GGplot 3: Advanced charts and graphs
This week, we’re going to be working on one larger project; re-creating and improving a spiral-shaped plot of COVID-19 cases, published In the New York Times several years ago.
There were several criticisms of the chart.
- The x axis is longer for each year;
the length of the line for 2020 is shorter than the one for 2021, making it hard to compare infection rates over time
- The spiral is off-center
and isn’t centered on the page, making the whole thing seem a bit janky.
While there have been several remixes of the original, we’ll today focus on recreating a version of the spiral chart.
This will involve integrating our data processing into the chart, as well as doing some data transformations to make ggplot do some things that aren’t “out of the box”.
Getting the data
To start, we’ll use the COVID19 package to get some data on COVID-19
cases. You can find better, more up-to-date data
elsewhere, but the data set
gets really big and unwieldy.
1 install.packages("COVID19")
The package contains one function, covid19(), which returns a data
frame with COVID-19 data. Let’s load the data and see what we’re working
with.
1 library(tidyverse)
2 library(COVID19)
3 4 covid_data = covid19()
1 covid_data |> glimpse()
Rows: 287,783 Columns: 47 $ id <chr> "8320791a", "8320791a", "8320791a"… $ date <date> 2020-03-13, 2020-03-14, 2020-03-1… $ confirmed <int> 2, 2, 2, 2, 8, 9, 9, 10, 10, 13, 1… $ deaths <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ recovered <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ tests <int64> NA, NA, NA, NA, NA, NA, NA, NA, … $ vaccines <int64> NA, NA, NA, NA, NA, NA, NA, NA, … $ people_vaccinated <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ people_fully_vaccinated <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ hosp <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ icu <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ vent <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ school_closing <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ workplace_closing <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ cancel_events <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ gatherings_restrictions <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ transport_closing <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ stay_home_restrictions <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ internal_movement_restrictions <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ international_movement_restrictions <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ information_campaigns <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ testing_policy <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ contact_tracing <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ facial_coverings <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ vaccination_policy <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ elderly_people_protection <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ government_response_index <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ stringency_index <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ containment_health_index <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ economic_support_index <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ administrative_area_level <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1… $ administrative_area_level_1 <chr> "Grand Princess", "Grand Princess"… $ administrative_area_level_2 <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ administrative_area_level_3 <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ latitude <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ longitude <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ population <int> 3533, 3533, 3533, 3533, 3533, 3533… $ iso_alpha_3 <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ iso_alpha_2 <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ iso_numeric <int> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ iso_currency <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ key_local <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ key_google_mobility <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ key_apple_mobility <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ key_jhu_csse <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ key_nuts <lgl> NA, NA, NA, NA, NA, NA, NA, NA, NA… $ key_gadm <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA…
Looking back on the chart we’re trying to recreate, we can see that really the only data we need is the date and the number of confirmed cases. This data set also lists every country, so we’ll filter it down to just one country. I’m going to go with Switzerland. We’ll then select the date and the number of confirmed cases, as that’s the only data we’ll need in the end.
1 ch_covid_data <- covid_data |>
2 filter(administrative_area_level_1 == "Switzerland") |>
3 select(date, confirmed)
4 5 ch_covid_data
date confirmed 1 2020-01-22 NA 2 2020-01-23 NA 3 2020-01-24 NA 4 2020-01-25 NA 5 2020-01-26 NA 6 2020-01-27 NA 7 2020-01-28 NA 8 2020-01-29 NA 9 2020-01-30 NA 10 2020-01-31 NA 11 2020-02-01 NA 12 2020-02-02 NA 13 2020-02-03 NA 14 2020-02-04 NA 15 2020-02-05 NA 16 2020-02-06 NA 17 2020-02-07 NA 18 2020-02-08 NA 19 2020-02-09 NA 20 2020-02-10 NA 21 2020-02-11 NA 22 2020-02-12 NA 23 2020-02-13 NA 24 2020-02-14 NA 25 2020-02-15 NA 26 2020-02-16 NA 27 2020-02-17 NA 28 2020-02-18 NA 29 2020-02-19 NA 30 2020-02-20 NA 31 2020-02-21 NA 32 2020-02-22 NA 33 2020-02-23 NA 34 2020-02-24 1 35 2020-02-25 2 36 2020-02-26 12 37 2020-02-27 22 38 2020-02-28 32 39 2020-02-29 45 40 2020-03-01 57 41 2020-03-02 87 42 2020-03-03 120 43 2020-03-04 181 44 2020-03-05 243 45 2020-03-06 316 46 2020-03-07 365 47 2020-03-08 434 48 2020-03-09 625 49 2020-03-10 835 50 2020-03-11 1168 51 2020-03-12 1525 52 2020-03-13 1956 53 2020-03-14 2373 54 2020-03-15 2699 55 2020-03-16 3759 56 2020-03-17 4841 57 2020-03-18 6047 58 2020-03-19 6881 59 2020-03-20 8020 60 2020-03-21 8710 61 2020-03-22 9257 62 2020-03-23 10719 63 2020-03-24 11961 64 2020-03-25 13029 65 2020-03-26 14143 66 2020-03-27 15448 67 2020-03-28 16170 68 2020-03-29 16602 69 2020-03-30 17909 70 2020-03-31 19041 71 2020-04-01 20054 72 2020-04-02 20931 73 2020-04-03 21859 74 2020-04-04 22345 75 2020-04-05 22626 76 2020-04-06 23554 77 2020-04-07 24267 78 2020-04-08 24879 79 2020-04-09 25439 80 2020-04-10 25749 81 2020-04-11 26007 82 2020-04-12 26226 83 2020-04-13 26471 84 2020-04-14 26895 85 2020-04-15 27222 86 2020-04-16 27543 87 2020-04-17 27832 88 2020-04-18 27977 89 2020-04-19 28063 90 2020-04-20 28340 91 2020-04-21 28553 92 2020-04-22 28708 93 2020-04-23 28915 94 2020-04-24 29082 95 2020-04-25 29169 96 2020-04-26 29230 97 2020-04-27 29403 98 2020-04-28 29557 99 2020-04-29 29677 100 2020-04-30 29778 101 2020-05-01 29861 102 2020-05-02 29908 103 2020-05-03 29930 104 2020-05-04 30019 105 2020-05-05 30092 106 2020-05-06 30171 107 2020-05-07 30217 108 2020-05-08 30276 109 2020-05-09 30315 110 2020-05-10 30330 111 2020-05-11 30377 112 2020-05-12 30419 113 2020-05-13 30456 114 2020-05-14 30493 115 2020-05-15 30534 116 2020-05-16 30549 117 2020-05-17 30561 118 2020-05-18 30595 119 2020-05-19 30631 120 2020-05-20 30660 121 2020-05-21 30671 122 2020-05-22 30691 123 2020-05-23 30704 124 2020-05-24 30714 125 2020-05-25 30729 126 2020-05-26 30747 127 2020-05-27 30772 128 2020-05-28 30802 129 2020-05-29 30823 130 2020-05-30 30833 131 2020-05-31 30840 132 2020-06-01 30843 133 2020-06-02 30865 134 2020-06-03 30888 135 2020-06-04 30905 136 2020-06-05 30920 137 2020-06-06 30930 138 2020-06-07 30938 139 2020-06-08 30963 140 2020-06-09 30974 141 2020-06-10 30998 142 2020-06-11 31023 143 2020-06-12 31057 144 2020-06-13 31066 145 2020-06-14 31078 146 2020-06-15 31110 147 2020-06-16 31131 148 2020-06-17 31149 149 2020-06-18 31167 150 2020-06-19 31204 151 2020-06-20 31236 152 2020-06-21 31249 153 2020-06-22 31305 154 2020-06-23 31342 155 2020-06-24 31406 156 2020-06-25 31455 157 2020-06-26 31548 158 2020-06-27 31584 159 2020-06-28 31608 160 2020-06-29 31769 161 2020-06-30 31908 162 2020-07-01 32005 163 2020-07-02 32132 164 2020-07-03 32228 165 2020-07-04 32254 166 2020-07-05 32290 167 2020-07-06 32396 168 2020-07-07 32503 169 2020-07-08 32600 170 2020-07-09 32697 171 2020-07-10 32808 172 2020-07-11 32879 173 2020-07-12 32926 174 2020-07-13 33046 175 2020-07-14 33175 176 2020-07-15 33298 177 2020-07-16 33404 178 2020-07-17 33519 179 2020-07-18 33599 180 2020-07-19 33641 181 2020-07-20 33768 182 2020-07-21 33906 183 2020-07-22 34023 184 2020-07-23 34166 185 2020-07-24 34313 186 2020-07-25 34408 187 2020-07-26 34465 188 2020-07-27 34658 189 2020-07-28 34847 190 2020-07-29 35067 191 2020-07-30 35279 192 2020-07-31 35443 193 2020-08-01 35540 194 2020-08-02 35620 195 2020-08-03 35816 196 2020-08-04 35997 197 2020-08-05 36149 198 2020-08-06 36305 199 2020-08-07 36490 200 2020-08-08 36620 201 2020-08-09 36720 202 2020-08-10 37001 203 2020-08-11 37256 204 2020-08-12 37502 205 2020-08-13 37776 206 2020-08-14 38008 207 2020-08-15 38157 208 2020-08-16 38294 209 2020-08-17 38579 210 2020-08-18 38873 211 2020-08-19 39158 212 2020-08-20 39507 213 2020-08-21 39821 214 2020-08-22 40013 215 2020-08-23 40149 216 2020-08-24 40453 217 2020-08-25 40791 218 2020-08-26 41153 219 2020-08-27 41505 220 2020-08-28 41882 221 2020-08-29 42097 222 2020-08-30 42241 223 2020-08-31 42588 224 2020-09-01 42952 225 2020-09-02 43369 226 2020-09-03 43807 227 2020-09-04 44234 228 2020-09-05 44529 229 2020-09-06 44699 230 2020-09-07 45166 231 2020-09-08 45553 232 2020-09-09 46015 233 2020-09-10 46522 234 2020-09-11 47055 235 2020-09-12 47321 236 2020-09-13 47559 237 2020-09-14 48063 238 2020-09-15 48520 239 2020-09-16 49036 240 2020-09-17 49535 241 2020-09-18 49980 242 2020-09-19 50264 243 2020-09-20 50464 244 2020-09-21 50859 245 2020-09-22 51285 246 2020-09-23 51674 247 2020-09-24 51984 248 2020-09-25 52343 249 2020-09-26 52565 250 2020-09-27 52685 251 2020-09-28 53054 252 2020-09-29 53503 253 2020-09-30 54043 254 2020-10-01 54645 255 2020-10-02 55322 256 2020-10-03 55834 257 2020-10-04 56146 258 2020-10-05 57165 259 2020-10-06 58263 260 2020-10-07 59640 261 2020-10-08 61113 262 2020-10-09 62836 263 2020-10-10 64088 264 2020-10-11 64899 265 2020-10-12 67457 266 2020-10-13 70145 267 2020-10-14 73143 268 2020-10-15 76537 269 2020-10-16 80217 270 2020-10-17 82596 271 2020-10-18 84380 272 2020-10-19 89778 273 2020-10-20 95456 274 2020-10-21 101912 275 2020-10-22 108684 276 2020-10-23 116186 277 2020-10-24 120848 278 2020-10-25 124058 279 2020-10-26 134185 280 2020-10-27 143929 281 2020-10-28 153208 282 2020-10-29 162182 283 2020-10-30 171125 284 2020-10-31 176253 285 2020-11-01 179879 286 2020-11-02 190396 287 2020-11-03 199993 288 2020-11-04 208705 289 2020-11-05 216532 290 2020-11-06 223908 291 2020-11-07 228095 292 2020-11-08 231074 293 2020-11-09 239753 294 2020-11-10 247142 295 2020-11-11 253501 296 2020-11-12 259464 297 2020-11-13 265145 298 2020-11-14 268349 299 2020-11-15 270664 300 2020-11-16 277242 301 2020-11-17 282658 302 2020-11-18 287513 303 2020-11-19 292003 304 2020-11-20 296331 305 2020-11-21 298904 306 2020-11-22 300776 307 2020-11-23 306210 308 2020-11-24 310643 309 2020-11-25 314936 310 2020-11-26 318885 311 2020-11-27 322952 312 2020-11-28 325262 313 2020-11-29 326953 314 2020-11-30 332220 315 2020-12-01 336472 316 2020-12-02 340895 317 2020-12-03 345214 318 2020-12-04 349773 319 2020-12-05 352291 320 2020-12-06 354230 321 2020-12-07 360057 322 2020-12-08 364726 323 2020-12-09 369966 324 2020-12-10 374603 325 2020-12-11 379281 326 2020-12-12 382144 327 2020-12-13 384230 328 2020-12-14 390145 329 2020-12-15 395311 330 2020-12-16 400129 331 2020-12-17 404515 332 2020-12-18 408857 333 2020-12-19 411534 334 2020-12-20 413489 335 2020-12-21 419112 336 2020-12-22 423718 337 2020-12-23 428193 338 2020-12-24 431241 339 2020-12-25 432962 340 2020-12-26 435319 341 2020-12-27 437435 342 2020-12-28 442470 343 2020-12-29 446933 344 2020-12-30 451142 345 2020-12-31 454197 346 2021-01-01 455881 347 2021-01-02 458608 348 2021-01-03 460833 349 2021-01-04 466120 350 2021-01-05 470298 351 2021-01-06 473652 352 2021-01-07 476829 353 2021-01-08 479884 354 2021-01-09 481570 355 2021-01-10 482866 356 2021-01-11 486333 357 2021-01-12 488944 358 2021-01-13 491435 359 2021-01-14 493674 360 2021-01-15 495832 361 2021-01-16 497169 362 2021-01-17 498063 363 2021-01-18 501006 364 2021-01-19 503340 365 2021-01-20 505618 366 2021-01-21 507675 367 2021-01-22 509716 368 2021-01-23 510817 369 2021-01-24 511548 370 2021-01-25 514040 371 2021-01-26 515878 372 2021-01-27 517753 373 2021-01-28 519534 374 2021-01-29 521301 375 2021-01-30 522321 376 2021-01-31 523064 377 2021-02-01 525229 378 2021-02-02 526881 379 2021-02-03 528614 380 2021-02-04 530055 381 2021-02-05 531561 382 2021-02-06 532448 383 2021-02-07 533121 384 2021-02-08 534872 385 2021-02-09 536222 386 2021-02-10 537481 387 2021-02-11 538651 388 2021-02-12 539781 389 2021-02-13 540500 390 2021-02-14 540985 391 2021-02-15 542290 392 2021-02-16 543427 393 2021-02-17 544642 394 2021-02-18 545651 395 2021-02-19 546745 396 2021-02-20 547436 397 2021-02-21 547964 398 2021-02-22 549371 399 2021-02-23 550628 400 2021-02-24 551705 401 2021-02-25 552824 402 2021-02-26 553904 403 2021-02-27 554627 404 2021-02-28 555174 405 2021-03-01 556593 406 2021-03-02 557823 407 2021-03-03 559063 408 2021-03-04 560128 409 2021-03-05 561407 410 2021-03-06 562173 411 2021-03-07 562755 412 2021-03-08 564457 413 2021-03-09 565845 414 2021-03-10 567230 415 2021-03-11 568570 416 2021-03-12 569912 417 2021-03-13 570835 418 2021-03-14 571538 419 2021-03-15 573461 420 2021-03-16 575115 421 2021-03-17 576940 422 2021-03-18 578587 423 2021-03-19 580132 424 2021-03-20 581244 425 2021-03-21 582032 426 2021-03-22 584295 427 2021-03-23 586289 428 2021-03-24 588220 429 2021-03-25 590362 430 2021-03-26 592361 431 2021-03-27 593568 432 2021-03-28 594562 433 2021-03-29 597094 434 2021-03-30 599187 435 2021-03-31 601319 436 2021-04-01 603304 437 2021-04-02 604509 438 2021-04-03 605826 439 2021-04-04 606848 440 2021-04-05 608138 441 2021-04-06 610828 442 2021-04-07 613215 443 2021-04-08 615705 444 2021-04-09 618121 445 2021-04-10 619803 446 2021-04-11 620984 447 2021-04-12 623726 448 2021-04-13 626101 449 2021-04-14 628315 450 2021-04-15 630462 451 2021-04-16 632583 452 2021-04-17 634013 453 2021-04-18 635061 454 2021-04-19 637805 455 2021-04-20 640196 456 2021-04-21 642505 457 2021-04-22 644770 458 2021-04-23 646935 459 2021-04-24 648390 460 2021-04-25 649578 461 2021-04-26 652019 462 2021-04-27 654057 463 2021-04-28 656066 464 2021-04-29 657777 465 2021-04-30 659454 466 2021-05-01 660613 467 2021-05-02 661445 468 2021-05-03 663462 469 2021-05-04 665197 470 2021-05-05 666845 471 2021-05-06 668329 472 2021-05-07 669902 473 2021-05-08 670963 474 2021-05-09 671742 475 2021-05-10 673461 476 2021-05-11 674844 477 2021-05-12 676285 478 2021-05-13 676957 479 2021-05-14 678126 480 2021-05-15 678928 481 2021-05-16 679581 482 2021-05-17 681252 483 2021-05-18 682596 484 2021-05-19 683808 485 2021-05-20 684848 486 2021-05-21 685832 487 2021-05-22 686455 488 2021-05-23 686925 489 2021-05-24 687506 490 2021-05-25 688654 491 2021-05-26 689577 492 2021-05-27 690426 493 2021-05-28 691177 494 2021-05-29 691677 495 2021-05-30 692034 496 2021-05-31 692888 497 2021-06-01 693540 498 2021-06-02 694091 499 2021-06-03 694559 500 2021-06-04 695050 501 2021-06-05 695363 502 2021-06-06 695550 503 2021-06-07 696134 504 2021-06-08 696551 505 2021-06-09 696926 506 2021-06-10 697230 507 2021-06-11 697548 508 2021-06-12 697742 509 2021-06-13 697863 510 2021-06-14 698191 511 2021-06-15 698424 512 2021-06-16 698615 513 2021-06-17 698771 514 2021-06-18 698920 515 2021-06-19 699023 516 2021-06-20 699101 517 2021-06-21 699266 518 2021-06-22 699388 519 2021-06-23 699515 520 2021-06-24 699628 521 2021-06-25 699726 522 2021-06-26 699786 523 2021-06-27 699845 524 2021-06-28 699995 525 2021-06-29 700110 526 2021-06-30 700269 527 2021-07-01 700438 528 2021-07-02 700603 529 2021-07-03 700732 530 2021-07-04 700871 531 2021-07-05 701181 532 2021-07-06 701470 533 2021-07-07 701790 534 2021-07-08 702106 535 2021-07-09 702433 536 2021-07-10 702688 537 2021-07-11 702918 538 2021-07-12 703489 539 2021-07-13 704009 540 2021-07-14 704599 541 2021-07-15 705222 542 2021-07-16 705867 543 2021-07-17 706321 544 2021-07-18 706716 545 2021-07-19 707573 546 2021-07-20 708310 547 2021-07-21 709184 548 2021-07-22 709908 549 2021-07-23 710651 550 2021-07-24 711189 551 2021-07-25 711669 552 2021-07-26 712620 553 2021-07-27 713360 554 2021-07-28 714145 555 2021-07-29 714927 556 2021-07-30 715730 557 2021-07-31 716386 558 2021-08-01 716921 559 2021-08-02 718155 560 2021-08-03 719145 561 2021-08-04 720215 562 2021-08-05 721335 563 2021-08-06 722488 564 2021-08-07 723534 565 2021-08-08 724543 566 2021-08-09 726863 567 2021-08-10 728893 568 2021-08-11 730978 569 2021-08-12 733145 570 2021-08-13 735353 571 2021-08-14 737032 572 2021-08-15 738683 573 2021-08-16 742634 574 2021-08-17 745472 575 2021-08-18 748182 576 2021-08-19 750694 577 2021-08-20 753254 578 2021-08-21 755002 579 2021-08-22 756635 580 2021-08-23 760486 581 2021-08-24 763402 582 2021-08-25 766167 583 2021-08-26 768861 584 2021-08-27 771457 585 2021-08-28 773134 586 2021-08-29 774519 587 2021-08-30 778081 588 2021-08-31 781055 589 2021-09-01 783997 590 2021-09-02 787048 591 2021-09-03 790123 592 2021-09-04 791883 593 2021-09-05 793282 594 2021-09-06 797277 595 2021-09-07 800321 596 2021-09-08 803381 597 2021-09-09 806150 598 2021-09-10 808772 599 2021-09-11 810323 600 2021-09-12 811429 601 2021-09-13 814193 602 2021-09-14 816489 603 2021-09-15 818731 604 2021-09-16 820729 605 2021-09-17 822420 606 2021-09-18 823447 607 2021-09-19 824149 608 2021-09-20 826029 609 2021-09-21 827732 610 2021-09-22 829263 611 2021-09-23 830641 612 2021-09-24 831984 613 2021-09-25 832870 614 2021-09-26 833477 615 2021-09-27 835138 616 2021-09-28 836346 617 2021-09-29 837533 618 2021-09-30 838727 619 2021-10-01 839865 620 2021-10-02 840651 621 2021-10-03 841124 622 2021-10-04 842403 623 2021-10-05 843437 624 2021-10-06 844446 625 2021-10-07 845456 626 2021-10-08 846376 627 2021-10-09 847014 628 2021-10-10 847495 629 2021-10-11 848667 630 2021-10-12 849631 631 2021-10-13 850594 632 2021-10-14 851551 633 2021-10-15 852536 634 2021-10-16 853236 635 2021-10-17 853916 636 2021-10-18 855487 637 2021-10-19 856776 638 2021-10-20 858206 639 2021-10-21 859579 640 2021-10-22 860931 641 2021-10-23 861885 642 2021-10-24 862612 643 2021-10-25 864636 644 2021-10-26 866335 645 2021-10-27 868142 646 2021-10-28 869945 647 2021-10-29 871809 648 2021-10-30 873068 649 2021-10-31 874149 650 2021-11-01 876492 651 2021-11-02 879473 652 2021-11-03 882309 653 2021-11-04 885151 654 2021-11-05 887906 655 2021-11-06 889837 656 2021-11-07 891374 657 2021-11-08 895578 658 2021-11-09 899316 659 2021-11-10 903285 660 2021-11-11 907237 661 2021-11-12 911381 662 2021-11-13 914090 663 2021-11-14 916255 664 2021-11-15 922388 665 2021-11-16 928014 666 2021-11-17 934195 667 2021-11-18 940249 668 2021-11-19 946443 669 2021-11-20 950580 670 2021-11-21 953983 671 2021-11-22 963147 672 2021-11-23 971150 673 2021-11-24 979619 674 2021-11-25 987606 675 2021-11-26 995573 676 2021-11-27 1000835 677 2021-11-28 1005077 678 2021-11-29 1016577 679 2021-11-30 1026576 680 2021-12-01 1036991 681 2021-12-02 1047144 682 2021-12-03 1056945 683 2021-12-04 1063202 684 2021-12-05 1068140 685 2021-12-06 1080802 686 2021-12-07 1091837 687 2021-12-08 1101788 688 2021-12-09 1112208 689 2021-12-10 1121964 690 2021-12-11 1128053 691 2021-12-12 1132448 692 2021-12-13 1144877 693 2021-12-14 1155403 694 2021-12-15 1164935 695 2021-12-16 1174078 696 2021-12-17 1182814 697 2021-12-18 1188690 698 2021-12-19 1193023 699 2021-12-20 1204492 700 2021-12-21 1215268 701 2021-12-22 1226276 702 2021-12-23 1237542 703 2021-12-24 1247174 704 2021-12-25 1253630 705 2021-12-26 1261001 706 2021-12-27 1278426 707 2021-12-28 1296094 708 2021-12-29 1315086 709 2021-12-30 1334832 710 2021-12-31 1350965 711 2022-01-01 1360363 712 2022-01-02 1373173 713 2022-01-03 1405443 714 2022-01-04 1438808 715 2022-01-05 1469458 716 2022-01-06 1495587 717 2022-01-07 1522852 718 2022-01-08 1540630 719 2022-01-09 1554015 720 2022-01-10 1588544 721 2022-01-11 1619464 722 2022-01-12 1649075 723 2022-01-13 1677694 724 2022-01-14 1705318 725 2022-01-15 1724607 726 2022-01-16 1739364 727 2022-01-17 1781159 728 2022-01-18 1820470 729 2022-01-19 1859564 730 2022-01-20 1897222 731 2022-01-21 1933859 732 2022-01-22 1958490 733 2022-01-23 1976825 734 2022-01-24 2025157 735 2022-01-25 2069280 736 2022-01-26 2110672 737 2022-01-27 2148381 738 2022-01-28 2184000 739 2022-01-29 2207279 740 2022-01-30 2225298 741 2022-01-31 2268530 742 2022-02-01 2305075 743 2022-02-02 2339133 744 2022-02-03 2368045 745 2022-02-04 2394373 746 2022-02-05 2412551 747 2022-02-06 2426968 748 2022-02-07 2459136 749 2022-02-08 2486108 750 2022-02-09 2511579 751 2022-02-10 2533812 752 2022-02-11 2553499 753 2022-02-12 2567488 754 2022-02-13 2578229 755 2022-02-14 2602892 756 2022-02-15 2623416 757 2022-02-16 2642512 758 2022-02-17 2656779 759 2022-02-18 2670221 760 2022-02-19 2679997 761 2022-02-20 2688043 762 2022-02-21 2707646 763 2022-02-22 2724639 764 2022-02-23 2742220 765 2022-02-24 2757990 766 2022-02-25 2772844 767 2022-02-26 2783921 768 2022-02-27 2793212 769 2022-02-28 2817029 770 2022-03-01 2839462 771 2022-03-02 2864293 772 2022-03-03 2888904 773 2022-03-04 2912505 774 2022-03-05 2930160 775 2022-03-06 2944539 776 2022-03-07 2979430 777 2022-03-08 3010203 778 2022-03-09 3041142 779 2022-03-10 3070532 780 2022-03-11 3098243 781 2022-03-12 3118522 782 2022-03-13 3134606 783 2022-03-14 3174978 784 2022-03-15 3208189 785 2022-03-16 3238832 786 2022-03-17 3265409 787 2022-03-18 3289152 788 2022-03-19 3305268 789 2022-03-20 3317464 790 2022-03-21 3348220 791 2022-03-22 3371243 792 2022-03-23 3390965 793 2022-03-24 3407762 794 2022-03-25 3422322 795 2022-03-26 3432136 796 2022-03-27 3439367 797 2022-03-28 3458051 798 2022-03-29 3471855 799 2022-03-30 3484001 800 2022-03-31 3492327 801 2022-04-01 3498709 802 2022-04-02 3502903 803 2022-04-03 3505953 804 2022-04-04 3514556 805 2022-04-05 3521571 806 2022-04-06 3528781 807 2022-04-07 3534923 808 2022-04-08 3540639 809 2022-04-09 3544305 810 2022-04-10 3546878 811 2022-04-11 3554099 812 2022-04-12 3559697 813 2022-04-13 3565229 814 2022-04-14 3569945 815 2022-04-15 3572716 816 2022-04-16 3576026 817 2022-04-17 3577901 818 2022-04-18 3580396 819 2022-04-19 3585851 820 2022-04-20 3589901 821 2022-04-21 3593527 822 2022-04-22 3596687 823 2022-04-23 3598709 824 2022-04-24 3600241 825 2022-04-25 3604123 826 2022-04-26 3606918 827 2022-04-27 3609651 828 2022-04-28 3612185 829 2022-04-29 3614482 830 2022-04-30 3615986 831 2022-05-01 3617009 832 2022-05-02 3620003 833 2022-05-03 3622151 834 2022-05-04 3624301 835 2022-05-05 3626219 836 2022-05-06 3628128 837 2022-05-07 3629310 838 2022-05-08 3630144 839 2022-05-09 3632642 840 2022-05-10 3634609 841 2022-05-11 3636455 842 2022-05-12 3638163 843 2022-05-13 3639708 844 2022-05-14 3640645 845 2022-05-15 3641341 846 2022-05-16 3643280 847 2022-05-17 3644756 848 2022-05-18 3646182 849 2022-05-19 3647451 850 2022-05-20 3648531 851 2022-05-21 3649304 852 2022-05-22 3649896 853 2022-05-23 3651355 854 2022-05-24 3652468 855 2022-05-25 3653552 856 2022-05-26 3654055 857 2022-05-27 3655218 858 2022-05-28 3655924 859 2022-05-29 3656584 860 2022-05-30 3658364 861 2022-05-31 3659945 862 2022-06-01 3661543 863 2022-06-02 3663128 864 2022-06-03 3664721 865 2022-06-04 3665867 866 2022-06-05 3666781 867 2022-06-06 3668071 868 2022-06-07 3671131 869 2022-06-08 3673779 870 2022-06-09 3676262 871 2022-06-10 3678716 872 2022-06-11 3680412 873 2022-06-12 3681706 874 2022-06-13 3685819 875 2022-06-14 3689473 876 2022-06-15 3693369 877 2022-06-16 3696907 878 2022-06-17 3700841 879 2022-06-18 3703418 880 2022-06-19 3705160 881 2022-06-20 3711120 882 2022-06-21 3716198 883 2022-06-22 3721451 884 2022-06-23 3726474 885 2022-06-24 3731097 886 2022-06-25 3734386 887 2022-06-26 3736723 888 2022-06-27 3745008 889 2022-06-28 3752253 890 2022-06-29 3759489 891 2022-06-30 3766645 892 2022-07-01 3773495 893 2022-07-02 3777997 894 2022-07-03 3780918 895 2022-07-04 3792146 896 2022-07-05 3801755 897 2022-07-06 3811553 898 2022-07-07 3820199 899 2022-07-08 3828116 900 2022-07-09 3833102 901 2022-07-10 3836408 902 2022-07-11 3847384 903 2022-07-12 3856855 904 2022-07-13 3865504 905 2022-07-14 3873216 906 2022-07-15 3880279 907 2022-07-16 3884586 908 2022-07-17 3887444 909 2022-07-18 3897073 910 2022-07-19 3904153 911 2022-07-20 3910489 912 2022-07-21 3916205 913 2022-07-22 3921499 914 2022-07-23 3924699 915 2022-07-24 3926683 916 2022-07-25 3933314 917 2022-07-26 3938153 918 2022-07-27 3942382 919 2022-07-28 3946037 920 2022-07-29 3949569 921 2022-07-30 3951890 922 2022-07-31 3953429 923 2022-08-01 3955237 924 2022-08-02 3960093 925 2022-08-03 3963751 926 2022-08-04 3966862 927 2022-08-05 3969837 928 2022-08-06 3971923 929 2022-08-07 3973315 930 2022-08-08 3977650 931 2022-08-09 3981009 932 2022-08-10 3984117 933 2022-08-11 3986849 934 2022-08-12 3989521 935 2022-08-13 3991064 936 2022-08-14 3992216 937 2022-08-15 3995528 938 2022-08-16 3998688 939 2022-08-17 4001556 940 2022-08-18 4004090 941 2022-08-19 4006425 942 2022-08-20 4007864 943 2022-08-21 4008871 944 2022-08-22 4012279 945 2022-08-23 4014875 946 2022-08-24 4017376 947 2022-08-25 4019654 948 2022-08-26 4021821 949 2022-08-27 4023186 950 2022-08-28 4024089 951 2022-08-29 4027438 952 2022-08-30 4030156 953 2022-08-31 4032739 954 2022-09-01 4034839 955 2022-09-02 4036876 956 2022-09-03 4038069 957 2022-09-04 4038850 958 2022-09-05 4041858 959 2022-09-06 4044247 960 2022-09-07 4046623 961 2022-09-08 4048623 962 2022-09-09 4050601 963 2022-09-10 4051766 964 2022-09-11 4052603 965 2022-09-12 4055596 966 2022-09-13 4057970 967 2022-09-14 4060341 968 2022-09-15 4062497 969 2022-09-16 4064498 970 2022-09-17 4065685 971 2022-09-18 4066442 972 2022-09-19 4069123 973 2022-09-20 4071620 974 2022-09-21 4074190 975 2022-09-22 4076669 976 2022-09-23 4079219 977 2022-09-24 4080851 978 2022-09-25 4082039 979 2022-09-26 4086466 980 2022-09-27 4090309 981 2022-09-28 4094100 982 2022-09-29 4097849 983 2022-09-30 4101567 984 2022-10-01 4103894 985 2022-10-02 4105520 986 2022-10-03 4111760 987 2022-10-04 4117277 988 2022-10-05 4123130 989 2022-10-06 4128739 990 2022-10-07 4134421 991 2022-10-08 4137824 992 2022-10-09 4139883 993 2022-10-10 4148095 994 2022-10-11 4154626 995 2022-10-12 4160989 996 2022-10-13 4166549 997 2022-10-14 4171903 998 2022-10-15 4175057 999 2022-10-16 4176964 1000 2022-10-17 4184504 1001 2022-10-18 4190229 1002 2022-10-19 4195507 1003 2022-10-20 4200028 1004 2022-10-21 4204270 1005 2022-10-22 4206711 1006 2022-10-23 4208264 1007 2022-10-24 4213940 1008 2022-10-25 4218441 1009 2022-10-26 4222493 1010 2022-10-27 4226075 1011 2022-10-28 4229456 1012 2022-10-29 4231441 1013 2022-10-30 4232752 1014 2022-10-31 4237494 1015 2022-11-01 4240661 1016 2022-11-02 4244691 1017 2022-11-03 4247955 1018 2022-11-04 4251057 1019 2022-11-05 4252807 1020 2022-11-06 4253908 1021 2022-11-07 4258323 1022 2022-11-08 4261937 1023 2022-11-09 4265364 1024 2022-11-10 4268482 1025 2022-11-11 4271599 1026 2022-11-12 4273360 1027 2022-11-13 4274542 1028 2022-11-14 4279164 1029 2022-11-15 4283007 1030 2022-11-16 4286636 1031 2022-11-17 4289973 1032 2022-11-18 4292879 1033 2022-11-19 4294703 1034 2022-11-20 4295852 1035 2022-11-21 4300266 1036 2022-11-22 4303705 1037 2022-11-23 4307108 1038 2022-11-24 4309975 1039 2022-11-25 4312678 1040 2022-11-26 4314234 1041 2022-11-27 4315217 1042 2022-11-28 4319216 1043 2022-11-29 4322481 1044 2022-11-30 4325759 1045 2022-12-01 4328551 1046 2022-12-02 4331311 1047 2022-12-03 4332868 1048 2022-12-04 4333803 1049 2022-12-05 4337605 1050 2022-12-06 4340817 1051 2022-12-07 4343655 1052 2022-12-08 4346021 1053 2022-12-09 4348319 1054 2022-12-10 4349583 1055 2022-12-11 4350394 1056 2022-12-12 4353564 1057 2022-12-13 4356006 1058 2022-12-14 4358271 1059 2022-12-15 4360415 1060 2022-12-16 4362328 1061 2022-12-17 4363360 1062 2022-12-18 4364049 1063 2022-12-19 4366599 1064 2022-12-20 4368798 1065 2022-12-21 4370799 1066 2022-12-22 4372595 1067 2022-12-23 4374210 1068 2022-12-24 4375085 1069 2022-12-25 4375567 1070 2022-12-26 4376250 1071 2022-12-27 4377532 1072 2022-12-28 4378584 1073 2022-12-29 4379585 1074 2022-12-30 4380463 1075 2022-12-31 4381033 1076 2023-01-01 4381252 1077 2023-01-02 NA 1078 2023-01-03 NA 1079 2023-01-04 NA 1080 2023-01-05 NA 1081 2023-01-06 NA 1082 2023-01-07 NA 1083 2023-01-08 4383454 1084 2023-01-09 NA 1085 2023-01-10 NA 1086 2023-01-11 NA 1087 2023-01-12 NA 1088 2023-01-13 NA 1089 2023-01-14 NA 1090 2023-01-15 4384791 1091 2023-01-16 NA 1092 2023-01-17 NA 1093 2023-01-18 NA 1094 2023-01-19 NA 1095 2023-01-20 NA 1096 2023-01-21 NA 1097 2023-01-22 4385717 1098 2023-01-23 NA 1099 2023-01-24 NA 1100 2023-01-25 NA 1101 2023-01-26 NA 1102 2023-01-27 NA 1103 2023-01-28 NA 1104 2023-01-29 4386564 1105 2023-01-30 NA 1106 2023-01-31 NA 1107 2023-02-01 NA 1108 2023-02-02 NA 1109 2023-02-03 NA 1110 2023-02-04 NA 1111 2023-02-05 4387476 1112 2023-02-06 NA 1113 2023-02-07 NA 1114 2023-02-08 NA 1115 2023-02-09 NA 1116 2023-02-10 NA 1117 2023-02-11 NA 1118 2023-02-12 4388447 1119 2023-02-13 NA 1120 2023-02-14 NA 1121 2023-02-15 NA 1122 2023-02-16 NA 1123 2023-02-17 NA 1124 2023-02-18 NA 1125 2023-02-19 4389665 1126 2023-02-20 NA 1127 2023-02-21 NA 1128 2023-02-22 NA 1129 2023-02-23 NA 1130 2023-02-24 NA 1131 2023-02-25 NA 1132 2023-02-26 4391025 1133 2023-02-27 NA 1134 2023-02-28 NA 1135 2023-03-01 NA 1136 2023-03-02 NA 1137 2023-03-03 NA 1138 2023-03-04 NA 1139 2023-03-05 4392515 1140 2023-03-06 NA 1141 2023-03-07 NA 1142 2023-03-08 NA 1143 2023-03-09 NA 1144 2023-03-10 NA 1145 2023-03-11 NA 1146 2023-03-12 4394313 1147 2023-03-13 NA 1148 2023-03-14 NA 1149 2023-03-15 NA 1150 2023-03-16 NA 1151 2023-03-17 NA 1152 2023-03-18 NA 1153 2023-03-19 4396099 1154 2023-03-20 NA 1155 2023-03-21 NA 1156 2023-03-22 NA 1157 2023-03-23 NA 1158 2023-03-24 NA 1159 2023-03-25 NA 1160 2023-03-26 4397683 1161 2023-03-27 NA 1162 2023-03-28 NA 1163 2023-03-29 NA 1164 2023-03-30 NA 1165 2023-03-31 NA 1166 2023-04-01 NA 1167 2023-04-02 4399119 1168 2023-04-03 NA 1169 2023-04-04 NA 1170 2023-04-05 NA 1171 2023-04-06 NA 1172 2023-04-07 NA 1173 2023-04-08 NA 1174 2023-04-09 4400299 1175 2023-04-10 NA 1176 2023-04-11 NA 1177 2023-04-12 NA 1178 2023-04-13 NA 1179 2023-04-14 NA 1180 2023-04-15 NA 1181 2023-04-16 4401413 1182 2023-04-17 NA 1183 2023-04-18 NA 1184 2023-04-19 NA 1185 2023-04-20 NA 1186 2023-04-21 NA 1187 2023-04-22 NA 1188 2023-04-23 4402475 1189 2023-04-24 NA 1190 2023-04-25 NA 1191 2023-04-26 NA 1192 2023-04-27 NA 1193 2023-04-28 NA 1194 2023-04-29 NA 1195 2023-04-30 4403491 1196 2023-05-01 NA 1197 2023-05-07 4404327
Classwork: Plot No. 1
Just to check that we have the data we need, let’s plot the number of confirmed cases over time.
This looks pretty good! However, we have to deal with a little bit of missing data.
First, we can see that towards the end of the pandemic, the number of confirmed cases was only reported about once a week. This means that we have to fill in the missing values with the last known value.
1 ch_covid_data |> tail(20)
date confirmed 1178 2023-04-13 NA 1179 2023-04-14 NA 1180 2023-04-15 NA 1181 2023-04-16 4401413 1182 2023-04-17 NA 1183 2023-04-18 NA 1184 2023-04-19 NA 1185 2023-04-20 NA 1186 2023-04-21 NA 1187 2023-04-22 NA 1188 2023-04-23 4402475 1189 2023-04-24 NA 1190 2023-04-25 NA 1191 2023-04-26 NA 1192 2023-04-27 NA 1193 2023-04-28 NA 1194 2023-04-29 NA 1195 2023-04-30 4403491 1196 2023-05-01 NA 1197 2023-05-07 4404327
Replacing missing values with `fill()`
We can use the fill() function to fill in the missing values. This
takes the last known value above and fills it in below, until the next
known value is found.
Here, we’re looking at the total confirmed cases, so filling in the missing values with the last known value is a good idea. If these were new cases, it would be a different story, we’d probably want to fill in the missing values with 0.
1 ch_covid_data <- ch_covid_data |>
2 fill(confirmed, .direction = "down")
3 4 ch_covid_data |> tail(20)
date confirmed 1178 2023-04-13 4400299 1179 2023-04-14 4400299 1180 2023-04-15 4400299 1181 2023-04-16 4401413 1182 2023-04-17 4401413 1183 2023-04-18 4401413 1184 2023-04-19 4401413 1185 2023-04-20 4401413 1186 2023-04-21 4401413 1187 2023-04-22 4401413 1188 2023-04-23 4402475 1189 2023-04-24 4402475 1190 2023-04-25 4402475 1191 2023-04-26 4402475 1192 2023-04-27 4402475 1193 2023-04-28 4402475 1194 2023-04-29 4402475 1195 2023-04-30 4403491 1196 2023-05-01 4403491 1197 2023-05-07 4404327
However, there are still some missing values at the beginning of the data set, before COVID was seen in Switzerland.
1 ch_covid_data |> head(20)
date confirmed 1 2020-01-22 NA 2 2020-01-23 NA 3 2020-01-24 NA 4 2020-01-25 NA 5 2020-01-26 NA 6 2020-01-27 NA 7 2020-01-28 NA 8 2020-01-29 NA 9 2020-01-30 NA 10 2020-01-31 NA 11 2020-02-01 NA 12 2020-02-02 NA 13 2020-02-03 NA 14 2020-02-04 NA 15 2020-02-05 NA 16 2020-02-06 NA 17 2020-02-07 NA 18 2020-02-08 NA 19 2020-02-09 NA 20 2020-02-10 NA
We can simply replace these missing values with 0, using replace_na().
1 ch_covid_data <- ch_covid_data |>
2 mutate(confirmed = replace_na(confirmed, 0))
3 4 ch_covid_data |> head(20)
date confirmed 1 2020-01-22 0 2 2020-01-23 0 3 2020-01-24 0 4 2020-01-25 0 5 2020-01-26 0 6 2020-01-27 0 7 2020-01-28 0 8 2020-01-29 0 9 2020-01-30 0 10 2020-01-31 0 11 2020-02-01 0 12 2020-02-02 0 13 2020-02-03 0 14 2020-02-04 0 15 2020-02-05 0 16 2020-02-06 0 17 2020-02-07 0 18 2020-02-08 0 19 2020-02-09 0 20 2020-02-10 0
Let’s plot the data again to confirm that we didn’t mess anything up.
Classwork: Plot No. 2
Yep! This looks about the same as the last one, but now we have no missing values.
Calculating new daily cases
However, this data set has a slight problem: it represents the total number of confirmed cases. We want to plot the number of new cases each day. We can calculate this by taking the difference between the number of confirmed cases today and the number of confirmed cases yesterday.
To get yesterday’s number of confirmed cases, we can use the lag()
function. This function shifts the column up by one row, so that the
value in the first row is now in the second row, the value in the second
row is now in the third row, and so on.
We’ll be left with 1 NA value in the first row, as there is no value
before the first row. We can replace this with 0 using the default
argument.
(The opposite of lag() is lead(), which shifts the column down by
one row.)
To check the data, I’ve used tail() and head() to look a little bit
further down the data set.
1 ch_covid_data <- ch_covid_data |>
2 mutate(yesterday_cases = lag(confirmed, 1, default = 0))
3 4 ch_covid_data |> tail(-30) |> head(20)
date confirmed yesterday_cases 31 2020-02-21 0 0 32 2020-02-22 0 0 33 2020-02-23 0 0 34 2020-02-24 1 0 35 2020-02-25 2 1 36 2020-02-26 12 2 37 2020-02-27 22 12 38 2020-02-28 32 22 39 2020-02-29 45 32 40 2020-03-01 57 45 41 2020-03-02 87 57 42 2020-03-03 120 87 43 2020-03-04 181 120 44 2020-03-05 243 181 45 2020-03-06 316 243 46 2020-03-07 365 316 47 2020-03-08 434 365 48 2020-03-09 625 434 49 2020-03-10 835 625 50 2020-03-11 1168 835
Now, we can calculate the number of new cases each day by taking the difference between the number of confirmed cases today and the number of confirmed cases yesterday.
1 ch_covid_data <- ch_covid_data |>
2 mutate(daily_cases = confirmed - yesterday_cases)
3 4 ch_covid_data |> tail(-30) |> head(20)
date confirmed yesterday_cases daily_cases 31 2020-02-21 0 0 0 32 2020-02-22 0 0 0 33 2020-02-23 0 0 0 34 2020-02-24 1 0 1 35 2020-02-25 2 1 1 36 2020-02-26 12 2 10 37 2020-02-27 22 12 10 38 2020-02-28 32 22 10 39 2020-02-29 45 32 13 40 2020-03-01 57 45 12 41 2020-03-02 87 57 30 42 2020-03-03 120 87 33 43 2020-03-04 181 120 61 44 2020-03-05 243 181 62 45 2020-03-06 316 243 73 46 2020-03-07 365 316 49 47 2020-03-08 434 365 69 48 2020-03-09 625 434 191 49 2020-03-10 835 625 210 50 2020-03-11 1168 835 333
Now, let’s make a line chart to visualize the number of new cases each day.
Classwork: Plot No. 3
Smoothing out the data
We can see another problem from this: the number of new cases each day is very noisy. Sometimes, new cases are reported in batches, often after a weekend. This is going to make it hard to see any trends in the data.
We can fix this by taking a rolling average of the number of new cases each day. This will smooth out the data and make it easier to see any trends.
The easiest way to do this is to use the rollmean() function from the
zoo package. You might need to install the package first.
1 install.packages("zoo")
In the rollmean() function, k is the number of days to average over.
Here, I’ve chosen 14 days, meaning that the new value will be the
average of the 7 days before and after the analyzed row. This should
smooth out the data enough to make it easier to see any trends.
1 library(zoo)
1 ch_covid_data <- ch_covid_data |>
2 mutate(year_day = yday(date)) |>
3 mutate(daily_cases_smoothed = rollmean(daily_cases, k=14, fill=0))
4 5 ch_covid_data |>
6 tail(-30) |>
7 head(20) |>
8 select(date, daily_cases, daily_cases_smoothed)
date daily_cases daily_cases_smoothed 31 2020-02-21 0 2.285714 32 2020-02-22 0 3.214286 33 2020-02-23 0 4.071429 34 2020-02-24 1 6.214286 35 2020-02-25 1 8.571429 36 2020-02-26 10 12.928571 37 2020-02-27 10 17.357143 38 2020-02-28 10 22.571429 39 2020-02-29 13 26.071429 40 2020-03-01 12 31.000000 41 2020-03-02 30 44.571429 42 2020-03-03 33 59.500000 43 2020-03-04 61 82.571429 44 2020-03-05 62 107.357143 45 2020-03-06 73 137.428571 46 2020-03-07 49 166.285714 47 2020-03-08 69 188.714286 48 2020-03-09 191 262.285714 49 2020-03-10 210 337.214286 50 2020-03-11 333 419.000000
To make sure this works, let’s plot the data again.
Classwork: Plot No. 4
This looks much better! The smoothed line is much easier to read than the noisy line.
Creating a ribbon plot
Our next step is to convert this from a line chart to a spiral chart.
We’ll start by creating a ribbon plot, where the area between two lines
is filled in. Using geom_ribbon(), we can fill in the area between two
lines.
geom_ribbon() takes two arguments, ymin and ymax, which are the
lower and upper bounds of the ribbon.
Here, we’ll set ymin to -daily_cases_smoothed and ymax to
daily_cases_smoothed. This will create a ribbon plot where the area
between -daily_cases_smoothed and daily_cases_smoothed is filled in.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = date,
6 ymin = -daily_cases_smoothed,
7 ymax = daily_cases_smoothed8 ),
9 fill = "#eeaaaa",
10 color = "#662222"
11 )
Using polar coordinates
This is the basic structure of the spiral chart. Out next step is to wrap it around a circle.
This can be done with coord_polar(), which uses polar coordinates
instead of Cartesian coordinates. This will wrap the chart around a
circle, creating a spiral chart.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = date,
6 ymin = -daily_cases_smoothed,
7 ymax = daily_cases_smoothed8 ),
9 fill = "#eeaaaa",
10 color = "#662222"
11 ) +
12 coord_polar()
What’s going on here? The chart still has X and Y coordinates, but now they’re arranged so that Y=0 is at the center of the circle, and the X axis wraps around the circle. This is similar to how coordinates work around a globe, where the south pole is at 0 degrees north, and the east-west axis wraps around the globe.
Transforming a circle to a spiral
Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0. ℹ Please use `linewidth` instead.
Let’s think about polar coordinates for a moment. In polar coordinates, the X axis is the angle around the circle, and the Y axis is the distance from the center of the circle. This means that the further out you go, the higher the Y value.
So if we plotted something with, for example, the X axis as the day of the year and the Y axis as the year, we’d get concentric circles, with each year being a circle. We can see this in the two examples, with Cartesian and polar coordinates, below.
Now: this might take some work to wrap your mind around: if we keep the X axis as the day of the year, but change the Y axis to be the number of days since 2020, we get a spiral.This is because the Y axis is now the distance from the center of the circle, and the further out you go, the higher the Y value. When we get to the end of the year, we jump to the next year, and the spiral continues.
So to convert our circle plot to a spiral plot, we need to change the Y axis from the year to the number of days since 2020.
We’ll first make a new column, year_num that is the number of years
since 2020. We’ll then make a new column, days_since_2020, that is the
number of days since 2020. This is the number of years since 2020 times
365, plus the day of the year.
1 ch_covid_data <- ch_covid_data |>
2 mutate(year_num = year(date) - 2020) |>
3 mutate(days_since_2020 = year_num * 365 + year_day)
Just to make sure everything is working, let’s just plot the center line of the spiral.
If we look closely, we can see a small issue! There’s a gap between the end of one year and the start of the next.
This is because 2020 was a leap year: it has 366 days. When we wrap back around, we’re missing a day.
We could potentially fix this with a whole bunch of math, but it’s much easier to just filter out the leap day. While this isn’t data-perfect solution, it’s good enough for this sort of visualization.
We have to do this before we calculate the days_since_2020 column, as
we’re using the year_day column in the calculation.
1 ch_covid_data <- ch_covid_data |>
2 mutate(year_num = year(date) - 2020) |>
3 filter(year_day <= 365) |>
4 mutate(days_since_2020 = year_num * 365 + year_day)
We’ll also convert the year_day column to a date, so that ggplot knows
how to label it: seeing the month of the year is much more informative
than just a random number. Note that we have to specify a year for the
origin, but this is arbitrary; we could have chosen any year.
1 ch_covid_data <- ch_covid_data |>
2 mutate(year_day = as_date(year_day, origin = "1970-01-01"))
Now when we plot the center line of the spiral, we can see that the gap is gone.
This is why it’s good to visualize your data as you go along: it’s much easier to spot mistakes when you can see them.
Making the spiral chart the easy way
A somewhat cheesy way to complete this spiral would then simply be to alter the size of the line based on the number of new cases each day. This will make the spiral look like it’s “breathing”, with the line getting thicker when there are more cases.
We can alter the relative size of the line using
scale_size_continuous(). This will make the line thicker or thinner
based on the value of daily_cases_smoothed.
This would be a valid way to do it! However, it doesn’t look quite as nice as the original spiral chart. The original chart has a ribbon plot, where the area between two lines is filled in. This makes it look like a spiral, rather than just a weird leech sitting in the water.
Making the spiral chart the hard way
So, we need to go back to our ribbon plot we did earlier in class, and think about ways to combine this with the spiral chart. In the ribbon plot, the Y axis is the total amount of daily cases, whereas in the spiral chart, the Y axis is the number of days since 2020.
This method was adapted from This page on Github, which was a much cleaner solution than the one I’d implemented on my own.
The number of daily cases is huge compared to the number of days since 2020. We need to rescale the number of daily cases to make it fit on the same scale as the number of days since 2020. We’re therefore going to re-scale the number of daily cases to be a smaller, arbitrary number.
First, we’ll re-scale the number of daily cases to be between 0 and 1, by dividing by the maximum number of daily cases. This will make the number of daily cases relative to the maximum number of daily cases.
We can use summary() to check that the maximum value of
rescaled_daily_cases is 1.
1 ch_covid_data <- ch_covid_data |>
2 mutate(rescaled_daily_cases = daily_cases_smoothed / max(daily_cases_smoothed))
3 4 ch_covid_data |> select(rescaled_daily_cases, daily_cases_smoothed) |> summary()
rescaled_daily_cases daily_cases_smoothed Min. :0.000000 Min. : 0.0 1st Qu.:0.007261 1st Qu.: 252.8 Median :0.044859 Median : 1561.6 Mean :0.105660 Mean : 3678.3 3rd Qu.:0.102114 3rd Qu.: 3554.8 Max. :1.000000 Max. :34812.2
Now, we can actually just add our two Y axes together, and we’ll get a spiral chart with a ribbon plot. It’s perfectly valid to do a little light calculation inside ggplot itself.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = year_day,
6 ymin = days_since_2020 - rescaled_daily_cases,
7 ymax = days_since_2020 + rescaled_daily_cases,
8 group = year_num9 ), fill = "#eeaaaa", color = "#662222") +
10 geom_line(aes(x = year_day, y = days_since_2020, group=year_num))
However, we still don’t see anything; the ribbon only goes between 0 and 1, which is too small to see. We need to re-scale the number of daily cases to be a larger number.
This requires a little experimentation, and in this case, you may wish to save our magic number as a variable, so that you can easily change it later. I can then play around with different numbers to see what looks good, and it will be automatically applied to different areas of the chart.
1 ribbon_multiplier = 300
2 3 ch_covid_data |>
4 ggplot() +
5 geom_ribbon(6 aes(7 x = year_day,
8 ymin = days_since_2020 - rescaled_daily_cases * ribbon_multiplier,
9 ymax = days_since_2020 + rescaled_daily_cases * ribbon_multiplier,
10 group = year_num11 ), fill = "#eeaaaa", color = "#662222") +
12 geom_line(aes(x = year_day, y = days_since_2020, group=year_num))
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = year_day,
6 ymin = days_since_2020 - rescaled_daily_cases * ribbon_multiplier,
7 ymax = days_since_2020 + rescaled_daily_cases * ribbon_multiplier,
8 group = year_num9 ), fill = "#eeaaaa", color = "#662222") +
10 geom_line(aes(x = year_day, y = days_since_2020, group=year_num)) +
11 coord_polar()
Cleaning up the chart
That’s basically it! We’ve rotated the data in our minds enough to form it into the shape of a spiral, having manipulated the data very far from the two columns we started with. Best of all, because we did this all through code, we could easily change the country, the smoothing, or the styling, and the chart would automatically update. Feel free to delete your code, go back to the beginning, and try it all again with a different country. You should get a different spiral chart, but with the same basic structure.
All that’s left to do is make it look a little cleaner and more presentable.
First, we notice that the line starts at the very center of the spiral, which gives us a bit of a strange appearance. We can fix this by changing limits of the Y axis to below 0; this will move the center of the spiral down a little bit.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = year_day,
6 ymin = days_since_2020 - rescaled_daily_cases * ribbon_multiplier,
7 ymax = days_since_2020 + rescaled_daily_cases * ribbon_multiplier,
8 group = year_num9 ), fill = "#eeaaaa", color = "#662222") +
10 geom_line(aes(x = year_day, y = days_since_2020, group=year_num)) +
11 coord_polar() +
12 lims(y = c(-300, 1300))
Next, I’ve changed our base theme to theme_minimal() and lightened the
center line of the spiral a little bit using alpha = 0.2, which
controls the transparency of the line.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = year_day,
6 ymin = days_since_2020 - rescaled_daily_cases * ribbon_multiplier,
7 ymax = days_since_2020 + rescaled_daily_cases * ribbon_multiplier,
8 group = year_num9 ), fill = "#eeaaaa", color = "#662222") +
10 geom_line(aes(x = year_day, y = days_since_2020, group=year_num), alpha =0.2) +
11 coord_polar() +
12 lims(y = c(-300, 1300)) +
13 theme_minimal()
Third, dividing a year into eighths is a little cursed Because we
changed the X axis to a date, we can use scale_x_date() to change the
breaks to be every two months.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = year_day,
6 ymin = days_since_2020 - rescaled_daily_cases * ribbon_multiplier,
7 ymax = days_since_2020 + rescaled_daily_cases * ribbon_multiplier,
8 group = year_num9 ), fill = "#eeaaaa", color = "#662222") +
10 geom_line(aes(x = year_day, y = days_since_2020, group=year_num), alpha =0.2) +
11 coord_polar() +
12 lims(y = c(-300, 1300)) +
13 theme_minimal() +
14 scale_x_date(date_breaks = "2 months", date_labels = "%b")
Finally, we can remove the Y axis, the Y axis labels, the Y axis ticks, and the Y axis title. This will clean up the chart a little bit, as we don’t need the Y axis for a spiral chart. This leaves us with a similar appearance to the original spiral chart.
1 ch_covid_data |>
2 ggplot() +
3 geom_ribbon(4 aes(5 x = year_day,
6 ymin = days_since_2020 - rescaled_daily_cases * ribbon_multiplier,
7 ymax = days_since_2020 + rescaled_daily_cases * ribbon_multiplier,
8 group = year_num9 ), fill = "#eeaaaa", color = "#662222") +
10 geom_line(aes(x = year_day, y = days_since_2020, group=year_num), alpha =0.2) +
11 coord_polar() +
12 lims(y = c(-300, 1300)) +
13 theme_minimal() +
14 scale_x_date(date_breaks = "2 months", date_labels = "%b") +
15 theme(legend.position = "none",
16 axis.text.y = element_blank(),
17 axis.ticks.y = element_blank(),
18 axis.title = element_blank(),
19 panel.grid.major = element_blank()
20 )
From here, we’d be just about ready to export the chart and share it with the world.
Just about as clean as the New York Times!