Absence of conflict of interest.
Citation
Highlights
- The study’s objective was to examine the implementation of the Employment Retention and Advancement Project which looked to promote employment stability and wage progression among low-income workers through preemployment job search, team-based case management, a stipend for welfare leavers, and intensive post-employment services at three sites in Texas.
- The study authors conducted an implementation evaluation using unemployment insurance, TANF, and food stamp records data, a 12-month survey, field research, data on stipend receipt, a time study of program staff, telephone discussions, and baseline demographic data.
- The study found that all three sites put significant effort into marketing the stipend with an overall low level of receipt among program groups. Of the three sites studied, Corpus Christi had the smoothest implementation and other sites adopted some of its strategies.
- The companion impact studies were reviewed by CLEAR in December 2015 and April 2016.
Intervention Examined
The Employment Retention and Advancement Project (ERA)
Features of the Intervention
- Type of organization: Texas Department of Human Services & Texas Workforce Commission
- Location/setting: Multi-site in Texas
- Population served and scale: Low Income Adults; 2,137 participants
- Industry focus: Social Services
- Intervention activities: Preemployment job search, case management, stipend for welfare leavers, and post-employment services
- Organizational Partnerships: Domestic violence and substance abuse organizations
- Cost: Not included
- Fidelity: Not included
The ERA program was developed in 1999 in response to a growing concern of the rate at which welfare cases “recycle”. “Recycling” is when an individual leaves welfare and ends up returning for services due to low wages and or poor benefits. The project was funded by the Administration for Children and Families within the Department of Health and Human Services with supplemental funding from the Department of Labor. The ERA program was implemented in 15 locations in 8 states with varying methods and goals.
Features of the Study
The ERA implementation study focused on three sites in Texas: Corpus Christi, Fort Worth, and Houston. The three sites offered significant differences in local economy and demographics which helped demonstrate the feasibility of the program as well as different challenges, successes, and solutions in different conditions. The vast majority of the ERA group participants were female, U.S. citizens who were proficient in English, had a high school diploma or less, were unemployed at baseline, had children, and were currently applying for cash assistance. The racial / ethnic composition of the groups varied by site; in Corpus Christi the majority of the participants were Hispanic and in Houston and Fort Worth the majority of participants were non-Hispanic Black. On average the participants were about 28 years old. The authors conducted an implementation analysis which included both quantitative and qualitative data analysis.
Findings
Intervention activities/services:
- The study found that in Corpus Christi and Fort Worth, there was a statistically significant increase in the number of contacts made with case managers among those in the intervention group.
- The study found that in Houston there was no impact on the amount of contact which may reflect strong case management in their Choices program (the control group).
- The study found that the programs in Corpus Christi and Fort Worth produced statistically significant increases in proportion of clients receiving job retention and advancement services.
- The study found that Corpus Christi had the highest rate of stipend receipt, with approximately 30% of the intervention group receiving it.
Fidelity:
-
The study did not include any fidelity measures or findings.
Implementation challenges and solutions:
- Those who choose to not receive the stipend, despite being eligible had varied reasons for non-receipt which included having the desire to discontinue involvement with a government program, lack of knowledge or understanding of the requirements associated with stipend receipt, or lack of attendance at required activities.
- Staff reported that job loss was more pervasive among program participants than expected and with that staff realized that they had to strengthen reemployment services to respond to that.
- The authors found that it was essential to have sufficient marketing for the stipend and noted that it may have had greater uptake if it was promoted/talked about on its own and not in conjunction with other program components / requirements.
- The site in Corpus Christi developed several strategies that were later implemented by the other sites such as stipend marketing strategies, designated case managers for stipend recipients and post-employment services. Some stipend related improvements included displaying posters and fliers throughout the offices, as well as posters and videos highlighting participant success stories, started referring to the stipend as an employment bonus and strong staff encouragement efforts. The post-employment services included regular employer site visits, the creation of specific metrics for staff, and support groups to meet stipend participation requirements.
Cost/ROI:
-
The study did not include any measures or findings related to cost or return on investment.
Considerations for Interpreting the Findings
During the period of the evaluation, a waiver which allowed individuals in Texas to participate in a wider range of employment and training activities to reach federal work participation requirements ended which meant that they now had to follow federal requirements. This placed greater limitations as to what counted towards meeting federal participation requirements and therefore resulted in a greater emphasis on immediate employment among participants in both the implementation and control groups.
The authors note that case manager contact may be undercounted among both the intervention and control groups as it was recorded via survey responses which were collected a year after random assignment and respondents may have forgotten contact that was earlier in the year and or not very intensive.