AKForrest
Reputation: 66
R: Converting a Wide Table to Narrow with Specific Columns
I have a wide table with 161 variables that I would like to transform into a narrow table with only 13 variables. As far as I can tell, I cannot use pivot_longer because I need to input an additional column for Year, and then only select certain columns from the wide table for each row.
I started with a blank data frame:
subset_gathered <- read.csv(text = "FIPS_State, Place_Code, Place_Name, Longitude, Latitude, County, Closest Metro, Year, Population_Density, Labor Force, Employed, Unemployed, Unemployment Rate",
colClasses = c("character", "integer", "character", "numeric", "numeric", "character", "numeric","integer", "numeric", "integer", "integer", "integer", "numeric") )
and then I tried a for loop to pick up each of the relevant columns (actual solution would repeat for years 1990-2018, which each variable is a different column)
for(i in subset[1:5,]){
temp1990 <- c(subset$`FIPS State`[i], subset$`Place Code`[i], subset$Name[i], subset$longitude[i], subset$latitude[i], subset$COUNTY[i], subset$closest_metro[i], 1990, subset$density1990[i], subset$`Labor Force.90`[i], subset$Employed.90[i], subset$`Unemployment Level.90`[i], subset$`Unemployment Rate.90`[i])
subset_gathered[nrow(subset_gathered)+1,]<- temp1990
temp1991 <- c(subset$`FIPS State`[i], subset$`Place Code`[i], subset$Name[i], subset$longitude[i], subset$latitude[i], subset$COUNTY[i], subset$closest_metro[i], 1991, subset$density1991[i], subset$`Labor Force.91`[i], subset$Employed.91[i], subset$`Unemployment Level.91`[i], subset$`Unemployment Rate.91`[i])
subset_gathered[nrow(subset_gathered)+1,]<- temp1991
}
I assume the results are wacky because it is running each vector at a time, and not the row
head(subset_gathered)
FIPS_State Place_Code Place_Name Longitude Latitude County Closest.Metro Year Population_Density Labor.Force
1 1 1 1 1 1 124 124 124 124 124
2 1 1 1 1 1 124 124 124 124 124
3 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
4 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
5 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
6 <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA> <NA>
Employed Unemployed Unemployment.Rate
1 Abbeville city Abbeville city Abbeville city
2 Abbeville city Abbeville city Abbeville city
3 <NA> <NA> <NA>
4 <NA> <NA> <NA>
5 <NA> <NA> <NA>
6 <NA> <NA> <NA>
If I were to isolate the rows outside of a for loop, I get exactly the results I want:
temp1990 <- c(subset$`FIPS State`[1], subset$`Place Code`[1], subset$Name[1], subset$longitude[1], subset$latitude[1], subset$COUNTY[1], subset$closest_metro[1], 1990, subset$density1990[1], subset$`Labor Force.90`[1], subset$Employed.90[1], subset$`Unemployment Level.90`[1], subset$`Unemployment Rate.90`[1])
subset_gathered[nrow(subset_gathered)+1,]<- temp1990
temp1991 <- c(subset$`FIPS State`[1], subset$`Place Code`[1], subset$Name[1], subset$longitude[1], subset$latitude[1], subset$COUNTY[1], subset$closest_metro[1], 1991, subset$density1991[1], subset$`Labor Force.91`[1], subset$Employed.91[1], subset$`Unemployment Level.91`[1], subset$`Unemployment Rate.91`[1])
subset_gathered[nrow(subset_gathered)+1,]<- temp1991
FIPS_State Place_Code Place_Name Longitude Latitude County Closest.Metro Year Population_Density
1 1 124 Abbeville city -85.2513 31.567949 Henry County 24.3361834333029 1990 7.85948198868711e-05
2 1 124 Abbeville city -85.2513 31.567949 Henry County 24.3361834333029 1991 7.96367966656743e-05
Labor.Force Employed Unemployed Unemployment.Rate
1 6867 6539 328 4.8
2 6648 6106 542 8.2
I then tried to write an apply function, but it returned only a single row, and it should have returned 5 (there are 5 rows with data in subset).
subset_gathered[nrow(subset_gathered)+1,]<- apply(subset, 1, function(x) c(x[1], x[2], x[3], x[37], x[38], x[41], x[40], 1990, x[7], x[118], x[119], x[120], x[121]))
FIPS_State Place_Code Place_Name Longitude Latitude County Closest.Metro Year Population_Density
1 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1990 7.859482e-05
Labor.Force Employed Unemployed Unemployment.Rate
1 6867 6539 328 4.8
Am I on the right track here? I think using a for loop would work in Python or SAS, but apply always seems to be the right answer in R, but I still cannot get the results I need.
dput:
subset<- structure(list(`FIPS State` = c(1, 1, 1, 1, 1), `Place Code` = c(124,
484, 676, 1228, 1396), Name = c("Abbeville city", "Addison town",
"Akron town", "Aliceville city", "Allgood town"), `1990 Pop` = c(3168,
639, 468, 3052, 461), GEOID = c(100124, 100484, 100676, 101228,
101396), USPS = c("AL", "AL", "AL", "AL", "AL"), density1990 = c(7.85948198868711e-05,
8.31165452653486e-05, 0.000325678496868476, 0.000262064228061137,
0.000171694599627561), density1991 = c(7.96367966656743e-05,
8.28563995837669e-05, 0.000329157967988866, 0.000260604499398935,
0.000176163873370577), density1992 = c(7.89669544507294e-05,
8.37669094693028e-05, 0.000331941544885177, 0.000259402369912416,
0.00017877094972067), density1993 = c(7.89421454798055e-05, 8.49375650364204e-05,
0.000331941544885177, 0.000256568778979907, 0.000183612662942272
), density1994 = c(7.87188647414905e-05, 8.6888657648283e-05,
0.0003312456506611, 0.000252189592993302, 0.000187709497206704
), density1995 = c(7.77381836037938e-05, 7.39452771986497e-05,
0.000332684436284319, 0.000248944989340191, 0.000195857866765266
), density1996 = c(7.63486725020343e-05, 7.4934144204278e-05,
0.000334772413917902, 0.000251349425249651, 0.000203304933942652
), density1997 = c(7.62246090108058e-05, 7.54835147629603e-05,
0.000334772413917902, 0.000249460225606504, 0.000213358474632124
), density1998 = c(7.60012947265945e-05, 7.58131370981698e-05,
0.000338252376640541, 0.0002473134078302, 0.000222667308603857
), density1999 = c(7.51328502879948e-05, 7.66921299920616e-05,
0.000338948369185069, 0.000244479608365479, 0.000230859082498983
), density2000 = c(7.41651550564123e-05, 7.66921299920616e-05,
0.000339644361729597, 0.000225673484645058, 0.000246497923571494
), density2001 = c(7.29741455406185e-05, 7.7021752327271e-05,
0.000329204473561679, 0.000225244121089797, 0.000245008510136017
), density2002 = c(7.21801391967559e-05, 7.77908711094264e-05,
0.00031598061521565, 0.000224900630245588, 0.000243891450059409
), density2003 = c(7.15101963441219e-05, 7.84501157798452e-05,
0.000307628704681315, 0.000221637467225606, 0.00024351909670054
), density2004 = c(7.08898788879792e-05, 7.93291086737371e-05,
0.00029649282396887, 0.000220091758426668, 0.000242029683265062
), density2005 = c(7.00527969547999e-05, 7.47440401111995e-05,
0.000233954174226123, 0.000214997442884744, 0.000239751372251362
), density2006 = c(6.97298585291218e-05, 7.59743946809311e-05,
0.000227205496123447, 0.00021432028558432, 0.000239007957143606
), density2007 = c(6.91585059298451e-05, 7.68971606082299e-05,
0.000222143987546439, 0.000212881326320918, 0.00023826454203585
), density2008 = c(6.8115166400731e-05, 7.71022197031852e-05,
0.000215957699285652, 0.000212542747670706, 0.000237149419374215
), density2009 = c(6.7419606714655e-05, 7.78199265355286e-05,
0.000207521851657307, 0.000211103788407304, 0.000234547466497069
), density2010 = c(6.67737298632987e-05, 7.7717396988051e-05,
0.00020021078371274, 0.00021042663110688, 0.000231202098512166
), density2011 = c(6.65004742723402e-05, 7.69996901557075e-05,
0.000194024495451953, 0.000205347951353697, 0.000231573806066044
), density2012 = c(6.53080846167217e-05, 7.62819207541575e-05,
0.000194238820289117, 0.000205286788530157, 0.000231202012572632
), density2013 = c(6.48857660254055e-05, 7.61794069120457e-05,
0.00019198677599591, 0.000202371348887106, 0.00023083030515692
), density2014 = c(6.44634620351942e-05, 7.58718184588342e-05,
0.000190860753849307, 0.000200340867125697, 0.000231202012572632
), density2015 = c(6.42895721568719e-05, 7.53591710368151e-05,
0.000189734731702703, 0.000199579436465168, 0.00023083030515692
), density2016 = c(6.39169509890384e-05, 7.52566415524112e-05,
0.000186919676336194, 0.000198225781957562, 0.00023083030515692
), density2017 = c(6.37430611107161e-05, 7.46414646459882e-05,
0.000186919781574101, 0.000197548954703759, 0.000231202012572632
), density2018 = c(6.36685368771494e-05, 7.4231339097472e-05,
0.000185793758793534, 0.000194672438875096, 0.000231202012572632
), minpop = c(2563, 637, 330, 2301, 461), longitude = c(-85.2513,
-87.177851, -87.738779, -88.154427, -86.516109), latitude = c(31.567949,
34.202689, 32.879495, 33.126276, 33.907623), is_metro = c(0,
0, 0, 0, 0), closest_metro = c(24.3361834333029, 28.0961047219205,
27.2224172144133, 34.7564812052357, 30.296520864832), COUNTY = c("Henry County",
"Winston County", "Hale County", "Pickens County", "Blount County"
), COUNTY2 = c(NA_character_, NA_character_, NA_character_, NA_character_,
NA_character_), `LAUS Code` = c("CN0106700000000", "CN0113300000000",
"CN0106500000000", "CN0110700000000", "CN0100900000000"), `State FIPS` = c("01",
"01", "01", "01", "01"), `County FIPS` = c("067", "133", "065",
"107", "009"), `Labor Force.00` = c(7634, 11687, 6976, 8597,
25106), Employed.00 = c(7259, 10776, 6557, 7952, 24231), `Unemployment Level.00` = c(375,
911, 419, 645, 875), `Unemployment Rate.00` = c(4.9, 7.8, 6,
7.5, 3.5), `Labor Force.01` = c(7542, 10737, 7019, 8440, 25305
), Employed.01 = c(7137, 9732, 6513, 7732, 24393), `Unemployment Level.01` = c(405,
1005, 506, 708, 912), `Unemployment Rate.01` = c(5.4, 9.4, 7.2,
8.4, 3.6), `Labor Force.02` = c(7530, 10478, 7054, 8248, 25757
), Employed.02 = c(7075, 9561, 6462, 7569, 24366), `Unemployment Level.02` = c(455,
917, 592, 679, 1391), `Unemployment Rate.02` = c(6, 8.8, 8.4,
8.2, 5.4), `Labor Force.03` = c(7494, 10407, 7086, 8138, 25900
), Employed.03 = c(7059, 9418, 6485, 7475, 24702), `Unemployment Level.03` = c(435,
989, 601, 663, 1198), `Unemployment Rate.03` = c(5.8, 9.5, 8.5,
8.1, 4.6), `Labor Force.04` = c(7548, 10337, 7049, 8027, 26208
), Employed.04 = c(7122, 9580, 6516, 7416, 25101), `Unemployment Level.04` = c(426,
757, 533, 611, 1107), `Unemployment Rate.04` = c(5.6, 7.3, 7.6,
7.6, 4.2), `Labor Force.05` = c(7469, 10365, 7053, 7960, 26446
), Employed.05 = c(7146, 9840, 6669, 7527, 25491), `Unemployment Level.05` = c(323,
525, 384, 433, 955), `Unemployment Rate.05` = c(4.3, 5.1, 5.4,
5.4, 3.6), `Labor Force.06` = c(7506, 10591, 7170, 8113, 26770
), Employed.06 = c(7204, 10067, 6792, 7703, 25902), `Unemployment Level.06` = c(302,
524, 378, 410, 868), `Unemployment Rate.06` = c(4, 4.9, 5.3,
5.1, 3.2), `Labor Force.07` = c(7540, 10287, 6976, 8015, 26629
), Employed.07 = c(7135, 9712, 6602, 7613, 25780), `Unemployment Level.07` = c(405,
575, 374, 402, 849), `Unemployment Rate.07` = c(5.4, 5.6, 5.4,
5, 3.2), `Labor Force.08` = c(7376, 9984, 6991, 7925, 26698),
Employed.08 = c(6807, 9120, 6473, 7359, 25453), `Unemployment Level.08` = c(569,
864, 518, 566, 1245), `Unemployment Rate.08` = c(7.7, 8.7,
7.4, 7.1, 4.7), `Labor Force.09` = c(7132, 9519, 6869, 8042,
26480), Employed.09 = c(6334, 7832, 5890, 6942, 23832), `Unemployment Level.09` = c(798,
1687, 979, 1100, 2648), `Unemployment Rate.09` = c(11.2,
17.7, 14.3, 13.7, 10), `Labor Force.10` = c(7259, 9883, 6445,
7699, 24906), Employed.10 = c(6469, 8304, 5473, 6687, 22460
), `Unemployment Level.10` = c(790, 1579, 972, 1012, 2446
), `Unemployment Rate.10` = c(10.9, 16, 15.1, 13.1, 9.8),
`Labor Force.11` = c(7270, 9819, 6296, 7570, 25123), Employed.11 = c(6563,
8492, 5426, 6663, 22939), `Unemployment Level.11` = c(707,
1327, 870, 907, 2184), `Unemployment Rate.11` = c(9.7, 13.5,
13.8, 12, 8.7), `Labor Force.12` = c(7030, 9607, 6212, 7467,
24960), Employed.12 = c(6447, 8647, 5499, 6709, 23244), `Unemployment Level.12` = c(583,
960, 713, 758, 1716), `Unemployment Rate.12` = c(8.3, 10,
11.5, 10.2, 6.9), `Labor Force.13` = c(6954, 9644, 6208,
7420, 24887), Employed.13 = c(6421, 8737, 5486, 6727, 23325
), `Unemployment Level.13` = c(533, 907, 722, 693, 1562),
`Unemployment Rate.13` = c(7.7, 9.4, 11.6, 9.3, 6.3), `Labor Force.14` = c(6775,
9630, 6032, 7760, 24527), Employed.14 = c(6267, 8802, 5436,
7124, 23023), `Unemployment Level.14` = c(508, 828, 596,
636, 1504), `Unemployment Rate.14` = c(7.5, 8.6, 9.9, 8.2,
6.1), `Labor Force.15` = c(6718, 9379, 6050, 8011, 24485),
Employed.15 = c(6262, 8673, 5580, 7469, 23163), `Unemployment Level.15` = c(456,
706, 470, 542, 1322), `Unemployment Rate.15` = c(6.8, 7.5,
7.8, 6.8, 5.4), `Labor Force.16` = c(6733, 9548, 5991, 7779,
24623), Employed.16 = c(6295, 8872, 5534, 7246, 23298), `Unemployment Level.16` = c(438,
676, 457, 533, 1325), `Unemployment Rate.16` = c(6.5, 7.1,
7.6, 6.9, 5.4), `Labor Force.17` = c(6713, 9693, 5945, 7718,
24725), Employed.17 = c(6385, 9203, 5590, 7302, 23726), `Unemployment Level.17` = c(328,
490, 355, 416, 999), `Unemployment Rate.17` = c(4.9, 5.1,
6, 5.4, 4), `Labor Force.90` = c(6867, 10482, 6204, 8606,
19168), Employed.90 = c(6539, 9241, 5725, 7818, 17955), `Unemployment Level.90` = c(328,
1241, 479, 788, 1213), `Unemployment Rate.90` = c(4.8, 11.8,
7.7, 9.2, 6.3), `Labor Force.91` = c(6648, 10435, 6111, 8775,
19132), Employed.91 = c(6106, 9113, 5547, 7906, 18021), `Unemployment Level.91` = c(542,
1322, 564, 869, 1111), `Unemployment Rate.91` = c(8.2, 12.7,
9.2, 9.9, 5.8), `Labor Force.92` = c(6762, 10927, 6425, 9052,
19553), Employed.92 = c(6231, 9895, 5762, 8061, 18334), `Unemployment Level.92` = c(531,
1032, 663, 991, 1219), `Unemployment Rate.92` = c(7.9, 9.4,
10.3, 10.9, 6.2), `Labor Force.93` = c(6553, 11880, 6972,
9171, 19929), Employed.93 = c(5905, 10978, 6284, 8110, 18868
), `Unemployment Level.93` = c(648, 902, 688, 1061, 1061),
`Unemployment Rate.93` = c(9.9, 7.6, 9.9, 11.6, 5.3), `Labor Force.94` = c(6395,
12306, 7192, 9006, 20263), Employed.94 = c(5911, 11567, 6505,
8237, 19408), `Unemployment Level.94` = c(484, 739, 687,
769, 855), `Unemployment Rate.94` = c(7.6, 6, 9.6, 8.5, 4.2
), `Labor Force.95` = c(6472, 12777, 7129, 8938, 20993),
Employed.95 = c(6018, 11981, 6453, 8140, 20055), `Unemployment Level.95` = c(454,
796, 676, 798, 938), `Unemployment Rate.95` = c(7, 6.2, 9.5,
8.9, 4.5), `Labor Force.96` = c(6440, 13038, 7136, 8877,
21631), Employed.96 = c(6004, 12211, 6549, 8041, 20918),
`Unemployment Level.96` = c(436, 827, 587, 836, 713), `Unemployment Rate.96` = c(6.8,
6.3, 8.2, 9.4, 3.3), `Labor Force.97` = c(6456, 13175, 7392,
8824, 22684), Employed.97 = c(6087, 12175, 6823, 8131, 22003
), `Unemployment Level.97` = c(369, 1000, 569, 693, 681),
`Unemployment Rate.97` = c(5.7, 7.6, 7.7, 7.9, 3), `Labor Force.98` = c(6412,
12661, 7368, 8699, 23611), Employed.98 = c(6087, 11934, 6811,
8002, 22839), `Unemployment Level.98` = c(325, 727, 557,
697, 772), `Unemployment Rate.98` = c(5.1, 5.7, 7.6, 8, 3.3
), `Labor Force.99` = c(6428, 12410, 7016, 8577, 23968),
Employed.99 = c(6093, 11623, 6488, 7683, 23297), `Unemployment Level.99` = c(335,
787, 528, 894, 671), `Unemployment Rate.99` = c(5.2, 6.3,
7.5, 10.4, 2.8), `Labor Force.18` = c(6766, 9781, 5991, 7805,
25006), `Unemployment Rate.18` = c(6466, 9371, 5686, 7449,
24128), `Unemployment Level` = c(300, 410, 305, 356, 878),
`Unemployment Rate` = c(4.4, 4.2, 5.1, 4.6, 3.5)), row.names = c(NA,
5L), class = "data.frame")
Upvotes: 0
Views: 279
Answers (1)
Uwe
Reputation: 42544
The difficulty is that the year is encoded in the column names in different ways, some columns use a 4 digit year, some columns a 2 digit year. (In addition, the names of the last 3 columns of subset seem to be misspelled at all.)
The year naming issue can be solved by reshaping all variables to long format, separating and completing the year from the variable names and then reshaping to the requested format:
library(data.table)
library(magrittr)
melt(setDT(subset),
id.vars = c("FIPS State", "Place Code", "Name", "longitude", "latitude",
"COUNTY", "closest_metro"),
measure.vars = patterns("\\d\\d$")) %>%
.[, c("variable", "Year") := tstrsplit(variable, "(?<=density)|\\.", perl = TRUE)] %>%
.[, Year := lubridate::ymd(Year, truncated = 2L) %>% year()] %>%
dcast(... ~ variable)
FIPS State Place Code Name longitude latitude COUNTY closest_metro Year Employed Labor Force 1: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1990 6539 6867 2: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1991 6106 6648 3: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1992 6231 6762 4: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1993 5905 6553 5: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1994 5911 6395 --- 141: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2014 23023 24527 142: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2015 23163 24485 143: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2016 23298 24623 144: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2017 23726 24725 145: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2018 NA 25006 Unemployment Level Unemployment Rate density 1: 328 4.8 7.859482e-05 2: 542 8.2 7.963680e-05 3: 531 7.9 7.896695e-05 4: 648 9.9 7.894215e-05 5: 484 7.6 7.871886e-05 --- 141: 1504 6.1 2.312020e-04 142: 1322 5.4 2.308303e-04 143: 1325 5.4 2.308303e-04 144: 999 4.0 2.312020e-04 145: NA 24128.0 2.312020e-04
Fixing the column names for 2018
The reshaped data show some values for 2018 to appear in the wrong columns or are missing at all. By inspecting subset, it seems that the last 3 columns have been given the wrong column names. This can be fixed by renaming the columns using data.tables's set_names() function beforehand:
library(data.table)
library(magrittr)
setDT(subset) %>%
setnames(c("Unemployment Rate.18", "Unemployment Level", "Unemployment Rate"),
c("Employed.18", "Unemployment Level.18", "Unemployment Rate.18")) %>%
melt(id.vars = c("FIPS State", "Place Code", "Name", "longitude", "latitude",
"COUNTY", "closest_metro"),
measure.vars = patterns("\\d\\d$")) %>%
.[, c("variable", "Year") := tstrsplit(variable, "(?<=density)|\\.", perl = TRUE)] %>%
.[, Year := lubridate::ymd(Year, truncated = 2L) %>% lubridate::year()] %>%
dcast(... ~ variable)
FIPS State Place Code Name longitude latitude COUNTY closest_metro Year Employed Labor Force 1: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1990 6539 6867 2: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1991 6106 6648 3: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1992 6231 6762 4: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1993 5905 6553 5: 1 124 Abbeville city -85.25130 31.56795 Henry County 24.33618 1994 5911 6395 --- 141: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2014 23023 24527 142: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2015 23163 24485 143: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2016 23298 24623 144: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2017 23726 24725 145: 1 1396 Allgood town -86.51611 33.90762 Blount County 30.29652 2018 24128 25006 Unemployment Level Unemployment Rate density 1: 328 4.8 7.859482e-05 2: 542 8.2 7.963680e-05 3: 531 7.9 7.896695e-05 4: 648 9.9 7.894215e-05 5: 484 7.6 7.871886e-05 --- 141: 1504 6.1 2.312020e-04 142: 1322 5.4 2.308303e-04 143: 1325 5.4 2.308303e-04 144: 999 4.0 2.312020e-04 145: 878 3.5 2.312020e-04
Upvotes: 1
Related Questions
- Wide to long with many different columns
- Wide table to long table
- How to convert dataframe from wide to long format with multiple columns r
- How to convert a complicated wide table into long table
- reshape long dataframe to wide for multiple columns
- reshape wide to long using data.table with multiple columns
- Transforming a data frame to wide table with R
- Long to wide on multiple columns by data.table
- Change a long table to wide table
- Converting a dataframe from "wide" format to "long" format in R