Topics covered in this article include:
- Selecting a Data Set
- Filter Data by Date
- Filter Data by Column Values
- Export Data
- Visualize Data
The analysis page displays a comparison of measurements for the selected tests across schedules. To compare data click the Analysis link in the main navigation toolbar.
Selecting a Data Set
To generate a comparison report, expand the desired section by clicking the arrow to the right:
Next select the individual test or test group. If there is data for the measurement a table will be created with the following columns:
- Site: site where the measurements were taken.
- Machine: machine from which the measurements were taken.
- Schedule: which schedule was used to record the measurements.
- Energy: for energy dependent measurements, this column will show the the energy at which the measurement was taken. For measurements that are not energy dependent, this column will not be displayed.
- Count: number of reports containing the measurement.
- Mean, Median, Max, Standard Deviation, Max, Min: statistics calculated for each row of the table.
The displayed result can be further refined by selecting the desired test within the group, selecting an energy for energy dependent tests, and by selecting one or more meta items. Below is an example of a measurement that is energy dependent:
Here is another example with a meta item:
Note that more than one meta item can be selected and there are tests that are both energy dependent and include meta items.
Filter Data by Date
You may optionally select a date range from which to pull data.
The default range is back 1 year from the current date. There are also several other options for date range:
Custom Range can be used to set any start or end date desired.
Filter Data by Column Values
The results may be filtered by any column, except for the statistic columns. Some or all of the columns may be filtered at the same time by selecting the desired values below those columns:
Multiple items may be selected for a column. To clear previously selected items, click the small x in the upper right of the filter box.
The table may be exported to a .csv file by clicking the Export Summary CSV button or the underlying measurements can be exported using the Export All Data as CSV button as shown below:
A Box and Whisker plot appears under the summary data to visualize data set variability. There must be a minimum of four data points (count column) in a data set to be displayed.
- The top of the box is the third quartile. The third quartile is the middle value between the overall median and the largest non-outlier value.
- The bottom of the box is the first quartile. The first quartile is the middle value between the overall median and the smallest non-outlier value.
- The line within the box within the box indicates the data's median value.
- The height of the box is the difference between the first and third quartiles; known as the interquartile range (IQR). The IQR determines the classification of data points as outliers. Points are outliers if they are higher than the third quartile plus 1.5 x IQR or less than the first quartile minus 1.5 x IQR.
- The top whisker represents the maximum non-outlier measurement.
- The bottom whisker represents the minimum non-outlier measurement.
- Open circles display outliers falling outside the whiskers.
In some cases the box or whiskers will appear compressed due to the same data values are repeated in the data set. General information on a box and whiskers plot may be found on Wikipedia in the Box plot article.