Data_Reshaping

Reshaping

#GATHER
set.seed(2137)

library(tidyr)

# Data
student_data <- data.frame(
  student_id = 1:50,
  name = paste0("Student", 1:50),
  midterm1 = sample(60:100, 50, replace = TRUE),
  midterm2 = sample(60:100, 50, replace = TRUE),
  final = sample(60:100, 50, replace = TRUE)
)

# Transform data to long format using gather:
student_data_long <- gather(student_data, "exam", "score", -student_id, -name)

# Print first few rows:
head(student_data_long)

##   student_id     name     exam score
## 1          1 Student1 midterm1    85
## 2          2 Student2 midterm1    76
## 3          3 Student3 midterm1    73
## 4          4 Student4 midterm1    76
## 5          5 Student5 midterm1    73
## 6          6 Student6 midterm1    95

#SEPARATE
library(tidyr)

# Data
student_data2 <- data.frame(
  student_id = 1:50,
  name_age = c("John_21", "Alice_20", "Bob_22", "Emily_23", "Michael_22"),
  exam_scores = c("midterm1_80,midterm2_85,final_75", "midterm1_75,midterm2_78,final_80", "midterm1_82,midterm2_80,final_85", "midterm1_88,midterm2_90,final_92", "midterm1_85,midterm2_86,final_88")
)

# Split variables name_age to 2 separate columns: name and age
student_data2 <- separate(student_data2, name_age, 
                         into = c("name", "age"), sep="_")

# Split variables exam_scores to separate columns for each exam type
student_data2 <- separate(student_data2, exam_scores, 
                         into = c("exam_1", "exam_2","exam_3"), sep=",")

student_data2 <- separate(student_data2, exam_1, 
                         into = c("exam_1", "midterm1"), sep="_")
student_data2 <- separate(student_data2, exam_2, 
                         into = c("exam_2", "midterm2"), sep="_")
student_data2 <- separate(student_data2, exam_3, 
                         into = c("exam_3", "final"), sep="_")

# Transform student_data2 to long format
student_data2 <- subset(student_data2, select = -c(exam_1, exam_2, exam_3))
student_data_long2 <- gather(student_data2, exam_type, score, -student_id, -name, -age)

head(student_data_long2)

##   student_id    name age exam_type score
## 1          1    John  21  midterm1    80
## 2          2   Alice  20  midterm1    75
## 3          3     Bob  22  midterm1    82
## 4          4   Emily  23  midterm1    88
## 5          5 Michael  22  midterm1    85
## 6          6    John  21  midterm1    80

#COMPLETE

library(tidyr)
library(dplyr)

## 
## Dołączanie pakietu: 'dplyr'

## Następujące obiekty zostały zakryte z 'package:stats':
## 
##     filter, lag

## Następujące obiekty zostały zakryte z 'package:base':
## 
##     intersect, setdiff, setequal, union

# Data
student_data3 <- data.frame(
  student_id = 1:50,
  name_age = c("John_21", "Alice_20", "Bob_22", "Emily_23", "Michael_22"),
  exam_scores = c("midterm1_80,midterm2_85,final_75", "midterm1_75,midterm2_78,final_80", "midterm1_82,midterm2_80,final_85", "midterm1_88,midterm2_90,final_92", "midterm1_85,midterm2_86,final_88")
)

student_data3 <- separate(student_data3, name_age, into = c("name", "age"), sep = "_")

student_data3 <- separate(student_data3, exam_scores, into = c("midterm1_name", "midterm1_score", "midterm2_name", "midterm2_score", "final_name", "final_score"), sep = "[_,]")

colnames(student_data3) <- c("student_id", "name", "age", "exam_name", "exam_score", "exam_name2", "exam_score2", "exam_name3", "exam_score3")

# Complete missing combinations for students
student_data3 <- complete(student_data3, student_id, nesting(name, age), fill = list(exam_name = NA, exam_score = NA, exam_name2 = NA, exam_score2 = NA, exam_name3 = NA, exam_score3 = NA))

head(student_data3)

## # A tibble: 6 × 9
##   student_id name   age   exam_name exam_score exam_name2 exam_score2 exam_name3
##        <int> <chr>  <chr> <chr>     <chr>      <chr>      <chr>       <chr>     
## 1          1 Alice  20    <NA>      <NA>       <NA>       <NA>        <NA>      
## 2          1 Bob    22    <NA>      <NA>       <NA>       <NA>        <NA>      
## 3          1 Emily  23    <NA>      <NA>       <NA>       <NA>        <NA>      
## 4          1 John   21    midterm1  80         midterm2   85          final     
## 5          1 Micha… 22    <NA>      <NA>       <NA>       <NA>        <NA>      
## 6          2 Alice  20    midterm1  75         midterm2   78          final     
## # ℹ 1 more variable: exam_score3 <chr>

#SPREAD
library(tidyr)

# Data
student_data <- data.frame(
  student_id = 1:50,
  name_age = c("John_21", "Alice_20", "Bob_22", "Emily_23", "Michael_22"),
  exam_scores = c("midterm1_80,midterm2_85,final_75", "midterm1_75,midterm2_78,final_80", "midterm1_82,midterm2_80,final_85", "midterm1_88,midterm2_90,final_92", "midterm1_85,midterm2_86,final_88")
)

# Transform data to wide format
student_data_wide <- student_data %>%
  separate(name_age, into = c("name", "age"), sep = "_") %>%
  separate_rows(exam_scores, sep = ",") %>%
  separate(exam_scores, into = c("exam", "score"), sep = "_") %>%
  pivot_wider(names_from = exam, values_from = score)

head(student_data_wide)

## # A tibble: 6 × 6
##   student_id name    age   midterm1 midterm2 final
##        <int> <chr>   <chr> <chr>    <chr>    <chr>
## 1          1 John    21    80       85       75   
## 2          2 Alice   20    75       78       80   
## 3          3 Bob     22    82       80       85   
## 4          4 Emily   23    88       90       92   
## 5          5 Michael 22    85       86       88   
## 6          6 John    21    80       85       75

#UNITE \/
student_data_long <- student_data_wide %>%
  unite(name_age, name, age, sep=",")
head(student_data_long)

## # A tibble: 6 × 5
##   student_id name_age   midterm1 midterm2 final
##        <int> <chr>      <chr>    <chr>    <chr>
## 1          1 John,21    80       85       75   
## 2          2 Alice,20   75       78       80   
## 3          3 Bob,22     82       80       85   
## 4          4 Emily,23   88       90       92   
## 5          5 Michael,22 85       86       88   
## 6          6 John,21    80       85       75