Both Ways

The Monthly (Australia) - - APRIL 2018 - by Os­car Schwartz

In Oc­to­ber last year, a three-minute story on ABC Ra­dio’s AM pro­gram started with a ques­tion: “Can a com­puter mark a piece of writ­ing?” Ac­cord­ing to Les Perel­man, an Amer­i­can aca­demic who taught com­po­si­tion at MIT for two decades, the an­swer was no. “Au­to­mated es­say scor­ing doesn’t work,” he said, be­cause al­go­rithms can­not grasp de­vel­op­ment, rea­son and logic in a piece of writ­ing. Perel­man’s claim was re­futed, with equal con­fi­dence, by Stan­ley Rabi­nowitz, a PhD in ed­u­ca­tional psy­chol­ogy and the gen­eral man­ager of the Aus­tralian Cur­ricu­lum, As­sess­ment and Re­port­ing Au­thor­ity (ACARA). He coun­tered that ex­ten­sive em­pir­i­cal re­search had al­ready demon­strated that au­to­mated sys­tems can be as “re­li­able and valid” as hu­man scor­ers. Perel­man and Rabi­nowitz were de­bat­ing al­go­rith­mic func­tion­al­ity on na­tional ra­dio be­cause ACARA, the body re­spon­si­ble for ad­min­is­ter­ing NAPLAN stan­dard­ised tests in Aus­tralian schools, was plan­ning to roll out au­to­mated scor­ing to as­sess per­sua­sive or nar­ra­tive es­says for Year 5, 7 and 9 stu­dents na­tion­ally the fol­low­ing year. ACARA’s plan was to trial the tech­nol­ogy by us­ing both com­puter and hu­man mark­ers in 2018, and com­par­ing their re­sults. If no ma­jor dis­crep­an­cies be­tween al­go­rithm and teacher showed up, they would then move to full au­to­ma­tion by 2020. Perel­man had been com­mis­sioned by the NSW Teach­ers Fed­er­a­tion to re­view ACARA’s plan and found that it re­lied on an “ex­tremely du­bi­ous method­ol­ogy and in­cor­rect in­for­ma­tion”. ACARA’s chair, Steven Schwartz, hit back, claim­ing that Perel­man’s doubts were un­founded. “Com­put­ers can an­a­lyse x-rays & blood tests, drive cars, take dic­ta­tion and trans­late lan­guages,” he tweeted, “but they are un­able to mark an 8-year-old’s es­say.” For many teach­ers, in­clud­ing Dan Yore, this ques­tion of al­go­rith­mic com­pe­tence was be­side the point. Like many of his col­leagues, Yore be­lieves that au­to­mated es­say scor­ing would fur­ther en­trench a ped­a­gog­i­cal model that dam­ages the most vi­tal as­pect of ed­u­ca­tion: the teacher–stu­dent re­la­tion­ship. Yore teaches a com­bined class of Years 7 to 9 at Yir­rkala Com­mu­nity School in East Arn­hem Land. The guiding prin­ci­ple of the school’s ped­a­gogy, ac­cord­ing to its di­rec­tor, Mer­rki Ganam­barr-Stubbs, is to pro­vide the 200-odd stu­dents with a “both-ways” ed­u­ca­tion – in Yol­ngu Matha and English, in the class­room and on coun­try. This both-ways ap­proach de­rives from the con­cept of garma. When salt wa­ter from the ocean and fresh wa­ter from the land meet in a river sys­tem, a white foam pre­cip­i­tates on the sur­face. Ganam­barr-Stubbs ex­plains that this spume, or garma, is a metaphor for how new forms of knowledge rise out of the meet­ing of the Yol­ngu and Aus­tralian cur­ric­ula in the class­room. Yore first vis­ited the Yir­rkala school in 2007 dur­ing a med­i­cal place­ment in the com­mu­nity. He was so im­pressed by the both-ways ap­proach to ed­u­ca­tion that he left the med­i­cal pro­fes­sion, en­rolled in the Teach for Aus­tralia pro­gram, moved to the North­ern Ter­ri­tory, and worked for a few years as a teacher in Dar­win be­fore an op­por­tu­nity opened up in Yir­rkala. After a year of teach­ing at the school, Yore still be­lieves that the both-ways method is one of the most ground­break­ing ap­proaches to ped­a­gogy in Aus­tralia. Yet he laments how, for the most part, it is over­looked in na­tional con­ver­sa­tions about in­no­va­tion in ed­u­ca­tion. This is in part due to a common mis­con­cep­tion that in­no­va­tion has to be tech­no­log­i­cal. But it is also be­cause many of the pos­i­tive out­comes of both-ways ed­u­ca­tion are mea­sured in con­ver­sa­tions be­tween teach­ers and com­mu­nity el­ders. “When we as­sess a stu­dent’s progress we don’t just look at lit­er­acy and nu­mer­acy,” he says. “We also speak to el­ders about whether the stu­dent is ful­fill­ing his or her cul­tural and lead­er­ship roles. For me, this is a more ac­cu­rate and im­por­tant marker for how a stu­dent is track­ing than whether they hit cer­tain key cri­te­ria in an es­say.”

Ac­cord­ing to Yore, the NAPLAN model of as­sess­ment can­not as­sim­i­late th­ese qual­i­ta­tive mea­sures of progress. “When it comes to as­sess­ing stu­dent out­comes on a na­tional scale,” he says, “there is the NAPLAN way or noth­ing, which means many of the sig­nif­i­cant ben­e­fits of our both-ways ed­u­ca­tion are over­looked or mis­un­der­stood.” Hav­ing worked as the data prac­ti­tioner at his for­mer school, Yore un­der­stands the value of stan­dard­ised data in ed­u­ca­tion. “I know how ob­ser­va­tions from a dis­tance can help schools pick up pat­terns and form hy­pothe­ses.” His con­cern is that as na­tional ed­u­ca­tion pol­icy moves fur­ther to­wards an al­go­rith­mic model of ed­u­ca­tion, other cru­cial mea­sures of stu­dent de­vel­op­ment are get­ting lost. To Yore, the idea of us­ing au­to­mated es­say scor­ing is em­blem­atic of this prob­lem. “It is the pin­na­cle of a re­duc­tion­ist model of ed­u­ca­tion that as­sesses stu­dents en­tirely out­side of their class­room con­text.”

Other cru­cial mea­sures of stu­dent de­vel­op­ment are get­ting lost.

To see how the both-ways method works in prac­tice, I spent a day in the class­room with Yore and Vanessa, a Yol­ngu ed­u­ca­tor who works closely with him. Stu­dents ar­rived be­tween 8 and 8.30am, a to­tal of five girls and two boys. Yore told me it would be a quiet day be­cause there was an im­por­tant burial cer­e­mony hap­pen­ing in a nearby com­mu­nity, and ful­fill­ing cul­tural du­ties is part of the stu­dents’ ed­u­ca­tion. Once ev­ery­one had ar­rived, Yore handed out pic­ture books about the plan­ets and our so­lar sys­tem. The class was in the mid­dle of a term-long unit about the cos­mos, which was be­ing taught through the dual lens of Western science and Yol­ngu mythol­ogy. (“Yol­ngu peo­ple have been telling sto­ries about the stars for a long time!” some­one had writ­ten on the white­board.) After read­ing time was over, I led a short cre­ativewrit­ing class in which the stu­dents had to write a portrait of an undis­cov­ered planet. One girl wrote about a planet made out of bones and hot choco­late rivers. The boys, who were in­sep­a­ra­ble, imag­ined a planet called N.W.A where they met an alien called Tu­pac who had green and red teeth. The next les­son was maths. Yore handed out work­sheets on an­gles, and the stu­dents took out their pro­trac­tors. There was a big ca­pac­ity gap in the class, which Yore ne­go­ti­ated by group­ing stu­dents of sim­i­lar abil­ity to­gether and then roam­ing the room to pro­vide one-onone time with each stu­dent. As the day pro­gressed, lessons be­came less di­rec­tive. We watched part of Apollo 13, lis­tened to mu­sic in lan­guage, and did yoga. Through­out the day, the way the garma metaphor works in the class­room be­came more ob­vi­ous. The re­la­tional, both-ways ap­proach in­flu­enced ev­ery­thing, from how Yore and Vanessa in­ter­acted with each other and the stu­dents to the types of posters on the walls. There were English gram­mar and spell­ing lists, but also com­plex ma­tri­ces out­lin­ing the kin­ship re­la­tion­ships be­tween stu­dents. There was a topo­log­i­cal world map, but also a huge hand-drawn map that traced a coastal trek the kids had com­pleted ear­lier in the year with com­mu­nity el­ders. A col­lab­o­ra­tive poem writ­ten in black marker on a large sheet of butcher’s pa­per caught my at­ten­tion. “As busy as a bee / And as ralpa as ants” read one of the lines. (Ralpa is a Yol­ngu word that is hard to trans­late, but “in­dus­tri­ous” comes clos­est.) The idea that soft­ware de­signed to as­sess lin­guis­tic pro­fi­ciency at a na­tional scale could ever com­pre­hend the dy­namic and some­what chaotic process of Yol­ngu and English co-cre­ation in this class was ab­surd. Ac­cord­ing to Kate Brady, who works as a teacher in Dar­win, us­ing al­go­rithms to mark NAPLAN es­says would be just as bad for teach­ers and stu­dents in other schools across the na­tion. Brady worked as a NAPLAN es­say marker in 2014 and 2016. She says that not only did she have the op­por­tu­nity to learn more about the me­chan­ics of writ­ing and as­sess­ment but she also de­vel­oped a bet­ter ap­pre­ci­a­tion of how much ef­fort most stu­dents put into their NAPLAN writ­ten re­sponses. “I would say that 80 to 90 per cent of the es­says I marked showed cre­ativ­ity and per­sonal re­flec­tion. If they re­placed us with ro­bots, my ques­tion would be, where is the ap­pre­ci­a­tion for the ef­fort the child puts in? Where is the em­pa­thy and com­pas­sion for the stu­dent and their ideas and sto­ries?” On Jan­uary 25 this year, ACARA an­nounced that it had re­ceived in­struc­tion from the fed­eral Ed­u­ca­tion Coun­cil not to go ahead with au­to­mated es­say scor­ing, at least for now. Yore is heart­ened to hear that peo­ple in po­si­tions of power are still think­ing care­fully be­fore in­tro­duc­ing new tech­nolo­gies into ed­u­ca­tion. “But to be hon­est,” he adds, “the fact that we’re even hav­ing this de­bate re­minds me of how far off course we’ve come in our ped­a­gog­i­cal think­ing in this coun­try. It’s not enough to just slightly re­di­rect the ship; we ac­tu­ally have to turn the Ti­tanic around.”

Newspapers in English

Newspapers from Australia

© PressReader. All rights reserved.