Table 3.
Design feature | Level | LCM | K-means | K-median |
---|---|---|---|---|
Sample size | N = 100 | .8133 | .8071 | .8309 |
N = 200 | .8339 | .8193 | .8366 | |
N = 400 | .8480 | .8255 | .8426 | |
Number of clusters | K = 2 | .9290 | .8830 | .9109 |
K = 3 | .8758 | .8668 | .8740 | |
K = 4 | .8390 | .8252 | .8419 | |
K = 5 | .7775 | .7710 | .7937 | |
K = 6 | .7373 | .7406 | .7631 | |
Number of variables | V = 6 | .7146 | .6778 | .6989 |
V = 9 | .8559 | .8566 | .8807 | |
V = 12 | .9247 | .9176 | .9305 | |
Cluster sizes | Equal | .8319 | .8435 | .8441 |
60% | .8332 | .8003 | .8381 | |
10% | .8300 | .8081 | .8280 | |
Error level | 5% | .9347 | .9364 | .9416 |
10% | .8485 | .8336 | .8512 | |
15% | .7119 | .6820 | .7173 | |
| ||||
Overall | .8317 | .8173 | .8367 |