Skip to main content

Improving effectiveness of Individual Training Accounts: Long-term findings from an experimental evaluation of three service delivery models (Perez-Johnson et al., 2011)

  • Findings

    See findings section of this profile.

    Evidence Rating

    Not Rated

Review Guidelines

Absence of conflict of interest. 


Perez-Johnson, I., Moore, Q., & Santillano, R. (2011). Improving effectiveness of Individual Training Accounts: Long-term findings from an experimental evaluation of three service delivery models. Mathematica Policy Research


  • The study’s objective was to examine the implementation of the three models of Workforce Investment Act (WIA) funded Individual Training Account (ITA) service delivery. 
  • The study authors conducted an implementation evaluation using administrative data, survey data, qualitative reviews of participant case files, and data from interviews and observations conducted during three site visits. 
  • The study found that the Structured Choice model of service delivery was not implemented as intended, unlike the other two models, because counselors felt discomfort being directive in their interactions with clients. 
  • The authors provided limited details about their data collection and analysis methods, which impedes review. 
  • The embedded impact study was reviewed by CLEAR in September 2016. 

Intervention Examined

The Individual Training Accounts (ITA) Models

Features of the Intervention

  • Type of organization: Local workforce investment boards 
  • Location/setting: Multi-site in the United States 
  • Population served and scale: WIA-eligible, work-authorized adults; 7,920 participants 
  • Industry Focus: Not included 
  • Intervention Activities: Career exploration and counseling, training program selection support, supportive funding for training activities 
  • Organizational Partnerships: Proprietary training operators, community colleges and public universities 
  • Cost: $47,000 benefit to society per customer switching to Structured Choice Model of ITA service delivery 
  • Fidelity: Not included 

The passage of WIA in 1998 led to the launch of ITAs, which are used to cover training expenses for individuals. Customers generally have a great deal of control over how they spend this money and regional variation in service delivery has emerged in the years after the law’s passage. This DOL-funded program is intended to examine how different service delivery models operate in practice and impact customers. ITAs are available for work authorized adults in the United States, but training can be particularly helpful for dislocated workers looking to reskill themselves. To participate in the study, customers had to meet ITA eligibility requirements and consent to participate in the evaluation. ITA counseling services were delivered to customers through WIA One-Stop Centers and satellite offices in the respective workforce development areas included in the study beginning in 2001 and continuing through 2004 (and beyond, in some cases). Customers were randomly assigned to the Structured Choice, Guided Choice, and Maximum Choice groups. Each group differed in terms of required counselling and level of counselor control over the training program selection process, with Maximum Choice clients having no counseling requirements and Structured Choice clients having a good deal of constraint regarding the type of training they could enroll in and many counseling requirements. Local educational institutions and private businesses provided the training to customers. 

Features of the Study

The eight sites were selected through a grant competition. Each differed slightly in terms of local context and the service delivery model practiced by site staff before the implementation, with some sites leaning towards directive practices and others being more hands off with customers. The 7,920 people that entered into the study were randomly assigned to one of three service delivery model groups. The majority of participants were dislocated workers earning about $20,000 per year. The average age was 41 and a majority of participants had kids and were women. The Maximum Choice group had slightly more GED holders while the Structured Choice group had slightly more bachelor’s degree holders. The implementation study was conducted using observational and interview data with staff, participants, and partners during three sites visits. Additional administrative data and survey data were used to measure participation. Researchers also qualitatively reviewed some customer case files. The study authors do not describe data collection or analysis methods in detail, and they did not conduct a fidelity assessment. 

Study Sites

  • Phoenix, AZ 
  • Maricopa County, AZ 
  • Bridgeport, CT 
  • Jacksonville, FL 
  • Atlanta, GA 
  • Northeast Region, GA 
  • North Cook County, IL 
  • Charlotte, NC 


Intervention activities/services: 

  • The study found that ITA counselors generally help customers consider the relative costs and benefits of different training programs. 
  • The study found that each service delivery model was effective in getting customers to engage in some counseling and training, but the Structured Choice model was ultimately not implemented as designed because counselors did not feel comfortable being directive in their work for a number of reasons. 

Implementation challenges and solutions:

  • The study found that counselors were resistant to directive counseling as required by the Structured Choice model because they wanted to respect customer choices and did not always feel qualified to give directive advice to customers. Training and technical assistance to grantees did help boost adherence to some aspects of the Structured Choice model. 


  • The study found that the benefits of switching to a Structured Choice model yielded a public benefit of $47,000 because Structure Choice led to more completion of more fruitful training programs, in spite of higher upfront costs. 

Considerations for Interpreting the Findings

The authors provided limited details about their data collection and analysis methods. 

Reviewed by CLEAR

July 2023

Topic Area