WtW programs will make full use of their resources only if they succeed, in collaboration with local partners, in generating a substantial flow of entrants to their programs. To some extent, WtW participants may be referred by TANF and other agencies. Evaluation site visits strongly suggest, however, that WtW programs also need to reach out energetically to those agencies, to encourage referrals, and to conduct their own outreach and recruitment.
The second grantee survey generally confirms the findings of the first survey on how grantees are going about recruiting participants (Table III.6). Overall, grantees are still most likely by far to name the TANF agency as their principal source of WtW referrals or recruits. Other means of recruiting, however, are also common. Importantly, two shifts in grantees' reported recruitment sources provide evidence to support observations from the exploratory site visits.
CHANGES IN WTW RECRUITMENT STRATEGIES: FIRST AND SECOND GRANTEE SURVEYS
|Grantees Planning to Recruit or Recruiting from These Sources
(Percentage of Grantees)a
|Estimated Share of WtW Participants to Be Recruited from Each Source
|First Survey||Second Survey||First Survey||Second Survey|
|Child Support Enforcement||54.5||41.9***||4.5||3.3|
|Grantee's Community Outreach||48.6||53.5||6.5||11.9**|
|Other Organizations' Community Outreach||34.4||33.8||6.1||6.3|
|Grantee's Existing Caseload||37.2||27.8***||6.3||4.9|
|Source: National Evaluation of the Welfare-to-Work Grants Program, First Grantee Survey (November 1998 - February 1999) and Second Grantee Survey (November 1999 - February 2000).
a Based on responses of grantees that identified their recruiting sources. They represent 85 percent of all responding grantees in the first survey and 99 percent of respondents to the second survey.
b Estimates based on the projected share of total participants grantees expect to recruit using each method weighted by total expected participation. Percentages may not sum to 100 due to data rounding.
* Differences between first and second grantee survey results are significantly different from zero at the .10 level, two-tailed test.
** Differences between first and second grantee survey results are significantly different from zero at the .05 level, two-tailed test.
*** Differences between first and second grantee survey results are significantly different from zero at the .01 level, two-tailed test.