In contrast to new subscriptions, cancellation recommendations occurred in all 33 curriculum cores. For purposes of simplicity, comparisons will be made of cancellations only to overall new subscriptions, and not to either high faculty score or mid-faculty score new subscriptions. The cancellation measures followed the usual highly skewed pattern. In terms of actual numbers, the top five cores in cancellation titles accounted for 130 (38.0%) of the 342 titles recommended for such treatment; the top five cores in cancellation faculty score accounted for 628 (47.1%) of the 1,334 faculty score points attributable to the cancellation titles; and the top five cores in cancellation costs accounted for $122,548 (55.1%) of the $222,409 in budgetary savings that would result from the cancellations.

Two cores—Biology and Mathematics—appeared among the five dominant cores on all three cancellation measures, and these cores were also among the five dominant cores on all three measures of total core size. This phenomenon was similar to the one that had occurred with the five dominant cores in overall new subscription measures, where the overlap had also taken place on three of the larger cores. Therefore it was not surprising to find that the cancellation measures in actual numbers correlated significantly with all three total core size measures, with Spearman rank-order coefficients that ranged from 0.40 to 0.92. However, this result was different from that of the analysis of the relationship of the overall new subscription measures in actual numbers to core size measures, where total core faculty score did not correlate significantly with overall new subscription titles and costs.

There were major similarities and one telling difference between cancellation and overall new subscription measures as percentages of their respective core measures. The similarities were the highly skewed distributions and the overlap at the low end of core size measures. Concerning the percentage of cancellation titles of core titles, the top five cores ranged from 34.1% to 53.8% with a mean of 41.8%, compared to a mean of 22.1% for all 33 cores. As for percentage of faculty score of core faculty score, the top five cores ranged from 5.6% to 12.5% with a mean of 8.4%, compared to a mean of 3.3% for all 33 cores. With respect to percentage of cancellation cost to core cost, the top five cores ranged from 46.1% to 80.5% with a mean 55.6%, compared to a mean of 26.6% for all 33 cores. Here the overlap occurred only on Biological and Agricultural Engineering, which was among the smallest cores on all three core size measures. This overlap resembled the one among the five dominant cores on overall new subscription measures as percentages of their respective core measures. The latter overlap had also taken place at the lower end of core size measures, although on three small cores instead of one.

Of great importance was the telling difference between overall new subscription and cancellation measures as percentages of their respective core measures in their relationship to core size measures. All three overall new subscription measures in these terms had significant negative correlations with total core faculty score only, demonstrating that the ST serials holdings of LSU Libraries tended to be damaged where faculty interest and political power were weak. However, there were no significant correlations of the cancellation measures in these terms with any core size measures, making it appear that the cancellations were independent in this respect.

The crucial difference between the cancellations and the overall new subscriptions recommended as a result of the exercise revealed itself in the matter of leverage, i.e., the ratio of percentage of ST value to the percentage of cost. Thus, the overall new subscriptions total leverage of a positive 1.46 to 1 was 11.2 times greater than the total cancellation leverage of a negative 0.13 to 1. The story was the same with average leverage. In this case, the overall new subscription average leverage of a positive 1.2 to 1 was 10.0 times more than the cancellation average leverage of a negative 0.12 to 1. On a core-by-core basis, cancellation leverage ranged from 0.0 to 1 in Human Nutrition and Food as well as Industrial and Manufacturing Systems Engineering, where only zero-class faculty score titles were recommended for cancellation, to 0.26 to 1 in Food Science. The mean core-by-core cancellation leverage was a negative 0.12 to 1, and the overall new subscription mean core-by-core leverage of 3.1 to 1 was 25.8 times higher than this.