As an interdisciplinary field of research, Needs Assessment “models” come from a variety of professions and applications. Below is a partial list of primary “models” in the field and a short description of each — each listed in alphabetical order by the priciple author of the “model”. If you have other “models” that we should add to the list, please email us using the Contact Us link to the left.

Note: Many of the model descriptions are taken from Watkins, R., Leigh, D., Platt, W., and Kaufman, R. (1998). Needs assessment: A digest, review, and comparison of needs assessment literature. Performance Improvement Journal, 37(7), 40–53. — and thus may not represent the most current works by several authors. Where possible we have provided current references and included recently contributed models.

Altschuld | Authur | Burton | Darraugh | Gilbert | Gordon | Hannum | Harless | Johnson | Kaufman | Mager | Ostroff | Robinson | Rossett | Rothwell | Rummler | Wedman | Witkin | Zemke

Authur’s Performance Evaluation Matrix
Authur, L. (1993). Improving Software Quality. New York: John Wiley and Sons.

Authur’s model for “investigating the problem” is provided within the context of software development and improvement.  The model suggests that problem assessment begins with an analysis of customers and suppliers.  The model focuses efforts around the requirements of end-users or those “who ultimately serve the external paying customer.”  After customer and supplier requirements are addressed, Arthur recommends that the processes be inspected to “identify defects created in the cycle, use data to identify the root causes of defects from each cycle, and improve the process each time you go through”.  The author furthers the development of the assessment model by adding a “Performance Evaluation Matrix” built around problem themes (or areas) as well as a guide for specifying problem themes.  While the model does not address societal or organizational results specifically, the quality focus on the requirements of both internal and external customers offers software developers tools for improving their products.  However, before improving software the developer should determine if the initial software development was derived from a results focused needs assessment.

Burton and Merrill’s Four Phase Model
Burton, J. & Merrill, P.  (1988) Needs Assessment: Goals, Needs and Priorities. In L. J. Briggs, K. L. Gustafson, M. H. Tillman, (Eds.), Instructional Design: Principles and Applications. Englewood Cliffs, NJ: Educational Technology Publications.

The four phase model for needs assessment proposed by Burton & Merrill is applicable for practitioners in a variety of disciplines, and recognizes both internal and external clients.  Additionally, this model focuses on “the application of needs assessment in the development of instructional materials at the level of a course” (pp. 25-26) and intentionally does not address societal and organizational results.  Doing so, the authors acknowledge, relies on the assumption that the acquisition of skills learned in a course will contribute to desired societal outcomes and organizational outputs.  This rolling-up – from inputs to process to products – may result in an internally efficient plan of operation where inputs and processes are linked to individual or small group payoffs and assumes, rather than ensures, that higher level results at the societal and organizational levels will follow.  Burton & Merrill’s model utilizes instructional goals, rather than measurable performance objectives, which are presumed to possess the specificity necessary for practical and reliable decision making, and to be accurate.

Darraugh’s Six Step Model
Darraugh, B. (1991).  It takes six. (six-step model for needs assessment). Training & Development Journal, v45, n3, p21(3)

The “six-step model for needs assessment” described by Darraugh in this short article appears to parallel Rossett’s (1987) Training Needs Assessment model in its determination of actuals, optimals, attitudes, and causes.  Darraugh does provide readers with fifteen questions that are recommended as essentials to the needs assessment process.  While these questions may provide an orientation to the purpose of a needs assessment, the six steps briefly described in the article are not linked to societal, organizational, nor individual accomplishments and provide little guidance in the procedures of conducting a useful needs assessment.

Gilbert’s Performance Matrix
Gilbert, T. (1978) Human Competence: Engineering Worthy Performance. New York: McGraw-Hill.

Gilbert provides an insightful discussion of why and how people differ in opinion concerning various solutions to performance discrepancies.  His focus is on process improvement – well before the popularity of quality management surfaced — as the means by which individual and organizational “performance measures” discrepancies can be closed most efficiently with available resources. In communicating his model, Gilbert provides two versions of a Performance Matrix: one full-scale and another truncated version for application. His proposed Performance Matrix does not specifically address how desired results can be assessed and linked at the societal, organizational, individual and small group levels.  Gilbert asserts that “at whatever level we ultimately wish to draw conclusions about performance, we must begin by identifying the context at a higher level” (pp.120).  However, his discussion of a holistic framework that starts at a “philosophical” level and cascades down to “tactical” and “logistical” levels is reduced in application to “simplified performance matrix” which does not extend beyond a conventional focus on one’s organization. Gilbert’s Performance Matrix does go beyond behavior to the achievement of internal accomplishments necessary for the closure of individual and organizational performance discrepancies.

Gordon’s Front-End Analysis Model
Gordon, S. (1994) Systematic Training Program Design: Maximizing Effectiveness and Minimizing Liability. Englewood Cliffs, NJ: Prentice Hall.

Gordon approaches needs assessment as an analysis activity, and does not so much identify and document gaps in results as discusses inputs and processes which the organization can employ when prescribing training and non-training solutions to its internal clients.  The Front-End Analysis Model rolls-down from desired individual results, though it does not formally address desired small group, organizational, or societal results.  Instead, it acts to identify resource unavailability and/or faulty processes for shortcomings in individual performance.  While the quality and availability of inputs and processes are essential for organizational success, Gordon does not acknowledge systemic errors which, according to W. Edward Deming, are the cause of 80% of performance discrepancies (in Clark, 1994; Triner, et. al., 1996).  Further, the model operates in a primarily reactive mode in which gaps between goals and actions (processes not performance) are shaped by end user’s preference and convenience, rather than data-based organizational requirements.

Hannum and Hansen’s Needs Analysis Model
Hannum, W. & Hansen, C. (1989) Instructional Systems Development in Large Organizations. Englewood Cliffs, N.J: Educational Technology Publications.

Hannum & Hansen’s support beginning with a top-down, societal needs assessment, but elect in their model to examine only gaps in results at the level of the individual performer and suggest that their model be solely used to document process inefficiencies.  Clearly, efficient processes are essential to the successful operation of any organization.  However, discrepancies that are not demonstrated to have a relationship with actual gaps in results may not necessarily yield effective results.  As such, Hannum & Hansen’s model functions in a “needs analysis” mode without the prior and (we suggest) essential phase of needs assessment which they choose to omit.  Nevertheless, the model is reasonably strong on research methods with guidelines for the collection of hard (independently verifiable) and soft (not independently verifiable) data which are applicable in a variety of settings.

Harlless’s Front-End Analysis Model
Harless, J. H. (1975). An ounce of analysis is worth a pound of objectives. Newnan, Georgia: Harless Performance Guild.

Harless concentrates on performance analysis and cause analysis as tools for finding the most efficient way to correct a performance problem. His focus on results (rather than solutions) aligns his work with the other predominant needs assessment models, and his focus on differentiating symptoms from performance problems can guide many useful assessments. Applications of the model use 13 questions to drive the analysis process; for instance: Do we have a performance problem? How will we know when the problem is solved? What are the alternate subclasses of solution?

Johnson’s Training Requirements Model
Johnson, D. (1996). Take two classes and call me in the morning: the case for training wellness. Hospital Materiel Management Quarterly, v17, n3, p21(8)

Johnson acknowledges that his needs assessment model is focused on “training requirements planning” and does not address the identification of performance problems that are likely candidates for training solutions.  But within this limited context and without a clear focus of performance, Johnson does offer several guidelines for determining the role of training within an organization.  The model suggests that the process begins the identification of management’s perception of training and the role of training within the organization.  Then in an approach that parallels other business processes, Johnson recommends the use of a “market survey” for the identification of desired training programs.  Based on the results of the “market survey” training requirements (including what knowledge, skills, and aptitudes to be taught) are defined and analyzed.  The “training requirements planning” continues through the evaluation of training with short-term and long-term feedback.

Kaufman’s Organizational Elements Model
Kaufman, R., Oakley-Brown, H., Watkins, R., and Leigh, D. (2003). Strategic planning for success: Aligning people, performance, and payoffs. San Francisco: Jossey-Bass.

Watkins, R. (2007). Performance By Design: The systematic selection, design, and development of performance technologies that produce useful results. Amherst, Ma:HRD Press and Silver Spring, Md:International Society for Performance Improvement.

Kaufman’s Organizational Elements Model (OEM) is the only needs assessment framework reviewed that formally addresses the linkages between every results focus (societal, organizational, small group and individual).  The OEM framework suggest that a needs assessment begin with a focus on societal results (referred to as Mega level results) and roll-down to organizational (Macro level) and individual or small group (Micro level) results.  Functional, both proactively and reactively, the OEM is a dynamic template that can be used to identify the impact of organizational action at all levels of results.  Though a direct distinction is not made between individuals and small groups, Kaufman’s outside-in approach to planning and inside-out approach to implementation yields findings which may be applied to data-based decision making.  Also included in his approach are several algorithms and toolkits which are intended to aid in the implementation of needs assessments.

Mager and Pipe’s Performance Analysis Model
Mager, R. F. & Pipe, P.  (1997)  Analyzing Performance Problems.  (3rd edit.) Atlanta, GA: The Center for Effective Performance, Inc.

The scope of Mager & Pipe’s model extends as far as the individual performer and, by extension, the organization itself.  Additionally, the model is fundamentally reactive, primarily intended for making only adjustments to the status quo at the level of individual and small group performance. Mager & Pipe hold that cost-benefit is the best approach to solution selection but do not explicitly detail a process by which alternate solutions are generated. Additionally, Mager & Pipe do not directly address formative evaluation and continuous improvement despite Quality Management’s emphasis on data-based decision making – a goal of the “Performance Analysis Flow Diagram”.

Ostroff and Ford’s Levels Perspective Model
Ostroff, C. & Ford, J.K (1989) Assessing Training Needs: Critical Levels of Analysis. In I. L. Goldstein, (Ed.), Training and Development in Organizations. San Francisco: Jossey-Bass Publishers.

Ostroff & Ford’s model is one of several models for needs assessment derived from McGehee & Thayer’s 1961 text Training in Business and Industry.  This text proposes that training requirements be analyzed according to three “content areas”: organizational, task, and person. Ostroff & Ford expand this framework by including a “levels” dimension (consisting of organizational, sub-unit, and individual) as well as an “application” dimension (incorporating the issues of conceptualization, operationalization, and interpretation).  Twenty-seven discreet analyses can be conducted based on similarities and differences of constructs between and across various “levels”.  In addition to the significant logistical difficulty of managing such an analysis, Ostroff & Ford provide no examples, tools, or methods regarding application of their model.  Further, their model approaches needs assessment as a rolling-up process, similar to Burton & Merrill, and does not link organizational means to ends and external consequences.  Without a doubt, the “Levels Perspective Model”is one of the most dense and theoretically conceived models in this study, however the model does not consider external impact.

Robinson and Robinson’s Performance Relationship Map
Robinson, D. G. & Robinson, J. C. (1995) Performance Consulting: Moving Beyond Training.  Berrett-Koehler.

Similar to Gordon’s “Front-End Analysis Model of Needs Assessment,” the Performance Relationship Map emphasizes both training and non-training solutions to individual and small group performance discrepancies.  The authors suggest that business goals, objectives, and strategies for a unit, division, department, or entire organization serve as the basis against which all performance requirements are to be anchored.  Applying a Performance Relationship Map includes involving a wide variety of stakeholders into defining performance problems, and bases solution selection in value-added for the individual performer and the organization as a whole, rather than for external clients and society.  Instead, key performance requirements are identified according to perceived importance of job practices and competencies versus the current skill level of performers.  While this approach is internally efficient it does not causally link individual performance to organizational or societal success.

Rossett’s Training Needs Assessment Model
Rossett, A. (1987)  Training Needs Assessment. Englewood Cliffs, NJ: Educational Technology Publishing Co.

Perhaps one of the most widely used training requirements analysis models currently in use by business and industry, Rossett’s reactive model seeks to lessen the gap between “optimal” and ‘actual” individual and small group performance.  The procedure for this activity involves responding to initiating performance discrepancies by first identifying the source of problems (causal analysis), then gathering opinions and ideas from primarily soft data sources using the largely qualitative methods of collection.  Rossett holds that findings are to be used for decision making, but does not demonstrate how individual and small group results ensure desired organizational and societal payoff.  Training Needs Assessment provides many useful tools which may be employed during needs analysis and can be quite helpful to practitioners new to data collection.

Rothwell and Kazanas’ Needs Assessment for Planning Model
Rothwell, W. J. & Kazanas, H.C. (1992) Mastering the Instructional Design Process: A Systematic Approach. San Francisco: Jossey-Bass Publishers.

Rothwell & Kazanas take great pains to establish operational definitions of the constructs associated with needs assessment.  Differentiating the planning process, from the tangible plan and the implementation of the needs assessment, the authors establish a model that is relatively consistent with the definitions used in this study.  Their guide to conducting a needs assessment functions well as a management and implementation plan and is strong concerning sampling and data collection.  As with Burton & Merrill’s “Four Phase Needs Assessment Model”, Rothwell & Kazanas’ model relies on two main assumptions.  First, the authors presuppose that intended results will necessarily follow from individual and small group application of skills.  Second, they assume that instructional goals possess the rigor necessary for decision-making, and will contribute to individual, small group, organizational and societal consequences.  Review of the Rothwell & Kazanas model may be useful in the development of a needs assessment management plan, though its utilization could lead an organization away from identified and required consequences of an intervention unless based in a rigorous results-based needs assessment.

Rummler and Brache’s Relationship Map
Rummler, G. A. & Brache, A. P.  (1990)  Improving performance: How to manage the white space on the organization chart.  San Francisco: Jossey-Bass Publishers.

The Relationship Map is a proposed improvement to the organizational maps common in many fields.  The major contribution of the Relationship Map is the provision of a horizontal systems perspective which includes recognition of internal and external clients, outputs delivered to customers outside the organization, and the flow of work which transforms inputs to products and outputs.  This mapping provides a systems view which can be useful to needs assessment initiatives.  The application of this tool with other models (especially the societal results focus of Kaufman’s OEM) may increase the utility of the needs assessment effort.  In a latter chapter of their text Rummler & Brache go on to explain how their Map can be used to design an efficient organizational structure by comparing current and desired processes.  While the authors discuss this activity from the organizational, process, and job/performer levels, their concern is primarily with resource availability and process efficiency.  Though external environmental factors are discussed as inputs to the organizational system, the model does not seem to address the reverse… the societal impact of organizational outputs.   Due to this disposition, the Relationship Map might infer that the results achieved by the organization will deliver beneficial results to external clients and society.

Wedman’s Performance Pyramid Model
Watkins, R., and Wedman, J. (2003). A process for aligning performance improvement resources and strategies. Performance Improvement Journal, 42(7), 9–17.

Watkins, R. and Leigh, D. (Eds.) (2009). The Handbook of Improving Performance in the Workplace – Volume 2: Selecting and Implementing Performance Interventions. San Francisco: Wiley-Jossey/Bass and Silver Spring, MD: ISPI.

John Wedman’s Performance Pyramid Model offers a comprehensive systems model, guiding performance improvement initiatives by building on both the foundational pyramid framework that is at the center and examining the supporting components that form a performance system—inputs, processes, and results. The result is a useful model for systemically accomplishing significant results, linking together the various systems (e.g., expectations & feedback; incentives, rewards & recognition; capacity; knowledge and skills; vision, mission & objectives) to provide a holistic performance systems perspective. You can use the Performance Pyramid to analyze current performance problems within your organization and to assess the various interventions as option making improvements.

Witkin and Altschuld’s Three Phase Needs Assessment Model
Altschuld, J. W. (2010). The Needs Assessment KIT. (ed.). Thousand Oaks, CA: SAGE Publications. [5 volume series with individual titles listed in the Books section of this site]

Witkin, B. R. & Altschuld, J. W. (1995) Planning and Conducting Needs Assessments: A Practical Guide.  Thousand Oaks, CA. Sage Publications.

The Three-Phase Model proposed by Witkin & Altschuld is actually an analysis, assessment and action plan framework embedded within one method.  These processes occur over three phases: preassessment (exploration), assessment (data gathering), and postassessment (utilization).  As the first and last phase go beyond the scope of needs assessment as defined here, the model is actually a plan for problem identification and resolution.  As a reactive model, the authors describe needs assessment as a means by which to cope with current and future problems and constraints via risk identification techniques, rather than anticipating and creating future opportunities.  This philosophical underpinning is significant because it contributes to the absence of formal concern with societal payoffs of organizational action.  Rather, the focus of Witkin & Altschuld’s model tends to be on process improvement and the achievement of the organization’s goals for individuals and small groups.  Organizational payoff is assumed to flow directly from the accomplishment of these goals with external clients functioning only in the context of demand and consumption of organizational outputs.

Zemke and Kramlinger’s Figuring Things Out Model
Zemke, R. and Kramlinger, T. (1985). Figuring Things Out: A trainers guide to needs and task analysis. Reading, MA: Addison-Wesley.

The Zemke and Kramlinger’s Figuring Things Out (FTO) model is composed of factors that affect performance in an organization.  The factors are detailed in three primary arenas: performer, local performance environment, and organizational climate.  The authors suggest that an FTO study begins by capturing the organizational climate and culture data, including how the mission and goals of the organization are translated into actions at the functional/operational levels.  Though from this point on, the FTO model mirrors many conventional task analyses.  Comparable to Rossett’s Training Needs Assessment, Zemke and Kramlinger devote most of their effort to providing tools and guides for collecting both hard and soft data within an organization. The FTO model, like many of the needs assessment models examined in this digest, infers that the results achieved by the performer will be beneficial to internal and external clients as well as society.

A situation in any organization that would require an integrated approach evaluation would be an analysis to determine if the front desk assistant requires training to a greater level of depth and detail than what is currently offered. This would be based on the main fact that the front desk assistant makes a high percentage of mistakes. A contributing fact is that when we evaluate the last three employees that have held this position, there appears to be a positive relationship with increased error rates in this particular assistant’s position after a period of approximately six months have passed. Each person in this position is then given more work to complete, and the error rates increase. This leads us to believe that greater training is needed for the tasks that are delegated after the initial six-month period has ended, even though the tasks require no special or specific level of education, degrees, or licenses of any type to complete.
The integrated approach evaluation to address this situation would utilize resources and assessments from several different areas. We would use job assessments, job evaluations, HR data regarding the position, and market research data from the same industry to compile a comprehensive report for our initial evaluation and analysis. We can then determine the best practices for our remaining steps to address our specific situation.
Formative evaluation methods examine the actual delivery of the training currently in use for the position. A formative method is based on the actual processes, programs, and methodology used in the training currently being delivered. A summative method examines the outputs of the processes. This evaluation method also addresses the cost vs. benefit issues based on the current training methodologies, programs, and processes being used. The summative method also takes multiple instances into consideration, and not just isolated incidents. It is more of a comprehensive method that is used in the organization. A confirmative evaluation method typically entails a comprehensive analysis of all related data. We would look at the training programs from a few standpoints. We would evaluate the program based on how it’s being delivered and the content of the program and processes. We would then continually monitor the program to determine if circumstances have changed, which may alter the cost/benefit and/or effectiveness of the training program. We would also determine how this information transfers to the actual business environment — we analyze how practical the information is in our specific business setting.
If I were to combine all methods for the situation described, I would do so in the following manner:
In using a formative method, I would conduct a thorough needs assessment and process evaluation of the training. I would combine this with summative methodologies that include performing a thorough cost – benefit analysis based on the data that I have generated and compiled thus far. I would also use a summative method to determine if the outcomes of the training are effective based on the current processes being employed. As to this point, I can safely say that they are not, or else our employee error rates in this position would be lower. Therefore, I can reasonably be assured that the training needs to be redeveloped in this area. I would combine the formative and summative methods with a confirmative method by continually evaluating and monitoring the training that is developed for this position. I would also use a confirmative method to analyze how applicable the training being offered is to the actual position. Based on the information we have gained, we can determine that the training is not applicable to the actual business setting because the error rates are high.

Posted in Uncategorized

Leave a Reply