What Is Training Program Evaluation?

Logic Models

Logic models are very easy tools to help you plan and evaluate your training program. A logic model provides a representation of a "theory of change" (if...then) that clearly aligns the training inputs and activities to the outputs and results.  Logic models are created in the planning phase and can be completed for programs, initiatives, individual courses, events and activities. There is an important distinction between training program evaluation and training evaluation. Training program evaluation examines the entire program including the training component, while training evaluation is assessment of the training course, process, or component within a program. For example, the Senior Executive Service Candidate Development Program (SESCDP) involves a recruitment, selection and certification process. These processes, in addition to the required 80 hours of training and 4-month developmental assignment, should also be consistently monitored and evaluated to determine progress and areas of improvement.  All processes within the program contribute to program success or failure, yet all processes are not "training". Knowledge, understanding and skill in program evaluation in necessary to effectively evaluate training and development programs. 

Contact us at SESDevelopment@opm.gov if you would like training on how to use the OPM Training Evaluation Field Guide and/or Logic Models in your agency.

For more detailed information on logic models, you can watch this video or take EPA's FREE Logic Modeling Course.

Agencies are required to evaluate their training programs annually to determine how well such plans and programs contribute to mission accomplishment and meet organizational performance goals (5 CFR 410.202).  The law authorizes OPM to require Federal agencies to report training data (Example: Sample Agency Training Report FY2010).  An important part of the evaluation process involves consideration of training costs, and other elements not directly addressed in the typical evaluation.  Agencies must track and report accurate training data on all completed training events to OPM as prescribed by the Final Rule on Training Reporting Requirements, published on May 17, 2006, in the Federal Register, The Federal Workforce Flexibility Act of 2004 (P.L. 108-411).

The Training Evaluation Field Guide uses the Kirkpatrick Model of evaluation as the basis of instruction.  However, there are other viable, well researched and more appropriate models available to evaluate your training program, not just the training component of your program. Agencies should use the model that more closely meets their needs.

• Daniel Stufflebeam's CIPP Model (Context, Input, Process, Product)The CIPP Model for evaluation is a comprehensive framework for guiding formative and summative evaluations of programs, projects, personnel, products, institutions, and systems. This model was introduced by Daniel Stufflebeam in 1966 to guide mandated evaluations of U.S. federally funded projects.http://srmo.sagepub.com/view/encyclopedia-of-evaluation/n82.xml

• Robert Stake's Responsive Evaluation Model: Robert Stake (1975) coined the term responsive evaluation. Responsive evaluation distinguishes four generations in the historical development of evaluation: measurement, description, judgment and negotiation. ‘Measurement’ includes the collection of quantitative data.http://mailer.fsu.edu/~sullivan/SEA_Newsletter/Responsive_Evaluation.pdf

• Robert Stake's Countenance Model: This model focuses on description and judgment. Stake wrote that greater emphasis should be placed on description, and that judgment was actually the collection of data. Stake wrote about connections in education between antecedents, transactions and contingencies (outcomes). He also noted connections between intentions and observations, which he called congruence. Stake developed matrices for the notation of data for the evaluation. Data is collected through these matrices.http://ged550.wikispaces.com/Robert+Stake's+Countenance+Model

• Kaufman's Five Levels of Evaluation: Modeled after University of Wisconsin professor Donald Kirkpatrick's four level evaluation method, Roger Kaufman's theory applies five levels. It is designed to evaluate a program from the trainee's perspective and assess the possible impacts implementing a new training program may have on the client and society.
  http://www.ehow.com/info_8582553_kaufmans-five-levels-evaluation.html#ixzz2ePyjynYl 

• CIRO (Context, Input, Reaction, Outcome): This model is based on four stages or types of evaluation – context, input, reactions and outcome – and is underpinned by a set of three questions, which, according to the authors, the trainer should always bear in mind.http://toolkit.goodpractice.com/mdt/resources/development-cycle/training-cycle-evaluation/the-ciro-model-of-evaluation

• PERT (Program Evaluation and Review Technique:The Program (or Project) Evaluation and Review Technique, commonly abbreviated PERT, is a statistical tool, used in project management that is designed to analyze and represent the tasks involved in completing a givenproject. First developed by the U.S. Navy in the 1950s, it is commonly used in conjunction with the critical path method (CPM).http://en.wikipedia.org/wiki/Program_Evaluation_and_Review_Technique

• Michael Scriven's Goal-Free Evaluation Approach: This approach is premised on the assumption that an evaluation should establish the value of a program by examining what it is doing rather than what it is trying to do.http://www.click4it.org/index.php/Goal-Free_Evaluation

• Provus' Discrepancy Model: Developed in 1969 this model provides information for program assessment and program improvement. Under this model evaluation is defined as the comparison of an actual performance to the desired standard. http://www.slideshare.net/AgroSmekad/provuss-discrepency-evaluation-model

• Illuminative Evaluation Model: An illuminative evaluation is a custom built research strategy which lacks in formal statements of objectives, avoids (but does not exclude) statistical procedures, employs subjective methods, and is primarily interested in the informing function of an evaluation, rather than the more usual inspectoral or grading functions of an evaluation.http://www.iisd.org/casl/CASLGuide/EvalModel.htm

Logic Model Template for Supervisory & Managerial Training:   This logic model provides agencies with a systematic and visual aid to identify a training program’s components: inputs, outputs, outcomes and evaluation methodologies, and illustrates the relationship among them.  The logic model has been prepopulated with Federal supervisory and managerial curriculum-specific examples for agencies to reference when creating their agency-specific logic model.  Once completed, the logic model will provide stakeholders with a roadmap that describes the sequence of related events.  Stakeholders will also have a clear understanding of the curriculum goals and evaluation. 

Note:  The Federal Managerial Curriculum Logic Model components have been color-coded to map to each of the stages included in the OPM Training Field Guide Supplement. 

Certifications: Both Kirkpatrick and Phillips offer training evaluation certificates although a certified evaluator is not necessary to evaluate the effectiveness of agency training. The 2011 OPM Training Evaluation Field Guide, books and/or training courses on performance measurement, program evaluation, and training evaluation should provide enough information to successfully evaluate your agency training.

Associations: The American Evaluation Association (AEA) is an international professional association of evaluators devoted to the application of many other forms of evaluation. Evaluation involves assessing the strengths and weaknesses of programs, policies, personnel, products, and organizations to improve their effectiveness. The AEA has approximately 5500 members representing all 50 states in the US as well as over 60 foreign countries.