www.peoplelab.hks.harvard.edu | 1
In collaboration with the California Policy Lab, the Los Angeles Mayor’s Ofce Innovation Team, and
the Los Angeles Economics Workforce and Development Department (EWDD), we conducted a
randomized experiment to test the impact of timely, actionable, and behaviorally informed text
messages aimed at increasing job seekers’ engagement with the city’s workforce development services.
In a 14-week intervention, job seekers received text messages reminding them to engage in job-
search related activities, utilize EWDD WorkSource Center resources, and set goals related to securing
employment. We then measured the effect of the text message campaign on engagement with Los
Angeles WorkSource Centers and subsequent employment outcomes.
INCREASING ENGAGEMENT &
EMPLOYMENT OUTCOMES IN
WORKFORCE DEVELOPMENT
PROGRAMS
Jessica Lasky-Fink, Elizabeth Linos, and Laura Schwartz
April 2024
KEY TAKEAWAYS
1
3
Clients who received
timely and behaviorally
informed text messages
were 3% more likely to
engage with Los Angeles
WorkSource Centers than
those who did not receive
communications.
Clients who received
text messages were 20%
more likely to secure
employment at the end
of the study compared to
those who did not receive
text messages.
A majority of clients who
received text messages
found them helpful and
expressed interest in
continuing to receive
similar messages in
the future.
2
CONTEXT
Unemployment and underemployment are strongly correlated with
poverty and have been shown to negatively affect psychological
well-being.
1
Many cities and states offer free employment services for
residents. However, these resources are only effective if they are used.
Los Angeles EWDD runs 16 WorkSource Centers (WSCs) that provide
free employment services to around 27,000 job seekers a year in
LA County. WorkSource Centers offer specialized services including
career counseling, skills workshops, resume guidance, job matching,
and employment referrals to enrolled clients. However, despite
this robust workforce development infrastructure, the WorkSource
Centers are underutilized by Los Angeles residents looking for
employment. In fact, over half of enrolled clients (56%) only visit their
local WorkSource Center once after enrolling.
Motivating sustained engagement over time is a common challenge
faced by many public sector agencies. While people often intend to
change and maintain their behavior, evidence suggests that doing
so is difcult. For instance, WorkSource Center clients may intend to
apply for a certain number of jobs each week. But, job seekers—like
most people—face numerous cognitive and psychological barriers
that can make it difcult to remember and motivate themselves to
follow through on their intentions. This intention–behavior gap may
contribute to low rates of client engagement with the WorkSource
Centers, as well as limited employment outcomes.
www.peoplelab.hks.harvard.edu | 2
RESEARCH
For 14 weeks from November 2019 to February 2020, we conducted a randomized experiment with 5,537
active WorkSource Center clients. Clients were randomly assigned to a control group that did not receive
communication as part of the study, or one of two treatment groups:
Reminders: Clients received weekly text messages with information about upcoming WSC events
and workshops, and reminders to engage in individual job-search activities like checking in with
case managers and utilizing the WSC computer centers.
Reminders + Plan-making: Clients received the same weekly text message reminders as the
“Reminders” group with additional language encouraging goal-setting and links to online plan-
making forms. These messages leveraged evidence showing that prompting people to set goals
and make plans can increase follow-through and help bridge the intention–behavior gap in areas
like voting and annual u shots.
At the end of the 14-week intervention, we evaluated the impact of receiving communications on
employment outcomes and individuals’ engagement with WorkSource Centers. Engagement was
measured as any interaction with a WorkSource Center including meeting with a case manager or
attending a workshop or recruitment event. Additionally, we measured whether clients were employed
in March 2020—approximately six weeks after the end of the intervention. Employment was captured in
EWDD administrative data.
A second cohort was planned for January–March 2020 but was cut short due to the outbreak of the
Covid-19 pandemic. This cohort is thus excluded from this analysis.
FIGURE 1
Sample text messages
sent to clients in the
Reminders and Reminders +
Plan-making groups
1
2
1 2
Reminders Reminders + Plan-making
www.peoplelab.hks.harvard.edu | 3
WHAT WE FOUND
To evaluate the impact of the text message campaign, we compared outcomes between clients who
were assigned to the no-communication control group and those who were assigned to either treatment
group. We found that clients who were assigned to receive any communication (Reminders or Reminders
+ Plan-making) were 3% more likely to interact with the WorkSource Centers: 45.9% of clients in the two
treatment groups interacted one or more times with the WSCs during the 14-week intervention period
compared to 44.5% of clients in the no-communication control group.
Additionally, clients who were assigned to receive text messages were 20% more likely to be employed
approximately six weeks after the intervention period: 9.4% of clients in the two treatment conditions were
employed according to EWDD administrative data, compared to 7.8% of clients in the control group.
At the end of the intervention, we also conducted a follow-up text message survey among all clients in
the experiment. Of 212 clients who were assigned to one of the treatment groups and responded to the
survey, 75% remembered receiving the text messages. Of these, 80% reported nding the text messages
helpful, and 82% reported being interested in enrolling in a similar program if offered in the future.
Because the second cohort was cut short, we are underpowered to detect effects between the two
treatment groups on any outcomes of interest.
FIGURE 2
Employment at the end of the 14-week intervention.
(total N=5,537)
0.0%
2.0%
4.0%
6.0%
8.0%
10.0%
12.0%
% employed
Treatment group
Control
7.8%
Pooled treatment
9.4%
About The People Lab
The People Lab aims to empower the public sector by producing cutting-
edge research on the people of government and the communities they
serve. Using evidence from public management and insights from
behavioral science, we study, design, and test strategies for solving
urgent public sector challenges in three core areas: strengthening the
government workforce; improving resident–government interactions;
and reimagining the production and use of evidence.
Contact Us
[email protected]d.edu
@HKS_PeopleLab
WHAT’S NEXT
A 14-week program of weekly personalized communications increased client engagement with
WorkSource Centers and improved employment outcomes. Additionally, approximately 80% of
jobseekers who received the communications reported nding them helpful and expressed interest
in enrolling in the program if offered again. These ndings suggest that job seekers may benet from
low-cost communications to prompt engagement with workforce development services and job-
search–related activities. Future research could consider similar information interventions in other
public sector contexts that require sustained behavior and engagement over time.
SOURCE
1. Theodossiou, I. (1988). The effects of low-pay and unemployment on psychological well-being: A logistic regression
approach, Journal of Health Economics 17(1), 85-104. https://doi.org/10.1016/s0167-6296(97)00018-0