WtW enrollees in all but the two JHU program sites were employed for an average of only between one-third and one-half of the follow-up year (Exhibit IV.3). Because the JHU program was designed to serve persons who already had jobs, it is not surprising that enrollees in the Baltimore County and St. Lucie County sites were employed for the largest proportions of the year nearly three-quarters, on average. The much lower proportions for the other nine sites reflect the interplay among three intermediate employment outcomes: (1) the fraction of enrollees who were never employed during the year,(37) (2) the average duration until the first job for those who became employed, and (3) the rate of job loss by enrollees who were employed at enrollment or who subsequently became employed. These intermediate outcomes may be influenced by WtW service strategies, as follows:
A strategy of rapid job entry would be expected to minimize the duration until the first job (intermediate outcome #2) and perhaps also reduce the fraction of enrollees who are never employed during the year following program entry (intermediate outcome #1). However, this strategy might carry an elevated risk that enrollees will not be strongly attached to the jobs that they obtain, resulting in an elevated rate of job loss (intermediate outcome #3).(38)
A strategy of extensive pre-employment services might result in better matches of enrollees with jobs, thus reducing the risk of job loss (intermediate outcome #3).(39),(40) However, the extensive pre-employment services would likely extend the time until the first job (intermediate outcome #1) and might also increase the risk that an enrollee would remain without work for the entire year following program entry (intermediate outcome #2).
The Yakima and West Virginia study sites illustrate the trade-offs between these two service strategies with respect to employment outcomes. Yakima pursued a rapid entry strategy; only half of the WtW enrollees in that site received job readiness training, with an average duration of just 9 days. In contrast, West Virginia pursued an extensive services strategy; three-fourths of its enrollees received job readiness training for an average duration of 24 days. The enrollees in these two sites were equally likely to be employed sometime during the year following program entry (80 percent, Exhibit IV.1). However, the mean duration until the first job was longer for enrollees in West Virginia than those in Yakima: 5.1 months versus 4.3 months (Exhibit IV.2). The incidence of job loss was 58 percent in West Virginia, versus 71 percent in Yakima (Exhibit IV.8). Despite the substantial differences in two of the three intermediate employment outcomes, the mean percentage of time enrollees had a job during the year following program entry was similar in the two sites 49 percent in Yakima and 44 percent in West Virginia.