Data Prep

Load Libraries

# if you haven't run this code before, you'll need to download the below packages first
# instructions on how to do this are included in the video
# but as a reminder, you use the packages tab to the right

library(psych) # for the describe() command
library(naniar) # for the gg_miss-upset() command
library(expss) # for the cross_cases() command
## Loading required package: maditr
## 
## To get total summary skip 'by' argument: take_all(mtcars, mean)
## 
## Use 'expss_output_rnotebook()' to display tables inside R Notebooks.
##  To return to the console output, use 'expss_output_default()'.
## 
## Attaching package: 'expss'
## The following object is masked from 'package:naniar':
## 
##     is_na

Import Data

# for the lab, you'll import the CSV file you downloaded along with the current file we're working in (an RMD file)
# for the homework, you'll download the CSV file from your chosen README page (should be titled arc_data_final.csv or eammi2_data_final.csv)
df <- read.csv(file="Data/fakedata.csv", header=T)

Viewing Data

# these are commands useful for viewing a dataframe
# you can also click the object in the environment tab to view it in a new window
names(df)
##  [1] "id"         "variable1"  "variable2"  "variable3"  "variable4" 
##  [6] "variable5"  "variable6"  "variable7"  "variable8"  "variable9" 
## [11] "variable10" "variable11" "variable12"
head(df)
##     id variable1 variable2 variable3 variable4 variable5 variable6 variable7
## 1 id_1   level b   level d   level a   level b  3.449743 3.4805072  1.039619
## 2 id_2   level b   level c   level a   level c  2.707133 2.6169258  2.280291
## 3 id_3   level b   level b   level b   level b  3.135185 3.2119597  1.854115
## 4 id_4   level c   level d   level b      <NA>  2.901656 0.9048516  1.358520
## 5 id_5   level b   level d   level a      <NA>  2.277099 2.9104901  1.343883
## 6 id_6   level b   level d   level b   level c  2.404385 2.5616353  1.670581
##   variable8 variable9 variable10 variable11 variable12
## 1  4.460219  1.153984   5.056802   1.546471   6.866127
## 2  4.754638  1.474888   4.214870   1.146760   7.208780
## 3  3.518784  1.218142   4.116025   2.746469   5.253641
## 4  4.617069  1.066567   4.670609   1.206538   5.847952
## 5  3.221986  1.374207   4.861695   1.575204   6.936754
## 6  4.892009  1.143127   4.459852   2.765479   6.697210
str(df)
## 'data.frame':    1000 obs. of  13 variables:
##  $ id        : chr  "id_1" "id_2" "id_3" "id_4" ...
##  $ variable1 : chr  "level b" "level b" "level b" "level c" ...
##  $ variable2 : chr  "level d" "level c" "level b" "level d" ...
##  $ variable3 : chr  "level a" "level a" "level b" "level b" ...
##  $ variable4 : chr  "level b" "level c" "level b" NA ...
##  $ variable5 : num  3.45 2.71 3.14 2.9 2.28 ...
##  $ variable6 : num  3.481 2.617 3.212 0.905 2.91 ...
##  $ variable7 : num  1.04 2.28 1.85 1.36 1.34 ...
##  $ variable8 : num  4.46 4.75 3.52 4.62 3.22 ...
##  $ variable9 : num  1.15 1.47 1.22 1.07 1.37 ...
##  $ variable10: num  5.06 4.21 4.12 4.67 4.86 ...
##  $ variable11: num  1.55 1.15 2.75 1.21 1.58 ...
##  $ variable12: num  6.87 7.21 5.25 5.85 6.94 ...

Subsetting Data

# for the HW: use the codebook you created in the codebook activity to get the names of your variables (first column)
# enter this list of names in the select=c() argument to subset those columns from the dataframe
# variables for the lab: id, variable2, variable3, variable5, variable8, variable10, variable11
d <- subset(df, select=c(id, variable2, variable3, variable5, variable8, variable10, variable11))
#replace insert variables here, with your own variables 
#id will be something else in HW for EAMMI2, I believe it will be R

Recoding Variables

# categorical variables need to be recoded as factors
# the content of the variable will stay the same, but R will treat the variable differently at times
d$variable2 <- as.factor(d$variable2)
d$variable3 <- as.factor(d$variable3)

str(d)
## 'data.frame':    1000 obs. of  7 variables:
##  $ id        : chr  "id_1" "id_2" "id_3" "id_4" ...
##  $ variable2 : Factor w/ 6 levels "level a","level b",..: 4 3 2 4 4 4 3 2 4 3 ...
##  $ variable3 : Factor w/ 2 levels "level a","level b": 1 1 2 2 1 2 2 2 2 2 ...
##  $ variable5 : num  3.45 2.71 3.14 2.9 2.28 ...
##  $ variable8 : num  4.46 4.75 3.52 4.62 3.22 ...
##  $ variable10: num  5.06 4.21 4.12 4.67 4.86 ...
##  $ variable11: num  1.55 1.15 2.75 1.21 1.58 ...

Missing Data

Talk about missing data here. We looked at the missinf data in our data set and found that about 12% of the particpants in our sample skipped at least one item, we dropped these particpants from our analysis, which is not advisable and runs the risk of dropping vulnerable groups or skewing results, however we will proceed for the sake of this class using the reduced data set.

# use the gg_miss_upset() command for a visualization of your missing data
gg_miss_upset(d[-1], nsets = 6)

# use the na.omit() command to create a new dataframe in which any participants with missing data are dropped from the dataframe
d2 <- na.omit(d)
889/1000
## [1] 0.889
#you can find out the percentage of how many people had missed data, by diving number from missed with total number, then the total is 12 percent dropped.

Exporting Data

# last step is to export the data after you've dropped NAs
# for the HW, the file you're exporting here is what you'll use for all future HW assignments (labs will use the files I provide you)
# make sure you give it a name that is memorable!
# and make sure you save it to your Data folder!
write.csv(d2, file="Data/fakedata_final.csv", row.names = F)

# since we've created a cleaned dataframe in d2, we'll use that for the rest of the lab/HW

Basic Statistics

Univariate Plots: Histograms & Tables

table(d2$variable2)
## 
## level a level b level c level d level e level f 
##      29     233     351     228      45       3
table(d2$variable3)
## 
## level a level b 
##     200     689
# replace categorical variables with your own categorical variables
hist(d2$variable5)

hist(d2$variable8)

hist(d2$variable10)

hist(d2$variable11)

# replace continious with your own continuous variables

Univariate Normality

Talk about skew and kurtosis here. We anaslyzed the skew and kurosis of our continious variables and all were within the accepted range (-2/+2). We analyzed the skew and kurtosis and most/some wewre within the accepted range (-2/+2) however some variables (varble ex) were outside of the accepted rnge. For this anaylsis, we will use them anuyway, but outside of this class this is bad practice.

describe(d2)
##            vars   n   mean     sd median trimmed    mad  min    max  range
## id*           1 889 445.00 256.78 445.00  445.00 329.14 1.00 889.00 888.00
## variable2*    2 889   3.04   0.94   3.00    3.02   1.48 1.00   6.00   5.00
## variable3*    3 889   1.78   0.42   2.00    1.84   0.00 1.00   2.00   1.00
## variable5     4 889   2.53   0.49   2.51    2.52   0.50 1.17   4.15   2.99
## variable8     5 889   3.89   0.65   3.96    3.93   0.68 1.36   5.00   3.63
## variable10    6 889   4.87   0.97   4.90    4.90   0.95 1.04   6.98   5.94
## variable11    7 889   1.82   0.61   1.72    1.76   0.64 1.00   3.99   2.99
##             skew kurtosis   se
## id*         0.00    -1.20 8.61
## variable2*  0.15    -0.30 0.03
## variable3* -1.32    -0.27 0.01
## variable5   0.14    -0.07 0.02
## variable8  -0.57     0.03 0.02
## variable10 -0.35     0.27 0.03
## variable11  0.83     0.15 0.02
# look and see if skew and kurtosis and if they are between the cutoffs of -2 and 2. Only look at your continuous variables

Bivariate Plots

Crosstabs

cross_cases(d2, variable2, variable3)
 variable3 
 level a   level b 
 variable2 
   level a  5 24
   level b  56 177
   level c  79 272
   level d  46 182
   level e  13 32
   level f  1 2
   #Total cases  200 689 
# replace categorical with your categorical variables.

Scatterplots

plot(d2$variable5, d2$variable8,
     main="Scatterplot of variable5 and variable8",
     xlab = "variable5",
     ylab = "variable8")

# x variable is the first variable listed, y is the second variable listed. 
# you only need to choose two of your variables, you don't need all 4.
plot(d2$variable10, d2$variable11,
     main="Scatterplot of variable10 and variable11",
     xlab = "variable10",
     ylab = "variable11")

Boxplots

boxplot(data=d2, variable5~variable2,
        main="Boxplot of variable2 and variable5",
        xlab = "x (variable2)",
        ylab = "Y (variable5)")

boxplot(data=d2, variable11~variable3,
        main="Boxplot of variable3 and variable11",
        xlab = "x (variable3)",
        ylab = "Y (variable11)")

# x is categorical, y is continuous. Continuous~Categorical