HAFTA_3_Calisma_Gunlugu
Asli_Ece_Kocak
2025-03-09
HAFTA 3 - OGRENME GUNLUGU: UC DEGERLER (OUTLIERS)
R PAKETLERI
VERI SETI
- Bu calisma gunlugu icin PISA 2022 USA ogrenci anketi veri setinde yer alan bazi degiskenler secilerek 7, 8,9, 95, 97, 99 gibi degerler kayip veri gibi davrandigi kabul edilerek → derste islenen SCREEN datasina benzer bir veri seti elde edilmistir.
- Veri seti yukleme ve veri setindeki degiskenlerin “etiketlerini” kaldirmak icin → “expss” paketinde yer alan → “drop_var_labs()” fonksiyonu kullanma:
orj_USA <- read_sav("C:/Users/User/Desktop/USA.SAV")
orj_USA <- expss::drop_var_labs(orj_USA)
head(orj_USA)## # A tibble: 6 × 34
## CNT CNTSTUID AGE GRADE IMMIG LANGN REPEAT MISSSC SKIPPING TARDYSD EXERPRAC
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 USA 84000002 15.4 0 1 313 0 0 0 0 10
## 2 USA 84000003 15.5 0 1 313 0 0 1 0 10
## 3 USA 84000004 16.1 0 1 313 0 0 0 0 5
## 4 USA 84000005 15.6 0 1 313 0 0 1 0 0
## 5 USA 84000006 16.3 1 1 313 0 0 0 2 5
## 6 USA 84000008 15.6 0 2 313 0 1 0 0 4
## # ℹ 23 more variables: STUDYHMW <dbl>, WORKPAY <dbl>, WORKHOME <dbl>,
## # EXPECEDU <dbl>, MATHPREF <dbl>, MATHEASE <dbl>, MATHMOT <dbl>,
## # DURECEC <dbl>, SISCO <dbl>, MISCED <dbl>, FISCED <dbl>, HISCED <dbl>,
## # PAREDINT <dbl>, BMMJ1 <dbl>, BFMJ2 <dbl>, HISEI <dbl>, ICTRES <dbl>,
## # HOMEPOS <dbl>, ESCS <dbl>, FCFMLRTY <dbl>, ICTAVSCH <dbl>, ICTHOME <dbl>,
## # ICTAVHOM <dbl>- Excel ile benim tarafimdan temizlenmeye calisilmis USA datasi “.csv” formatinda yuklenerek →calisma icin en uygun olacak sekilde umarim kullanilacak hale getirilmesi:
## CNT CNTSTUID AGE GRADE IMMIG LANGN REPEAT MISSSC SKIPPING TARDYSD EXERPRAC
## 1 USA 84000002 15.42 0 1 313 0 0 0 0 10
## 2 USA 84000003 15.50 0 1 313 0 0 1 0 10
## 3 USA 84000004 16.08 0 1 313 0 0 0 0 5
## 4 USA 84000005 15.58 0 1 313 0 0 1 0 0
## 5 USA 84000006 16.33 1 1 313 0 0 0 2 5
## 6 USA 84000008 15.58 0 2 313 0 1 0 0 4
## STUDYHMW WORKPAY WORKHOME EXPECEDU MATHPREF MATHEASE MATHMOT DURECEC SISCO
## 1 2 0 2 9 0 0 0 2 1
## 2 6 0 0 4 0 0 0 NA 0
## 3 9 0 5 8 1 0 0 2 1
## 4 3 10 10 4 0 0 0 3 1
## 5 1 5 0 9 0 0 0 2 1
## 6 10 0 0 9 NA 0 NA 4 0
## MISCED FISCED HISCED PAREDINT BMMJ1 BFMJ2 HISEI ICTRES HOMEPOS ESCS
## 1 7 9 9 16 NA 79.05 79.05 0.5500 1.1179 1.2582
## 2 6 6 6 12 73.91 59.89 73.91 0.4946 0.7300 0.3488
## 3 9 9 9 16 67.94 82.41 82.41 1.0020 1.1761 1.3463
## 4 5 3 5 12 24.98 NA 24.98 -0.7480 -0.9389 -1.3108
## 5 5 5 5 12 NA 16.50 16.50 1.7606 0.2333 -0.4500
## 6 10 3 10 16 85.85 51.92 85.85 1.4757 0.4713 1.1127
## FCFMLRTY ICTAVSCH ICTHOME ICTAVHOM
## 1 11 7 0.3346 6
## 2 9 7 0.3346 6
## 3 15 7 0.3346 6
## 4 10 7 0.3346 6
## 5 10 7 0.3346 6
## 6 0 3 -1.5118 5- Eksik veri duzenlemesi icin:
USA <- USA %>%
mutate(IMMIG = ifelse(is.na(IMMIG), mean(IMMIG, na.rm=T), IMMIG)) %>% na.omit()
summary(USA) ## CNT CNTSTUID AGE GRADE
## Length:1571 Min. :84000004 Min. :15.33 Min. :-2.0000
## Class :character 1st Qu.:84002126 1st Qu.:15.58 1st Qu.: 0.0000
## Mode :character Median :84004153 Median :15.83 Median : 0.0000
## Mean :84004109 Mean :15.83 Mean : 0.1337
## 3rd Qu.:84006116 3rd Qu.:16.08 3rd Qu.: 0.0000
## Max. :84008157 Max. :16.33 Max. : 2.0000
## IMMIG LANGN REPEAT MISSSC
## Min. :1.000 Min. :156.0 Min. :0.0000 Min. :0.00000
## 1st Qu.:1.000 1st Qu.:313.0 1st Qu.:0.0000 1st Qu.:0.00000
## Median :1.000 Median :313.0 Median :0.0000 Median :0.00000
## Mean :1.239 Mean :316.2 Mean :0.0471 Mean :0.04583
## 3rd Qu.:1.000 3rd Qu.:313.0 3rd Qu.:0.0000 3rd Qu.:0.00000
## Max. :3.000 Max. :859.0 Max. :1.0000 Max. :1.00000
## SKIPPING TARDYSD EXERPRAC STUDYHMW
## Min. :0.0000 Min. :0.0000 Min. : 0.000 Min. : 0.000
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.: 2.000 1st Qu.: 3.000
## Median :0.0000 Median :0.0000 Median : 5.000 Median : 5.000
## Mean :0.3514 Mean :0.5086 Mean : 4.745 Mean : 4.959
## 3rd Qu.:1.0000 3rd Qu.:1.0000 3rd Qu.: 8.000 3rd Qu.: 7.000
## Max. :1.0000 Max. :2.0000 Max. :10.000 Max. :10.000
## WORKPAY WORKHOME EXPECEDU MATHPREF
## Min. : 0.000 Min. : 0.000 Min. :2.000 Min. :0.0000
## 1st Qu.: 0.000 1st Qu.: 1.000 1st Qu.:6.000 1st Qu.:0.0000
## Median : 0.000 Median : 4.000 Median :7.000 Median :0.0000
## Mean : 1.282 Mean : 4.122 Mean :7.073 Mean :0.1693
## 3rd Qu.: 2.000 3rd Qu.: 7.000 3rd Qu.:8.000 3rd Qu.:0.0000
## Max. :10.000 Max. :10.000 Max. :9.000 Max. :1.0000
## MATHEASE MATHMOT DURECEC SISCO
## Min. :0.0000 Min. :0.0000 Min. :0.000 Min. :0.0000
## 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:2.000 1st Qu.:1.0000
## Median :0.0000 Median :0.0000 Median :2.000 Median :1.0000
## Mean :0.1254 Mean :0.0261 Mean :2.266 Mean :0.8351
## 3rd Qu.:0.0000 3rd Qu.:0.0000 3rd Qu.:3.000 3rd Qu.:1.0000
## Max. :1.0000 Max. :1.0000 Max. :6.000 Max. :1.0000
## MISCED FISCED HISCED PAREDINT
## Min. : 1.00 Min. : 1.000 Min. : 1.000 Min. : 3.00
## 1st Qu.: 5.00 1st Qu.: 5.000 1st Qu.: 6.000 1st Qu.:12.00
## Median : 8.00 Median : 7.000 Median : 8.000 Median :16.00
## Mean : 7.17 Mean : 6.751 Mean : 7.622 Mean :14.62
## 3rd Qu.: 9.00 3rd Qu.: 9.000 3rd Qu.: 9.000 3rd Qu.:16.00
## Max. :10.00 Max. :10.000 Max. :10.000 Max. :16.00
## BMMJ1 BFMJ2 HISEI ICTRES
## Min. :11.56 Min. :11.01 Min. :11.56 Min. :-2.5500
## 1st Qu.:30.90 1st Qu.:25.95 1st Qu.:47.83 1st Qu.:-0.2720
## Median :59.18 Median :51.50 Median :68.70 Median : 0.3344
## Mean :53.03 Mean :48.52 Mean :60.87 Mean : 0.3359
## 3rd Qu.:70.50 3rd Qu.:70.57 3rd Qu.:76.49 3rd Qu.: 0.8511
## Max. :88.70 Max. :88.96 Max. :88.96 Max. : 4.1742
## HOMEPOS ESCS FCFMLRTY ICTAVSCH
## Min. :-2.5965 Min. :-3.3280 Min. : 0.000 Min. :0.000
## 1st Qu.:-0.2490 1st Qu.:-0.1888 1st Qu.: 4.000 1st Qu.:7.000
## Median : 0.3100 Median : 0.5272 Median : 7.000 Median :7.000
## Mean : 0.3315 Mean : 0.3416 Mean : 7.181 Mean :6.683
## 3rd Qu.: 0.8680 3rd Qu.: 0.9649 3rd Qu.:10.000 3rd Qu.:7.000
## Max. : 4.9111 Max. : 2.6262 Max. :16.000 Max. :7.000
## ICTHOME ICTAVHOM
## Min. :-6.3292 Min. :0.000
## 1st Qu.: 0.3346 1st Qu.:6.000
## Median : 0.3346 Median :6.000
## Mean : 0.0949 Mean :5.796
## 3rd Qu.: 0.3346 3rd Qu.:6.000
## Max. : 0.3456 Max. :6.000UC DEGERLER (OUTLIERS)
- Veri setinde outliers bulunma nedenleri:
- Veri dosyasina isleme hatasi
- Yanlis kodlama
- Evren-orneklem uyusmazligi
- Orneklemin evrenden cekilmis olmasina ragmen → NORMAL DAGILIMa uygun olmayan → uc/aykiri/asiri degerlere sahip olmasi
- Tek Degiskenli vs. Cok Degiskenli Uc Deger → farkli
UC DEGERLERIN BELIRLENMESI
Kategorik Degiskenlerde Uc Deger
- MISSSC DEGISKENI:
## ### Frequencies
## #### USA$MISSSC
## **Type:** Integer
##
## | | Freq | % | % Cum. |
## |----------:|-----:|--------:|--------:|
## | **0** | 1499 | 95.417 | 95.417 |
## | **1** | 72 | 4.583 | 100.000 |
## | **Total** | 1571 | 100.000 | 100.000 |- Tabloda USA veri setindeki “MISSSC” degiskenine ait
frekans dagilimi hesaplanarak → Markdown formatinda raporlandi.
- MISSSC degiskeni 2 kategoriye sahip → 0 ve 1
- 1499 gozlem (%95.417) → 0 kategorisinde → BASKIN kategori
- 72 gozlem (%4.583) → 1 kategorisinde
- Buna gore → MISSSC degiskeninde bir taraf yaklasik %95 iken diger taraf yaklasik %5 oldugu icin → iyi temsil EDILMIYOR!
- SKIPPING DEGISKENI:
library(kableExtra)
freq(USA$SKIPPING, report.nas = F) %>%
kable(format='markdown',
caption="Frekans Tablosu", digits = 3) %>%
kable_styling(full_width = T, font_size = 14, bootstrap_options = "striped") %>%
row_spec(0, background = "white", color = "black")| Freq | % Valid | % Valid Cum. | % Total | % Total Cum. | |
|---|---|---|---|---|---|
| 0 | 1019 | 64.863 | 64.863 | 64.863 | 64.863 |
| 1 | 552 | 35.137 | 100.000 | 35.137 | 100.000 |
| <NA> | 0 | NA | NA | 0.000 | 100.000 |
| Total | 1571 | 100.000 | 100.000 | 100.000 | 100.000 |
- SKIPPING degiskenine ait dagilim incelendiginde:
- 0 degeri %65 iken 1 degeri %35 oldugu icin (65/35 orani) → KABUL edilebilir temsil durumu vardir.
Surekli Degiskenlerde Uc Deger
Tek Degiskenli Uc Deger Belirleme
- Degiskene ait butun degerlerin ORTALAMA 0 ve STANDART SAPMA 1 olacak
sekilde → standartlastirmak icin z puanlarina donusturme
- TEK degiskenli uc degerler → COK BUYUK z puanlarina sahiptir.
- outliers paketinde yer alan → scores() fonksiyonu ile → z degerleri hesaplama:
## BMMJ1 BFMJ2 HISEI
## 1 0.67 1.46 1.08
## 2 0.78 -0.90 0.48
## 3 0.94 1.09 0.65
## 4 0.77 0.96 0.50
## 5 -1.26 -0.63 -1.36
## 6 -0.98 -0.57 -1.28BMMJ1, BMMJ2 ve HISEI degiskenleri icin (25:27) secilerek bu degiskenler incelenmistir.
summarytools paketinde yer alan → descr() fonksiyonu ile z degerlerinin → en dusuk ve en yuksek degerlerini hesaplama:
##
## Min Max
## ----------- ------- ------
## BFMJ2 -1.61 1.74
## BMMJ1 -1.86 1.60
## HISEI -2.47 1.41- Elde edilen degerlere gore uc deger bulunmamaktadir → cünkü → en DUSUK ve en YUKSEK degerler -3 ile +3 araligindadir.
NOTE: Veri setinde bir sorun oldugunu fark ederek USA veri setinin icerisine farkli surekli degiskenler eklenmesi gerektigini anliyorum :(
Veri Setine Surekli Degisken Ekleme
set.seed(123)
n <- nrow(USA)
USA$NEW_VAR <- runif(n, min = 0, max = 200)
outliers <- c(-150, 450)
USA$NEW_VAR[sample(1:n, 2)] <- outliers
summary(USA$NEW_VAR)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -150.00 49.97 97.82 99.21 147.98 450.00set.seed(123)
n <- nrow(USA)
USA$NEW_VAR_2 <- runif(n, min = 0, max = 300)
outliers <- c(-200, 550)
USA$NEW_VAR_2[sample(1:n, 2)] <- outliers
summary(USA$NEW_VAR_2)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -200.00 74.95 146.73 148.76 221.98 550.00## CNT CNTSTUID AGE GRADE IMMIG LANGN REPEAT MISSSC SKIPPING TARDYSD EXERPRAC
## 3 USA 84000004 16.08 0 1 313 0 0 0 0 5
## 8 USA 84000011 16.08 0 3 313 0 0 0 0 8
## 9 USA 84000014 16.25 1 1 313 0 0 0 1 10
## 10 USA 84000015 15.83 0 1 313 0 0 1 1 8
## 12 USA 84000019 16.25 0 1 313 1 0 0 0 5
## 13 USA 84000021 15.83 0 1 313 0 1 1 1 0
## STUDYHMW WORKPAY WORKHOME EXPECEDU MATHPREF MATHEASE MATHMOT DURECEC SISCO
## 3 9 0 5 8 1 0 0 2 1
## 8 10 0 5 9 0 0 0 1 1
## 9 10 10 10 8 0 0 0 3 1
## 10 1 2 0 7 0 0 0 1 0
## 12 6 4 6 4 0 0 0 1 1
## 13 0 3 6 6 0 0 0 3 1
## MISCED FISCED HISCED PAREDINT BMMJ1 BFMJ2 HISEI ICTRES HOMEPOS ESCS
## 3 9 9 9 16.0 67.94 82.41 82.41 1.0020 1.1761 1.3463
## 8 9 8 9 16.0 70.50 27.52 70.50 0.2867 0.2929 0.7473
## 9 10 10 10 16.0 73.91 73.91 73.91 2.2545 1.1559 1.1772
## 10 8 9 9 16.0 70.10 70.89 70.89 -0.3588 0.6069 0.8877
## 12 5 5 5 12.0 25.04 33.76 33.76 0.7541 0.4548 -0.5548
## 13 6 7 7 14.5 31.08 35.34 35.34 0.8701 0.0469 -0.2746
## FCFMLRTY ICTAVSCH ICTHOME ICTAVHOM NEW_VAR NEW_VAR_2
## 3 15 7 0.3346 6 57.51550 86.27326
## 8 12 7 0.3346 6 157.66103 236.49154
## 9 11 7 0.3346 6 81.79538 122.69308
## 10 5 7 -0.3881 5 176.60348 264.90522
## 12 8 7 0.3346 6 188.09346 282.14019
## 13 7 7 0.3346 6 9.11130 13.66695## NEW_VAR NEW_VAR_2
## 1 -0.72 -0.72
## 2 1.01 1.01
## 3 -0.30 -0.30
## 4 1.33 1.34
## 5 1.53 1.54
## 6 -1.55 -1.56##
## Min Max
## --------------- ------- ------
## NEW_VAR -4.29 6.04
## NEW_VAR_2 -4.03 4.63- NEW_VAR ve NEW_VAR_2 sayesinde artik uc degerlere sahip surekli degisken elde ettim.
Frekans Tablosu
- DT paketinde yer alan → datatable() fonksiyonu:
DT::datatable(z.scores_USA_2,
options = list(pageLength = 5,
scrollX = T,
searching = T,
autoWidth = F))- Yukaridaki tablo ile → 2 surekli degiskene ait → z puanlarini gorebiliyorum.
Histogram Grafigi
- ggplot2 paketi ile HISTOGRAM GRAFIGI elde etme:
ggplot2::ggplot(USA, aes(x = NEW_VAR)) +
geom_histogram(bins = 30L, fill = "red", color = "black") +
theme_minimal() +
xlim(min(USA$NEW_VAR), max(USA$NEW_VAR)) +
coord_cartesian(expand = T) +
labs(title = "NEW_VAR Histogramı", x = "NEW_VAR Degerleri", y = "Frekans") +
theme(plot.title = element_text(hjust = 0.5, size = 14))
- NEW_VAR degiskeninin → ortalamasini elde ederek → grafige eklenmesi icin:
## [1] 99.21278- Grafik ile ORTALAMANIN gosterilmesi:
ggplot2::ggplot(USA, aes(x = NEW_VAR)) +
geom_histogram(bins = 30, fill = "lightblue", color = "black", alpha = 0.7) +
geom_vline(xintercept = 99.21278, color = "red", linetype = "dashed", size = 1) +
annotate("text", label = "Ort = 99.21278", x = 12, y = max(table(cut(USA$NEW_VAR, breaks = 30))),
color ="black", size = 5, fontface = "bold") +
theme_minimal() +
labs(title = "NEW_VAR Degiskeninin Histogram Grafigi",
subtitle = "Ortalama 99.21278 olarak isaretlendi",
x = "NEW_VAR Degerleri",
y = "Frekans") +
theme(plot.title = element_text(hjust = 0.5, size = 14, face = "bold"),
plot.subtitle = element_text(hjust = 0.5, size = 12),
axis.text = element_text(size = 12),
axis.title = element_text(size = 13))
Yogunluk Grafigi
ggplot2::ggplot(USA, aes(x = NEW_VAR)) +
geom_histogram(aes(y = ..density..), bins = 30, fill = "lightblue", color = "black", alpha = 0.6) +
geom_density(alpha = 0.5, fill = "red", color = "darkred", size = 1.2) +
geom_vline(xintercept = mean(USA$NEW_VAR, na.rm = TRUE), color = "blue",
linetype = "dashed", size = 1) +
annotate("text", label = paste("Ortalama =", round(mean(USA$NEW_VAR, na.rm = T), 2)),
x = mean(USA$NEW_VAR, na.rm = T) + 5,
y = max(density(USA$NEW_VAR, na.rm = T)$y) * 0.9,
color = "blue", size = 5, fontface = "bold") +
labs(title = "NEW_VAR Degiskeninin Histogrami ve Yogunluk Grafigi",
subtitle = "Histogram (lightblue) ve Yogunluk Egrisi (red)",
x = "NEW_VAR Degerleri",
y = "Yogunluk") +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5, size = 14, face = "bold"),
plot.subtitle = element_text(hjust = 0.5, size = 12),
axis.text = element_text(size = 12),
axis.title = element_text(size = 13))
Interaktif Grafik
plot_ly(
x = USA$NEW_VAR,
type = "histogram",
histnorm = "probability",
nbinsx = 30,
marker = list(color = "red", line = list(color = "black", width = 1.2))
) %>%
layout(
title = "NEW_VAR Degiskeninin Histogrami",
xaxis = list(title = "NEW_VAR Degerleri", showgrid = F),
yaxis = list(title = "Olasilik", showgrid = F),
plot_bgcolor = "white"
)Kutu Grafigi
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## -150.00 49.97 97.82 99.21 147.98 450.00ggplot(USA, aes(x = "", y = NEW_VAR)) +
geom_boxplot(fill = "lightblue", color = "black", outlier.color = "red",
outlier.shape = 16, outlier.size = 3, na.rm = T) +
stat_summary(fun = mean, geom = "point", shape = 18, size = 4,
color = "blue", fill = "blue", na.rm = T) +
theme_minimal() +
labs(title = "NEW_VAR Degiskeninin Boxplot Grafigi",
subtitle = "Kutu grafigi ile dagilimin gorsellestirilmesi",
y = "NEW_VAR Degerleri") +
theme(plot.title = element_text(hjust = 0.5, size = 14, face = "bold"),
plot.subtitle = element_text(hjust = 0.5, size = 12),
axis.text = element_text(size = 12),
axis.title = element_text(size = 13))
- boxplot.stats() fonksiyonu → out bileseni → UC DEGERLERI verir:
## [1] -150 450## [1] 403 978## CNT CNTSTUID AGE GRADE IMMIG LANGN REPEAT MISSSC SKIPPING TARDYSD
## 1182 USA 84002165 15.67 0 1 313 0 0 0 0
## 2858 USA 84005114 15.50 0 2 859 0 0 0 0
## EXERPRAC STUDYHMW WORKPAY WORKHOME EXPECEDU MATHPREF MATHEASE MATHMOT
## 1182 5 2 0 2 9 0 0 0
## 2858 3 5 0 5 7 0 0 0
## DURECEC SISCO MISCED FISCED HISCED PAREDINT BMMJ1 BFMJ2 HISEI ICTRES
## 1182 1 1 8 8 8 16 70.34 62.39 70.34 -0.6000
## 2858 2 1 8 7 8 16 47.83 57.37 57.37 0.7374
## HOMEPOS ESCS FCFMLRTY ICTAVSCH ICTHOME ICTAVHOM NEW_VAR NEW_VAR_2
## 1182 0.2592 0.7300 6 7 0.3346 6 -150 -200
## 2858 0.7674 0.7004 8 7 0.3346 6 450 550- USA veri setinde NEW_VAR degiskeni icin 2 adet uc
deger var:
- Bu uc degerler sirasi ile → SATIR 1182 ve 2858’de yer almaktadir.
- CNTSTUID (Ogrenci_Kimligi) → 84002165 ve 84005114
- AGE (Yas) → 15.67 ve 15.50
- IMMIG (Gocmenlik Durumu) → 1 ve 2
- LANGN (Dil Kodu) → 313 ve 859
- Bu 2 uc degerlerin NEW_VAR degiskeninde hangi gozlemlerde/bireylerde oldugunu belirleme:
## [1] 403 978Interaktif Kutu Grafigi
plot_ly(y = ~USA$NEW_VAR, type = 'box',
boxpoints = "all",
jitter = 0.3,
pointpos = -1.8,
marker = list(color = 'red', size = 5, opacity = 0.6),
line = list(color = 'black')) %>%
layout(title = "NEW_VAR Degiskeninin Boxplot Grafigi",
yaxis = list(title = "NEW_VAR Degerleri"),
plot_bgcolor = "white")- Interaktif kutu grafiginde gorulen uc degerlerin IDlerinin isaretlenmesi:
out_ind <- which(USA$NEW_VAR %in% outliers)
plot_ly(y = ~USA$NEW_VAR, type = 'box',
boxpoints = "all", jitter = 0.3, pointpos = -1.5,
marker = list(color = "red", size = 6, opacity = 0.6),
line = list(color = "black")) %>%
layout(title = "NEW_VAR Degiskeninin Boxplot Grafigi",
yaxis = list(title = "NEW_VAR Degerleri"),
plot_bgcolor = "white",
annotations = lapply(1:length(outliers), function(i) {
list(
x = -0.2,
y = outliers[i],
text = paste("İndeks:", out_ind[i]),
showarrow = F,
xanchor = "right",
font = list(size = 10, color = "blue")
)
})
)NEW_VAR Degiskeninin NEW_VAR_2 Degiskenine gore Incelenmesi
ggplot(USA, aes(x = factor(NEW_VAR_2),
y = NEW_VAR,
fill = factor(NEW_VAR_2))) +
geom_boxplot(outlier.color = "red", outlier.shape = 16, outlier.size = 3) +
scale_fill_brewer(palette = "Set2") +
theme_minimal() +
labs(title = "NEW_VAR Degiskeninin Kategorik Boxplot Grafigi",
subtitle = "NEW_VAR_2 degiskenine gore gruplanmis",
x = "NEW_VAR_2 Kategorileri",
y = "NEW_VAR Degerleri",
fill = "NEW_VAR_2") +
theme(plot.title = element_text(hjust = 0.5, size = 14, face = "bold"),
plot.subtitle = element_text(hjust = 0.5, size = 12),
axis.text = element_text(size = 12),
axis.title = element_text(size = 13),
legend.position = "top")
NEW_VAR Degiskeninin MISCED Degiskenine gore Incelenmesi
## [1] 7.169955ggplot(USA, aes(x = MISCED)) +
geom_histogram(bins = 10, fill = "#0c4c8a", color = "white", alpha = 0.8) +
geom_vline(aes(xintercept = mean(MISCED, na.rm = T)),
color = "red", linetype = "dashed", size = 1) +
annotate("text", x = mean(USA$MISCED, na.rm = T) + 1, y = 10,
label = paste("Ortalama:", round(mean(USA$MISCED, na.rm = T), 2)),
color = "red", size = 5, fontface = "bold") +
theme_minimal() +
labs(title = "MISCED Degiskeninin Histogrami",
subtitle = "Bireylerin anne egitim seviyesinin dagilimi",
x = "MISCED (Anne Egitim Seviyesi)",
y = "Frekans") +
theme(plot.title = element_text(hjust = 0.5, size = 14, face = "bold"),
plot.subtitle = element_text(hjust = 0.5, size = 12),
axis.text = element_text(size = 12),
axis.title = element_text(size = 13))
MAHALANOBIS UZAKLIGI
Mahalanobis Uzakligi Hesaplama:
library(psych)
veri <- USA[,3:13]
md <- mahalanobis(veri, center = colMeans(veri), cov = cov(veri))
head(md, 20)## 3 8 9 10 12 13 14 23
## 5.207553 18.846258 20.993176 5.879155 24.881852 25.627941 1.669207 8.206659
## 30 31 32 33 36 39 40 43
## 6.919917 3.857980 3.342804 4.997115 28.018506 10.388325 9.631875 4.049845
## 48 51 52 54
## 13.647818 6.397026 1.283293 7.905329Mahalanobis Uzakligi Kritik Deger Belirleme
## [1] 31.26413- Mahalonobis uzakligi degerleri ki-kare ile
değerlendirilir.
- Serbestlik derecesi BAGIMSIZ degisken sayisina ESIT
- 31.26413 kritik degerinden BUYUK olan degerler 0.001 alfa duzeyinde istatistiksel olarak MANIDAR seklinde degerlendirilmektedir.
## AGE GRADE IMMIG LANGN REPEAT MISSSC SKIPPING TARDYSD EXERPRAC
## 129 16.17 0 1.000000 313 1 0 0 2 10
## 206 15.58 0 3.000000 859 0 0 1 0 7
## 218 15.83 -1 1.000000 313 1 1 1 2 0
## 226 15.50 0 2.000000 156 1 1 1 2 4
## 249 16.25 1 3.000000 859 0 0 0 1 5
## 336 15.42 -1 1.000000 859 0 1 1 1 0
## 400 15.50 -1 1.000000 313 0 1 1 0 10
## 426 15.75 0 2.000000 156 0 1 0 1 1
## 537 15.50 0 2.000000 156 0 1 0 1 0
## 659 15.42 -2 1.000000 313 1 0 1 2 10
## 669 15.42 0 3.000000 859 0 0 1 0 3
## 830 15.50 0 3.000000 156 0 0 1 2 4
## 908 15.92 0 2.000000 156 1 0 0 1 4
## 967 15.58 -1 1.000000 313 1 0 0 2 10
## 1010 15.42 0 3.000000 859 0 0 0 0 0
## 1013 15.58 0 3.000000 859 0 1 1 0 4
## 1065 16.17 0 1.000000 156 0 1 0 2 2
## 1366 16.17 0 1.000000 313 1 0 0 0 10
## 1470 15.50 -1 3.000000 156 1 0 1 1 8
## 1576 15.42 -1 1.000000 313 1 1 0 1 3
## 1641 16.33 1 2.000000 156 0 1 0 0 5
## 1724 16.17 0 1.000000 156 0 1 1 2 10
## 1749 16.17 -1 1.000000 313 1 0 0 2 5
## 1756 16.08 0 1.284064 859 0 0 0 0 10
## 1764 16.08 1 3.000000 156 0 0 1 0 4
## 1844 15.67 0 1.000000 156 1 1 0 0 10
## 1970 16.00 0 2.000000 859 0 1 1 2 10
## 1987 16.08 0 3.000000 859 0 0 1 2 5
## 2004 16.00 0 2.000000 156 1 1 1 2 10
## 2084 15.42 -1 1.000000 313 1 0 0 0 10
## 2156 15.83 1 3.000000 859 0 0 0 0 5
## 2196 15.42 -1 1.000000 313 1 0 0 2 2
## 2241 16.08 0 2.000000 156 0 1 0 0 0
## 2287 15.67 1 2.000000 313 1 0 0 0 2
## 2371 16.25 1 1.284064 859 0 0 1 0 0
## 2423 15.67 -1 1.000000 313 1 1 1 1 0
## 2439 16.33 0 2.000000 156 1 0 0 0 4
## 2448 16.08 0 2.000000 859 0 0 1 2 4
## 2490 16.25 0 2.000000 156 1 0 0 1 0
## 2581 16.08 -1 1.000000 313 0 0 0 1 10
## 2583 16.33 0 1.000000 313 0 1 0 2 10
## 2599 15.92 -1 1.000000 313 1 0 1 0 10
## 2613 16.17 0 1.000000 313 1 0 0 0 0
## 2631 15.92 -1 1.000000 313 1 1 0 1 10
## 2665 15.42 0 1.000000 313 0 1 0 2 0
## 2689 15.33 0 3.000000 859 0 0 0 2 0
## 2704 15.42 -1 1.000000 313 0 1 1 1 0
## 2730 16.08 0 1.000000 313 0 1 0 1 10
## 2843 16.25 0 1.000000 313 1 0 0 0 0
## 2879 15.50 -1 1.000000 313 1 0 1 2 0
## 2885 16.17 0 1.000000 313 0 1 1 1 0
## 2914 16.08 0 2.000000 859 0 0 0 2 10
## 2998 15.50 -1 3.000000 859 1 0 0 0 5
## 3117 15.50 -1 1.000000 859 0 0 0 2 2
## 3124 16.08 0 3.000000 859 0 0 1 1 0
## 3131 15.75 0 1.000000 859 1 0 1 1 10
## 3272 15.92 0 1.284064 313 1 1 0 0 10
## 3299 16.17 1 2.000000 156 0 1 1 1 10
## 3415 15.67 0 3.000000 859 0 0 0 0 5
## 3440 16.33 0 3.000000 156 1 0 1 0 6
## 3453 16.25 -1 1.000000 313 0 0 0 2 10
## 3488 15.58 0 1.000000 313 0 1 0 1 0
## 3507 15.50 -1 1.000000 313 1 0 0 2 8
## 3517 15.58 0 1.000000 313 1 0 1 0 10
## 3776 15.92 -1 1.000000 313 1 1 1 2 1
## 3799 16.00 1 3.000000 859 0 0 0 1 10
## 3887 15.83 1 3.000000 156 0 1 0 1 8
## 3986 16.17 0 1.000000 313 1 1 0 2 0
## 4001 16.17 0 1.000000 313 0 1 1 2 10
## 4054 16.00 0 1.000000 859 0 0 0 1 0
## 4067 15.92 -1 1.000000 313 1 0 0 1 10
## 4083 16.08 0 2.000000 313 1 1 1 1 8
## 4088 16.33 1 1.000000 313 1 0 1 0 1
## 4147 16.33 0 1.000000 313 0 1 1 2 2
## 4197 15.50 0 1.000000 313 1 1 0 2 8
## 4460 15.92 0 3.000000 859 0 0 0 0 2
## 4492 16.08 0 3.000000 859 0 0 0 1 5
## 4504 16.00 0 1.000000 313 1 1 1 0 1
## 4522 15.58 -1 3.000000 156 0 1 0 2 4
## STUDYHMW WORKPAY
## 129 3 0
## 206 8 0
## 218 0 0
## 226 2 3
## 249 5 0
## 336 2 0
## 400 10 0
## 426 10 0
## 537 5 7
## 659 10 7
## 669 5 0
## 830 9 9
## 908 10 0
## 967 8 0
## 1010 4 0
## 1013 10 0
## 1065 10 5
## 1366 6 10
## 1470 2 0
## 1576 4 0
## 1641 8 1
## 1724 8 0
## 1749 4 2
## 1756 8 10
## 1764 10 10
## 1844 10 10
## 1970 10 10
## 1987 5 0
## 2004 3 10
## 2084 10 10
## 2156 10 0
## 2196 5 0
## 2241 3 0
## 2287 6 0
## 2371 6 6
## 2423 7 0
## 2439 10 0
## 2448 2 8
## 2490 4 4
## 2581 10 10
## 2583 6 2
## 2599 0 0
## 2613 10 0
## 2631 10 0
## 2665 4 0
## 2689 3 0
## 2704 0 0
## 2730 0 0
## 2843 0 8
## 2879 6 0
## 2885 7 9
## 2914 3 0
## 2998 5 0
## 3117 4 3
## 3124 8 0
## 3131 8 4
## 3272 4 8
## 3299 6 0
## 3415 6 0
## 3440 10 0
## 3453 10 0
## 3488 0 10
## 3507 8 6
## 3517 0 10
## 3776 0 10
## 3799 10 6
## 3887 4 4
## 3986 6 3
## 4001 6 6
## 4054 0 10
## 4067 10 10
## 4083 4 0
## 4088 3 2
## 4147 2 0
## 4197 3 2
## 4460 4 0
## 4492 0 0
## 4504 1 1
## 4522 4 0## AGE GRADE IMMIG LANGN REPEAT MISSSC SKIPPING TARDYSD EXERPRAC
## 129 16.17 0 1.000000 313 1 0 0 2 10
## 206 15.58 0 3.000000 859 0 0 1 0 7
## 218 15.83 -1 1.000000 313 1 1 1 2 0
## 226 15.50 0 2.000000 156 1 1 1 2 4
## 249 16.25 1 3.000000 859 0 0 0 1 5
## 336 15.42 -1 1.000000 859 0 1 1 1 0
## 400 15.50 -1 1.000000 313 0 1 1 0 10
## 426 15.75 0 2.000000 156 0 1 0 1 1
## 537 15.50 0 2.000000 156 0 1 0 1 0
## 659 15.42 -2 1.000000 313 1 0 1 2 10
## 669 15.42 0 3.000000 859 0 0 1 0 3
## 830 15.50 0 3.000000 156 0 0 1 2 4
## 908 15.92 0 2.000000 156 1 0 0 1 4
## 967 15.58 -1 1.000000 313 1 0 0 2 10
## 1010 15.42 0 3.000000 859 0 0 0 0 0
## 1013 15.58 0 3.000000 859 0 1 1 0 4
## 1065 16.17 0 1.000000 156 0 1 0 2 2
## 1366 16.17 0 1.000000 313 1 0 0 0 10
## 1470 15.50 -1 3.000000 156 1 0 1 1 8
## 1576 15.42 -1 1.000000 313 1 1 0 1 3
## 1641 16.33 1 2.000000 156 0 1 0 0 5
## 1724 16.17 0 1.000000 156 0 1 1 2 10
## 1749 16.17 -1 1.000000 313 1 0 0 2 5
## 1756 16.08 0 1.284064 859 0 0 0 0 10
## 1764 16.08 1 3.000000 156 0 0 1 0 4
## 1844 15.67 0 1.000000 156 1 1 0 0 10
## 1970 16.00 0 2.000000 859 0 1 1 2 10
## 1987 16.08 0 3.000000 859 0 0 1 2 5
## 2004 16.00 0 2.000000 156 1 1 1 2 10
## 2084 15.42 -1 1.000000 313 1 0 0 0 10
## 2156 15.83 1 3.000000 859 0 0 0 0 5
## 2196 15.42 -1 1.000000 313 1 0 0 2 2
## 2241 16.08 0 2.000000 156 0 1 0 0 0
## 2287 15.67 1 2.000000 313 1 0 0 0 2
## 2371 16.25 1 1.284064 859 0 0 1 0 0
## 2423 15.67 -1 1.000000 313 1 1 1 1 0
## 2439 16.33 0 2.000000 156 1 0 0 0 4
## 2448 16.08 0 2.000000 859 0 0 1 2 4
## 2490 16.25 0 2.000000 156 1 0 0 1 0
## 2581 16.08 -1 1.000000 313 0 0 0 1 10
## 2583 16.33 0 1.000000 313 0 1 0 2 10
## 2599 15.92 -1 1.000000 313 1 0 1 0 10
## 2613 16.17 0 1.000000 313 1 0 0 0 0
## 2631 15.92 -1 1.000000 313 1 1 0 1 10
## 2665 15.42 0 1.000000 313 0 1 0 2 0
## 2689 15.33 0 3.000000 859 0 0 0 2 0
## 2704 15.42 -1 1.000000 313 0 1 1 1 0
## 2730 16.08 0 1.000000 313 0 1 0 1 10
## 2843 16.25 0 1.000000 313 1 0 0 0 0
## 2879 15.50 -1 1.000000 313 1 0 1 2 0
## 2885 16.17 0 1.000000 313 0 1 1 1 0
## 2914 16.08 0 2.000000 859 0 0 0 2 10
## 2998 15.50 -1 3.000000 859 1 0 0 0 5
## 3117 15.50 -1 1.000000 859 0 0 0 2 2
## 3124 16.08 0 3.000000 859 0 0 1 1 0
## 3131 15.75 0 1.000000 859 1 0 1 1 10
## 3272 15.92 0 1.284064 313 1 1 0 0 10
## 3299 16.17 1 2.000000 156 0 1 1 1 10
## 3415 15.67 0 3.000000 859 0 0 0 0 5
## 3440 16.33 0 3.000000 156 1 0 1 0 6
## 3453 16.25 -1 1.000000 313 0 0 0 2 10
## 3488 15.58 0 1.000000 313 0 1 0 1 0
## 3507 15.50 -1 1.000000 313 1 0 0 2 8
## 3517 15.58 0 1.000000 313 1 0 1 0 10
## 3776 15.92 -1 1.000000 313 1 1 1 2 1
## 3799 16.00 1 3.000000 859 0 0 0 1 10
## 3887 15.83 1 3.000000 156 0 1 0 1 8
## 3986 16.17 0 1.000000 313 1 1 0 2 0
## 4001 16.17 0 1.000000 313 0 1 1 2 10
## 4054 16.00 0 1.000000 859 0 0 0 1 0
## 4067 15.92 -1 1.000000 313 1 0 0 1 10
## 4083 16.08 0 2.000000 313 1 1 1 1 8
## 4088 16.33 1 1.000000 313 1 0 1 0 1
## 4147 16.33 0 1.000000 313 0 1 1 2 2
## 4197 15.50 0 1.000000 313 1 1 0 2 8
## 4460 15.92 0 3.000000 859 0 0 0 0 2
## 4492 16.08 0 3.000000 859 0 0 0 1 5
## 4504 16.00 0 1.000000 313 1 1 1 0 1
## 4522 15.58 -1 3.000000 156 0 1 0 2 4
## STUDYHMW WORKPAY
## 129 3 0
## 206 8 0
## 218 0 0
## 226 2 3
## 249 5 0
## 336 2 0
## 400 10 0
## 426 10 0
## 537 5 7
## 659 10 7
## 669 5 0
## 830 9 9
## 908 10 0
## 967 8 0
## 1010 4 0
## 1013 10 0
## 1065 10 5
## 1366 6 10
## 1470 2 0
## 1576 4 0
## 1641 8 1
## 1724 8 0
## 1749 4 2
## 1756 8 10
## 1764 10 10
## 1844 10 10
## 1970 10 10
## 1987 5 0
## 2004 3 10
## 2084 10 10
## 2156 10 0
## 2196 5 0
## 2241 3 0
## 2287 6 0
## 2371 6 6
## 2423 7 0
## 2439 10 0
## 2448 2 8
## 2490 4 4
## 2581 10 10
## 2583 6 2
## 2599 0 0
## 2613 10 0
## 2631 10 0
## 2665 4 0
## 2689 3 0
## 2704 0 0
## 2730 0 0
## 2843 0 8
## 2879 6 0
## 2885 7 9
## 2914 3 0
## 2998 5 0
## 3117 4 3
## 3124 8 0
## 3131 8 4
## 3272 4 8
## 3299 6 0
## 3415 6 0
## 3440 10 0
## 3453 10 0
## 3488 0 10
## 3507 8 6
## 3517 0 10
## 3776 0 10
## 3799 10 6
## 3887 4 4
## 3986 6 3
## 4001 6 6
## 4054 0 10
## 4067 10 10
## 4083 4 0
## 4088 3 2
## 4147 2 0
## 4197 3 2
## 4460 4 0
## 4492 0 0
## 4504 1 1
## 4522 4 0