The Wiki page on Grapheme–color synesthesia says:
... one recent study has documented a case of synesthesia in which synesthetic associations could be traced back to colored refrigerator magnets.
Wiki points out that this might be an individual case, but let's assume that synesthesia is something you can train. Let's further assume that everybody is more or (most of us) less a synesthesist.
Can the level of synesthesia be measured by the following test:
You are shown a sequence of colored characters, i.e. number and letters, or the characters shown are placed on a colored background.
After that you are asked to give the sequence.
Should there be a measurable statistical effect in remembering the sequences when they are colored in your specific color scheme?