Reproducible Research Assignment, Coursera Data Science

Data taken from course website.

Description (from course website):

This assignment makes use of data from a personal activity monitoring device. This device collects data at 5 minute intervals through out the day. The data 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.

Code to Read the Data

activ <- read.csv("activity.csv")

Processing Code

intervalID <- rep(1:288,61) 
dateID <- rep(1:61, each = 288)
activ <- cbind(activ, dateID, intervalID)

Analysis Code

daysums <- c()
for (i in 1:61){
        value <- sum(activ[activ$dateID == i,1], na.rm=TRUE)
        daysums <- c(daysums, value)
}
intervalmeans <- c()
for (i in 1:288){
        value2 <- mean(activ[activ$intervalID == i,1], na.rm=TRUE)
        intervalmeans <- c(intervalmeans,value2)
}

Plotting Code

# create histogram of total steps per day
par(mar = c(4,4,3,3)) 
hist (daysums, breaks =8, col ="steelblue", 
      main = "Histogram of Total Steps Taken per Day",
      xlab = "Total Steps per Day", ylab = "Frequency in Days",
      xlim = c(0, 25000))

mymean <- round(mean(daysums), 2)
mymedian <- median(daysums)
meanpaste <- paste ("Mean total amount of steps per day is", mymean)
medianpaste <- paste ("Median total amount of steps per day is", mymedian)
print (meanpaste)
## [1] "Mean total amount of steps per day is 9354.23"
print (medianpaste)
## [1] "Median total amount of steps per day is 10395"
plot(intervalmeans, type="l", col="palegreen2",
     lwd = 3,
     main = "Average Steps Taken,\n Intervals 1-288",
     xlab = "Interval ID", 
     ylab = "Average number of steps",
     xlim = c(0,288))

maxinterval <- which(intervalmeans == max(intervalmeans))
maxintvalue <- activ[maxinterval,3]
maxpaste <- paste ("The interval that, on average, contains the largest amount of steps is ", 
                maxintvalue, ". It is represented as IntervalID", maxinterval, 
                "on the attached plot.")
print(maxpaste)
## [1] "The interval that, on average, contains the largest amount of steps is  835 . It is represented as IntervalID 104 on the attached plot."

Imputation Strategy

# 1. calculate and report total number of missing values

missing <- which(is.na(activ$steps)) # gives rows in which steps measurements are NA
totalmiss <- length(missing)
missingpaste <- paste ("There are", totalmiss, "missing values.")
print (missingpaste)
## [1] "There are 2304 missing values."
# 2. Devise a strategy for filling in missing values

#       I opt for the mean of the interval in question.  I cannot rely on measures
#       for the DAY in question as the missing values tend to be a result of
#       entire days of missing data.

# 3. Create a new dataset that is equal to the original dataset
#    but with the missing data filled in.

#       for every row in "missing":
#               identify the interval containing the missing data, using intervalID
#               get the value of intervalmeans[that particular intervalID]
#               replace the NA for that row with the obtained value

activ2 <- activ
for (i in 1:length(missing)) {
        val1 <- missing[i] # get the row number
        val2 <- activ2[val1, "intervalID"] # identify the interval
        val3 <- intervalmeans[val2] # get the mean for that interval
        activ2[val1, "steps"] <- round(val3, 1) # change the NA value to the mean taken above
}

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

daysums2 <- c()
for (i in 1:61){
        value2 <- sum(activ2[activ2$dateID == i,1], na.rm=TRUE)
        daysums2 <- c(daysums2, value2)
}
 
hist (daysums2, breaks =8, col ="dodgerblue2", 
      main = "Histogram of Total Steps Taken per Day \n(NAs Imputed)",
      xlab = "Total Steps per Day", ylab = "Frequency in Days",
      xlim = c(0, 25000), ylim = c(0,25))

mymean2 <- round(mean(daysums), 2)
mymedian2 <- median(daysums)
meanpaste2 <- paste ("With NAs imputed, mean total amount of steps per day is", mymean2)
medianpaste2 <- paste ("With NAs imputed, median total amount of steps per day is", mymedian2)
print (meanpaste2)
## [1] "With NAs imputed, mean total amount of steps per day is 9354.23"
print (medianpaste2)
## [1] "With NAs imputed, median total amount of steps per day is 10395"

Differences in Activity Patterns

# 1. Create a new factor variable with levels "weekday"
# and "weekend" indicating status of any given day.


#daycode <- factor(c("Weekday", "Weekday", "Weekday", "Weekday", 
#                    "Weekday", "Weekend","Weekend"))

#daytype <- rep(daycode, each = 288)
#daytype2 <- rep(daytype, 8)
#dayextra <- rep(daycode[1], 1440)
#daycategory <- c(daytype2, dayextra)
myweek <- rep("Weekday", 1440)
myweekend <- rep("Weekend", 576)
wholeweek <- c(myweek, myweekend)
daycategory <- factor(c(rep(wholeweek,8), myweek))

activ2 <- cbind(activ2,  daycategory)



# 2. Make a panel plot containing 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 weekday days or weekend days (y-axis).


intervalmeans.wkdy <- c()
intervalmeans.wknd <- c()
for (i in 1:288){
        value.wkdy <- mean(activ2[activ2$intervalID == i & activ2$daycategory == "Weekday", 1])
        intervalmeans.wkdy <- c(intervalmeans.wkdy,value.wkdy)
        value.wknd <- mean(activ2[activ2$intervalID == i & activ2$daycategory == "Weekend", 1])
        intervalmeans.wknd <- c(intervalmeans.wknd,value.wknd)
}


plot(intervalmeans.wkdy, type="l", col="darkseagreen4",
     main = "Average Steps by Interval, WEEKDAY",
     xlab = "Interval ID", 
     ylab = "Average number of steps",
     xlim = c(0,288), ylim = c(0,250))

plot(intervalmeans.wknd, type="l", col="darkorchid4",
     main = "Average Steps by Interval, WEEKEND",
     xlab = "Interval ID", 
     ylab = "Average number of steps",
     xlim = c(0,288), ylim = c(0, 250))

intervalmeans3 <- c()
for (i in 1:288){
        value3 <- mean(activ2[activ2$intervalID == i,1])
        intervalmeans3 <- c(intervalmeans3,value3)
}
library(lattice)
mypanel <- function(x,y){
       panel.xyplot(x,y,pch=19, type = "l")
       }
xyplot(steps~intervalID|daycategory, data = activ2,
      layout = c(1,2),
      type = "l",
      main="Activity Comparison: Weekday vs. Weekend",
      xlab="Interval ID",
      ylab="Average Steps",
      panel = mypanel)