Author Description

This historic book may have numerous typos and missing text. Purchasers can download a free scanned copy of the original book (without typos) from the publisher. Not indexed. Not illustrated. 1979-10 edition. Excerpt: ...their prison work and training experiences and their postrelease jobs to find out which prison experiences were useful later. The Lmw summary accurately reports the findings from Glaser's Table 11.5 (p. 256), which shows the percentage of each group using various categories of prison experience on postrelease jobs. It fails to point out that the two groups differ in other ways than the success-failure dimension, as discussed above. It neglects to include the first three lines of Glaser's table, which show that, compared to the failures, relatively more of the successes had postrelease jobs that required training and relatively fewer were unemployed. The first three lines put the rest of the table in perspective. Unfortunately, none of Glaser's statistical tests can be repeated because of the form in which the data are reported. The Lmw annotation is equally remiss in its failure to point out other differences in the two groups and in its neglect of the first three lines of the table. The Lmw summary then describes the second study. It involved a 10-percent systematic sample of all federal prison releasees in 1956. This is the sample that the Lmw annotation improperly describes as common to both studies. Table 11.6 (p. 257) in Glaser shows the failure rates of these releasees for each of seven categories of the final prison work assignment. If one treats these data as a two-way contingency table, recidivism by work assignment, one finds no reason to assume any relationship between work assignment and failure rate. Glaser's contention that the lowest failure rates were associated with semiskilled work assignments and the highest failure rates were associated with influential jobs is not justified. (The statement is based on only one out of a set...