1 Stress Data

We’re looking at information from 477 people, each represented as one row in the dataset. Every column is a different piece of information we collected about them. Some columns capture demographics (like age, gender, and ethnicity), while others are responses to questionnaires.
There are two main groups of questionnaire items here:
- Recovery questions (recovery1 through recovery16): these ask about how well someone bounces back from stress or fatigue.
- PANAS questions (PANAS1 through PANAS20): this stands for Positive and Negative Affect Schedule, a common survey that measures how positive or negative a person’s emotions are.
Think of the dataset as a giant spreadsheet where each row is a person and each column is a question they answered.

# Read from the same folder as this .Rmd
stress <- read.csv("Stress.csv", na.strings = c("", "NA"))

# Sanity checks
dim(stress)        # rows x columns

## [1] 477  40

names(stress)      # column names

##  [1] "age"        "gender"     "ethnicity"  "recovery1"  "recovery2" 
##  [6] "recovery3"  "recovery4"  "recovery5"  "recovery6"  "recovery7" 
## [11] "recovery8"  "recovery9"  "recovery10" "recovery11" "recovery12"
## [16] "recovery13" "recovery14" "recovery15" "recovery16" "AttnChk"   
## [21] "PANAS1"     "PANAS2"     "PANAS3"     "PANAS4"     "PANAS5"    
## [26] "PANAS6"     "PANAS7"     "PANAS8"     "PANAS9"     "PANAS10"   
## [31] "PANAS11"    "PANAS12"    "PANAS13"    "PANAS14"    "PANAS15"   
## [36] "PANAS16"    "PANAS17"    "PANAS18"    "PANAS19"    "PANAS20"

head(stress, 5)    # first 5 rows

##   age gender ethnicity recovery1 recovery2 recovery3 recovery4 recovery5
## 1  34      2         1        NA        NA        NA        NA        NA
## 2  22      1         1        NA        NA        NA        NA        NA
## 3  29      2         1        NA        NA        NA        NA        NA
## 4  34      2         1        NA        NA        NA        NA        NA
## 5  34      1         1        NA        NA        NA        NA        NA
##   recovery6 recovery7 recovery8 recovery9 recovery10 recovery11 recovery12
## 1        NA        NA        NA        NA         NA         NA         NA
## 2        NA        NA        NA        NA         NA         NA         NA
## 3        NA        NA        NA        NA         NA         NA         NA
## 4        NA        NA        NA        NA         NA         NA         NA
## 5        NA        NA        NA        NA         NA         NA         NA
##   recovery13 recovery14 recovery15 recovery16 AttnChk PANAS1 PANAS2 PANAS3
## 1         NA         NA         NA         NA      NA     NA     NA     NA
## 2         NA         NA         NA         NA      NA     NA     NA     NA
## 3         NA         NA         NA         NA      NA     NA     NA     NA
## 4         NA         NA         NA         NA      NA     NA     NA     NA
## 5         NA         NA         NA         NA      NA     NA     NA     NA
##   PANAS4 PANAS5 PANAS6 PANAS7 PANAS8 PANAS9 PANAS10 PANAS11 PANAS12 PANAS13
## 1     NA     NA     NA     NA     NA     NA      NA      NA      NA      NA
## 2     NA     NA     NA     NA     NA     NA      NA      NA      NA      NA
## 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
##   PANAS14 PANAS15 PANAS16 PANAS17 PANAS18 PANAS19 PANAS20
## 1      NA      NA      NA      NA      NA      NA      NA
## 2      NA      NA      NA      NA      NA      NA      NA
## 3      NA      NA      NA      NA      NA      NA      NA
## 4      NA      NA      NA      NA      NA      NA      NA
## 5      NA      NA      NA      NA      NA      NA      NA

1.1 Adding an ID column

Because we have hundreds of rows, it helps to assign each person a unique ID number starting at 1. This isn’t information about the person themselves, it’s just like putting a name tag on each row so we can keep track of them.

stress$ID <- seq_len(nrow(stress))
head(stress[, c("ID", names(stress)[1:5])], 5)

##   ID age gender ethnicity recovery1 recovery2
## 1  1  34      2         1        NA        NA
## 2  2  22      1         1        NA        NA
## 3  3  29      2         1        NA        NA
## 4  4  34      2         1        NA        NA
## 5  5  34      1         1        NA        NA

1.2 Gender and ethinicty recoding

In the original file, gender and ethnicity are stored as numbers (like 1, 2, 3…). While its easier to track data with numbers it can be confusing for people. We recode these into categories, so instead of treating them like mathematical values, we treat them as labels. That way when we summarize or group the data, it makes sense to us.

num_cols <- grep("^(recovery\\d+|PANAS\\d+)$", names(stress), value = TRUE)
for (cn in num_cols) {
  # Coerce only if not already numeric
  if (!is.numeric(stress[[cn]])) {
    suppressWarnings(stress[[cn]] <- as.numeric(stress[[cn]]))
  }
}
str(stress[num_cols])

## 'data.frame':    477 obs. of  36 variables:
##  $ recovery1 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery2 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery3 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery4 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery5 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery6 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery7 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery8 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery9 : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery10: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery11: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery12: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery13: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery14: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery15: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ recovery16: int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS1    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS2    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS3    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS4    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS5    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS6    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS7    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS8    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS9    : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS10   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS11   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS12   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS13   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS14   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS15   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS16   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS17   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS18   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS19   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ PANAS20   : int  NA NA NA NA NA NA NA NA NA NA ...

# Safe, codebook-agnostic factorization
if ("gender" %in% names(stress)) {
  stress$gender <- factor(stress$gender)
}
if ("ethnicity" %in% names(stress)) {
  stress$ethnicity <- factor(stress$ethnicity)
}

# Example labeled version (use only if you know your coding):
# stress$gender <- factor(stress$gender, levels = c(1, 2), labels = c("Female", "Male"))
# stress$ethnicity <- factor(stress$ethnicity,
#                            levels = c(1, 2, 3, 4, 5), 
#                            labels = c("Group1","Group2","Group3","Group4","Group5"))

summary(stress[, intersect(c("gender","ethnicity"), names(stress))])

##   gender    ethnicity 
##  1   :275   1   :377  
##  2   :197   2   : 44  
##  3   :  2   3   :  5  
##  NA's:  3   4   : 28  
##             5   :  1  
##             6   : 19  
##             NA's:  3

1.3 Audit of missing values

When looking at real-world data there is often times missing data points. To clean the data we have to decide what to do with the missing data.
- First, we need to check how many points are missing.
- Then, decide how to handle those points, For numbers, we can fill in missing spots with the median, which is less sensitive to extreme highs or lows. For categories like gender, we can add a label called “Missing” so it’s clear they didn’t answer.

This way, we’re not throwing away data, but we’re also not pretending the blanks don’t exist.

# NA counts by column
na_by_col <- sapply(stress, function(x) sum(is.na(x)))
na_by_col[na_by_col > 0]

##        age     gender  ethnicity  recovery1  recovery2  recovery3  recovery4 
##          3          3          3         25         25         25         25 
##  recovery5  recovery6  recovery7  recovery8  recovery9 recovery10 recovery11 
##         25         25         25         25         25         25         25 
## recovery12 recovery13 recovery14 recovery15 recovery16    AttnChk     PANAS1 
##         25         25         25         25         25         25         51 
##     PANAS2     PANAS3     PANAS4     PANAS5     PANAS6     PANAS7     PANAS8 
##         52         51         52         51         51         53         51 
##     PANAS9    PANAS10    PANAS11    PANAS12    PANAS13    PANAS14    PANAS15 
##         51         52         51         52         53         51         52 
##    PANAS16    PANAS17    PANAS18    PANAS19    PANAS20 
##         51         52         52         52         52

# Total number of NA cells
sum(is.na(stress))

## [1] 1467

# How many rows contain any missing values?
rows_with_any_na <- sum(!complete.cases(stress))
rows_with_any_na

## [1] 66

1.4 Using Data Wrangling

This is where we put the cleaned data into action that allows funtions like
- selecting specific columns
- arranging them in order
- Grouping

# Load dplyr for the assignment's "include some dplyr" requirement
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

# Identify numeric and factor columns
num_cols    <- names(Filter(is.numeric, stress))
factor_cols <- names(Filter(is.factor,  stress))

# Create a cleaned copy with explicit, reproducible rules
stress_clean <-
  stress %>%
  mutate(
    across(all_of(num_cols),    ~ ifelse(is.na(.x), median(.x, na.rm = TRUE), .x)),
    across(all_of(factor_cols), ~ {
      x <- as.character(.x)
      x[is.na(x)] <- "Missing"
      factor(x)
    })
  )

# Verify that missingness is addressed
na_by_col_after <- sapply(stress_clean, function(x) sum(is.na(x)))
na_by_col_after[na_by_col_after > 0]

## named integer(0)

# Compact demographic snapshot
stress_clean %>%
  select(ID, age, gender, ethnicity) %>%
  arrange(ID) %>%
  head(10)

##    ID age gender ethnicity
## 1   1  34      2         1
## 2   2  22      1         1
## 3   3  29      2         1
## 4   4  34      2         1
## 5   5  34      1         1
## 6   6  29      1         2
## 7   7  51      2         1
## 8   8  23      1         1
## 9   9  31      2         2
## 10 10  32      1         2

# Example grouped summary (swap metrics as needed)
stress_clean %>%
  group_by(gender) %>%
  summarise(
    n = n(),
    mean_age = mean(age, na.rm = TRUE),
    .groups = "drop"
  )

## # A tibble: 4 × 3
##   gender      n mean_age
##   <fct>   <int>    <dbl>
## 1 1         275     32.1
## 2 2         197     36.0
## 3 3           2     22.5
## 4 Missing     3     31

1.5 Summary Table

After all these steps, we end up with a polished version of the dataset called stress_v1. It’s the same people and questions as before, but now:
- Each person has an ID.
- Gender and ethnicity are proper categories.
- Missing values are treated consistently.
It’s easy to summarize, plot, and analyze.

stress_v1 <- stress_clean

# Brief summary
dim(stress_v1)

## [1] 477  41

summary(stress_v1[, intersect(c("ID","age","gender","ethnicity"), names(stress_v1))])

##        ID           age            gender      ethnicity  
##  Min.   :  1   Min.   :19.00   1      :275   1      :377  
##  1st Qu.:120   1st Qu.:27.00   2      :197   2      : 44  
##  Median :239   Median :31.00   3      :  2   3      :  5  
##  Mean   :239   Mean   :33.66   Missing:  3   4      : 28  
##  3rd Qu.:358   3rd Qu.:38.00                 5      :  1  
##  Max.   :477   Max.   :70.00                 6      : 19  
##                                              Missing:  3

head(stress_v1, 8)

##   age gender ethnicity recovery1 recovery2 recovery3 recovery4 recovery5
## 1  34      2         1         3         2         4         4         4
## 2  22      1         1         3         2         4         4         4
## 3  29      2         1         3         2         4         4         4
## 4  34      2         1         3         2         4         4         4
## 5  34      1         1         3         2         4         4         4
## 6  29      1         2         3         2         4         4         4
## 7  51      2         1         3         2         4         4         4
## 8  23      1         1         3         2         4         4         4
##   recovery6 recovery7 recovery8 recovery9 recovery10 recovery11 recovery12
## 1         4         4         4         4          4          4          4
## 2         4         4         4         4          4          4          4
## 3         4         4         4         4          4          4          4
## 4         4         4         4         4          4          4          4
## 5         4         4         4         4          4          4          4
## 6         4         4         4         4          4          4          4
## 7         4         4         4         4          4          4          4
## 8         4         4         4         4          4          4          4
##   recovery13 recovery14 recovery15 recovery16 AttnChk PANAS1 PANAS2 PANAS3
## 1          4          4          4          4       3      4      4      2
## 2          4          4          4          4       3      4      4      2
## 3          4          4          4          4       3      4      4      2
## 4          4          4          4          4       3      4      4      2
## 5          4          4          4          4       3      4      4      2
## 6          4          4          4          4       3      4      4      2
## 7          4          4          4          4       3      4      4      2
## 8          4          4          4          4       3      4      4      2
##   PANAS4 PANAS5 PANAS6 PANAS7 PANAS8 PANAS9 PANAS10 PANAS11 PANAS12 PANAS13
## 1      1    3.5      2      1      3      2       4       4       2       1
## 2      1    3.5      2      1      3      2       4       4       2       1
## 3      1    3.5      2      1      3      2       4       4       2       1
## 4      1    3.5      2      1      3      2       4       4       2       1
## 5      1    3.5      2      1      3      2       4       4       2       1
## 6      1    3.5      2      1      3      2       4       4       2       1
## 7      1    3.5      2      1      3      2       4       4       2       1
## 8      1    3.5      2      1      3      2       4       4       2       1
##   PANAS14 PANAS15 PANAS16 PANAS17 PANAS18 PANAS19 PANAS20 ID
## 1       2       3       4       2       2       3       4  1
## 2       2       3       4       2       2       3       4  2
## 3       2       3       4       2       2       3       4  3
## 4       2       3       4       2       2       3       4  4
## 5       2       3       4       2       2       3       4  5
## 6       2       3       4       2       2       3       4  6
## 7       2       3       4       2       2       3       4  7
## 8       2       3       4       2       2       3       4  8

stopifnot(exists("stress_v1"))
str(stress_v1[ , intersect(c("ID","age","gender","ethnicity"), names(stress_v1))])

## 'data.frame':    477 obs. of  4 variables:
##  $ ID       : int  1 2 3 4 5 6 7 8 9 10 ...
##  $ age      : num  34 22 29 34 34 29 51 23 31 32 ...
##  $ gender   : Factor w/ 4 levels "1","2","3","Missing": 2 1 2 2 1 1 2 1 2 1 ...
##  $ ethnicity: Factor w/ 7 levels "1","2","3","4",..: 1 1 1 1 1 2 1 1 2 2 ...

keep_cols <- unique(c("ID", "age", "gender", "ethnicity",
                      "recovery_mean", "PANAS_pos_mean", "PANAS_neg_mean"))
keep_cols <- keep_cols[keep_cols %in% names(stress)]

summary(stress[keep_cols])

##        ID           age         gender    ethnicity 
##  Min.   :  1   Min.   :19.00   1   :275   1   :377  
##  1st Qu.:120   1st Qu.:27.00   2   :197   2   : 44  
##  Median :239   Median :31.00   3   :  2   3   :  5  
##  Mean   :239   Mean   :33.68   NA's:  3   4   : 28  
##  3rd Qu.:358   3rd Qu.:38.00              5   :  1  
##  Max.   :477   Max.   :70.00              6   : 19  
##                NA's   :3                  NA's:  3

# If you want a small sample display:
head(stress[keep_cols], 10)

##    ID age gender ethnicity
## 1   1  34      2         1
## 2   2  22      1         1
## 3   3  29      2         1
## 4   4  34      2         1
## 5   5  34      1         1
## 6   6  29      1         2
## 7   7  51      2         1
## 8   8  23      1         1
## 9   9  31      2         2
## 10 10  32      1         2

1.6 Exectuive Summary Table

# ---- table_polished -------------------------------------------------------
knitr::kable({
  cols <- intersect(c("ID","age","gender","ethnicity","recovery_mean","PANAS_pos_mean","PANAS_neg_mean"),
                    names(stress_v1))
  df <- stress_v1[cols]
  # round numeric columns for readability
  num <- sapply(df, is.numeric)
  df[num] <- lapply(df[num], function(x) round(x, 2))
  # order by ID if present
  if ("ID" %in% names(df)) df <- df[order(df$ID), ]
  head(df, 25)
}, caption = "Final dataset (clean preview: first 25 rows, rounded)")

Final dataset (clean preview: first 25 rows, rounded)
ID	age	gender	ethnicity
1	34	2	1
2	22	1	1
3	29	2	1
4	34	2	1
5	34	1	1
6	29	1	2
7	51	2	1
8	23	1	1
9	31	2	2
10	32	1	2
11	19	2	2
12	25	1	1
13	50	2	1
14	49	2	1
15	48	1	1
16	31	2	2
17	30	1	1
18	42	2	1
19	29	1	1
20	32	2	6
21	23	1	1
22	49	2	4
23	29	1	1
24	32	1	2
25	26	2	1

Stress Dataset Report

Nat Martinov

1 Stress Data

1.1 Adding an ID column

1.2 Gender and ethinicty recoding

1.3 Audit of missing values

1.4 Using Data Wrangling

1.5 Summary Table

1.6 Exectuive Summary Table

ID	age	gender	ethnicity
1	34	2	1
2	22	1	1
3	29	2	1
4	34	2	1
5	34	1	1
6	29	1	2
7	51	2	1
8	23	1	1
9	31	2	2
10	32	1	2
11	19	2	2
12	25	1	1
13	50	2	1
14	49	2	1
15	48	1	1
16	31	2	2
17	30	1	1
18	42	2	1
19	29	1	1
20	32	2	6
21	23	1	1
22	49	2	4
23	29	1	1
24	32	1	2
25	26	2	1

ID	age	gender	ethnicity
1	34	2	1
2	22	1	1
3	29	2	1
4	34	2	1
5	34	1	1
6	29	1	2
7	51	2	1
8	23	1	1
9	31	2	2
10	32	1	2
11	19	2	2
12	25	1	1
13	50	2	1
14	49	2	1
15	48	1	1
16	31	2	2
17	30	1	1
18	42	2	1
19	29	1	1
20	32	2	6
21	23	1	1
22	49	2	4
23	29	1	1
24	32	1	2
25	26	2	1

ID	age	gender	ethnicity
1	34	2	1
2	22	1	1
3	29	2	1
4	34	2	1
5	34	1	1
6	29	1	2
7	51	2	1
8	23	1	1
9	31	2	2
10	32	1	2
11	19	2	2
12	25	1	1
13	50	2	1
14	49	2	1
15	48	1	1
16	31	2	2
17	30	1	1
18	42	2	1
19	29	1	1
20	32	2	6
21	23	1	1
22	49	2	4
23	29	1	1
24	32	1	2
25	26	2	1