Meeting Youth Where They’re At

How a group of government technologists designed a more accessible summer employment system for Boston youth

Youth Jobs Platform that a Civic Tech Data Collaborative designed to improve access to summer jobs

In 2015, our team of government data and technology experts from MAPC and the City of Boston formed the Boston branch of the Civic Tech Data Collaborative (CTDC) to tackle the design of a more accessible youth employment system. After two years of cross-sector collaboration, we released a final open source Youth Jobs Platform in November of 2017. Ongoing, the City of Boston will update and use this platform for the 2018 job season and beyond.

Why a summer job matters, in the long run

Summer jobs provide immediate benefits to youth: they help develop an understanding of financial management, offer real-world work experience, expose youth to positive adult role models, can help build a resumé or create networks for future employment opportunities. Researchers have also found a positive correlation between having a summer job and achieving much longer term life outcomes that make a huge difference in life-long success: reduced incarceration rates and higher educational achievement. In Boston, Northeastern Labor Economist, Alicia Sasser-Modestino, is conducting a multi-year study to measure such benefits:

“In the long-run, having a summer job can make crime a relatively less attractive option by providing youth with the tools and experience needed to navigate the job market on their own and raising their aspirations for future career options and/or postsecondary education.”

Recently summer employment has been declining for teens (ages 16–19) across the country, which is raising concerns about fewer future pathways to employment for Black and Latino youth living in places with few existing employment opportunities. Youth in Boston have organized around increasing access to the jobs that are available. Last February, over 900 youth convened a rally to increase job opportunities and decrease juvenile incarceration, organized by the Youth Jobs Coalition, an alliance of 40 youth community groups from across the state.

Youth at the Youth Jobs Coalition rally for youth employment, February 2016

In response to these issues, municipalities across the United States have refocused energy around Summer Youth Employment Programs, which provide subsidized jobs to help prepare young people for the workforce and engage local institutions in the hiring process.

Boston’s youth employment program

The City of Boston first began funding summer jobs for youth in 1990, and today works to place over 10,000 youth across the city through non-profit, public sector and private industry partners. Youth work up to 25 hours a week for six weeks in the summer. Boston’s program has been recognized as a national leader for its high share of private sector placements and intensive career-readiness curriculum.

In recent years, the Division of Youth Engagement and Employment (DYEE) in the City of Boston, has struggled with staff capacity to tackle the enormous job. The Mayor’s Summer Jobs Program places youth in 10,000 positions, and about one third (3,200 in 2017) of those are administered directly through DYEE in a program called SuccessLink. Each year, staff conduct direct youth outreach to thousands of applicants each summer (5,000 teens in 2017). At 15 minutes per phone call, outreach staff spent months tracking down young people to offer youth a chance at a summer job.

Youth-centered design

In 2015, the collaborative began a design research process to fully understand the current business practice behind the SuccessLink Program. Through a series of expert interviews and group brainstorms, the team identified critical pain points in the experience youth had while finding, applying for and starting a SuccessLink summer job. This was done primarily by involving the users (youth) at every step of the research and design phase, where teens were asked what was important in finding and accepting a summer job offer. In many cases, this job was their very first introduction to the workforce.

Youth brainstorming barriers to finding a summer job

Youth shared that access to a transit pass, the amount of time it took to travel to the job, and how closely the job matched their own interests or passions were important factors that staff weren’t taking into consideration systematically in 2015. Outreach staff had limited access to information about youth and the job application didn’t inquire about preferences or access to transit.

The collaborative worked with the City to redevelop a new SuccessLink application form to include job industry preferences, transit access information and whether youth found job proximity or alignment to their interests more important. This data is stored in a way that was easily retrievable by City staff to use in outreach. After the new application was rolled out in the summer of 2016, relevant information was also made available to researchers at Northeastern University through an official data sharing agreement, enabling the City of Boston to fully understand long-term benefits of their program and potentially participate in a Pay for Success program in future years.

Algorithmic job match

With the City having access to better data about the needs of youth applicants, the team grappled with another challenge: there were far fewer jobs available than the number of youth who had applied. To provide the best possible jobs to the most youth we adapted a widely used matching algorithm called Gale Shapley to solve the stable marriage problem. A stable marriage or ‘match’ in this case, is the ability to make a repeatable and reliable match between two equally sized sets of data (in this case jobs and youth applicants).

How the Gale–Shapley algorithm works for youth jobs matching

The algorithm generates a score for each applicant for “travel time” based on the time and actual route a young person could take to get from home to a job site, relative to all other job sites. This is created using an open source public transit routing engine called Graphopper. A “job interest score” is also calculated, which reflects if a job matches the interest area youth indicate on their application form. These two scores are merged to form a composite match score, which is used to assign the most youth to jobs they desire and can reach.

By embedding youth feedback into the algorithmic design, the job matching process greatly improved the quality of the offers made to young people, and provided staff with an automated, data-driven hiring process.

Improving the experience for youth, end-to-end

Now that job offers were made with better information, the collaborative wanted to ensure that the user experience teens went through was painless and enjoyable, which began with how we communicated. In 2015, Pew reported that 55% of teens text message as a form of communication with friends every day, almost twice as often as they speak in person or by phone, and nearly six times as more frequently as emailing socially.

In the summer of 2016, the City piloted the matching algorithm using only email-based job offers. We quickly learned that youth don’t typically use email during summer months, which led us to pivot toward a hybrid strategy of sending job offers to youth in text and emails. This simple change resulted in an extraordinary acceptance rate increase without human intervention. 56% more youth accepted their digital offer in 2017 and the overall number of teens hired increased by 20%.

Lessons

After two years of development and iteration, the CTDC team measured results before and after the civic technology intervention. The group found the Youth Job Platform successfully employed more teens and placed them more effectively into summer jobs. Designed with youth, for youth, the project revealed some incredible outcomes:

  1. Increasing City Staff Capacity through Data-driven Program Delivery

In 2017, over 3,000 job offers were made through email and text. In the past, without text reminders and email based offers, staff spent (on average) 15 minutes per offer on the phone. Digital job offers in 2017 saved City staff approximately 95 days or 19 work weeks. By using an iteratively designed, data-driven process to design technology, the collaborative was able to greatly reduce the amount of time City of Boston outreach staff needed to reach youth in the hiring process so that they could spend time on other activities to expand the employment funnel or think creatively about reaching youth who they aren’t reaching currently.

2. Higher completion rates from co-designed technology: 75% of those who accepted an offer digitally also completed onboarding

In past years there had been a dropoff problem: youth either didn’t receive the message that they had a job offer, or they may have accepted a job offer but did not make it through the full onboarding process. Embedding the needs of youth in communications technology and the quality of the offer itself decreased the number of youth who dropped out of the process. 58% of those offered a job, also responded to the offer. 75% of applicants who accepted a job through email/text alone also completed the hiring process in entirety.

3. Agile technology development led to agile public policy development

As the team codified and translated policy decisions into code and began to measure data that hadn’t been collected before, DYEE was given a surprising opportunity to refine and improve on existing policies. Through an iterative technology development process, we were able to test and measure the impact of decisions before they were rolled out to the general population, providing a safe environment to ensure employment policies were equitable and an equal chance at a job was given to every young person in Boston.

4. More meaningful jobs led to more youth being hired

Youth were placed into a job by a matching algorithm that considered personalized job preferences like job type, access to transit and distance from home to job. In past years, only availability of job was taken into consideration. As a result, the rate of job placement increased. In 2016, 30% of applicants completed the entirety of the hiring process. In 2017, after email and text-based offers, 60% of youth completed hiring. In absolute terms, 20% more youth were hired 2017 than in 2016.

Given the great success of a two year pilot, the City of Boston’s Department of Innovation and Technology, currently led by Interim CIO Patricia Boyle-McKenna, a key partner in this work, have absorbed the platform into the city’s digital roadmap. The City of Boston Digital Team will support the tool in future years.

Want to Re-use this technology in your city?

Since launching, municipalities across the globe have expressed interest in launched this type of program in their own city. To reuse code and learn more, please visit the Metropolitan Area Planning Council’s Github repository.

Watch a short video documenting the process

The Youth Jobs Platform was built through a partnership between the Metropolitan Area Planning Council (MAPC, Boston’s Regional Planning Agency), the City of Boston Division of Youth Engagement and Employment (DYEE) and Department Innovation & Technology (DoIT), local volunteer-based civic technologists (Code for Boston), the Boston Foundation, and faculty researchers at MIT.

The Civic Tech and Data Collaborative (CTDC) is a joint initiative of Living Cities, Code for America and the National Neighborhood Indicators Partnership, supported by a grant from the John D. and Catherine T. MacArthur Foundation. Locally in Boston, this project was generously supported by BNY Mellon