5.1 Viewing univariate statistics
5.1.1 Skim in SkimR
The video and the OpenStax textbook explain the key measure of dispersion and measures of central tendency. They are easy to view in R and there are many tools to do this. The most complete is the skim command in the skimr package.
Note: For reasons I can’t figure out, skim won’t run inside this document. I will post examples in Blackboard.
#skim_without_charts(my_gapminder_data_2003)skim groups variables by type:
- character or text data
- factor or categorical data
- numeric or continuous data
As you can see above, skim provides a lot of information in a small space. For continuous data (numerics in R) skim provides:
- variable name (variable),
- missing values (n_missing),
- the number of observations or cases (complete and n),
- the mean (mean),
- standard deviation (sd),
- the minimum value (p0),
- the 25th percentile (p25),
- the 50th percentile the median (p50),
- the 75th percentile (p75),
- the maximum value (p100),
- and a small histogram (noted as hist).
For categorical data (factors in R) skim provides a count of each category.
It also lists character data.
5.1.2 descr in sjmisc
An alternative is the descr command in the sjmisc package.
## Descriptive Statistics
## my_gapminder_data_2003
## N: 163
##
## co2EmissionsPerPerson giniCoefficent mortalityRate population
## ----------------- ----------------------- ---------------- --------------- ---------------
## Mean 4.86 40.04 49.08 35912207.98
## Std.Dev 7.26 7.84 51.31 138090625.04
## Min 0.02 25.20 3.20 28600.00
## Q1 0.46 33.70 9.51 1450000.00
## Median 2.00 40.00 24.80 7110000.00
## Q3 7.02 43.70 81.70 22500000.00
## Max 60.60 64.10 217.00 1320000000.00
## MAD 2.69 7.71 27.81 9530152.80
## IQR 6.56 9.95 71.85 20510000.00
## CV 1.49 0.20 1.05 3.85
## Skewness 3.69 0.58 1.15 7.88
## SE.Skewness 0.19 0.19 0.19 0.19
## Kurtosis 21.21 0.19 0.19 65.43
## N.Valid 161.00 163.00 163.00 163.00
## Pct.Valid 98.77 100.00 100.00 100.00
##
## Table: Table continues below
##
##
##
## studentSpendingPerCapita year
## ----------------- -------------------------- ---------
## Mean 0.16 2003.00
## Std.Dev 0.05 0.00
## Min 0.05 2003.00
## Q1 0.11 2003.00
## Median 0.15 2003.00
## Q3 0.20 2003.00
## Max 0.28 2003.00
## MAD 0.06 0.00
## IQR 0.08 0.00
## CV 0.34 0.00
## Skewness 0.20 NaN
## SE.Skewness 0.31 0.19
## Kurtosis -0.90 NaN
## N.Valid 61.00 163.00
## Pct.Valid 37.42 100.00It lists similar statistics as skim.
- variable name (var),
- type of data (type),
- label for the variable if available (label)
- the number of observations or cases (n),
- missing values (NA_prc),
- the mean (mean),
- standard deviation (sd),
- standard error (se),
- median (md),
- the trimmed mean (trimmed),
- minimum value, maximum value, and range (range),
- iqr (the interquartile range),
- a measure of skewness (skew).
5.1.3 Summary in Base R
Base R also has a similar command, summary. It does not provide standard deviation. It does rank the top categories of categorical data.
Below I ran a my small WVS dataset in summary.
## N_REGION_WVS
## ET-OR Oromia :261
## ET-AM Amara :245
## ET-SO Somali :208
## ET-SN Southern Nations, Nationalities and Peoples:120
## ET-TI Tigrai :113
## ET-AA Addis Ababa : 92
## (Other) :191
## N_TOWN G_TOWNSIZE G_TOWNSIZE2
## Boloso Sore - Wereda : 50 5,000-10,000 :358 Under 5,000 :196
## Hawassa - City administation: 36 10,000-20,000 :287 5000-20000 :645
## Omonada - Wereda : 31 20,000-50,000 :199 20000-100000 :357
## Dedo- Wereda : 30 50,000-100,000:158 100000-500000 : 23
## Dire Dawa-Urban : 30 2,000-5,000 :103 500000 and more: 9
## Gedeb - Wereda : 30 Under 2,000 : 93
## (Other) :1023 (Other) : 32
## H_SETTLEMENT H_URBRURAL
## Capital city :101 Urban:296
## Regional center :115 Rural:934
## District center :284
## Another city, town (not a regional or district center): 99
## Village :631
##
##
## Q34
## Agree strongly :541
## Agree :372
## Neither agree nor disagree: 43
## Disagree :206
## Disagree strongly : 65
## NA's : 3
##
## Q241
## An essential characteristic of democracy :531
## Not an essential characteristic of democracy:299
## 9 : 87
## 8 : 79
## 7 : 49
## (Other) :142
## NA's : 43
## Q242
## Not an essential characteristic of democracy:416
## An essential characteristic of democracy :336
## 2 : 75
## 9 : 69
## 3 : 55
## (Other) :230
## NA's : 49
## Q243
## An essential characteristic of democracy :808
## 9 :147
## 8 : 92
## Not an essential characteristic of democracy: 53
## 7 : 29
## (Other) : 42
## NA's : 59
## Q244
## An essential characteristic of democracy :547
## Not an essential characteristic of democracy:166
## 9 :122
## 8 : 96
## 7 : 67
## (Other) :182
## NA's : 50
## Q245
## An essential characteristic of democracy :431
## Not an essential characteristic of democracy:317
## 9 : 96
## 8 : 72
## 7 : 49
## (Other) :196
## NA's : 69
## Q246
## An essential characteristic of democracy :628
## 9 :158
## Not an essential characteristic of democracy: 95
## 8 : 94
## 7 : 52
## (Other) :110
## NA's : 93
## Q247
## An essential characteristic of democracy :459
## Not an essential characteristic of democracy:262
## 9 : 97
## 8 : 75
## 2 : 58
## (Other) :231
## NA's : 48
## Q248
## An essential characteristic of democracy :690
## 9 :152
## 8 :110
## Not an essential characteristic of democracy: 82
## 7 : 58
## (Other) : 92
## NA's : 46
## Q249 Q250
## An essential characteristic of democracy :817 Absolutely important:1044
## 9 :173 9 : 47
## 8 : 84 5 : 44
## Not an essential characteristic of democracy: 36 8 : 35
## 7 : 29 Not at all important: 14
## (Other) : 31 (Other) : 36
## NA's : 60 NA's : 10
## Q251 Q252
## 5 :282 Not satisfied at all:328
## Not at all democratic:218 5 :242
## Completely democratic:198 Completely satisfied:137
## 6 :123 6 : 96
## 7 : 94 7 : 91
## (Other) :299 (Other) :317
## NA's : 16 NA's : 19
## Q253 Q254 Q255
## A great deal of respect:169 Very proud :1061 Very close :683
## Fairly much respect :468 Quite proud : 129 Close :371
## Not much respect :371 Not very proud : 25 Not very close :107
## No respect at all :207 Not at all proud: 14 Not close at all: 64
## NA's : 15 NA's : 1 NA's : 5
##
##
## Q256 Q257 Q258
## Very close :542 Very close :355 Very close :172
## Close :457 Close :512 Close :290
## Not very close :165 Not very close :246 Not very close :400
## Not close at all: 60 Not close at all:104 Not close at all:266
## NA's : 6 NA's : 13 NA's :102
##
##
## Q259 Q260 Q261 E1_LITERACY
## Very close :159 Male :622 1990 : 89 Literate :826
## Close :226 Female:608 2000 : 83 Illiterate:404
## Not very close :374 1998 : 77
## Not close at all:349 1995 : 72
## NA's :122 1985 : 66
## 1992 : 61
## (Other):782
## Q275R Q251_recode
## Lower :803 Not Democratic :318
## Middle:241 Somewhat Democratic:572
## Higher:180 Democratic :324
## NA's : 6 NA's : 16
##
##
## 5.1.4 Create a Codebook Using the Summarytools package
Use the following command to create a nice codebook. It will open as a file in the bottom right quadant of RStudio. It creates an HTML file that can be opened in any web browser and printed.
view(dfSummary(ethiopia_small))