/lecture_notes/Data_Reshaping.R

https://github.com/SISBID/Module1 · R · 116 lines · 54 code · 41 blank · 21 comment · 0 complexity · 43515f539703da57d5ea4a4ef296c448 MD5 · raw file 

    

  1. ## ---- echo = FALSE, include=FALSE---------------------------------------------
    2. 

  2. library(knitr)
    3. 

  3. library(readr)
    4. 

  4. opts_chunk$set(comment = "")
    5. 

  5. library(tidyverse)
    6. 

  6. 

  7. 

  8. ## ---- echo = FALSE, message=FALSE---------------------------------------------
    9. 

  9. library(tidyverse)
    10. 

  10. 

  11. 

  12. ## ---- echo = FALSE------------------------------------------------------------
    13. 

  13. ex_wide = tibble(State = "Alabama",
    14. 

  14.                  June_vacc_rate = "37.2%",
    15. 

  15.                  May_vacc_rate = "36.0%",
    16. 

  16.                  April_vacc_rate = "32.4%"
    17. 

  17.                      )
    18. 

  18. ex_long = pivot_longer(ex_wide, cols = c(June_vacc_rate, May_vacc_rate, April_vacc_rate))
    19. 

  19. 

  20. 

  21. ## ---- echo = FALSE------------------------------------------------------------
    22. 

  22. ex_wide
    23. 

  23. 

  24. 

  25. ## ---- echo = FALSE------------------------------------------------------------
    26. 

  26. ex_long
    27. 

  27. 

  28. 

  29. ## ---- echo = FALSE------------------------------------------------------------
    30. 

  30. ex_wide = tibble(State = c("Alabama", "Alaska"),
    31. 

  31.                  June_vacc_rate = c("37.2%", "47.5%"),
    32. 

  32.                  May_vacc_rate = c("36.0%", "46.2%"),
    33. 

  33.                  April_vacc_rate = c("32.4%", "41.7%")
    34. 

  34.                      )
    35. 

  35. ex_long = pivot_longer(ex_wide, cols = c(June_vacc_rate, May_vacc_rate, April_vacc_rate))
    36. 

  36. 

  37. 

  38. ## ---- echo = FALSE------------------------------------------------------------
    39. 

  39. ex_wide
    40. 

  40. 

  41. 

  42. ## ---- echo = FALSE------------------------------------------------------------
    43. 

  43. ex_long
    44. 

  44. 

  45. 

  46. ## ---- fig.alt="Wide versus long data rearanges the position of column names and row content.", out.width = "60%", echo = FALSE, fig.align='center'----
    47. 

  47. knitr::include_graphics("media/Reshape_pivot.jpg")
    48. 

  48. 

  49. 

  50. ## ---- fig.alt="Wide versus long data rearanges the position of column names and row content, animated.", out.width = "50%", echo = FALSE, fig.align='center'----
    51. 

  51. knitr::include_graphics("media/Reshape_pivot_animated.gif")
    52. 

  52. 

  53. 

  54. ## ---- echo = FALSE------------------------------------------------------------
    55. 

  55. ex_wide
    56. 

  56. 

  57. 

  58. ## ---- echo = FALSE------------------------------------------------------------
    59. 

  59. ex_long
    60. 

  60. 

  61. 

  62. ## ---- message = FALSE---------------------------------------------------------
    63. 

  63. circ = read_csv(
    64. 

  64.   paste0("http://jhudatascience.org/intro_to_r/",
    65. 

  65.          "data/Charm_City_Circulator_Ridership.csv"))
    66. 

  66. head(circ, 5)
    67. 

  67. 

  68. 

  69. ## -----------------------------------------------------------------------------
    70. 

  70. long = circ %>% 
    71. 

  71.   pivot_longer(starts_with(c("orange","purple","green","banner")),
    72. 

  72.                names_to = "var", values_to = "number")
    73. 

  73. long
    74. 

  74. 

  75. 

  76. ## -----------------------------------------------------------------------------
    77. 

  77. long = circ %>% pivot_longer(!c(day, date, daily),
    78. 

  78.                     names_to = "var", values_to = "number")
    79. 

  79. long
    80. 

  80. 

  81. 

  82. ## -----------------------------------------------------------------------------
    83. 

  83. long %>% count(var)
    84. 

  84. 

  85. 

  86. ## -----------------------------------------------------------------------------
    87. 

  87. long = long %>% mutate(
    88. 

  88.   var = str_replace(var, "Board", "_Board"),
    89. 

  89.   var = str_replace(var, "Alight", "_Alight"),
    90. 

  90.   var = str_replace(var, "Average", "_Average") 
    91. 

  91. )
    92. 

  92. long
    93. 

  93. 

  94. 

  95. ## -----------------------------------------------------------------------------
    96. 

  96. long = 
    97. 

  97.   long %>% 
    98. 

  98.   separate(var, into = c("line", "type"), sep = "_")
    99. 

  99. long
    100. 

  100. 

  101. 

  102. ## -----------------------------------------------------------------------------
    103. 

  103. reunited = long %>% 
    104. 

  104.   unite(var, line, type, sep = "_")  
    105. 

  105. reunited
    106. 

  106. 

  107. 

  108. ## -----------------------------------------------------------------------------
    109. 

  109. wide = long %>% pivot_wider(names_from = "type", 
    110. 

  110.                             values_from = "number") 
    111. 

  111. wide
    112. 

  112. 

  113. 

  114. ## ---- fig.alt="Long form sandwich", out.width = "50%", echo = FALSE, fig.align='center'----
    115. 

  115. knitr::include_graphics("media/Reshape_pivot_sandwich.png")
    116.