Author: Andria Hall

Introduction:

This report is a Reproducible Research in a single R Markdown document that was processed by ‘knitr’ and transformed into a HTML file. It was created from a dataset that consists of 17568 observations and 3 variables. This dataset was generated from a personal activity monitoring device that collects data at 5-minute intervals throughout the day by an anonymous individual during the month of October and November 2012.

Prepare the working evironment

Since report requires writing code chunks in the R markdown document, I have set echo = TRUE and set results = hold as global options so that the reader can read the code.

library(knitr)
opts_chunk$set(echo = TRUE, results = 'hold')

Load required packages

# load packages
suppressMessages(require("data.table"))
suppressMessages(require("dplyr"))
suppressMessages(require("tidyr"))
suppressMessages(require("ggplot2"))

Loading and preprocessing the data

Here the activity dataset is loaded and processed for analysis

# load the activity data
setwd("/Users/Andria/data/RepData_PeerAssessment1")
f <- file.path(getwd(), "activity.zip")
activity <- tbl_df(read.csv(unz(f, "activity.csv"),header= TRUE, sep= ","))

# explore activity data
head(select(activity, steps:interval))
## Source: local data frame [6 x 3]
## 
##   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       25

What is mean total number of steps taken per day?

Here the missing values are ignored and the total number of steps computed

steps_taken <- aggregate(steps ~ date, activity, sum, na.rm = TRUE)
head(steps_taken)
##         date steps
## 1 2012-10-02   126
## 2 2012-10-03 11352
## 3 2012-10-04 12116
## 4 2012-10-05 13294
## 5 2012-10-06 15420
## 6 2012-10-07 11015
  1. Now we can plot an histogram from steps_taken showing the total number of steps taken per day, using the appropriate bin interval.
g <- ggplot(steps_taken, aes(x=steps)) + 
        geom_histogram(fill=heat.colors(1), col="black", binwidth=1000) +
        labs(x = "Number of Steps per Day") + 
        labs(y = "Number of Times per Day") +
        labs(title = "Histogram of Total Number of Steps Taken per Day") +
        theme_bw()
print(g)

  1. Now we calculate the mean and median total number of steps taken per day
steps_taken %>% summarise(steps_mean= mean(steps), steps_median = median(steps))
##   steps_mean steps_median
## 1   10766.19        10765
  1. The mean number of steps is: 10766.19
  2. The median number of steps is: 10765

What is the average daily activity pattern?

  1. We will now make a time series plot of the average number of steps taken across all days(y-axis), of the 5-minute interval (x-axis).
interval_steps <- activity %>% group_by(interval) %>% summarise(avg_steps= mean(steps, na.rm=TRUE))

# time series plot of the 5-minute interval
g <- ggplot(interval_steps, aes(x=interval, y=avg_steps)) +
        geom_line(color="blue", size=1) +
        labs(title= "Average Daily Activity Pattern", x = "5-minute Interval", y="Number of Steps Taken") +
        theme_bw()
print(g)

  1. We will now calculate which 5-minute interval, on average across all the days in interval_steps, contains the maximum number of steps.
interval_steps[which.max(interval_steps$avg_steps),]
## Source: local data frame [1 x 2]
## 
##   interval avg_steps
## 1      835  206.1698

We can see from the report that the 835th, 5-minute interval, has the average maximum of 206 steps.

Imputing missing values

We will now calculate and report the total number of rows with NA.from the activity dataset created earlier.

sum(is.na(activity))
## [1] 2304
  1. The total number of rows with NA is: 2304

Now we will create a new dataset called fill_activity, by filling in all of the missing NA values in the activity dataset of corresponding interval and the mean for 5-minute interval in the interval_steps dataset.

fill_activity <- activity # make a copy of original dataset
sapply(unique(activity$interval), # make interval distinct
       function(x)
               fill_activity[is.na(fill_activity) & (fill_activity$interval == x), 1] <<- interval_steps$avg_steps[interval_steps$interval == x])
##   [1]   1.7169811   0.3396226   0.1320755   0.1509434   0.0754717
##   [6]   2.0943396   0.5283019   0.8679245   0.0000000   1.4716981
##  [11]   0.3018868   0.1320755   0.3207547   0.6792453   0.1509434
##  [16]   0.3396226   0.0000000   1.1132075   1.8301887   0.1698113
##  [21]   0.1698113   0.3773585   0.2641509   0.0000000   0.0000000
##  [26]   0.0000000   1.1320755   0.0000000   0.0000000   0.1320755
##  [31]   0.0000000   0.2264151   0.0000000   0.0000000   1.5471698
##  [36]   0.9433962   0.0000000   0.0000000   0.0000000   0.0000000
##  [41]   0.2075472   0.6226415   1.6226415   0.5849057   0.4905660
##  [46]   0.0754717   0.0000000   0.0000000   1.1886792   0.9433962
##  [51]   2.5660377   0.0000000   0.3396226   0.3584906   4.1132075
##  [56]   0.6603774   3.4905660   0.8301887   3.1132075   1.1132075
##  [61]   0.0000000   1.5660377   3.0000000   2.2452830   3.3207547
##  [66]   2.9622642   2.0943396   6.0566038  16.0188679  18.3396226
##  [71]  39.4528302  44.4905660  31.4905660  49.2641509  53.7735849
##  [76]  63.4528302  49.9622642  47.0754717  52.1509434  39.3396226
##  [81]  44.0188679  44.1698113  37.3584906  49.0377358  43.8113208
##  [86]  44.3773585  50.5094340  54.5094340  49.9245283  50.9811321
##  [91]  55.6792453  44.3207547  52.2641509  69.5471698  57.8490566
##  [96]  56.1509434  73.3773585  68.2075472 129.4339623 157.5283019
## [101] 171.1509434 155.3962264 177.3018868 206.1698113 195.9245283
## [106] 179.5660377 183.3962264 167.0188679 143.4528302 124.0377358
## [111] 109.1132075 108.1132075 103.7169811  95.9622642  66.2075472
## [116]  45.2264151  24.7924528  38.7547170  34.9811321  21.0566038
## [121]  40.5660377  26.9811321  42.4150943  52.6603774  38.9245283
## [126]  50.7924528  44.2830189  37.4150943  34.6981132  28.3396226
## [131]  25.0943396  31.9433962  31.3584906  29.6792453  21.3207547
## [136]  25.5471698  28.3773585  26.4716981  33.4339623  49.9811321
## [141]  42.0377358  44.6037736  46.0377358  59.1886792  63.8679245
## [146]  87.6981132  94.8490566  92.7735849  63.3962264  50.1698113
## [151]  54.4716981  32.4150943  26.5283019  37.7358491  45.0566038
## [156]  67.2830189  42.3396226  39.8867925  43.2641509  40.9811321
## [161]  46.2452830  56.4339623  42.7547170  25.1320755  39.9622642
## [166]  53.5471698  47.3207547  60.8113208  55.7547170  51.9622642
## [171]  43.5849057  48.6981132  35.4716981  37.5471698  41.8490566
## [176]  27.5094340  17.1132075  26.0754717  43.6226415  43.7735849
## [181]  30.0188679  36.0754717  35.4905660  38.8490566  45.9622642
## [186]  47.7547170  48.1320755  65.3207547  82.9056604  98.6603774
## [191] 102.1132075  83.9622642  62.1320755  64.1320755  74.5471698
## [196]  63.1698113  56.9056604  59.7735849  43.8679245  38.5660377
## [201]  44.6603774  45.4528302  46.2075472  43.6792453  46.6226415
## [206]  56.3018868  50.7169811  61.2264151  72.7169811  78.9433962
## [211]  68.9433962  59.6603774  75.0943396  56.5094340  34.7735849
## [216]  37.4528302  40.6792453  58.0188679  74.6981132  85.3207547
## [221]  59.2641509  67.7735849  77.6981132  74.2452830  85.3396226
## [226]  99.4528302  86.5849057  85.6037736  84.8679245  77.8301887
## [231]  58.0377358  53.3584906  36.3207547  20.7169811  27.3962264
## [236]  40.0188679  30.2075472  25.5471698  45.6603774  33.5283019
## [241]  19.6226415  19.0188679  19.3396226  33.3396226  26.8113208
## [246]  21.1698113  27.3018868  21.3396226  19.5471698  21.3207547
## [251]  32.3018868  20.1509434  15.9433962  17.2264151  23.4528302
## [256]  19.2452830  12.4528302   8.0188679  14.6603774  16.3018868
## [261]   8.6792453   7.7924528   8.1320755   2.6226415   1.4528302
## [266]   3.6792453   4.8113208   8.5094340   7.0754717   8.6981132
## [271]   9.7547170   2.2075472   0.3207547   0.1132075   1.6037736
## [276]   4.6037736   3.3018868   2.8490566   0.0000000   0.8301887
## [281]   0.9622642   1.5849057   2.6037736   4.6981132   3.3018868
## [286]   0.6415094   0.2264151   1.0754717
  1. Even though the report shows no missing values are in dataset, let us reaffirm.
sum(is.na(fill_activity))
## [1] 0
  1. The total number of rows with NA is: 0

We will now plot an histogram of the total number of steps taken each day from the filled-in fill_activity dataset.

fill_steps_taken <- aggregate(steps ~ date, fill_activity, sum, na.rm = TRUE)

# use the ggplot to make an hisgram of the total numbers of steps each day  
g <- ggplot(fill_steps_taken, aes(x=steps)) + 
        geom_histogram(fill=heat.colors(1), col="black", binwidth=1000) +
        labs(x = "Number of Steps per Day") + 
        labs(y = "Number of Times per Day") +
        labs(title = "Histogram of Total Number of Steps Taken per Day with Filled-in Data") +
        theme_bw()
print(g)

We will now calculate the and report the mean and median total number of steps taken per day computed in the dataset fill_steps_taken.

fill_steps_taken %>% summarise(steps_mean= mean(steps), steps_median = median(steps))
##   steps_mean steps_median
## 1   10766.19     10766.19
  1. The mean number of steps is: 10766.19

  2. The median number of steps is: 10766.19

  3. We can see that the mean 10766.19 is the same for calculated steps in the dataset with missing values and dataset without missing values, while the median is showing a slight difference of 1.19 in favor of the filled-in dataset, that is equivalent to the 10766.19 of its calculated mean value.

# dataset of filled-in values for the average 5-minute interval of steps taken
fill_interval_steps <- fill_activity %>% group_by(interval) %>% summarise(fill_avg_steps= mean(steps, na.rm=TRUE))

#calculated 5-minute interval on average of maximum number of steps from filled-in dataset
fill_interval_steps[which.max(fill_interval_steps$fill_avg_steps),]
## Source: local data frame [1 x 2]
## 
##   interval fill_avg_steps
## 1      835       206.1698
  1. From the report the 835th, 5-minute interval, has the same average maximum of 206 steps for the filled-in dataset as for the missing dataset, therefore the impact of imputting missing data in the missing dataset remains the same for the maximim average numbers of steps.

Are there differences in activity patterns between weekdays and weekends?

  1. We will create a new factor variable in the dataset with two levels - weekday and weekend indicating whether a given date is a weekday or weekend day.
fill_activity <- fill_activity %>%  
        mutate(days = as.factor(ifelse(weekdays(as.Date(date)) %in% c("Saturday","Sunday"), "Weekend", "Weekday"))) %>%
        group_by(days)
  1. Then we will make a panel plot containing a time series plot of the 5-minute interval on the (x-axis) and the average number of steps taken, averaged across all weekday days or weekend days on the (y-axis)
week_days <- fill_activity %>%
        group_by(interval, days) %>%
        summarise(avg_steps = mean(steps))

g <- ggplot(week_days, aes(x=interval, y=avg_steps)) +
        geom_line(color="blue") +
        facet_wrap(~ days, nrow=2, ncol=1) +
        labs(x="interval", y="Number of Steps") +
        theme_bw()
print(g)

We observe from the graph above that more activities were done across all weekends than on weekdays, but the maximum number of activities on average are between 500 and 1000 intervals that were done on weekdays. This could be based on the fact that activities on weekdays between these intervals, mostly followed a consistent activity pattern.

The rest of the intervals shows on average, a higher but constant distribution of activity done on weekends than on weekdays, once again could be based on a consistent activity pattern, that gradually decreases as the activity period comes to an end.

Conclusion

This report concludes by asking this vital question. What would have been the results of the activity patterns in all analysis, if the original dataset did not consist of 2304 missing values?