{"id":45,"date":"2020-09-03T16:38:49","date_gmt":"2020-09-03T13:38:49","guid":{"rendered":"http:\/\/acikkitap.com.tr\/oaek\/chapter\/bolum-4-olcme-ve-ogrenme-analitiginde-kullanimi\/"},"modified":"2020-09-03T16:38:49","modified_gmt":"2020-09-03T13:38:49","slug":"bolum-4-olcme-ve-ogrenme-analitiginde-kullanimi","status":"publish","type":"chapter","link":"https:\/\/acikkitap.com.tr\/oaek\/chapter\/bolum-4-olcme-ve-ogrenme-analitiginde-kullanimi\/","title":{"raw":"B\u00d6l\u00fcm 4 \u00d6l\u00e7me ve \u00d6\u011frenme Analiti\u011finde Kullan\u0131m\u0131","rendered":"B\u00d6l\u00fcm 4 \u00d6l\u00e7me ve \u00d6\u011frenme Analiti\u011finde Kullan\u0131m\u0131"},"content":{"raw":"\n<p align=\"justify\"><a name=\"_Toc27652707\"><\/a> <span style=\"font-family: Source Sans Pro Light, serif;\"><span style=\"font-size: medium;\">Paul Prinsloo<sup>1<\/sup>, Sharon Slade<sup>2<\/sup><\/span><\/span><\/p>\n<span style=\"font-family: Source Sans Pro Light, serif;\"><span style=\"font-size: small;\"><sup>1 <\/sup>\u0130\u015fletme Y\u00f6netimi B\u00f6l\u00fcm\u00fc, G\u00fcney Afrika \u00dcniversitesi, G\u00fcney Afrika<\/span><\/span>\n\n<span style=\"font-family: Source Sans Pro Light, serif;\"><span style=\"font-size: small;\"><sup>2 <\/sup>\u0130\u015fletme ve Hukuk Fak\u00fcltesi, A\u00e7\u0131k \u00dcniversite, \u0130ngiltere<\/span><\/span>\n\n<span style=\"font-family: Source Sans Pro, serif;\"><span style=\"font-size: small;\">DOI: 10.18608\/hla17.004<\/span><\/span>\n<h2 class=\"western\">\u00d6Z<\/h2>\n<span style=\"font-size: small;\">\u00d6\u011frenme analiti\u011fi alan\u0131 olgunla\u015ft\u0131k\u00e7a ve kapsam\u0131, tan\u0131m\u0131, zorluklar\u0131 ve \u00f6\u011frenme analiti\u011fi f\u0131rsatlar\u0131n\u0131 \u00e7evreleyen s\u00f6ylemler daha da artt\u0131k\u00e7a bunlara ili\u015fkin eklemeler etik meselelerle ilgili ne kadar yol ald\u0131\u011f\u0131m\u0131z\u0131 g\u00f6zden ge\u00e7irmemizde ve gelece\u011fi g\u00f6z \u00f6n\u00fcnde bulundurmam\u0131zda fayda vard\u0131r. Bu b\u00f6l\u00fcm, kendi d\u00fc\u015f\u00fcncemizin nas\u0131l geli\u015fti\u011fine ve alandaki daha geni\u015f geli\u015fmelere kar\u015f\u0131 yolculu\u011fumuzu haritaland\u0131rmaya genel bir bak\u0131\u015f sa\u011flar. Teknolojik geli\u015fmeler ve her tarafa n\u00fcfuz eden g\u00f6zetim konusundaki endi\u015felerin artmas\u0131 ve algoritmalar\u0131n \u00f6nemi ve istenmeyen sonu\u00e7lar\u0131 ile ilgili olarak,etik ve ahlaki bir uygulama olarak \u00f6\u011frenme analiti\u011fine y\u00f6nelik ara\u015ft\u0131rmalar\u0131n geli\u015ftirilmesi, korkular\u0131n ve ger\u00e7eklerin zengin bir resmini sunar. Daha da \u00f6nemlisi, eti\u011fi ve mahremiyeti, \u00f6\u011frenme analiti\u011fi i\u00e7in \u00e7ok \u00f6nemli bir kolayla\u015ft\u0131r\u0131c\u0131 unsur olarak g\u00f6rmeye ba\u015fl\u0131yoruz. Bu b\u00f6l\u00fcm, bireysel ara\u015ft\u0131rma yolculu\u011fumuzu izlemeden \u00f6nce, geni\u015f ba\u011flamda y\u00fcksek\u00f6\u011frenimi \u015fekillendiren g\u00fc\u00e7ler ile veri ve kan\u0131t rollerine de\u011finerek alandaki mevcut \u00e7al\u0131\u015fmalar\u0131 vurgulayarak ve gelecekteki sorunlar\u0131 haritaland\u0131rarak, \u00f6\u011frenme analiti\u011fi eti\u011fini ele almaktad\u0131r.<\/span>\n\n<span style=\"font-size: small;\"><span style=\"font-family: Source Sans Pro Black, serif;\">Anahtar Kelimeler:<\/span>Etik, yapay zek\u00e2, veri, b\u00fcy\u00fck veri, \u00f6\u011frenciler<\/span>\n<p align=\"justify\">2011'de, Yeni Medya Konsorsiyumu'nun Ufuk Raporu (YMK, 2011), orta s\u0131n\u0131f bir teknolojiden ya da trendden bir y\u0131l yada daha az bir zaman diliminde geli\u015fimi fark edilecek bir seviyeye <span style=\"font-family: Source Serif Pro Light, serif;\"><i>ula\u015fan<\/i><\/span> \u00f6\u011frenme analiti\u011finin artan \u00f6nemine i\u015faret etmi\u015ftir. (YMK 2016, s. 38). \u00d6\u011frenme analiti\u011fi ile daha geli\u015fmi\u015f e\u011fitsel veri madencili\u011fi alan\u0131 aras\u0131nda a\u00e7\u0131k ba\u011flant\u0131lar olmas\u0131na ra\u011fmen otomasyon, ama\u00e7lar, k\u00f6kenler, teknikler ve y\u00f6ntemler ile di\u011ferleri aras\u0131nda da \u00f6nemli ayr\u0131mlar vard\u0131r. (Siemens ve Baker, 2012). \u00d6\u011frenme analiti\u011fi alan\u0131, farkl\u0131 bir ara\u015ft\u0131rma ve uygulama alan\u0131 olarak geli\u015ftik\u00e7e (bk. Van Barneveld, Arnold ve Campbell, 2012), etik meseleleri d\u00fc\u015f\u00fcnmek de yava\u015f yava\u015f \u00e7evreden merkeze do\u011fru ta\u015f\u0131nmaya ba\u015flam\u0131\u015ft\u0131r. Slade ve Prinsloo (2013), \u00f6\u011frenme analiti\u011finde etik odakl\u0131 bir \u015fekilde geli\u015ftirilen en eski \u00e7er\u00e7evelerden birini kurmu\u015ftur. O zamandan beri, bu alt alanda yay\u0131n yapan yazarlar\u0131n say\u0131s\u0131 \u00f6nemli \u00f6l\u00e7\u00fcde artm\u0131\u015ft\u0131r; bu da artan say\u0131da \u00e7er\u00e7eve, uygulama esaslar\u0131, taksonomilerle ve rehberle sonu\u00e7lanm\u0131\u015ft\u0131r (Ga\u0161evi\u0107, Dawson &amp; Jovanovi\u0107, 2016).<\/p>\n<p align=\"justify\">Geni\u015f kamuoyu kapsam\u0131nda artan g\u00f6zetimi ve g\u00fcvence alt\u0131na al\u0131n(ma)m\u0131\u015f ki\u015fisel verilerin toplanmas\u0131, analiz edilmesi ve kullan\u0131lmas\u0131yla ilgili, \u201ckorku ve ger\u00e7ekler genellikle ay\u0131rt edilemez \u015fekilde kar\u0131\u015ft\u0131r\u0131larak potansiyel faydalan\u0131c\u0131lar aras\u0131nda bir belirsizlik ortam\u0131na yol a\u00e7maktad\u0131r\u201d (Drachsler ve Greller, 2016, s. 89). Ga\u0161evi\u0107 vd.(2016) ayr\u0131ca, \u201ce\u011fitim uygulamalar\u0131na daha fazla destek olmak ve entegrasyona yard\u0131mc\u0131 olmak i\u00e7in ele al\u0131nmas\u0131 gereken zorluklar\u0131n bulundu\u011funu ve etik ile mahremiyeti, \u00f6\u011frenme analiti\u011fi alan\u0131ndaki \u201cengelden ziyade\u201d \u00f6nemli bir destekleyici olarak g\u00f6rd\u00fcklerini ileri s\u00fcrmektedir (s. 2).<\/p>\n<p align=\"justify\">Alandaki ki\u015fisel ara\u015ft\u0131rma yolculu\u011fumuzu haritalamadan \u00f6nce, \u00f6\u011frenme analitiklerinin etiksel \u00e7\u0131kar\u0131mlar\u0131n\u0131 g\u00f6z \u00f6n\u00fcnde bulundurma ba\u011flam\u0131n\u0131 k\u0131saca a\u00e7\u0131kl\u0131yoruz. Daha sonra son geli\u015fmeleri g\u00f6z \u00f6n\u00fcnde bulundurup daha geni\u015f ve d\u00e2hil edilmesi gereken daha \u00f6nemli bir dizi konuyu se\u00e7erek sonu\u00e7land\u0131r\u0131yoruz.<\/p>\n\n<h2 class=\"western\">BA\u011eLAMIN KURULMASI: ET\u0130\u011e\u0130N KONUYLA \u0130LG\u0130S\u0130<\/h2>\n<p align=\"justify\">\u00d6\u011frenimin gelece\u011finin, e\u011fitimin \u201cya\u015fam kalitesini ve anlaml\u0131 istihdam\u0131 sa\u011flamas\u0131 anlam\u0131nda (a) ola\u011fan\u00fcst\u00fc kalite ara\u015ft\u0131rmalar\u0131n\u0131n; (b) karma\u015f\u0131k veri toplaman\u0131n ve (c) ileri makine \u00f6\u011frenmesi ve insan \u00f6\u011frenmesinin analizi \/ deste\u011fi\" ile dijital, da\u011f\u0131t\u0131lm\u0131\u015f ve veri odakl\u0131 olaca\u011f\u0131 konusunda bir fikir birli\u011fi vard\u0131r (Siemens, 2016, Slayt 2). Her ne kadar \u00f6\u011frenme analitiklerinin etik etkilerine ili\u015fkin d\u00fc\u015f\u00fcnceler ba\u015flang\u0131\u00e7ta alanda geri planda kalsa da eti\u011fin \u00f6n plana \u00e7\u0131kmas\u0131 uzun bir yol kat etmi\u015ftir ve giderek daha da \u00f6n plana \u00e7\u0131kmaktad\u0131r (Ga\u0161evi\u0107 vd., 2016). Artan veri toplama i\u015fleminden kaynaklanan daha ba\u015far\u0131l\u0131 \u00f6\u011frenme deneyimlerinin (hem \u00f6\u011frenciler hem de kurum i\u00e7in) potansiyel <span style=\"font-family: Source Serif Pro Light, serif;\"><i>ekonomik<\/i><\/span> fayda ile ilgili \u00e7ok \u015fey s\u00f6ylenebilecek bir ba\u011flamda, \u201cveri proxy'sinden elde edilen ayr\u0131nt\u0131l\u0131 bir \u015fekilde somutla\u015ft\u0131r\u0131lm\u0131\u015f referans\u0131n\u0131 dezavantajl\u0131 hale getirdi\u011fi veri - vek\u00e2let kaynakl\u0131 zorluklar\u0131n bulunma ihtimalini g\u00f6z ard\u0131 etmemeliyiz\"(Smith, 2016, s. 16; ayr\u0131ca bk. Ruggiero, 2016; Strauss, 2016b; ve Watters, 2016).<\/p>\n<p align=\"justify\">\u00d6\u011frenci verilerinin toplanmas\u0131, analizi ve kullan\u0131lmas\u0131yla ilgili etik \u00e7\u0131kar\u0131mlar, potansiyel \u00e7\u0131kar \u00e7at\u0131\u015fmas\u0131 s\u00f6z konusu oldu\u011funda, \u00f6\u011frenciler ve kurumlar gibi bir dizi payda\u015f\u0131n talepleri dikkate al\u0131nmal\u0131d\u0131r. \u00d6\u011frenci verilerinin toplanmas\u0131ndan, analiz edilmesinden ve kullan\u0131lmas\u0131ndan kaynaklanan fayda, risk ve zarar potansiyeli hakk\u0131ndaki g\u00f6r\u00fc\u015fler, belirli bir payda\u015f\u0131n \u00e7\u0131kar\u0131na ve alg\u0131s\u0131na ba\u011fl\u0131 olacakt\u0131r. Bu b\u00f6l\u00fcmde, \u00f6ncelikle \u00f6\u011frencilerin ve kurumlar\u0131n farkl\u0131 konumsall\u0131klar\u0131, talepleri ve \u00e7\u0131karlar\u0131 hakk\u0131nda \u00f6ng\u00f6r\u00fc sunmay\u0131 umuyoruz.<\/p>\n\n<h2 class=\"western\">ET\u0130K \u0130LKELER\u0130N KURULMASI: NE KADAR YOL KATETT\u0130K?<\/h2>\n<p align=\"justify\">G\u00fcn\u00fcm\u00fczde etik yakla\u015f\u0131m ve \u00f6\u011frenci verilerinin <span style=\"font-family: Source Serif Pro Light, serif;\"><i>nas\u0131l<\/i><\/span> ve hangi \u015fartlar alt\u0131nda kullan\u0131laca\u011f\u0131na ili\u015fkin sorgulama ihtiyac\u0131 daha \u00e7ok kabul g\u00f6rm\u00fc\u015f ve daha sa\u011flam bir yer edinmi\u015f olsa da alan\u0131n ilk y\u0131llar\u0131nda bu konular \u00e7ok k\u0131y\u0131da k\u00f6\u015fede kalm\u0131\u015f konulard\u0131. \u00d6\u011frenme analiti\u011fi ile ilgili daha geni\u015f konular\u0131 ke\u015ffetmeye y\u00f6nelik ilk giri\u015fimler Vancouver'daki LAK '12'de sunuldu. Bu konferanstaki oturumlar\u0131n b\u00fcy\u00fck \u00e7o\u011funlu\u011fu geli\u015fimsel \u00e7al\u0131\u015fmaya odaklanm\u0131\u015f olarak kald\u0131. Drachsler ve Greller (2012) ba\u015fta olmak \u00fczere, \u00f6\u011frenci verilerinin nas\u0131l kullan\u0131laca\u011f\u0131na ili\u015fkin payda\u015f alg\u0131lar\u0131ndan da bahsedilmi\u015fti ancak makalelerinde ankete kat\u0131lan payda\u015flar\u0131n de\u011ferlendirmelerinin b\u00fcy\u00fck oranda gizlilik odakl\u0131 oldu\u011fu ve \u00f6zellikle tart\u0131\u015fmal\u0131 olarak kabul edilmedi\u011fi \u00f6ne s\u00fcr\u00fclm\u00fc\u015ft\u00fc. Bir ba\u015fka makale (Prinsloo, Slade ve Galpin, 2012), \u00f6\u011frencilerin \u00f6\u011frenme ba\u015far\u0131lar\u0131n\u0131 art\u0131rmak i\u00e7in, onlar\u0131n \u00f6\u011frenme yolculuklar\u0131nda <span style=\"font-family: Source Serif Pro Light, serif;\"><i>t\u00fcm<\/i><\/span> payda\u015flar\u0131n etkilerinin d\u00fc\u015f\u00fcn\u00fclmesi gerekti\u011fine de\u011finilmi\u015ftir. \u201c\u00dc\u00e7l\u00fc alan\u201d kavram\u0131, zorluklar\u0131n ve f\u0131rsatlar\u0131n haritalanmas\u0131nda yararl\u0131 bir bulgusal alan sa\u011flaman\u0131n yan\u0131 s\u0131ra, \u00f6\u011frenme analiti\u011finin \u00e7eli\u015fkilerini ve bunun \u00f6\u011frenci ba\u015far\u0131s\u0131 ve kal\u0131c\u0131 \u00f6\u011frenme \u00fczerindeki potansiyel etkisini de ortaya koymu\u015ftur. Ayn\u0131 konferansta, Campbell, DeBlois ve Oblinger'in (2007) ilk d\u00f6nem \u00e7al\u0131\u015fmalar\u0131na dayanan ve farkl\u0131 payda\u015flar\u0131n bak\u0131\u015f a\u00e7\u0131lar\u0131ndan bir dizi ilgili etik meseleyi g\u00f6z \u00f6n\u00fcnde bulundurmay\u0131 ve geni\u015fletmeyi ama\u00e7layan bir ara\u015ft\u0131rma \u00e7al\u0131\u015ftay\u0131 (Slade ve Galpin, 2012) yap\u0131lm\u0131\u015ft\u0131r.<\/p>\n<p align=\"justify\">2013 y\u0131l\u0131nda yap\u0131lan \u00e7al\u0131\u015fmalarda, verilerin nas\u0131l kullan\u0131laca\u011f\u0131 ve korunaca\u011f\u0131 konusundaki ama\u00e7lar\u0131 belirleyen mevcut kurumsal politika \u00e7er\u00e7evelerinin incelenmesine ba\u015flanm\u0131\u015ft\u0131r (Prinsloo ve Slade, 2013). B\u00fcy\u00fcyen ve geli\u015fen \u00f6\u011frenme analiti\u011fi, \u00f6\u011frenci verilerinin kullan\u0131mlar\u0131n\u0131n da h\u0131zla artt\u0131\u011f\u0131n\u0131 g\u00f6rm\u00fc\u015ft\u00fcr. Genel olarak, \u00f6\u011frenci verilerinin kurumsal kullan\u0131m\u0131na ili\u015fkin politikalar bu art\u0131\u015fa ayak uyduramam\u0131\u015ft\u0131r; \u00f6zellikle veri y\u00f6netimine, veri g\u00fcvenli\u011fine ve gizlilik konular\u0131na odaklanarak etik kayg\u0131lar\u0131 g\u00f6z \u00f6n\u00fcne alma ihtiyac\u0131na gereken \u00f6nemi vermemi\u015ftir. \u0130nceleme, mevcut politikalardaki bo\u015fluklar\u0131 ve yetersizlikleri tespit etmi\u015ftir.<\/p>\n<p align=\"justify\">Slade ve Prinsloo (2013), \u00f6\u011frenme analiti\u011finin kullan\u0131m\u0131na dair sosyo-ele\u015ftirel bir bak\u0131\u015f a\u00e7\u0131s\u0131 kullanarak \u00f6\u011frenme analiti\u011finin etik kullan\u0131m\u0131n\u0131n kapsam\u0131n\u0131 ve tan\u0131m\u0131n\u0131 etkileyen bir dizi konuyu ele alm\u0131\u015ft\u0131r. Bir dizi etik sorun birbiriyle \u00f6rt\u00fc\u015fen \u00fc\u00e7 geni\u015f kategoride grupland\u0131r\u0131lm\u0131\u015ft\u0131r. Bunlar:<\/p>\n\n<ul>\n \t<li>\n<p align=\"justify\">Verilerin yeri ve yorumlanmas\u0131<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">A\u00e7\u0131k r\u0131za, mahremiyet ve verilerin anonimle\u015ftirilmesi<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Verilerin y\u00f6netimi, s\u0131n\u0131fland\u0131r\u0131lmas\u0131 ve depolanmas\u0131d\u0131r<\/p>\n<\/li>\n<\/ul>\n<p align=\"justify\">Slade ve Prinsloo (2013) a\u015fa\u011f\u0131daki alt\u0131 ilkeye dayanan bir \u00e7er\u00e7eve \u00f6nermi\u015ftir:<\/p>\n\n<ol>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenme Analiti\u011finin ahlaki uygulamas\u0131n\u0131n sadece neyin ge\u00e7erli oldu\u011fu hususunun yan\u0131s\u0131ra neyin uygun ve ahlaki olarak gerekli oldu\u011fu \u00fczerinde odaklanmas\u0131<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frencilerin sadece m\u00fcdahale ve hizmetlerin al\u0131c\u0131s\u0131 olmay\u0131p ayn\u0131 zamanda ortak ve birli\u011fi yap\u0131lacak arac\u0131lar olmas\u0131<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Ge\u00e7ici dinamik yap\u0131lar olarak \u00f6\u011frenci kimli\u011fi ve performans\u0131 analiti\u011fin belirli bir zamanda ve ba\u011flamda \u00f6\u011frencinin anl\u0131k g\u00f6r\u00fcnt\u00fcs\u00fcn\u00fc sa\u011flad\u0131\u011f\u0131n\u0131n fark\u0131nda olunmas\u0131<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00c7ok boyutlu, karma\u015f\u0131k bir olgu olarak \u00f6\u011frenci ba\u015far\u0131s\u0131<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Bireyin kimli\u011finin korunmas\u0131, verilere eri\u015fim, hangi verilerin hangi \u015fartlar alt\u0131nda hangi ama\u00e7lar i\u00e7in kullan\u0131laca\u011f\u0131 konusunda \u00f6nem ta\u015f\u0131yan \u015feffafl\u0131k<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Y\u00fcksek\u00f6\u011frenimin veri kullanmamay\u0131 g\u00f6ze alamamas\u0131<\/p>\n<\/li>\n<\/ol>\n<p align=\"justify\">Bu ilkeler, faydal\u0131 bir ba\u015flang\u0131\u00e7 pozisyonu sunar ancak (hangi \u015fartlar alt\u0131nda) kimin yarar\u0131na oldu\u011fu hususunda tam bir anlay\u0131\u015f\u0131n geli\u015ftirilmesi gibi birtak\u0131m pratik hususlar dikkate al\u0131narak desteklenmelidir. Onay al\u0131narak kurumsal pozisyonlar\u0131n olu\u015fturulmas\u0131, tan\u0131mlar\u0131n belirlenmesi ve reddedilmesi konusunda; g\u00fcvenlik a\u00e7\u0131\u011f\u0131 ve zarar\u0131na ili\u015fkin sorunlar (\u00f6r. yanl\u0131\u015fl\u0131kla etiketleme), tazminat sistemleri (hem \u00f6\u011frenci hem de kurum i\u00e7in), veri toplama, analizler, eri\u015fim ve depolama (\u00f6r. g\u00fcvenlik sorunlar\u0131 ve \u00f6nyarg\u0131n\u0131n s\u00fcreklili\u011finden ka\u00e7\u0131nma) ile y\u00f6netim ve kaynak tahsisi kar\u015f\u0131lanmal\u0131d\u0131r.<\/p>\n<p align=\"justify\">Kaynak tahsisinin bu ikinci y\u00f6n\u00fc, e\u011fitim triyaj\u0131 kavram\u0131 g\u00f6z \u00f6n\u00fcne al\u0131narak daha sonraki bir makalede daha ayr\u0131nt\u0131l\u0131 incelenmi\u015ftir (Prinsloo ve Slade, 2014a). \u00d6\u011frenme analiti\u011fi, Y\u00d6K'lere (y\u00fcksek\u00f6\u011fretim kurumlar\u0131) ba\u015far\u0131s\u0131z veya b\u0131rakma riski alt\u0131nda olan \u00f6\u011frencileri proaktif olarak tan\u0131tmak ve desteklemek i\u00e7in <span style=\"font-family: Source Serif Pro Light, serif;\"><i>teorik<\/i><\/span> f\u0131rsatlar sunsa da bunlar\u0131 kaynaklar\u0131n gittik\u00e7e s\u0131n\u0131rland\u0131\u011f\u0131 bir ba\u011flamda yapar. O halde zorluk, destek kaynaklar\u0131n\u0131 y\u00f6nlendirmede en <span style=\"font-family: Source Serif Pro Light, serif;\"><i>iyi yerin neresi<\/i><\/span> oldu\u011fu ve bu karar\u0131n hangi temele dayanarak verildi\u011finin bilinmesidir. Deste\u011fin \u201chayatta kalmas\u0131\u201d muhtemel olan \u00f6\u011frencilere y\u00f6nelik y\u00f6nlendirmenin bir arac\u0131 olarak e\u011fitim \u00f6nceli\u011fi kavram\u0131, \u00f6\u011frenci \u00f6zerkli\u011fine sayg\u0131 g\u00f6sterme ve ayn\u0131 zamanda kurumun uzun vadeli s\u00fcrd\u00fcr\u00fclebilirli\u011fini sa\u011flama aras\u0131ndaki denge, yararlanma (her zaman \u00f6\u011frencinin \u00e7\u0131kar\u0131na ilgi g\u00f6stermek i\u00e7in), zarar vermeme ihtiyac\u0131 (en az zararla en faydal\u0131 sonuca varmak) ve da\u011f\u0131t\u0131c\u0131 bir adalet duygusu sa\u011flama (demografik \u00f6zelliklerin sa\u011flanan deste\u011fi, varsay\u0131mlar\u0131, bunlar\u0131 tan\u0131ma ve \u00e7\u00f6zme ihtiyac\u0131n\u0131 etkiledi\u011fini anlamak) gibi bir dizi karma\u015f\u0131k konunun dikkatlice de\u011ferlendirilmesini gerektirir.<\/p>\n<p align=\"justify\">\u00d6\u011frencinin haberinin olmas\u0131n\u0131 gerektirmeden <span style=\"font-family: Source Serif Pro Light, serif;\"><i>bir \u015feyler yapman\u0131n bir yolu<\/i><\/span> olarak \u00f6\u011frenme analiti\u011fi bilincindeki art\u0131\u015f; g\u00f6zlem, \u00f6\u011frenci mahremiyeti ve kurumsal g\u00fcvenilirlik konular\u0131n\u0131n daha da ara\u015ft\u0131r\u0131lmas\u0131n\u0131 tetiklemi\u015ftir (Prinsloo ve Slade, 2014b). Sonu\u00e7ta ortaya \u00e7\u0131kan tart\u0131\u015fma \u00f6\u011frenci \u00f6\u011frenmesinin do\u011fru, nesnel, tamamen eksiksiz resimlerinin bir \u00fcreticisi olarak \u00f6\u011frenme analiti\u011fi etraf\u0131ndaki varsay\u0131mlara meydan okuyarak kurum ve \u00f6\u011frenci aras\u0131ndaki e\u015fit olmayan olas\u0131 ili\u015fkiyi yeniden ele alm\u0131\u015ft\u0131r. Ki\u015fisel verilerin kullan\u0131m\u0131 ve analizi ile ilgili mevcut \u00e7er\u00e7eveler g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, \u00e7al\u0131\u015fma, <span style=\"font-family: Source Serif Pro Light, serif;\"><i>\u00f6\u011frenci merkezli<\/i><\/span> bir \u00f6\u011frenme analiti\u011fi i\u00e7in temel olu\u015fturabilecek alt\u0131 unsur \u00f6nermektedir:<\/p>\n\n<ol>\n \t<li>\n<p align=\"justify\">K\u00fcmelenmi\u015f, ki\u015fiselle\u015ftirilmemi\u015f verilerin kullan\u0131m\u0131, etkili ve uygun bir \u00f6\u011fretme ve \u00f6\u011frenme sa\u011flamak i\u00e7in esast\u0131r ancak \u00f6\u011frenciler giri\u015f \/ \u00e7\u0131k\u0131\u015f kararlar\u0131n\u0131 bilin\u00e7li bir \u015fekilde yapabilmelidirler<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenciler hangi verilerin topland\u0131\u011f\u0131 ve nas\u0131l kullan\u0131ld\u0131\u011f\u0131 konusunda tam bilgi sahibi olmal\u0131d\u0131rlar<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenciler ki\u015fisel veri kay\u0131tlar\u0131n\u0131n eksiksiz ve g\u00fcncel oldu\u011fundan emin olmal\u0131d\u0131rlar<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Etkinliklerin izlenmesi ve verilerin toplanmas\u0131 \u00f6\u011frenci ilerlemesine zarar vermemelidir<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Algoritmik \u00e7\u0131kt\u0131 (potansiyel) insan incelemesine tabi tutulmal\u0131 ve gerekirse d\u00fczeltilmelidir<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenme analitikleri temel olarak ba\u011flam ve zamana \u00f6zg\u00fc, ge\u00e7ici, tamamlanmam\u0131\u015f \u00f6\u011frencilerin portrelerini sa\u011flamal\u0131, algoritmalar s\u0131kl\u0131kla g\u00f6zden ge\u00e7irilmeli ve do\u011frulanmal\u0131d\u0131r<\/p>\n<\/li>\n<\/ol>\n<p align=\"justify\">G\u00f6zetime ili\u015fkin sorunlar ve \u00f6\u011frencilerin kendi verilerinin kullan\u0131m\u0131n\u0131n aktif faili olarak tan\u0131nmas\u0131 gereklili\u011fi, A\u00e7\u0131k \u00dcniversitenin (A\u00dc; 2014) \u00f6\u011frenci analiti\u011finin \u00f6\u011frenci verilerinin etik kullan\u0131m\u0131 politikas\u0131n\u0131n geli\u015ftirilmesi kapsam\u0131nda a\u00e7\u0131k\u00e7a ele al\u0131nm\u0131\u015ft\u0131r. Payda\u015flarla isti\u015farenin bir par\u00e7as\u0131 olarak, 50 \u00f6\u011frenciden olu\u015fan bir temsilci grubu, \u00fc\u00e7 haftal\u0131k bir s\u00fcre zarf\u0131nda \u00e7al\u0131\u015fma hedeflerini tamamlamada \u00f6\u011frencilere destek olmak i\u00e7in verilerin nas\u0131l kullan\u0131ld\u0131\u011f\u0131na dair yakla\u015f\u0131mlar\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Bir anket ara\u015ft\u0131rmas\u0131 (Slade ve Prinsloo, 2014), \u00f6\u011frencilerin, verilerin aktif olarak toplanma ve kullan\u0131lma derecelerinin b\u00fcy\u00fck \u00f6l\u00e7\u00fcde fark\u0131nda olmad\u0131klar\u0131n\u0131 ve birtak\u0131m endi\u015feleri dile getirdiklerini ortaya koymu\u015ftur. \u00d6\u011frencilerin \u00e7o\u011funlu\u011fuyla aktif olarak r\u0131za g\u00f6sterme (ya da g\u00f6stermeme) potansiyeli ile ilgili temel kayg\u0131, sonradan bu tercihlerinden vazge\u00e7ebilme iste\u011finin olmas\u0131d\u0131r. \u00d6\u011frencilerin bu do\u011frudan kat\u0131l\u0131m\u0131, \u00f6\u011frenme analiti\u011fi eti\u011fiyle ilgili bir politikay\u0131 \u015fekillendirmede s\u00f6z konusudur; \u00f6\u011frencilerin verilerini dikkatli bir \u015fekilde korunmas\u0131n\u0131 ve hatta daha dikkatli bir \u015fekilde uygulanmalar\u0131 i\u00e7in de\u011ferli bir varl\u0131k olarak g\u00f6r\u00fclmesi y\u00f6n\u00fcnde benzersiz bir fikir vermi\u015ftir. Slade ve Prinsloo'daki (2014) \u00f6rneklemin toplam evreni tam olarak temsil edemeyebilece\u011fi g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, sonu\u00e7lar kurumsal ve jeopolitik ba\u011flamda genellenemez.<\/p>\n<p align=\"justify\">Bu artan \u00f6\u011frenci kayg\u0131s\u0131 konusundaki fark\u0131ndal\u0131\u011fa cevaben, Prinsloo ve Slade (2015), \u00f6\u011frencilerin mahremiyetle ilgili tutumlar\u0131n\u0131 \u00e7evreleyen varsay\u0131mlar\u0131m\u0131z\u0131n ve anlay\u0131\u015f\u0131m\u0131z\u0131n hem halk\u0131n ya\u015famlar\u0131n\u0131n detaylar\u0131n\u0131 payla\u015fmakta g\u00f6r\u00fclen belirgin kolayl\u0131ktan hem de b\u00fcy\u00fck \u00f6l\u00e7\u00fcde ataerkil kurum k\u00fclt\u00fcrlerinden etkilenip etkilenmedi\u011fini sorgulam\u0131\u015ft\u0131r. \u00c7al\u0131\u015fma, \u00f6\u011frencilerin verilerini izlemelerine izin vermemelerini veya se\u00e7memelerini sa\u011flamak i\u00e7in onay verme ve g\u00f6r\u00fcn\u00fc\u015fte basit bir se\u00e7im konusunda ya\u015fad\u0131klar\u0131 sorunlar\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Tart\u0131\u015fman\u0131n temeli olarak, \u00fc\u00e7 kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i ders (KA\u00c7D) tedarik\u00e7isinin \u015fart ve ko\u015fullar\u0131, kullan\u0131c\u0131lara verilerinin kullan\u0131m\u0131yla ilgili verilen bilgileri belirlemek i\u00e7in g\u00f6zden ge\u00e7irildi. Bu Y\u00d6K'lerin \u00f6\u011frencileri ki\u015fisel verilerindeki \u00f6\u011frenme analiti\u011fi \u00fczerindeki etkileri hakk\u0131nda tam olarak bilgilendiren bir yakla\u015f\u0131ma do\u011fru nas\u0131l hareket edebileceklerine dair bir tart\u0131\u015fmaya evrilmi\u015ftir. Benzer bir tema Prinsloo ve Slade'de (2016a) izlenmi\u015ftir. Bu makale, bir\u00e7ok Y\u00d6K'\u00fcn \u00f6\u011frenci verilerine otoriter bir yakla\u015f\u0131m benimseme e\u011filimine itiraz etmi\u015ftir. \u00d6\u011frenme analitiklerinin yay\u0131lmas\u0131ndaki h\u0131zl\u0131 b\u00fcy\u00fcmeye ra\u011fmen, baz\u0131 Y\u00d6K'lerin d\u00fczenleyici \u00e7er\u00e7eveleri vard\u0131r ve \/ veya toplanan, analiz edilen, kullan\u0131lan ve payla\u015f\u0131lan \u00f6\u011frenci verilerinin kapsam\u0131 konusunda tamamen \u015feffaft\u0131rlar. \u00d6\u011frenci hassasl\u0131\u011f\u0131 \u00f6\u011frenme analiti\u011fi potansiyelini ger\u00e7ekle\u015ftirmek ve Y\u00d6K'lerin g\u00fc\u00e7 ili\u015fkileri, asimetrik bilgi ve \u00f6\u011frencilerle g\u00fc\u00e7 ili\u015fkileri ba\u011flam\u0131nda ve \u00f6\u011frenci eylemlili\u011fini \u00e7evreleyen \u00f6\u011frenme analiti\u011fi konusundaki karma\u015f\u0131kl\u0131klar aras\u0131ndaki ba\u011fda incelenmi\u015ftir. Ama\u00e7, \u00f6\u011frenci k\u0131r\u0131lganl\u0131\u011f\u0131n\u0131n nas\u0131l ele al\u0131nabilece\u011fini, \u00f6\u011frenci eylemlili\u011fini art\u0131rarak ve onlar\u0131 nicelikli veri nesnelerinden nitelikli ve kalifiye ki\u015filere do\u011fru ilerleyen \u00f6\u011frenme analiti\u011fi aktif kat\u0131l\u0131mc\u0131lar\u0131 olarak g\u00fc\u00e7lendirme yollar\u0131n\u0131 d\u00fc\u015f\u00fcnmektir (ayr\u0131ca bk. Prinsloo ve Slade, 2016b).<\/p>\n\n<h2 class=\"western\">ET\u0130K \u00c7ER\u00c7EVELER \u0130LE \u0130LG\u0130L\u0130 SON GEL\u0130\u015eMELER<\/h2>\n<p align=\"justify\">Verinin, y\u00fckselen de\u011fi\u015f toku\u015f edilebilme de\u011feriyle birlikte, payla\u015f\u0131labilir ticari bir \u00fcr\u00fcn olarak artan de\u011ferinin, yasal ve geleneksel etik \u00e7er\u00e7evelerimizi a\u015ft\u0131\u011f\u0131 genel kabul g\u00f6rmektedir (Zhang, 2016). \u201cDerin ekonomik bask\u0131lar, ba\u011flant\u0131n\u0131n yo\u011funla\u015fmas\u0131n\u0131 ve \u00e7evrimi\u00e7i izlemenin g\u00fc\u00e7lenmesini sa\u011flamaktad\u0131r\u201d (Couldry, 2016, b\u00f6l.13) ve \u201cihtiya\u00e7 duyulan \u015fey, bizim etik ya\u015fam olanaklar\u0131m\u0131z i\u00e7in kapitalizmin yeni veri ili\u015fkileri maliyetleri olup (Couldry, 2016, b\u00f6l. 35) daha kolektif bir yans\u0131mas\u0131d\u0131r\". Bu nedenle, \u00f6\u011frenme analiti\u011fi etik uygulamalar\u0131 ile u\u011fra\u015fmak i\u00e7in farkl\u0131 jeopolitik ve kurumsal ba\u011flamlarda giri\u015fimler olmu\u015ftur. Sclater, Peasgood ve Mullan (2016), \u00f6rne\u011fin, Amerika Birle\u015fik Devletleri, Avustralya ve Birle\u015fik Krall\u0131k'ta y\u00fcksek\u00f6\u011fretimdeki uygulamalar\u0131 g\u00f6zden ge\u00e7irmi\u015ftir. Bulgular\u0131n\u0131, \u00f6\u011frenme analitiklerinin 1) kalite g\u00fcvencesi ve kalite iyile\u015ftirmede 2) tutma oranlar\u0131n\u0131n artt\u0131r\u0131lmas\u0131nda; 3) \u00f6\u011frenci pop\u00fclasyonu aras\u0131ndaki farkl\u0131 sonu\u00e7lar\u0131 de\u011ferlendirme ve bunlara etki etmede ve 4) uyarlanabilir \u00f6\u011frenmenin geli\u015fimi ve tan\u0131t\u0131lmas\u0131nda \u00f6nemli katk\u0131larda bulundu\u011funu belirterek \u00f6zetlemektedirler. Rapor bir\u00e7ok imk\u00e2n\u0131 tan\u0131sa da ayn\u0131 zamanda \u201cetik ve veri gizlili\u011fi sorunlar\u0131, \"a\u015f\u0131r\u0131 analiz\" ve sonu\u00e7lar\u0131n genelle\u015ftirilememesi, modellerin yanl\u0131\u015f s\u0131n\u0131fland\u0131r\u0131lmas\u0131 ve \u00e7eli\u015fkili bulgular\u201d gibi tehditleri de vurgulamaktad\u0131r (s. 16). \u0130ncelemelerinde, \u00f6\u011frenme analiti\u011findeki etik kayg\u0131lar\u0131 ele almak i\u00e7in politika d\u00fczeyinde bir teklifin kurumsal bir \u00f6rne\u011fi A\u00e7\u0131k \u00dcniversite (\u0130ngiltere) dir. 2014 y\u0131l\u0131nda, A\u00dc, toplanan ve analiz edilen verilerin niteli\u011fini ve kapsam\u0131n\u0131 s\u0131n\u0131rlayan \u201c\u00f6\u011frenme analiti\u011fi i\u00e7in \u00f6\u011frenci verilerinin etik kullan\u0131m\u0131 politikas\u0131\u201d n\u0131 ve toplanan \u00f6\u011frenme analiti\u011fi i\u00e7in <span style=\"font-family: Source Serif Pro Light, serif;\"><i>kullan\u0131lmayacak<\/i><\/span> verilerin a\u00e7\u0131k bir spesifikasyonunu yay\u0131nlam\u0131\u015ft\u0131r. Politika a\u015fa\u011f\u0131daki sekiz prensibi i\u00e7erir (s. 6):<\/p>\n\n<ol>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenme analiti\u011fi, lisans d\u00fczeyinde \u00e7al\u0131\u015fmaya a\u00e7\u0131k giri\u015f gibi temel d\u00fczenleme ilkeleri ile uyumlu olmas\u0131 gereken etik bir uygulamad\u0131r.<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">A\u00dc, m\u00fcmk\u00fcn oldu\u011funda \u00f6\u011frencilerin yarar\u0131na, \u00f6\u011frenci verilerinden anlam \u00e7\u0131karmak ve bunlar\u0131 kullanmak i\u00e7in t\u00fcm payda\u015flara kar\u015f\u0131 sorumludur.<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenciler tamamen g\u00f6zle g\u00f6r\u00fclebilir veriler veya bu verileri yorumlamam\u0131zla tan\u0131mlanmamal\u0131d\u0131r. [Dahas\u0131, bu ilke normlara uyan \u00f6\u011frencilere kar\u015f\u0131 uyar\u0131 verir ve genel kal\u0131plara uymayan bireyleri kabul eder. Bu ilke ayr\u0131ca, \u00e7al\u0131\u015fanlar\u0131n analizin g\u00fcvenilirli\u011fini ciddi \u015fekilde etkileyebilecek t\u00fcm veri k\u00fcmesine eri\u015femeyebilece\u011fini de a\u00e7\u0131k\u00e7a ortaya koymaktad\u0131r.]<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenme analitiklerinin kullan\u0131m\u0131na ili\u015fkin ama\u00e7 ve s\u0131n\u0131rlar iyi tan\u0131mlanm\u0131\u015f ve g\u00f6r\u00fcn\u00fcr olmal\u0131d\u0131r.<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00dcniversite veri toplama konusunda \u015feffaft\u0131r ve \u00f6\u011frencilere d\u00fczenli aral\u0131klarla kendi verilerini g\u00fcncelleme f\u0131rsat\u0131 sunar.<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u00d6\u011frenciler \u00f6\u011frenme analiti\u011finin uygulanmas\u0131nda etkin \u00f6zne olarak kullan\u0131lmal\u0131d\u0131r (\u00f6r. A\u00e7\u0131k r\u0131za, ki\u015fiselle\u015ftirilmi\u015f \u00f6\u011frenme yollar\u0131, m\u00fcdahaleler).<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Verilerin analizine dayanan modelleme ve m\u00fcdahaleler sa\u011flam ve \u00f6nyarg\u0131s\u0131z olmal\u0131d\u0131r.<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">A\u00dc i\u00e7indeki \u00f6\u011frenme analitiklerini benimsemek, de\u011ferlerin ve faydalar\u0131n (kurum k\u00fclt\u00fcr\u00fc) geni\u015f kabul\u00fcn\u00fc ve kurum genelinde uygun becerilerin geli\u015ftirilmesini gerektirir.<\/p>\n<\/li>\n<\/ol>\n<p align=\"justify\">Bu politika ve ilkeleri \u00f6\u011frenci verilerinin toplanmas\u0131nda, analiz edilmesinde ve kullan\u0131lmas\u0131ndaki etik etkilere ilk kurumsal tepkilerden biri olarak, ke\u015ffedilmemi\u015f b\u00f6lgeleri haritalamaya \u00e7al\u0131\u015fm\u0131\u015ft\u0131r. \u00d6zel ilgi alan\u0131, \u201c\u00f6\u011frencinin, verisinin bir k\u0131sm\u0131n\u0131n veya tamam\u0131n\u0131n \u00f6\u011frenme analiti\u011fi i\u00e7in kullan\u0131labilece\u011fi ve r\u0131za g\u00f6sterdi\u011fi ama\u00e7lar\u0131n bilincinde oldu\u011fu s\u00fcrece yap\u0131lan \u201ca\u00e7\u0131k r\u0131za \u201dtan\u0131m\u0131d\u0131r. Bilgilendirilmi\u015f onay, bir mod\u00fcl veya yeterlilik i\u00e7in rezervasyon veya kay\u0131t s\u0131ras\u0131nda uygulan\u0131r \u201d(A\u00e7\u0131k \u00dcniversite, 2014, s. 3). Politika, verilerin toplanmas\u0131n\u0131, analiz edilmesini ve kullan\u0131lmamas\u0131n\u0131 tercih eden \u00f6\u011frencilerin olas\u0131l\u0131\u011f\u0131n\u0131 ele almamaktad\u0131r (Engelfriet, Manderveld ve Jeunink, 2015; Sclater, 2015; ayr\u0131ca bk. Shacklock, 2016).<\/p>\n<p align=\"justify\">Avustralya ba\u011flam\u0131nda analitik uygulamalar\u0131n\u0131n \u00f6\u011frenilmesine genel bir bak\u0131\u015fta, Dawson, Ga\u0161evi\u0107 ve Rogers (2016) \u201c\u00e7al\u0131\u015fmalar boyunca eti\u011fe verilen g\u00f6receli sessizli\u011fin \u00f6nemli oldu\u011funu\u201d (s. 3) ve bunun \u201c sekt\u00f6r\u00fcn bu konular\u0131 dikkate almas\u0131 gereken ciddiyet\u201d (s. 33), oldu\u011funu belirtmi\u015ftir. Kurumlar olgunla\u015ft\u0131k\u00e7a, etik kayg\u0131lar\u0131n daha da belirginle\u015fmekte oldu\u011fu savunulsa da rapora g\u00f6re y\u00fcksek\u00f6\u011frenim sekt\u00f6r\u00fcn\u00fcn daha \u00f6nce b\u00f6yle bir sohbete haz\u0131r olmad\u0131\u011f\u0131 muhtemeldir (s. 33)<\/p>\n<p align=\"justify\">Ayr\u0131ca Avustralya ba\u011flam\u0131nda, Galce ve McKinney (2015) \u201cdisiplinin kurumlar, pratisyenler ve teknoloji sat\u0131c\u0131lar\u0131 ile g\u00f6receli olgunla\u015fmam\u0131\u015fl\u0131\u011f\u0131\u201d ba\u011flam\u0131nda, \u201cneyin i\u015fe yarad\u0131\u011f\u0131n\u0131\u201d ve uygulamalar\u0131n s\u0131n\u0131rlar\u0131n\u0131 bulmay\u0131 ve k\u00f6t\u00fcye kullanma potansiyelini bulma ihtiyac\u0131na dikkat \u00e7ekmektedir (s. 588). \u00dcniversitenin a) \u00f6\u011frenme ve \u00f6\u011fretme ile do\u011frudan ilgili olmayan b) \u00f6\u011frencilerin \u00fcniversitenin b\u00f6yle bir veri koleksiyonu olu\u015fturmas\u0131n\u0131 beklemedi\u011fi veri kaynaklar\u0131n\u0131 bar\u0131nd\u0131ran bir \u00f6\u011frenme analiti\u011fi kullanmayaca\u011f\u0131na dair taahh\u00fcd\u00fc \u00f6nemlidir (s. 590). \u00d6\u011frenci verileri, yaln\u0131zca s\u00f6z konusu verilerin topland\u0131\u011f\u0131 orijinal ama\u00e7 ba\u011flam\u0131nda kullan\u0131lacak olup a\u015fa\u011f\u0131daki \u015fartlar alt\u0131nda kullan\u0131m\u0131 devam edebilir:<\/p>\n<p align=\"justify\">A\u00e7\u0131k \u015fekilde bilgilendirilmi\u015f onay \u00f6l\u00e7\u00fcme konu olanlardan sa\u011flan\u0131r. Bilgilendirilmi\u015f onay \u015fu anlama gelir: (a) hangi verilerin topland\u0131\u011f\u0131 veya toplanabilece\u011fi, neden ve nas\u0131l topland\u0131\u011f\u0131, nas\u0131l depoland\u0131\u011f\u0131 ve nas\u0131l kullan\u0131ld\u0131\u011f\u0131 hakk\u0131nda a\u00e7\u0131k ve do\u011fru bilgiler verilir ve (b) a\u00e7\u0131klanan uygulamaya \/ uygulamalara serbest\u00e7e izin verilir (s.590).<\/p>\n<p align=\"justify\">Yukar\u0131daki ilkeler, \u00f6\u011fretme ve \u00f6\u011frenmeyi geli\u015ftirmek ve \u00f6\u011frencilere \u201c\u00f6\u011frenmeleri \u00fczerinde daha fazla kontrol ve sorumluluk sa\u011flamak\u201d i\u00e7in toplanan verilerin nas\u0131l kullan\u0131lmas\u0131 gerekti\u011fine ili\u015fkin olarak kalan iki ilkeyle birlikte okunmal\u0131d\u0131r (s. 591). Tam bir tart\u0131\u015fma i\u00e7in, bk. Galce ve McKinney (2015).<\/p>\n<p align=\"justify\">Drachsler ve Greller (2016) etik, mahremiyet ve ilgili yasal \u00e7er\u00e7evelere genel bir bak\u0131\u015f sa\u011flar ve veri toplay\u0131c\u0131 ile veri nesnesi aras\u0131ndaki asimetrik g\u00fc\u00e7 ili\u015fkisi \u0131\u015f\u0131\u011f\u0131nda ger\u00e7ek kullan\u0131m olas\u0131l\u0131\u011f\u0131, m\u00fclkiyet sorunlar\u0131, anonimlik ve veri g\u00fcvenli\u011fi, gizlilik ve veri kimli\u011finin yan\u0131 s\u0131ra \u015feffafl\u0131k ve g\u00fcven unsurlar\u0131 hakk\u0131nda da bilgi verir. \u00d6\u011frenme analitiklerinin \u201cveri toplama ve i\u015fleme politikalar\u0131na ba\u011fl\u0131 korkular\u0131n \u00fcstesinden gelmek i\u00e7in\u201d kabul edilebilir ve uyumlu bir \u015fekilde ilerlemesini sa\u011flamak amac\u0131yla bir kontrol listesi (DELICATE \u00a9) sunarlar (s. 96).<\/p>\n<p align=\"justify\">Sclater (2015), \u00fcst d\u00fczey y\u00f6netim, analiz komitesi, veri bilimcileri, e\u011fitim ara\u015ft\u0131rmac\u0131lar\u0131, BT ve \u00f6\u011frenciler gibi bir dizi payda\u015f\u0131n \u00f6\u011frenme analiti\u011finden nas\u0131l etkilendi\u011fi ve ona ili\u015fkin nas\u0131l sorumluluk ald\u0131\u011f\u0131na dair genel bir bak\u0131\u015f ile \u00f6\u011frenme analiti\u011finde etik, yasal ve lojistik konulara dair bir (taslak) taksonomi \u00f6nerir. Taslak, ba\u015fka \u015feylerin yan\u0131 s\u0131ra, r\u0131za; kimlik; kapsam d\u0131\u015f\u0131 b\u0131rakman\u0131n olas\u0131 etkileri; kurum ve \u00f6\u011frenciler aras\u0131ndaki asimetrik ili\u015fki; (s\u0131n\u0131rlar) \u00f6\u011frenci verilerinin izin verilen kullan\u0131mlar\u0131; \u015feffafl\u0131k; kullan\u0131m d\u00e2hil (ve hari\u00e7) veriler ve di\u011ferleri aras\u0131nda \u00f6\u011frenci \u00f6zerkli\u011fini kapsar. Etik kayg\u0131lar\u0131n tam listesi i\u00e7in Sclater (2015)'a bak\u0131n\u0131z.<\/p>\n<p align=\"justify\">Hollanda'daki y\u00fcksek\u00f6\u011frenim ba\u011flam\u0131nda Engelfriet vd. (2015), Ki\u015fisel Bilgilerin Korunmas\u0131 Kanununun, \u00f6\u011frenme analiti\u011fi \u00fczerindeki etkilerini g\u00f6z \u00f6n\u00fcne almaktad\u0131r. Bunlar, izin ihtiyac\u0131n\u0131 (ve r\u0131zan\u0131n al\u0131nmas\u0131ndan do\u011facak sorumlulu\u011fu) ve bir hizmet sa\u011flay\u0131c\u0131 ile al\u0131c\u0131 aras\u0131nda, hizmetin sa\u011flanmas\u0131 i\u00e7in ihtiya\u00e7 duyulan herhangi bir ki\u015fisel bilgiyi kullanabilece\u011fi konusunda yap\u0131lan onay anla\u015fmas\u0131n\u0131n sonu\u00e7lar\u0131n\u0131 i\u00e7erir. Yasa, <span style=\"font-family: Source Serif Pro Light, serif;\"><i>temel<\/i><\/span> bilgiler ile \u201ckullan\u0131\u015fl\u0131\u201d bilgiler aras\u0131nda ayr\u0131m yapar. Engelfriet vd. (2015) \u00f6\u011frenme analitiklerinin yeni ortaya \u00e7\u0131kan bir uygulama olarak g\u00f6r\u00fcld\u00fc\u011f\u00fc g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, g\u00fcvenli bir \u015fekilde \u201ckullan\u0131\u015fl\u0131\u201d bilgi toplanmas\u0131 olarak kabul edilebilece\u011fi ve belki de kurum ve \u00f6\u011frenciler aras\u0131ndaki fikir birli\u011fi anla\u015fmas\u0131ndan muaf tutulabilece\u011fi y\u00f6n\u00fcnde muhalefetli bir g\u00f6r\u00fc\u015f benimsemektedir. Yazarlar bu d\u00f6rt ilkenin \u00f6\u011frenme analiti\u011fini y\u00f6nlendirmesi gerekti\u011fini \u00f6ne s\u00fcrmektedirler:<\/p>\n\n<ul>\n \t<li>\n<p align=\"justify\">Ki\u015fisel bilgiler, yaln\u0131zca temin edilen ba\u011flamda ve ama\u00e7larda kullan\u0131labilir<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Bu t\u00fcr verilerin daha sonra kullan\u0131lmas\u0131, orijinal ba\u011flam ve ama\u00e7 ile uyumlu olmal\u0131d\u0131r<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Veriler dikkatlice toplanmal\u0131 ve analiz edilmelidir ve \u201csinsi\u201d (Hollandaca \u201cstiekeme\u201d) analitik kullan\u0131m\u0131na izin verilmez; Bu saydaml\u0131k, \u00f6\u011frenci dan\u0131\u015fmanl\u0131\u011f\u0131 ve sat\u0131n alma konusundaki bir ihtiyac\u0131 vurgulamaktad\u0131r<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Veriler sadece toplanan verilerin amac\u0131 \/ kullan\u0131m\u0131 a\u00e7\u0131k\u00e7a belirtildi\u011fi zaman toplanabilir<\/p>\n<\/li>\n<\/ul>\n<p align=\"justify\">Engelfriet vd. (2015), verilerinin y\u00f6netimi konusundaki \u00f6\u011frenci haklar\u0131n\u0131 a\u015fa\u011f\u0131dakiler de d\u00e2hil olmak \u00fczere ara\u015ft\u0131rm\u0131\u015flard\u0131r:<\/p>\n\n<ul>\n \t<li>\n<p align=\"justify\">Toplanan bilgilere kolay eri\u015fim<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">Yanl\u0131\u015f bilgileri (veya bundan kaynaklanan yorumlar\u0131) d\u00fczeltme hakk\u0131<\/p>\n<\/li>\n \t<li>\n<p align=\"justify\">\u0130lgisiz bilgileri kald\u0131rma hakk\u0131<\/p>\n<\/li>\n<\/ul>\n<p align=\"justify\">\u00d6zellikle ilgi \u00e7ekici olan, algoritmik karar vermek i\u00e7in etik sonu\u00e7lar\u0131n ara\u015ft\u0131r\u0131lmas\u0131d\u0131r ve yazarlar, Hollanda yasalar\u0131yla olas\u0131 \u00e7at\u0131\u015fmaya yol a\u00e7an \u00f6rneklere i\u015faret etmektedirler. Bunun anlam\u0131, insanlar\u0131n algoritmik karar verme sorumlulu\u011funu \u00fcstlenmesi ve g\u00f6zetim alt\u0131nda tutmas\u0131 gerekti\u011fidir. Algoritmalar, en fazla, \u00fcniversite veya destek personelinin dikkatini \u00e7ekmek i\u00e7in \u00f6zel davran\u0131\u015flara i\u015faret edebilir. Ayr\u0131ca, \u00f6\u011frencilerin ki\u015fisel verilerinin analizlerine dayanarak verilen kararlara itiraz etme haklar\u0131 vard\u0131r. Y\u00d6K'lerin yaz\u0131l\u0131m geli\u015ftiricilere ta\u015feronluk yapt\u0131\u011f\u0131 durumlarda, nihai sorumluluk ve g\u00f6zetim kurumda g\u00fcvende kal\u0131r ve delege edilemez (bk. Engelfriet vd., 2015).<\/p>\n\n<h2 class=\"western\">GELECEKTEK\u0130 BAZI HUSUSLAR<\/h2>\n<p align=\"justify\">\u00d6\u011frenci verileri ile teknolojideki ilerlemeler ve analiz metotlar\u0131 aras\u0131ndaki kesi\u015fmelerdeki karma\u015f\u0131kl\u0131k ve pratiklikleri anlamam\u0131zdaki mevcut ve gelecekteki bo\u015fluklar\u0131 haritaland\u0131rmak bu b\u00f6l\u00fcm\u00fcn kapsam\u0131 d\u0131\u015f\u0131nda kalmaktad\u0131r. Bununla birlikte, gelecekteki de\u011ferlendirmeler i\u00e7in baz\u0131 \u00f6nerilerde bulunmak istiyoruz.<\/p>\n<p align=\"justify\">Etkili, uygun, uygun maliyetli \u00f6\u011frenme deneyimleri sa\u011flamak ve \u00f6\u011frencilerin ba\u015far\u0131l\u0131 olmalar\u0131n\u0131 desteklemek i\u00e7in y\u00fcksek\u00f6\u011fretim kurumlar\u0131n\u0131n g\u00f6rev s\u00fcreleri g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, kurumlar\u0131n \u00f6\u011frenci bilgilerini toplama ve kullanma hakk\u0131na sahip oldu\u011fu konusunda geni\u015f bir anla\u015fma vard\u0131r. Bununla birlikte, \u00f6\u011frencilerin izinlerinin toplanmas\u0131ndan, analiz edilmesinden ve verilerinin kullan\u0131lmamas\u0131n\u0131 tercih etmelerine izin vermede onay konusunda mutab\u0131k bir pozisyon bulunmamaktad\u0131r. Onay ile ilgili \u00f6\u011frencilerin durumlar\u0131 tamamen mant\u0131kl\u0131 ya da ak\u0131lc\u0131 olmayan konulardan etkilenebilir. Faydalar\u0131n, maliyetlerin ve risklerin s\u0131kl\u0131kla \u00f6rt\u00fcl\u00fc olarak hesaplanmas\u0131, di\u011ferlerinin yan\u0131 s\u0131ra, \u00f6nceki deneyimler, ihtiya\u00e7 ve alg\u0131lanan faydalar gibi bir dizi fakt\u00f6re ba\u011fl\u0131 olacakt\u0131r (bak\u0131n\u0131z, \u00f6rne\u011fin, O'Brien'daki Daniel Pink, 2010).<\/p>\n<p align=\"justify\">Son zamanlarda \u00e7ekilmenin bir \u00f6rne\u011fi, \u00f6\u011frencileri devlet taraf\u0131ndan zorunlu hale getirilmi\u015f standart testler almay\u0131 reddetmeye te\u015fvik eden ABD'deki Ulusal Adil ve A\u00e7\u0131k Test Merkezi taraf\u0131ndan g\u00f6sterilmi\u015ftir. 2014-2015 \u00f6\u011fretim y\u0131l\u0131nda (FairTest, nd) yakla\u015f\u0131k 650.000 \u00f6\u011frenci, ABD E\u011fitim Bakanl\u0131\u011f\u0131 taraf\u0131ndan finanse edilmeme tehdidine cevaben \u00e7ekilmi\u015ftir (Strauss, 2016a).<\/p>\n<p align=\"justify\">\u00d6\u011frencilerin kayg\u0131lar\u0131, ayr\u0131lma haklar\u0131 ve y\u00fcksek\u00f6\u011fretimin \u00f6\u011frenci verilerini bireysel bir d\u00fczeyde m\u00fcdahalelerde bulunmak i\u00e7in kullanma yetkisinin olmas\u0131n\u0131n sonu\u00e7lar\u0131 aras\u0131ndaki olas\u0131 \u00e7at\u0131\u015fmalar\u0131 ara\u015ft\u0131rmak i\u00e7in daha fazla ara\u015ft\u0131rmaya ihtiya\u00e7 vard\u0131r. Bu konunun merkezinde \u201ckim yararlan\u0131r?\u201d sorusu bulunmaktad\u0131r. (bk. Watters, 2016). \u00d6\u011frenci verilerinin toplanmas\u0131, analizi ve kullan\u0131m\u0131yla ilgili etik kurallar\u0131n (gerek \u00f6\u011frenme analiti\u011finde gerek resm\u00ee de\u011ferlendirmelerde olsun), tart\u0131\u015fmal\u0131 iddialar\u0131 ve kazan\u0131lm\u0131\u015f ilgi alanlar\u0131n\u0131 da tan\u0131mas\u0131 gerekir.<\/p>\n<p align=\"justify\">Daha geni\u015f bir \u00e7evrimi\u00e7i ara\u015ft\u0131rma ba\u011flam\u0131nda, Vitak, Shilton ve Ashktorab (2016), \u00e7evrimi\u00e7i ba\u011flamlardaki etik ara\u015ft\u0131rma uygulamalar\u0131 ile ilgili olarak, yeniden kimlik tan\u0131mlamas\u0131 yap\u0131lmas\u0131 konusunda gittik\u00e7e artarak devam eden endi\u015feler gibi \u00e7e\u015fitli zorluklara ki; \u201cara\u015ft\u0131rmac\u0131lar hala \u00e7evrimi\u00e7i veri k\u00fcmeleri kullan\u0131m\u0131yla ara\u015ft\u0131rma etikleri aras\u0131nda dengeyi bulmakta zorlanmaktad\u0131r\u201d i\u015faret etmektedir (s. 1). \u0130lgin\u00e7tir ki, bulgular\u0131 da bir\u00e7ok ara\u015ft\u0131rmac\u0131n\u0131n Belmont ilkelerinin \u00f6tesine ge\u00e7ti\u011fini (\u00e7\u0131kt\u0131lar\u0131n, ara\u015ft\u0131rmalar\u0131n neden oldu\u011fu olas\u0131 zararlardan daha a\u011f\u0131r basmas\u0131n\u0131 sa\u011flaman\u0131n \u00f6nemini vurgulayarak) g\u00f6stermektedir. Bu ilkeler: \u201c(1) kat\u0131l\u0131mc\u0131larla \u015feffafl\u0131k, (2) i\u015f arkada\u015flar\u0131 ile etik konular hakk\u0131nda m\u00fczakere etmek (3) sonu\u00e7lar\u0131n payla\u015f\u0131m\u0131nda dikkatli olmakt\u0131r (b\u00f6l.66).<\/p>\n<p align=\"justify\">Yapay zek\u00e2 (YZ), makine \u00f6\u011frenmesi ve b\u00fcy\u00fck verilerle ilgili iyimserli\u011fi dengeleme konusundaki endi\u015fe gittik\u00e7e artmaktad\u0131r. \u00d6rne\u011fin, ABD Ba\u015fkanl\u0131\u011f\u0131 Y\u00f6netim Ofisi, faydalar\u0131 vurgulayan (Munoz, Smith ve Patil, 2016) ancak ayn\u0131 zamanda b\u00fcy\u00fck verilerin kullan\u0131m\u0131nda do\u011fabilecek potansiyel zararlarla ilgili endi\u015feleri de ele alan bir rapor yay\u0131nlad\u0131. Rapor, e\u011fer \u201cbu teknolojiler [algoritmik sistemler] dikkatli bir \u015fekilde uygulanmazlarsa, zararl\u0131 ayr\u0131mc\u0131l\u0131\u011f\u0131 s\u00fcrd\u00fcrebileceklerini, k\u00f6t\u00fcle\u015ftirebileceklerini veya maskeleyebileceklerini\u201d kabul eder (s. 5). Algoritmik ayr\u0131mc\u0131l\u0131\u011f\u0131n azalt\u0131lmas\u0131, sa\u011flam ve \u015feffaf algoritmalar\u0131n geli\u015ftirilmesi ve kullan\u0131lmas\u0131n\u0131n te\u015fvik edilmesi, algoritmik denetim, veri bilimindeki \u201cak\u0131c\u0131l\u0131\u011f\u0131n\u201d iyile\u015ftirilmesi ve veri kullan\u0131m\u0131nda uygulama kodlar\u0131n\u0131 belirlemek hususunda devlet ve \u00f6zel sekt\u00f6r\u00fcn rolleri ve ara\u015ft\u0131rmalara yat\u0131r\u0131m konusunda baz\u0131 \u00f6nerilerde bulunmaktad\u0131r.<\/p>\n<p align=\"justify\">Benzer \u015fekilde, Birle\u015fik Krall\u0131k H\u00fckumeti k\u0131sa bir s\u00fcre \u00f6nce \u201cyasalar\u0131n d\u0131\u015f\u0131nda kalan etik konularla ilgili\u201d rehberlik sa\u011flayan bir \u201cVeri bilimi etik \u00e7er\u00e7evesi\u201d (Kabine Ofisi, 2016) yay\u0131nlam\u0131\u015ft\u0131r (s. 3). Bu \u00e7er\u00e7eve, ki\u015fisel verilerin toplanmas\u0131n\u0131n, analiz edilmesinin ve kullan\u0131lmas\u0131n\u0131n faydalar\u0131n\u0131n niteli\u011fi; izinsiz giri\u015fin kapsam\u0131 ve niteli\u011fi; verilerin kalitesi ve toplanan verilerle ilgili kararlar\u0131n otomasyonu; istenmeyen sonu\u00e7lar\u0131n riski; veri nesnelerinin toplama ve analiz i\u00e7in mutab\u0131k kal\u0131p kalmad\u0131\u011f\u0131; g\u00f6zetimin niteli\u011fi ve kapsam\u0131 ve toplanan verilerin g\u00fcvenli\u011fi gibi konular\u0131 i\u00e7erir. \u00c7er\u00e7eve ayr\u0131ca veri bilimcilere projenin yararlar\u0131n\u0131n gizlilik ve olumsuz sonu\u00e7lara kar\u015f\u0131n risklerin ne kadar fazla oldu\u011fu, riskleri en aza indirmek ve do\u011fru yorumlamay\u0131 sa\u011flamak i\u00e7in at\u0131lan ad\u0131mlar ve veri nesnelerinin \/ kamuoyunun projeye ili\u015fkin g\u00f6r\u00fc\u015flerinin ne \u00f6l\u00e7\u00fcde dikkate al\u0131nd\u0131\u011f\u0131 gibi \u201czorlu meselelere\u201d (s.6) a\u00e7\u0131kl\u0131k getirmelerini gerektiren bir \u201cGizlilik Etki De\u011ferlendirmesi\u201d \u00f6nermektedir (bk. Kabine Ofisi, 2016).<\/p>\n<p align=\"justify\">Y\u00fcksek\u00f6\u011frenimdeki algoritmik d\u00f6n\u00fc\u015f\u00fcm ile veri ve sinir bilim aras\u0131ndaki s\u0131n\u0131rlar\u0131n bulan\u0131kla\u015fmas\u0131 ba\u011flam\u0131nda, efsane, karma\u015f\u0131kl\u0131k ve y\u00f6ntemler arac\u0131l\u0131\u011f\u0131yla yol al\u0131rken \u00f6\u011frenme analitiklerinin etik \u00e7\u0131kar\u0131mlar\u0131n\u0131 g\u00f6z \u00f6n\u00fcne almak i\u00e7in kritik bir yakla\u015f\u0131ma ihtiyac\u0131m\u0131z vard\u0131r (Ziewitz, 2016). \u00d6rne\u011fin, Williamson (2016a) \u201ce\u011fitsel veri bilimini, \u00e7ocuklar\u0131n bedensel, duygusal ve g\u00f6m\u00fcl\u00fc ya\u015famlar\u0131n\u0131n de\u011ferlendirilmesi ve y\u00f6netimine odaklanan \"<span style=\"font-family: Source Serif Pro Light, serif;\"><i>biyopolitik bir strateji<\/i><\/span>\" olarak g\u00f6rmektedir (s. 401, vurgu). Bu nedenle, \u201c\u00e7ocuklar hakk\u0131nda bilgi sistemleri \u00fcretmek ve onlar\u0131 m\u00fcdahalenin konusu ve nesnesi olarak tan\u0131mlamak i\u00e7in me\u015fru yetkiye sahip\u201d e\u011fitsel veri bilim adamlar\u0131n\u0131n temelini ve kapsam\u0131n\u0131 g\u00f6z \u00f6n\u00fcnde bulundurmal\u0131y\u0131z (Williamson, 2016a, s. 401). Gelecekte \u00f6\u011frenme analiti\u011fi, temelde algoritmalar ve makine \u00f6\u011frenmesine dayal\u0131 olacak ve y\u00f6nlendirilecektir; bu nedenle algoritmalar\u0131n \u201csosyal d\u00fcnyay\u0131, bilgiyi ve bilgi ile kar\u015f\u0131 kar\u015f\u0131ya gelenlerin g\u00f6r\u00fc\u015flerini nas\u0131l g\u00fc\u00e7lendirdi\u011fini, s\u00fcrd\u00fcrd\u00fc\u011f\u00fcn\u00fc ve hatta yeniden \u015fekillendirdi\u011fini\u201d g\u00f6z \u00f6n\u00fcnde bulundurmal\u0131y\u0131z (Williamson, 2016b, s. 4). G\u00fcvenilirlik, \u015feffafl\u0131k ve d\u00fczenleyici \u00e7er\u00e7eveler, <span style=\"font-family: Source Serif Pro Light, serif;\"><i>etik<\/i><\/span> \u00f6\u011frenme analiti\u011fini sa\u011flayan \u00e7er\u00e7evelerde temel unsurlar olacakt\u0131r (bk. Prinsloo, 2016; Taneja, 2016).<\/p>\n<p align=\"justify\">Bu b\u00f6l\u00fcm, \u00f6\u011frenci verilerinin toplanmas\u0131, analizi ve kullan\u0131m\u0131yla ilgili etik sonu\u00e7lar\u0131 g\u00f6z \u00f6n\u00fcnde bulundurmadaki ilerlemenin haritas\u0131n\u0131 \u00e7\u0131kar\u0131rken, g\u00fcvenilirli\u011fi ve \u015feffafl\u0131\u011f\u0131 sa\u011flamak i\u00e7in kurumsal s\u00fcre\u00e7leri daha fazla dikkate almadan zarar verme potansiyelinin ele al\u0131nmayaca\u011f\u0131 a\u00e7\u0131kt\u0131r. Willis, Slade ve Prinsloo'nun (2016) belirtti\u011fi gibi, \u00f6\u011frenme analiti\u011fi genellikle kurumsal inceleme kurullar\u0131 (K\u0130K) taraf\u0131ndan sa\u011flanan s\u00fcre\u00e7lerin ve g\u00f6zetimin d\u0131\u015f\u0131nda kalmaktad\u0131r. Bu a\u015famada \u00f6\u011frenme analiti\u011finin etik etkilerinin kim taraf\u0131ndan ve nas\u0131l sa\u011flanaca\u011f\u0131 net de\u011fildir.<\/p>\n\n<h2 class=\"western\">SONU\u00c7LAR<\/h2>\n<p align=\"justify\">2011 y\u0131l\u0131nda \u00f6\u011frenme analiti\u011finin ortaya \u00e7\u0131kmas\u0131ndan bu yana, alan ilerlemekle kalmay\u0131p ayn\u0131 zamanda \u00f6\u011frenci verilerinin toplanmas\u0131nda, analiz edilmesinde ve kullan\u0131lmas\u0131ndaki etik sonu\u00e7lar\u0131n korkular\u0131 ve ger\u00e7ekleri g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda giderek daha fazla ilgi g\u00f6rmeye ba\u015flam\u0131\u015ft\u0131r. Bu b\u00f6l\u00fcmde, alandaki daha geni\u015f geli\u015fmelerin yan\u0131 s\u0131ra kendi d\u00fc\u015f\u00fcncemizin nas\u0131l geli\u015fti\u011fine de genel bir bak\u0131\u015f sunuyoruz. Teknolojik geli\u015fmeler ve yayg\u0131n g\u00f6zetim konusundaki endi\u015felerin artmas\u0131 ve y\u00fcksek e\u011fitimin gelece\u011finin dijital, da\u011f\u0131t\u0131lm\u0131\u015f ve veri odakl\u0131 olaca\u011f\u0131 konusunda artan bir fikir birli\u011fine kar\u015f\u0131, bu b\u00f6l\u00fcm \u00f6\u011frenme analiti\u011finin etik anlam\u0131n\u0131 \u00e7evreleyen s\u00f6ylemlerin ne kadar ileri geldi\u011fini ve gelecekte dikkate al\u0131nmas\u0131 gereken noktalar\u0131 g\u00f6stermektedir.<\/p>\n<p align=\"justify\">Tart\u0131\u015f\u0131lan \u00f6\u011frenme analitiklerindeki etik sonu\u00e7lar\u0131n \u00e7er\u00e7evelerinin, uygulama kodlar\u0131n\u0131n ve kavramsal haritalamalar\u0131n her biri ekonomik ve etik yollarla \u00f6\u011fretmenin ve \u00f6\u011frenmenin etkinli\u011fini ve uygunlu\u011funu artt\u0131rmak i\u00e7in \u00f6\u011frenci veri proxylerini kullanma yolunda nas\u0131l ilerleyebilece\u011fimizi daha zengin bir anlay\u0131\u015fla sunar. Bu anlay\u0131\u015f\u0131n pratik uygulamas\u0131 b\u00fcy\u00fck \u00f6l\u00e7\u00fcde tamamlanmam\u0131\u015f olsa da tamam\u0131yla yerindedir.<\/p>\n\n<h2 class=\"western\">TE\u015eEKK\u00dcR B\u00d6L\u00dcM\u00dc<\/h2>\n<p align=\"justify\">Edit\u00f6r ekibinden ve \u00f6zellikle bu b\u00f6l\u00fcm\u00fcn g\u00f6zden ge\u00e7irenlerden gelen yorumlara, \u00f6nemli girdilere ve verdikleri deste\u011fe te\u015fekk\u00fcr\u00fcm\u00fcz\u00fc arz ederiz.<\/p>\n\n<h2 class=\"western\">KAYNAK\u00c7A<\/h2>\n<span style=\"font-size: small;\">Cabinet Office. (2016, 19 May). Data science ethical framework. https:\/\/www.gov.uk\/government\/uploads\/ system\/uploads\/attachment_data\/file\/524298\/Data_science_ethics_framework_v1.0_for_publication__1_.pdf <\/span>\n\n<span style=\"font-size: small;\">Campbell, J. P., DeBlois, P. B., &amp; Oblinger, D. G. (2007, July\/August) Academic analytics: A new tool, a new era. <i>EDUCAUSE Review<\/i>. http:\/\/net.educause.edu\/ir\/library\/pdf\/erm0742.pdf <\/span>\n\n<span style=\"font-size: small;\">Couldry, N. (2016, September 23). The price of connection: \u201cSurveillance capitalism.\u201d [Web log post]. https:\/\/ theconversation.com\/the-price-of-connection-surveillance-capitalism-64124 <\/span>\n\n<span style=\"font-size: small;\">Dawson, S., Ga\u0161evi\u0107, D., &amp; Rogers, T. (2016). Student retention and learning analytics: A snapshot of Australian practices and a framework for advancement. Australian Government. http:\/\/he-analytics.com\/wp-content\/uploads\/SP13_3249_Dawson_Report_2016-3.pdf <\/span>\n\n<span style=\"font-size: small;\">Drachsler, H., &amp; Greller, W. (2012). The pulse of learning analytics understandings and expectations from the stakeholders. <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 120\u2013129). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Drachsler, H., &amp; Greller, W. (2016, April). Privacy and analytics: It\u2019s a DELICATE issue \u2014 a checklist for trusted learning analytics. <i>Proceedings of the 6th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201916), 25\u201329 April 2016, Edinburgh, UK (pp. 89\u201398). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Engelfriet, A., Manderveld, J., &amp; Jeunink, E. (2015). Learning analytics onder de Wet bescherming persoonsgegevens. SURFnet. https:\/\/www.surf.nl\/binaries\/content\/assets\/surf\/nl\/kennisbank\/2015\/surf_learning-analytics-onder-de-wet-wpb.pdf <\/span>\n\n<span style=\"font-size: small;\">FairTest. (n.d.). Just say no to the test. http:\/\/www.fairtest.org\/get-involved\/opting-out <\/span>\n\n<span style=\"font-size: small;\">Ga\u0161evi\u0107, D., Dawson, S., &amp; Jovanovi\u0107, J. (2016). Ethics and privacy as enablers of learning analytics. <i>Journal of Learning Analytics, 3<\/i>(1), 1\u20134. http:\/\/dx.doi.org\/10.18608\/jla.2016.31.1 <\/span>\n\n<span style=\"font-size: small;\">Munoz, C., Smith, M., &amp; Patil, D. J. (2016). Big data: A report on algorithmic systems, opportunity, and civil rights. Executive Office of the President, USA. https:\/\/www.whitehouse.gov\/sites\/default\/files\/microsites\/ostp\/2016_0504_data_discrimination.pdf <\/span>\n\n<span style=\"font-size: small;\">NMC (New Media Consortium). (2011). The NMC Horizon Report. http:\/\/www.educause.edu\/Resources\/2011HorizonReport\/223122 <\/span>\n\n<span style=\"font-size: small;\">NMC (New Media Consortium). (2016). The NMC Horizon Report. http:\/\/www.nmc.org\/publication\/ nmc-horizon-report-2016-higher-education-edition\/ <\/span>\n\n<span style=\"font-size: small;\">O\u2019Brien, A. (2010, September 29). Predictably irrational: A conversation with best-selling author Dan Ariely. [Web log post]. http:\/\/www.learningfirst.org\/predictably-irrational-conversation-best-selling-author-dan-ariely <\/span>\n\n<span style=\"font-size: small;\">Open University. (2014). Policy on ethical use of student data for learning analytics. http:\/\/www.open.ac.uk\/students\/charter\/sites\/www.open.ac.uk.students.charter\/files\/files\/ecms\/web-content\/ethical-use-of-student-data-policy.pdf <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P. (2016, September 22). Fleeing from Frankenstein and meeting Kafka on the way: Algorithmic decision-making in higher education. Presentation at NUI, Galway. http:\/\/www.slideshare.net\/prinsp\/feeling-from-frankenstein-and-meeting-kafka-on-the-way-algorithmic-decisionmaking-in-higher-education <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., Slade, S., &amp; Galpin, F. (2012) Learning analytics: Challenges, paradoxes and opportunities for mega open distance learning institutions. <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 130\u2013133). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2013). An evaluation of policy frameworks for addressing ethical considerations in learning analytics. <i>Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 240\u2013244). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2014a). Educational triage in higher online education: Walking a moral tightrope. <i>International Review of Research in Open Distributed Learning <\/i>(IRRODL)<i>, 14<\/i>(4), 306\u2013331. http:\/\/www.irrodl.org\/ index.php \/ irrodl \/ article \/ view \/ 1881 <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2014b). Student privacy and institutional accountability in an age of surveillance. In M. E. Menon, D. G. Terkla, &amp; P. Gibbs (Eds.), <i>Using data to improve higher education: Research, policy and practice <\/i>(pp. 197\u2013214). Global Perspectives on Higher Education (29). Rotterdam: Sense Publishers. <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2015). Student privacy self-management: Implications for learning analytics. <i>Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 83\u201392). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2016a). Student vulnerability, agency, and learning analytics: An exploration. <i>Journal of Learning Analytics, 3<\/i>(1), 159\u2013182. <\/span>\n\n<span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2016b). Here be dragons: Mapping student responsibility in learning analytics. In M. Anderson &amp; C. Gavan (Eds.), <i>Developing effective educational experiences through learning analytics <\/i>(pp. 170\u2013188). Hershey, PA: IGI Global. <\/span>\n\n<span style=\"font-size: small;\">Ruggiero, D. (2016, May 18). What metrics don\u2019t tell us about the way students learn. <i>The Conversation<\/i>. http:\/\/ theconversation.com\/what-metrics-dont-tell-us-about-the-way-students-learn-59271 <\/span>\n\n<span style=\"font-size: small;\">Sclater, N. (2015, March 3). Effective learning analytics. A taxonomy of ethical, legal and logistical issues in learning analytics v1.0. JISC. https:\/\/analytics.jiscinvolve.org\/wp\/2015\/03\/03\/a-taxonomy-of-ethical-legal-and-logistical-issues-of-learning-analytics-v1-0\/ <\/span>\n\n<span style=\"font-size: small;\">Sclater, N., Peasgood, A., &amp; Mullan, J. (2016). Learning analytics in higher education. A review of UK and international practice. JISC. https:\/\/www.jisc.ac.uk\/reports\/learning-analytics-in-higher-education <\/span>\n\n<span style=\"font-size: small;\">Shacklock, X. (2016). From bricks to clicks: The potential of data and analytics in higher education. Higher Education Commission. http:\/\/www.policyconnect.org.uk\/hec\/sites\/site_hec\/files\/report\/419\/fieldreportdownload\/frombrickstoclicks-hecreportforweb.pdf <\/span>\n\n<span style=\"font-size: small;\">Siemens, G. (2016, April 28). Reflecting on learning analytics and SoLAR. [Web log post]. http:\/\/www.elearnspace.org\/blog\/2016\/04\/28\/reflecting-on-learning-analytics-and-solar\/<\/span>\n\n<span style=\"font-size: small;\">Siemens, G., &amp; Baker, R. (2012, April). Learning analytics and educational data mining: Towards communication and collaboration. <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 252\u2013254). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Slade, S., &amp; Galpin, F. (2012) Learning analytics and higher education: Ethical perspectives (workshop). <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 16\u201317). New York: ACM. <\/span>\n\n<span style=\"font-size: small;\">Slade, S., &amp; Prinsloo, P. (2013). Learning analytics: Ethical issues and dilemmas. <i>American Behavioral Scientist, 57<\/i>(1), 1509\u20131528. <\/span>\n\n<span style=\"font-size: small;\">Slade, S., &amp; Prinsloo, P. (2014). Student perspectives on the use of their data: Between intrusion, surveillance and care. <i>Proceedings of the European Distance and E-Learning Network 2014 Research Workshop <\/i>(EDEN 2014), 27\u201328 October 2014, Oxford, UK (pp. 291\u2013300). <\/span>\n\n<span style=\"font-size: small;\">Smith, G.J. (2016). Surveillance, data and embodiment: On the work of being watched. <i>Body &amp; Society<\/i>, 1\u201332. doi:10.1177\/1357034X15623622 <\/span>\n\n<span style=\"font-size: small;\">Strauss, V. (2016a, January 28). U.S. Education Department threatens to sanction states over test opt-outs. The Washington Post. https:\/\/www.washingtonpost.com\/news\/answer-sheet\/wp\/2016\/01\/28\/u-s-education-department-threatens-to-sanction-states-over-test-opt-outs\/ <\/span>\n\n<span style=\"font-size: small;\">Strauss, V. (2016b, May 9). \u201cBig data\u201d was supposed to fix education. It didn\u2019t. It\u2019s time for \u201csmall data.\u201d The Washington Post. https:\/\/www.washingtonpost.com\/news\/answer-sheet\/wp\/2016\/05\/09\/big-data-was-supposed-to-fix-education-it-didnt-its-time-for-small-data\/ <\/span>\n\n<span style=\"font-size: small;\">Taneja, H. (2016, September 8). The need for algorithmic accountability. <i>TechCrunch<\/i>. https:\/\/techcrunch.com\/2016\/09\/08\/the-need-for-algorithmic-accountability\/ <\/span>\n\n<span style=\"font-size: small;\">van Barneveld, A., Arnold, K., &amp; Campbell, J. (2012). Analytics in higher education: Establishing a common language. <i>EDUCAUSE Learning Initiative, 1<\/i>, 1\u201311. <\/span>\n\n<span style=\"font-size: small;\">Vitak, J., Shilton, K., &amp; Ashktorab, Z. (2016). Beyond the Belmont principles: Ethical challenges, practices, and beliefs in the online data research community. <i>Proceedings of the 19th ACM Conference on Computer Supported Cooperative Work &amp; Social Computing <\/i>(CSCW\u201916), 27 February\u20132 March 2016, San Francisco, CA, USA. New York: ACM. https:\/\/terpconnect.umd.edu\/~kshilton\/pdf\/VitaketalCSCWpreprint.pdf <\/span>\n\n<span style=\"font-size: small;\">Watters, A. (2016, May 7). Identity, power, and education\u2019s algorithms. [Web log post]. http:\/\/hackeducation.com\/2016\/05\/07\/identity-power-algorithms <\/span>\n\n<span style=\"font-size: small;\">Welsh, S., &amp; McKinney, S. (2015). Clearing the fog: A learning analytics code of practice. In T. Reiners et al. (Eds.), Globally connected, digitally enabled. <i>Proceedings of the 32nd Annual Conference of the Australasian Society for Computers in Learning in Tertiary Education <\/i>(ASCILITE 2015), 29 November\u20132 December 2015, Perth, Western Australia (pp. 588\u2013592). http:\/\/research.moodle.net\/80\/ <\/span>\n\n<span style=\"font-size: small;\">Williamson, B. (2016a). Coding the biodigital child: The biopolitics and pedagogic strategies of educational data science. <i>Pedagogy, Culture &amp; Society, 24<\/i>(3), 401\u2013416. doi:10.1080\/14681366.2016.1175499 <\/span>\n\n<span style=\"font-size: small;\">Williamson, B. (2016b). Computing brains: Learning algorithms and neurocomputation in the smart city. <i>Information, Communication &amp; Society, 20<\/i>(1), 81\u201399. doi:10.1080\/1369118X.2016.1181194 <\/span>\n\n<span style=\"font-size: small;\">Willis, J., Slade, S., &amp; Prinsloo, P. (2016). Ethical oversight of student data in learning analytics: A typology derived from a cross-continental, cross-institutional perspective. Educational Technology Research and Development. doi:10.1007\/s11423-016-9463-4 <\/span>\n\n<span style=\"font-size: small;\">Zhang, S. (2016, May 20). Scientists are just as confused about the ethics of big data research as you. Wired. http:\/\/www.wired.com\/2016\/05\/scientists-just-confused-ethics-big-data-research\/ <\/span>\n\n<a name=\"_Hlk25065891\"><\/a> <span style=\"font-size: small;\">Ziewitz, M. (2016). Governing algorithms: Myth, mess, and methods. <i>Science, Technology &amp; Human Values, 41<\/i>(1), 3\u201316.<\/span>\n","rendered":"<p style=\"text-align: justify;\"><a name=\"_Toc27652707\" id=\"_Toc27652707\"><\/a> <span style=\"font-family: Source Sans Pro Light, serif;\"><span style=\"font-size: medium;\">Paul Prinsloo<sup>1<\/sup>, Sharon Slade<sup>2<\/sup><\/span><\/span><\/p>\n<p><span style=\"font-family: Source Sans Pro Light, serif;\"><span style=\"font-size: small;\"><sup>1 <\/sup>\u0130\u015fletme Y\u00f6netimi B\u00f6l\u00fcm\u00fc, G\u00fcney Afrika \u00dcniversitesi, G\u00fcney Afrika<\/span><\/span><\/p>\n<p><span style=\"font-family: Source Sans Pro Light, serif;\"><span style=\"font-size: small;\"><sup>2 <\/sup>\u0130\u015fletme ve Hukuk Fak\u00fcltesi, A\u00e7\u0131k \u00dcniversite, \u0130ngiltere<\/span><\/span><\/p>\n<p><span style=\"font-family: Source Sans Pro, serif;\"><span style=\"font-size: small;\">DOI: 10.18608\/hla17.004<\/span><\/span><\/p>\n<h2 class=\"western\">\u00d6Z<\/h2>\n<p><span style=\"font-size: small;\">\u00d6\u011frenme analiti\u011fi alan\u0131 olgunla\u015ft\u0131k\u00e7a ve kapsam\u0131, tan\u0131m\u0131, zorluklar\u0131 ve \u00f6\u011frenme analiti\u011fi f\u0131rsatlar\u0131n\u0131 \u00e7evreleyen s\u00f6ylemler daha da artt\u0131k\u00e7a bunlara ili\u015fkin eklemeler etik meselelerle ilgili ne kadar yol ald\u0131\u011f\u0131m\u0131z\u0131 g\u00f6zden ge\u00e7irmemizde ve gelece\u011fi g\u00f6z \u00f6n\u00fcnde bulundurmam\u0131zda fayda vard\u0131r. Bu b\u00f6l\u00fcm, kendi d\u00fc\u015f\u00fcncemizin nas\u0131l geli\u015fti\u011fine ve alandaki daha geni\u015f geli\u015fmelere kar\u015f\u0131 yolculu\u011fumuzu haritaland\u0131rmaya genel bir bak\u0131\u015f sa\u011flar. Teknolojik geli\u015fmeler ve her tarafa n\u00fcfuz eden g\u00f6zetim konusundaki endi\u015felerin artmas\u0131 ve algoritmalar\u0131n \u00f6nemi ve istenmeyen sonu\u00e7lar\u0131 ile ilgili olarak,etik ve ahlaki bir uygulama olarak \u00f6\u011frenme analiti\u011fine y\u00f6nelik ara\u015ft\u0131rmalar\u0131n geli\u015ftirilmesi, korkular\u0131n ve ger\u00e7eklerin zengin bir resmini sunar. Daha da \u00f6nemlisi, eti\u011fi ve mahremiyeti, \u00f6\u011frenme analiti\u011fi i\u00e7in \u00e7ok \u00f6nemli bir kolayla\u015ft\u0131r\u0131c\u0131 unsur olarak g\u00f6rmeye ba\u015fl\u0131yoruz. Bu b\u00f6l\u00fcm, bireysel ara\u015ft\u0131rma yolculu\u011fumuzu izlemeden \u00f6nce, geni\u015f ba\u011flamda y\u00fcksek\u00f6\u011frenimi \u015fekillendiren g\u00fc\u00e7ler ile veri ve kan\u0131t rollerine de\u011finerek alandaki mevcut \u00e7al\u0131\u015fmalar\u0131 vurgulayarak ve gelecekteki sorunlar\u0131 haritaland\u0131rarak, \u00f6\u011frenme analiti\u011fi eti\u011fini ele almaktad\u0131r.<\/span><\/p>\n<p><span style=\"font-size: small;\"><span style=\"font-family: Source Sans Pro Black, serif;\">Anahtar Kelimeler:<\/span>Etik, yapay zek\u00e2, veri, b\u00fcy\u00fck veri, \u00f6\u011frenciler<\/span><\/p>\n<p style=\"text-align: justify;\">2011&#8217;de, Yeni Medya Konsorsiyumu&#8217;nun Ufuk Raporu (YMK, 2011), orta s\u0131n\u0131f bir teknolojiden ya da trendden bir y\u0131l yada daha az bir zaman diliminde geli\u015fimi fark edilecek bir seviyeye <span style=\"font-family: Source Serif Pro Light, serif;\"><i>ula\u015fan<\/i><\/span> \u00f6\u011frenme analiti\u011finin artan \u00f6nemine i\u015faret etmi\u015ftir. (YMK 2016, s. 38). \u00d6\u011frenme analiti\u011fi ile daha geli\u015fmi\u015f e\u011fitsel veri madencili\u011fi alan\u0131 aras\u0131nda a\u00e7\u0131k ba\u011flant\u0131lar olmas\u0131na ra\u011fmen otomasyon, ama\u00e7lar, k\u00f6kenler, teknikler ve y\u00f6ntemler ile di\u011ferleri aras\u0131nda da \u00f6nemli ayr\u0131mlar vard\u0131r. (Siemens ve Baker, 2012). \u00d6\u011frenme analiti\u011fi alan\u0131, farkl\u0131 bir ara\u015ft\u0131rma ve uygulama alan\u0131 olarak geli\u015ftik\u00e7e (bk. Van Barneveld, Arnold ve Campbell, 2012), etik meseleleri d\u00fc\u015f\u00fcnmek de yava\u015f yava\u015f \u00e7evreden merkeze do\u011fru ta\u015f\u0131nmaya ba\u015flam\u0131\u015ft\u0131r. Slade ve Prinsloo (2013), \u00f6\u011frenme analiti\u011finde etik odakl\u0131 bir \u015fekilde geli\u015ftirilen en eski \u00e7er\u00e7evelerden birini kurmu\u015ftur. O zamandan beri, bu alt alanda yay\u0131n yapan yazarlar\u0131n say\u0131s\u0131 \u00f6nemli \u00f6l\u00e7\u00fcde artm\u0131\u015ft\u0131r; bu da artan say\u0131da \u00e7er\u00e7eve, uygulama esaslar\u0131, taksonomilerle ve rehberle sonu\u00e7lanm\u0131\u015ft\u0131r (Ga\u0161evi\u0107, Dawson &amp; Jovanovi\u0107, 2016).<\/p>\n<p style=\"text-align: justify;\">Geni\u015f kamuoyu kapsam\u0131nda artan g\u00f6zetimi ve g\u00fcvence alt\u0131na al\u0131n(ma)m\u0131\u015f ki\u015fisel verilerin toplanmas\u0131, analiz edilmesi ve kullan\u0131lmas\u0131yla ilgili, \u201ckorku ve ger\u00e7ekler genellikle ay\u0131rt edilemez \u015fekilde kar\u0131\u015ft\u0131r\u0131larak potansiyel faydalan\u0131c\u0131lar aras\u0131nda bir belirsizlik ortam\u0131na yol a\u00e7maktad\u0131r\u201d (Drachsler ve Greller, 2016, s. 89). Ga\u0161evi\u0107 vd.(2016) ayr\u0131ca, \u201ce\u011fitim uygulamalar\u0131na daha fazla destek olmak ve entegrasyona yard\u0131mc\u0131 olmak i\u00e7in ele al\u0131nmas\u0131 gereken zorluklar\u0131n bulundu\u011funu ve etik ile mahremiyeti, \u00f6\u011frenme analiti\u011fi alan\u0131ndaki \u201cengelden ziyade\u201d \u00f6nemli bir destekleyici olarak g\u00f6rd\u00fcklerini ileri s\u00fcrmektedir (s. 2).<\/p>\n<p style=\"text-align: justify;\">Alandaki ki\u015fisel ara\u015ft\u0131rma yolculu\u011fumuzu haritalamadan \u00f6nce, \u00f6\u011frenme analitiklerinin etiksel \u00e7\u0131kar\u0131mlar\u0131n\u0131 g\u00f6z \u00f6n\u00fcnde bulundurma ba\u011flam\u0131n\u0131 k\u0131saca a\u00e7\u0131kl\u0131yoruz. Daha sonra son geli\u015fmeleri g\u00f6z \u00f6n\u00fcnde bulundurup daha geni\u015f ve d\u00e2hil edilmesi gereken daha \u00f6nemli bir dizi konuyu se\u00e7erek sonu\u00e7land\u0131r\u0131yoruz.<\/p>\n<h2 class=\"western\">BA\u011eLAMIN KURULMASI: ET\u0130\u011e\u0130N KONUYLA \u0130LG\u0130S\u0130<\/h2>\n<p style=\"text-align: justify;\">\u00d6\u011frenimin gelece\u011finin, e\u011fitimin \u201cya\u015fam kalitesini ve anlaml\u0131 istihdam\u0131 sa\u011flamas\u0131 anlam\u0131nda (a) ola\u011fan\u00fcst\u00fc kalite ara\u015ft\u0131rmalar\u0131n\u0131n; (b) karma\u015f\u0131k veri toplaman\u0131n ve (c) ileri makine \u00f6\u011frenmesi ve insan \u00f6\u011frenmesinin analizi \/ deste\u011fi&#8221; ile dijital, da\u011f\u0131t\u0131lm\u0131\u015f ve veri odakl\u0131 olaca\u011f\u0131 konusunda bir fikir birli\u011fi vard\u0131r (Siemens, 2016, Slayt 2). Her ne kadar \u00f6\u011frenme analitiklerinin etik etkilerine ili\u015fkin d\u00fc\u015f\u00fcnceler ba\u015flang\u0131\u00e7ta alanda geri planda kalsa da eti\u011fin \u00f6n plana \u00e7\u0131kmas\u0131 uzun bir yol kat etmi\u015ftir ve giderek daha da \u00f6n plana \u00e7\u0131kmaktad\u0131r (Ga\u0161evi\u0107 vd., 2016). Artan veri toplama i\u015fleminden kaynaklanan daha ba\u015far\u0131l\u0131 \u00f6\u011frenme deneyimlerinin (hem \u00f6\u011frenciler hem de kurum i\u00e7in) potansiyel <span style=\"font-family: Source Serif Pro Light, serif;\"><i>ekonomik<\/i><\/span> fayda ile ilgili \u00e7ok \u015fey s\u00f6ylenebilecek bir ba\u011flamda, \u201cveri proxy&#8217;sinden elde edilen ayr\u0131nt\u0131l\u0131 bir \u015fekilde somutla\u015ft\u0131r\u0131lm\u0131\u015f referans\u0131n\u0131 dezavantajl\u0131 hale getirdi\u011fi veri &#8211; vek\u00e2let kaynakl\u0131 zorluklar\u0131n bulunma ihtimalini g\u00f6z ard\u0131 etmemeliyiz&#8221;(Smith, 2016, s. 16; ayr\u0131ca bk. Ruggiero, 2016; Strauss, 2016b; ve Watters, 2016).<\/p>\n<p style=\"text-align: justify;\">\u00d6\u011frenci verilerinin toplanmas\u0131, analizi ve kullan\u0131lmas\u0131yla ilgili etik \u00e7\u0131kar\u0131mlar, potansiyel \u00e7\u0131kar \u00e7at\u0131\u015fmas\u0131 s\u00f6z konusu oldu\u011funda, \u00f6\u011frenciler ve kurumlar gibi bir dizi payda\u015f\u0131n talepleri dikkate al\u0131nmal\u0131d\u0131r. \u00d6\u011frenci verilerinin toplanmas\u0131ndan, analiz edilmesinden ve kullan\u0131lmas\u0131ndan kaynaklanan fayda, risk ve zarar potansiyeli hakk\u0131ndaki g\u00f6r\u00fc\u015fler, belirli bir payda\u015f\u0131n \u00e7\u0131kar\u0131na ve alg\u0131s\u0131na ba\u011fl\u0131 olacakt\u0131r. Bu b\u00f6l\u00fcmde, \u00f6ncelikle \u00f6\u011frencilerin ve kurumlar\u0131n farkl\u0131 konumsall\u0131klar\u0131, talepleri ve \u00e7\u0131karlar\u0131 hakk\u0131nda \u00f6ng\u00f6r\u00fc sunmay\u0131 umuyoruz.<\/p>\n<h2 class=\"western\">ET\u0130K \u0130LKELER\u0130N KURULMASI: NE KADAR YOL KATETT\u0130K?<\/h2>\n<p style=\"text-align: justify;\">G\u00fcn\u00fcm\u00fczde etik yakla\u015f\u0131m ve \u00f6\u011frenci verilerinin <span style=\"font-family: Source Serif Pro Light, serif;\"><i>nas\u0131l<\/i><\/span> ve hangi \u015fartlar alt\u0131nda kullan\u0131laca\u011f\u0131na ili\u015fkin sorgulama ihtiyac\u0131 daha \u00e7ok kabul g\u00f6rm\u00fc\u015f ve daha sa\u011flam bir yer edinmi\u015f olsa da alan\u0131n ilk y\u0131llar\u0131nda bu konular \u00e7ok k\u0131y\u0131da k\u00f6\u015fede kalm\u0131\u015f konulard\u0131. \u00d6\u011frenme analiti\u011fi ile ilgili daha geni\u015f konular\u0131 ke\u015ffetmeye y\u00f6nelik ilk giri\u015fimler Vancouver&#8217;daki LAK &#8217;12&#8217;de sunuldu. Bu konferanstaki oturumlar\u0131n b\u00fcy\u00fck \u00e7o\u011funlu\u011fu geli\u015fimsel \u00e7al\u0131\u015fmaya odaklanm\u0131\u015f olarak kald\u0131. Drachsler ve Greller (2012) ba\u015fta olmak \u00fczere, \u00f6\u011frenci verilerinin nas\u0131l kullan\u0131laca\u011f\u0131na ili\u015fkin payda\u015f alg\u0131lar\u0131ndan da bahsedilmi\u015fti ancak makalelerinde ankete kat\u0131lan payda\u015flar\u0131n de\u011ferlendirmelerinin b\u00fcy\u00fck oranda gizlilik odakl\u0131 oldu\u011fu ve \u00f6zellikle tart\u0131\u015fmal\u0131 olarak kabul edilmedi\u011fi \u00f6ne s\u00fcr\u00fclm\u00fc\u015ft\u00fc. Bir ba\u015fka makale (Prinsloo, Slade ve Galpin, 2012), \u00f6\u011frencilerin \u00f6\u011frenme ba\u015far\u0131lar\u0131n\u0131 art\u0131rmak i\u00e7in, onlar\u0131n \u00f6\u011frenme yolculuklar\u0131nda <span style=\"font-family: Source Serif Pro Light, serif;\"><i>t\u00fcm<\/i><\/span> payda\u015flar\u0131n etkilerinin d\u00fc\u015f\u00fcn\u00fclmesi gerekti\u011fine de\u011finilmi\u015ftir. \u201c\u00dc\u00e7l\u00fc alan\u201d kavram\u0131, zorluklar\u0131n ve f\u0131rsatlar\u0131n haritalanmas\u0131nda yararl\u0131 bir bulgusal alan sa\u011flaman\u0131n yan\u0131 s\u0131ra, \u00f6\u011frenme analiti\u011finin \u00e7eli\u015fkilerini ve bunun \u00f6\u011frenci ba\u015far\u0131s\u0131 ve kal\u0131c\u0131 \u00f6\u011frenme \u00fczerindeki potansiyel etkisini de ortaya koymu\u015ftur. Ayn\u0131 konferansta, Campbell, DeBlois ve Oblinger&#8217;in (2007) ilk d\u00f6nem \u00e7al\u0131\u015fmalar\u0131na dayanan ve farkl\u0131 payda\u015flar\u0131n bak\u0131\u015f a\u00e7\u0131lar\u0131ndan bir dizi ilgili etik meseleyi g\u00f6z \u00f6n\u00fcnde bulundurmay\u0131 ve geni\u015fletmeyi ama\u00e7layan bir ara\u015ft\u0131rma \u00e7al\u0131\u015ftay\u0131 (Slade ve Galpin, 2012) yap\u0131lm\u0131\u015ft\u0131r.<\/p>\n<p style=\"text-align: justify;\">2013 y\u0131l\u0131nda yap\u0131lan \u00e7al\u0131\u015fmalarda, verilerin nas\u0131l kullan\u0131laca\u011f\u0131 ve korunaca\u011f\u0131 konusundaki ama\u00e7lar\u0131 belirleyen mevcut kurumsal politika \u00e7er\u00e7evelerinin incelenmesine ba\u015flanm\u0131\u015ft\u0131r (Prinsloo ve Slade, 2013). B\u00fcy\u00fcyen ve geli\u015fen \u00f6\u011frenme analiti\u011fi, \u00f6\u011frenci verilerinin kullan\u0131mlar\u0131n\u0131n da h\u0131zla artt\u0131\u011f\u0131n\u0131 g\u00f6rm\u00fc\u015ft\u00fcr. Genel olarak, \u00f6\u011frenci verilerinin kurumsal kullan\u0131m\u0131na ili\u015fkin politikalar bu art\u0131\u015fa ayak uyduramam\u0131\u015ft\u0131r; \u00f6zellikle veri y\u00f6netimine, veri g\u00fcvenli\u011fine ve gizlilik konular\u0131na odaklanarak etik kayg\u0131lar\u0131 g\u00f6z \u00f6n\u00fcne alma ihtiyac\u0131na gereken \u00f6nemi vermemi\u015ftir. \u0130nceleme, mevcut politikalardaki bo\u015fluklar\u0131 ve yetersizlikleri tespit etmi\u015ftir.<\/p>\n<p style=\"text-align: justify;\">Slade ve Prinsloo (2013), \u00f6\u011frenme analiti\u011finin kullan\u0131m\u0131na dair sosyo-ele\u015ftirel bir bak\u0131\u015f a\u00e7\u0131s\u0131 kullanarak \u00f6\u011frenme analiti\u011finin etik kullan\u0131m\u0131n\u0131n kapsam\u0131n\u0131 ve tan\u0131m\u0131n\u0131 etkileyen bir dizi konuyu ele alm\u0131\u015ft\u0131r. Bir dizi etik sorun birbiriyle \u00f6rt\u00fc\u015fen \u00fc\u00e7 geni\u015f kategoride grupland\u0131r\u0131lm\u0131\u015ft\u0131r. Bunlar:<\/p>\n<ul>\n<li>\n<p style=\"text-align: justify;\">Verilerin yeri ve yorumlanmas\u0131<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">A\u00e7\u0131k r\u0131za, mahremiyet ve verilerin anonimle\u015ftirilmesi<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Verilerin y\u00f6netimi, s\u0131n\u0131fland\u0131r\u0131lmas\u0131 ve depolanmas\u0131d\u0131r<\/p>\n<\/li>\n<\/ul>\n<p style=\"text-align: justify;\">Slade ve Prinsloo (2013) a\u015fa\u011f\u0131daki alt\u0131 ilkeye dayanan bir \u00e7er\u00e7eve \u00f6nermi\u015ftir:<\/p>\n<ol>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenme Analiti\u011finin ahlaki uygulamas\u0131n\u0131n sadece neyin ge\u00e7erli oldu\u011fu hususunun yan\u0131s\u0131ra neyin uygun ve ahlaki olarak gerekli oldu\u011fu \u00fczerinde odaklanmas\u0131<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frencilerin sadece m\u00fcdahale ve hizmetlerin al\u0131c\u0131s\u0131 olmay\u0131p ayn\u0131 zamanda ortak ve birli\u011fi yap\u0131lacak arac\u0131lar olmas\u0131<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Ge\u00e7ici dinamik yap\u0131lar olarak \u00f6\u011frenci kimli\u011fi ve performans\u0131 analiti\u011fin belirli bir zamanda ve ba\u011flamda \u00f6\u011frencinin anl\u0131k g\u00f6r\u00fcnt\u00fcs\u00fcn\u00fc sa\u011flad\u0131\u011f\u0131n\u0131n fark\u0131nda olunmas\u0131<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00c7ok boyutlu, karma\u015f\u0131k bir olgu olarak \u00f6\u011frenci ba\u015far\u0131s\u0131<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Bireyin kimli\u011finin korunmas\u0131, verilere eri\u015fim, hangi verilerin hangi \u015fartlar alt\u0131nda hangi ama\u00e7lar i\u00e7in kullan\u0131laca\u011f\u0131 konusunda \u00f6nem ta\u015f\u0131yan \u015feffafl\u0131k<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Y\u00fcksek\u00f6\u011frenimin veri kullanmamay\u0131 g\u00f6ze alamamas\u0131<\/p>\n<\/li>\n<\/ol>\n<p style=\"text-align: justify;\">Bu ilkeler, faydal\u0131 bir ba\u015flang\u0131\u00e7 pozisyonu sunar ancak (hangi \u015fartlar alt\u0131nda) kimin yarar\u0131na oldu\u011fu hususunda tam bir anlay\u0131\u015f\u0131n geli\u015ftirilmesi gibi birtak\u0131m pratik hususlar dikkate al\u0131narak desteklenmelidir. Onay al\u0131narak kurumsal pozisyonlar\u0131n olu\u015fturulmas\u0131, tan\u0131mlar\u0131n belirlenmesi ve reddedilmesi konusunda; g\u00fcvenlik a\u00e7\u0131\u011f\u0131 ve zarar\u0131na ili\u015fkin sorunlar (\u00f6r. yanl\u0131\u015fl\u0131kla etiketleme), tazminat sistemleri (hem \u00f6\u011frenci hem de kurum i\u00e7in), veri toplama, analizler, eri\u015fim ve depolama (\u00f6r. g\u00fcvenlik sorunlar\u0131 ve \u00f6nyarg\u0131n\u0131n s\u00fcreklili\u011finden ka\u00e7\u0131nma) ile y\u00f6netim ve kaynak tahsisi kar\u015f\u0131lanmal\u0131d\u0131r.<\/p>\n<p style=\"text-align: justify;\">Kaynak tahsisinin bu ikinci y\u00f6n\u00fc, e\u011fitim triyaj\u0131 kavram\u0131 g\u00f6z \u00f6n\u00fcne al\u0131narak daha sonraki bir makalede daha ayr\u0131nt\u0131l\u0131 incelenmi\u015ftir (Prinsloo ve Slade, 2014a). \u00d6\u011frenme analiti\u011fi, Y\u00d6K&#8217;lere (y\u00fcksek\u00f6\u011fretim kurumlar\u0131) ba\u015far\u0131s\u0131z veya b\u0131rakma riski alt\u0131nda olan \u00f6\u011frencileri proaktif olarak tan\u0131tmak ve desteklemek i\u00e7in <span style=\"font-family: Source Serif Pro Light, serif;\"><i>teorik<\/i><\/span> f\u0131rsatlar sunsa da bunlar\u0131 kaynaklar\u0131n gittik\u00e7e s\u0131n\u0131rland\u0131\u011f\u0131 bir ba\u011flamda yapar. O halde zorluk, destek kaynaklar\u0131n\u0131 y\u00f6nlendirmede en <span style=\"font-family: Source Serif Pro Light, serif;\"><i>iyi yerin neresi<\/i><\/span> oldu\u011fu ve bu karar\u0131n hangi temele dayanarak verildi\u011finin bilinmesidir. Deste\u011fin \u201chayatta kalmas\u0131\u201d muhtemel olan \u00f6\u011frencilere y\u00f6nelik y\u00f6nlendirmenin bir arac\u0131 olarak e\u011fitim \u00f6nceli\u011fi kavram\u0131, \u00f6\u011frenci \u00f6zerkli\u011fine sayg\u0131 g\u00f6sterme ve ayn\u0131 zamanda kurumun uzun vadeli s\u00fcrd\u00fcr\u00fclebilirli\u011fini sa\u011flama aras\u0131ndaki denge, yararlanma (her zaman \u00f6\u011frencinin \u00e7\u0131kar\u0131na ilgi g\u00f6stermek i\u00e7in), zarar vermeme ihtiyac\u0131 (en az zararla en faydal\u0131 sonuca varmak) ve da\u011f\u0131t\u0131c\u0131 bir adalet duygusu sa\u011flama (demografik \u00f6zelliklerin sa\u011flanan deste\u011fi, varsay\u0131mlar\u0131, bunlar\u0131 tan\u0131ma ve \u00e7\u00f6zme ihtiyac\u0131n\u0131 etkiledi\u011fini anlamak) gibi bir dizi karma\u015f\u0131k konunun dikkatlice de\u011ferlendirilmesini gerektirir.<\/p>\n<p style=\"text-align: justify;\">\u00d6\u011frencinin haberinin olmas\u0131n\u0131 gerektirmeden <span style=\"font-family: Source Serif Pro Light, serif;\"><i>bir \u015feyler yapman\u0131n bir yolu<\/i><\/span> olarak \u00f6\u011frenme analiti\u011fi bilincindeki art\u0131\u015f; g\u00f6zlem, \u00f6\u011frenci mahremiyeti ve kurumsal g\u00fcvenilirlik konular\u0131n\u0131n daha da ara\u015ft\u0131r\u0131lmas\u0131n\u0131 tetiklemi\u015ftir (Prinsloo ve Slade, 2014b). Sonu\u00e7ta ortaya \u00e7\u0131kan tart\u0131\u015fma \u00f6\u011frenci \u00f6\u011frenmesinin do\u011fru, nesnel, tamamen eksiksiz resimlerinin bir \u00fcreticisi olarak \u00f6\u011frenme analiti\u011fi etraf\u0131ndaki varsay\u0131mlara meydan okuyarak kurum ve \u00f6\u011frenci aras\u0131ndaki e\u015fit olmayan olas\u0131 ili\u015fkiyi yeniden ele alm\u0131\u015ft\u0131r. Ki\u015fisel verilerin kullan\u0131m\u0131 ve analizi ile ilgili mevcut \u00e7er\u00e7eveler g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, \u00e7al\u0131\u015fma, <span style=\"font-family: Source Serif Pro Light, serif;\"><i>\u00f6\u011frenci merkezli<\/i><\/span> bir \u00f6\u011frenme analiti\u011fi i\u00e7in temel olu\u015fturabilecek alt\u0131 unsur \u00f6nermektedir:<\/p>\n<ol>\n<li>\n<p style=\"text-align: justify;\">K\u00fcmelenmi\u015f, ki\u015fiselle\u015ftirilmemi\u015f verilerin kullan\u0131m\u0131, etkili ve uygun bir \u00f6\u011fretme ve \u00f6\u011frenme sa\u011flamak i\u00e7in esast\u0131r ancak \u00f6\u011frenciler giri\u015f \/ \u00e7\u0131k\u0131\u015f kararlar\u0131n\u0131 bilin\u00e7li bir \u015fekilde yapabilmelidirler<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenciler hangi verilerin topland\u0131\u011f\u0131 ve nas\u0131l kullan\u0131ld\u0131\u011f\u0131 konusunda tam bilgi sahibi olmal\u0131d\u0131rlar<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenciler ki\u015fisel veri kay\u0131tlar\u0131n\u0131n eksiksiz ve g\u00fcncel oldu\u011fundan emin olmal\u0131d\u0131rlar<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Etkinliklerin izlenmesi ve verilerin toplanmas\u0131 \u00f6\u011frenci ilerlemesine zarar vermemelidir<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Algoritmik \u00e7\u0131kt\u0131 (potansiyel) insan incelemesine tabi tutulmal\u0131 ve gerekirse d\u00fczeltilmelidir<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenme analitikleri temel olarak ba\u011flam ve zamana \u00f6zg\u00fc, ge\u00e7ici, tamamlanmam\u0131\u015f \u00f6\u011frencilerin portrelerini sa\u011flamal\u0131, algoritmalar s\u0131kl\u0131kla g\u00f6zden ge\u00e7irilmeli ve do\u011frulanmal\u0131d\u0131r<\/p>\n<\/li>\n<\/ol>\n<p style=\"text-align: justify;\">G\u00f6zetime ili\u015fkin sorunlar ve \u00f6\u011frencilerin kendi verilerinin kullan\u0131m\u0131n\u0131n aktif faili olarak tan\u0131nmas\u0131 gereklili\u011fi, A\u00e7\u0131k \u00dcniversitenin (A\u00dc; 2014) \u00f6\u011frenci analiti\u011finin \u00f6\u011frenci verilerinin etik kullan\u0131m\u0131 politikas\u0131n\u0131n geli\u015ftirilmesi kapsam\u0131nda a\u00e7\u0131k\u00e7a ele al\u0131nm\u0131\u015ft\u0131r. Payda\u015flarla isti\u015farenin bir par\u00e7as\u0131 olarak, 50 \u00f6\u011frenciden olu\u015fan bir temsilci grubu, \u00fc\u00e7 haftal\u0131k bir s\u00fcre zarf\u0131nda \u00e7al\u0131\u015fma hedeflerini tamamlamada \u00f6\u011frencilere destek olmak i\u00e7in verilerin nas\u0131l kullan\u0131ld\u0131\u011f\u0131na dair yakla\u015f\u0131mlar\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Bir anket ara\u015ft\u0131rmas\u0131 (Slade ve Prinsloo, 2014), \u00f6\u011frencilerin, verilerin aktif olarak toplanma ve kullan\u0131lma derecelerinin b\u00fcy\u00fck \u00f6l\u00e7\u00fcde fark\u0131nda olmad\u0131klar\u0131n\u0131 ve birtak\u0131m endi\u015feleri dile getirdiklerini ortaya koymu\u015ftur. \u00d6\u011frencilerin \u00e7o\u011funlu\u011fuyla aktif olarak r\u0131za g\u00f6sterme (ya da g\u00f6stermeme) potansiyeli ile ilgili temel kayg\u0131, sonradan bu tercihlerinden vazge\u00e7ebilme iste\u011finin olmas\u0131d\u0131r. \u00d6\u011frencilerin bu do\u011frudan kat\u0131l\u0131m\u0131, \u00f6\u011frenme analiti\u011fi eti\u011fiyle ilgili bir politikay\u0131 \u015fekillendirmede s\u00f6z konusudur; \u00f6\u011frencilerin verilerini dikkatli bir \u015fekilde korunmas\u0131n\u0131 ve hatta daha dikkatli bir \u015fekilde uygulanmalar\u0131 i\u00e7in de\u011ferli bir varl\u0131k olarak g\u00f6r\u00fclmesi y\u00f6n\u00fcnde benzersiz bir fikir vermi\u015ftir. Slade ve Prinsloo&#8217;daki (2014) \u00f6rneklemin toplam evreni tam olarak temsil edemeyebilece\u011fi g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, sonu\u00e7lar kurumsal ve jeopolitik ba\u011flamda genellenemez.<\/p>\n<p style=\"text-align: justify;\">Bu artan \u00f6\u011frenci kayg\u0131s\u0131 konusundaki fark\u0131ndal\u0131\u011fa cevaben, Prinsloo ve Slade (2015), \u00f6\u011frencilerin mahremiyetle ilgili tutumlar\u0131n\u0131 \u00e7evreleyen varsay\u0131mlar\u0131m\u0131z\u0131n ve anlay\u0131\u015f\u0131m\u0131z\u0131n hem halk\u0131n ya\u015famlar\u0131n\u0131n detaylar\u0131n\u0131 payla\u015fmakta g\u00f6r\u00fclen belirgin kolayl\u0131ktan hem de b\u00fcy\u00fck \u00f6l\u00e7\u00fcde ataerkil kurum k\u00fclt\u00fcrlerinden etkilenip etkilenmedi\u011fini sorgulam\u0131\u015ft\u0131r. \u00c7al\u0131\u015fma, \u00f6\u011frencilerin verilerini izlemelerine izin vermemelerini veya se\u00e7memelerini sa\u011flamak i\u00e7in onay verme ve g\u00f6r\u00fcn\u00fc\u015fte basit bir se\u00e7im konusunda ya\u015fad\u0131klar\u0131 sorunlar\u0131 ara\u015ft\u0131rm\u0131\u015ft\u0131r. Tart\u0131\u015fman\u0131n temeli olarak, \u00fc\u00e7 kitlesel a\u00e7\u0131k \u00e7evrimi\u00e7i ders (KA\u00c7D) tedarik\u00e7isinin \u015fart ve ko\u015fullar\u0131, kullan\u0131c\u0131lara verilerinin kullan\u0131m\u0131yla ilgili verilen bilgileri belirlemek i\u00e7in g\u00f6zden ge\u00e7irildi. Bu Y\u00d6K&#8217;lerin \u00f6\u011frencileri ki\u015fisel verilerindeki \u00f6\u011frenme analiti\u011fi \u00fczerindeki etkileri hakk\u0131nda tam olarak bilgilendiren bir yakla\u015f\u0131ma do\u011fru nas\u0131l hareket edebileceklerine dair bir tart\u0131\u015fmaya evrilmi\u015ftir. Benzer bir tema Prinsloo ve Slade&#8217;de (2016a) izlenmi\u015ftir. Bu makale, bir\u00e7ok Y\u00d6K&#8217;\u00fcn \u00f6\u011frenci verilerine otoriter bir yakla\u015f\u0131m benimseme e\u011filimine itiraz etmi\u015ftir. \u00d6\u011frenme analitiklerinin yay\u0131lmas\u0131ndaki h\u0131zl\u0131 b\u00fcy\u00fcmeye ra\u011fmen, baz\u0131 Y\u00d6K&#8217;lerin d\u00fczenleyici \u00e7er\u00e7eveleri vard\u0131r ve \/ veya toplanan, analiz edilen, kullan\u0131lan ve payla\u015f\u0131lan \u00f6\u011frenci verilerinin kapsam\u0131 konusunda tamamen \u015feffaft\u0131rlar. \u00d6\u011frenci hassasl\u0131\u011f\u0131 \u00f6\u011frenme analiti\u011fi potansiyelini ger\u00e7ekle\u015ftirmek ve Y\u00d6K&#8217;lerin g\u00fc\u00e7 ili\u015fkileri, asimetrik bilgi ve \u00f6\u011frencilerle g\u00fc\u00e7 ili\u015fkileri ba\u011flam\u0131nda ve \u00f6\u011frenci eylemlili\u011fini \u00e7evreleyen \u00f6\u011frenme analiti\u011fi konusundaki karma\u015f\u0131kl\u0131klar aras\u0131ndaki ba\u011fda incelenmi\u015ftir. Ama\u00e7, \u00f6\u011frenci k\u0131r\u0131lganl\u0131\u011f\u0131n\u0131n nas\u0131l ele al\u0131nabilece\u011fini, \u00f6\u011frenci eylemlili\u011fini art\u0131rarak ve onlar\u0131 nicelikli veri nesnelerinden nitelikli ve kalifiye ki\u015filere do\u011fru ilerleyen \u00f6\u011frenme analiti\u011fi aktif kat\u0131l\u0131mc\u0131lar\u0131 olarak g\u00fc\u00e7lendirme yollar\u0131n\u0131 d\u00fc\u015f\u00fcnmektir (ayr\u0131ca bk. Prinsloo ve Slade, 2016b).<\/p>\n<h2 class=\"western\">ET\u0130K \u00c7ER\u00c7EVELER \u0130LE \u0130LG\u0130L\u0130 SON GEL\u0130\u015eMELER<\/h2>\n<p style=\"text-align: justify;\">Verinin, y\u00fckselen de\u011fi\u015f toku\u015f edilebilme de\u011feriyle birlikte, payla\u015f\u0131labilir ticari bir \u00fcr\u00fcn olarak artan de\u011ferinin, yasal ve geleneksel etik \u00e7er\u00e7evelerimizi a\u015ft\u0131\u011f\u0131 genel kabul g\u00f6rmektedir (Zhang, 2016). \u201cDerin ekonomik bask\u0131lar, ba\u011flant\u0131n\u0131n yo\u011funla\u015fmas\u0131n\u0131 ve \u00e7evrimi\u00e7i izlemenin g\u00fc\u00e7lenmesini sa\u011flamaktad\u0131r\u201d (Couldry, 2016, b\u00f6l.13) ve \u201cihtiya\u00e7 duyulan \u015fey, bizim etik ya\u015fam olanaklar\u0131m\u0131z i\u00e7in kapitalizmin yeni veri ili\u015fkileri maliyetleri olup (Couldry, 2016, b\u00f6l. 35) daha kolektif bir yans\u0131mas\u0131d\u0131r&#8221;. Bu nedenle, \u00f6\u011frenme analiti\u011fi etik uygulamalar\u0131 ile u\u011fra\u015fmak i\u00e7in farkl\u0131 jeopolitik ve kurumsal ba\u011flamlarda giri\u015fimler olmu\u015ftur. Sclater, Peasgood ve Mullan (2016), \u00f6rne\u011fin, Amerika Birle\u015fik Devletleri, Avustralya ve Birle\u015fik Krall\u0131k&#8217;ta y\u00fcksek\u00f6\u011fretimdeki uygulamalar\u0131 g\u00f6zden ge\u00e7irmi\u015ftir. Bulgular\u0131n\u0131, \u00f6\u011frenme analitiklerinin 1) kalite g\u00fcvencesi ve kalite iyile\u015ftirmede 2) tutma oranlar\u0131n\u0131n artt\u0131r\u0131lmas\u0131nda; 3) \u00f6\u011frenci pop\u00fclasyonu aras\u0131ndaki farkl\u0131 sonu\u00e7lar\u0131 de\u011ferlendirme ve bunlara etki etmede ve 4) uyarlanabilir \u00f6\u011frenmenin geli\u015fimi ve tan\u0131t\u0131lmas\u0131nda \u00f6nemli katk\u0131larda bulundu\u011funu belirterek \u00f6zetlemektedirler. Rapor bir\u00e7ok imk\u00e2n\u0131 tan\u0131sa da ayn\u0131 zamanda \u201cetik ve veri gizlili\u011fi sorunlar\u0131, &#8220;a\u015f\u0131r\u0131 analiz&#8221; ve sonu\u00e7lar\u0131n genelle\u015ftirilememesi, modellerin yanl\u0131\u015f s\u0131n\u0131fland\u0131r\u0131lmas\u0131 ve \u00e7eli\u015fkili bulgular\u201d gibi tehditleri de vurgulamaktad\u0131r (s. 16). \u0130ncelemelerinde, \u00f6\u011frenme analiti\u011findeki etik kayg\u0131lar\u0131 ele almak i\u00e7in politika d\u00fczeyinde bir teklifin kurumsal bir \u00f6rne\u011fi A\u00e7\u0131k \u00dcniversite (\u0130ngiltere) dir. 2014 y\u0131l\u0131nda, A\u00dc, toplanan ve analiz edilen verilerin niteli\u011fini ve kapsam\u0131n\u0131 s\u0131n\u0131rlayan \u201c\u00f6\u011frenme analiti\u011fi i\u00e7in \u00f6\u011frenci verilerinin etik kullan\u0131m\u0131 politikas\u0131\u201d n\u0131 ve toplanan \u00f6\u011frenme analiti\u011fi i\u00e7in <span style=\"font-family: Source Serif Pro Light, serif;\"><i>kullan\u0131lmayacak<\/i><\/span> verilerin a\u00e7\u0131k bir spesifikasyonunu yay\u0131nlam\u0131\u015ft\u0131r. Politika a\u015fa\u011f\u0131daki sekiz prensibi i\u00e7erir (s. 6):<\/p>\n<ol>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenme analiti\u011fi, lisans d\u00fczeyinde \u00e7al\u0131\u015fmaya a\u00e7\u0131k giri\u015f gibi temel d\u00fczenleme ilkeleri ile uyumlu olmas\u0131 gereken etik bir uygulamad\u0131r.<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">A\u00dc, m\u00fcmk\u00fcn oldu\u011funda \u00f6\u011frencilerin yarar\u0131na, \u00f6\u011frenci verilerinden anlam \u00e7\u0131karmak ve bunlar\u0131 kullanmak i\u00e7in t\u00fcm payda\u015flara kar\u015f\u0131 sorumludur.<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenciler tamamen g\u00f6zle g\u00f6r\u00fclebilir veriler veya bu verileri yorumlamam\u0131zla tan\u0131mlanmamal\u0131d\u0131r. [Dahas\u0131, bu ilke normlara uyan \u00f6\u011frencilere kar\u015f\u0131 uyar\u0131 verir ve genel kal\u0131plara uymayan bireyleri kabul eder. Bu ilke ayr\u0131ca, \u00e7al\u0131\u015fanlar\u0131n analizin g\u00fcvenilirli\u011fini ciddi \u015fekilde etkileyebilecek t\u00fcm veri k\u00fcmesine eri\u015femeyebilece\u011fini de a\u00e7\u0131k\u00e7a ortaya koymaktad\u0131r.]<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenme analitiklerinin kullan\u0131m\u0131na ili\u015fkin ama\u00e7 ve s\u0131n\u0131rlar iyi tan\u0131mlanm\u0131\u015f ve g\u00f6r\u00fcn\u00fcr olmal\u0131d\u0131r.<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00dcniversite veri toplama konusunda \u015feffaft\u0131r ve \u00f6\u011frencilere d\u00fczenli aral\u0131klarla kendi verilerini g\u00fcncelleme f\u0131rsat\u0131 sunar.<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u00d6\u011frenciler \u00f6\u011frenme analiti\u011finin uygulanmas\u0131nda etkin \u00f6zne olarak kullan\u0131lmal\u0131d\u0131r (\u00f6r. A\u00e7\u0131k r\u0131za, ki\u015fiselle\u015ftirilmi\u015f \u00f6\u011frenme yollar\u0131, m\u00fcdahaleler).<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Verilerin analizine dayanan modelleme ve m\u00fcdahaleler sa\u011flam ve \u00f6nyarg\u0131s\u0131z olmal\u0131d\u0131r.<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">A\u00dc i\u00e7indeki \u00f6\u011frenme analitiklerini benimsemek, de\u011ferlerin ve faydalar\u0131n (kurum k\u00fclt\u00fcr\u00fc) geni\u015f kabul\u00fcn\u00fc ve kurum genelinde uygun becerilerin geli\u015ftirilmesini gerektirir.<\/p>\n<\/li>\n<\/ol>\n<p style=\"text-align: justify;\">Bu politika ve ilkeleri \u00f6\u011frenci verilerinin toplanmas\u0131nda, analiz edilmesinde ve kullan\u0131lmas\u0131ndaki etik etkilere ilk kurumsal tepkilerden biri olarak, ke\u015ffedilmemi\u015f b\u00f6lgeleri haritalamaya \u00e7al\u0131\u015fm\u0131\u015ft\u0131r. \u00d6zel ilgi alan\u0131, \u201c\u00f6\u011frencinin, verisinin bir k\u0131sm\u0131n\u0131n veya tamam\u0131n\u0131n \u00f6\u011frenme analiti\u011fi i\u00e7in kullan\u0131labilece\u011fi ve r\u0131za g\u00f6sterdi\u011fi ama\u00e7lar\u0131n bilincinde oldu\u011fu s\u00fcrece yap\u0131lan \u201ca\u00e7\u0131k r\u0131za \u201dtan\u0131m\u0131d\u0131r. Bilgilendirilmi\u015f onay, bir mod\u00fcl veya yeterlilik i\u00e7in rezervasyon veya kay\u0131t s\u0131ras\u0131nda uygulan\u0131r \u201d(A\u00e7\u0131k \u00dcniversite, 2014, s. 3). Politika, verilerin toplanmas\u0131n\u0131, analiz edilmesini ve kullan\u0131lmamas\u0131n\u0131 tercih eden \u00f6\u011frencilerin olas\u0131l\u0131\u011f\u0131n\u0131 ele almamaktad\u0131r (Engelfriet, Manderveld ve Jeunink, 2015; Sclater, 2015; ayr\u0131ca bk. Shacklock, 2016).<\/p>\n<p style=\"text-align: justify;\">Avustralya ba\u011flam\u0131nda analitik uygulamalar\u0131n\u0131n \u00f6\u011frenilmesine genel bir bak\u0131\u015fta, Dawson, Ga\u0161evi\u0107 ve Rogers (2016) \u201c\u00e7al\u0131\u015fmalar boyunca eti\u011fe verilen g\u00f6receli sessizli\u011fin \u00f6nemli oldu\u011funu\u201d (s. 3) ve bunun \u201c sekt\u00f6r\u00fcn bu konular\u0131 dikkate almas\u0131 gereken ciddiyet\u201d (s. 33), oldu\u011funu belirtmi\u015ftir. Kurumlar olgunla\u015ft\u0131k\u00e7a, etik kayg\u0131lar\u0131n daha da belirginle\u015fmekte oldu\u011fu savunulsa da rapora g\u00f6re y\u00fcksek\u00f6\u011frenim sekt\u00f6r\u00fcn\u00fcn daha \u00f6nce b\u00f6yle bir sohbete haz\u0131r olmad\u0131\u011f\u0131 muhtemeldir (s. 33)<\/p>\n<p style=\"text-align: justify;\">Ayr\u0131ca Avustralya ba\u011flam\u0131nda, Galce ve McKinney (2015) \u201cdisiplinin kurumlar, pratisyenler ve teknoloji sat\u0131c\u0131lar\u0131 ile g\u00f6receli olgunla\u015fmam\u0131\u015fl\u0131\u011f\u0131\u201d ba\u011flam\u0131nda, \u201cneyin i\u015fe yarad\u0131\u011f\u0131n\u0131\u201d ve uygulamalar\u0131n s\u0131n\u0131rlar\u0131n\u0131 bulmay\u0131 ve k\u00f6t\u00fcye kullanma potansiyelini bulma ihtiyac\u0131na dikkat \u00e7ekmektedir (s. 588). \u00dcniversitenin a) \u00f6\u011frenme ve \u00f6\u011fretme ile do\u011frudan ilgili olmayan b) \u00f6\u011frencilerin \u00fcniversitenin b\u00f6yle bir veri koleksiyonu olu\u015fturmas\u0131n\u0131 beklemedi\u011fi veri kaynaklar\u0131n\u0131 bar\u0131nd\u0131ran bir \u00f6\u011frenme analiti\u011fi kullanmayaca\u011f\u0131na dair taahh\u00fcd\u00fc \u00f6nemlidir (s. 590). \u00d6\u011frenci verileri, yaln\u0131zca s\u00f6z konusu verilerin topland\u0131\u011f\u0131 orijinal ama\u00e7 ba\u011flam\u0131nda kullan\u0131lacak olup a\u015fa\u011f\u0131daki \u015fartlar alt\u0131nda kullan\u0131m\u0131 devam edebilir:<\/p>\n<p style=\"text-align: justify;\">A\u00e7\u0131k \u015fekilde bilgilendirilmi\u015f onay \u00f6l\u00e7\u00fcme konu olanlardan sa\u011flan\u0131r. Bilgilendirilmi\u015f onay \u015fu anlama gelir: (a) hangi verilerin topland\u0131\u011f\u0131 veya toplanabilece\u011fi, neden ve nas\u0131l topland\u0131\u011f\u0131, nas\u0131l depoland\u0131\u011f\u0131 ve nas\u0131l kullan\u0131ld\u0131\u011f\u0131 hakk\u0131nda a\u00e7\u0131k ve do\u011fru bilgiler verilir ve (b) a\u00e7\u0131klanan uygulamaya \/ uygulamalara serbest\u00e7e izin verilir (s.590).<\/p>\n<p style=\"text-align: justify;\">Yukar\u0131daki ilkeler, \u00f6\u011fretme ve \u00f6\u011frenmeyi geli\u015ftirmek ve \u00f6\u011frencilere \u201c\u00f6\u011frenmeleri \u00fczerinde daha fazla kontrol ve sorumluluk sa\u011flamak\u201d i\u00e7in toplanan verilerin nas\u0131l kullan\u0131lmas\u0131 gerekti\u011fine ili\u015fkin olarak kalan iki ilkeyle birlikte okunmal\u0131d\u0131r (s. 591). Tam bir tart\u0131\u015fma i\u00e7in, bk. Galce ve McKinney (2015).<\/p>\n<p style=\"text-align: justify;\">Drachsler ve Greller (2016) etik, mahremiyet ve ilgili yasal \u00e7er\u00e7evelere genel bir bak\u0131\u015f sa\u011flar ve veri toplay\u0131c\u0131 ile veri nesnesi aras\u0131ndaki asimetrik g\u00fc\u00e7 ili\u015fkisi \u0131\u015f\u0131\u011f\u0131nda ger\u00e7ek kullan\u0131m olas\u0131l\u0131\u011f\u0131, m\u00fclkiyet sorunlar\u0131, anonimlik ve veri g\u00fcvenli\u011fi, gizlilik ve veri kimli\u011finin yan\u0131 s\u0131ra \u015feffafl\u0131k ve g\u00fcven unsurlar\u0131 hakk\u0131nda da bilgi verir. \u00d6\u011frenme analitiklerinin \u201cveri toplama ve i\u015fleme politikalar\u0131na ba\u011fl\u0131 korkular\u0131n \u00fcstesinden gelmek i\u00e7in\u201d kabul edilebilir ve uyumlu bir \u015fekilde ilerlemesini sa\u011flamak amac\u0131yla bir kontrol listesi (DELICATE \u00a9) sunarlar (s. 96).<\/p>\n<p style=\"text-align: justify;\">Sclater (2015), \u00fcst d\u00fczey y\u00f6netim, analiz komitesi, veri bilimcileri, e\u011fitim ara\u015ft\u0131rmac\u0131lar\u0131, BT ve \u00f6\u011frenciler gibi bir dizi payda\u015f\u0131n \u00f6\u011frenme analiti\u011finden nas\u0131l etkilendi\u011fi ve ona ili\u015fkin nas\u0131l sorumluluk ald\u0131\u011f\u0131na dair genel bir bak\u0131\u015f ile \u00f6\u011frenme analiti\u011finde etik, yasal ve lojistik konulara dair bir (taslak) taksonomi \u00f6nerir. Taslak, ba\u015fka \u015feylerin yan\u0131 s\u0131ra, r\u0131za; kimlik; kapsam d\u0131\u015f\u0131 b\u0131rakman\u0131n olas\u0131 etkileri; kurum ve \u00f6\u011frenciler aras\u0131ndaki asimetrik ili\u015fki; (s\u0131n\u0131rlar) \u00f6\u011frenci verilerinin izin verilen kullan\u0131mlar\u0131; \u015feffafl\u0131k; kullan\u0131m d\u00e2hil (ve hari\u00e7) veriler ve di\u011ferleri aras\u0131nda \u00f6\u011frenci \u00f6zerkli\u011fini kapsar. Etik kayg\u0131lar\u0131n tam listesi i\u00e7in Sclater (2015)&#8217;a bak\u0131n\u0131z.<\/p>\n<p style=\"text-align: justify;\">Hollanda&#8217;daki y\u00fcksek\u00f6\u011frenim ba\u011flam\u0131nda Engelfriet vd. (2015), Ki\u015fisel Bilgilerin Korunmas\u0131 Kanununun, \u00f6\u011frenme analiti\u011fi \u00fczerindeki etkilerini g\u00f6z \u00f6n\u00fcne almaktad\u0131r. Bunlar, izin ihtiyac\u0131n\u0131 (ve r\u0131zan\u0131n al\u0131nmas\u0131ndan do\u011facak sorumlulu\u011fu) ve bir hizmet sa\u011flay\u0131c\u0131 ile al\u0131c\u0131 aras\u0131nda, hizmetin sa\u011flanmas\u0131 i\u00e7in ihtiya\u00e7 duyulan herhangi bir ki\u015fisel bilgiyi kullanabilece\u011fi konusunda yap\u0131lan onay anla\u015fmas\u0131n\u0131n sonu\u00e7lar\u0131n\u0131 i\u00e7erir. Yasa, <span style=\"font-family: Source Serif Pro Light, serif;\"><i>temel<\/i><\/span> bilgiler ile \u201ckullan\u0131\u015fl\u0131\u201d bilgiler aras\u0131nda ayr\u0131m yapar. Engelfriet vd. (2015) \u00f6\u011frenme analitiklerinin yeni ortaya \u00e7\u0131kan bir uygulama olarak g\u00f6r\u00fcld\u00fc\u011f\u00fc g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, g\u00fcvenli bir \u015fekilde \u201ckullan\u0131\u015fl\u0131\u201d bilgi toplanmas\u0131 olarak kabul edilebilece\u011fi ve belki de kurum ve \u00f6\u011frenciler aras\u0131ndaki fikir birli\u011fi anla\u015fmas\u0131ndan muaf tutulabilece\u011fi y\u00f6n\u00fcnde muhalefetli bir g\u00f6r\u00fc\u015f benimsemektedir. Yazarlar bu d\u00f6rt ilkenin \u00f6\u011frenme analiti\u011fini y\u00f6nlendirmesi gerekti\u011fini \u00f6ne s\u00fcrmektedirler:<\/p>\n<ul>\n<li>\n<p style=\"text-align: justify;\">Ki\u015fisel bilgiler, yaln\u0131zca temin edilen ba\u011flamda ve ama\u00e7larda kullan\u0131labilir<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Bu t\u00fcr verilerin daha sonra kullan\u0131lmas\u0131, orijinal ba\u011flam ve ama\u00e7 ile uyumlu olmal\u0131d\u0131r<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Veriler dikkatlice toplanmal\u0131 ve analiz edilmelidir ve \u201csinsi\u201d (Hollandaca \u201cstiekeme\u201d) analitik kullan\u0131m\u0131na izin verilmez; Bu saydaml\u0131k, \u00f6\u011frenci dan\u0131\u015fmanl\u0131\u011f\u0131 ve sat\u0131n alma konusundaki bir ihtiyac\u0131 vurgulamaktad\u0131r<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Veriler sadece toplanan verilerin amac\u0131 \/ kullan\u0131m\u0131 a\u00e7\u0131k\u00e7a belirtildi\u011fi zaman toplanabilir<\/p>\n<\/li>\n<\/ul>\n<p style=\"text-align: justify;\">Engelfriet vd. (2015), verilerinin y\u00f6netimi konusundaki \u00f6\u011frenci haklar\u0131n\u0131 a\u015fa\u011f\u0131dakiler de d\u00e2hil olmak \u00fczere ara\u015ft\u0131rm\u0131\u015flard\u0131r:<\/p>\n<ul>\n<li>\n<p style=\"text-align: justify;\">Toplanan bilgilere kolay eri\u015fim<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">Yanl\u0131\u015f bilgileri (veya bundan kaynaklanan yorumlar\u0131) d\u00fczeltme hakk\u0131<\/p>\n<\/li>\n<li>\n<p style=\"text-align: justify;\">\u0130lgisiz bilgileri kald\u0131rma hakk\u0131<\/p>\n<\/li>\n<\/ul>\n<p style=\"text-align: justify;\">\u00d6zellikle ilgi \u00e7ekici olan, algoritmik karar vermek i\u00e7in etik sonu\u00e7lar\u0131n ara\u015ft\u0131r\u0131lmas\u0131d\u0131r ve yazarlar, Hollanda yasalar\u0131yla olas\u0131 \u00e7at\u0131\u015fmaya yol a\u00e7an \u00f6rneklere i\u015faret etmektedirler. Bunun anlam\u0131, insanlar\u0131n algoritmik karar verme sorumlulu\u011funu \u00fcstlenmesi ve g\u00f6zetim alt\u0131nda tutmas\u0131 gerekti\u011fidir. Algoritmalar, en fazla, \u00fcniversite veya destek personelinin dikkatini \u00e7ekmek i\u00e7in \u00f6zel davran\u0131\u015flara i\u015faret edebilir. Ayr\u0131ca, \u00f6\u011frencilerin ki\u015fisel verilerinin analizlerine dayanarak verilen kararlara itiraz etme haklar\u0131 vard\u0131r. Y\u00d6K&#8217;lerin yaz\u0131l\u0131m geli\u015ftiricilere ta\u015feronluk yapt\u0131\u011f\u0131 durumlarda, nihai sorumluluk ve g\u00f6zetim kurumda g\u00fcvende kal\u0131r ve delege edilemez (bk. Engelfriet vd., 2015).<\/p>\n<h2 class=\"western\">GELECEKTEK\u0130 BAZI HUSUSLAR<\/h2>\n<p style=\"text-align: justify;\">\u00d6\u011frenci verileri ile teknolojideki ilerlemeler ve analiz metotlar\u0131 aras\u0131ndaki kesi\u015fmelerdeki karma\u015f\u0131kl\u0131k ve pratiklikleri anlamam\u0131zdaki mevcut ve gelecekteki bo\u015fluklar\u0131 haritaland\u0131rmak bu b\u00f6l\u00fcm\u00fcn kapsam\u0131 d\u0131\u015f\u0131nda kalmaktad\u0131r. Bununla birlikte, gelecekteki de\u011ferlendirmeler i\u00e7in baz\u0131 \u00f6nerilerde bulunmak istiyoruz.<\/p>\n<p style=\"text-align: justify;\">Etkili, uygun, uygun maliyetli \u00f6\u011frenme deneyimleri sa\u011flamak ve \u00f6\u011frencilerin ba\u015far\u0131l\u0131 olmalar\u0131n\u0131 desteklemek i\u00e7in y\u00fcksek\u00f6\u011fretim kurumlar\u0131n\u0131n g\u00f6rev s\u00fcreleri g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda, kurumlar\u0131n \u00f6\u011frenci bilgilerini toplama ve kullanma hakk\u0131na sahip oldu\u011fu konusunda geni\u015f bir anla\u015fma vard\u0131r. Bununla birlikte, \u00f6\u011frencilerin izinlerinin toplanmas\u0131ndan, analiz edilmesinden ve verilerinin kullan\u0131lmamas\u0131n\u0131 tercih etmelerine izin vermede onay konusunda mutab\u0131k bir pozisyon bulunmamaktad\u0131r. Onay ile ilgili \u00f6\u011frencilerin durumlar\u0131 tamamen mant\u0131kl\u0131 ya da ak\u0131lc\u0131 olmayan konulardan etkilenebilir. Faydalar\u0131n, maliyetlerin ve risklerin s\u0131kl\u0131kla \u00f6rt\u00fcl\u00fc olarak hesaplanmas\u0131, di\u011ferlerinin yan\u0131 s\u0131ra, \u00f6nceki deneyimler, ihtiya\u00e7 ve alg\u0131lanan faydalar gibi bir dizi fakt\u00f6re ba\u011fl\u0131 olacakt\u0131r (bak\u0131n\u0131z, \u00f6rne\u011fin, O&#8217;Brien&#8217;daki Daniel Pink, 2010).<\/p>\n<p style=\"text-align: justify;\">Son zamanlarda \u00e7ekilmenin bir \u00f6rne\u011fi, \u00f6\u011frencileri devlet taraf\u0131ndan zorunlu hale getirilmi\u015f standart testler almay\u0131 reddetmeye te\u015fvik eden ABD&#8217;deki Ulusal Adil ve A\u00e7\u0131k Test Merkezi taraf\u0131ndan g\u00f6sterilmi\u015ftir. 2014-2015 \u00f6\u011fretim y\u0131l\u0131nda (FairTest, nd) yakla\u015f\u0131k 650.000 \u00f6\u011frenci, ABD E\u011fitim Bakanl\u0131\u011f\u0131 taraf\u0131ndan finanse edilmeme tehdidine cevaben \u00e7ekilmi\u015ftir (Strauss, 2016a).<\/p>\n<p style=\"text-align: justify;\">\u00d6\u011frencilerin kayg\u0131lar\u0131, ayr\u0131lma haklar\u0131 ve y\u00fcksek\u00f6\u011fretimin \u00f6\u011frenci verilerini bireysel bir d\u00fczeyde m\u00fcdahalelerde bulunmak i\u00e7in kullanma yetkisinin olmas\u0131n\u0131n sonu\u00e7lar\u0131 aras\u0131ndaki olas\u0131 \u00e7at\u0131\u015fmalar\u0131 ara\u015ft\u0131rmak i\u00e7in daha fazla ara\u015ft\u0131rmaya ihtiya\u00e7 vard\u0131r. Bu konunun merkezinde \u201ckim yararlan\u0131r?\u201d sorusu bulunmaktad\u0131r. (bk. Watters, 2016). \u00d6\u011frenci verilerinin toplanmas\u0131, analizi ve kullan\u0131m\u0131yla ilgili etik kurallar\u0131n (gerek \u00f6\u011frenme analiti\u011finde gerek resm\u00ee de\u011ferlendirmelerde olsun), tart\u0131\u015fmal\u0131 iddialar\u0131 ve kazan\u0131lm\u0131\u015f ilgi alanlar\u0131n\u0131 da tan\u0131mas\u0131 gerekir.<\/p>\n<p style=\"text-align: justify;\">Daha geni\u015f bir \u00e7evrimi\u00e7i ara\u015ft\u0131rma ba\u011flam\u0131nda, Vitak, Shilton ve Ashktorab (2016), \u00e7evrimi\u00e7i ba\u011flamlardaki etik ara\u015ft\u0131rma uygulamalar\u0131 ile ilgili olarak, yeniden kimlik tan\u0131mlamas\u0131 yap\u0131lmas\u0131 konusunda gittik\u00e7e artarak devam eden endi\u015feler gibi \u00e7e\u015fitli zorluklara ki; \u201cara\u015ft\u0131rmac\u0131lar hala \u00e7evrimi\u00e7i veri k\u00fcmeleri kullan\u0131m\u0131yla ara\u015ft\u0131rma etikleri aras\u0131nda dengeyi bulmakta zorlanmaktad\u0131r\u201d i\u015faret etmektedir (s. 1). \u0130lgin\u00e7tir ki, bulgular\u0131 da bir\u00e7ok ara\u015ft\u0131rmac\u0131n\u0131n Belmont ilkelerinin \u00f6tesine ge\u00e7ti\u011fini (\u00e7\u0131kt\u0131lar\u0131n, ara\u015ft\u0131rmalar\u0131n neden oldu\u011fu olas\u0131 zararlardan daha a\u011f\u0131r basmas\u0131n\u0131 sa\u011flaman\u0131n \u00f6nemini vurgulayarak) g\u00f6stermektedir. Bu ilkeler: \u201c(1) kat\u0131l\u0131mc\u0131larla \u015feffafl\u0131k, (2) i\u015f arkada\u015flar\u0131 ile etik konular hakk\u0131nda m\u00fczakere etmek (3) sonu\u00e7lar\u0131n payla\u015f\u0131m\u0131nda dikkatli olmakt\u0131r (b\u00f6l.66).<\/p>\n<p style=\"text-align: justify;\">Yapay zek\u00e2 (YZ), makine \u00f6\u011frenmesi ve b\u00fcy\u00fck verilerle ilgili iyimserli\u011fi dengeleme konusundaki endi\u015fe gittik\u00e7e artmaktad\u0131r. \u00d6rne\u011fin, ABD Ba\u015fkanl\u0131\u011f\u0131 Y\u00f6netim Ofisi, faydalar\u0131 vurgulayan (Munoz, Smith ve Patil, 2016) ancak ayn\u0131 zamanda b\u00fcy\u00fck verilerin kullan\u0131m\u0131nda do\u011fabilecek potansiyel zararlarla ilgili endi\u015feleri de ele alan bir rapor yay\u0131nlad\u0131. Rapor, e\u011fer \u201cbu teknolojiler [algoritmik sistemler] dikkatli bir \u015fekilde uygulanmazlarsa, zararl\u0131 ayr\u0131mc\u0131l\u0131\u011f\u0131 s\u00fcrd\u00fcrebileceklerini, k\u00f6t\u00fcle\u015ftirebileceklerini veya maskeleyebileceklerini\u201d kabul eder (s. 5). Algoritmik ayr\u0131mc\u0131l\u0131\u011f\u0131n azalt\u0131lmas\u0131, sa\u011flam ve \u015feffaf algoritmalar\u0131n geli\u015ftirilmesi ve kullan\u0131lmas\u0131n\u0131n te\u015fvik edilmesi, algoritmik denetim, veri bilimindeki \u201cak\u0131c\u0131l\u0131\u011f\u0131n\u201d iyile\u015ftirilmesi ve veri kullan\u0131m\u0131nda uygulama kodlar\u0131n\u0131 belirlemek hususunda devlet ve \u00f6zel sekt\u00f6r\u00fcn rolleri ve ara\u015ft\u0131rmalara yat\u0131r\u0131m konusunda baz\u0131 \u00f6nerilerde bulunmaktad\u0131r.<\/p>\n<p style=\"text-align: justify;\">Benzer \u015fekilde, Birle\u015fik Krall\u0131k H\u00fckumeti k\u0131sa bir s\u00fcre \u00f6nce \u201cyasalar\u0131n d\u0131\u015f\u0131nda kalan etik konularla ilgili\u201d rehberlik sa\u011flayan bir \u201cVeri bilimi etik \u00e7er\u00e7evesi\u201d (Kabine Ofisi, 2016) yay\u0131nlam\u0131\u015ft\u0131r (s. 3). Bu \u00e7er\u00e7eve, ki\u015fisel verilerin toplanmas\u0131n\u0131n, analiz edilmesinin ve kullan\u0131lmas\u0131n\u0131n faydalar\u0131n\u0131n niteli\u011fi; izinsiz giri\u015fin kapsam\u0131 ve niteli\u011fi; verilerin kalitesi ve toplanan verilerle ilgili kararlar\u0131n otomasyonu; istenmeyen sonu\u00e7lar\u0131n riski; veri nesnelerinin toplama ve analiz i\u00e7in mutab\u0131k kal\u0131p kalmad\u0131\u011f\u0131; g\u00f6zetimin niteli\u011fi ve kapsam\u0131 ve toplanan verilerin g\u00fcvenli\u011fi gibi konular\u0131 i\u00e7erir. \u00c7er\u00e7eve ayr\u0131ca veri bilimcilere projenin yararlar\u0131n\u0131n gizlilik ve olumsuz sonu\u00e7lara kar\u015f\u0131n risklerin ne kadar fazla oldu\u011fu, riskleri en aza indirmek ve do\u011fru yorumlamay\u0131 sa\u011flamak i\u00e7in at\u0131lan ad\u0131mlar ve veri nesnelerinin \/ kamuoyunun projeye ili\u015fkin g\u00f6r\u00fc\u015flerinin ne \u00f6l\u00e7\u00fcde dikkate al\u0131nd\u0131\u011f\u0131 gibi \u201czorlu meselelere\u201d (s.6) a\u00e7\u0131kl\u0131k getirmelerini gerektiren bir \u201cGizlilik Etki De\u011ferlendirmesi\u201d \u00f6nermektedir (bk. Kabine Ofisi, 2016).<\/p>\n<p style=\"text-align: justify;\">Y\u00fcksek\u00f6\u011frenimdeki algoritmik d\u00f6n\u00fc\u015f\u00fcm ile veri ve sinir bilim aras\u0131ndaki s\u0131n\u0131rlar\u0131n bulan\u0131kla\u015fmas\u0131 ba\u011flam\u0131nda, efsane, karma\u015f\u0131kl\u0131k ve y\u00f6ntemler arac\u0131l\u0131\u011f\u0131yla yol al\u0131rken \u00f6\u011frenme analitiklerinin etik \u00e7\u0131kar\u0131mlar\u0131n\u0131 g\u00f6z \u00f6n\u00fcne almak i\u00e7in kritik bir yakla\u015f\u0131ma ihtiyac\u0131m\u0131z vard\u0131r (Ziewitz, 2016). \u00d6rne\u011fin, Williamson (2016a) \u201ce\u011fitsel veri bilimini, \u00e7ocuklar\u0131n bedensel, duygusal ve g\u00f6m\u00fcl\u00fc ya\u015famlar\u0131n\u0131n de\u011ferlendirilmesi ve y\u00f6netimine odaklanan &#8220;<span style=\"font-family: Source Serif Pro Light, serif;\"><i>biyopolitik bir strateji<\/i><\/span>&#8221; olarak g\u00f6rmektedir (s. 401, vurgu). Bu nedenle, \u201c\u00e7ocuklar hakk\u0131nda bilgi sistemleri \u00fcretmek ve onlar\u0131 m\u00fcdahalenin konusu ve nesnesi olarak tan\u0131mlamak i\u00e7in me\u015fru yetkiye sahip\u201d e\u011fitsel veri bilim adamlar\u0131n\u0131n temelini ve kapsam\u0131n\u0131 g\u00f6z \u00f6n\u00fcnde bulundurmal\u0131y\u0131z (Williamson, 2016a, s. 401). Gelecekte \u00f6\u011frenme analiti\u011fi, temelde algoritmalar ve makine \u00f6\u011frenmesine dayal\u0131 olacak ve y\u00f6nlendirilecektir; bu nedenle algoritmalar\u0131n \u201csosyal d\u00fcnyay\u0131, bilgiyi ve bilgi ile kar\u015f\u0131 kar\u015f\u0131ya gelenlerin g\u00f6r\u00fc\u015flerini nas\u0131l g\u00fc\u00e7lendirdi\u011fini, s\u00fcrd\u00fcrd\u00fc\u011f\u00fcn\u00fc ve hatta yeniden \u015fekillendirdi\u011fini\u201d g\u00f6z \u00f6n\u00fcnde bulundurmal\u0131y\u0131z (Williamson, 2016b, s. 4). G\u00fcvenilirlik, \u015feffafl\u0131k ve d\u00fczenleyici \u00e7er\u00e7eveler, <span style=\"font-family: Source Serif Pro Light, serif;\"><i>etik<\/i><\/span> \u00f6\u011frenme analiti\u011fini sa\u011flayan \u00e7er\u00e7evelerde temel unsurlar olacakt\u0131r (bk. Prinsloo, 2016; Taneja, 2016).<\/p>\n<p style=\"text-align: justify;\">Bu b\u00f6l\u00fcm, \u00f6\u011frenci verilerinin toplanmas\u0131, analizi ve kullan\u0131m\u0131yla ilgili etik sonu\u00e7lar\u0131 g\u00f6z \u00f6n\u00fcnde bulundurmadaki ilerlemenin haritas\u0131n\u0131 \u00e7\u0131kar\u0131rken, g\u00fcvenilirli\u011fi ve \u015feffafl\u0131\u011f\u0131 sa\u011flamak i\u00e7in kurumsal s\u00fcre\u00e7leri daha fazla dikkate almadan zarar verme potansiyelinin ele al\u0131nmayaca\u011f\u0131 a\u00e7\u0131kt\u0131r. Willis, Slade ve Prinsloo&#8217;nun (2016) belirtti\u011fi gibi, \u00f6\u011frenme analiti\u011fi genellikle kurumsal inceleme kurullar\u0131 (K\u0130K) taraf\u0131ndan sa\u011flanan s\u00fcre\u00e7lerin ve g\u00f6zetimin d\u0131\u015f\u0131nda kalmaktad\u0131r. Bu a\u015famada \u00f6\u011frenme analiti\u011finin etik etkilerinin kim taraf\u0131ndan ve nas\u0131l sa\u011flanaca\u011f\u0131 net de\u011fildir.<\/p>\n<h2 class=\"western\">SONU\u00c7LAR<\/h2>\n<p style=\"text-align: justify;\">2011 y\u0131l\u0131nda \u00f6\u011frenme analiti\u011finin ortaya \u00e7\u0131kmas\u0131ndan bu yana, alan ilerlemekle kalmay\u0131p ayn\u0131 zamanda \u00f6\u011frenci verilerinin toplanmas\u0131nda, analiz edilmesinde ve kullan\u0131lmas\u0131ndaki etik sonu\u00e7lar\u0131n korkular\u0131 ve ger\u00e7ekleri g\u00f6z \u00f6n\u00fcne al\u0131nd\u0131\u011f\u0131nda giderek daha fazla ilgi g\u00f6rmeye ba\u015flam\u0131\u015ft\u0131r. Bu b\u00f6l\u00fcmde, alandaki daha geni\u015f geli\u015fmelerin yan\u0131 s\u0131ra kendi d\u00fc\u015f\u00fcncemizin nas\u0131l geli\u015fti\u011fine de genel bir bak\u0131\u015f sunuyoruz. Teknolojik geli\u015fmeler ve yayg\u0131n g\u00f6zetim konusundaki endi\u015felerin artmas\u0131 ve y\u00fcksek e\u011fitimin gelece\u011finin dijital, da\u011f\u0131t\u0131lm\u0131\u015f ve veri odakl\u0131 olaca\u011f\u0131 konusunda artan bir fikir birli\u011fine kar\u015f\u0131, bu b\u00f6l\u00fcm \u00f6\u011frenme analiti\u011finin etik anlam\u0131n\u0131 \u00e7evreleyen s\u00f6ylemlerin ne kadar ileri geldi\u011fini ve gelecekte dikkate al\u0131nmas\u0131 gereken noktalar\u0131 g\u00f6stermektedir.<\/p>\n<p style=\"text-align: justify;\">Tart\u0131\u015f\u0131lan \u00f6\u011frenme analitiklerindeki etik sonu\u00e7lar\u0131n \u00e7er\u00e7evelerinin, uygulama kodlar\u0131n\u0131n ve kavramsal haritalamalar\u0131n her biri ekonomik ve etik yollarla \u00f6\u011fretmenin ve \u00f6\u011frenmenin etkinli\u011fini ve uygunlu\u011funu artt\u0131rmak i\u00e7in \u00f6\u011frenci veri proxylerini kullanma yolunda nas\u0131l ilerleyebilece\u011fimizi daha zengin bir anlay\u0131\u015fla sunar. Bu anlay\u0131\u015f\u0131n pratik uygulamas\u0131 b\u00fcy\u00fck \u00f6l\u00e7\u00fcde tamamlanmam\u0131\u015f olsa da tamam\u0131yla yerindedir.<\/p>\n<h2 class=\"western\">TE\u015eEKK\u00dcR B\u00d6L\u00dcM\u00dc<\/h2>\n<p style=\"text-align: justify;\">Edit\u00f6r ekibinden ve \u00f6zellikle bu b\u00f6l\u00fcm\u00fcn g\u00f6zden ge\u00e7irenlerden gelen yorumlara, \u00f6nemli girdilere ve verdikleri deste\u011fe te\u015fekk\u00fcr\u00fcm\u00fcz\u00fc arz ederiz.<\/p>\n<h2 class=\"western\">KAYNAK\u00c7A<\/h2>\n<p><span style=\"font-size: small;\">Cabinet Office. (2016, 19 May). Data science ethical framework. https:\/\/www.gov.uk\/government\/uploads\/ system\/uploads\/attachment_data\/file\/524298\/Data_science_ethics_framework_v1.0_for_publication__1_.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">Campbell, J. P., DeBlois, P. B., &amp; Oblinger, D. G. (2007, July\/August) Academic analytics: A new tool, a new era. <i>EDUCAUSE Review<\/i>. http:\/\/net.educause.edu\/ir\/library\/pdf\/erm0742.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">Couldry, N. (2016, September 23). The price of connection: \u201cSurveillance capitalism.\u201d [Web log post]. https:\/\/ theconversation.com\/the-price-of-connection-surveillance-capitalism-64124 <\/span><\/p>\n<p><span style=\"font-size: small;\">Dawson, S., Ga\u0161evi\u0107, D., &amp; Rogers, T. (2016). Student retention and learning analytics: A snapshot of Australian practices and a framework for advancement. Australian Government. http:\/\/he-analytics.com\/wp-content\/uploads\/SP13_3249_Dawson_Report_2016-3.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">Drachsler, H., &amp; Greller, W. (2012). The pulse of learning analytics understandings and expectations from the stakeholders. <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 120\u2013129). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Drachsler, H., &amp; Greller, W. (2016, April). Privacy and analytics: It\u2019s a DELICATE issue \u2014 a checklist for trusted learning analytics. <i>Proceedings of the 6th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201916), 25\u201329 April 2016, Edinburgh, UK (pp. 89\u201398). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Engelfriet, A., Manderveld, J., &amp; Jeunink, E. (2015). Learning analytics onder de Wet bescherming persoonsgegevens. SURFnet. https:\/\/www.surf.nl\/binaries\/content\/assets\/surf\/nl\/kennisbank\/2015\/surf_learning-analytics-onder-de-wet-wpb.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">FairTest. (n.d.). Just say no to the test. http:\/\/www.fairtest.org\/get-involved\/opting-out <\/span><\/p>\n<p><span style=\"font-size: small;\">Ga\u0161evi\u0107, D., Dawson, S., &amp; Jovanovi\u0107, J. (2016). Ethics and privacy as enablers of learning analytics. <i>Journal of Learning Analytics, 3<\/i>(1), 1\u20134. http:\/\/dx.doi.org\/10.18608\/jla.2016.31.1 <\/span><\/p>\n<p><span style=\"font-size: small;\">Munoz, C., Smith, M., &amp; Patil, D. J. (2016). Big data: A report on algorithmic systems, opportunity, and civil rights. Executive Office of the President, USA. https:\/\/www.whitehouse.gov\/sites\/default\/files\/microsites\/ostp\/2016_0504_data_discrimination.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">NMC (New Media Consortium). (2011). The NMC Horizon Report. http:\/\/www.educause.edu\/Resources\/2011HorizonReport\/223122 <\/span><\/p>\n<p><span style=\"font-size: small;\">NMC (New Media Consortium). (2016). The NMC Horizon Report. http:\/\/www.nmc.org\/publication\/ nmc-horizon-report-2016-higher-education-edition\/ <\/span><\/p>\n<p><span style=\"font-size: small;\">O\u2019Brien, A. (2010, September 29). Predictably irrational: A conversation with best-selling author Dan Ariely. [Web log post]. http:\/\/www.learningfirst.org\/predictably-irrational-conversation-best-selling-author-dan-ariely <\/span><\/p>\n<p><span style=\"font-size: small;\">Open University. (2014). Policy on ethical use of student data for learning analytics. http:\/\/www.open.ac.uk\/students\/charter\/sites\/www.open.ac.uk.students.charter\/files\/files\/ecms\/web-content\/ethical-use-of-student-data-policy.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P. (2016, September 22). Fleeing from Frankenstein and meeting Kafka on the way: Algorithmic decision-making in higher education. Presentation at NUI, Galway. http:\/\/www.slideshare.net\/prinsp\/feeling-from-frankenstein-and-meeting-kafka-on-the-way-algorithmic-decisionmaking-in-higher-education <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., Slade, S., &amp; Galpin, F. (2012) Learning analytics: Challenges, paradoxes and opportunities for mega open distance learning institutions. <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 130\u2013133). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2013). An evaluation of policy frameworks for addressing ethical considerations in learning analytics. <i>Proceedings of the 3rd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201913), 8\u201312 April 2013, Leuven, Belgium (pp. 240\u2013244). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2014a). Educational triage in higher online education: Walking a moral tightrope. <i>International Review of Research in Open Distributed Learning <\/i>(IRRODL)<i>, 14<\/i>(4), 306\u2013331. http:\/\/www.irrodl.org\/ index.php \/ irrodl \/ article \/ view \/ 1881 <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2014b). Student privacy and institutional accountability in an age of surveillance. In M. E. Menon, D. G. Terkla, &amp; P. Gibbs (Eds.), <i>Using data to improve higher education: Research, policy and practice <\/i>(pp. 197\u2013214). Global Perspectives on Higher Education (29). Rotterdam: Sense Publishers. <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2015). Student privacy self-management: Implications for learning analytics. <i>Proceedings of the 5th International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201915), 16\u201320 March 2015, Poughkeepsie, NY, USA (pp. 83\u201392). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2016a). Student vulnerability, agency, and learning analytics: An exploration. <i>Journal of Learning Analytics, 3<\/i>(1), 159\u2013182. <\/span><\/p>\n<p><span style=\"font-size: small;\">Prinsloo, P., &amp; Slade, S. (2016b). Here be dragons: Mapping student responsibility in learning analytics. In M. Anderson &amp; C. Gavan (Eds.), <i>Developing effective educational experiences through learning analytics <\/i>(pp. 170\u2013188). Hershey, PA: IGI Global. <\/span><\/p>\n<p><span style=\"font-size: small;\">Ruggiero, D. (2016, May 18). What metrics don\u2019t tell us about the way students learn. <i>The Conversation<\/i>. http:\/\/ theconversation.com\/what-metrics-dont-tell-us-about-the-way-students-learn-59271 <\/span><\/p>\n<p><span style=\"font-size: small;\">Sclater, N. (2015, March 3). Effective learning analytics. A taxonomy of ethical, legal and logistical issues in learning analytics v1.0. JISC. https:\/\/analytics.jiscinvolve.org\/wp\/2015\/03\/03\/a-taxonomy-of-ethical-legal-and-logistical-issues-of-learning-analytics-v1-0\/ <\/span><\/p>\n<p><span style=\"font-size: small;\">Sclater, N., Peasgood, A., &amp; Mullan, J. (2016). Learning analytics in higher education. A review of UK and international practice. JISC. https:\/\/www.jisc.ac.uk\/reports\/learning-analytics-in-higher-education <\/span><\/p>\n<p><span style=\"font-size: small;\">Shacklock, X. (2016). From bricks to clicks: The potential of data and analytics in higher education. Higher Education Commission. http:\/\/www.policyconnect.org.uk\/hec\/sites\/site_hec\/files\/report\/419\/fieldreportdownload\/frombrickstoclicks-hecreportforweb.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">Siemens, G. (2016, April 28). Reflecting on learning analytics and SoLAR. [Web log post]. http:\/\/www.elearnspace.org\/blog\/2016\/04\/28\/reflecting-on-learning-analytics-and-solar\/<\/span><\/p>\n<p><span style=\"font-size: small;\">Siemens, G., &amp; Baker, R. (2012, April). Learning analytics and educational data mining: Towards communication and collaboration. <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 252\u2013254). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Slade, S., &amp; Galpin, F. (2012) Learning analytics and higher education: Ethical perspectives (workshop). <i>Proceedings of the 2nd International Conference on Learning Analytics and Knowledge <\/i>(LAK\u201912), 29 April\u20132 May 2012, Vancouver, BC, Canada (pp. 16\u201317). New York: ACM. <\/span><\/p>\n<p><span style=\"font-size: small;\">Slade, S., &amp; Prinsloo, P. (2013). Learning analytics: Ethical issues and dilemmas. <i>American Behavioral Scientist, 57<\/i>(1), 1509\u20131528. <\/span><\/p>\n<p><span style=\"font-size: small;\">Slade, S., &amp; Prinsloo, P. (2014). Student perspectives on the use of their data: Between intrusion, surveillance and care. <i>Proceedings of the European Distance and E-Learning Network 2014 Research Workshop <\/i>(EDEN 2014), 27\u201328 October 2014, Oxford, UK (pp. 291\u2013300). <\/span><\/p>\n<p><span style=\"font-size: small;\">Smith, G.J. (2016). Surveillance, data and embodiment: On the work of being watched. <i>Body &amp; Society<\/i>, 1\u201332. doi:10.1177\/1357034X15623622 <\/span><\/p>\n<p><span style=\"font-size: small;\">Strauss, V. (2016a, January 28). U.S. Education Department threatens to sanction states over test opt-outs. The Washington Post. https:\/\/www.washingtonpost.com\/news\/answer-sheet\/wp\/2016\/01\/28\/u-s-education-department-threatens-to-sanction-states-over-test-opt-outs\/ <\/span><\/p>\n<p><span style=\"font-size: small;\">Strauss, V. (2016b, May 9). \u201cBig data\u201d was supposed to fix education. It didn\u2019t. It\u2019s time for \u201csmall data.\u201d The Washington Post. https:\/\/www.washingtonpost.com\/news\/answer-sheet\/wp\/2016\/05\/09\/big-data-was-supposed-to-fix-education-it-didnt-its-time-for-small-data\/ <\/span><\/p>\n<p><span style=\"font-size: small;\">Taneja, H. (2016, September 8). The need for algorithmic accountability. <i>TechCrunch<\/i>. https:\/\/techcrunch.com\/2016\/09\/08\/the-need-for-algorithmic-accountability\/ <\/span><\/p>\n<p><span style=\"font-size: small;\">van Barneveld, A., Arnold, K., &amp; Campbell, J. (2012). Analytics in higher education: Establishing a common language. <i>EDUCAUSE Learning Initiative, 1<\/i>, 1\u201311. <\/span><\/p>\n<p><span style=\"font-size: small;\">Vitak, J., Shilton, K., &amp; Ashktorab, Z. (2016). Beyond the Belmont principles: Ethical challenges, practices, and beliefs in the online data research community. <i>Proceedings of the 19th ACM Conference on Computer Supported Cooperative Work &amp; Social Computing <\/i>(CSCW\u201916), 27 February\u20132 March 2016, San Francisco, CA, USA. New York: ACM. https:\/\/terpconnect.umd.edu\/~kshilton\/pdf\/VitaketalCSCWpreprint.pdf <\/span><\/p>\n<p><span style=\"font-size: small;\">Watters, A. (2016, May 7). Identity, power, and education\u2019s algorithms. [Web log post]. http:\/\/hackeducation.com\/2016\/05\/07\/identity-power-algorithms <\/span><\/p>\n<p><span style=\"font-size: small;\">Welsh, S., &amp; McKinney, S. (2015). Clearing the fog: A learning analytics code of practice. In T. Reiners et al. (Eds.), Globally connected, digitally enabled. <i>Proceedings of the 32nd Annual Conference of the Australasian Society for Computers in Learning in Tertiary Education <\/i>(ASCILITE 2015), 29 November\u20132 December 2015, Perth, Western Australia (pp. 588\u2013592). http:\/\/research.moodle.net\/80\/ <\/span><\/p>\n<p><span style=\"font-size: small;\">Williamson, B. (2016a). Coding the biodigital child: The biopolitics and pedagogic strategies of educational data science. <i>Pedagogy, Culture &amp; Society, 24<\/i>(3), 401\u2013416. doi:10.1080\/14681366.2016.1175499 <\/span><\/p>\n<p><span style=\"font-size: small;\">Williamson, B. (2016b). Computing brains: Learning algorithms and neurocomputation in the smart city. <i>Information, Communication &amp; Society, 20<\/i>(1), 81\u201399. doi:10.1080\/1369118X.2016.1181194 <\/span><\/p>\n<p><span style=\"font-size: small;\">Willis, J., Slade, S., &amp; Prinsloo, P. (2016). Ethical oversight of student data in learning analytics: A typology derived from a cross-continental, cross-institutional perspective. Educational Technology Research and Development. doi:10.1007\/s11423-016-9463-4 <\/span><\/p>\n<p><span style=\"font-size: small;\">Zhang, S. (2016, May 20). Scientists are just as confused about the ethics of big data research as you. Wired. http:\/\/www.wired.com\/2016\/05\/scientists-just-confused-ethics-big-data-research\/ <\/span><\/p>\n<p><a name=\"_Hlk25065891\" id=\"_Hlk25065891\"><\/a> <span style=\"font-size: small;\">Ziewitz, M. (2016). Governing algorithms: Myth, mess, and methods. <i>Science, Technology &amp; Human Values, 41<\/i>(1), 3\u201316.<\/span><\/p>\n","protected":false},"author":1,"menu_order":4,"template":"","meta":{"pb_show_title":"on","pb_short_title":"","pb_subtitle":"","pb_authors":[],"pb_section_license":""},"chapter-type":[48],"contributor":[],"license":[],"class_list":["post-45","chapter","type-chapter","status-publish","hentry","chapter-type-numberless"],"part":31,"_links":{"self":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/45","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters"}],"about":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/types\/chapter"}],"author":[{"embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/users\/1"}],"version-history":[{"count":0,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/45\/revisions"}],"part":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/parts\/31"}],"metadata":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapters\/45\/metadata\/"}],"wp:attachment":[{"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/media?parent=45"}],"wp:term":[{"taxonomy":"chapter-type","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/pressbooks\/v2\/chapter-type?post=45"},{"taxonomy":"contributor","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/contributor?post=45"},{"taxonomy":"license","embeddable":true,"href":"https:\/\/acikkitap.com.tr\/oaek\/wp-json\/wp\/v2\/license?post=45"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}