Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 114
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 114
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
6
Procedures Used to Produce the Fourth
Edition Dictionary of Occupational Titles
Because the Dictionary of Occupational Titles is a dictionary and a
reference manual, and because it is widely used and relied on to provide
comprehensive, authoritative information about occupations, it is especially
important that the information it contains be reliable and accurate. The
comprehensiveness, reliability, and accuracy of the DOT are in large part a
function of the data collection and analysis procedures used to produce it. This
chapter describes these procedures step by step, from the initial selection of
industries for study to the final stage of writing definitions for the occupational
titles that appear in the published volume. This description, along with the more
technical evaluation in chapter 7 of the implications of these procedures for the
quality and characteristics of data contained in the DOT, is intended to provide
information that will enable users of the DOT to make informed judgments about
its value.
In the course of our analysis it became clear that the production of the DOT
is seriously underdocumented. Because of the lack of published technical
information or documentation of procedures, the description in this chapter
draws heavily on information gathered during interviews with staff at the
national office and at seven field centers (Arizona, California, Michigan,
Missouri, North Carolina, New York, and Texas) as well as on
Pamela S.Cain had primary responsibility for the preparation of this chapter.
OCR for page 115
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 115
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
information found in instructional manuals developed primarily for internal use,
the Handbook for Analyzing Jobs (U.S. Department of Labor, 1972) and the
Definition Writer's Manual (U.S. Department of Labor, 1974).
HISTORICAL BACKGROUND
The procedures used to produce the fourth edition DOT were originally
developed by two different research programs dating from 1934 and 1949. The
first program of occupational research was initiated in response to the
depression. At that time the large number of unemployed, qualified workers
made job matching on the basis of work performed, by comparing an applicant's
work history with a job order, relatively easy. Occupational research at this
early stage hence focused primarily on job content. The research begun in 1934
formulated major principles of job analysis methodology for the DOT and set in
place sampling techniques and on-site procedures for the study of business
establishments that are still used in the program today.
World War II and postwar economic recovery prompted an influx into the
labor force of workers who had little or no prior labor force experience or up-to-
date job skills. Increasingly, it was necessary to match jobs to workers on the
basis of an applicant's potential to perform a given job rather than his or her
demonstrated expertise. In 1949 the functional occupational research project
was begun to investigate systematically various dimensions of worker potential.
This research, which culminated in the third edition DOT (1965), introduced
worker functions in relation to data, people, and things into the DOT
classification scheme, standardized definition writing, and developed various
scales for rating worker traits. Since publication of the third edition DOT in 1965,
few, if any, modifications appear to have been made in the basic methodology
of the occupational analysis program.
SAMPLING FOR THE DOT
Sampling for the fourth edition DOT was by no means straightforward. The
sampling universe was all jobs in the national economy. Conceptually, the
economy was categorized by occupations. For purposes of collecting data about
occupations, however, the economy was categorized by industries; that is,
industries were the basic units by which assignments were made to the field
centers. Once field centers received their industry assignments, they were
responsible for obtaining complete coverage of the jobs within their assigned
industries. In order to study jobs, however, the business
OCR for page 116
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 116
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
establishments in which jobs are found had to be located. Only after this had
been done were jobs finally selected and analyzed.
This multistage selection process identified an increasingly disaggregated
unit of analysis; the process then shifted direction: for purposes of compiling
the DOT, data about jobs were aggregated to form occupations. Occupations in
the DOT are not intended to be either firm or region specific, and in some cases
not even industry specific. Rather, the descriptions are intended to reflect the
occupation as it typically occurs in the national economy, although what is
meant by “typical” is not explicitly defined. The various steps involved in the
creation of the fourth edition DOT, from industry assignment to occupational
composite, are described in the subsequent sections of this chapter.
ASSIGNMENT OF INDUSTRIES TO FIELD CENTERS
Shortly after the completion of the third edition DOT, industry assignments
were made for the fourth edition. Most centers were assigned the same set of
industries for which they were responsible in the third edition. Although there is
no documentation of how assignments were made, in matching centers and
industries, national office staff appear to have relied on a combination of
common sense and general knowledge as well as on their own experience in
occupational analysis.
It was impossible to determine conclusively the basis or criteria of industry
assignment. When national office staff were queried as to how assignments had
been made, responses were vague. North Carolina, for example, was said to be
the “logical place for machine textiles and wood furniture”; similarly,
Washington was said to be “the natural place” for aeronautics, given the
location of Boeing in Seattle; while it was “pretty obvious that Texas doesn't
have logging.”
For relatively widespread or highly visible industries the rationale for
assignment was of this common sense variety. For other, more obscure,
industries the assignment of which was not so obvious, various documents were
consulted. The annually published Thomas Industrial Register, for example,
was consulted frequently for this purpose. It was said to be of limited
usefulness, however, because it lists a company's home office rather than its
production sites. In addition, various state and local industrial or manufacturing
directories were consulted on a case-by-case basis, as were assorted federally
produced censuses of manufacturers and businesses. Generally, industries were
assigned to centers in whose state or region they were thought likely, or
sometimes actually determined, to be located.
Large industries, such as banking and financial institutions, were usually
assigned to more than one field center to ensure adequate coverage and to
OCR for page 117
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 117
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
detect regional variations in jobs. Groups of related occupations that are not
industry specific (e.g., the “any industry” designation) were assigned en bloc to
a given center. Assignments were made using the 229 industry designations of
the third edition. The number of industries assigned per field center is given in
Table 6-1. Fourteen industries (6 percent of the total) were unassigned, and no
new job analyses were collected from them; 40 percent were assigned to one
field center; and the remaining 54 percent were assigned to two or more field
centers. On average, each center was responsible for 42 industries, ranging from
11 for the Arizona special project to 84 for California. Certain field centers
appear to have had greater responsibility for industry coverage than others.
California, for example, was assigned the largest number of industries and was
solely responsible for 9 percent of the 232 industries in the fourth edition DOT
industry designation. New York had the second largest number of industries and
was solely responsible for 8 percent of the total. Washington was assigned a
smaller number of industries and had sole responsibility for
TABLE 6-1 Field Center Industry Assignments
Centera Number of Industries Number Uniquely Assigned
Assigned
Arizona (2) 11 1
California (16) 84 20
Florida (8) 28 0
Michigan (12) 35 17
Missouri (12) 62 13
New York (18) 58 18
North Carolina (21) 39 12
Texas (7) 25 2
Utah (9) 45 4
Washington (12) 44 1
Wisconsin (12) 33 5
Unassigned 14 —
478b 93
TOTAL
aNumber
in parentheses following field center is number of analysts on staff.
bTotal
does not equal 232 because some industries were assigned to more than one field center.
SOURCE: Booz, Allen & Hamilton, Inc. (1979: Exhibit IV-3).
OCR for page 118
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 118
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
only one industry. Since these three field centers have approximately the same
number of analysts, the reason for this apparently uneven distribution of
responsibility is unclear.
ESTABLISHMENT SELECTION
Once field centers received their assignments, they independently
determined how best to fulfill them. Although the Handbook for Analyzing Jobs
(U.S. Department of Labor, 1972) provides relatively detailed information on
how to conduct a job analysis study within a firm, it contains virtually no
guidance on how to go about studying an industry per se or on how to select
firms within an industry. In most field centers each analyst was given sole or
lead responsibility for a set of industries. The number assigned per analyst
varied by the size of the industries involved and by other factors such as the size
of the field center and the extent of its involvement in activities other than job
analysis (see chapter 5 for a discussion of field center organization and
activities).
Analysts pursued their assignments relatively autonomously. There are
indications that field centers varied in the degree of supervision or direction
given to analysts. Several centers instituted informal procedures whereby
analysts submitted a general study design to the center supervisor or
supervisory analyst before proceeding, but this was not required by the national
office. Analysts typically began by doing background research on the industry
in question in order to acquaint themselves with its processes, products, and
jobs. This usually involved library research and reference to the appropriate
trade publications, the Standard Industrial Classification (U.S. Department of
Commerce, 1972), the Occupational Outlook Handbook (U.S. Department of
Labor, 1978a), and others. Some analysts also consulted old job analysis
schedules and occupational code requests as background material. The Standard
Industrial Classification appears to have been especially useful at this point.
Because it is much more detailed than the DOT's industry designations, it
provided a means by which analysts could determine an industry's subgroupings
and hence develop a framework for pursuing their assignment.
As soon as analysts were satisfied that they were familiar with the industry
and the types of jobs found in it, they selected establishments that seemed likely
to have jobs typical of the industry. In locating establishments, analysts relied
on various sources: industrial registers (e.g., Thomas; Dun & Bradstreet), the
classified sections of telephone directories, ES-202 forms (submitted to state
Employment Service offices by companies that contribute to unemployment
insurance), directories of
OCR for page 119
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 119
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
trade and professional associations, indexes of local manufacturers, and
publications of the Better Business Bureau and the Chamber of Commerce.
Many of these publications contain a good deal of information on the firms
available for study in the field center's area, often listing the firm's SIC code,
address, the names of company officers, the number of employees (sometimes
disaggregated by sex), and the products manufactured.
Despite the absence of guidelines, there was substantial uniformity among
the field centers on the criteria used for the selection of establishments. The
primary criterion appears to have been proximity to the center. For some field
centers, proximity meant the larger federal region in which they were located;
others stayed within their state borders (those with state-imposed restrictions on
out-of-state travel); and others stayed within the immediate metropolitan area.
One field center supervisor reported, for example, that analysts traveled to firms
outside the city in which the field center was located “only as a last resort.”
Overall, the bias appears to have been toward staying as close as possible to the
field center in fulfilling assignments.
Analysts attempted to select at least one small, one medium, and one large
establishment within an industry or (for large industries) subindustry. Analysts
repeatedly expressed the opinion that size was an especially important source of
variation in jobs. In their experience, what was a single job in a small company
would often be broken up into several jobs in a large firm. Size was apparently
assessed by referring to information on the relative number of employees in
various establishments in the area. Although this assessment appears to have
been a rather casual one, in at least one field center the supervisor reported that
the size distribution of local establishments in a particular industry was obtained
from locally available publications and divided into thirds so that one or two
establishments could be selected for study within each third. Analysts also tried
to select establishments that they believed might be employing new or emerging
technologies, on the assumption that new jobs would be available for job
analysis.
Having completed background research on the industry in question and
having selected several possible establishments for study, analysts then
attempted to gain the employer's consent to go on site for job analysis. Thus
availability, predicated on employers' consent, was the final criterion of
establishment selection. Various approaches were used to gain access. The
supervisor of one field center reported having the local Employment Service
office set up the initial meeting between the analyst and company personnel. At
most field centers, however, analysts contacted the appropri
OCR for page 120
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 120
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
ate company officer and arranged for the study on their own. Without
exception, analysts reported that they tried to talk to the most senior person in
the company and were more successful in gaining access when they did so. To
obtain an employer's consent, analysts used a “sales pitch,” emphasizing the
benefits to the employer of the data to be collected. In unionized companies,
union consent sometimes was required, although it was usually sufficient
merely to notify unions when analysts were to be on site.
It was difficult to determine the degree to which or the circumstances
under which employers cooperated or refused to cooperate in a study. Although
at least one field center maintained a file of all establishment contacts and their
outcome (including, when applicable, the reason for refusal), field centers were
not required to do so. As a result, there appears to be no systematic way of
assessing the extent, nature, or implications of employers' noncooperation.
We received mixed reports about the rate of refusal. There appears to have
been a good deal of variation across field centers. No clear picture emerged as
to the type of company most likely to have refused. Analysts at two field
centers reported that small companies were more likely to do so and attributed
this to the independent, “get-off-my-back” style of many such firms and to their
general antipathy toward government intervention of any sort. At another field
center, however, analysts could discern no pattern to nonresponse, saying that
small companies were no more likely to refuse than large ones. There was some
feeling that large companies were more cooperative than small ones. Gaining
access to large companies often entailed extensive negotiations and permission
from the company's home office (often located in another state or region), but
several analysts said that if they persevered, they were usually successful.
JOB ANALYSIS PROCEDURES
Before beginning an establishment study, analysts discussed the proposed
study with the employer or supervisory personnel, explaining its purpose and
intent, going over procedures, and learning the physical layout of the firm. Even
when employers agreed to allow a study to be conducted, they sometimes
imposed certain restrictions on analysts' activities. It was reported, for example,
that employers (and unions) typically did not permit analysts to use tape
recorders or decibel meters in the analysis of production jobs. Some jobs were
declared off limits for security reasons, usually to protect a unique aspect of a
manufacturer's production process.
OCR for page 121
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 121
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
STAFFING SCHEDULE AND ORGANIZATION AND
PROCESS FLOW CHARTS
In the next step of a typical study, analysts completed a form called a
staffing schedule, a copy of which is shown in Figure 6-1. Instructions for
completing the staffing schedule are laid out in some detail in the Handbook for
Analyzing Jobs (U.S. Department of Labor (1972), hereafter referred to as the
Handbook). Schedules were completed for every division of an establishment if
it was analyzed in its entirety or, if only a portion was analyzed, for the relevant
divisions or departments.
The staffing schedule was usually prepared with the assistance of the
establishment's personnel office or supervisor. If a company had a personnel
office, analysts usually copied its records verbatim to complete portions of the
staffing schedule. Otherwise, they compiled the necessary data anew. The
staffing schedule has two parts, a face sheet and a title sheet. On the face sheet,
analysts entered identifying information about the establishment (a unique
control number assigned by the field center and a Standard Industrial
Classification code) and names of products manufactured or services rendered.
Analysts then completed one or more title sheets. To do so, they listed company
job titles by organizational units for the entire establishment or for the divisions
in which they were interested. For each job the total number of workers was
recorded, as were subtotals broken down by sex and by shift, if applicable. If
the job was entry level, this was also noted. Although it is nowhere stated in the
Handbook, it was reported to us that part-time and trainee jobs were not listed
on the title sheet and hence were excluded from study.
On the basis of a preliminary inspection during a plant tour and/or in
consultation with management or other personnel, analysts then determined for
each job whether it was substantially similar in basic tasks and requirements to
a job definition in the third edition DOT. If so, a third edition title and
preliminary code were entered in the appropriate space on the staffing schedule.
If a job was a combination of two or more DOT occupations, all the applicable
titles and codes were entered; if a job could not be converted to a third edition
code, the space was left blank.
This comparison of jobs in an establishment against third edition titles
determined the scope and direction of the ensuing study. Jobs that could be
converted to a DOT definition but were not specific to the industry being studied
(e.g., clerical jobs in manufacturing industries) were not supposed to be
analyzed. Instructions in the Handbook are ambiguous about whether all
remaining jobs are supposed to be analyzed. It appears that they should be, but
departures from complete top-to-botton studies were not uncommon.
OCR for page 122
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.OCCUPATIONAL TITLES
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF
122
OCR for page 123
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.OCCUPATIONAL TITLES
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF
123
OCR for page 124
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 124
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
For jobs that could not be converted to a third edition title, a complete job
analysis was undertaken. A slightly abbreviated analysis was made of jobs that
could be converted as well as of jobs that could not be converted but that the
analyst knew had already been analyzed in other establishments. At a later date,
after they had finished analyzing jobs in the establishment, analysts entered
“treatment codes” for the remaining jobs to indicate how thoroughly each had
been studied.
In the final step before beginning the intensive analysis of particular jobs,
analysts diagramed the company's organization arrangements, indicating lines
of authority between different divisions. They also prepared a “process flow
chart” to depict the “sequence of procedures or processes at the establishment:
(a) if industrial, from arrival of raw materials to shipment of finished product;
(b) if service, from entry of client or material into the service until the service is
completed on the individual or item” (Handbook, p. 55). Process flow charts
were usually prepared only for company divisions that were directly engaged in
production and not for supporting departments such as quality control, sales,
accounting, etc. These departments, however, were included in the organization
chart. The Handbook offers very little instruction about the completion of either
the organization or the process flow charts, and it seems that frequently these
charts were not prepared.
JOB ANALYSIS
All of the information obtained for the staffing schedules, organization
arrangements, and process flow charts was intended to assist analysts in
identifying jobs for study and in keeping track of their progress as the
establishment study proceeded. The Handbook (p. 3) defines “job” as “a group
of positions which are identical with respect to their major or significant tasks
and sufficiently alike to justify their being covered by a single analysis. There
may be one or many persons employed in the same job.” The various
components of a job according to the methodology used in the occupational
analysis program are elements, tasks, and positions. “Element” is “the smallest
step into which it is practicable to subdivide any work activity”; “task” is “one
or more elements” that form a distinct activity or step in the performance of
work; and “position” is “a collection of tasks constituting the total work
assignment of a single worker” (Handbook, p. 3). Thus there is a one-to-one
correspondence between a worker and a position. Despite these conceptual
distinctions between a job and its component parts, analysts generally accepted
the establishment's definition of which positions constituted a job.
Job analysis for the DOT is based on a combination of observation and
OCR for page 137
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 137
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
and seven environmental conditions. Physical demands were (1) strength
(lifting, carrying, pushing, and/or pulling), (2) climbing and/or balancing, (3)
stooping, kneeling, crouching, and/or crawling, (4) reaching, handling,
fingering, and/or feeling, (5) talking and/or hearing, and (6) seeing.
Environmental conditions were (1) work location, (2) extreme cold with or
without temperature changes, (3) extreme heat with or without temperature
changes, (4) wetness and/or humidity, (5) noise and/or vibration, (6) hazards,
and (7) atmospheric conditions.
Time4
Level
1 Short demonstration only
2 Anything beyond short demonstration up to and including 30 days
3 Over 30 days up to and including 3 months
4 Over 3 months up to and including 6 months
5 Over 6 months up to and including 1 year
6 Over 1 year up to and including 2 years
7 Over 2 years up to and including 4 years
8 Over 4 years up to and including 10 years
9 Over 10 years
FIGURE 6-4 Scale for specific vocational preparation (SVP). Source: Handbook
for Analyzing Jobs (U.S. Department of Labor, 1972:220).
4Time spent in general educational development is not considered in estimating specific vocational
preparation.
The analyst indicated simply the presence or absence of physical demands
2 through 6 and environmental conditions 2 through 7. Strength was rated
according to one of five levels to reflect sedentary, light, medium, heavy, or
very heavy work. Work location was rated to indicate whether the job was
performed primarily indoors, outdoors, or both. Many of these factors could
have been measured objectively. Because analysts did not take thermometers,
decibel meters, or other instruments and gauges with them on site—in fact, as
was mentioned previously, they
OCR for page 138
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 138
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
G Intelligence K Motor Coordination
V Verbal F Finger Dexterity
N Numerical M Manual Dexterity
S Spatial E Eye-Hand-Foot Coordination
P Form Perception C Color Discrimination
Q Clerical Perception
Quintiles for Rating Aptitudes
1. The top 10 percent of the population. This segment of the
population possesses an extremely high degree of the aptitude.
2. The highest third exclusive of the top 10 percent of the population.
This segment of the population possesses an above average of high
degree of the aptitude.
3. The middle third of the population. This segment of the population
possesses a medium degree of the aptitude, ranging from slightly
below to slightly above average.
4. The lowest third exclusive of the bottom 10 percent of the
population. This segment of the population possesses a below
average or low degree of the aptitude.
5. The lowest 10 percent of the population. This segment of the
population possesses a negligible degree of the aptitude.
FIGURE 6-5 Aptitude factors and rating scale. Source: Handbook for
Analyzing Jobs (U.S. Department of Labor, 1972:233).
D DCP (direction, control, and planning)
F FIF (feelings, ideas, or facts)
I INFLU (influencing)
J SJC (sensory or judgmental criteria)
M MVC (measurable or verifiable criteria)
P DEPL (dealing with people)
R REPCON (repetitive, continuous)
S PUS (performing under stress)
T STS (set limits, tolerances, or standards)
V VARCH (variety and change)
FIGURE 6-6 Temperament factors. Source: Handbook for Analyzing Jobs (U.S.
Department of Labor, 1972:295).
OCR for page 139
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 139
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
A preference for A preference for activities
1a. activities dealing with vs. 1b. concerned with the
things and objects. communication of data.
A preference for
A preference for activities of
activities involving
2a. vs. 2b. a scientific and technical
business contact with
nature.
people.
A preference for
activities of a routine, A preference for activities of
3a. vs. 3b.
concrete, organized an abstract and creative nature.
nature.
A preference for activities
A preference for
that are carried on in relation
4a. working for the vs. 4b.
to processes, machines, and
presumed good of people.
techniques.
A preference for
A preference for activities
activities resulting in
5a. vs. 5b. resulting in tangible,
prestige or the esteem of
productive satisfaction.
others.
FIGURE 6-7 Interest factors. Source: Handbook for Analyzing Jobs (U.S.
Department of Labor, 1972:317).
were often prohibited by employers from doing so—these ratings too were
based on analysts' judgments. The analyst queried workers closely about the
processes, machines, and materials they worked with in order to determine
environmental conditions. To assess physical demands, job tasks were usually
merely observed.
COMPLETING AN ESTABLISHMENT STUDY
After completing the establishment staffing schedule, organization and
process flow charts, and job analysis schedules for each of the jobs being
analyzed in the establishment, the analyst summarized the study in a narrative
report. As its title implies, the narrative report is essentially descriptive.
Although there is no prescribed format for writing these reports, the Handbook
(p. 59) suggests that they be organized to include information on (1) the purpose
and history of the establishment and the scope of the job analysis study, (2)
environmental conditions, (3) the organization and operations or activities of
the firm, and (4) its personnel policies and practices.
As the study proceeded or at its completion, the analyst submitted the
materials produced to either a lead analyst or the field center supervisor for
review. Once the study had been checked and approved, the analyst often sent a
note of appreciation to the employer and, depending on the
OCR for page 140
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 140
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
field center's practice, either gave the employer a copy of all materials or
provided copies on request.
MODIFICATIONS OF PROCEDURES
The steps described above for conducting an establishment study were in
effect during the normal course of fourth edition production and reflect standard
operating procedures. During certain periods of production, however, and
sometimes in the normal course of events these procedures were modified.
Several years prior to the anticipated publication of the fourth edition,
national office staff decided that in order to increase coverage and expedite the
production process, modifications to traditional operating procedures were
necessary. From 1974 to 1976, analysts were directed to concentrate their
efforts on verifying jobs against existing job schedules for similar jobs in other
establishments or against the DOT definition if the job could be converted to a
third edition code. In this way much of the time-consuming writing entailed in
completing the job analysis schedule was eliminated. Evidently, this directive
was variously interpreted by the field centers. Analysts at some field centers
continued to produce job analysis schedules according to Handbook procedures,
which require a complete study of the job being verified; others resorted to
shortcuts, telephoning an establishment or trade association, for example, in
order to verify descriptions of jobs being analyzed in other establishments or to
confirm a third edition definition. When this procedure was followed, staffing
schedules were not produced.
At about the same time that the directive to change the standard procedure
was issued, field centers were also asked to review the status of their industry
assignments. As part of this review, field centers submitted lists of jobs
(primarily unanalyzed third edition jobs) in their assigned industries that they
had been unable to analyze. All such jobs were compiled in a so-called Not
Available and Obsolete (NA&O) list that was then circulated among the field
centers. Field center staff were requested to try to locate and analyze those jobs
on the list that were available in their region; to do so, a complete establishment
study was not required. Thus most of the jobs on the list were picked up
piecemeal, analysts often entering an establishment to analyze one or two jobs
rather than the entire company or a division within it. Again, in this
circumstance, staffing schedules were rarely produced.
These changes in procedure—verification and NA&O—were instituted
under the pressure of an approaching deadline. Many analysts reported that
these pressures resulted in a drop in the quality of the studies and schedules
produced during this period. The modifications, however,
OCR for page 141
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 141
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
TABLE 6-3 Definition Writing Assignments
Center Occupation Category Percentage of Base Titles
California, national office 0–1 12
New York 2 8
Florida 3 4
Washington State 4 2
North Carolina 5–9 74
— 100
TOTAL
appear to have had the intended effect of increasing the quantity of jobs
analyzed: Booz, Allen & Hamilton, Inc. (1979) reports that 30 percent of the
schedules supporting the fourth edition were produced during the 1974–1976
period in which these procedures were in effect.
Finally, no attempt was made to observe certain types of jobs, including
some professional jobs, seasonal jobs, and jobs involving a wide variety of tasks
spread over long periods of time. The methodology used by the occupational
analysis program, because it relies heavily on the direct observation of jobs to
collect information about them, is not feasible for the analysis of such jobs on
site. Instead, analysts contacted trade and professional associations, employers,
or industry representatives for information. On the basis of information
obtained from these sources, job analysis schedules were completed in the usual
manner.
DEFINITION WRITING FOR THE DOT
The job analysis schedules produced from 1965 to 1976, intended for use
in compiling the fourth edition, were filed in the North Carolina field center by
third edition DOT code, along with all the other materials resulting from
establishment studies. Definition writing was not an ongoing process, and
fourth edition definitions were written during 1976, the year preceding
publication. (The Definition Writer's Manual, a technical manual to assist in
this process, was issued a little more than a year earlier, in November 1974.)
The national office assigned definition writing to the field centers by
occupational categories (see Table 6-3). As is readily apparent, the North
Carolina field center wrote the major portion of the DOT: 74 percent of the base
title definitions.
OCR for page 142
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.OCCUPATIONAL TITLES
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF
FIGURE 6-8 Worksheet for definition writing.
142
OCR for page 143
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 143
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
In order to write definitions, analysts were provided with all the source
materials available for the occupations in their assigned categories, which
consisted primarily of job analysis schedules. For some titles there were also
occupational code requests (job descriptions submitted to the Employment
Service for coding to the DOT); copies of third edition definitions; and materials
from trade associations, employers, or unions. An analysis of fourth edition
source data, discussed in detail in chapter 7, reveals that the material available
for each occupation varied greatly in both quantity (from zero to hundreds of
job analysis schedules) and quality (from complete job descriptions to the
notation “same as third edition”).
Analysts wrote definitions according to procedures contained in the
Definition Writer's Manual (U.S. Department of Labor (1974), hereafter
referred to as the Manual). Definition writing involved four basic steps, each of
which is discussed below: (1) evaluation of source data, (2) identification of
related jobs for possible combination to form an occupation, (3) classification
and rating of the occupation, and (4) definition writing.
In step 1, if more than two or three source documents were available,
analysts usually made up a worksheet (see Figure 6-8) to organize and array the
data for easy inspection. At this point, the Manual (p. 6) advises analysts to
“correct ratings that are clearly in error.” The meaning and basis for assessing
“error” are not spelled out, but, apparently, analysts were permitted to use their
judgment to ensure that the job descriptions in the source data supported the
ratings assigned. Since guidelines for determining error were not provided, the
analyst writing the composite definition appears to have had a great deal of
discretion in assessing and overriding the field work of other analysts.
In step 2, source data were combined. The primary basis for doing so was
“common work objective, work field, similarity of tasks, and the level of skill
and responsibility involved (worker functions)” (Manual, p. 6). Jobs with the
same first three digits of the preliminary DOT code and the same work fields and/
or generic titles were likely candidates for combination. Analysts were also
advised to check jobs in parallel occupational divisions or groups to locate jobs
for possible combination, for example, jobs in division 56 (processing of wood
and wood products) with jobs in division 66 (wood machining).
Although combined jobs did not have to be identical, they were supposed
to be similar. With regard to worker functions in relation to data, people, and
things, for instance, the Manual (p. 7) offers the following guidelines:
OCR for page 144
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 144
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
Any decision to combine jobs of three points difference or more should be
reviewed carefully, for if the ratings assigned to the source data are correct, a
range of as much as three points difference for a significant worker function
would likely mean that the source data are not sufficiently similar to warrant
combination.
Similarly, with regard to GED, SVP, and other worker traits the Manual
emphasizes that jobs should be combined only if their ratings are very similar.
Instructions about GED (p. 8), for example, advise that
Usually the GED level (of combined jobs) should be the same. However,
because of differences between raters, the strong influence of employer
requirements (which may vary from place-to-place and with supply of labor),
and problems of rating borderline situations, a difference of one level but rarely
two may be considered for combination when the other factors (especially SVP
and aptitudes) support the decision.
Thus in deciding to aggregate jobs to form occupations, analysts used no
single criterion. Analysts made the initial determination on the basis of the jobs'
preliminary DOT codes and titles and then took account, within fairly narrow
guidelines, of worker traits and the particular relationships among them.
Having decided how data about individual jobs would be combined, in step
3 analysts assigned the occupation its DOT code and rated it for worker traits
according to instructions in the Handbook. Because “erroneous” ratings had
already been corrected and because jobs were aggregated on the basis of the
similarity of their worker trait ratings, there should have been little within-
occupation variation on each of these traits. Presumably, then, the rating was
straightforward. Analysts appear to have eyeballed the raw data and chosen the
most frequently occurring level for each trait (i.e., the modal value) rather than
to have calculated other measures of central tendency (mean or median). As our
analysis of the reliability of ratings in the next chapter shows, however, there
appears to be rather more variability among job descriptions combined to form
occupations than would be expected from this description of procedures.
In step 4, which involved writing the actual definition, analysts drew most
heavily on information contained in the job summary and description of tasks
on the job analysis schedule. If a third edition title could be used, it was, as was
its definition, with modification as required. Definitions were written according
to the structured format described in chapter 2. In defining an occupation,
analysts described its important tasks in detail, with particular emphasis on
“responsibilities and requirements imposed upon the worker” (Manual, p. 38).
Significant worker functions were also supposed to be reflected in the
definition, but worker traits did not have to be referred to directly unless
OCR for page 145
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 145
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
they were especially pertinent to job performance. The Manual (p. 56) does
counsel, however, that a “relationship must be maintained between the trait
rating and the definition. For example, if temperament factor J (JUDGMENT) was
rated, some indication should appear in the definition that describes the nature
of the judgment required.”
Definitions were sent to the North Carolina field center, where they were
reviewed by teams of analysts selected by the national office from each of the
field centers. Review teams worked for a week at a time and attempted to
review as many definitions as were available. Definitions written by one field
center were reviewed by analysts from another. Aside from this restriction,
analysts chose which occupations they would review and were not required to
be particularly knowledgeable about them. Reviewers were permitted to make
changes in the definitions they reviewed; changes from the third edition
definition, however, had to be specially justified on an occupational definition
transmittal form (used to record the final definition approved for the fourth
edition). According to analysts involved in definition writing, reviewers were
quite conservative and resistant to incorporating changes in existing third
edition definitions. This claim is borne out by the Booz, Allen & Hamilton
study. According to results from a random sample of 307 DOT base title
occupations, 81 percent of fourth edition definitions were identical to those in
the third edition.
CONCLUSION
Results from our field center site visits, coupled with inspection of the
technical manuals used to assist analysts in the production of the DOT, lead us to
certain general impressions about this process.
First, instructions about how to study jobs appear to have been insufficient
and inadequate. Major steps in the job analysis process did not have sufficient
guidance (e.g., establishment selection). Furthermore, the manual of basic data
collection procedures, the Handbook for Analyzing Jobs (U.S. Department of
Labor, 1972), was not published until midway through production of the fourth
edition. In the absence of clear, explicit instructions, decisions about major
aspects of data collection were left entirely up to individual field centers and
analysts. Although there appears to have been surprising uniformity among
field centers in the way they conducted job analysis studies and fulfilled their
industry assignments, there was certainly room for considerable variation,
which may have adversely affected the quality and comparability of the data
collected.
Second, procedures used to produce the DOT were insufficiently
documented. For virtually every step in the production process, there was
OCR for page 146
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 146
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
little or no information available about the criteria used in decision making or
how a particular decision had been reached. Not only did this lack of
documentation make it difficult to determine what was actually done, but also it
means that nothing is known—or easily found out—about the sampling
properties, quality, and characteristics of the resulting data. The indeterminate
nature of the data potentially compromises the usefulness of the DOT.
Third, production of the DOT was poorly planned and coordinated. The
modified procedures implemented prior to publication reportedly resulted in job
analyses of lower quality than those produced during the period when standard
procedures were in effect. (This conclusion is discussed in chapter 7.)
Definitions were written especially hurriedly, with the likely result that source
data were not fully explored or perhaps evenly and consistently aggregated or
combined.
Fourth, considering the procedures followed, there is some question as to
whether, in developing material for the fourth edition, adequate coverage was
obtained of newly emerging industries and occupations. The previous edition
served, in effect, as the sampling frame for the fourth edition. Industries were
assigned by using the third edition industry designations, and a major portion of
the total data collection effort was spent trying to verify or update third edition
occupations. These practices were efficient in some ways, but they were also
rather conservative, minimizing the probability of incorporating newly
emerging jobs in the DOT or of picking up changes in existing jobs. They may
also have perpetuated the flaws, if any, of the previous edition. Furthermore, by
relying on the third edition, little effort seems to have been made to explore the
possibility of developing new or better sampling strategies. Data are available
on industries and establishments at the local and national levels, for example,
that might have been exploited to assign industries to field centers more
systematically and to aid in the selection of establishments at the regional level.
Fifth, no attempt was made to develop new job analysis methodology or to
adapt existing methods to deal with a changing occupational structure.
Although the methodology used provides a standardized and relatively
objective means of obtaining job data, it is time consuming and not suitable for
all jobs. In particular, it can be applied most practically to manufacturing jobs
or, more generally, to any type of structured job that can be broken down into
discrete tasks and performed over a limited amount of time. It is less suited to
unstructured jobs, such as certain service jobs that entail widely varying tasks.
The lack of a well-developed method for analyzing jobs that are not amenable
to the usual procedures may have impaired the DOT's coverage of these jobs as
well as the
OCR for page 147
PROCEDURES USED TO PRODUCE THE FOURTH EDITION DICTIONARY OF 147
OCCUPATIONAL TITLES
original typesetting files. Page breaks are true to the original; line lengths, word breaks, heading styles, and other typesetting-specific formatting, however, cannot be
About this PDF file: This new digital representation of the original work has been recomposed from XML files created from the original paper book, not from the
retained, and some typographic errors may have been accidentally inserted. Please use the print version of this publication as the authoritative version for attribution.
comparability of the resulting data for different types of jobs, especially
structured versus unstructured ones. Existing alternative job analysis
methodologies, such as task inventories and structured job analysis, should be
explored in an effort to improve the consistency of job analysis and to delineate
the boundaries of occupations more precisely. See McCormick (1979) and Prien
and Ronan (1971) for recent summaries and reviews of these approaches.1
These points raise a number of questions about the quality and
characteristics of the data on which the DOT is based, in particular, the
following: Were industries and jobs adequately covered? Are the data valid and
reliable? To what extent did variations in procedures across field centers and
analysts introduce bias and lack of comparability into the resulting data? What
is the distribution of the occupational characteristics data (worker functions and
worker traits)? What are the interrelationships among them? These questions
and others are systematically addressed, to the extent possible given the data
available to do so, in chapter 7.
1The California field center produced a collection of task analysis inventories in 1973.
This collection was designed to be used as an aid in job analysis, to substitute an
abbreviated data collection process for the standard procedure, and to provide guidelines
for those not trained in job analysis.
Representative terms from entire chapter:
fourth edition
