Recap
R is a language, Rstudio let's us use it
Mardown - Grey = code, white = notes
We set a working directory and read in data
R is sensitive (cases, NAs)
Functions learned
data$Gender # format: name of data $ column name setwd() # sets working directory read.csv() # brings in data head() tail() # peak at the first or last 6 rows View() # capital V! opens spreadsheet summary() # max, min, mean str() # structure of data names() # lists all the columns table() # summary table for a column
Class 2 : Cleaning data
Unclean data
table(data$`Case Type`)
## ## Brain Tumor CSDH ## 1 66 ## CSDH, epidural CSDH, Epidural ## 1 2 ## CSDH,other,Extensive brain hypoxia Epidural ## 1 1 ## Infection Other, Cervial injury ## 1 1 ## Other, Extradural lesion Other, Scalp Abscess ## 1 1 ## Other,Brain Tumor Others ## 1 6 ## Others skull fracture Others, Stroke ## 1 1 ## SDH Spine ## 3 19 ## TBI Tumor ## 626 41 ## Tumor(Thalamus) Tumor, Other,HCP ## 1 1 ## unknown ## 1
Clean data
table(data$`Case Type`)
## ## CSDH Other Spine TBI Tumor ## 70 12 20 630 45
Unclean data
names(data2)
## [1] "Patien's Age" "Admission Date" ## [3] "Total1" "Heart Rate/Min" ## [5] "Resp Rate/min" "Blood Pressure" ## [7] "/" "Discharge GCS" ## [9] "Discharge GOS" "Alcohol status" ## [11] "Gender" "Pupil Reactive?" ## [13] "Temp" "Spo2" ## [15] "Did planned operation occur" "Management" ## [17] "Case Type"
Spaces are bad
Symbols cause problems
Clean data
underscore instead of spaces
no symbols
names(data2)
## [1] "Age" "adm_date" "GCS" "HR" "RR" ## [6] "Sys_BP" "Dia_BP" "dc_GCS" "dc_GOS" "etoh" ## [11] "male" "pupils" "temp" "spo2" "tbi_surg" ## [16] "tbi_surg2" "dx"
Time to code
letters <- c( "a", "b", "c")
numbers <- c(1 , 2, 3)
mydata <- data.frame(letters, numbers)
mydata$chicken <- c("hard", "soft", "chewy")
mydata$food <- mydata$chicken
Back to our data
#load in data if you have not already done so #Make a new dataframe with only columns 13, 14, 8 data2 <- data[ , c(8, 9, 3)]
or
#data2 <- select(data, "Patien's Age", Gender , Case Type) data2 <- select(data, "Patien's Age", Gender , "Case Type")
head(data2)
## # A tibble: 6 x 3 ## `Patien's Age` Gender `Case Type` ## <int> <chr> <chr> ## 1 28 Male TBI ## 2 26 Female CSDH ## 3 0 <NA> TBI ## 4 24 Male TBI ## 5 19 Male TBI ## 6 0 Male TBI
Change column names
colnames(data2) <- c("age",
"gender",
"case_type")
head(data2)
## # A tibble: 6 x 3 ## age gender case_type ## <int> <chr> <chr> ## 1 28 Male TBI ## 2 26 Female CSDH ## 3 0 <NA> TBI ## 4 24 Male TBI ## 5 19 Male TBI ## 6 0 Male TBI
summary(data2$age)
## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's ## 0.00 16.00 26.00 29.64 40.00 280.00 2
Unclean column
table(data2$case_type)
## ## Brain Tumor CSDH ## 1 66 ## CSDH, epidural CSDH, Epidural ## 1 2 ## CSDH,other,Extensive brain hypoxia Epidural ## 1 1 ## Infection Other, Cervial injury ## 1 1 ## Other, Extradural lesion Other, Scalp Abscess ## 1 1 ## Other,Brain Tumor Others ## 1 6 ## Others skull fracture Others, Stroke ## 1 1 ## SDH Spine ## 3 19 ## TBI Tumor ## 626 41 ## Tumor(Thalamus) Tumor, Other,HCP ## 1 1 ## unknown ## 1
Recode function
data2$case_type <- car::recode(data2$case_type, "'Brain Tumor' = 'Tumor';
'Tumor(Thalamus)' = 'Tumor';
'Tumor, Other,HCP' = 'Tumor'")
## ## CSDH CSDH, epidural ## 66 1 ## CSDH, Epidural CSDH,other,Extensive brain hypoxia ## 2 1 ## Epidural Infection ## 1 1 ## Other, Cervial injury Other, Extradural lesion ## 1 1 ## Other, Scalp Abscess Other,Brain Tumor ## 1 1 ## Others Others skull fracture ## 6 1 ## Others, Stroke SDH ## 1 3 ## Spine TBI ## 19 626 ## Tumor unknown ## 44 1
Recode the rest of the column
copy paste my code
data2$case_type <- car::recode(data2$case_type, "'Brain Tumor' = 'Tumor';
'Tumor(Thalamus)' = 'Tumor';
'Tumor, Other,HCP' = 'Tumor';
'CSDH, epidural' = 'CSDH';
'CSDH, Epidural' = 'CSDH';
'CSDH,other,Extensive brain hypoxia' = 'CSDH';
'Other,Brain Tumor' = 'Tumor';
'Others skull fracture' = 'Other';
'Others, Stroke' = 'Other';
'Other, Scalp Abscess' = 'Other';
'Epidural' = 'TBI';
'Other, Cervial injury' = 'Spine';
'Other, Extradural lesion' = 'Other';
'Others' = 'Other';
'Infection' = 'Other';
'unknown' = 'Other';
'SDH' = 'TBI'")
table(data2$case_type)
## ## CSDH Other Spine TBI Tumor ## 70 12 20 630 45