ca­tion­al out­comes. It is un­clear, how­ev­er, whether such in­ter­ven­tions were suc­cess­ful be­cause of the spe­cif­ic in­for­ma­tion con­veyed, tai­lored to the cir­cum­stances of the re­cip­i­ent, or be­cause such mes­sages make par­tic­u­lar is­sues top-of-mind, in­duc­ing re­cip­i­ents to re­fo­cus their at­ten­tion on that par­tic­u­lar is­sue. If the lat­ter is true, then salience in­ter­ven­tions that do not re­quire re­cip­i­ent-spe­cif­ic in­for­ma­tion (such as nudges) may pre­form just as well - and per­haps even bet­ter. In fact, it may even be the case that such in­ter­ven­tions in­duce in­di­vid­u­als to col­lect the rel­e­vant data them­selves, up­dat­ing be­liefs in the cor­rect di­rec­tion in much the same way as in­for­ma­tion­al in­ter­ven­tions would do but at a much low­er cost (no mi­crochips need­ed!). This pa­per presents first-hand ev­i­dence sup­port­ing this mech­a­nism out­side of the lab.

In the ex­per­i­ment – across 287 schools in São Paulo, Brazil, en­com­pass­ing 19,300 ninth graders – math teach­ers pro­vide week­ly in­for­ma­tion about their stu­dents’ be­hav­ior (at­ten­dance, punc­tu­al­i­ty and home­work com­ple­tion) through a plat­form over the course of 18 weeks. We ran­dom­ly as­sign par­ents to dif­fer­ent mes­sages with­in each class­room, shared via text mes­sages (SMS). Some par­ents re­ceive child-spe­cif­ic in­for­ma­tion (e.g., “Nina missed be­tween 3 and 5 math class­es over the last three weeks”), some a salience mes­sage em­pha­siz­ing the im­por­tance of pay­ing at­ten­tion to that di­men­sion (e.g., “It is im­por­tant that Nina at­tends every math class”), and still oth­ers re­ceive no mes­sage at all (the con­trol group). We find that in­for­ma­tion makes par­ents more ac­cu­rate about stu­dent at­ten­dance and sig­nif­i­cant­ly im­pacts their test scores and grade pro­mo­tion rel­a­tive to the con­trol group.

Even though salience mes­sages, in con­trast, do not make par­ents more ac­cu­rate about at­ten­dance lev­els, learn­ing out­comes in the salience group im­prove by at least as much. Why? We show that treat­ed par­ents across both con­di­tions be­come more ac­cu­rate about changes in their chil­dren’s grades over time, al­though not about grade lev­els. Such coarse be­lief up­dat­ing is con­sis­tent with in­de­pen­dent in­for­ma­tion ac­qui­si­tion in re­sponse to salience ef­fects from both in­ter­ven­tions.

The loop: un­der-ed­u­ca­tion, as pri­ma­ry care­givers do not fol­low their chil­dren’s school life close­ly.

Break­ing the loop: redi­rect­ing care­givers’ at­ten­tion to their chil­dren’s school life im­proves learn­ing out­comes even in the ab­sence of child-spe­cif­ic in­for­ma­tion.

Re­search Team

Prof. Dr. Eric Bettinger

University of Stanford

Nina Cunha

FHI 360

Guilherme Lichand

UZH

More Infos

Ricardo Madeira

Universidade de São Paulo

Carlos Belchior

UZH