Factor 51 (The maturity of the organization's system capabilities for network defense), an example unimodal distribution, has a high level of agreement with most ratings clustered at the high level, indicating that this factor is very important to most participants.
Clusters 1, 2 and 3 show children with linear growth in ratings, clusters 4, 5, and 6 identified subgroups with delay and a power function, clusters 6, 7 and 8 represent groups with little increase.
Table 1 provides more detailed information concerning individual items contributing to the ratings among two highly-rated clusters ('Addiction Risks', 'Treatment Problems'). Figure 3 is a cluster rating map depicting participants' perspectives concerning the factors' relative amenability to intervention (to optimize NA utilization).
For rating-based cluster target selection, success rates range up to (83%), indicating a good support for evolving information needs.
In the latter case, quality ratings were clustered on a smaller portion of the scale.
In each case, statistically significant differences among the mean ratings across clusters were observed.
Accounting for teacher ratings and clustering effects, moderate associations were found between GOLD® scale scores and a direct assessment measure.
Nine of the ten criteria (i.e., those that were dichotomous) had ratings that clustered around the same score (0, 1, or 3).
For the rating-based clusters, success rates are higher and range up to (83%).
For rating-based clustering, we apply k-means based on the rating vectors for each item and select (k = |I| / 3).
