Now the group is ready to examine each cluster in turn and try to arrive at consensus on a label for the cluster. For this activity, it is useful to provide participants with a printed listing of the statements grouped by cluster along with copies of the point and cluster maps.
The facilitator should direct the process.
Give the participants 5 to 10 minutes to read individually through the sets of statements for each cluster and come up with a name to describe each set of statements.
Ask participants to assign names to clusters individually before doing so as a group; this will minimize any interpersonal pressures in the group.
Have the group work cluster by cluster in an attempt to achieve group consensus on acceptable cluster names. This is often an interesting negotiating task.
The group may experience some difficulty assigning an agreed-upon name for clusters that are not clearly defined.
If the statements in a cluster represent several different ideas, suggest that the group use a hybrid name, perhaps by combining titles from several individuals.
Tell the group, in any event, that these names are tentative and may be revised later.
If consensus on a cluster name is not forthcoming, tell the group that the cluster label can be filled in at a later time.
The facilitator can provide other relevant information that might help the group attain consensus on cluster names.
Provide cluster labels the computer analysis selected (from the sort category labels). Point out that these computer-assigned labels are not necessarily desirable; they are only suggestions.
Introduce the idea of bridging values. Each statement and cluster is assigned a bridging value calculated automatically in the computer analysis. The bridging value is higher for statements or clusters that tend to bridge or span several areas of the map. Statements or clusters with low bridging values indicate more accurately the meaning of those particular points on the map. Encourage the group to place more emphasis on statements or clusters with low bridging values when assigning labels to clusters.
Consider Cluster Labels
At this point, the facilitator might ask the group to write the cluster names that the group arrived at next to the appropriate clusters on their paper version of the cluster map. Or, the facilitator might enter the labels directly on the screen so that all can see them.
Ask participants to examine this cluster map to see whether it makes sense. Remind participants that, in general, just as individual statements that are similar appear close to one another, clusters that are similar conceptually should be closer together on the cluster map than those that are quite different from one another.. Ask participants to assess whether this seems to be true of the completed map.
As with the point map, have participants begin at some point on the map and, thinking of a geographic map, travel across the map, reading each cluster in turn to see whether the visual structure makes sense.
At this point, the group should have at least rough consensus on the cluster labels and should be developing some broader theories about the connections or relationships among clusters. They might see meaningful patterns of clusters as they move clockwise or counterclockwise around the map. They might see opposite or conflicting issues on opposite sides of the map. Or, they might see regions (or clusters of clusters) that represent even more general categories that are meaningful. Thus participants are moving toward interpretation of the cluster map.


