Activity Monitoring on Wearables

By: DicksonC
Data source: http://github.com/rdpeng/RepData_PeerAssessment1

It is now possible to collect a large amount of data about personal movement using activity monitoring devices such as a Fitbit, Nike Fuelband, or Jawbone Up. These type of devices are part of the “quantified self” movement – a group of enthusiasts who take measurements about themselves regularly to improve their health, to find patterns in their behavior.

This project aims to illustrate the activity patterns according to the measurements on wearables.

Importing the required library

library(readr)
library(dplyr)
library(ggplot2)
library(lubridate)
library(lattice)

Loading and preprocessing the data

# Reading dataset
data <- read.csv(unz("../data/activity.zip", "activity.csv"), na.strings = "NA")
# Formatting date column
data$date <- ymd(as.character(data$date))
summary(data)

##      steps             date               interval     
##  Min.   :  0.00   Min.   :2012-10-01   Min.   :   0.0  
##  1st Qu.:  0.00   1st Qu.:2012-10-16   1st Qu.: 588.8  
##  Median :  0.00   Median :2012-10-31   Median :1177.5  
##  Mean   : 37.38   Mean   :2012-10-31   Mean   :1177.5  
##  3rd Qu.: 12.00   3rd Qu.:2012-11-15   3rd Qu.:1766.2  
##  Max.   :806.00   Max.   :2012-11-30   Max.   :2355.0  
##  NA's   :2304

What is mean total number of steps taken per day?

# Grouping data by date and calculate sum
totalsteps <- data %>%
              select(date, steps) %>%
              group_by(date) %>%
              summarize(total.steps = sum(steps))
# Plotting histogram
with(totalsteps, qplot(total.steps, geom = "histogram", ylab = "days"))

# Calculate mean and median of the total number of steps taken per day
summary <- summary(totalsteps)
noquote(summary[c(4, 3),2])

##                                 
## Mean   :10766   Median :10765

What is the average daily activity pattern?

# Grouping data by interval and calculate mean
avgsteps <- data %>%
            select(interval, steps) %>%
            group_by(interval) %>%
            summarize(average.steps = mean(steps, na.rm = TRUE))
# Plotting points
with(avgsteps, qplot(interval, average.steps, geom = "path"))

# Maximum number of steps in 5-minute interval on average across the days
index <- which.max(avgsteps$average.steps)
as.integer(avgsteps[index, "interval"])

## [1] 835

Imputing missing values

# Total number of missing values in the dataset
sum(is.na(data))

## [1] 2304

# Replacing missing values with mean by interval
cleandata <- data %>%
             group_by(interval) %>%
             mutate(average.steps = mean(steps, na.rm = TRUE)) %>%
             mutate(steps = ifelse(is.na(steps), average.steps, steps)) %>%
             select(steps, date)
# Checking if any missing values in the new dataset
sum(is.na(cleandata))

## [1] 0

# Grouping cleandata by date and calculate sum
totalcleansteps <- cleandata %>%
                   select(date, steps) %>%
                   group_by(date) %>%
                   summarize(total.steps = sum(steps))
# Plotting histogram
with(totalcleansteps, qplot(total.steps, geom = "histogram", ylab = "days"))

# Calculate mean and median of the total number of steps taken per day
cleansummary <- summary(totalcleansteps)
# Difference with missing values
data.frame(Data = c("With NAs", "Without NAs"), 
           Mean = c(parse_number(summary[4, 2]), parse_number(cleansummary[4, 2])),
           Median = c(parse_number(summary[3,2]), parse_number(cleansummary[3, 2])))

##          Data  Mean Median
## 1    With NAs 10766  10765
## 2 Without NAs 10766  10766

Are there differences in activity patterns between weekdays and weekends?

wdwe <- cleandata %>%
        # Creating new variable for weekdays/weekends in cleandata
        mutate(weekday.weekend = factor(ifelse(wday(date) %in% c(1,7), "weekend", "weekday"))) %>%
        # Grouping cleandata by weekdays/weekends and interval before calculate average
        group_by(weekday.weekend, interval) %>%
        summarize(Average.steps = mean(steps))
with(wdwe, qplot(interval, Average.steps, facets = weekday.weekend~., geom = "line"))