We support a number of different chart presentations. In general, charting is a complex topic. As the designer of a chart, you typically have to make three choices:
- How many individual series 'lines' do you draw?
- What values are on the X and Y axes?
- What chart display type do you use?-- line, bar chart, etc.
We have tried to keep it simple by limiting the possible answers for each of these choices:
This type of chart is ideal for segmenting and drilling down a dataset. A histogram is a special kind of bar chart to show aggregate distributions. For example, if you want to see the number of customers who purchased each automobile model, you specify the column whose values will be aggregated. The X axis gets one entry for each unique value in this column and the Y axis shows the count of the number of rows that have that value in that column. You can specify multiple levels of drill-down by specifying multiple columns. You can also represent the aggregated values as a pie chart.
Row Series Chart
This type of chart is appropriate for spreadsheet data where most of the data is numeric on a uniform scale (eg: monthly sales data). Every row forms a series, each of which is identified by the row key. You choose a subset of the column names that will be charted on the X axis. The Y axis is based on the numeric values in the row cells. You can choose a line chart, a bar chart, or a stacked bar chart as the display type.
A good example of a Row Series chart is the 'By Month' view in the Sales Report sample.
Column Series Chart
A column series chart is appropriate for spreadsheet data with just a few rows corresponding to items to compare (eg: monthly sales data by category). Every column forms a series, each of which is identified by the column name. Each row (using its row key column) forms one entry on the X axis. The Y axis is based on the numeric values in the row cells. You can choose a line chart, a bar chart or a stacked bar chart as the display type.
If the column definition specifies explicit MinValue and MaxValue properties, they are used as the lower and upper ends of the Y-range. Otherwise, we use the min and max data values in the column as the lower and upper ends of the Y-range.
A good example of a Column Series chart is the "By Type" view in the Sales Report sample.
A pie chart is appropriate for representing part-to-whole comparisons and for getting a quick read on approximate percentages. These work best when there are a very small number of categories (two or three). Every row (using its row key column) creates one category, one slice of the pie. The angle of the slice is determined by numeric values in the row cells - the percentage of the total value of the categories.
You can see an example in the Sales Report sample.
The table and detail views accessed by clicking the Data button or the individual bars of the chart can be configured by creating Reference Views or modifying those generated by the system. Over time, we will add more chart types while trying to keep the concepts simple.