If the data panel is not already open, you can show it
by clicking the Data button, .
The data panel provides a structured list of all available data columns in a data table, that is, an overview of your data. It can be the starting point for configuring the visualizations, because from the panel, you can drag columns to drop targets in the middle of the visualizations and to the visualization axes. In addition, filters are associated with the columns, so you can filter the data from the panel.
The columns are by default organized into categories to make them easy to locate in the list. For example, loading the Microsoft Excel file below results in the data panel to the right. Columns that contain numerical data suitable to aggregate are gathered in the NUMBERS section, columns related to time in the TIME section, and columns useful for splitting the data in the CATEGORIES section. If more than one data table is available in the analysis, you first select, in the drop-down menu at the top of the data panel, which data table to display.
The suggested category for a column may not be the preferred category.
For example, if the values in a column are integers like 4633, 4637, and 4638, the suggested category might be NUMBERS. This means that you can calculate sums or averages, or use other aggregation methods. However, these numbers could just as well represent, for example, employee numbers or purchase order numbers, and any aggregations would not be relevant. In these cases, the column values should instead be handled as IDENTIFIERS or CATEGORIES.
Then you can change the categorization. Right-click the column, open the Change Categorization menu, and select the desired category, and the column will appear beneath the selected category. Alternatively, drag the column to the header of the preferred category as shown below. This method works only when the wanted category is already represented in the panel.
Tip: If you want to change the category for several columns simultaneously, select the columns in question, and they will all be included in the move to the new category using the methods above.
These methods also work for changing the category of one column at a time in the Recommended visualizations window.
Note: The category of a column will affect which kinds of visualizations are advised in Recommended visualizations.
The Change categorization menu also offers the Reset to default option. If you select this option, the organization of the columns into categories will be totally based on the column data types.
Selecting columns on an axis
You can drag a column from the data panel to a visualization and drop it on any of the column selectors indicated with a blue color or on a drop target. The visualization will immediately reflect the new setting.
Note: More than one column can be selected on an axis. You can drag another column from the panel and drop it on the drop-down menu next to the first selected column on the axis.
If a hierarchy filter has been created, it is possible to drag the entire hierarchy or any of its sub-levels to an axis.
Filtering data in columns
Depending on whether you are working with in-memory or in-database data, you get access to the filters in the data panel in slightly different ways.
For in-memory
data, filters are created automatically. Hover with the cursor over
a column in the panel, and click the Show filter button that appears. Use the opened
filter to limit the data directly.
For in-database
data, the filters must first be created. Hover with the cursor over
a column, click the Show filter button that appears,
and click to create a filter. Use the created filter to limit the
data.
Note: Depending on the
database you are using, this may take some time.
Details about which columns have been filtered, and to which values, are displayed at the bottom of the data panel. If you want to modify what has been filtered in a filter, you can open the filter from here by simply clicking it.
Above, the expanded view is shown. To collapse the filter
details, click the button.
Filtering can be reset. Click the
button to reset separate filters, and the
button to
reset all filters.
Filtering can also be handled using the filters panel, which offers more filtering options.
Data Panel Views
The sections in the data panel differ slightly depending
on the data source and the data content. Some examples of data panels
are described below. For most of the different types of views it is possible
to click on the Expand data panel for tools and details
button, , to get an
expanded view with more
information about the data.
In-Memory Data and In-Database Relational Data from a Single Table
Data in in-memory data tables, or in-database data that has a single source table from a relational database, is simply displayed as a list of the categorized columns in the selected data table. The columns are categorized in any of the following sections: NUMBERS, CURRENCY, TIME, LOCATION, CATEGORIES, IDENTIFIERS, TEXT, IMAGES, and BINARY.
Number |
Section |
Description |
1 |
Data table selector |
[Only available if more than one data table have been added to the analysis.] Lists all data tables in the analysis. Choose a different data table to view the columns in that data table instead. |
2 |
Expand button |
Expands the data panel so you can view tools and details about the selected data table or specific columns. If one column is selected in the Data panel, the expanded panel will show details about that column. If more than one column is selected, you will see an overview of the selected columns, and, if no column is selected, you get an overview of the entire data table. |
3 |
Columns |
Lists all columns available in the selected data table. If the columns are shown 'Categorized', then columns of similar type are grouped together. You can change how columns are displayed here
from the data table view of the expanded data panel, by clicking
on the Sort order in data panel button,
|
Note: Using the right-click menu in the Columns field, you can easily specify that a column contains geocoding information, that is, information that can be used to position data on a map.
In-Database Relational Data from Joined Tables
If data tables from in-db databases have been joined with relations in the Views in Connection dialog, they can be treated as a single, virtual data table within Spotfire. In this case, there will be an additional field showing the source structure in the data panel as seen below. If no relations have been defined, each data table in the external data connection will be a separate data table within Spotfire as shown above.
Number |
Section |
Description |
1 |
Data table selector |
[Only available if more than one data table have been added to the analysis.] Lists all data tables in the analysis. Choose a different data table to view the columns in that data table instead. |
2 |
Expand button |
Expands the data panel so you can view tools and details about the selected data table or specific columns. If one column is selected in the Data panel, the expanded panel will show details about that column. If more than one column is selected, you will see an overview of the selected columns, and, if no column is selected, you get an overview of the entire data table. |
3 |
Source structure |
[Only available if the selected data table in Spotfire is built from more than one tables from the original database in a virtual join view.] Lists the underlying source structure from the database so that you can see which columns come from which original database table. |
4 |
Columns |
Lists all columns available in the selected data table. If the columns are shown 'Categorized', then columns of similar type are grouped together. You can change how columns are displayed here
from the data table view of the expanded data panel, by clicking
on the Sort order in data panel button,
|
Cube Data
When you are working with cube data from Microsoft SQL Server Analysis Services, Oracle Essbase or SAP BW you will have the option to show the underlying structure of the cube in the Data panel. This view is called 'Hierarchical' in the column list menu.
Note: Since the hierarchical structure varies with the data source, you may see a different structure than the one below, which shows an example of an Oracle Essbase cube.
Number |
Section |
Description |
1 |
Data table selector |
[Only available if more than one data table have been added to the analysis.] Lists all data tables in the analysis. Choose a different data table to view the columns in that data table instead. (The expanded data panel does not support cube data sources.) |
2 |
Source structure |
If the data is shown with a hierarchical structure, the underlying structure of the data source is shown here. Top levels can be expanded to reveal, for example, the different levels in a hierarchy. See Icon Descriptions for an explanation of the icons in the hierarchical view. It is recommended to use the hierarchical view when working with cube data, to avoid mixing unrelated columns in one visualization. See the section Cubes in Spotfire under Connectors or the separate document Working with Cubes (available under Spotfire Analyst on the TIBCO Product Documentation web portal) for more information about different systems. |
Note: When working with in-db cube data it is not possible to create filters for measures or sets, only for dimension columns. This is because the cube calculates the measures in the context of the selected dimensions.