Reproducible Research
Set the working directory and read in the data
wd <- getwd()
current_proj_dir <- "/Reproducible Research/RepData_PeerAssessment1"
wd1 <- paste0(wd, "/activity/activity.csv")
activity <- read.csv(wd1)Check some stats about the data
head(activity)## steps date interval
## 1 NA 2012-10-01 0
## 2 NA 2012-10-01 5
## 3 NA 2012-10-01 10
## 4 NA 2012-10-01 15
## 5 NA 2012-10-01 20
## 6 NA 2012-10-01 25str(activity)## 'data.frame': 17568 obs. of 3 variables:
## $ steps : int NA NA NA NA NA NA NA NA NA NA ...
## $ date : Factor w/ 61 levels "2012-10-01","2012-10-02",..: 1 1 1 1 1 1 1 1 1 1 ...
## $ interval: int 0 5 10 15 20 25 30 35 40 45 ...names(activity)## [1] "steps" "date" "interval"Convert dates to date objects
library(lubridate)##
## Attaching package: 'lubridate'## The following object is masked from 'package:base':
##
## dateactivity$date <- ymd(activity$date)
str(activity)## 'data.frame': 17568 obs. of 3 variables:
## $ steps : int NA NA NA NA NA NA NA NA NA NA ...
## $ date : Date, format: "2012-10-01" "2012-10-01" ...
## $ interval: int 0 5 10 15 20 25 30 35 40 45 ...head(activity)## steps date interval
## 1 NA 2012-10-01 0
## 2 NA 2012-10-01 5
## 3 NA 2012-10-01 10
## 4 NA 2012-10-01 15
## 5 NA 2012-10-01 20
## 6 NA 2012-10-01 25Calculate the total number of steps taken per day
ttl_walks_by_day <- aggregate(activity["steps"], by=list(Date=activity$date), FUN=sum)
ttl_walks_by_day## Date steps
## 1 2012-10-01 NA
## 2 2012-10-02 126
## 3 2012-10-03 11352
## 4 2012-10-04 12116
## 5 2012-10-05 13294
## 6 2012-10-06 15420
## 7 2012-10-07 11015
## 8 2012-10-08 NA
## 9 2012-10-09 12811
## 10 2012-10-10 9900
## 11 2012-10-11 10304
## 12 2012-10-12 17382
## 13 2012-10-13 12426
## 14 2012-10-14 15098
## 15 2012-10-15 10139
## 16 2012-10-16 15084
## 17 2012-10-17 13452
## 18 2012-10-18 10056
## 19 2012-10-19 11829
## 20 2012-10-20 10395
## 21 2012-10-21 8821
## 22 2012-10-22 13460
## 23 2012-10-23 8918
## 24 2012-10-24 8355
## 25 2012-10-25 2492
## 26 2012-10-26 6778
## 27 2012-10-27 10119
## 28 2012-10-28 11458
## 29 2012-10-29 5018
## 30 2012-10-30 9819
## 31 2012-10-31 15414
## 32 2012-11-01 NA
## 33 2012-11-02 10600
## 34 2012-11-03 10571
## 35 2012-11-04 NA
## 36 2012-11-05 10439
## 37 2012-11-06 8334
## 38 2012-11-07 12883
## 39 2012-11-08 3219
## 40 2012-11-09 NA
## 41 2012-11-10 NA
## 42 2012-11-11 12608
## 43 2012-11-12 10765
## 44 2012-11-13 7336
## 45 2012-11-14 NA
## 46 2012-11-15 41
## 47 2012-11-16 5441
## 48 2012-11-17 14339
## 49 2012-11-18 15110
## 50 2012-11-19 8841
## 51 2012-11-20 4472
## 52 2012-11-21 12787
## 53 2012-11-22 20427
## 54 2012-11-23 21194
## 55 2012-11-24 14478
## 56 2012-11-25 11834
## 57 2012-11-26 11162
## 58 2012-11-27 13646
## 59 2012-11-28 10183
## 60 2012-11-29 7047
## 61 2012-11-30 NAMake a histogram of the total number of steps taken each day
hist(ttl_walks_by_day$steps
, breaks = 20
, col = "pink"
, main = "Total Steps By Day"
,xlab = "Steps"
,labels = TRUE)
3a Calculate and report the mean of the total number of steps taken per day
daily_mean <- aggregate(activity["steps"], by=list(Date=activity$date), FUN=mean)
daily_mean## Date steps
## 1 2012-10-01 NA
## 2 2012-10-02 0.4375000
## 3 2012-10-03 39.4166667
## 4 2012-10-04 42.0694444
## 5 2012-10-05 46.1597222
## 6 2012-10-06 53.5416667
## 7 2012-10-07 38.2465278
## 8 2012-10-08 NA
## 9 2012-10-09 44.4826389
## 10 2012-10-10 34.3750000
## 11 2012-10-11 35.7777778
## 12 2012-10-12 60.3541667
## 13 2012-10-13 43.1458333
## 14 2012-10-14 52.4236111
## 15 2012-10-15 35.2048611
## 16 2012-10-16 52.3750000
## 17 2012-10-17 46.7083333
## 18 2012-10-18 34.9166667
## 19 2012-10-19 41.0729167
## 20 2012-10-20 36.0937500
## 21 2012-10-21 30.6284722
## 22 2012-10-22 46.7361111
## 23 2012-10-23 30.9652778
## 24 2012-10-24 29.0104167
## 25 2012-10-25 8.6527778
## 26 2012-10-26 23.5347222
## 27 2012-10-27 35.1354167
## 28 2012-10-28 39.7847222
## 29 2012-10-29 17.4236111
## 30 2012-10-30 34.0937500
## 31 2012-10-31 53.5208333
## 32 2012-11-01 NA
## 33 2012-11-02 36.8055556
## 34 2012-11-03 36.7048611
## 35 2012-11-04 NA
## 36 2012-11-05 36.2465278
## 37 2012-11-06 28.9375000
## 38 2012-11-07 44.7326389
## 39 2012-11-08 11.1770833
## 40 2012-11-09 NA
## 41 2012-11-10 NA
## 42 2012-11-11 43.7777778
## 43 2012-11-12 37.3784722
## 44 2012-11-13 25.4722222
## 45 2012-11-14 NA
## 46 2012-11-15 0.1423611
## 47 2012-11-16 18.8923611
## 48 2012-11-17 49.7881944
## 49 2012-11-18 52.4652778
## 50 2012-11-19 30.6979167
## 51 2012-11-20 15.5277778
## 52 2012-11-21 44.3993056
## 53 2012-11-22 70.9270833
## 54 2012-11-23 73.5902778
## 55 2012-11-24 50.2708333
## 56 2012-11-25 41.0902778
## 57 2012-11-26 38.7569444
## 58 2012-11-27 47.3819444
## 59 2012-11-28 35.3576389
## 60 2012-11-29 24.4687500
## 61 2012-11-30 NA3b Calculate and report the median of the total number of steps taken per day
daily_median <- aggregate(activity["steps"], by=list(Date=activity$date), FUN=median)
daily_median## Date steps
## 1 2012-10-01 NA
## 2 2012-10-02 0
## 3 2012-10-03 0
## 4 2012-10-04 0
## 5 2012-10-05 0
## 6 2012-10-06 0
## 7 2012-10-07 0
## 8 2012-10-08 NA
## 9 2012-10-09 0
## 10 2012-10-10 0
## 11 2012-10-11 0
## 12 2012-10-12 0
## 13 2012-10-13 0
## 14 2012-10-14 0
## 15 2012-10-15 0
## 16 2012-10-16 0
## 17 2012-10-17 0
## 18 2012-10-18 0
## 19 2012-10-19 0
## 20 2012-10-20 0
## 21 2012-10-21 0
## 22 2012-10-22 0
## 23 2012-10-23 0
## 24 2012-10-24 0
## 25 2012-10-25 0
## 26 2012-10-26 0
## 27 2012-10-27 0
## 28 2012-10-28 0
## 29 2012-10-29 0
## 30 2012-10-30 0
## 31 2012-10-31 0
## 32 2012-11-01 NA
## 33 2012-11-02 0
## 34 2012-11-03 0
## 35 2012-11-04 NA
## 36 2012-11-05 0
## 37 2012-11-06 0
## 38 2012-11-07 0
## 39 2012-11-08 0
## 40 2012-11-09 NA
## 41 2012-11-10 NA
## 42 2012-11-11 0
## 43 2012-11-12 0
## 44 2012-11-13 0
## 45 2012-11-14 NA
## 46 2012-11-15 0
## 47 2012-11-16 0
## 48 2012-11-17 0
## 49 2012-11-18 0
## 50 2012-11-19 0
## 51 2012-11-20 0
## 52 2012-11-21 0
## 53 2012-11-22 0
## 54 2012-11-23 0
## 55 2012-11-24 0
## 56 2012-11-25 0
## 57 2012-11-26 0
## 58 2012-11-27 0
## 59 2012-11-28 0
## 60 2012-11-29 0
## 61 2012-11-30 NA- Make a time series plot (i.e. type = “l”) of the 5-minute interval (x-axis) and the average number of steps taken, averaged across all days (y-axis)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:lubridate':
##
## intersect, setdiff, union## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionactivity_new <- activity %>%
group_by(interval) %>%
summarise(AvgSteps = mean(steps, na.rm = TRUE))
plot(activity_new,
type ="l",
xlab= "5-minute interval",
ylab="averaged steps across all days",
main ="Average Daily Steps Over Each 5 Minute Interval",
col = "orange")
library(ggplot2)
ggplot(activity_new, aes(interval, AvgSteps,col = AvgSteps)) + geom_line() +
xlab("5-minute interval") + ylab("averaged steps across all days")
check if there are all NAs
sum(!is.na(activity_new$AvgSteps))## [1] 288- which interval on average contains the maximum number of steps
activity_new$interval[which.max(activity_new$AvgSteps)]## [1] 835Imputing missing values
1 Calculate and report the total number of missing values in the dataset (i.e. the total number of rows with NAs)
impute MISSING VALUES with the interval average steps
activity_imputed <- activity %>%
group_by(interval) %>%
mutate(AvgSteps = ifelse( is.na(steps), mean(steps, na.rm = TRUE), steps))
#check if there are any NAs in the average
sum(is.na(activity_imputed$AvgSteps))## [1] 04a. Calculate the total number of steps taken per day Make a histogram of the total number of steps taken each day
activity_imputed_ttl_daily_steps <- aggregate(activity_imputed["AvgSteps"], by=list(Date=activity_imputed$date), FUN=sum)
activity_imputed_ttl_daily_steps## Date AvgSteps
## 1 2012-10-01 10766.19
## 2 2012-10-02 126.00
## 3 2012-10-03 11352.00
## 4 2012-10-04 12116.00
## 5 2012-10-05 13294.00
## 6 2012-10-06 15420.00
## 7 2012-10-07 11015.00
## 8 2012-10-08 10766.19
## 9 2012-10-09 12811.00
## 10 2012-10-10 9900.00
## 11 2012-10-11 10304.00
## 12 2012-10-12 17382.00
## 13 2012-10-13 12426.00
## 14 2012-10-14 15098.00
## 15 2012-10-15 10139.00
## 16 2012-10-16 15084.00
## 17 2012-10-17 13452.00
## 18 2012-10-18 10056.00
## 19 2012-10-19 11829.00
## 20 2012-10-20 10395.00
## 21 2012-10-21 8821.00
## 22 2012-10-22 13460.00
## 23 2012-10-23 8918.00
## 24 2012-10-24 8355.00
## 25 2012-10-25 2492.00
## 26 2012-10-26 6778.00
## 27 2012-10-27 10119.00
## 28 2012-10-28 11458.00
## 29 2012-10-29 5018.00
## 30 2012-10-30 9819.00
## 31 2012-10-31 15414.00
## 32 2012-11-01 10766.19
## 33 2012-11-02 10600.00
## 34 2012-11-03 10571.00
## 35 2012-11-04 10766.19
## 36 2012-11-05 10439.00
## 37 2012-11-06 8334.00
## 38 2012-11-07 12883.00
## 39 2012-11-08 3219.00
## 40 2012-11-09 10766.19
## 41 2012-11-10 10766.19
## 42 2012-11-11 12608.00
## 43 2012-11-12 10765.00
## 44 2012-11-13 7336.00
## 45 2012-11-14 10766.19
## 46 2012-11-15 41.00
## 47 2012-11-16 5441.00
## 48 2012-11-17 14339.00
## 49 2012-11-18 15110.00
## 50 2012-11-19 8841.00
## 51 2012-11-20 4472.00
## 52 2012-11-21 12787.00
## 53 2012-11-22 20427.00
## 54 2012-11-23 21194.00
## 55 2012-11-24 14478.00
## 56 2012-11-25 11834.00
## 57 2012-11-26 11162.00
## 58 2012-11-27 13646.00
## 59 2012-11-28 10183.00
## 60 2012-11-29 7047.00
## 61 2012-11-30 10766.19#4bhistory gram
hist(activity_imputed_ttl_daily_steps$AvgSteps,
breaks = 20,
col="green",
main = "Total Steps By Day After inputing Missen values",
xlab = "Steps",
labels = T)
4c and Calculate and report the mean and median total number of steps taken per day. ###4ci mean
daily_mean_after_imputation <- aggregate(activity_imputed["AvgSteps"], by=list(Date=activity_imputed$date), FUN=mean)
daily_mean_after_imputation## Date AvgSteps
## 1 2012-10-01 37.3825996
## 2 2012-10-02 0.4375000
## 3 2012-10-03 39.4166667
## 4 2012-10-04 42.0694444
## 5 2012-10-05 46.1597222
## 6 2012-10-06 53.5416667
## 7 2012-10-07 38.2465278
## 8 2012-10-08 37.3825996
## 9 2012-10-09 44.4826389
## 10 2012-10-10 34.3750000
## 11 2012-10-11 35.7777778
## 12 2012-10-12 60.3541667
## 13 2012-10-13 43.1458333
## 14 2012-10-14 52.4236111
## 15 2012-10-15 35.2048611
## 16 2012-10-16 52.3750000
## 17 2012-10-17 46.7083333
## 18 2012-10-18 34.9166667
## 19 2012-10-19 41.0729167
## 20 2012-10-20 36.0937500
## 21 2012-10-21 30.6284722
## 22 2012-10-22 46.7361111
## 23 2012-10-23 30.9652778
## 24 2012-10-24 29.0104167
## 25 2012-10-25 8.6527778
## 26 2012-10-26 23.5347222
## 27 2012-10-27 35.1354167
## 28 2012-10-28 39.7847222
## 29 2012-10-29 17.4236111
## 30 2012-10-30 34.0937500
## 31 2012-10-31 53.5208333
## 32 2012-11-01 37.3825996
## 33 2012-11-02 36.8055556
## 34 2012-11-03 36.7048611
## 35 2012-11-04 37.3825996
## 36 2012-11-05 36.2465278
## 37 2012-11-06 28.9375000
## 38 2012-11-07 44.7326389
## 39 2012-11-08 11.1770833
## 40 2012-11-09 37.3825996
## 41 2012-11-10 37.3825996
## 42 2012-11-11 43.7777778
## 43 2012-11-12 37.3784722
## 44 2012-11-13 25.4722222
## 45 2012-11-14 37.3825996
## 46 2012-11-15 0.1423611
## 47 2012-11-16 18.8923611
## 48 2012-11-17 49.7881944
## 49 2012-11-18 52.4652778
## 50 2012-11-19 30.6979167
## 51 2012-11-20 15.5277778
## 52 2012-11-21 44.3993056
## 53 2012-11-22 70.9270833
## 54 2012-11-23 73.5902778
## 55 2012-11-24 50.2708333
## 56 2012-11-25 41.0902778
## 57 2012-11-26 38.7569444
## 58 2012-11-27 47.3819444
## 59 2012-11-28 35.3576389
## 60 2012-11-29 24.4687500
## 61 2012-11-30 37.38259964cii median
daily_median_imputation <- aggregate(activity_imputed["AvgSteps"], by=list(Date=activity_imputed$date), FUN=median)
daily_median_imputation## Date AvgSteps
## 1 2012-10-01 34.11321
## 2 2012-10-02 0.00000
## 3 2012-10-03 0.00000
## 4 2012-10-04 0.00000
## 5 2012-10-05 0.00000
## 6 2012-10-06 0.00000
## 7 2012-10-07 0.00000
## 8 2012-10-08 34.11321
## 9 2012-10-09 0.00000
## 10 2012-10-10 0.00000
## 11 2012-10-11 0.00000
## 12 2012-10-12 0.00000
## 13 2012-10-13 0.00000
## 14 2012-10-14 0.00000
## 15 2012-10-15 0.00000
## 16 2012-10-16 0.00000
## 17 2012-10-17 0.00000
## 18 2012-10-18 0.00000
## 19 2012-10-19 0.00000
## 20 2012-10-20 0.00000
## 21 2012-10-21 0.00000
## 22 2012-10-22 0.00000
## 23 2012-10-23 0.00000
## 24 2012-10-24 0.00000
## 25 2012-10-25 0.00000
## 26 2012-10-26 0.00000
## 27 2012-10-27 0.00000
## 28 2012-10-28 0.00000
## 29 2012-10-29 0.00000
## 30 2012-10-30 0.00000
## 31 2012-10-31 0.00000
## 32 2012-11-01 34.11321
## 33 2012-11-02 0.00000
## 34 2012-11-03 0.00000
## 35 2012-11-04 34.11321
## 36 2012-11-05 0.00000
## 37 2012-11-06 0.00000
## 38 2012-11-07 0.00000
## 39 2012-11-08 0.00000
## 40 2012-11-09 34.11321
## 41 2012-11-10 34.11321
## 42 2012-11-11 0.00000
## 43 2012-11-12 0.00000
## 44 2012-11-13 0.00000
## 45 2012-11-14 34.11321
## 46 2012-11-15 0.00000
## 47 2012-11-16 0.00000
## 48 2012-11-17 0.00000
## 49 2012-11-18 0.00000
## 50 2012-11-19 0.00000
## 51 2012-11-20 0.00000
## 52 2012-11-21 0.00000
## 53 2012-11-22 0.00000
## 54 2012-11-23 0.00000
## 55 2012-11-24 0.00000
## 56 2012-11-25 0.00000
## 57 2012-11-26 0.00000
## 58 2012-11-27 0.00000
## 59 2012-11-28 0.00000
## 60 2012-11-29 0.00000
## 61 2012-11-30 34.11321Do these values differ from the estimates from the first part of the assignment?
ANS: yes, they do, the missen values are well filled out which impacts the distribution of the steps
What is the impact of imputing missing data on the estimates of the total daily number of steps?
ANS: it makes it more clear that on average more steps were taken which counted around between 10000 and 15000
activity_imputed_wkdays <- activity_imputed %>%
mutate(wkday = as.factor(ifelse( weekdays(date) %in% c("Saturday", "Sunday"), "weekend","weekday")))
table(activity_imputed$wkday)## Warning: Unknown or uninitialised column: 'wkday'.## < table of extent 0 >g <- ggplot(data = activity_imputed_wkdays, aes(x= interval, y=AvgSteps, col = wkday) ) + geom_line()
g + facet_grid(wkday ~ .)
5 Are there differences in activity patterns between weekdays and weekends?
ANS: Yes. On average, the are more steps over the 600 steps during weekdays than there is during the weekend.