The City of Los Angeles has 16 WorkSource Centers that serve approximately 30,000 residents per year with the ultimate goal of connecting them to meaningful employment and occupational training. The WorkSource Centers achieve this by providing clients free employment services such as career guidance, resume and interview skill building, and employment referrals.Of WorkSource Center clients, around 80% are low income, 80% are people of color, and 40% experience other barriers to employment such as being homeless, justice involved, or substance abuse issues.This randomized experiment evaluates the effect of sending WorkSource Center clients regular communications via text message on client engagement, persistence, and employment and training outcomes.
Experimental design
All current and active clients will be randomized into one of three conditions: 1. Control: Clients randomized to the control condition will not receive any communications as part of this study. They will continue to receive any routine WSC communications that are part of the current status quo. 2. Reminders: Clients randomized to the Reminders condition will receive weekly text message reminders about upcoming WSC events, workshops, and activities to further their job search. 3. Implementation Intentions: Clients randomized to the Implementation Intentions condition will receive weekly text message reminders about upcoming WSC events, workshops, and activities to further their job search, as well as messages encouraging goal-setting and plan-making as part of their job search.
Clients randomized to the Reminders or Implementation Intentions condition will receive weekly messages for 10 weeks.
Design details
Randomization method
Randomization done by computer in stata.
Randomization unit
Randomization will be clustered at the household level. Forty-percent of the sample will be randomized to the control condition; 30% will be randomized to each treatment condition. The randomization will be stratified by WSC and enrollment date, defined as enrollment before/after July 1, 2019, which is the start of LA’s fiscal year.
Randomization will be conducted in two waves:
Cohort 1: At the start of the intervention, all active clients who enrolled in one of Los Angeles’ 16 WSCs between Jan. 1, 2019 and Oct. 31, 2019, and who have at least one recorded visit or engagement with a WSC since May 1, 2019, will be randomized into one of the three conditions outlined above.
Cohort 2: Every four weeks from intervention start, all newly enrolled clients at any one of Los Angeles’ 16 WSCs will be randomized into one of the three conditions outlined above.
Clusters
Cohort 1 will have approximately 5,400 households (clusters). Will will continue adding newly enrolled clients once per month as feasible until we reach the end of the intervention period or the maximum target sample size of ~16,000 clients or ~15,500 households.
Observations
The initial sample size for randomization cohort 1 will be around 5,600 clients. We will continue adding newly enrolled clients once per month as feasible until we reach the end of the intervention period or the maximum target sample size of ~16,000 clients.
Arms
40% of the sample will be randomized to the control condition; 30% will be randomized to each treatment condition. The randomization will be stratified by WSC and enrollment date, defined as enrollment before/after July 1, 2019, which is the start of LA’s fiscal year.
Primary outcomes
1. Workshop/WSC attendance. Data collected via sign-in sheets at WSC events.
2. WSC engagement, as measured by the number of visits to a WSC. Data are recorded by case managers in the WorkSource Centers' administrative database.
3. WSC "touch-points," as measured by clients' self-reported number of conversations or other interactions with their WSC. Data collected via text surveys.
4. Clients' number of outside job-search activities, as measured by clients' self-reported frequency of engagement in non-WSC job-search activities. Data collected via text surveys.
5. Employment outcomes. Data collected by WorkSource Centers and stored in the WSC administrative database.
Cited by
No citation data available for this registration yet.