Special Examination Report No. 15-15

December 2015

Why we did this review
This special examination of Communities In Schools (CIS) was conducted at the request of the House Appropriations Committee. The committee asked us to provide information on which school systems are participating in CIS, the services being provided through the initiative, and how state grant funds appropriated for CIS are being used. In addition, the Committee asked us to determine the state's return on its investment in CIS, as well as compare graduation rates of CIS participants to rates of non-participants and statewide rates.
About CIS
CIS is a national initiative established in the 1970s that promotes academic achievement and educational attainment of at-risk students by coordinating a system of supports to address academic and non-academic barriers to learning. During the school year 2015, 25 states and the District of Columbia had CIS networks in place.
To promote the delivery of needed services to students in the state, the General Assembly has appropriated funding to Communities In Schools of Georgia (CISGa) since 1999. In school year 2015, CISGa was awarded $1.03 million in state grant funds. During the year, CISGa and a network of 32 local affiliates managed CIS programs in 41 school systems containing 207 sites. Approximately 120,000 students (including 10,580 who were identified as at-risk) received services during the year.

Communities In Schools
Requested information on CIS activities
and effectiveness
What we found CIS has been implemented in one-fifth of the state's school systems (41 of 201 total systems) to improve academic and non-academic outcomes of students deemed to be at risk of dropping out of school. CIS programs and services were funded in part by a state grant to CISGa totaling $1.03 million in fiscal year 2015, which is only a small percentage of the total cost of operating CIS programs and services.
Based on current funding levels, the state's return on investment is estimated to be $18:$1 for a high school graduate over his or her lifetime. For the state to breakeven on its investment in CIS in fiscal year 2015, only 138 CIS students would have needed to graduate that year, the equivalent to increasing the 2014-15 statewide graduation rate by 0.2 percentage points.
While we were unable to establish a statistical correlation between CIS and various outcomes due to data limitations, we assessed CIS's relative effectiveness in Georgia, by comparing (1) graduation rates of a sample of CIS schools to the statewide average and a sample of similar non-CIS schools and (2) CIS schools' graduation rates before and after the implementation of CIS. We found that CIS's average graduation rates were higher than statewide rates and average rates for similar non-CIS schools. We also found that average graduation rates in CIS schools improved one and two years after CIS's implementation.
We also identified additional independent, third-party research studies on CIS to better understand CIS's impact on academic outcomes. We focused on research that had been identified by Child Trends--a nonprofit, nonpartisan research center--as meeting high methodological standards. Child Trends' assessment

270 Washington Street, SW, Suite 1-156

Atlanta, Georgia 30334

Phone: (404)656-2180

www.audits.ga.gov

of the research studies found that CIS programs and services have a positive effect on academic outcomes such as student progress (dropout rate and credit completion), school attendance, and math achievement. In addition, a national evaluation of CIS (also included as part of the Child Trends review) concluded that CIS is linked with slight increases in graduation rates, particularly among CIS schools that use the full CIS model and use their resources to serve more students with fewer services.
In addition, CISGa's survey of principals found CIS to have a positive influence on improving outcomes. Principals credited the various programs and services delivered through CIS as helping to improve attendance, student discipline, academic performance, and progress toward graduation. Currently, CIS services fall into several categories: case management, basic needs and resources, academic assistance, life skills/social development, family engagement/strengthening, physical health, mental health, behavior interventions, college/career preparation, community service/service learning, and enrichment/motivation. These services, targeted to meet the needs of students, parents/families, and schools, are specifically aimed at promoting positive outcomes in such areas as dropout and graduation, academic achievement, attendance, and behavior.
What we recommend This report is intended primarily to provide answers to questions posed by the House Appropriations Committee. While we were unable to establish a statistical correlation due to data limitations, our analyses showed improvement in schools where CIS had been implemented. Should the General Assembly be interested in future evaluations of the effectiveness of CIS, the Georgia Department of Education (GaDOE) should consider adding student participation in CIS as an identifier (to be included as part of the detail elements of the Student Level Record) in its annual data collection process. This would enable GaDOE and state auditors to link CIS programs and services to individual student-level outcomes.
CISGa's Response: In its response, CISGa indicated its general agreement with the findings presented and provided technical comments. CISGa also noted throughout its response that "due to local interpretation of federal privacy laws for students under FERPA (Family Educational Rights and Privacy Act)," CIS affiliates are not permitted to "use official student identification numbers or testing identification (GTIDs) in their recordkeeping." In addition, CISGa indicated that affiliates' access to student information is limited to the current year's case managed students. As a result, "CISGa lacks consistent access to [data on CIS participants and students in districts not served by CIS] needed to compare the outcomes of CIS participants to school systems and the state average."
GaDOE Response: In its response, GaDOE had no disagreements with the report.

Communities In Schools

i

Table of Contents

Purpose of the Special Examination

1

Background

1

Communities In Schools

1

The CIS Model

2

CIS Participants

3

CIS Funding

4

Requested Information

7

Which school systems are participating in Communities In Schools?

7

What services are being provided through Communities In Schools?

9

How are state grant funds appropriated through GaDOE being used?

11

What is the state's return on investment in Communities In Schools?

14

How do graduation rates for Communities In Schools participants compare to school

systems and the state?

16

Appendicies

21

Appendix A: Objectives, Scope, and Methodology

21

Appendix B: CIS Affiliates and School Systems School Years 2011 - 2015

24

Appendix C: Levels of Service by Affiliate and Service Category School Year 2015 26

Appendix D: CIS Affiliate Fundraising, Fiscal Year 2015

27

Communities In Schools

1

Purpose of the Special Examination
This examination was conducted at the request of the House Appropriations Committee. In its request, the Committee indicated that Communities In Schools had been a long-term investment partner with the state and expressed an interest in an overall update of Communities In Schools activities and its effectiveness. Specifically, our examination addressed the following questions:
1. Which school systems are participating in Communities In Schools? 2. What services are being provided through Communities In Schools? 3. How are state grant funds being used by Communities In Schools of Georgia? 4. What is the return on the state's investment in Communities In Schools? 5. How do graduation rates for Communities In Schools participants compare to
school systems and statewide graduation rates?
A description of the objectives, scope, and methodology used in this review is included in Appendix A. The examination primarily focused on the activities and effectiveness of the traditional Communities In Schools model. While elements of the Performance Learning Center (PLC) model are discussed throughout the report, we did not evaluate PLCs separately due to time constraints. A draft of the report was provided to Communities In Schools of Georgia and the Georgia Department of Education for their review, and pertinent responses were incorporated into the report.
Background

CIS refers to the dropout prevention program. The CIS national office provides support for individual state offices, while state offices provide support for local
affiliates. CISGa is Georgia's state office.

Communities In Schools
The first Communities In Schools (CIS) non-profit organization started in Atlanta in 1971. Communities In Schools, Inc., the national organization, was formed in 1977 to replicate the Atlanta-based program nationally. The stated mission of CIS is "to surround students with a community of support and empower them to stay in school and achieve in life." It developed a dropout prevention strategy based on the premise that there are multiple reasons a student may drop out of school. This strategy, referred to as the "CIS model," serves at-risk students by bringing needed social services such as mental health counseling, assistance in obtaining governmental financial benefits, family strengthening initiatives, and mentoring to a centralized setting (typically, the school). Traditional academic assistance such as tutoring, special classes, and literacy skills development is also provided.1 By coordinating a system of supports (or integrated student supports) for students, families, and schools, CIS targets students' academic and non-academic barriers to learning. According to the CIS national office, 25 states and the District of Columbia had CIS networks in place during the 2015 school year and provided programs and services to approximately 1.5 million students.2

Communities In Schools of Georgia (CISGa) is the state's CIS organization. The goal of CISGa is "to assist communities throughout Georgia in implementing locally-defined, comprehensive stay-in-school programs which result in an increase in the number of

1 Both academic and non-academic services coincide with the core ideals of CIS: a personal relationship with a caring adult; a safe place; a healthy start; a marketable skill; and a chance to give back.
2 CIS network refers to the state offices and independent 501(c)(3) organizations (local affiliates) that deliver services under the CIS model.

Communities In Schools

2

children who continue their education at least through high school and are able to take their place in the workforce and in their communities." CISGa and its network of local affiliate organizations work together to achieve this goal. Local school systems interested in implementing CIS programs organize their own affiliate offices with CISGa assistance and many provide funding to support local affiliate operations. CISGa provides technical support to local affiliates which in turn provide direct support and programs to students and schools in their communities. Each affiliate office is an independent 501(c)(3) organization governed by a local board of directors and is headed by an executive director who is responsible for managing the office's daily operations and obtaining additional funding for the affiliate's operations. For school year 2015, the program had 32 affiliate offices.
The CIS Model
As shown in Exhibit 1, the CIS model is implemented by a CIS site team led by a trained site coordinator. The coordinator oversees implementation of the primary components and processes of the CIS model, including an annual needs assessment; planning with school leadership; delivery of programs and services; regular monitoring and adjustment of plans; evaluation of effectiveness in achieving school and student goals; and reporting. These processes are conducted throughout the school year.
Exhibit 1 CIS Model Flow Chart

CIS Site Team

Annual Needs Assessment

Annual Site Operations Planning

Connect Level 1 services to address school-wide needs
Connect Level 2 services based on individual needs

Monitor and
adjust services

Evaluate effectiveness of achieving
goals

Annual Reporting

Source: ICF National CIS Evaluation Report
Needs Assessment The CIS team conducts an annual needs assessment to determine the needs of the school. Needs may vary on a school-by-school and student-by-student basis. Needs are determined based on an analysis of multiple sources of data.
Site Planning Based on the needs assessment, site coordinators lead their site team in the development of a comprehensive site operations plan designed to address identified and prioritized academic and non-academic needs.
Identification and Referral School counselors, teachers, and school staff make most of the student referrals to CIS. Although specific criteria for referral can vary from one affiliate to the next, students can generally be referred for a

Communities In Schools

3

According to a national CIS study, high implementers are
characterized as schools that provide both Level 1
and Level 2 services, compared to partial implementers that offer only Level 1 or Level 2
services.
Although dropouts typically occur during high school
years, many CIS programs provide early intervention services to students prior
to their entry into high school.

variety of academic and non-academic reasons. For example, academic reasons can include poor academic achievement, poor attendance, disruptive behavior, or learning disability. Non-academic reasons for referral can include teen parenting, limited parental involvement, homelessness, lack of basic needs (food, clothing, health services), and suspected drug abuse or gang affiliation. If the site coordinator considers the student eligible to receive targeted support, they will obtain parental consent and develop an appropriate plan. Students deemed eligible in one school year may not need services the following year and students may be referred for and enrolled in CIS services throughout their school years as needed.
Programs and Services Based on the results of the needs assessment, each local affiliate office has the flexibility to develop and administer its own selection of programs and services to address the reasons students in their communities drop out of school. (Specific services provided by CISGa's local affiliates are discussed on page 10.) These services are delivered by schools, business partners, and community-based organizations. Programs and services can include whole school activities, long-term and well-defined programs (called sustained services), and short-term and intense interventions (called acute services). Programs and services are divided into two levels:
Level 1 includes whole school services. Whole school services are provided to all students at a CIS school. Examples include assemblies on substance abuse, career fairs, and field trips.
Level 2 includes sustained and acute services. Sustained services are designed to achieve one or more outcomes such as improved academic performance, attendance, or behavior. Acute services typically last only a few hours to a few days and are provided on an as-needed basis. Some examples include providing clothing, school supplies, eye glasses, dental care, or grief counseling to students.
In addition to these service levels, CISGa established Performance Learning Centers (PLCs) which are academy-style high schools for students who are not succeeding in the traditional school environment. PLCs provide opportunities for students to make up credits through online classes, with facilitation from on-site PLC teachers.
Monitoring and Adjustment The CIS site team continuously monitors student and school progress and adjusts supports to optimize results.
Evaluation and Annual Reporting Continuous assessments of partners and student supports are conducted by the CIS affiliates to demonstrate results and improve practice.
CIS Participants
In school year 2015, 41 school systems were serviced by the 32 local affiliate offices (some affiliates served multiple county and city school systems). Across school systems, more than 120,000 students attending 207 schools/sites were provided CIS programs and services, including the 10,580 who received sustained services. As shown in Exhibit 2, 53% (5,597) of students receiving sustained services were high school students, including those enrolled in alternative high schools and PLCs. An additional 44% (4,611) of participants were middle school (22%) and elementary school (22%) students. Of the

Communities In Schools

4

remaining 4% (372) of participants (which includes students at all grade levels), some were enrolled in combined schools or receiving services at community sites.3
Exhibit 2 CIS Student Participants, School Year 2015
0.6% 2% 1%

53%

22% 22%

Pre-K Elementary School Middle School High School1 Combined School Community

1 High school includes traditional high schools, alternative schools, and Performance Learning Centers. Source: CISGa services data

CIS Funding
Since 1999, the General Assembly has appropriated funds to CISGa totaling approximately $20 million.4 This funding, provided to CISGa through a grant administered by the Georgia Department of Education (GaDOE), comprises a very small portion (5% on average in fiscal years 2011 to 2015) of total funding to CISGa and local affiliates on an annual basis; most funding is generated through the fund raising efforts of local affiliates. As shown in Exhibit 3, state grant funds have remained relatively constant since fiscal year 2011 while funding from other sources has decreased, primarily due to reductions in the amount generated through the fund raising efforts of local affiliates. According to CISGa, this is due in part to the recession and reductions in the amount of federal funding for human services and education support available to nonprofit organizations.

3 Combined schools are those with grades K-12 or some combination of elementary, middle, and high school grades. 4 CISGa also received demonstration site funding for two local offices in 1998.

Communities In Schools

5

Exhibit 3 State Grant Funds Comprise a Small Percentage of Total CIS Funding, Fiscal Years 2011-2015
$35,000,000

$30,000,000 $25,000,000 $20,000,000 $15,000,000 $10,000,000
$5,000,000

93%

93%

91%

All Other Funds Other State Funds GaDOE Grant

90%

86%

$2011

2012

2013

2014

2015

GaDOE Grant Funds Other State Funds Other Funds Total Funds

2011

2012

2013

2014

2015

$ 933,100 4% $ 914,438 3% $ 905,107 5% $ 941,093 5% $ 1,033,100 7%

$ 627,892 3% $ 1,133,059 4% $ 754,525 4% $ 781,021 4% $ 940,888 7%

$21,155,159 93% $27,767,369 93% $16,283,164 91% $16,013,250 90% $12,056,176 86%

$22,716,151

$29,814,866

$17,942,795

$17,735,364

$14,030,164

Source: CISGa Financial Data

Other state funds include grants from other state agencies or universities, such as the Governor's Office for Children and Families and Middle Georgia State College. In fiscal year 2015, CISGa was awarded approximately $255,000 in state grant funds by the Governor's Office of Student Achievement.
Other funding sources accounted for approximately 86% to 93% of CISGa's total revenues in fiscal years 2011 through 2015. These sources include contributions from federal grants, private donors (individuals, foundations, and corporations), local school systems, local governments, and other sources. CISGa and its affiliates also benefit from in-kind donations (such as school supplies, office space, or guest speakers for events) provided by these same sources. In fiscal years 2011 through 2015, in-kind services were valued at $6 million. During the same period, CIS affiliates recorded a combined total of 427,000 volunteer service hours, worth $9.5 million.5

5 Value of volunteer hours was calculated based on the dollar value of volunteering in 2015 in Georgia ($22.25), according to the Corporation for National & Community Service.

Communities In Schools

6

CISGa Grant Requirements
GaDOE contracts with CISGa to support local affiliates (the primary providers of CIS services) in their comprehensive dropout prevention plans. The contract requires CISGa to provide training and technical assistance to local affiliates to enable them to create comprehensive stay-in-school plans, provide support to students and schools, develop a board of directors, meet established stay-in-school goals, improve attendance and academic achievement, reduce discipline problems, and increase volunteer participation in their local school system. CISGa must also submit regular reports to GaDOE each year, with each report including such elements as descriptions of plans for achieving stay-in-school goals and for increasing community and parent involvement, services provided to schools and students, and performance data for each affiliate. CISGa collects the required information from all affiliates and sends the consolidated reports to GaDOE.

GaDOE's dropout rate calculation is the
number of students in grades 9-12 with a withdrawal code corresponding to a
dropout divided by the number of students that
attended the school. The adjusted cohort
graduation rate indicates the
percentage of students who graduated on time
(or with their cohort).

As part of the state's strategic planning requirements, CISGa also compiles and reports information on several performance indicators relating to the provision of services by local affiliates. Exhibit 4 shows the measures reported to GaDOE in fiscal years 2011 to 2015. Two of CISGa's measures (dropout rate and graduation rate) are calculated differently than similarly named rates calculated by GaDOE, as explained below:
Annual dropout rate for students served by CIS This measure indicates the frequency of dropouts among individual students that receive Level 2 CIS services (sustained or acute services) in a single school year. CISGa's dropout rate is calculated by dividing the total number of Level 2 students who dropped out by the total number of Level 2 students enrolled in CIS participating high schools during that school year.
Graduation rate for students served by CIS This measure indicates the frequency of graduation among individual high school seniors that receive Level 2 CIS services in a single school year. Unlike GaDOE's adjusted cohort rate, CISGa's graduation rate is calculated by dividing the total number of Level 2 high school seniors who graduated by the total number of Level 2 high school seniors enrolled in CIS participating high schools during that school year. According to CISGa, while it would like to be able to calculate a cohort graduation rate for caseload (Level 2) students, it is unable to do so due to local school systems' restrictions on CIS's use of student-level data.

Exhibit 4 Governor's Budget Report Performance Measures, Fiscal Years 2011 - 2015

Performance Measures Annual dropout rate for students served by CIS

FY 2011 FY 2012 FY 2013 FY 2014 FY 2015

2.4%

3.1%

3.0%

5.4%

4.6%

Graduation rate for students served by CIS

93.2%

82.5%

86.9%

81.6%

89.5%

Number of at-risk students receiving services

17,570

14,539

10,092

10,225

10,580

Percentage of school districts with campuses participating in CIS

30%

26%

26%

26%

23%

Average amount of state funds spent per student served Total dollars leveraged1

$53.11

$62.90

$89.69

$91.26

$97.65

$16,289,161 $19,342,312 $14,528,791 $14,813,871 $12,045,405

1Includes in-kind contributions

Source: FY 2014 Governor's Office of Planning & Budget Agency Performance Measures Report & CISGa reported performance measures

Communities In Schools

7

Requested Information

Which school systems are participating in Communities In Schools?

Approximately one-fifth (41 of 201 total systems) of school systems across the state had CIS during school year 2015.6 As shown in Exhibit 5, some affiliates served multiple school systems. For example, the Carroll affiliate services both Carrollton City and Carroll County school systems. Each school system may have multiple schools/sites where CIS programs and services are delivered. In school year 2015, the number of sites managed by affiliate offices ranged from 1 to 34, including PLC sites. See Appendix B for a comprehensive list of affiliates and school systems served in school years 2011-2015.

Exhibit 5 School Systems Participating in CIS, School Year 2015

Catoosa

Dade Walker

Whitfield

Murr ay

Chattooga Fl oy d

Gordon Bartow

Fannin

Union

Towns

Rabun

Gilmer
Pi ck ens Cherokee

Lumpkin Dawson

White Habersham Stephens

Banks

Franklin

Forsy th

Hall

Jack son

Madis on

Hart Elber t

County Participated County & City Participated PLC

Polk Haralson

Paulding

Cobb

Douglas

Gwinnett Dekalb

Barr ow Walton

Clark e Oconee

Oglethor pe

Wi lk es

Lincoln

Carroll Heard

Fulton Clayton

Newton

Morgan

Fayette

Henry

Coweta

Spalding

Butts

Jasper

Putnam

Greene

Taliaferro

McDuffie

Warren

Hanc oc k

Gl as cock

Columbia Ric hmond

Troup

Meriwether

Lamar Pike
Monroe

Jones

Bal dwi n

Washington

Jeffer son

Burk e

Harris

Upson Talbot

Crawford

Bibb

Twiggs

Wi lk inson

Johns on

Jenkins

Sc reven

Muscogee

Taylor

Chattahoochee Marion Sc hl ey

Macon

Stewart

Webster

Quitman

Sumter

Peach Houston

Bleckley

Dooly

Pulaski

Dodge

Laurens

Treutlen

Wheeler

Emanuel Candler

Bulloch

Toombs

Ev ans

Crisp

Wi lc ox

Telfair

Tattnall

Effingham

Br yan

Chatham

Terrell Lee
Randolph

Clay

Calhoun

Dougher ty

Early

Baker

Miller Seminole
Decatur

Mitchell Grady

Turner Worth

Ben Hill Irwin

Colquitt

Tift Berr ien
Cook

Jeff Davis

Appling

Long

Liberty

Coffee Atk inson

Bacon

Wayne

Pierce

Ware

Brantley

McIntosh Glynn

Thomas

Br ooks

Lanier

Cli nc h

Lowndes

Ec hol s

Charlton

Camden

Affiliates 1. Albany/Dougherty 2. Athens/Clarke 3. Atlanta 4. Augusta/Richmond 5. Baldwin 6. Barrow 7. Ben Hill 8. Berrien 9. Burke 10. Candler 11. Carroll 12. Catoosa 13. Central Georgia 14. Cochran/Bleckley 15. Coweta 16. Dodge 17. Douglas 18. Glascock 19. Glynn 20. Hancock 21. Hart 22. Henry 23. Laurens 24. Marietta/Cobb 25. Rome/Floyd 26. Savannah/Chatham 27. Troup 28. Turner 29. Walton 30. Warren 31. Washington 32. Wilkes

Source: CISGa data

6 Figure includes systems operated by Department of Juvenile Justice, Department of Corrections, and Department of Human Services, as well as charter schools and state schools.

Communities In Schools

8

As shown in Exhibit 6, overall participation in CIS has declined since 2011. The number of affiliates, school systems, and sites all declined between 20% and 29% during the fiveyear period since 2011. The number of students participating in CIS declined by 17% over the same time period. Over the five-year period, 13 affiliates closed (representing 17 school systems), and 5 affiliates opened.7 Two of the five openings were affiliates that closed and reopened during the time period examined.8

Exhibit 6 CIS Participation Decreased between School Years 2011 and 2015
15%

10%

5%

Percent Change

0%

-5%

-10%

-15%

-20%

2011 to 2012 Affiliates

2012 to 2013 School Systems

2013 to 2014

2014 to 2015

Sites All Students

Number of CIS Affiliates1

2011 402

2012 39

2013 40

2014 34

2015 32

Number of School Systems with CIS

53

44

47

42

41

Number of CIS Sites3

292

257

228

219

207

Number of Students Receiving CIS Services (Levels 1 & 2)

146,152 124,904 125,574 108,280 121,260

Students Receiving Level 2 Services

17,777

14,539

10,136

10,225

10,580

1Includes the number of affiliates that served students in each year. 2 Includes an affiliate that closed in the prior year, but continued to serve students into school year
2011. 3The majority of CIS sites are schools; however, these numbers also include community sites.

Source: CISGa services data

CISGa staff stated that the reductions in participation occurred for various reasons. For example, the Barrow, North Georgia, and Jefferson affiliates closed due to loss of funding and/or staff. In addition, CISGa staff noted that Atlanta Public Schools (APS) stepped away from all third party agreements--including CIS--after the test cheating scandal.
7 Three PLCs also closed during this period, representing three additional school systems. 8 The Stephens affiliate closed, but services were still offered in Stephens County in 2015 through the Hart affiliate.

Communities In Schools

9

CISGa staff also noted that several affiliates (e.g., Bulloch, Effingham, Jenkins, McDuffie, Stephens) were not making progress toward meeting Total Quality System standards and, therefore, officially withdrew from the network in 2014.9 While expansion and increases also occurred between 2011 and 2015, CISGa staff stated that they have prioritized providing more comprehensive services in areas where CIS already exists over expanding to new areas.
What services are being provided through Communities In Schools?
CIS services fall into the following categories: case management, basic needs and resources, academic assistance, life skills/social development, family engagement/ strengthening, physical health, mental health, behavior interventions, college/career preparation, community service/service learning, and enrichment/motivation. These services are targeted to meet the needs of students, parents/families, and schools, and are intended to promote positive outcomes in such areas as dropout and graduation, academic achievement, attendance, and behavior.
Exhibit 7 describes the categories of services provided at sites across the state's 32 CIS affiliates during school year 2015. Within each category, there are a number of different strategies designed to address the risk factors associated with dropping out and other adverse outcomes. According to CISGa, these services and strategies follow best practices and are proven interventions in dropout prevention.10 For example, according to research studies, alternative education programs, mentoring or counseling, vocational training, college-oriented programming, and case management were consistently associated with reductions in dropping out.
These services are not available in every affiliate or CIS site, however. CISGa notes that the services and strategies available at a particular site depends on the needs of the students and the school, as well as the resources available to meet those needs. As Exhibit 7 shows, case management, life skills/social development, basic needs and family services were provided by all 32 affiliates while service to others programs and enrichment activities were provided by fewer affiliates. The availability of services also varied across the 207 CIS sites--even those sites served by the same affiliate had varying services because they are addressing different needs at the individual student and site level.
Surveys of staff at affiliate offices and principals at CIS sites support CISGa's point that, to some extent, funding determines the availability of services. According to CISGa's principal survey, 15% (21 of 139 respondents) of principal respondents stated that they would like to see additional services implemented at their school, and 9% stated they would like more funding for CIS.11 Similarly, in response to a survey of CIS affiliate staff conducted by the audit team, 8 of 30 (27%) respondents said they would like to utilize additional CIS services; however, most indicated that funding was a barrier to the provision of additional services.

9 The CIS Total Quality System (TQS) provides an integrated set of standards and policies for affiliates and sites to ensure fidelity to the CIS model. 10 Communities In Schools of Georgia, Mid-Year Progress Report to The Georgia Department of Education, FY2015. 11 CISGa conducts an annual survey of principals as part of its annual evaluation efforts.

Communities In Schools

10

Exhibit 7 CIS Provides Services in Ten Categories, School Year 2015

Service Category

Description

Case Management

Consists of an in-school CIS coordinator who works intensively with a caseload of students referred for assistance. Coordinator ensures students receive needed social services and may refer the student to outside agencies for assistance.

Basic Needs

Includes resources such as housing assistance, food banks, clothing, childcare and school supplies. May also include a direct school need such as providing students with school uniforms, school supplies, caps and gowns for graduation, or transportation to school.

Affiliates (n=32) 32
32

Sites1 (n=207)
157
172

Includes mentorship, leadership skills, school safety, personal

skills development and abstinence education programs.

Life Skills

Affiliates follow best practices identified in the area and may

32

188

use a particular curricula. These generally help students build

resiliency.

Includes programs that seek to increase parent involvement,

Family

give parents updates on student progress, and train parents in

a number of research-based parent education programs (e.g.,

32

148

Services

Parents as Teachers and Parents Assuring Student's

Success).

Includes health screenings and education, drug and alcohol

Physical &

prevention, and linkages to dental and eye care providers.

29

123

Mental Health May also include schools hosting health clinics and services to

promote healthy living.

Behavior

Includes services such as anger management and conflict

resolution, as well as gang, bullying, and violence prevention

29

136

Interventions programs.

College & Career Readiness

Includes career planning, internships, training programs, college exploration, college application support and scholarships.

26

111

Academic

Includes afterschool programs, tutoring, homework assistance,

30

151

Supports

and test tutoring.

Integrates youth into the community and school by providing

Service to Others

opportunities for youth to be resources through community service projects, service learning, peer mentoring, and peer

24

122

tutoring.

Enrichment Activities

Includes speakers and workshops, events, clubs, recreation and sports activities.

26

117

1 Sites include Pre-K, elementary, middle, and high schools, combined (e.g., elementary/middle, middle/high, or K-12), alternative schools, PLCs, and community-based locations.

Source: CISGa service information

Communities In Schools

11

Level 1 services are services and resources available to the entire school. Level 2 services are aimed at a subset of students who are most at risk of dropping out of
school.

Where services do exist, they may vary in their intensity. As shown in Exhibit 8, our review found that most service categories included programs and services provided at Level 1 (low-intensity, generally) and Level 2 (intensive). Depending on school and student needs, the same service may be offered at Level 1 or Level 2. For example, school supplies (categorized as a Basic Need) can be provided to all students in a school, while a student whose family needs assistance with shelter, food, and clothing would require more resource intensive services. Appendix C shows the levels of services provided across the 32 affiliates.
Exhibit 8 CIS Service Levels Vary, School Year 2015

Enrichment Activities Service to Others
Academic Supports College & Career Readiness
Behavior Interventions Physical & Mental Health
Family Services Life Skills
Basic Needs Case Management

0

5 10 15 20 25 30 35

Number of Affiliates

Level 1 Only Level 2 Only Level 1 & Level 2

Source: CISGa services data

How are state grant funds appropriated through GaDOE being used?
All state grant funds appropriated for CIS are passed through from GaDOE to CISGa. As shown in Exhibit 9, the majority of state funds are then passed through from CISGa headquarters to the local affiliates. Of the $1.03 million in state grant funds received in fiscal year 2015, CISGa distributed $693,775 (67%) to local affiliates. CISGa reserved an additional $80,430 (8%) to pass through to local affiliates in fiscal year 2016. After a cost share for the AmeriCorps12 program ($65,000 or 6%), the remaining $193,895 (19%) was retained by CISGa.

12 The CISGa AmeriCorps*VISTA Mentoring Project focuses on anti-poverty programming (including mentoring and family strengthening) and develops the capacity for CIS sites to better serve youth, families, and schools.

Communities In Schools

12

Exhibit 9 Distribution of GaDOE Grant Funds, Fiscal Years 2011 2015

FY 2011

FY 2012

FY 2013

FY 2014

FY 2015

GaDOE Grant Funds Passed Through Retained by CISGa AmeriCorps Cost Share Reserved Funds1 Total Grant Funds

$691,000 (74%) 197,100 (21%) 45,000 (5%) 0 (0%)
$933,100 (100%)

$692,000 (76%) 177,438 (19%) 45,000 (5%) 0 (0%)
$914,438 (100%)

$684,665 (76%) 175,442 (19%) 45,000 (5%) 0 (0%)
$905,107 (100%)

$696,000 (74%) 200,093 (21%) 45,000 (5%) 0 (0%)
$941,093 (100%)

$693,775 (67%) 193,895 (19%) 65,000 (6%) 80,430 (8%)
$1,033,100 (100%)

Distributions to Local Affiliates

Number of Affiliates2 Maximum Amount

39 $33,000

413 $20,000

40 $20,000

34 $20,000

32 $66,0004

Minimum Amount

$17,000

$16,000

$12,000

$16,000

$9,0005

Average Amount

$17,700

$16,900

$17,100

$18,100

$21,7006

1 This amount is intended for distribution to affiliates in subsequent years. 2 This includes the number of affiliates receiving funding in each fiscal year. 3 Includes two new affiliates that received funding prior to serving students. 4 The Augusta/Richmond affiliate received two payments in fiscal year 2015: one for $16,000 and an additional payment of $50,000 for

expansion. 5 The Barrow affiliate closed mid-year and, therefore, only received partial funding in fiscal year 2015. 6 Based on the total cash payments in fiscal year 2015 ($693,775).

Source: CISGa financial data

Local Affiliates Use of State Grant Funds
As shown in Exhibit 9, state grants to local affiliates averaged approximately $21,700, with individual grant amounts ranging from $9,000 to $66,000 in fiscal year 2015. According to CISGa staff, each affiliate receives a minimum amount each year, depending on the total state grant amount. In addition, affiliates may receive additional funds if they meet certain criteria, such as expanding or becoming accredited.13 It should be noted that state grant funding provided to each affiliate has increased from $17,700 to $21,700, on average, over the last five years as the number of affiliate offices has decreased.
In fiscal year 2015, local affiliates used 53% ($368,000) of state grant funds to defray personnel costs, while the remaining 47% (approximately $326,000) was designated for program and administrative costs. (See Exhibit 10.) State grant funds spent on salaries and benefits covered a portion of the costs of local affiliate executive directors. Information for school year 2015 indicated that the compensation range for executive directors (including salaries and benefits) ranged from $24,000 to $120,000, with an average of $60,000. Essentially, the state grant would cover about 35% of an executive director's salary, on average.

13 Becoming accredited refers to CIS's Total Quality System (TQS).

Communities In Schools

13

Exhibit 10 CIS Affiliates Spent 53% of State Grant Funds on Salaries & Benefits, Fiscal Year 2015

$124,783 (18%)
$201,161 (29%)

$367,831 (53%)

Salaries & Benefits
1
Program Costs
2
Administrative Costs

1Program costs includes the "Other Personnel" category (25%), which may include Site Coordinator expenses. 2Administrative Costs include: Travel (5%), Consultants/Contract may also include Site Coordinator expenses (5%), Office Supplies (2%), Office Equipment (1%), Phone/Internet (2%), Rent (0.2%), and All Other (2%).
Source: FY 2015 CISGa financial data
It should be noted that the primary purpose of the state grant is to provide partial funding for an executive director at each local affiliate office. In the past, executive directors were required to seek out and obtain additional funding that at least matched the amount of the state grant. While no longer a requirement, most affiliate offices obtain at least $1 of matching funds for every $1 of state grant funds they received. In fiscal year 2015, the 32 affiliate offices raised an average of $18.07 per state dollar granted. Appendix D shows the funds raised by each affiliate office in fiscal year 2015.
CISGa's Use of State Grant Funds The remaining grant funds that were not passed through to local affiliates were used by CISGa's state office to pay personnel, program, and administrative costs. As shown in Exhibit 11, 40% ($137,000) of funds remaining in fiscal year 2015 were used for salaries and benefits of personnel providing training and assistance to local affiliates. As previously noted, grant funds also covered approximately $145,000 (43%) in program costs, which consisted of funds reserved for pass through to local affiliates in fiscal year 2016 and funds to cover CISGa's portion of AmeriCorps program costs. The remaining 17% ($57,000) of state grant funds covered CISGa's administrative costs.

Communities In Schools

14

Exhibit 11 CISGa Spent 40% of Remaining State Grant Funds on Salaries & Benefits, Fiscal Year 2015

$56,653 (17%)
$145,448 (43%)

$137,243 (40%)

Salaries & Benefits
Program Costs
1
Administrative Costs

1Administrative Costs include: Travel/Parking (6.5%), Training/Meetings/Conferences (4%), Contracted Services (2%), and All Other (8%). All Other includes: audit fees, telephone, office supplies, rent/utilities, staff development, postage, insurance, printing/design, and computers/technology. Source: FY 2015 CISGa financial data
What is the state's return on investment in Communities In Schools?
Research indicates that while the payback may take time to accrue, the benefits gained from models such as CIS exceed the relative costs. A 2012 study assessing the economic impact of CIS estimated the benefit-cost ratio to be 11.6--that is, there is an estimated return of $11.60 for every one dollar invested. In Georgia, the state's annual investment in CIS of $1 million is relatively small compared to other sources of funding for CISGA and local affiliates. Based on current funding levels and number of CIS students, the state's return on investment is estimated to be $18:$1 for a high school graduate over his or her lifetime.14
Our analysis of the state's return on investment includes: (1) the dollar amount of state funds spent per CIS student, (2) the amount of state taxes paid by high school dropouts compared to the amount paid by high school graduates, and (3) the estimated increase in the state's graduation rate needed to generate a positive return on investment.
14 Although CIS served more than 120,000 students in fiscal year 2015, our analysis is based on the 10,580 case-managed (Level 2) students.

Communities In Schools

15

State Cost per CIS Student Based on the number of CIS students who received sustained (Level 2) services in school year 2015 (10,580 at all grade levels), the average state grant amount per student was $98.15 As previously noted, state grant funds of $1 million comprised only 7% of CISGa's total revenue in fiscal year 2015.

Estimated Additional Taxes Generated Our estimate of the benefit to the state is derived from additional tax revenue collections for each CIS student who attains a high school diploma. Our analysis of earnings of high school graduates compared to non-graduates indicates that each additional high school graduate produces an estimated $7,500 in tax benefits to the state over the working lifetimes of the students.16 These benefits are based on the additional income and sales tax revenues generated by high school graduates each year. When we considered the earnings differential between dropouts and those completing some college and higher, the benefits to the state increase significantly. See Exhibit 12 below.

Exhibit 12 Additional Income and Sales Tax Benefits to State by Educational Attainment

$140,000 $120,000 $100,000
$80,000 $60,000 $40,000 $20,000
$0

$114,658
$80,766
$43,920 $33,332 $7,497 $11,212 $14,148

Source: U.S. Census Bureau's Current Population Survey and Georgia tax rates
Breakeven AnalysisTo generate returns equivalent to the value of the fiscal year 2015 state grant ($1,033,100), it is estimated that 138 CIS participating students would have needed to graduate high school that year. This "breakeven point" is the equivalent of increasing the school year 2015 statewide graduation
15 Using this cost per student ratio only considers the portion of students that received more intensive services; CIS provided Level 1 services to approximately 120,000 total students. If these students were included in the analysis, the state's cost per student would decrease and the return on the investment would increase. 16 This analysis compared a high school dropout's annual income over 35 years to a high school graduate's annual income over 35 years.

Communities In Schools

16

rate from 78.8% to 79.0%, an increase of 0.2 percentage points. As shown in Exhibit 13, any additional CIS graduates would create a positive return on the state's investment in CIS.

Exhibit 13 138 CIS Graduates Allows the State to Breakeven on its Investment, Fiscal Year 2015
$4,000,000

$3,500,000

Tax Revenue Generated

$3,000,000 $2,500,000 $2,000,000

Breakeven to Cover State Grant Cost =
138 Students

$1,500,000

$1,000,000

$500,000

$-

0

100

200

300

400

500

Number of CIS Graduates

FY 2015 Grant

Tax Revenue Generated

Source: U.S. Census Bureau's Current Population Survey and Georgia tax rates

Additional analyses of CIS would further help inform policy decisions. An assessment of CIS's impact on academic outcomes is discussed on the following pages. However, a review of CIS's relative cost-effectiveness compared to other dropout prevention programs is not currently possible without specific student-level data on various interventions they receive.

How do graduation rates for Communities In Schools participants compare to school systems and the state?

Graduation rate describes the proportion of high school freshmen graduating with regular diplomas four years later. This rate measures on-
time graduation, with adjustments for students
who transfer in/out of school or leave for other
reasons.

Both GaDOE and CISGa (due to local restrictions on CIS's use of student-level data) lack student-specific data needed to compare the outcomes of CIS participants to school systems and the state average as requested. Alternately, we assessed CIS's impact on graduation rates by comparing average rates of a sample of CIS schools to statewide rates and to a sample of non-CIS schools matched on similar demographics. In addition, we analyzed the change in graduation rates of CIS schools pre- and post-CIS implementation. As discussed below, we found that schools with CIS had improved graduation rates during the time periods reviewed and, comparatively, their graduation rates were higher than the statewide average and the matched non-CIS schools. However, we could not establish a correlation between the graduation rates and the presence of CIS and could not project the results to a larger population based on available data. The details of our analyses are discussed below.

Communities In Schools

17

Comparison of CIS Schools' and Statewide Rates
In this analysis, we compared average graduation rates for a sample of 35 schools that participated in CIS to statewide rates.17 As shown in Exhibit 14, CIS schools' average graduation rates were higher than statewide rates by 1.3 to 2.9 percentage points in school years 2011 to 2014. Year-to-year average graduation rates improved for CIS schools and statewide by 1 to 2 percentage points.

As previously discussed, we estimate that CIS would have to influence a relatively small number of students to progress through school and ultimately graduate to achieve a breakeven at current funding levels. For example, we estimate that 125 CIS graduates were needed to breakeven on the state's investment in school year 2014. We found that in the 35 CIS sample schools alone (out of 60 total CIS high schools), approximately 7,600 students graduated.18 While we were unable to isolate the number directly influenced by CIS, it is reasonable to assume that the state at least broke even on its investment.

Exhibit 14
CIS Schools had Higher Average Graduation Rates Compared to Statewide Rates, School Years 2011 20141

80%

75%

+2.3

+2.8

+1.3

+2.9

70%

65% 60% 55%

70.4% 67.5%

72.5% 69.7%

73.1% 71.8%

74.8% 72.5%

50%

2011

2012

2013

2014

CIS Statewide

1 Statewide rate includes all public schools in Georgia, ranging from 424 to 475 schools each year, over the four years. CIS schools are included in the statewide rate.

Source: GaDOE graduation rate analysis & GOSA reported data

Comparison of CIS and Non-CIS Schools' Rates
In this analysis, we compared average graduation rates for a sample of 70 schools, consisting of 35 CIS schools and 35 non-CIS schools. The non-CIS schools were matched on demographic factors such as enrollment, locale, poverty, race/ethnicity, and special

17 Based on statewide graduation rates as reported annually by GaDOE in the Governor's Budget Report and dropout rates calculated by the audit team from GaDOE's dropout rate data. 18 Consists of schools where a portion of students received Level 1 or Level 2 services, or both.

Communities In Schools

18

education (some of which are also associated with dropping out).19 Our review found that the sample of CIS schools experienced higher average graduation rates each year since school year 2011, with average rates for CIS schools exceeding non-CIS school rates by 0.8 to 1.9 percentage points each year. CIS schools' graduation rates improved by 1 to 2 percentage points each year. While graduation rates for non-CIS schools improved overall by 5 percentage points, the rates slightly declined between school years 2012 and 2013.

Exhibit 15 CIS Schools had Higher Average Graduation Rates Compared to Similar Schools, School Years 2011 2014

80%

75%

+1.5

+0.8

+1.9

+1.8

70%

65%

60% 55%

70.4% 68.6%

72.5% 71.7%

73.1% 71.1%

74.8% 73.2%

50%

2011

2012

2013

CIS Non-CIS

Source: GaDOE graduation rate analysis & GaDOE reported data

2014

Pre- and Post-Implementation Analysis of CIS In this analysis, we compared school graduation rates one year prior to the year of CIS implementation to rates in the first and second year following CIS implementation. There were 65 schools that implemented CIS between school years 2000 and 2011 with sufficient data to conduct the analysis.
Our analysis revealed that the majority of schools saw improved graduation rates after CIS implementation. As shown in Exhibit 16, prior to implementation, schools in our sample had an average graduation rate of 66.9%. The average increased to 70.2% after the first year of implementation, and to 70.6% after the second year of implementation.

19 Some non-CIS schools may have participated in CIS in the past, but we limited the group to those that had been removed from CIS participation for at least eight years to limit the potential for any lingering impacts of CIS to skew the results.

Communities In Schools

19

Exhibit 16 CIS Schools' Average Graduation Rate Improved After CIS Implementation

75% 70%

+3.3

+0.4

65%

60% 55%

66.9%

70.2%

70.6%

50%

Year Prior

Year 1

Source: GaDOE graduation rate analysis & GaDOE reported data

Year 2

Other Studies of CIS
We identified additional independent, third party research studies on CIS to better understand CIS's impact on academic outcomes. We focused on research that had been identified as meeting high methodological standards. A February 2014 research study conducted by Child Trends20--a nonprofit, nonpartisan research center--identified four rigorous outcome evaluations of CIS programs based on factors such as research design, presence of statistically significant results, and equivalence of comparison groups at baseline.21
According to the Child Trends report, research studies found CIS to have a positive impact on certain outcome measures assessed. 22 These studies found that CIS had a positive effect on academic achievement, based on measures associated with student progress, school attendance, and math achievement.23 For example, according to the Child Trends report, a national evaluation of CIS found significant, positive effects on promoting power (a proxy for dropout rate) after three years of implementing the CIS model.24 While not noted in the Child Trends report, the national evaluation also concluded that CIS is linked with slight increases in graduation rates, particularly among CIS schools that use the full CIS model and use their resources to serve more students with fewer services. Two other evaluations found CIS's case management services had positive, significant impacts on another measure of student progress-- credit completion--initially; however, the impacts lessened over time. The studies also
20 Child Trends, Making the Grade: Assessing the Evidence for Integrated Student Supports, February 2014. 21 Standards varied according to the type of research design. 22 Evaluations were of CIS in three cities (Jacksonville, Florida, Wichita, Kansas, and Austin, Texas) as well as a national evaluation. 23 Alternately, the evaluations found less definitive evidence of CIS's impact on non-academic outcomes, such as student behavior. 24 Promoting power is a proxy measure of dropout rate and determines the proportion of students who are promoted to the next grade on time. The measure was developed by Johns Hopkins University and is used to classify schools as "dropout factories."

Communities In Schools

20

found CIS had significant impacts on attendance rates and math test scores. It should be noted that due to the small number of evaluations assessed, the Child Trends report indicates that these conclusions are tentative.

CISGa has its own evaluation process. CISGa monitors student improvement as it relates to academic outcomes shown in Exhibit 17. In addition, its survey of school principals found CIS to have a positive influence on improving outcomes in the areas of attendance, behavior, academics, and progress toward graduation. As the exhibit shows, approximately 74% of principals thought CIS had helped to improve the rate of disciplinary referrals, and a higher percentage found CIS to be helpful in improving academic performance, attendance, and progress toward graduation.

Exhibit 17 Reported Improvements in Other Outcomes

Outcome Measures

2014 Annual Report
(% of CIS students showing improvement)

2015 Principal Survey Results
(% of respondents that found CIS to be "extremely helpful" or "very helpful" in
improving outcomes)

Attendance

64.6%

Discipline

65.6%

Academic Performance Elementary School Students
Middle School Students High School Students
Progress Toward Graduation

66.6% 61.3% 61.5%
94.6%1

1Represents the High School Stay In School Rate Source: CISGa FY 2014 annual report & FY 2015 Principal Survey data

80.8% 74.1% 78.3%
82.0%

CISGa's Response: In its response, CISGa indicated that its "affiliates collect and report a lot of student level data to [them], but whether affiliates are allowed to use the full student ID or GTID that would allow matching to the GaDOE student records is at the discretion of district leadership based on the local interpretation of FERPA requirements."
CISGa also noted that its "evaluation department is not currently given access to data from the GaDOE student data collection." Any student-level data it obtains comes directly "from the local districts and [they] are only given access to data for caseload students (those with parent permission granting access)." They stated that, as a result, they "would not be able to do comparative analysis [themselves]" because they do not "have access to comparison data." CISGa added that "the student records are not verified until months after [their] required reporting to GaDOE."

Communities In Schools

21

Appendix A: Objectives, Scope, and Methodology
Objectives This report examines the Communities In Schools (CIS) program, funded through a grant from the Georgia Department of Education (GaDOE). Specifically, our examination set out to determine the following:
1. Which school systems are participating in Communities In Schools? 2. What services are being provided through Communities In Schools? 3. How are state grant funds being used by Communities In Schools? 4. What is the return on the state's investment in Communities In Schools? 5. How do graduation rates for Communities In Schools participants compare to
school systems and statewide graduation rates?
Scope This special examination generally covered activity related to CISGa that occurred in fiscal years (or school years) 2011 through 2015, with consideration of earlier or later periods when relevant. Information used in this report was obtained by reviewing CISGa and GaDOE documents, conducting interviews with program and agency staff, analyzing data provided by CISGa and GaDOE, and reviewing research studies.
We obtained data from CISGa on participation in CIS which included a list of sites/schools that participate in CIS, along with total student enrollment at the school by level of service. While we concluded that the information was sufficiently reliable for the purposes of our review, we did not independently verify the data.
We also obtained school-level demographic data from GaDOE in order to conduct analyses related to graduation rates. Data provided included enrollment, diplomas, special education, attendance, dropout rate, and free/reduced lunch. The data were assessed for issues with reliability and were found to be generally reliable for the purposes of our review.
Both GaDOE and CISGa lack student-specific data needed to compare the outcomes of CIS participants to school systems and the state average as requested. Due to local school systems' interpretation of the Family Educational Rights and Privacy Act (FERPA), CISGa is unable to access student information (e.g., student identification numbers or student testing identification numbers (GTID)) needed to conduct studentlevel analyses. In addition, GaDOE does not track student participation in CIS. As a result, we based our analysis on school-level data.
Methodology To determine which school systems participate in CIS, we analyzed CISGa's services participation data. This data includes the affiliate, school system, and school/site. We used enrollment data to determine which sites, school systems, and affiliates had students enrolled in Level 2 services each year.
To determine what services are being provided through CIS, we reviewed documents describing CIS services, and analyzed data showing the number of participants in CIS and level of services provided. We also conducted a brief web-based survey of 32 local affiliates and received 30 responses. The survey sought to identify what services are

Communities In Schools

22

currently provided and needed by affiliates, as well as what barriers to providing those services might be. We obtained similar information from our review of CISGa's surveys of school principals.
To determine how GaDOE grant funds are being used by CIS, we reviewed and analyzed GaDOE budget documents and CISGa financial data (state office data and local affiliate funding/budget data).
To determine the extent to which the state receives a return on its investment in CIS, we reviewed research studies, obtained and analyzed income data from the U.S. Census Bureau's Current Population Survey (CPS), analyzed state income and sales tax rates, and calculated the present value of lifetime earnings. We calculated the state's return on investment by dividing the difference between additional tax benefits associated with a high school graduate over his or her lifetime and the cost per CIS student by the state's cost per CIS student.
To determine how graduation rates of CIS participants compare to those of school systems and the state, we conducted several analyses of CISGa and GaDOE data.
First, we conducted a pre- and post-CIS implementation analysis to determine the change in school-level graduation rates. We obtained historical information from CISGa on school participation in CIS (from 1999 to 2014) and their graduation rates. We filtered the data to isolate a group of schools that received direct CIS services for two consecutive years (year 1 and year 2 of CIS) and that had complete graduation data for those years and the year prior to implementation of CIS. We considered the point at which the state's method of calculating graduation rates changed to ensure graduation rates analyzed had been calculated consistently. We identified a sample of 65 schools. We analyzed the change in rates over the two-year period for each school in our sample.
Second, using school-level data obtained from GaDOE, we filtered the data to identify high schools with complete demographic data. Using information obtained from CISGa on school participation in CIS, we coded the GaDOE data to identify CIS-participating schools. We used a statistical software to identify a group of CIS schools and a comparison group of non-CIS schools, based on the demographic data. This matching process identified 35 CIS schools and a matched set of 35 non-CIS schools. We compared graduation rates for the two groups over a four-year period (2011-2014). Our sample was based on the availability of data needed to conduct the analysis and is not representative sample of all CIS high schools.
Third, we compared graduation rates of the sample of 35 CIS schools identified above to statewide graduation rates.
Due to the lack of student-level data, we were limited to conducting school-level analyses. As such, our analyses describe school-level performance, potentially masking CIS's impact on individual student performance. In addition, because of the variety of factors that may be influencing school-level rates, we were unable to isolate the impact of CIS.
This special examination was not conducted in accordance with generally accepted government auditing standards (GAGAS) given the timeframe in which the report was needed. However, it was conducted in accordance with Performance Audit Division

Communities In Schools

23

policies and procedures for non-GAGAS engagements. These policies and procedures require that we plan and perform the engagement to obtain sufficient, appropriate evidence to provide a reasonable basis for the information reported and that data limitations be identified for the reader.

Communities In Schools

24

Appendix B: CIS Affiliates and School Systems School Years 2011 - 2015

Affiliate Name Albany Dougherty Athens Clarke
Atlanta
Augusta Richmond Baldwin Barrow Ben Hill Berrien Bulloch Burke Candler
Carroll
Catoosa Central Georgia Cochran-Bleckley Coweta Decatur Dodge Douglas Effingham Elbert Forsyth Glascock Glynn Hancock Harris Hart Henry Jefferson Jenkins
Laurens
Marietta/Cobb

System Name Dougherty County Clarke County Atlanta Public Schools DeKalb County Fulton County Richmond County Baldwin County Barrow County Ben Hill County Berrien County Bulloch County Burke County Candler County Carroll County Carrollton City Catoosa County Bibb County Bleckley County Coweta County Decatur County Dodge County Douglas County Effingham County Elbert County Forsyth County Glascock County Glynn County Hancock County Harris County Hart County Henry County Jefferson County Jenkins County Dublin City Laurens County Twiggs County Cobb County Marietta City

2011 X X X X
X X X X X X X X
X
X X X X X X X X X X X X X
X X X X
X X

2012 X X X X
X X X X X X X X
X X X
X X X X
X X X X X
X * X X X
X X

2013 X X X X X X X X X X X X X X X X X X
X X X
X X X X X X X X X X
X X

2014 X X X X X X X X X X X * X X X X X X X
X X X
X X X
X X
X X X X X X

2015 X X X X X X X X X X
X X X X X X X X
X X
X X X
X X
X X X X X

Communities In Schools

25

Affiliate Name

System Name

2011

2012

2013

2014

2015

McDuffie

McDuffie County

X

X

X

Muscogee

Muscogee County

X

Gilmer County

X

Lumpkin County

X *

Pickens County

X

North Georgia

Towns County

X

Union County

X

White County

X

Rome Floyd

Floyd County Rome City

X

X

X

X

X

X

X

X

X

X

Savannah-Chatham

Chatham County

X

X

X

X

Screven

Screven County

X

X *

X

Stephens

Stephens County

X

X

X

X

X

Sumter

Sumter County

X

X

X

Troup

Troup County

X

X

X

X

X

Turner

Turner County

X

X

X

X

X

Walton

Social Circle City Walton County

X

X

X

X

X

X

X

X

X

X

Warren

Warren County

X

X

X

X

X

Washington

Washington County

X

X

X

X

X

Wilkes

Wilkes County

X

X

X

X

X

Total

53

44

47

42

41

*Indicates that the two data sources (Level 2 Service enrollment and Number of Site Coordinators) conflicted. In these cases, we counted the system as participating if one data source showed participation. Source: CISGa service data

Communities In Schools

26

Appendix C: Levels of Service by Affiliate and Service Category School Year 2015

Case Management Basic Needs Life Skills Family Services Physical & Mental Health Behavior Interventions College & Caree r Readiness Academic Supports Service to Others Enrichment Activities

Affiliate Albany/Dougherty Athens/Clarke Atlanta Augusta/Richmond Baldwin Barrow Ben Hill Berrien Burke Candler Carroll Catoosa Central Georgia Cochran/Bleckley Coweta Dodge Douglas Glascock Glynn Hancock Hart Henry Laurens Marietta/Cobb Rome/Floyd Savannah/Chatham Troup Turner Walton Warren Washington Wilkes Level 1 Only (1) Level 2 Only (2) Level 1 & Level 2 (3) Total

2

3

1

1

1

1

3

2

2

3

2

3

3

3

3

3

3

3

3

3

2

3

3

3

3

3

3

3

3

3

2

3

3

3

3

1

3

2

3

1

2

3

3

1

3

3

2

3

2

3

2

3

3

3

3

3

2

2

2

3

3

3

3

3

3

3

2

3

2

3

3

3

3

3

3

3

3

1

2

1

3

1

1

3

1

3

1

2

3

3

1

3

3

3

3

1

1

2

3

3

3

3

3

3

3

3

1

2

3

3

3

3

1

3

2

3

3

2

3

3

3

3

3

3

3

3

3

2

3

3

1

1

1

2

1

2

3

3

2

3

3

3

3

3

3

2

3

2

1

3

2

3

3

3

3

3

3

1

2

3

3

3

3

3

3

2

3

3

2

3

2

3

3

3

3

2

3

3

3

3

3

3

2

3

2

1

3

3

2

3

3

3

3

3

1

3

1

1

2

1

3

3

3

3

3

3

3

3

2

3

3

3

3

2

3

3

3

1

2

3

2

1

1

1

2

1

1

2

3

2

3

2

2

2

3

2

3

3

3

3

3

3

3

3

3

2

3

3

3

3

1

3

3

3

1

2

3

3

3

3

3

3

3

2

3

2

3

3

3

3

3

3

3

3

3

2

3

3

2

2

2

3

2

3

2

3

3

3

3

3

0

2

1

8

4

6

2

32

0

4

3

2

4

3

3

3

1

4

9

5

6

0

0

30 27 21 13 19

21

24 14 17

32 32 32 32 29 29

26

30 24 26

Communities In Schools

27

Appendix D: CIS Affiliate Fundraising, Fiscal Year 2015

Affiliate Name

Amount of State Grant

Total Funds Raised or Received
(Outside State Grant)

Total Funds Raised or
Received/State Grant

Albany/Dougherty

$21,000

$145,245

$6.92

Athens/Clarke

21,000

973,665

46.37

Atlanta

21,000

1,678,516

79.93

Augusta/Richmond

66,0001

114,770

1.74

Baldwin

21,000

409,864

19.52

Barrow

9,0002

51,250

5.69

Ben Hill

21,000

180,518

8.60

Berrien

18,000

96,641

5.37

Burke

21,000

146,193

6.96

Candler

21,000

92,489

4.40

Carroll

21,000

160,544

7.64

Catoosa

21,000

507,566

24.17

Central Georgia

23,500

511,195

21.75

Cochran/Bleckley

21,000

288,144

13.72

Coweta

16,000

188,010

11.75

Dodge

21,000

414,088

19.72

Douglas

21,000

2,100,873

100.04

Glascock

21,000

476,705

22.70

Glynn

21,000

407,455

19.40

Hancock

18,000

74,271

4.13

Hart

21,000

75,000

3.57

Henry

16,000

506,489

31.66

Laurens

36,000

304,492

8.46

Marietta/Cobb

21,000

640,181

30.48

Rome/Floyd

21,000

131,787

6.28

Savannah/Chatham

11,275

82,559

7.32

Troup

18,000

136,007

7.56

Turner

21,000

140,929

6.71

Walton

21,000

239,617

11.41

Warren

21,000

366,100

17.43

Washington

21,000

116,364

5.54

Wilkes

21,000

235,887

11.23

Total

$693,775

$12,035,414

$17.35

Average

$21,680

$364,710

$18.07

1 The Augusta/Richmond affiliate received two payments in fiscal year 2015, one for $16,000 and the second for

$50,000 for expansion.

2 The Barrow affiliate closed mid-year, and therefore only received partial funding.

Source: CISGa financial data

The Performance Audit Division was established in 1971 to conduct in-depth reviews of state-funded programs. Our reviews determine if programs are meeting goals and objectives; measure program results and effectiveness; identify alternate methods to meet goals; evaluate efficiency of resource allocation; assess compliance with laws and regulations; and provide credible management information to decision-makers. For more information, contact
us at (404)656-2180 or visit our website at www.audits.ga.gov.