Exploring the Data from Fitness Device

Overview

This data is taken from the Coursera Assignment. It comes from the activity monitoring devices such as Fitbit, Nike Fuelband, or Jawbone Up. This device collects data at 5 minute intervals through out the day. It consists of two months of data from an anonymous individual collected during the months of October and November, 2012 and include the number of steps taken in 5 minute intervals each day.

Load libraries

library(ggplot2)
library(dplyr)
library(lubridate)

Loading and preprocessing the data

1. Loading the data.

data <- read.csv("./activity.csv")

2. Preprocessing the data.

data$date <- as.Date(data$date)

What is the mean total number of steps taken per day?

stepSumPerDay <- data %>% group_by(date) %>% summarise(stepSum = sum(steps, na.rm = TRUE))

plot1 <- ggplot(stepSumPerDay, aes(x = stepSum)) + geom_histogram(bins = 50) +
        labs(x = "Total Steps Taken per Day", y = "Frequency")
print(plot1)

stepMean <- mean(stepSumPerDay$stepSum)
stepMedian <- median(stepSumPerDay$stepSum)

1. Mean of total steps taken per day : 9354.2295082
2. Median of total steps taken per day : 10395

What is the average daily activity pattern?

1. Time Series plot of the 5-minute interval.

stepSumPerInterval <- data %>% group_by(interval) %>% summarise(stepMeanInterval = mean(steps, na.rm = TRUE))

plot2 <- ggplot(stepSumPerInterval, aes(x = interval, y = stepMeanInterval)) + geom_line() +
        labs(x = "Interval", y = "Average Step")
print(plot2)

2. Interval with the maximum number of steps.

maximumStep <- which.max(stepSumPerInterval$stepMeanInterval)
intervalMax <- stepSumPerInterval[maximumStep, 1]

Maximum step occured at @ 8:35

Imputing missing values

1. Calculate and report the total number of missing values in the dataset.

totalNA <- length(data$steps[is.na(data$steps)])

Total number of NAs : 2304

2. Will use the mean of the 5-minute interval as replacement for the NAs.
3. Create a new dataset that is equal to the original dataset but with the missing data filled in.

stepSumPerInterval <- data %>% group_by(interval) %>% summarise(stepMeanInterval = mean(steps, na.rm = TRUE))

data_2 <- merge(data, stepSumPerInterval)
data_2 <- data_2[order(data_2$date),]
data_2$steps <- ifelse(is.na(data_2$steps), data_2$stepMeanInterval, data_2$steps)
data_2 <- data_2[, c(2,3,1)]

4. Make a histogram of the total number of steps taken each day. Calculate and report the mean and median total number of steps taken per day.

stepSumPerDay_2 <- data_2 %>% group_by(date) %>% summarise(stepSum = sum(steps, na.rm = TRUE))

plot3 <- ggplot(stepSumPerDay_2, aes(x = stepSum)) + geom_histogram(bins = 50) +
        labs(x = "Total Steps Taken per Day", y = "Frequency")
print(plot3)

Noticeably, the chart looks more “bell-shaped” compared to the first chart.

stepSumPerDay_2 <- data_2 %>% group_by(date) %>% summarise(stepSum = sum(steps, na.rm = TRUE))

stepMean_2 <- mean(stepSumPerDay_2$stepSum)
stepMedian_2 <- median(stepSumPerDay_2$stepSum)

1. Mean of total steps taken per day : 1.076618910^{4}
   
2. Median of total steps taken per day : 1.076618910^{4}

Are there differences in activity patterns between weekdays and weekends?

data_2$weekday <- wday(data_2$date)
data_2$daytype[data_2$weekday > 1 | data_2$weekday < 7] <- "weekdays"
data_2$daytype[data_2$weekday == 1 | data_2$weekday == 7] <- "weekends"

stepSumPerInterval_2 <- data_2 %>% group_by(daytype, interval) %>% summarise(stepMeanInterval = mean(steps, na.rm = TRUE))

plot4 <- ggplot(stepSumPerInterval_2, aes(x = interval, y = stepMeanInterval)) + geom_line() + facet_wrap(~daytype, nrow = 2, scales = "free_y") +
        labs(x = "Interval", y = "Average Steps")
print(plot4)

The graph shows difference between weekdays and weekends.