PISA Analizi Uzerine Notlar ve R paketi intsvy nin tanitimi
Not: Bu sayfalari olusturmak için kullandigimiz R Markdown bütün Türkçe karakterleri desteklemedigi için yazimda sikintilar olusmakta. O yüzden özür dileriz.
Hatirlatmalar
Bu bölümde PISA analiz edilirken dikkat edilmesi gereken birkaç noktaya dikkat çekip, PISA veri setlerini R programinda dogru bir sekilde analiz etmek için gelistirilmis olan intsvy paketinin kisa bir tanitimini yapacagiz.
Öncelikle hatirlanmasi gereken nokta PISA örnekleminin nasil seçildigidir. PISA’da önce okullar, sonrasinda da okul içinde ögrenciler seçilmektedir. Okullar seçilirken genelde çesitli nedenlerden dolayi tabakali örnekleme denen bir yöntem kullanilmaktadir. Bu durumda bazi tur okullarin diger tür okullara göre PISA çalismasinda daha fazla veya daha az temsil edilmesi mümkün olabilmektedir. Ayni zamanda bir okulun seçilme olasiligi okulun 15 yas ögrenci sayisi ile dogru orantilidir. Yani 15 yas ögrenci sayisi çok olan okullarin çalismaya katilmasi daha olasi iken az olan okullarin seçilmesi daha az olasidir. Okullarda ögrencilerin yaslari ile ilgili bilgilerin dogru veya güncel olmamasi durumunda örneklemde hatalar olusabilmektedir. Son olarak da, seçilen okulda yine rastgele seçilen her ögrenci PISA çalismasina katilmayi kabul etmeyebilmektedir. Mesela, bir okulda PISA’ya katilmasi için seçilen ögrencilerden kiz ögrenciler çalismaya katilmaya daha istekli olup erkek ögrenciler katilmak istemeyebilirler. Dolayisi ile pratikte farkli sebeblerden dolayi isteyerek veya istemeyerek seçilen örneklem ülkedeki 15 yas popülasyonunu tam olarak temsil etmeyebilir. Bu yüzden bütün bu degisik faktörler hesaba katilarak herbir okul ve ögrenci için bir örneklem agirlik katsayisi (sample weights) hesaplanmaktadir. Yapilan analizlerde kestirilen parametrelerin yanli olmamasi, ve popülasyonu dogru temsil etmesi için bu agirlik katsayilarinin kullanilmasi gerekmektedir.
Bir baska nokta da ogrenciler direkt olarak rastgele seçilmediginden ve ayni okulun içindeki ögrenciler bir çok demografik degisken bakimindan birbirlerine benzer olabileceginden bu durum istatiksel olarak dikkate alinmalidir. Aksi takdirde kestirilen parametrelerin standard hatalarin oldugundan küçük hesaplanmakta ve bu durumda yapilan çikarimlarda hata yapma orani artmaktadir. Bunun için yine PISA veri setlerinde farkli agirlik katsayilari hesaplanmaktadir (replicate weights), ve analizler yine bu katsayilari dikkate alarak yapilmalidir.
Son olarak, PISA da bir ögrencinin bir konuda tek bir basari puani hesaplanmamaktadir. Mesela 2015 yilinda her ögrenci için matematik, fen, ve okuma alaninda 10 farkli basari puani bulunmaktadiar (plausible values). Herhangi bir analiz herbir puan üzerinden 10 defa yapilmali, ve sonuçlar daha sonra özel bir metodla biraraya getirilip parametreler ve standard hatalari kestirilmelidir. Bunun amaci hem örneklem hatasinin hem de test sorularindan kaynaklanan ölçme hatalarinin beraber dikkate alinmasini saglamaktir.
Hiç unutulmamasi gereken bir nokta ise yapilan bütün analizlerde PISA ve benzeri veri setlerinden hiçbir sekilde dogrudan “causal” yani nedensellik içeren çikarimlarda bulunulmamasidir. Bu veri setlerinden nedensellik içeren çikarimlarda bulunabilmek için çok komplike yöntemler kullanilmasi gerekir. Mesela, ilgilenenler su makalede sinif büyüklügünün basariya etkisinin nasil analiz edildigini okuyabilir. Kaldiki bu tür yöntemlerin de metodolojik olarak hala eksiklikleri bulunmaktadir.
- Li, W., & Konstantopoulos, S. (2016). Class Size Effects on Fourth Grade Mathematics Achievement: Evidence From TIMSS 2011. Journal of Research on Educational Effectiveness, 9(4)), 503-530.
R paketi intsvy’nin kullanimi
PISA ve benzeri veri setlerinin analizinde dikkat edilmesi gereken noktalari çok basit bir sekilde hatirlattiktan sonra bunlarin nasil yapabilecegi ustunde durabiliriz. Öncelikle söylemek gerekir ki PISA kendi websitesinde Data Explorer veya PISA Education GPS gibi araçlarla analizler yapilmasina imkan vermektedir. Ancak bu tür analizler ne yazik ki betimsel istatistiklerden öteye gitmemektedir. PISA gibi çok zengin bir veri setini sadece betimsel istatistikler üzerinden tartismanin elmasi bit pazarinda satmaya çalismakdan farki yoktur (örnek olarak sadece ortalamalar üzerinde TR’nin 2015 yilindaki düsüsü üzerinde yapilan tartismalar gibi). Ayni zamanda bu araçlarla genelde belli bir senenin verileri analiz edilmekte ve seneler arasindaki trend analizleri gözden kaçmaktadir. PISA’nin yayinlanan raporlarinda ise belli basli degiskenler yer almakta ve onlarca diger degisken kullanilmamakta veya tartisilmamaktadir. Bu yüzden ham veri setlerinin SPSS, SAS, veya R gibi yazilimlarla detayli bir sekilde analizlerin yapilmasi önemlidir. SPSS ve SAS lisansli yazilimlar olup kullanmak için para ödenmesi gerekmektedir. Birçok egitimci ve arastirmacinin bu paralari ödeme lüksü yoktur. Ayni zamanda, SPSS ve SAS’da bu verilerin analizi yine çok basit degildir, ve kullanicinin “syntax” diline hakim olmasi gerekmektedir. Bunlarin alternatifi olarak R programi bedavadir. Hiç bilmeyenlerin ilk basta zorlanmasi normal olmakla birlikte kisa süre içinde kullanicilar kendilerini gelistirip yetkin hale gelebilirler. R ayni zamanda SPSS ve SAS’a oranla grafik kapasitesi olarak daha esnek ve güçlü bir programdir. Dahasi, dünyanin dört bir yaninda alaninda uzman kisiler kullanilmasi için paketler yazip bunlari yine bedava erisilebilir hale getirmektedir.
Iste burda tanitacagimiz intsvy yazilimi da iki arastirmacinin PISA ve benzeri uluslararasi veri setlerini analiz etmek için gelistirdigi bir paket program. Detayli içerigine su linkten ulasabilirsiniz https://cran.r-project.org/web/packages/intsvy/intsvy.pdf. Simdi basit bir örnekle bu paketi nasil kullanabilecegimizi görelim.
- önceki yazida tarif edildigi sekilde RStudio programini açip PISA veri setlerini yüklemeniz gerekiyor.
- Daha sonra menüden File –> New File –> R Script seçeneklerini tikladigimizda sol üst tarafta bos bir panel açildigini göreceksiniz.
- Sonrasinda yapmaniz gereken intsvy paketini bilgisayariniza yüklemek. R studio bilgisayara kuruldugunda bilgisayarinizda bir yerde “library” adinda bir dosya olusturur. Bunun içinde kendiliginden gelen ve basit analizleri yapmak için kullanilabilecek paketler olur. Ancak diger paketleri kullanmak istediginizde bu paketleri bilgisayariniza indirmeniz gerekmekte. Sol üst kösedeki interaktif olmayan panele asagidaki kodu yazip, hepsini seçtikten sonra panelin sag üst kösesindeki Run dügmesine basarsaniz sag üst taraftaki panelde kodun çalistigini ve simdilik gerekli olan paketleri yükledigini görebilirsiniz.
install.packages("devtools")
library(devtools)
install_github("eldafani/intsvy",dependencies=TRUE)
library(intsvy)
- Simdi örnek olmasi açisindan 2015 yilindaki Türkiye örnekleminde “ESCS” degiskenini inceleyip, ortalamasini hesaplayalim. Burda ortalamasini hesaplarken örnek olmasi açisindan iki turlu hesaplayacagiz. Veri setindeki önceden bahsettigimiz agirlik katsayilarini hesaba katmayarak ve hesaba katarak.
# Ilk olarak Turkiye verisini suzup ayiralim
pisa_2015_stu_TUR <- subset(pisa_2015_stu,CNT=='Turkey')
dim(pisa_2015_stu_TUR)## [1] 5895 921 # Suzdugumuz veride 5895 ogrenci ve 921 degisken var.
# Bu suzdugumuz verideki ESCS ortalamasinin ve standard
# hatasinin normal olarak hesaplanmasi
mean(pisa_2015_stu_TUR$ESCS,na.rm=TRUE)## [1] -1.447651 sd(pisa_2015_stu_TUR$ESCS,na.rm=TRUE)/sqrt(5895)## [1] 0.01518541# Bu suzdugumuz verideki yas ortalamasinin agirlik
# katsayilarinin hesaba katilarak intsvy paketindeki ozel bir fonksiyonla hesaplanmasi
pisa2015.mean(variable="ESCS",data=pisa_2015_stu_TUR)## Freq Mean s.e. SD s.e
## 1 5859 -1.43 0.05 1.17 0.02 # Burda kullanilan "pisa2015.mean" intsvy paketinde
# bulunan ve bu amacla hazirlanmis ozel bir fonksiyon
Görüldügü gibi kestirilen ortalama her iki yontemde çok farkli olmamakla beraber agirlik katsayilari kullanildiginda standard hata daha yüksektir.
- Simdi de meseleyi bir adim daha ileriye götürüp, Türkiye’den sinava katilan ögrencilerin ESCS degiskeni üzerinden seneler açisinda nasil degistigine bakalim. Bu degisken PISA’nin her ögrenci için hesapladigi ekonomik, sosyal, ve kültürel statü indeksidir. PISA bu degiskeni hesaplarken ögrencilerin ebeveynlerinin egitim durumlari, mesleki durumlari, ailenin varlik durumu, evde bulunan egitimsel kaynaklar ve diger esyalar gibi bilgileri kullaniyor. Standard hale getirilmis bir skor oldugundan genel olarak -3 ve 3 arasinda degerlere sahip. Bu degisken hesaplanirken OECD ülkelerinin ortalamasi 0 olacak sekilde ve standard sapmasi da 1 olacak sekilde hesaplaniyor. OECD bu degiskenin hesaplanmasini yillar arasinda degistirdiginden 2015 yilinda kullandigi yöntemi kullanarak geçmise dönük bu degiskeni her yil için tekrar hesaplayip web sitesine koydu. Böylece seneler arasinda ESCS degerleri ayni ölçekte karsilastirilabilir sekilde sunuldu. Simdi asagidaki kodu kullanarak Türkiyenin seneler arasinda ESCS degiskenindeki degisimi inceleyelim.
# OECD nin yayinladigi gecmis senelere ait yeni ESCS degerleri ayri veri setlerinde
# oldugundan, oncelikle yapmamiz gereken bunlari eski senelerin veri setleri ile entegre etmek.
# 2003, 2006, 2009, ve 2012 yillarina ait yeni hesaplanmis ESCS degerlerini bu senelere
# ait orjinal veri setleri ile birlestirelim.
# Dikkat etmemiz gereken nokta yeni veri setinde ki ID lerinin
# isimlerinde ufak bi degisiklik var. Bunu orjinal verideki hali ile ayni hale
# getirmemiz lazim.
colnames(escs_2003)[1:3] <- c("CNT","SCHOOLID","STIDSTD")
# Simdi iki veri setini birlestirebiliriz.
pisa_2003_stu <- merge(pisa_2003_stu,escs_2003,by=c("CNT","SCHOOLID","STIDSTD") ,all=TRUE)
# Simdi bunu ayni sekilde 2006, 2009, ve 2012 seneleri icin tekrar edelim.
# Yalniz burda bir problemimiz var. 2003 senesinde CNT 3 karekterli bir kod iken,
# 2006, 2009, 2012 senelerine ait CNT ulkenin tam isminden olusuyor. Bu da veri setlerini
# birlestirirken problem olusturuyor. O yuzden once yeni escs veri setlerindeki
# uc karakterli ulke kodlarini ulkelerin isimleri ile degistirmemiz lazim.
# 2006
levels(escs_2006$cnt)## [1] "ARG" "AUS" "AUT" "BEL" "BGR" "BRA" "CAN" "CHE" "CHL" "COL" "CZE"
## [12] "DEU" "DNK" "ESP" "EST" "FIN" "FRA" "GBR" "GRC" "HKG" "HRV" "HUN"
## [23] "IDN" "IRL" "ISL" "ISR" "ITA" "JOR" "JPN" "KOR" "LTU" "LUX" "LVA"
## [34] "MAC" "MEX" "MNE" "NLD" "NOR" "NZL" "POL" "PRT" "QAT" "ROU" "RUS"
## [45] "SVK" "SVN" "SWE" "TAP" "THA" "TUN" "TUR" "URY" "USA" levels(escs_2006$cnt) <- c("Argentina","Australia","Austria","Belgium","Bulgaria"
,"Brazil","Canada","Switzerland","Chile","Colombia",
"Czech Republic","Germany","Denmark","Spain","Estonia",
"Finland","France","United Kingdom","Greece",
"Hong Kong-China","Croatia","Hungary","Indonesia","Ireland",
"Iceland","Israel", "Italy","Jordan","Japan","Korea",
"Lithuania","Luxembourg","Latvia","Macao-China",
"Mexico","Montenegro","Netherlands","Norway","New Zealand",
"Poland","Portugal","Qatar","Romania","Russian Federation",
"Slovak Republic","Slovenia ","Sweden",
"Chinese Taipei","Thailand","Tunisia","Turkey","Uruguay",
"United States")
levels(escs_2006$cnt)## [1] "Argentina" "Australia" "Austria"
## [4] "Belgium" "Bulgaria" "Brazil"
## [7] "Canada" "Switzerland" "Chile"
## [10] "Colombia" "Czech Republic" "Germany"
## [13] "Denmark" "Spain" "Estonia"
## [16] "Finland" "France" "United Kingdom"
## [19] "Greece" "Hong Kong-China" "Croatia"
## [22] "Hungary" "Indonesia" "Ireland"
## [25] "Iceland" "Israel" "Italy"
## [28] "Jordan" "Japan" "Korea"
## [31] "Lithuania" "Luxembourg" "Latvia"
## [34] "Macao-China" "Mexico" "Montenegro"
## [37] "Netherlands" "Norway" "New Zealand"
## [40] "Poland" "Portugal" "Qatar"
## [43] "Romania" "Russian Federation" "Slovak Republic"
## [46] "Slovenia " "Sweden" "Chinese Taipei"
## [49] "Thailand" "Tunisia" "Turkey"
## [52] "Uruguay" "United States" colnames(escs_2006)[1:3] <- c("CNT","SCHOOLID","STIDSTD")
pisa_2006_stu <- merge(pisa_2006_stu,escs_2006,by=c("CNT","SCHOOLID","STIDSTD") ,all=TRUE)
# 2009
levels(escs_2009$cnt)## [1] "ALB" "ARE" "ARG" "AUS" "AUT" "BEL" "BGR" "BRA" "CAN" "CHE" "CHL"
## [12] "COL" "CRI" "CZE" "DEU" "DNK" "ESP" "EST" "FIN" "FRA" "GBR" "GEO"
## [23] "GRC" "HKG" "HRV" "HUN" "IDN" "IRL" "ISL" "ISR" "ITA" "JOR" "JPN"
## [34] "KAZ" "KOR" "LTU" "LUX" "LVA" "MAC" "MDA" "MEX" "MLT" "MNE" "MYS"
## [45] "NLD" "NOR" "NZL" "PER" "POL" "PRT" "QAT" "ROU" "RUS" "SGP" "SVK"
## [56] "SVN" "SWE" "TAP" "THA" "TTO" "TUN" "TUR" "URY" "USA" levels(escs_2009$cnt) <- c("Albania","United Arab Emirates","Argentina","Australia",
"Austria","Belgium","Bulgaria","Brazil","Canada",
"Switzerland","Chile","Colombia","Costa Rica",
"Czech Republic","Germany","Denmark",
"Spain","Estonia","Finland","France","United Kingdom",
"Georgia","Greece","Hong Kong-China","Croatia","Hungary",
"Indonesia","Ireland","Iceland","Israel", "Italy","Jordan",
"Japan","Kazakhstan","Korea","Lithuania","Luxembourg",
"Latvia","Macao-China","Republic of Moldova","Mexico",
"Malta","Montenegro","Malaysia","Netherlands","Norway",
"New Zealand","Peru","Poland","Portugal","Qatar","Romania",
"Russian Federation","Singapore","Slovak Republic",
"Slovenia","Sweden","Chinese Taipei","Thailand",
"Trinidad and Tobago","Tunisia","Turkey","Uruguay",
"United States")
levels(escs_2009$cnt)## [1] "Albania" "United Arab Emirates" "Argentina"
## [4] "Australia" "Austria" "Belgium"
## [7] "Bulgaria" "Brazil" "Canada"
## [10] "Switzerland" "Chile" "Colombia"
## [13] "Costa Rica" "Czech Republic" "Germany"
## [16] "Denmark" "Spain" "Estonia"
## [19] "Finland" "France" "United Kingdom"
## [22] "Georgia" "Greece" "Hong Kong-China"
## [25] "Croatia" "Hungary" "Indonesia"
## [28] "Ireland" "Iceland" "Israel"
## [31] "Italy" "Jordan" "Japan"
## [34] "Kazakhstan" "Korea" "Lithuania"
## [37] "Luxembourg" "Latvia" "Macao-China"
## [40] "Republic of Moldova" "Mexico" "Malta"
## [43] "Montenegro" "Malaysia" "Netherlands"
## [46] "Norway" "New Zealand" "Peru"
## [49] "Poland" "Portugal" "Qatar"
## [52] "Romania" "Russian Federation" "Singapore"
## [55] "Slovak Republic" "Slovenia" "Sweden"
## [58] "Chinese Taipei" "Thailand" "Trinidad and Tobago"
## [61] "Tunisia" "Turkey" "Uruguay"
## [64] "United States" colnames(escs_2009)[1:3] <- c("CNT","SCHOOLID","StIDStd")
pisa_2009_stu <- merge(pisa_2009_stu,escs_2009,by=c("CNT","SCHOOLID","StIDStd") ,all=TRUE)
# 2012
levels(escs_2012$cnt)## [1] "ALB" "ARE" "ARG" "AUS" "AUT" "BEL" "BGR" "BRA" "CAN" "CHE" "CHL"
## [12] "COL" "CRI" "CZE" "DEU" "DNK" "ESP" "EST" "FIN" "FRA" "GBR" "GRC"
## [23] "HKG" "HRV" "HUN" "IDN" "IRL" "ISL" "ISR" "ITA" "JOR" "JPN" "KAZ"
## [34] "KOR" "LTU" "LUX" "LVA" "MAC" "MEX" "MNE" "MYS" "NLD" "NOR" "NZL"
## [45] "PER" "POL" "PRT" "QAT" "QUC" "ROU" "RUS" "SGP" "SVK" "SVN" "SWE"
## [56] "TAP" "THA" "TUN" "TUR" "URY" "USA" "VNM" levels(escs_2012$cnt) <- c("Albania","United Arab Emirates","Argentina","Australia",
"Austria","Belgium",
"Bulgaria","Brazil","Canada",
"Switzerland","Chile","Colombia","Costa Rica",
"Czech Republic","Germany","Denmark",
"Spain","Estonia","Finland","France","United Kingdom",
"Greece","Hong Kong-China",
"Croatia","Hungary","Indonesia","Ireland","Iceland",
"Israel","Italy","Jordan","Japan","Kazakhstan","Korea",
"Lithuania","Luxembourg","Latvia","Macao-China",
"Mexico","Montenegro","Malaysia","Netherlands","Norway",
"New Zealand","Peru","Poland","Portugal","Qatar",
"Shanghai-China","Romania","Russian Federation","Singapore",
"Slovak Republic","Slovenia","Sweden","Chinese Taipei",
"Thailand","Tunisia","Turkey",
"Uruguay","United States of America","Viet Nam")
levels(escs_2012$cnt)## [1] "Albania" "United Arab Emirates"
## [3] "Argentina" "Australia"
## [5] "Austria" "Belgium"
## [7] "Bulgaria" "Brazil"
## [9] "Canada" "Switzerland"
## [11] "Chile" "Colombia"
## [13] "Costa Rica" "Czech Republic"
## [15] "Germany" "Denmark"
## [17] "Spain" "Estonia"
## [19] "Finland" "France"
## [21] "United Kingdom" "Greece"
## [23] "Hong Kong-China" "Croatia"
## [25] "Hungary" "Indonesia"
## [27] "Ireland" "Iceland"
## [29] "Israel" "Italy"
## [31] "Jordan" "Japan"
## [33] "Kazakhstan" "Korea"
## [35] "Lithuania" "Luxembourg"
## [37] "Latvia" "Macao-China"
## [39] "Mexico" "Montenegro"
## [41] "Malaysia" "Netherlands"
## [43] "Norway" "New Zealand"
## [45] "Peru" "Poland"
## [47] "Portugal" "Qatar"
## [49] "Shanghai-China" "Romania"
## [51] "Russian Federation" "Singapore"
## [53] "Slovak Republic" "Slovenia"
## [55] "Sweden" "Chinese Taipei"
## [57] "Thailand" "Tunisia"
## [59] "Turkey" "Uruguay"
## [61] "United States of America" "Viet Nam" colnames(escs_2012)[1:3] <- c("CNT","SCHOOLID","StIDStd")
pisa_2012_stu <- merge(pisa_2012_stu,escs_2012,by=c("CNT","SCHOOLID","StIDStd") ,all=TRUE)
# Simdide 2003, 2006, 2009, 2012 yillarinda 2015 icin yaptigimiz gibi Turkiye verisini suzelim.
pisa_2012_stu_TUR <- subset(pisa_2012_stu,CNT=="Turkey")
pisa_2009_stu_TUR <- subset(pisa_2009_stu,CNT=='Turkey')
pisa_2006_stu_TUR <- subset(pisa_2006_stu,CNT=='Turkey')
pisa_2003_stu_TUR <- subset(pisa_2003_stu,CNT=='TUR')
# Simdi 2003, 2006, 2009, 2012, 2015 yillarinda Turkiyenin ortalama ESCS degerini bulalim.
escs.tur <- rbind(pisa.mean(variable="escs_trend",data=pisa_2003_stu_TUR),
pisa.mean(variable="escs_trend",data=pisa_2006_stu_TUR),
pisa.mean(variable="escs_trend",data=pisa_2009_stu_TUR),
pisa.mean(variable="escs_trend",data=pisa_2012_stu_TUR),
pisa2015.mean(variable="ESCS",data=pisa_2015_stu_TUR))
rownames(escs.tur) <- c("2003","2006","2009","2012","2015")
escs.tur## Freq Mean s.e. SD s.e
## 2003 4845 -1.71 0.08 1.32 0.04
## 2006 4934 -1.67 0.05 1.16 0.03
## 2009 4965 -1.54 0.04 1.19 0.02
## 2012 4806 -1.62 0.04 1.19 0.02
## 2015 5859 -1.43 0.05 1.17 0.02Görüldügü gibi 2003 yilindan itibaren Türkiye’nin okullardaki 15 yas popUlasyonunun ESCS degeri ile ölçülen sosyoekonomik kültürel statüsünde artma egilimi olmakla birlikte belirgin bir degisme yok.
- Simdi biraz daha ilerleyip, ayni dönel=mlerde OECD ve Avrupa Birligi ülkelerinde bunun nasil degistigini inceleyelim.
# Bu veri setlerinde her ulke icin OECD uyesi olup olmadigini gosteren bir degisken var.
# Bunu kullanarak her bir sene icin OECD uilkelerine ait veriyi suzelim.
pisa_2015_stu_OECD <- subset(pisa_2015_stu,OECD=="Yes")
pisa_2012_stu_OECD <- subset(pisa_2012_stu,OECD=="OECD")
pisa_2009_stu_OECD <- subset(pisa_2009_stu,OECD=="OECD")
pisa_2006_stu_OECD <- subset(pisa_2006_stu,OECD=="OECD")
pisa_2003_stu_OECD <- subset(pisa_2003_stu,OECD=="OECD country")
# Simdi 2003, 2006, 2009, 2012, 2015 yillarinda her bir OECD ulkesinin ortalama ESCS
# degerini bulalim.
# Once her ulkenin ortalama ESCS degerini hesaplayip, sonra bunlarin ortalamasini alacagiz
escs.oecd03 <- pisa.mean(variable="escs_trend",data=pisa_2003_stu_OECD,by="CNT")
escs.oecd06 <- pisa.mean(variable="escs_trend",data=pisa_2006_stu_OECD,by="CNT")
escs.oecd09 <- pisa.mean(variable="escs_trend",data=pisa_2009_stu_OECD,by="CNT")
escs.oecd12 <- pisa.mean(variable="escs_trend",data=pisa_2012_stu_OECD,by="CNT")
escs.oecd15 <- pisa2015.mean(variable="ESCS",data=pisa_2015_stu_OECD,by="CNT")
escs.oecd03## CNT Freq Mean s.e. SD s.e
## 1 AUS 12387 -0.02 0.02 0.98 0.01
## 2 AUT 4548 -0.29 0.03 0.97 0.01
## 3 BEL 8551 -0.12 0.03 1.14 0.01
## 4 CAN 26590 0.20 0.02 0.95 0.01
## 5 CHE 8342 -0.27 0.03 1.08 0.02
## 6 CZE 6155 -0.19 0.03 0.95 0.01
## 7 DEU 4399 -0.08 0.03 1.17 0.02
## 8 DNK 4178 0.07 0.03 1.02 0.02
## 9 ESP 10686 -0.42 0.06 1.51 0.02
## 10 FIN 5778 -0.03 0.02 1.01 0.01
## 11 FRA 4244 -0.54 0.03 1.07 0.02
## 12 GBR 9232 -0.25 0.03 1.02 0.02
## 13 GRC 4619 -0.65 0.05 1.17 0.02
## 14 HUN 4744 -0.58 0.03 1.04 0.02
## 15 IRL 3830 -0.47 0.04 1.07 0.02
## 16 ISL 3320 0.44 0.02 0.97 0.01
## 17 ITA 11607 -0.63 0.03 1.20 0.02
## 18 JPN 4671 -0.66 0.02 0.84 0.01
## 19 KOR 5420 -0.42 0.03 0.96 0.02
## 20 LUX 3864 -0.14 0.02 1.28 0.01
## 21 MEX 29798 -1.72 0.06 1.31 0.02
## 22 NLD 3868 -0.03 0.03 1.02 0.02
## 23 NOR 4012 0.33 0.02 0.98 0.02
## 24 NZL 4400 -0.14 0.02 1.08 0.01
## 25 POL 4383 -0.61 0.03 1.02 0.01
## 26 PRT 4578 -1.15 0.05 1.42 0.03
## 27 SVK 7335 -0.47 0.03 0.98 0.02
## 28 SWE 4586 0.07 0.03 1.03 0.01
## 29 TUR 4845 -1.71 0.08 1.32 0.04
## 30 USA 5392 -0.02 0.03 1.08 0.02 escs.oecd06## CNT Freq Mean s.e. SD s.e
## 1 Australia 13988 0.06 0.02 0.82 0.01
## 2 Austria 4914 -0.11 0.02 0.85 0.02
## 3 Belgium 8776 -0.04 0.02 0.95 0.01
## 4 Canada 22136 0.20 0.02 0.82 0.01
## 5 Switzerland 12136 -0.09 0.02 0.93 0.01
## 6 Czech Republic 5903 -0.36 0.02 0.78 0.01
## 7 Germany 4686 0.04 0.03 0.94 0.01
## 8 Denmark 4495 0.30 0.03 0.85 0.01
## 9 Spain 19497 -0.33 0.04 1.34 0.02
## 10 Finland 4697 0.07 0.02 0.79 0.01
## 11 France 4600 -0.43 0.03 0.89 0.01
## 12 United Kingdom 12806 -0.06 0.02 0.84 0.01
## 13 Greece 4862 -0.39 0.04 0.99 0.01
## 14 Hungary 4474 -0.47 0.03 0.90 0.01
## 15 Ireland 4501 -0.05 0.03 0.93 0.01
## 16 Iceland 3745 0.50 0.01 0.86 0.01
## 17 Italy 21680 -0.46 0.02 0.99 0.01
## 18 Japan 5862 -0.29 0.02 0.72 0.01
## 19 Korea 5168 -0.38 0.02 0.76 0.01
## 20 Luxembourg 4486 -0.14 0.01 1.14 0.01
## 21 Mexico 30875 -1.38 0.04 1.33 0.02
## 22 Netherlands 4838 0.08 0.03 0.87 0.01
## 23 Norway 4602 0.30 0.02 0.75 0.01
## 24 New Zealand 4727 -0.06 0.02 0.85 0.01
## 25 Poland 5520 -0.75 0.02 0.85 0.01
## 26 Portugal 5091 -0.98 0.04 1.28 0.02
## 27 Slovak Republic 4723 -0.41 0.02 0.90 0.02
## 28 Sweden 4392 0.17 0.02 0.82 0.02
## 29 Turkey 4934 -1.67 0.05 1.16 0.03
## 30 United States 5568 0.04 0.04 0.96 0.02 escs.oecd09## CNT Freq Mean s.e. SD s.e
## 1 Australia 13926 0.17 0.01 0.77 0.01
## 2 Austria 6420 -0.05 0.02 0.83 0.01
## 3 Belgium 8381 0.08 0.02 0.92 0.01
## 4 Canada 22616 0.34 0.02 0.84 0.01
## 5 Switzerland 11738 0.10 0.02 0.89 0.01
## 6 Chile 5580 -0.79 0.03 1.10 0.02
## 7 Czech Republic 6026 -0.21 0.01 0.71 0.01
## 8 Germany 4561 0.10 0.02 0.88 0.01
## 9 Denmark 5796 0.42 0.02 0.80 0.01
## 10 Spain 25631 -0.04 0.04 1.37 0.02
## 11 Estonia 4703 -0.12 0.02 0.77 0.01
## 12 Finland 5776 0.36 0.02 0.72 0.01
## 13 France 4227 -0.28 0.03 0.84 0.01
## 14 United Kingdom 11936 0.04 0.02 0.81 0.01
## 15 Greece 4948 -0.24 0.03 1.00 0.01
## 16 Hungary 4588 -0.40 0.03 0.94 0.02
## 17 Ireland 3841 -0.03 0.03 0.85 0.01
## 18 Iceland 3595 0.61 0.01 0.81 0.01
## 19 Israel 5596 -0.06 0.02 0.87 0.01
## 20 Italy 30810 -0.25 0.01 1.00 0.01
## 21 Japan 5984 -0.22 0.01 0.72 0.01
## 22 Korea 4982 -0.24 0.03 0.73 0.01
## 23 Luxembourg 4539 0.04 0.01 1.09 0.01
## 24 Mexico 38083 -1.32 0.03 1.29 0.01
## 25 Netherlands 4714 0.12 0.03 0.76 0.02
## 26 Norway 4620 0.48 0.02 0.69 0.01
## 27 New Zealand 4553 0.03 0.02 0.81 0.01
## 28 Poland 4867 -0.55 0.02 0.87 0.01
## 29 Portugal 6272 -0.59 0.04 1.17 0.02
## 30 Slovak Republic 4536 -0.31 0.02 0.79 0.01
## 31 Slovenia 6082 -0.09 0.01 0.85 0.01
## 32 Sweden 4512 0.28 0.02 0.81 0.01
## 33 Turkey 4965 -1.54 0.04 1.19 0.02
## 34 United States 5190 0.11 0.04 0.95 0.02 escs.oecd12## CNT Freq Mean s.e. SD s.e
## 1 Australia 14110 0.23 0.01 0.78 0.01
## 2 Austria 0 NaN NaN 0.00 0.00
## 3 Belgium 8414 0.13 0.02 0.89 0.01
## 4 Canada 21087 0.40 0.02 0.86 0.01
## 5 Switzerland 11137 0.17 0.02 0.87 0.01
## 6 Chile 6764 -0.68 0.04 1.12 0.02
## 7 Czech Republic 5295 -0.16 0.02 0.73 0.01
## 8 Germany 4141 0.16 0.02 0.85 0.01
## 9 Denmark 7298 0.46 0.02 0.81 0.01
## 10 Spain 25121 0.13 0.03 1.32 0.01
## 11 Estonia 4727 -0.05 0.01 0.79 0.01
## 12 Finland 8685 0.37 0.02 0.73 0.01
## 13 France 4489 -0.12 0.02 0.77 0.01
## 14 United Kingdom 12367 0.15 0.02 0.80 0.01
## 15 Greece 5091 -0.15 0.03 1.00 0.01
## 16 Hungary 4751 -0.33 0.03 0.96 0.01
## 17 Ireland 4973 0.09 0.02 0.85 0.01
## 18 Iceland 3400 0.68 0.01 0.77 0.01
## 19 Israel 4866 0.09 0.03 0.84 0.02
## 20 Italy 30873 -0.15 0.01 0.96 0.01
## 21 Japan 6185 -0.19 0.02 0.71 0.01
## 22 Korea 5022 -0.13 0.02 0.70 0.01
## 23 Luxembourg 5124 0.10 0.01 1.10 0.01
## 24 Mexico 33598 -1.25 0.02 1.26 0.01
## 25 Netherlands 4376 0.24 0.02 0.75 0.01
## 26 Norway 4574 0.41 0.02 0.71 0.01
## 27 New Zealand 4177 0.03 0.02 0.84 0.01
## 28 Poland 4560 -0.38 0.03 0.88 0.01
## 29 Portugal 5623 -0.58 0.05 1.17 0.02
## 30 Slovak Republic 4629 -0.32 0.03 0.89 0.01
## 31 Slovenia 5833 -0.01 0.01 0.83 0.01
## 32 Sweden 4616 0.30 0.01 0.79 0.01
## 33 Turkey 4806 -1.62 0.04 1.19 0.02
## 34 United States of America 4915 0.12 0.04 0.97 0.02 escs.oecd15## CNT Freq Mean s.e. SD s.e
## 1 Australia 13989 0.27 0.01 0.78 0.01
## 2 Austria 6939 0.09 0.02 0.85 0.01
## 3 Belgium 9452 0.16 0.02 0.90 0.01
## 4 Canada 19424 0.53 0.02 0.80 0.01
## 5 Switzerland 5794 0.14 0.02 0.91 0.01
## 6 Chile 6949 -0.49 0.03 1.08 0.01
## 7 Czech Republic 6788 -0.21 0.01 0.79 0.01
## 8 Germany 5630 0.12 0.02 0.88 0.01
## 9 Denmark 6985 0.59 0.02 0.86 0.01
## 10 Spain 6678 -0.51 0.04 1.19 0.01
## 11 Estonia 5499 0.05 0.01 0.76 0.01
## 12 Finland 5812 0.25 0.02 0.75 0.01
## 13 France 5941 -0.14 0.02 0.79 0.01
## 14 United Kingdom 13516 0.21 0.02 0.84 0.01
## 15 Greece 5492 -0.08 0.03 0.96 0.01
## 16 Hungary 5570 -0.23 0.02 0.95 0.01
## 17 Ireland 5667 0.16 0.02 0.84 0.01
## 18 Iceland 3283 0.73 0.01 0.72 0.01
## 19 Israel 6501 0.16 0.03 0.84 0.02
## 20 Italy 11330 -0.07 0.02 0.94 0.01
## 21 Japan 6557 -0.18 0.01 0.70 0.01
## 22 Korea 5548 -0.20 0.02 0.68 0.01
## 23 Luxembourg 5183 0.07 0.01 1.09 0.01
## 24 Latvia 4817 -0.44 0.02 0.91 0.01
## 25 Mexico 7507 -1.22 0.04 1.21 0.02
## 26 Netherlands 5324 0.16 0.02 0.76 0.01
## 27 Norway 5286 0.48 0.02 0.72 0.01
## 28 New Zealand 4334 0.17 0.02 0.77 0.01
## 29 Poland 4446 -0.39 0.02 0.82 0.01
## 30 Portugal 7225 -0.39 0.03 1.14 0.01
## 31 Slovak Republic 6257 -0.11 0.02 0.94 0.02
## 32 Slovenia 6343 0.03 0.01 0.81 0.01
## 33 Sweden 5313 0.33 0.02 0.81 0.01
## 34 Turkey 5859 -1.43 0.05 1.17 0.02
## 35 United States 5638 0.10 0.04 1.00 0.02 escs.oecd <- rbind(mean(escs.oecd03[,3]),
mean(escs.oecd06[,3]),
mean(escs.oecd09[,3]),
mean(escs.oecd12[,3],na.rm=TRUE),
mean(escs.oecd15[,3]))
rownames(escs.oecd) <- c("2003","2006","2009","2012","2015")
escs.oecd <- round(escs.oecd,2)
escs.oecd## [,1]
## 2003 -0.35
## 2006 -0.24
## 2009 -0.12
## 2012 -0.06
## 2015 -0.04# Veri setlerinde AB uyesi olup olmadigini gosteren bir degisken olmadigindan
# bunlari tek tek bizim yazip o sekilde suzmemiz lazim.
# Bunu yapmak icin su linkteki AB uyesi ulkelerin listesini kullandik
# https://europa.eu/european-union/about-eu/countries_en
AB <- c("AUT","BEL","BGR","HRV","CYP","CZE","DNK","EST","FIN","FRA","DEU","GRC","HUN",
"IRL","ITA","LVA","LTU","LUX","MLT","NLD","POL","PRT","ROU","SVK","SVN","ESP",
"SWE")
AB2 <- c("Austria","Belgium","Bulgaria","Croatia","Cyprus","Czech Republic","Denmark",
"Estonia","Finland","France","Germany","Greece","Hungary","Ireland","Italy",
"Latvia","Lithuania","Luxembourg","Malta","Netherlands","Poland","Portugal",
"Romania","Slovak Republic","Slovenia","Spain","Sweden")
pisa_2003_stu_AB <- subset(pisa_2003_stu, CNT %in% AB)
pisa_2006_stu_AB <- subset(pisa_2006_stu, CNT %in% AB2)
pisa_2009_stu_AB <- subset(pisa_2009_stu, CNT %in% AB2)
pisa_2012_stu_AB <- subset(pisa_2012_stu, CNT %in% AB2)
pisa_2015_stu_AB <- subset(pisa_2015_stu, CNT %in% AB2)
escs.AB03 <- pisa.mean(variable="escs_trend",data=pisa_2003_stu_AB,by="CNT")
escs.AB06 <- pisa.mean(variable="escs_trend",data=pisa_2006_stu_AB,by="CNT")
escs.AB09 <- pisa.mean(variable="escs_trend",data=pisa_2009_stu_AB,by="CNT")
escs.AB12 <- pisa.mean(variable="escs_trend",data=pisa_2012_stu_AB,by="CNT")
escs.AB15 <- pisa2015.mean(variable="ESCS",data=pisa_2015_stu_AB,by="CNT")
escs.AB03## CNT Freq Mean s.e. SD s.e
## 1 AUT 4548 -0.29 0.03 0.97 0.01
## 2 BEL 8551 -0.12 0.03 1.14 0.01
## 3 CZE 6155 -0.19 0.03 0.95 0.01
## 4 DEU 4399 -0.08 0.03 1.17 0.02
## 5 DNK 4178 0.07 0.03 1.02 0.02
## 6 ESP 10686 -0.42 0.06 1.51 0.02
## 7 FIN 5778 -0.03 0.02 1.01 0.01
## 8 FRA 4244 -0.54 0.03 1.07 0.02
## 9 GRC 4619 -0.65 0.05 1.17 0.02
## 10 HUN 4744 -0.58 0.03 1.04 0.02
## 11 IRL 3830 -0.47 0.04 1.07 0.02
## 12 ITA 11607 -0.63 0.03 1.20 0.02
## 13 LUX 3864 -0.14 0.02 1.28 0.01
## 14 LVA 4595 -0.38 0.03 0.82 0.01
## 15 NLD 3868 -0.03 0.03 1.02 0.02
## 16 POL 4383 -0.61 0.03 1.02 0.01
## 17 PRT 4578 -1.15 0.05 1.42 0.03
## 18 SVK 7335 -0.47 0.03 0.98 0.02
## 19 SWE 4586 0.07 0.03 1.03 0.01 escs.AB06## CNT Freq Mean s.e. SD s.e
## 1 Austria 4914 -0.11 0.02 0.85 0.02
## 2 Belgium 8776 -0.04 0.02 0.95 0.01
## 3 Bulgaria 4396 -0.62 0.05 1.03 0.02
## 4 Czech Republic 5903 -0.36 0.02 0.78 0.01
## 5 Germany 4686 0.04 0.03 0.94 0.01
## 6 Denmark 4495 0.30 0.03 0.85 0.01
## 7 Spain 19497 -0.33 0.04 1.34 0.02
## 8 Estonia 4853 -0.28 0.02 0.81 0.01
## 9 Finland 4697 0.07 0.02 0.79 0.01
## 10 France 4600 -0.43 0.03 0.89 0.01
## 11 Greece 4862 -0.39 0.04 0.99 0.01
## 12 Croatia 5205 -0.49 0.02 0.86 0.01
## 13 Hungary 4474 -0.47 0.03 0.90 0.01
## 14 Ireland 4501 -0.05 0.03 0.93 0.01
## 15 Italy 21680 -0.46 0.02 0.99 0.01
## 16 Lithuania 4719 -0.44 0.03 0.90 0.01
## 17 Luxembourg 4486 -0.14 0.01 1.14 0.01
## 18 Latvia 4691 -0.50 0.02 0.84 0.01
## 19 Netherlands 4838 0.08 0.03 0.87 0.01
## 20 Poland 5520 -0.75 0.02 0.85 0.01
## 21 Portugal 5091 -0.98 0.04 1.28 0.02
## 22 Romania 5110 -0.83 0.04 0.91 0.03
## 23 Slovak Republic 4723 -0.41 0.02 0.90 0.02
## 24 Sweden 4392 0.17 0.02 0.82 0.02 escs.AB09## CNT Freq Mean s.e. SD s.e
## 1 Austria 6420 -0.05 0.02 0.83 0.01
## 2 Belgium 8381 0.08 0.02 0.92 0.01
## 3 Bulgaria 4406 -0.37 0.04 0.96 0.02
## 4 Czech Republic 6026 -0.21 0.01 0.71 0.01
## 5 Germany 4561 0.10 0.02 0.88 0.01
## 6 Denmark 5796 0.42 0.02 0.80 0.01
## 7 Spain 25631 -0.04 0.04 1.37 0.02
## 8 Estonia 4703 -0.12 0.02 0.77 0.01
## 9 Finland 5776 0.36 0.02 0.72 0.01
## 10 France 4227 -0.28 0.03 0.84 0.01
## 11 Greece 4948 -0.24 0.03 1.00 0.01
## 12 Croatia 4987 -0.33 0.02 0.87 0.01
## 13 Hungary 4588 -0.40 0.03 0.94 0.02
## 14 Ireland 3841 -0.03 0.03 0.85 0.01
## 15 Italy 30810 -0.25 0.01 1.00 0.01
## 16 Lithuania 4481 -0.32 0.02 0.90 0.01
## 17 Luxembourg 4539 0.04 0.01 1.09 0.01
## 18 Latvia 4481 -0.46 0.02 0.80 0.01
## 19 Malta 3356 -0.27 0.02 0.96 0.01
## 20 Netherlands 4714 0.12 0.03 0.76 0.02
## 21 Poland 4867 -0.55 0.02 0.87 0.01
## 22 Portugal 6272 -0.59 0.04 1.17 0.02
## 23 Romania 4766 -0.65 0.03 0.86 0.02
## 24 Slovak Republic 4536 -0.31 0.02 0.79 0.01
## 25 Slovenia 6082 -0.09 0.01 0.85 0.01
## 26 Sweden 4512 0.28 0.02 0.81 0.01 escs.AB12## CNT Freq Mean s.e. SD s.e
## 1 Austria 0 NaN NaN 0.00 0.00
## 2 Belgium 8414 0.13 0.02 0.89 0.01
## 3 Bulgaria 5183 -0.31 0.03 1.01 0.02
## 4 Czech Republic 5295 -0.16 0.02 0.73 0.01
## 5 Germany 4141 0.16 0.02 0.85 0.01
## 6 Denmark 7298 0.46 0.02 0.81 0.01
## 7 Spain 25121 0.13 0.03 1.32 0.01
## 8 Estonia 4727 -0.05 0.01 0.79 0.01
## 9 Finland 8685 0.37 0.02 0.73 0.01
## 10 France 4489 -0.12 0.02 0.77 0.01
## 11 Greece 5091 -0.15 0.03 1.00 0.01
## 12 Croatia 4991 -0.30 0.02 0.84 0.01
## 13 Hungary 4751 -0.33 0.03 0.96 0.01
## 14 Ireland 4973 0.09 0.02 0.85 0.01
## 15 Italy 30873 -0.15 0.01 0.96 0.01
## 16 Lithuania 4581 -0.18 0.02 0.90 0.01
## 17 Luxembourg 5124 0.10 0.01 1.10 0.01
## 18 Latvia 4268 -0.39 0.02 0.83 0.01
## 19 Netherlands 4376 0.24 0.02 0.75 0.01
## 20 Poland 4560 -0.38 0.03 0.88 0.01
## 21 Portugal 5623 -0.58 0.05 1.17 0.02
## 22 Romania 5055 -0.55 0.04 0.91 0.02
## 23 Slovak Republic 4629 -0.32 0.03 0.89 0.01
## 24 Slovenia 5833 -0.01 0.01 0.83 0.01
## 25 Sweden 4616 0.30 0.01 0.79 0.01 escs.AB15## CNT Freq Mean s.e. SD s.e
## 1 Austria 6939 0.09 0.02 0.85 0.01
## 2 Belgium 9452 0.16 0.02 0.90 0.01
## 3 Bulgaria 5793 -0.08 0.03 0.98 0.02
## 4 Czech Republic 6788 -0.21 0.01 0.79 0.01
## 5 Germany 5630 0.12 0.02 0.88 0.01
## 6 Denmark 6985 0.59 0.02 0.86 0.01
## 7 Spain 6678 -0.51 0.04 1.19 0.01
## 8 Estonia 5499 0.05 0.01 0.76 0.01
## 9 Finland 5812 0.25 0.02 0.75 0.01
## 10 France 5941 -0.14 0.02 0.79 0.01
## 11 Greece 5492 -0.08 0.03 0.96 0.01
## 12 Croatia 5728 -0.24 0.02 0.82 0.01
## 13 Hungary 5570 -0.23 0.02 0.95 0.01
## 14 Ireland 5667 0.16 0.02 0.84 0.01
## 15 Italy 11330 -0.07 0.02 0.94 0.01
## 16 Lithuania 6334 -0.06 0.02 0.86 0.01
## 17 Luxembourg 5183 0.07 0.01 1.09 0.01
## 18 Latvia 4817 -0.44 0.02 0.91 0.01
## 19 Malta 3602 -0.05 0.01 0.95 0.01
## 20 Netherlands 5324 0.16 0.02 0.76 0.01
## 21 Poland 4446 -0.39 0.02 0.82 0.01
## 22 Portugal 7225 -0.39 0.03 1.14 0.01
## 23 Romania 4873 -0.58 0.04 0.87 0.02
## 24 Slovak Republic 6257 -0.11 0.02 0.94 0.02
## 25 Slovenia 6343 0.03 0.01 0.81 0.01
## 26 Sweden 5313 0.33 0.02 0.81 0.01 escs.AB <- rbind(mean(escs.AB03[,3]),
mean(escs.AB06[,3]),
mean(escs.AB09[,3]),
mean(escs.AB12[,3],na.rm=TRUE),
mean(escs.AB15[,3]))
rownames(escs.AB) <- c("2003","2006","2009","2012","2015")
escs.AB <- round(escs.AB,2)
escs.AB## [,1]
## 2003 -0.35
## 2006 -0.31
## 2009 -0.16
## 2012 -0.08
## 2015 -0.06Istersek bunlari bir tabloya koyabiliriz
Sosyo-ekonomik Kulturel Duzey
| Yil | Turkiye | OECD | AB |
|---|---|---|---|
| 2003 | -1.71(0.08) | -0.35 | -0.35 |
| 2006 | -1.67(0.05) | -0.24 | -0.31 |
| 2009 | -1.54(0.04) | -0.12 | -0.16 |
| 2012 | -1.62(0.04) | -0.06 | -0.08 |
| 2015 | -1.43(0.05) | -0.04 | -0.06 |
Bu rakamlari bir grafikte gostermek istersek asagidaki kodu kullanabiliriz. Bunun icin oncelikle “ggplot2” paketi yuklememiz gerekmekter.
install.packages("ggplot2",dependencies=TREUE)
install.packages("gridExtra",dependencies=TREUE)
install.packages("grid",dependencies=TREUE)require(ggplot2)## Loading required package: ggplot2## Warning: package 'ggplot2' was built under R version 3.2.5library(gridExtra)## Warning: package 'gridExtra' was built under R version 3.2.5library(grid)
escs <- rbind(escs.tur,escs.tur,escs.tur)
escs$tip <- c(rep("TUR",5),rep("OECD",5),rep("AB",5))
escs[6:10,2]=escs.oecd[,1]
escs[6:10,3]=0
escs[11:15,2]=escs.AB[,1]
escs[11:15,3]=0
escs$year = 1:5
ggplot(escs, aes(x=year, y=Mean,shape=tip)) +
theme_bw() +
geom_errorbar(aes(ymin=Mean-1.96*s.e., ymax=Mean+1.96*s.e.), width=.05,lty=2) +
geom_point(size=4, fill="black") +
geom_line()+
scale_x_discrete(limit = 1:5,labels=c("2003","2006","2009","2012","2015"))+
scale_y_continuous(limit = c(-2,.5)) +
labs(title = " PISA ESCS Indeksinin Yillara Gore Degisimi",
x = "YIL", y = "PISA Sosyoekonomik Kulturel Indeks",
shape=" ")+
theme(axis.title= element_text(size = 15),
axis.text= element_text(size = 12),
title = element_text(size = 12),
legend.justification=c(-0.5,-0.2),
legend.position=c(0,0),
legend.text=element_text(size = 12)
) +
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 1, xmax = 1, ymin = .3, ymax = .3)
ESCS acisindan 5% ve 95% lik dilimler arasindaki fark
# 2015
quantile(pisa_2015_stu_TUR$ESCS,probs=c(0.05,.95),na.rm=TRUE)## 5% 95%
## -3.21623 0.60585 quantile(pisa_2012_stu_TUR$escs_trend,probs=c(0.05,.95),na.rm=TRUE)## 5% 95%
## -3.2935630 0.5938632 quantile(pisa_2009_stu_TUR$escs_trend,probs=c(0.05,.95),na.rm=TRUE)## 5% 95%
## -3.1814873 0.6501741 quantile(pisa_2006_stu_TUR$escs_trend,probs=c(0.05,.95),na.rm=TRUE)## 5% 95%
## -3.2325891 0.5644762 quantile(pisa_2003_stu_TUR$escs_trend,probs=c(0.05,.95),na.rm=TRUE)## 5% 95%
## -3.4901217 0.6972358| Yil | 5% | 95% | Fark |
|---|---|---|---|
| 2003 | -3.49 | 0.70 | 4.20 |
| 2006 | -3.23 | 0.56 | 3.79 |
| 2009 | -3.18 | 0.65 | 3.83 |
| 2012 | -3.29 | 0.59 | 3.88 |
| 2015 | -3.22 | 0.61 | 3.83 |
# Bireysel Baglamda Diger Demografik Degiskenlerin Incelenmesi
## Sinif Seviyesi sinif.2003 <- pisa.table(variable="ST01Q01",data=pisa_2003_stu_TUR)
sinif.2006 <- pisa.table(variable="ST01Q01",data=pisa_2006_stu_TUR)
sinif.2009 <- pisa.table(variable="ST01Q01",data=pisa_2009_stu_TUR)
sinif.2012 <- pisa.table(variable="ST01Q01",data=pisa_2012_stu_TUR)
sinif.2015 <- pisa2015.table(variable="ST001D01T",data=pisa_2015_stu_TUR)
colnames(sinif.2015)[1] <- "ST01Q01"
levels(sinif.2015$ST01Q01) <- 7:12
sinif <- rbind(sinif.2003,sinif.2006,sinif.2009,sinif.2012,sinif.2015)
sinif$year <- factor(c(rep(2003,6),rep(2006,5),rep(2009,6),rep(2012,6),rep(2015,6)))
sinif## ST01Q01 Freq Percentage Std.err. year
## 7 7 27 0.84 0.33 2003
## 8 8 92 4.39 1.60 2003
## 9 9 191 3.20 0.67 2003
## 10 10 2863 52.12 2.18 2003
## 11 11 1670 39.19 2.37 2003
## 12 12 12 0.25 0.09 2003
## 71 7 23 0.82 0.32 2006
## 81 8 93 4.51 0.90 2006
## 91 9 2007 38.38 1.73 2006
## 101 10 2671 53.66 1.88 2006
## 111 11 148 2.63 0.27 2006
## 72 7 24 0.70 0.14 2009
## 82 8 113 3.50 0.76 2009
## 92 9 1258 25.17 1.29 2009
## 102 10 3393 66.62 1.52 2009
## 112 11 196 3.79 0.26 2009
## 121 12 12 0.22 0.07 2009
## 73 7 21 0.50 0.19 2012
## 83 8 99 2.16 0.31 2012
## 93 9 1317 27.60 1.20 2012
## 103 10 3202 65.47 1.23 2012
## 113 11 194 3.97 0.28 2012
## 122 12 15 0.29 0.07 2012
## 1 7 16 0.60 0.14 2015
## 2 8 105 2.62 0.36 2015
## 3 9 1273 20.73 0.99 2015
## 4 10 4308 72.93 1.19 2015
## 5 11 186 3.00 0.32 2015
## 6 12 7 0.12 0.04 2015 ggplot(sinif, aes(x=year, y=Percentage,fill=ST01Q01)) +
geom_bar(stat='identity',position=position_dodge()) +
geom_errorbar(aes(ymin=Percentage-1.96*Std.err., ymax=Percentage+1.96*Std.err.),
width=.2,position=position_dodge(.9)) +
scale_y_continuous(limit = c(0,100))+
theme_bw()+
geom_text(aes(y=Percentage,label=scales::percent(sinif$Percentage/100)),
stat= "identity", position=position_dodge(1),vjust = -3)+
labs(title =" Ogrencilerin Sinif Duzeyi Dagilimi",
x = "YIL", y = "Yuzde",
fill=" ")+
theme(axis.title= element_text(size = 15),
axis.text= element_text(size = 12),
title = element_text(size = 18),
legend.text=element_text(size = 12)) +
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 1, xmax = 1, ymin = 90, ymax = 90)
Sinif Duzeyinde katilim orani
| Yil | 7.Sinif | 8.Sinif | 9.Sinif | 10.Sinif | 11.Sinif | 12.Sinif |
|---|---|---|---|---|---|---|
| 2003 | %0.84 | %4.39 | %3.20 | %52.12 | %39.19 | %0.25 |
| 2006 | %0.82 | %4.51 | %38.38 | %53.66 | %2.63 | %0 |
| 2009 | %0.70 | %3.50 | %25.17 | %66.62 | %3.79 | %0.22 |
| 2012 | %0.50 | %2.16 | %27.60 | %65.47 | %3.97 | %0.29 |
| 2015 | %0.60 | %2.62 | %20.73 | %72.93 | %3.00 | %0.12 |
Cinsiyet
cins.2003 <- pisa.table(variable="ST03Q01",data=pisa_2003_stu_TUR)
cins.2006 <- pisa.table(variable="ST04Q01",data=pisa_2006_stu_TUR)
cins.2009 <- pisa.table(variable="ST04Q01",data=pisa_2009_stu_TUR)
cins.2012 <- pisa.table(variable="ST04Q01",data=pisa_2012_stu_TUR)
cins.2015 <- pisa2015.table(variable="ST004D01T",data=pisa_2015_stu_TUR)
colnames(cins.2015)[1] <- "ST04Q01"
colnames(cins.2003)[1] <- "ST04Q01"
cins <- rbind(cins.2003,cins.2006,cins.2009,cins.2012,cins.2015)
cins$year <- factor(c(rep(2003,2),rep(2006,2),rep(2009,2),rep(2012,2),rep(2015,2)))
cins## ST04Q01 Freq Percentage Std.err. year
## 1 Female 2090 44.99 1.95 2003
## 2 Male 2765 55.01 1.95 2003
## 11 Female 2290 45.30 1.92 2006
## 21 Male 2652 54.70 1.92 2006
## 12 Female 2445 48.40 1.71 2009
## 22 Male 2551 51.60 1.71 2009
## 13 Female 2370 49.45 1.98 2012
## 23 Male 2478 50.55 1.98 2012
## 14 Female 2938 49.97 1.55 2015
## 24 Male 2957 50.03 1.55 2015 ggplot(cins, aes(x=year, y=Percentage,fill=ST04Q01)) +
geom_bar(stat='identity',position=position_dodge()) +
geom_errorbar(aes(ymin=Percentage-1.96*Std.err., ymax=Percentage+1.96*Std.err.),
width=.2,position=position_dodge(.9)) +
scale_y_continuous(limit = c(0,100))+
theme_bw()+
geom_text(aes(y=Percentage,label=scales::percent(cins$Percentage/100)),
stat= "identity", position=position_dodge(1),vjust = -3)+
labs(title =" Ogrencilerin Cinsiyet Dagilimi",
x = "YIL", y = "Yuzde",
fill=" ")+
theme(axis.title= element_text(size = 15),
axis.text= element_text(size = 12),
title = element_text(size = 18),
legend.text=element_text(size = 12)) +
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 1, xmax = 1, ymin = 90, ymax = 90)
Gocmenlik (Dogum yeriniz baska bir ulke mi?)
goc.2003 <- pisa.table(variable="ST15Q01",data=pisa_2003_stu_TUR)
goc.2006 <- pisa.table(variable="ST11Q01",data=pisa_2006_stu_TUR)
goc.2009 <- pisa.table(variable="ST17Q01",data=pisa_2009_stu_TUR)
goc.2012 <- pisa.table(variable="ST20Q01",data=pisa_2012_stu_TUR)
goc.2015 <- pisa2015.table(variable="ST019AQ01T",data=pisa_2015_stu_TUR)
colnames(goc.2015)[1] <- "goc"
colnames(goc.2003)[1] <- "goc"
colnames(goc.2006)[1] <- "goc"
colnames(goc.2009)[1] <- "goc"
colnames(goc.2012)[1] <- "goc"
levels(goc.2003$goc) <- levels(goc.2015$goc)
levels(goc.2006$goc) <- levels(goc.2015$goc)
goc <- rbind(goc.2003,goc.2006,goc.2009,goc.2012,goc.2015)
goc$year <- factor(c(rep(2003,2),rep(2006,2),rep(2009,2),rep(2012,2),rep(2015,2)))
goc## goc Freq Percentage Std.err. year
## 1 Country of test 4771 98.99 0.18 2003
## 2 Other country 62 1.01 0.18 2003
## 11 Country of test 4834 98.62 0.26 2006
## 21 Other country 64 1.38 0.26 2006
## 12 Country of test 4922 99.26 0.13 2009
## 22 Other country 35 0.74 0.13 2009
## 13 Country of test 4763 99.20 0.14 2012
## 23 Other country 38 0.80 0.14 2012
## 14 Country of test 5765 99.27 0.13 2015
## 24 Other country 46 0.73 0.13 2015Gocmenlik Durumu (Turkiye’de mi dogdunuz?)
| Yil | Evet | Hayir |
|---|---|---|
| 2003 | %98.99 | %1.01 |
| 2006 | %98.62 | %1.38 |
| 2009 | %99.26 | %0.74 |
| 2012 | %99.20 | %0.80 |
| 2015 | %99.27 | %0.73 |
Haftalik Ders Saati
# Language
pisa_2009_stu_TUR$lang.min <- (pisa_2009_stu_TUR$ST29Q01*pisa_2009_stu_TUR$ST28Q01)
pisa_2012_stu_TUR$lang.min <- (pisa_2012_stu_TUR$ST70Q01*pisa_2012_stu_TUR$ST69Q01)
pisa_2015_stu_TUR$lang.min <- (pisa_2015_stu_TUR$ST059Q01TA*pisa_2015_stu_TUR$ST061Q01NA)
tur.2009 <- pisa.mean(variable="lang.min",data=pisa_2009_stu_TUR)
tur.2012 <- pisa.mean(variable="lang.min",data=pisa_2012_stu_TUR)
tur.2015 <- pisa2015.mean(variable="lang.min",data=pisa_2015_stu_TUR)
tur <- rbind(tur.2009,tur.2012,tur.2015)
# OECD
pisa_2009_stu_OECD$lang.min <- (pisa_2009_stu_OECD$ST29Q01*pisa_2009_stu_OECD$ST28Q01)
pisa_2012_stu_OECD$lang.min <- (pisa_2012_stu_OECD$ST70Q01*pisa_2012_stu_OECD$ST69Q01)
pisa_2015_stu_OECD$lang.min <- (pisa_2015_stu_OECD$ST059Q01TA*pisa_2015_stu_OECD$ST061Q01NA)
tur.2009_oecd <- pisa.mean(variable="lang.min",data=pisa_2009_stu_OECD,by="CNT")
tur.2012_oecd <- pisa.mean(variable="lang.min",data=pisa_2012_stu_OECD,by="CNT")
tur.2015_oecd <- pisa2015.mean(variable="lang.min",data=pisa_2015_stu_OECD,by="CNT")
# AB
pisa_2009_stu_AB$lang.min <- (pisa_2009_stu_AB$ST29Q01*pisa_2009_stu_AB$ST28Q01)
pisa_2012_stu_AB$lang.min <- (pisa_2012_stu_AB$ST70Q01*pisa_2012_stu_AB$ST69Q01)
pisa_2015_stu_AB$lang.min <- (pisa_2015_stu_AB$ST059Q01TA*pisa_2015_stu_AB$ST061Q01NA)
tur.2009_AB <- pisa.mean(variable="lang.min",data=pisa_2009_stu_AB,by="CNT")
tur.2012_AB <- pisa.mean(variable="lang.min",data=pisa_2012_stu_AB,by="CNT")
tur.2015_AB <- pisa2015.mean(variable="lang.min",data=pisa_2015_stu_AB,by="CNT")
tur <- rbind(tur,tur,tur)
tur$tip <- c(rep("TUR",3),rep("OECD",3),rep("AB",3))
tur$year <- 1:3
tur[4:6,2]=c(mean(tur.2009_oecd[,3]),mean(tur.2012_oecd[,3]),mean(tur.2015_oecd[,3]))
tur[4:6,3]=0
tur[7:9,2]=c(mean(tur.2009_AB[,3]),mean(tur.2012_AB[,3]),mean(tur.2015_AB[,3],na.rm=TRUE))
tur[7:9,3]=0
ggplot(tur, aes(x=year, y=Mean,shape=tip)) +
theme_bw() +
geom_errorbar(aes(ymin=Mean-1.96*s.e., ymax=Mean+1.96*s.e.), width=.05,lty=2) +
geom_point(size=4, fill="black") +
geom_line()+
scale_x_discrete(limit = 1:3,labels=c("2009","2012","2015"))+
scale_y_continuous(limit = c(150,250)) +
labs(title = " Haftalik Ortalama Toplam Turkce Ders Saati",
x = "YIL", y = "Dakika",
shape=" ")+
theme(axis.title= element_text(size = 18),
axis.text= element_text(size = 15),
title = element_text(size = 20),
legend.justification=c(-0.5,-0.2),
legend.position=c(0,0),
legend.text=element_text(size = 18)
) +
geom_text(aes(y=Mean,label=round(tur$Mean,2)), position=position_dodge(0.2),vjust = -2)+
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 3.5, xmax = 3.5, ymin = 240, ymax = 240)
# Matematik
pisa_2009_stu_TUR$lang.min_mat <- (pisa_2009_stu_TUR$ST29Q02*pisa_2009_stu_TUR$ST28Q02)
pisa_2012_stu_TUR$lang.min_mat <- (pisa_2012_stu_TUR$ST70Q02*pisa_2012_stu_TUR$ST69Q02)
pisa_2015_stu_TUR$lang.min_mat <- (pisa_2015_stu_TUR$ST059Q02TA*pisa_2015_stu_TUR$ST061Q01NA)
tur.2009 <- pisa.mean(variable="lang.min_mat",data=pisa_2009_stu_TUR)
tur.2012 <- pisa.mean(variable="lang.min_mat",data=pisa_2012_stu_TUR)
tur.2015 <- pisa2015.mean(variable="lang.min_mat",data=pisa_2015_stu_TUR)
tur <- rbind(tur.2009,tur.2012,tur.2015)
# OECD
pisa_2009_stu_OECD$lang.min_mat <- (pisa_2009_stu_OECD$ST29Q02*pisa_2009_stu_OECD$ST28Q02)
pisa_2012_stu_OECD$lang.min_mat <- (pisa_2012_stu_OECD$ST70Q02*pisa_2012_stu_OECD$ST69Q02)
pisa_2015_stu_OECD$lang.min_mat <- (pisa_2015_stu_OECD$ST059Q02TA*pisa_2015_stu_OECD$ST061Q01NA)
tur.2009_oecd <- pisa.mean(variable="lang.min_mat",data=pisa_2009_stu_OECD,by="CNT")
tur.2012_oecd <- pisa.mean(variable="lang.min_mat",data=pisa_2012_stu_OECD,by="CNT")
tur.2015_oecd <- pisa2015.mean(variable="lang.min_mat",data=pisa_2015_stu_OECD,by="CNT")
# AB
pisa_2009_stu_AB$lang.min_mat <- (pisa_2009_stu_AB$ST29Q02*pisa_2009_stu_AB$ST28Q02)
pisa_2012_stu_AB$lang.min_mat <- (pisa_2012_stu_AB$ST70Q02*pisa_2012_stu_AB$ST69Q02)
pisa_2015_stu_AB$lang.min_mat <- (pisa_2015_stu_AB$ST059Q02TA*pisa_2015_stu_AB$ST061Q01NA)
tur.2009_AB <- pisa.mean(variable="lang.min_mat",data=pisa_2009_stu_AB,by="CNT")
tur.2012_AB <- pisa.mean(variable="lang.min_mat",data=pisa_2012_stu_AB,by="CNT")
tur.2015_AB <- pisa2015.mean(variable="lang.min_mat",data=pisa_2015_stu_AB,by="CNT")
tur <- rbind(tur,tur,tur)
tur$tip <- c(rep("TUR",3),rep("OECD",3),rep("AB",3))
tur$year <- 1:3
tur[4:6,2]=c(mean(tur.2009_oecd[,3]),mean(tur.2012_oecd[,3]),mean(tur.2015_oecd[,3]))
tur[4:6,3]=0
tur[7:9,2]=c(mean(tur.2009_AB[,3]),mean(tur.2012_AB[,3]),mean(tur.2015_AB[,3],na.rm=TRUE))
tur[7:9,3]=0
ggplot(tur, aes(x=year, y=Mean,shape=tip)) +
theme_bw() +
geom_errorbar(aes(ymin=Mean-1.96*s.e., ymax=Mean+1.96*s.e.), width=.05,lty=2) +
geom_point(size=4, fill="black") +
geom_line()+
scale_x_discrete(limit = 1:3,labels=c("2009","2012","2015"))+
scale_y_continuous(limit = c(150,250)) +
labs(title = " Haftalik Ortalama Toplam Matematik Ders Saati",
x = "YIL", y = "Dakika",
shape=" ")+
theme(axis.title= element_text(size = 18),
axis.text= element_text(size = 15),
title = element_text(size = 20),
legend.justification=c(-0.5,-0.2),
legend.position=c(0,0),
legend.text=element_text(size = 18)
) +
geom_text(aes(y=Mean,label=round(tur$Mean,2)), position=position_dodge(0),vjust =1,hjust=-0.25)+
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 3.5, xmax = 3.5, ymin = 240, ymax = 240)
# Fen ve Bilim
pisa_2009_stu_TUR$lang.min_sci <- (pisa_2009_stu_TUR$ST29Q03*pisa_2009_stu_TUR$ST28Q03)
pisa_2012_stu_TUR$lang.min_sci <- (pisa_2012_stu_TUR$ST70Q03*pisa_2012_stu_TUR$ST69Q03)
pisa_2015_stu_TUR$lang.min_sci <- (pisa_2015_stu_TUR$ST059Q03TA*pisa_2015_stu_TUR$ST061Q01NA)
tur.2009 <- pisa.mean(variable="lang.min_sci",data=pisa_2009_stu_TUR)
tur.2012 <- pisa.mean(variable="lang.min_sci",data=pisa_2012_stu_TUR)
tur.2015 <- pisa2015.mean(variable="lang.min_sci",data=pisa_2015_stu_TUR)
tur <- rbind(tur.2009,tur.2012,tur.2015)
# OECD
pisa_2009_stu_OECD$lang.min_sci <- (pisa_2009_stu_OECD$ST29Q03*pisa_2009_stu_OECD$ST28Q03)
pisa_2012_stu_OECD$lang.min_sci <- (pisa_2012_stu_OECD$ST70Q03*pisa_2012_stu_OECD$ST69Q03)
pisa_2015_stu_OECD$lang.min_sci <- (pisa_2015_stu_OECD$ST059Q03TA*pisa_2015_stu_OECD$ST061Q01NA)
tur.2009_oecd <- pisa.mean(variable="lang.min_sci",data=pisa_2009_stu_OECD,by="CNT")
tur.2012_oecd <- pisa.mean(variable="lang.min_sci",data=pisa_2012_stu_OECD,by="CNT")
tur.2015_oecd <- pisa2015.mean(variable="lang.min_sci",data=pisa_2015_stu_OECD,by="CNT")
# AB
pisa_2009_stu_AB$lang.min_sci <- (pisa_2009_stu_AB$ST29Q03*pisa_2009_stu_AB$ST28Q03)
pisa_2012_stu_AB$lang.min_sci <- (pisa_2012_stu_AB$ST70Q03*pisa_2012_stu_AB$ST69Q03)
pisa_2015_stu_AB$lang.min_sci <- (pisa_2015_stu_AB$ST059Q03TA*pisa_2015_stu_AB$ST061Q01NA)
tur.2009_AB <- pisa.mean(variable="lang.min_sci",data=pisa_2009_stu_AB,by="CNT")
tur.2012_AB <- pisa.mean(variable="lang.min_sci",data=pisa_2012_stu_AB,by="CNT")
tur.2015_AB <- pisa2015.mean(variable="lang.min_sci",data=pisa_2015_stu_AB,by="CNT")
tur <- rbind(tur,tur,tur)
tur$tip <- c(rep("TUR",3),rep("OECD",3),rep("AB",3))
tur$year <- 1:3
tur[4:6,2]=c(mean(tur.2009_oecd[,3]),mean(tur.2012_oecd[,3]),mean(tur.2015_oecd[,3]))
tur[4:6,3]=0
tur[7:9,2]=c(mean(tur.2009_AB[,3]),mean(tur.2012_AB[,3]),mean(tur.2015_AB[,3],na.rm=TRUE))
tur[7:9,3]=0
ggplot(tur, aes(x=year, y=Mean,shape=tip)) +
theme_bw() +
geom_errorbar(aes(ymin=Mean-1.96*s.e., ymax=Mean+1.96*s.e.), width=.05,lty=2) +
geom_point(size=4, fill="black") +
geom_line()+
scale_x_discrete(limit = 1:3,labels=c("2009","2012","2015"))+
scale_y_continuous(limit = c(115,215)) +
labs(title = " Haftalik Ortalama Toplam Fen ve Bilim Ders Saati",
x = "YIL", y = "Dakika",
shape=" ")+
theme(axis.title= element_text(size = 18),
axis.text= element_text(size = 15),
title = element_text(size = 20),
legend.justification=c(-0.5,-0.2),
legend.position=c(0,0),
legend.text=element_text(size = 18)
) +
geom_text(aes(y=Mean,label=round(tur$Mean,2)), position=position_dodge(0),vjust=1,hjust=1.1)+
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 3.2, xmax = 3.2, ymin = 140, ymax = 140)
YAS
# TURKIYE
tur.2003 <- pisa.mean(variable="AGE",data=pisa_2003_stu_TUR)
tur.2006 <- pisa.mean(variable="AGE",data=pisa_2006_stu_TUR)
tur.2009 <- pisa.mean(variable="AGE",data=pisa_2009_stu_TUR)
tur.2012 <- pisa.mean(variable="AGE",data=pisa_2012_stu_TUR)
tur.2015 <- pisa2015.mean(variable="AGE",data=pisa_2015_stu_TUR)
tur <- rbind(tur.2003,tur.2006,tur.2009,tur.2012,tur.2015)
# OECD
oecd.2003 <- pisa.mean(variable="AGE",data=pisa_2003_stu_OECD)
oecd.2006 <- pisa.mean(variable="AGE",data=pisa_2006_stu_OECD)
oecd.2009 <- pisa.mean(variable="AGE",data=pisa_2009_stu_OECD)
oecd.2012 <- pisa.mean(variable="AGE",data=pisa_2012_stu_OECD)
oecd.2015 <- pisa2015.mean(variable="AGE",data=pisa_2015_stu_OECD)
oecd <- rbind(oecd.2003,oecd.2006,oecd.2009,oecd.2012,oecd.2015)
# AB
AB.2003 <- pisa.mean(variable="AGE",data=pisa_2003_stu_AB)
AB.2006 <- pisa.mean(variable="AGE",data=pisa_2006_stu_AB)
AB.2009 <- pisa.mean(variable="AGE",data=pisa_2009_stu_AB)
AB.2012 <- pisa.mean(variable="AGE",data=pisa_2012_stu_AB)
AB.2015 <- pisa2015.mean(variable="AGE",data=pisa_2015_stu_AB)
AB <- rbind(AB.2003,AB.2006,AB.2009,AB.2012,AB.2015)
cbind(tur,oecd,AB)## Freq Mean s.e. SD s.e Freq Mean s.e. SD s.e Freq Mean s.e.
## 1 4855 15.91 0.00 0.29 0 223330 15.80 0 0.29 0 107861 15.78 0
## 2 4942 15.90 0.01 0.29 0 251262 15.78 0 0.30 0 152305 15.77 0
## 3 4996 15.82 0.00 0.28 0 298454 15.77 0 0.29 0 179830 15.79 0
## 4 4848 15.81 0.00 0.28 0 295330 15.78 0 0.29 0 180379 15.79 0
## 5 5895 15.82 0.00 0.28 0 248620 15.80 0 0.29 0 162386 15.81 0
## SD s.e
## 1 0.29 0
## 2 0.29 0
## 3 0.29 0
## 4 0.29 0
## 5 0.29 0Okul disinda Diger Ogrenim Aktiviteleri icin harcanan haftalik toplam sure
# TURKIYE
tur.2012 <- pisa.mean(variable="OUTHOURS",data=pisa_2012_stu_TUR)
tur.2015 <- pisa2015.mean(variable="OUTHOURS",data=pisa_2015_stu_TUR)
tur <- rbind(tur.2012,tur.2015)
# OECD
oecd.2012 <- pisa.mean(variable="OUTHOURS",data=pisa_2012_stu_OECD,by="CNT")
oecd.2015 <- pisa2015.mean(variable="OUTHOURS",data=pisa_2015_stu_OECD,by="CNT")
# AB
AB.2012 <- pisa.mean(variable="OUTHOURS",data=pisa_2012_stu_AB,,by="CNT")
AB.2015 <- pisa2015.mean(variable="OUTHOURS",data=pisa_2015_stu_AB,,by="CNT")
tur <- rbind(tur,tur,tur)
tur$tip <- c(rep("TUR",2),rep("OECD",2),rep("AB",2))
tur$year <- 1:2
tur[3:4,2]=c(mean(oecd.2012[,3]),mean(oecd.2015[,3]))
tur[3:4,3]=0
tur[5:6,2]=c(mean(AB.2012[,3]),mean(AB.2015[,3],na.rm=TRUE))
tur[5:6,3]=0
ggplot(tur, aes(x=year, y=Mean,shape=tip)) +
theme_bw() +
geom_point(size=2, fill="black") +
geom_line()+
scale_x_discrete(limit = 1:2,labels=c("2012","2015")) +
scale_y_continuous(limit = c(5,30)) +
labs(title = "Okul Disinda Diger Tur Ogrenim Aktiviteleri Icin Harcanan Haftalik Toplam Sure",
x = "YIL", y = "Saat",
shape=" ")+
theme(axis.title= element_text(size = 18),
axis.text= element_text(size = 15),
title = element_text(size = 20),
legend.justification=c(-0.5,-0.2),
legend.position=c(0,0),
legend.text=element_text(size = 18)
) +
annotation_custom(grob = textGrob("@pisa_turkiye"),
xmin = 2.5, xmax = 2.5, ymin = 25, ymax = 25)
Okul Turleri
okul.2003 <- pisa.table(variable="PROGN",data=pisa_2003_stu_TUR)## Warning in `levels<-`(`*tmp*`, value = if (nl == nL) as.character(labels)
## else paste0(labels, : duplicated levels in factors are deprecated okul.2006 <- pisa.table(variable="PROGN",data=pisa_2006_stu_TUR)## Warning in `levels<-`(`*tmp*`, value = if (nl == nL) as.character(labels)
## else paste0(labels, : duplicated levels in factors are deprecated okul.2009 <- pisa.table(variable="PROGN",data=pisa_2009_stu_TUR)## Warning in `levels<-`(`*tmp*`, value = if (nl == nL) as.character(labels)
## else paste0(labels, : duplicated levels in factors are deprecated okul.2012 <- pisa.table(variable="progn",data=pisa_2012_stu_TUR)
okul.2015 <- pisa2015.table(variable="PROGN",data=pisa_2015_stu_TUR)
colnames(okul.2012)[1] <- "PROGN"
okul.2003## PROGN Freq Percentage
## 201 TUR: Primary education (lower sec.) 119 5.23
## 202 TUR: General high school (upper sec.) 2414 51.21
## 203 TUR: Anatolian high school (upper sec.) 200 5.63
## 204 TUR: High school with foreign language (upper sec.) 624 10.76
## 205 TUR: Science high schools (upper sec.) 63 1.75
## 206 TUR: Vocational high schools 619 15.65
## 207 TUR: Anatolian vocational high schools 435 7.09
## 208 TUR: Technical high schools 123 1.39
## 209 TUR: Anatolian technical high schools 258 1.28
## Std.err.
## 201 1.84
## 202 3.98
## 203 2.36
## 204 1.69
## 205 1.13
## 206 2.65
## 207 1.86
## 208 0.88
## 209 0.40 okul.2006## PROGN Freq
## 312 TUR: PRIMARY EDUCATION 116
## 313 TUR: GENERAL HIGH SCHOOL 2266
## 314 TUR: ANATOLIAN HIGH SCHOOL 549
## 315 TUR: HIGH SCHOOL WITH INTENSIVE FOREIGN LANGUAGE TEACHING 9
## 316 TUR: SCIENCE HIGH SCHOOLS 35
## 317 TUR: VOCATIONAL HIGH SCHOOLS 1510
## 318 TUR: ANATOLIAN VOCATIONAL HIGH SCHOOLS 179
## 319 TUR: SECONDARY AND VOCATIONAL HIGH SCHOOL 278
## Percentage Std.err.
## 312 5.33 1.16
## 313 40.24 2.82
## 314 12.81 2.59
## 315 0.21 0.21
## 316 0.77 0.77
## 317 29.45 2.24
## 318 4.05 1.61
## 319 7.14 1.79 okul.2009## PROGN Freq Percentage Std.err.
## 379 TUR: Primary school 137 4.20 0.85
## 380 TUR: General high school 1877 36.37 2.02
## 381 TUR: Anatolian high school 715 13.76 1.77
## 382 TUR: Science high school 100 2.37 1.19
## 384 TUR: Anatolian Teacher Training High School 67 1.35 0.96
## 385 TUR: Anatolian Fine Arts High School 32 0.67 0.67
## 387 TUR: Vocational High School 1254 25.14 1.61
## 388 TUR: Anatolian Vocational High School 356 7.26 1.47
## 389 TUR: Technical High School 53 1.08 0.19
## 390 TUR: Anatolian Technical High Schoo 137 2.79 0.51
## 391 TUR: Multi Programme High School 268 5.02 1.73 okul.2012## PROGN Freq Percentage
## 316 Turkey: Primary school 120 2.66
## 317 Turkey: General high school 1462 30.74
## 318 Turkey: Anatolian high school 1050 22.47
## 319 Turkey: Science high school 35 0.73
## 320 Turkey: Social Sciences High School 35 0.76
## 321 Turkey: Anatolian Teacher Training High School 207 4.47
## 323 Turkey: Vocational High School 1216 24.67
## 324 Turkey: Anatolian Vocational High School 279 5.73
## 325 Turkey: Technical High School 75 1.53
## 326 Turkey: Anatolian Technical High School 123 2.47
## 327 Turkey: Multi Programme High School 178 3.73
## 328 Turkey: Police High School 68 0.04
## Std.err.
## 316 0.40
## 317 2.68
## 318 2.25
## 319 0.73
## 320 0.54
## 321 1.55
## 323 1.79
## 324 1.23
## 325 0.33
## 326 0.58
## 327 1.56
## 328 0.00 okul.2015## PROGN Freq Percentage
## 417 Turkey: Basic Education 121 3.22
## 418 Turkey: General Secondary Education 3241 55.75
## 419 Turkey: Vocational and Technical Secondary Education 2533 41.03
## Std.err.
## 417 0.48
## 418 1.92
## 419 1.94