Unless mandated by higher, external powers, organizations do not undertake “job analyses” without reason. The reason is typically a desire to remove a current or pending problem. These problems (evident to management) may be external (market pressures, technology shifts, competition, etc), internal (heavy turnover, loss of product quality, inability to fill vacancies, company finances, etc), or a combination of the two (employee dissatisfaction due to comparison with external references). In any, event, it is assumed that any study is undertaken with the expectation of intervention to follow.
In designing any change or intervention, trade-offs will be made to achieve an acceptable result. The training and discipline of the design analyst will determine which of the available domains of activity will be considered as primary or anchored, and which domains are secondary and therefore primary candidates for making concessions.
Various disciplines exist in spearheading interventions in the workplace. The one I am personally most familiar with is dubbed “Occupational Analysis”. As with many terms, this is really a literary example of synecdoche. This is like denoting the modern architectural & building industry as the “concrete pouring business.” In modern times, we have taken to using the term “Job Structuring Technology” or JST for short. Occupational analysis, typically task-based, occupational analysis is a tool that is used to establish a solid foundation for the strategic decisions to follow. These decisions may be anywhere in the personnel life cycle (recruitment, selection, assignment/classification, training, promotion, tiger (ad-hoc) teams, mentoring, and out-placement). In addition, these decisions may address organization issues as well, anywhere in the organizational life cycle (conceptualization, formulation, organization, resource allocation, staffing, launching, integrating, stabilizing, growing, downsizing, disengaging, liquidating, and ceasing operation). OA does not claim to tackle external issues, it only provides a clear vision of the internal resources and capabilities of the organization’s workforce.
Because of its roots in the military, “occupational analysts” are trained to respect and maintain the trust of the majority of workers who make products possible. Occupational analysts are taught that the organization is working today because of, not in spite of its employees. Occupational analysts believe that the training & experience investment made, to date, in a worker’s life is very valuable. Further, the worker should have the expectation that, to the maximum extent possible, work will be modified to build upon his or her current capabilities and provide a career ladder that encourages, guides, and promotes growth in responsibility and position within the organization.
Occupational analysis is the antithesis of Time-and-Motion interventions. In cartoon form, “job design” specialists of the time-and-motion era give rise to the visions of Metropolis, the silent film with its two-class society. The desires of the workers are not at all a consideration in maintaining productivity. Neither is Occupational Analysis (OA) the opposite extreme – anarchy where every worker chooses to continue doing the same old thing forever. OA is not a communist system nor necessarily a Theory Y endeavor. No, OA is not a Utopian daydream, but a scientific method that has its roots in the real-world, high-tech industry – the United States Air Force where failure to perform the mission is not an option.
Because of it origins, task-anchored Occupational Analysis has been criticized by academic sources as being “without theory.” The military focus was on ensuring that TRAINING IS ON TARGET. They brought in high school graduates and had to teach them to be jet engine mechanics. They brought in college graduates and had to educate them to be jet pilots and navigators. Billions of dollars were being spent annually on training and the aircraft kept changing over time. In addition to training the backbone of the Air Force, many support and sustainment forces had to be trained, posted, and rotated to keep an active and viable Air Force.
It was recognized early on that numerous management issues would arise – recruitment, selection, assignment/classification, promotion, etc. It was also noted that academics like to work from theoretic models that give meaning to their data – unfortunately, different applications fall under the domain of different theories and data collection from one effort was not compatible with data from other related efforts. The decision was made to collect data at the lowest, unambiguous level so it could not only be reconfigured later for any application, but all applications could be linked back to a common framework. I attribute this quintessential insight to my mentor, Dr. Raymond E. Christal and his staff of personnel psychologists on staff in 1959 following their review of the then available systems in the U.S. and allied governments. In other words, pouring the concrete was without theory, but every building would have an architect with a clear vision of the future goal.
Because funds were available to the military, large-scale tools were written to handle very large data sets, ridiculously large regression problems, and unwieldy task inventories of thousands of items. The Air Force made good use of these software tools for Manpower, Personnel, and Training (MPT) decision-making. The most notable of these tools is an package known as CODAP – The Comprehensive Occupational Data Analysis Programs. CODAP is one of the Air Force’s success stories in Technology Transfer and is still in use today, after thirty five years.
For the first many years, few organizations could even afford the computer power needed to conduct scientific OA. Some criticisms may have been from jealously – it was like a junior college denying the existence of sub-atomic particles because they couldn’t fund an atom smasher. Today, however, the power of CODAP is available on a laptop. This is not a paired-down version, but a system with capabilities far in excess of what the Air Force used for its first 10 years. The software package is called “atCODAP” meaning “anchored to CODAP, but not limited by it.” atCODAP can handle task lists, functional job requirement lists, competency lists, knowledge lists, etc -- no longer tied to the "task" as the only backbone in an analysis. atCODAP evolved the way it did for a reason. In 1979 I was sent to the Federal Laboratory Technology Transfer Exposition to give away the mainframe CODAP system to state and local governments. Although we had successfully transferred the CODAP technology to Universities, Non-profit Institutions, and other military services (both US and abroad), the state and local governments gave me numerous reasons why they would need changes to CODAP before they could use it. The Air Force declined to make any changes and would only release it on an “as is” basis. We had few takers. Users wanted more than training priorities on task lists. atCODAP can handle up to 99 concurrent lists (tasks, competencies, knowledge, skills, abilities, tools, etc.) in a single study, perform hierarchical clustering on any or all of them and provide inter-rater reliability results for any factor.
I left civil service in 1982 and founded Sensible Systems, Inc. Originally we started out writing software to the support the mainframe CODAP system – automated survey software, improved mapping software for linking tasks to training or testing, simplified report formatting for personnel officers, etc. By 1987 we had not only extended the software to support mainframe CODAP, we had programmed the essential core of the CODAP capabilities and extended output formats to facilitate data transfer into other operational programs via Microsoft Office products like ACCESS, EXCEL, and PowerPoint.
The atCODAP system has been used by city and county governments. While that was our target market, most of work has come from federal clients – those who HAD the mainframe CODAP system on their own computers. We can produce products faster, cheaper, better, and with better links to ultimate users (in Excel and Access formats, not to mention HTML reports for the web). But even more than technical excellence, Sensible Systems, Inc. has always been willing to upgrade atCODAP to meet new and exciting needs. In addition, Sensible has underwritten the cost of projects for graduate students over the years. These projects include diverse topics such as “Job Design from Hierarchical Clustering”, “A Point-Based Factor System for Compensation”, “Community College Officers in the State of Texas”, and “Educational Needs for Systems Engineers in the 21st Century.”
As mentioned earlier, the OA/CODAP acts as a foundation for many possible human resource or organizational activities. Job design lies at the intersection of human resource and organizational concerns. The OA/CODAP perspective is to find out “what is” in terms of the current work and worker. This is done at the task and tool level at a minimum and may include attitudinal data and competency requirements. The CODAP methodology discusses from whom data should be collected – “what is done” from incumbents, and “how important/difficult to train” from supervisors/trainers. Once the data are collected, any number of scenarios exist for using the data. Historically, high-level managers would review the data and bring in Subject Matter Experts (SMEs) to advise them in policy making. On occasion, an SME would dispute the data as wrong and irrelevant. Because the data are linked back to people and locations, the SME's mistaken impression could be corrected, in short order, directly from the source. An SME's lack of contact with the real-world Air Force was no longer a source for poor policy.
In recent times, a new methodology was developed in the last days of the Air Force Human Resources Laboratory's (AFHRL's) long history in personnel management issues. The new Job Structuring Technology was designed to start with “what is” data and realign job boundaries based on reference to an external academic/training outline. This outline represents where and how people are trained to acquire proficiency. The more “pure” a job, the easier it is to focus training for new personnel. The “what is” data also provides information about what tasks are currently performed together (i.e., by an individual) The extent to which the current workforce is properly allocated (people have assignments that have common competencies), will leave portions of work together in the redesign. The JST system uses task learning difficulty ratings to “balance” workload and uses the current “months on the job” to strategically map career paths. When the JST system is done running, it has identified a series of realistic positions, linked to available training, and linked to a career path of increasing responsibility.
Issues of workload, however, represents a different problem. Once jobs are defined and flow is balanced, stresses and strains on workers can still mount if the workload is excessive. Last year we did a workload study for Atomic Energy Inspectors which included specialized measures for REAL HOURs per month on functional job areas and relative time spent on actual tasks. Rather than dealing from a sampling of tasks, we dealt with the person's entire job -- which, in case, included extensive, non-productive travel requirements. Depending on the job-type found in the cluster merger diagram, some people could spend no more than 20% on their core job while 80% was spent in logistics. Other people, with the same job TITLE, had 70% of their time available for core responsibilities and only 30% was spent in logistics. These types of differences are lost on analysts who believe that a job title is anything more than a convenience for the payroll clerk. A job is what a person is called upon to do, not what a person is called on paper. Historically, the Air Force has done studies on this type of work breakdown analysis in the medical area, particularly studying the role of nurses versus medical technicians (McFarland). Similar techniques are used in evaluating "equal pay" for "equal work." In another project for the United Nations Office of Project Services (UNOPS) we evaluated pay equity for globally dispersed Project Managers and associated personnel. Work was equated at the task level, across an inventory of 2600+ tasks and 300+ functional job areas. Again, real hours were collected for the functional job areas and relative time was collected for tasks. It was demonstrated, that regardless of job title, regional office, or currency paid in, there was equal pay for equal work.
Over the years, many projects have been designed, minor additional data collected, and analyzed in the context of an appropriate theoretical model. There is no over-all model save for the scientific method. The CODAP data define "what is" (empirically defined work requirements and personnel subgroups - along with routine job satisfaction data). While most routine projects simply defined training needs for specific career fields, research projects entailed the application of theoretical structure and the collection of data to address current issues -- usually executed by universities such as Purdue in the early days, and later by on-staff PhD personnel. I have the final reports on most of these projects in my library and can scan them into PDF format if required. If you go to the Master CODAP Bibliography linked below you can find references to a FEW of the reports along with retrieval numbers for and links to the National & Defense Information Services...
Link Papers Available Online, IJOA.ORG/Papers and Master CODAP Bibliography
[24 August 2004 Note: Since this paper was written, a new non-profit organization has been formed to put the entire technical report library of the Air Force Human Resources Laboratory onto a free-access web site. This growing site may be found at:
<http://www.icodap.org/papers>
This was written in response to two emails:
X-Mailer: Novell GroupWise Internet Agent 6.0.1
Date: Tue, 30 Jul 2002 11:43:37 -0500
From: "James Gilbreath" <gilbreaj@ipfw.edu>
To: <johnnyw@codap.com>
Subject: Need your help - article manuscript
X-Loop-Detect: 1
Hello there. I am writing an article, part of which deals with job
re-design. I would like to make sure my readers know about people in
your area of expertise (i.e., know that your approach exists). My
problem is that I don't know what to call people who do what you do.
Here's an excerpt from the manuscript that may explain better than I am:
There are, of course, other approaches to job design. Industrial
engineers, human factors engineers, occupational analysts, and
ergonomists bring their own perspectives and goals to job-design
interventions (Campion, 1988). Although their recommendations sometimes
clash, they all can make positive contributions. An interdisciplinary
approach to job design, therefore, is preferable (Campion).
I put in the term occupational analysts, but am not sure that's right. I
sure will appreciate your help on this!
Thank you.
Brad Gilbreath, Ph.D.
Division of Organizational Leadership & Supervision
288 Neff Hall
Indiana University-Purdue University Fort Wayne
2101 E. Coliseum Boulevard
Fort Wayne, Indiana 46805-1499
(260) 481-6496
Fax: (260) 481-6417
Date: Fri, 26 Jul 2002 16:18:23 -0500
From: Daniel Muraida <DMuraida@verdict.uthscsa.edu>
Subject: Functional Job Analysis proposal
To: "'johnnyw@codap.com'" <johnnyw@codap.com>
X-Mailer: Internet Mail Service (5.5.2653.19)
X-Loop-Detect: 1
Hi Johnny! Congratulations on your brand new grandchild, barely two weeks
old!
We have gotten feedback on our proposal entitled "Evidence-based Approaches
to Primary Care Staffing: Job Analysis. As I'm sure you remember, we are
attempting to apply Functional Job Analysis to the the problem of optimal
task allocation in primary medical care settings. We we criticized for not
having a solid "conceptual framework for staffing issues and workforce
availability". Our proposal used the works of management theorists such as
Senge and Nonaka to structure our approach to the problem. These writings
were selected by previous members of the research team who are no longer
with us. Much of what we were using as our theoretical base stemmed from
case studies. We need a conceptual framework that is closely aligned with
empirical research. Did a theoretical framework emerge in parallel with the
CODAP technology (Perhaps I have seen it in the literature not recognized it
as such). Does the CODAP literature address the mechanisms that underlie
job performance? Lastly, isn't there a report on the application of CODAP
in an Air Force Medical setting?
We have been directed to resubmit the proposal after addressing the
conceptual framework isue and correcting some other problems. I asked all
these questions because in speaking to my boss on Wednesday about the
rewrite of the proposal she agreed that CODAP should figure more prominently
this time, and that we should point out any medical applications of CODAP.
We are going to start the revision process next week since we have a 1
November deadline. I hope you can provide some help on these questions
because I still see a strong potential for us to work together.
Dan
P.S. I enclosed the following reaction to CODAP from Dr. Frank Moore (a
senior investigator on this project) when I shared your responses to my
questions last April:
Wow!, I feel like I have just been on "fast forward". I am more and more
certain that we should link with this group and the CODAP technology. There
does not seem to be any incompatibility with task data gathered using FJA
and the manipulation/applications developed by Weissmuller, et.al.
Thanks for following up on this. It will strengthen
the proposal.
Frank I. Moore
VA VERDICT: 210.617.5300 x 6637; Fax: 617.5234
SCHOOL OF PUB. HEALTH: 210.567.5930; Fax:567.5942
HOME: 210.698.2838; Fax: 698.5441; Mobile:210.313.6406
E-mail: mooref@uthscsa.edu