PA1_template.Rmd

# Reproducible Research 
## Peer Assessment 1

### Libraries needed for this asignment.
```{r}
library(lattice)
```

## Loading and preprocessing the data
### 1. Loads activity data for analysis.
```{r}
activityData <- read.csv("activity.csv", stringsAsFactors = FALSE)
```

### 2. Performs pre-processing of activity data for usability in the subsequent analysis.
```{r}
totalStepsDate <- aggregate(steps ~ date, activityData, sum, na.rm = TRUE)
avgStepsInterval <- aggregate(steps ~ interval, activityData, mean, na.rm = TRUE)
maxStepsCount <- which.max(avgStepsInterval$steps)
```

## What is mean total number of steps taken per day?
### 1. A histogram of the total number of steps taken each day.
```{r}
hist(totalStepsDate$steps, col = "blue",
     main = "Total Number Steps Per Day", xlab = "Steps Per Day")
```

### 2. Calculates and reports the mean and median step counts.
```{r}
meanStepsDay <- mean(totalStepsDate$steps)
medianStepsDay <- median(totalStepsDate$steps)
```

The Mean steps is `r meanStepsDay`. The Median steps is `r medianStepsDay`.

## What is the average daily activity pattern?
### 1. A time series plot demonstrating a 5 minute interval of the average steps taken across all days.  
```{r}
plot(avgStepsInterval$interval, avgStepsInterval$steps,
     type = "l", frame = FALSE, col = "blue",
     main = "Average Activity Per Day",
     xlab = "Interval (5 Minutes)", ylab = "Average Number Steps")
```

### 2. The interval, on average across all days, with the maximum number of steps.
```{r}
maxStepsInterval <- avgStepsInterval[maxStepsCount,]$interval
```

Interval `r maxStepsInterval` has the most steps.

## Imputing missing values
### 1. Calculates and reports the number of missing values in the dataset
```{r}
missingValuesTotal <- sum(is.na(activityData$steps))
```

A total of `r missingValuesTotal` missing values exist in the dataset.

### 2. & 3. The strategy for filling in missing values in the dataset.  Loop through a copy of the dataset looking for missing step values.  When found, replace the missing step value with the average number of steps for the matching interval.
```{r}
activityDataComplete <- activityData
activityCount = 0
for (i in 1:nrow(activityDataComplete)) {
    if (is.na(activityDataComplete[i, ]$steps)) {
        activityDataComplete[i, ]$steps <- 
            avgStepsInterval[avgStepsInterval$interval == activityDataComplete[i, ]$interval, ]$steps
        activityCount <- activityCount + 1
    }
}
```

### 4. The same histogram as above, this time using the complete dataset.  
```{r}
totalStepsDateComplete <- aggregate(steps ~ date, data = activityDataComplete, sum)
hist(totalStepsDateComplete$steps, col = "green",
     main = "Total Number Steps Per Day (No Missing Values)", xlab = "Steps Per Day")

meanStepsDayComplete <- mean(totalStepsDateComplete$steps)
medianStepsDayComplete <- median(totalStepsDateComplete$steps)
```

The Mean value remains `r meanStepsDayComplete` due to the strategy of using mean values for replacement.
The Median value has become `r medianStepsDayComplete` due to the mean value inserted as an actual value.

## Are there differences in activity patterns between weekdays and weekends?
### 1. A new factor column indicating whether a given date is a "weekday" or a "weekend".
```{r}
activityDataCompleteDay <- activityDataComplete
activityDataCompleteDay$day <- weekdays(as.Date(activityDataCompleteDay$date)) %in% c("Saturday","Sunday")
activityDataCompleteDay$day <- ifelse(activityDataCompleteDay$day, "weekend", "weekday")
```

### 2. Performs preprocessing of data to cacluate averages by interval and by weekday or weekend. The resulting panel plot demonstrates 5 minute intervals of the average total steps across all weekdays and weekends
```{r}
avgStepsIntervalDay <- aggregate(steps ~ interval + day, activityDataCompleteDay, mean)
xyplot(steps ~ interval | factor(day), data = avgStepsIntervalDay, 
       xlab = "Intervals (5 Minutes)",
       ylab = "Average Steps Across Weekday/Weekend", 
       main = "Differences in Activity Patterns Between Weekdays and Weekends",
       type = "l")
```

The plot indicates late morning activity is highest on the weekdays.  However, overall activity remains higher throught the entire day on the weekends.