Conditional formats calculate and visualise information in your map.
The buttons in this section apply conditional formats to the map after it has (optionally) been transformed in the analysis section. Each filter can only be applied once, so when you click an inactive button in the bottom row it becomes active and moves to the top row, and when you delete it from the top row it appears again in the bottom row.
You can now conditionally colour the text of your factor labels. For example, this can be useful to draw attention to factors which are opposites:
There is a dedicated shortcut button just for this:
You can label your factors, as well as your links. Labelling your factors is a useful way of adding detail and clarity to your maps. Click on the label factors buttons to open up the filter panel. The app will then ask you to choose which label(s) to add to your l factors. Options include frequency, size, and zoom level.
Normally you will want to add this additional information to the existing factor label, but if you want you can check the box to
Clear previous labels?
You can now also choose whether or not to include the name of the field in your factor labels.
- if the field is numerical, the values of that field will be assigned to a colour gradient with the given low, medium and high colours. If you specify white or grey as the low point, the mid point will be ignored.
- otherwise, if the field is not numerical, the values of that field will be assigned random colours up to a maximum of eight.
Sometimes you might want to set a specific colour for all your factors, or links, or factor borders. This is a bit tricky to set using the buttons, but you can specify a colour easily using the advanced editor:
color factors fixed=pink
color borders fixed=gray
color links fixed=coral
You can use any html colour like red, pink, aliceblue, beige … and also hex colours like #eee, #00000033 etc.
The default palette for discrete fields like gender or village is nice, but maybe you want something different. With continuous fields like source count, you can specify the high, low and mid colours; now you can change the palette for discrete fields too, by adding
pal=3 like this:
color links field = #area fun=literal pal=6
Palette 2 is a good choice if you don’t like the paler colours in palette 1 (the default).See also the special sections on conditional formatting with percentages and conditional formatting with surprise.
Also, much of this formatting only makes sense in terms of bundles of links from one factor to another. The app will bundle the links for you in these cases.
The app adds some pre-calculated fields for you and adds them to the tables.
You can use these in various ways to construct filters. For example you can add
frequency to your factor labels, or filter the map to show only bundles of links consisting of only one link (
find links field=source_frequency value=2 operator=less).
Some examples of calculated fields:
- betweenness: the number of paths going through the factor. Betweenness is a measure of how central a factor is in the map, how many paths go through it. Betweenness is similar to the frequency, but a factor with lots of links on the edge of the map will have a high frequency but won’t have a high betweenness.
- betweenness_rank: the rank of the betweenness.
- in_degree: the number of incoming links.
- out_degree: the number of outgoing links.
- role: the number of incoming links minus the number of outgoing links. High values are drivers, low values are outcomes
- frequency: the number of links.
- driver_score: how strongly is this factor a driver? (
- outcome_score: how strongly is this factor an outcome?
- driver_rank: rank of driver_score.
- outcome_rank: rank of outcome_score.
- is_opposable: does the factor label contain a ~.
- zoom_level: number of ; separators in factor label, plus 1.
- top_level_label: the label of the factor’s ultimate parent in the hierarchy, if any.
- top_level_frequency: the number of links to and from the top level factor.
- is_in_hierarchy: whether the factor is part of any hierarchy or is just a “singleton” factor.
You can use
is_in_hierarchy to focus only on your hierarchical structure, eg you can add a filter
find factors field=is_in_hierarchy value=true operator=equals
to exclude singletons.
When you combine opposites, the app gives you additional combined fields (in the links and factors tables) which you can use to display and filter your maps in various ways. For example, you can keep only links in which come from or to positively formulated factors:
find links field=flipped_bundle value=FALSE|FALSE operator=equals .
There are many more.
These fields are calculated at the point in the chain of filters where each filter is applied, so if you, say, filter for bundles with at least 10 links and then add filters which filter out, say, all the older people, you might get bundles with fewer links remaining.