Steve Aos, Elizabeth Drake - August 2010

Can knowledge about “what works” to reduce crime be used to help states achieve a win-win outcome of lower crime and lower taxpayer spending?

The Washington State Institute for Public Policy has constructed an analytical tool for the Washington legislature to help identify evidence-based sentencing and programming policy options to reduce crime and taxpayer criminal justice costs. With additional financial assistance from the MacArthur Foundation, The Pew Charitable Trusts contracted with WSIPP to: (1) develop the tool, (2) apply it to a policy process currently underway in Washington State, and (3) help Pew make the tool available to other interested states.

This report describes the tool (as of August 2010) in detail and illustrates its use by applying it to two hypothetical sentencing policy options in Washington State. The tool assesses benefits, costs, and risks. Results from the two hypothetical examples point to possible win-win policy combinations.

Mason Burley, Jim Mayfield - August 2010

The Washington State Institute for Public Policy was directed by the 2001 Washington State Legislature to “conduct a longitudinal study of long-term [mental health] client outcomes to assess any changes in client status at two, five, and ten years.” This latest report discusses supported employment and housing outcomes for public mental health consumers in Washington State.

For this analysis, we selected all adults who received public mental health services in 2006. We then analyzed all service episodes for these individuals between 2002 and 2007. Based on these records, we examined how employment and housing outcomes related to treatment patterns. According to this analysis, and a review of national literature:

• Employment history, diagnosis, and functioning all predict likelihood of employment after treatment.
• Earnings levels for employed adults remain low; eligibility guidelines for benefit programs (such as SSI) may also constrain employment decisions.
• Supported employment and permanent supportive housing programs can improve outcomes for consumers, if implemented according to recommended guidelines.

Annie Pennucci - May 2010

General Assistance-Unemployable (GA-U) is a Washington State-funded program that provides cash grants and medical benefits to adults who have temporary incapacities that prevent them from working. The GA-U caseload has increased from about 10,000 individuals in 1998 to over 16,000 in 2009. As of April 2009, the caseload was forecast to continue growing to over 20,000 people by 2013. Since about 2003, the GA-U caseload has risen faster than the state population and appears to correlate with growth in the number of adults in poverty and without health insurance.

In 2010, the GA-U was renamed the “Disability Lifeline Program.” The Disability Lifeline Program is similar to GA-U in most eligibility rules and benefits, except that the Disability Lifeline Program has a time limit during which individuals can receive benefits (up to 24 months within a five-year period). In the 2010 legislation that created the Disability Lifeline Program, the Institute was directed to “analyze the experiences of persons who have been terminated from disability lifeline benefits” due to time limits or other reasons. The Institute will track the number of former benefit recipients who transitioned to SSI benefits or became employed, in addition to other outcomes. This report is due in December, 2012.

Steve Aos, Elizabeth Drake - April 2010

Can knowledge about “what works” to reduce crime be used to help a state achieve a win-win outcome of: (1) lower crime, and (2) lower taxpayer spending? This progress report describes the work underway by the Washington State Institute for Public Policy (WSIPP) to develop an analytical tool for Washington, and perhaps other states, to identify evidence-based policy options to reduce crime rates and lower the taxpayer costs of the criminal justice system.

The Pew Charitable Trusts contracted with WSIPP to: (1) develop the tool, (2) apply it to the policy process currently underway in Washington State, and (3) help Pew make the tool available to other states.

This progress report describes the structure of the tool being constructed. The current plan calls for initial estimates by August 2010. In addition, the tool will be used to support the work of the legislatively directed study being conducted by the Washington State Sentencing Guidelines Commission.

Annie Pennucci - April 2010

Washington State’s Education Advocacy Program (EAP) helps over 2,000 foster youth annually to resolve problems in school. EAP coordinators provide direct advocacy, working with school staff to address issues such as youth academic performance. The coordinators also consult with foster youth, caregivers, and social workers to help youth and their caregivers advocate for themselves in the K—12 system. The EAP also provides information and referrals to local resources.

The EAP is structured around four goals: 1) improved access to services, 2) youth stay enrolled in school and improve attendance, 3) youth maintain academic progress, and 4) the reduction of school disciplinary actions. Of EAP youth who received direct advocacy or consultation services in 2007—08 and 2008—09, over half sought to improve access to services—often special education—and over a quarter worked to maintain school enrollment or academic progress. Approximately one-fifth of EAP youth addressed more than one goal.

This report describes factors—youth characteristics and foster care placement history—that influence which goals youth address in the EAP. Using statistical analyses, we found that boys were more likely than girls to address school discipline, and that the more time youth spent in foster care, the more likely they were to seek help to improve access to services. We also identified regional differences in which goals EAP youth addressed. We do not know whether regional differences were due to variation in youth issues, local practices, or both.

Mason Burley - April 2010

This report examines results from the Washington Assessment of Student Learning (WASL) for foster students completing this test in 2005 and 2008. For this study, we follow two cohorts of foster youth and analyze changes in assessment scores between 4th and 7th grades and 7th and 10th grades.

Overall, the “met standard” rates for foster youth completing the WASL are between 15 and 30 percentage points lower than for other students. Foster youth, however, have much higher rates of grade retention, school mobility, and other factors associated with poor test scores. As this analysis shows, a student’s previous test scores remain the strongest predictor of future assessment results. Between 69 and 77 percent of foster youth who took the math WASL in 2005 and 2008 failed to meet standards in both years (40 percent did not meet reading standards for both tests).

About half of foster youth who completed the WASL in 2008 were not in foster care three years earlier. Youth in care typically come into the child welfare system with educational deficits. Across all students, foster youth scored in the 29th and 23rd percentile on the reading and math WASL, respectively. Given the size of these gaps, even successful interventions are unlikely to bring educational outcomes to the student average (50th percentile). We found that for younger students, the number of months in foster care was associated with modest gains in WASL scores. The greatest improvements in WASL scores occurred for students in foster placements lasting between 20 and 22 months.

Marna Miller, Tali Klima, Corey Nunlist - February 2010

In 2008, the Legislature directed the Institute to study the truancy provisions of the 1995 "Becca Bill." The bill changed several aspects of the compulsory school attendance laws in Washington. In particular, the bill requires that school districts file truancy petitions in juvenile court when students accumulate a specified number of unexcused absences.

From a policy standpoint, it is of interest to know whether the Becca truancy laws have had a causal impact on key student outcomes, such as graduation rates.

Unfortunately, despite our best attempt to analyze this question with rigorous statistical methods, we cannot provide a scientific answer as to whether the law is having a positive, negative, or no effect on student outcomes. Sometimes research can provide answers to central questions, and sometimes it cannot; this is a case of the latter.

The 1995 Becca laws were implemented statewide and a random assignment study—the type of study offering the best scientific evidence—was never possible or envisioned. In addition, the historical data available for our study do not allow us to measure a vital aspect of the Becca laws: the number of unexcused absences from school. Without this information, it is impossible to employ appropriate statistical methods to study the question of the Becca Bill’s effectiveness.

We do know from our analysis that students who receive truancy petitions are at very high risk of academic failure as well as future criminal involvement. For example, of all students in Washington during the 2002–03 school year, only 20 percent of students with petitions graduated from high school by 2008, compared with 77 percent of their non-petitioned peers. Similarly, 20 percent of the students with petitions were subsequently convicted of a crime in Washington compared to 4 percent of students without petitions.

Our inability to analyze the effectiveness of the Becca Bill stems from that fact that, even before their truancy filing, petitioned students were already at much higher risk for negative outcomes. For example, prior to receiving the truancy filing, petitioned youth had a GPA of just 1.3, compared with a GPA of 2.7 for non-petitioned students. And, prior to the filing, 31 percent of the petitioned students had previously been convicted of a crime compared with 8 percent of non petitioned students.

Given these differences, and without additional information, it is not possible to identify an appropriate comparison group with which to judge—in a scientifically rigorous way—whether the Becca Bill has had the effects the Legislature intended.

Elizabeth Drake, Steve Aos, Robert Barnoski - January 2010

The 1999 Offender Accountability Act (OAA) directs the Department of Corrections (DOC) to perform a formal assessment of each offender’s risk for recidivism and then to allocate agency resources accordingly. The law also requires the Institute to evaluate the OAA and provide results by 2010.

This report presents our findings on whether the OAA has had an effect on recidivism. On average, offenders today have a greater risk for recidivism than historically; the general rise in recidivism over the last 20 years is largely explained by the increased underlying risk of DOC’s offender population. Since the OAA was implemented, however, something favorable has happened to cause recidivism rates to be lower than expected. Unfortunately, our statistical analysis does not allow us to identify whether this beneficial change can be attributed specifically to the OAA or other policies, or other unknown factors that occurred during the same time period. Regardless, the good news from our evaluation is that, after at least a decade of increasing recidivism, Washington is now beginning to observe improvements in adult felony recidivism.

Shawn Whiteman, Roxanne Lieb, Mason Burley - January 2010

The 2009 Legislature directed the Institute to “evaluate the adequacy of and access to financial aid and independent living programs for youth in foster care. The examination shall include opportunities to improve efficiencies within these programs.” In the past decade, the number of programs focused on Washington foster youth transitioning to adulthood has grown from three to 15. We estimate that 3,365 youth accessed one or more of these programs in 2009—roughly 60 percent of those eligible for the state’s Independent Living program (for foster youth ages 15 to 21). In this report, we review the research evidence on Independent Living programs.

In Fiscal Year 2009, approximately $3,300 per youth was spent on foster youth transition programs. Over $11 million in total was spent on these programs; the state spent close to $5 million. Almost half the state funding went toward housing programs, over a quarter toward college preparation and student financial aid, nearly a fifth toward health insurance, and 5 percent toward helping youth finish high school.

In Washington, 34 percent of students in foster care graduate on-time from high school, compared with 71 percent of their non-foster peers. We recommend the legislature consider reallocating funding to help more foster youth finish high school. The 2009 legislation (HB 2106) directing performance contracts for child welfare services offers an opportunity to consolidate foster youth transition services into the smallest number of contracts and emphasize key outcome measures.

Mason Burley, Stephanie Lee - January 2010

In 2006, the Washington State Legislature passed 2SHB 2002, which allowed up to 50 youth (per year) to remain in a foster care placement until they reached age 21. In order to be eligible for an extended foster care placement, youth in the Foster Care to 21 program must be enrolled in a post-high school academic or vocational program.

The 2006 Legislature also directed the Institute to “conduct a study measuring outcomes for foster youth who have received continued support,” and “include measurements of any savings to the state and local government.” This evaluation compares outcomes for Foster Care to 21 participants with a matched group of foster youth who completed high school before the program was available. Based on this analysis, we found that youth enrolled in Foster Care to 21 attended college for a longer period in the first two years after high school graduation, received food stamps for fewer total months, and were less likely to be arrested for a misdemeanor or felony crime. Positive outcomes associated with Foster Care to 21 saved taxpayers $1.35 for every dollar spent on the program, according to our estimates.