CHAPTER 3

JOB ANALYSIS

Job analysis is a necessary first step before recruiting, hiring, training employees and
evaluating their performance.

1. INTRODUCTION

It’s important to know what the job requirements are. For instance, let’s say you are
working for a cable sales company and were asked to recruit for new customer service
employees and develop procedures for hiring them. You would have a number of
questions before getting started. For example, what are the tasks and responsibilities for
the customer service job? What knowledge, skills, abilities, and other characteristics
(KSAOs) does a person need to be able to take on the job tasks and responsibilities? And
what are the critical issues that a customer service employee might face on the job?​
This is where you would turn to a job analysis, the systematic process which helps you
identify the job tasks and responsibilities, KSAOs, and critical incidents faced on the job.
We will give an overview of how job analysis data are used in organizations, as well as
how to conduct good job analyses. We will describe some of the commonly used job
analysis methods as well as the related concepts of competency modeling and job
evaluation. And we hope you’ll get a chance to see just how interesting it can be to do a
job analysis – you get to learn about jobs that you may have always wondered about.

Job analysis: The systematic process which helps you identify the job tasks and responsibilities,
KSAOs, and critical incidents faced on the job.

2. HOW ARE JOB ANALYSIS DATA USED?

Job description: An overview of a job, typically one to two pages outlining what the job entails.

Job specifications: A brief overview of the requirements for doing the job, including the
minimum qualifications necessary to do a job.

Qualifications: The skills and experience required to do a job.

Recruitment: A method for increasing your applicant pool in an attempt to find the best people
for the job.

Valid selection procedures: Methods, such as tests and interviews, which can be used in order
to assist in hiring the best applicant for a job.

Given what we’ve described above, it is no surprise that job analysis has been described
as the basis of other HR functions. In other words, a good job analysis forms the basis
for a lot of the functions I/O psychologists and HR professionals are concerned with.

First, the job analysis can become the basis for job descriptions and job specifications.
You might be wondering what the differences are between job analyses, job descriptions,
and job specifications. A job analysis usually implies a deep analytical process that is
used to describe the job and what a person needs in order to perform the job. A good
job analysis can allow a person who is relatively unfamiliar with a job to have a solid
understanding of it. A detailed job analysis, for example, might describe hundreds of
tasks. In contrast, a job description is a simpler document that can be provided to
employees so that they understand their main responsibilities, and is perhaps only a
page or two long. Similarly, job specifications give a relatively brief overview of the
characteristics needed to do a job, rather than a lengthy list, including the minimum
qualifications. In any case, the job analysis can be used to create these job descriptions
and job specifications.

Second, if an organization is planning to hire new employees, they need a job analysis to
understand what qualifications (skills, experience) are required to do the job. Moreover,
to be able to select the best employees, the organization will want to build up a large
pool of qualified applicants to choose from through recruitment. The job analysis can
provide important information about the job before starting to try to recruit the best
people, that is, to increase your applicant pool. Consider how much easier it would be to
write a job ad or posting for a job if you had detailed information about what the job
entails. Alternatively, consider how difficult it is to try to write a good job posting to
attract good applicants if you had little information about the job. In addition, a job
analysis is necessary to choose and develop valid selection procedures, such as tests and
interviews, used to hire employees. For example, if you wanted to develop good,
job-related interview questions to select the best customer service workers, a detailed
job analysis would help you to do so. In fact, in the US, legally defensible selection
systems should be based on a job analysis (Uniform Guidelines on Employee Selection
Procedures, 1978). Further, SIOP’s Principles for the Validation and Use of Selection
Procedures note that job analysis is a necessary part of developing a good selection
system in most situations.

At the same time, job analysis data are necessary to develop strong criterion measures,
that is, measures of job performance. For example, the job analysis might help you to
determine what a satisfactory level of performance would be for the customer service
specialists, such as the number of customers they help on a given day. Similarly, the
performance appraisals wherein supervisors or others evaluate employees’ performance
should be based on a job analysis. In fact, some research has shown that performance
appraisals that are based on job analysis may be a key part of making performance
appraisals less susceptible to legal challenges.

Training employees is a key part to organizations remaining competitive. As we will also

see, job analysis is a critical part of a thorough training needs assessment. Consider it
this way: How effective can someone be at training employees if they do not know the
knowledge, skills, and abilities that a person needs to do the job? This would be bad for
the organization but also bad for the person doing the job as he or she would struggle to
do well without the necessary personal resources to do a good job.

There are at least two more areas of I/O psychology practice that rely on job analysis.
First, job design, or a systematic analysis of the organization of work, often includes job
analysis to identify the best way to allocate various tasks and responsibilities among
different jobs. Finally, one particular type of job analysis, referred to as job evaluation, is
also used to determine the relative value or pay that jobs have in an organization.
Although it is generally the focus of specialists in compensation (pay) issues, we will
discuss some of the basic principles behind job evaluation. In summary, as you can see
from this brief overview, job analysis is necessary for carrying out many of the processes
we will discuss in this book.

3. SOME JOB ANALYSIS TERMINOLOGY

Task: A basic element that can be used to describe a job. Together, a large set of tasks makes up
a job.

Functional categories: A group of tasks that serve a similar purpose. These groups of tasks are
sometimes said to be grouped into larger “responsibilities”.

Knowledge, Skills, and Abilities (KSAs): Used to describe the characteristics an employee
needs to do the job. Knowledge is generally something that someone can learn, for example from
a book.​
A skill is something that you can learn how to do, such as handle a customer complaint properly.
An ability is more innate, and something that the person brings with them to the job, such as a
mechanical skill.

Subject matter expert (SME): A job expert with a great deal of knowledge about and/or
experience of the job, these are the people from whom we get the job analysis information.
Typically an incumbent or supervisor.

First, a task is a basic element that can be used to describe a job. Together, a large set
of tasks makes up the job. Often, tasks are stated in terms of an action verb, object, and
purpose (although sometimes tasks can be quite long and elaborate). An example of a
task for the customer service job would be “Answers (action verb) telephone calls
(object) to resolve customers’ issues and concerns (purpose).” Often tasks may be
grouped into functional categories, or groups of tasks that serve a similar purpose.
(These groups of tasks are sometimes said to be grouped into larger “responsibilities”).
For example, the following tasks might all be grouped into the functional category of
“Responding to customer complaints”:

●​ answers telephone calls to resolve customers’ issues and concerns

●​ writes e-mails to address customers’ inquiries
●​ engages in chat sessions with customers to answer their specific concerns​
about products and services.

Another commonly used term in job analysis is knowledge, skills, and abilities (KSAs).
Whereas tasks are used to describe the job, KSAs are used to describe the
characteristics the employee needs to do the job. It can be difficult to differentiate
between knowledge, skill, and ability, but typically, knowledge is generally something
that someone can learn, as in from a book. An example might be “knowledge of
company rules and procedures”. A skill is something that they can learn how to do, such
as “ability to handle customer complaints”; and an ability is something more long-lasting
or innate that the person brings with them to the job, such as “mechanical skill”. There
are two important things to note. First, rather than trying to differentiate between
knowledge, skills, and abilities, it may be easier just to think of them, collectively, as
what a person needs to do the job tasks. Second, many I/O psychologists use the term
“KSAOs,” meaning knowledge, skills, abilities, and other characteristics (including, for
instance, personality), although the two terms, KSA and KSAO, are frequently used
interchangeably.​
Another important term that is used frequently in job analysis is subject matter expert
(typically called an “SME”). An SME is an expert about the job in question, that is, the
person from whom we get the job analysis information.

In the field of job analysis this is usually the person doing the job (incumbents) and their
supervisors. Incumbents and supervisors each can provide important information about
the job. Incumbents, of course, actually do the job, and thus they are the most familiar
with the job. On the other hand, supervisors of the job in question may have a better
idea of how the job fits into the overall organization. Together, these incumbents and
supervisors can provide complementary information about the job. A final term is the job
analyst, or the person conducting the job analysis. This is typically an I/O psychologist;
an HR specialist who works at the organization or who was hired from outside the
organization; or it may be someone from the organization’s human resources
department.

4. GETTING STARTED: WHAT ARE SOME SOURCES OF JOB ANALYSIS DATA TO

USE AS A STARTING POINT?

Let’s say that you have been tasked with doing a job analysis for the customer service
workers in your company. How might you get started with doing a job analysis? This is
where you need to think about the different ways you have available to collect job
analysis data, and there are quite a few.

4.1. Existing Job Analysis Data

First, one source of job analysis data is to look at old job analyses. In other words, if the
company has already done a job analysis, there would be no reason to start from
scratch. Instead, you might use the old job analysis as a starting point. Clearly, your
organization has decided to do a job analysis, so we will assume that the job analysis
needs to be updated. But the existing job analysis can give you a good place to start.

Similarly, you might be able to get a job analysis for a similar job from another
organization. Although this is not very common in the private sector because of the
competition between organizations, it is more common in the public sector. For example,
if you worked for a city government and had to do a job analysis for the job of firefighter,
you might go to a city of similar size and with similar types of buildings and layout and
ask if they have an existing job analysis available. This was a common practice when one
of the authors of this book worked in a city government. Because city governments are
not in competition with each other, they are often willing to share these types of
information with each other.

4.2. Government Sources

Dictionary of Occupational Titles (DOT): Published by the US government in paper form, this
contains short job descriptions of nearly every job. The last edition was published in 1991.

Other sources of job analysis information are published by the US government. The first
of these, the Dictionary of Occupational Titles or DOT, has been around since the 1930s.
During the Great Depression, the Roosevelt Administration wanted to support businesses
by helping to provide job analysis information. That is when the DOT was developed. It
eventually grew to the size of two large telephone books and contained thousands of
short job descriptions. The DOT contains just about every job you can imagine, from
secretary to leather tanner. One important thing to keep in mind: The DOT listings are
not job analyses in themselves; what a secretary does at one company could be quite
different from what secretaries do at another. However, the DOT provides a reasonably
generic start to doing a job analysis. The most recent version of the DOT was published
in 1991.

One of the challenges with the DOT was that it came only in paper form and did not
change as new information was learned about jobs. It was also hard to update and did
not provide additional information that might be available about jobs. To solve this, the
US Department of Labor developed the Occupational Information Network or O*NET, an
online job analysis database which is available for free online. If you go to the O*NET
website and type in a job title that is of interest to you, you will find considerable
information about the job. The O*NET database describes worker characteristics and
requirements (which includes the KSAs), the experience needed for the job, the type of
work activities, and the occupation-specific information such as tasks. Thus, it describes
both characteristics of the job and what a worker needs to be able to do the job. Getting
back to our customer service example, if you type “customer service” into the O*NET
database, you will see that there is a job called “customer service representative” which
is the closest match to your job. The key here is to find the job that seems aligned with
the job you want to analyze. One of the things that may be very interesting to you as a
student is the workforce characteristics information, which provides information about
what the outlook is expected to be for the job.
Occupational Information Network (O*NET):​
An online database developed by the US Department of Labor. The database contains job analyses
from various job titles, and provides considerable information on work characteristics and
requirements, as well as the experience needed for the job.

A few points about O*NET. First, you may be wondering how this information “magically”
gets into the O*NET database. The data are collected via surveys from employees
around the US who are in these particular occupations. In other words, O*NET provides
information about what the average employee does in this occupation. But this raises a
second point. Although O*NET is far more detailed than the DOT, it still does not tell you
what a particular job is like in your organization. That is, detailed though the O*NET
information is, simply looking at an O*NET listing does not tell you what a particular job
is like in your organization. However, O*NET gives you a place to start in understanding
factors related to conducting a job analysis in your organization.

5. JOB ANALYSIS DATA COLLECTION METHODS

Once you’ve looked at different existing sources for job analysis information such as old
job analyses and O*NET, it’s time to think about how you might actually collect job
analysis data in your organization. There are a number of ways to do this, and no one
method is best – each provides useful information for doing the job analysis.

One of the most basic ways to learn about a job is through observations of SMEs doing
the work. This is especially useful for times when the job is fairly technical and the job
analyst cannot completely understand all of the technical terms that might come up in
an interview. As an example, one of the authors of this book did a number of job
analyses related to firefighting. Without actually observing people doing the job and
demonstrating the use of the equipment, it would be difficult to understand what actually
takes place on the job. A related type of observational job analysis method is referred to
as a “ride-along”, which is a common job analysis method when much of the work is
done in the field, such as in police work. Again, riding along with patrol officers may be
necessary to thoroughly understand the details of police procedures.

Perhaps one of the most common job analysis data collection methods is the job analysis
interview. With this method, the job analyst meets with SMEs to ask questions about the
job, such as what are the typical responsibilities and tasks performed on the job, what
KSAs are needed for a person to do the job effectively, critical incidents faced on the job,
and what types of qualifications and experience a person needs to be effective on the
job. One question is whether it is better to do interviews individually or with groups of
two or more SMEs. In fact, some job analysts use focus groups in which groups of SMEs
are asked structured sets of questions about their jobs. Each of these methods yields
something slightly different: The group interview might allow SMEs to build off each
other in trying to describe the job, but group interviews may cause SMEs not to say
certain things in front of each other, or to tailor what they say to the other SMEs in the
group. Consider doing a job analysis for the job of police officer, where you are
interviewing an officer and his or her supervisor. The power difference between different
ranks in a police department might be such that the officer would only agree with his
boss, rather than coming up with his or her own description of the job.

Interviews for job analyses provide rich information about the job. However, what do you
do if you are in a large organization that may have thousands of employees in a single
position? It would be difficult to interview all of these employees. This is why a number
of organizations use job analysis surveys, in which large numbers of employees complete
questionnaires about the job. As we will see, many job analysis approaches, such as
task-KSA analysis and the PAQ, rely a good bit on surveys. The main thing in conducting
job analysis surveys of large numbers of employees is to be sure that the survey
adequately samples employees in the job in terms of demographics such as gender and
ethnicity, and that it adequately samples employees from different work shifts and
geographical areas of the company.

Job analysis survey: A questionnaire given to a large number of employees about the job in
order to conduct the job analysis.

Given these different choices in job analysis methods, which is best? The answer is that
some combination of these methods may be the best, as each has unique advantages
and disadvantages. Morgeson and Campion (1997) identify and describe a number of
different types of bias that may affect the quality of different job analysis methods.
These biases include a range of issues, from impression management (such as wanting
to look like a diligent worker in front of others) to cognitive overload (when a job
analysis method is too long or tiring for the SME). Different job analysis methods seem
to be susceptible to different levels of these biases. For example, task questionnaires
which may be very long are particularly susceptible to cognitive overload. Perhaps the
best advice here is to be cognizant of the different biases and challenges associated with
each job analysis data collection method and try to reduce them as much as possible.
Another bit of advice may be to use multiple methods for data collection.
Observations: One of the most basic ways to learn about a job, observations are done by
watching incumbents and SMEs doing their job.

Job analysis interview: When a job analyst asks SMEs questions about job responsibilities, tasks
performed, critical incidents faced, and what KSAOs, experience, and qualifications are needed to
effectively perform the job.

Focus groups: When a job analyst gathers groups of SMEs and asks structured sets of questions
regarding their jobs.

5.1. Preparing SMEs for the Job Analysis

Suppose you are working in a company as a sales person. One day you hear from your
boss that someone from HR wants to come to interview you about “what you do on your
job.” Without additional information than that, how would you feel about this visit from
someone in HR? This might be a very frightening thing to hear, as you might think that
the company is concerned about your performance. Or let’s say that one day you are
asked to complete a survey about your job, without any explanation as to what the
survey is for. How would it feel not to know why people are asking you all of these
questions about your job?

Although job analysis may seem like a very straightforward process from the job
analyst’s perspective – simply learning more about the job so they can develop selection
procedures or training – this needs to be clearly communicated to employees. SMEs may
feel threatened by the job analysis if they don’t understand what it is for. Or they may
not take the process seriously (e.g., they might be careless when filling out surveys) if
they don’t realize that this process will affect who is hired into the organization. Perhaps
even worse, without any explanation, employees may begin to draw their own
conclusions about what the job analysis results will be used for – and they may conclude
that the job analysis is for some “sinister” reason. Under these conditions, it may affect
morale, or employees may even try to sabotage the job analysis process. Our point here
is that explaining the purpose of the job analysis to SMEs can be critical to making the
job analysis process a success. Further, SMEs should feel that they had the opportunity
to participate in the job analysis and that everyone’s opinion was heard. If all employees
are not chosen to participate in the job analysis – and it is commonplace for them not to
be when there are large numbers of SMEs – there should be some explanation as to how
SMEs were chosen for participation. Finally, it is important to let SMEs know the kinds of
things you will be asking them about – their jobs, what a person needs to do the job –
well in advance of any actual job analysis meeting. This gives the SMEs time to think
about what their job involves so that they are prepared to give complete information to
the job analyst. Moreover, letting the SME know that the job analysis is simply to get
information from them will make them less likely to inflate their job responsibilities.

6. JOB ANALYSIS FRAMEWORKS

Next we will discuss some examples of different approaches or frameworks for collecting
job analysis information. Each of these frameworks takes a slightly different approach,
such as focusing more on the job elements (e.g., tasks) versus focusing more on the
person elements (e.g., KSAOs). As a result, each of these methods results in a different
kind of job analysis data, and thus each method is more appropriate depending on what
you are doing the job analysis for (e.g., developing a selection test, developing a training
program). Note that each also requires a different amount of resources.

Here we present only a few of the most important examples of job analysis methods
such that the reader will have a basic idea of the range of different approaches you can
take to doing a job analysis. If you want to know more about the details of a particular
job analysis method, or would like to know about other job analysis possibilities, we
direct you towards classic treatments of job analysis methods, including the details of
how to actually conduct one.

6.1. Work-Oriented Job Analysis Methods

Some job analysis methods are classified as “work-oriented” in that the primary unit of
analysis is the characteristics of the job. Although there are several of these methods,
we will focus primarily on just a couple of common examples.

Work-oriented job analysis method: A job analysis method in which the primary unit of
analysis is the characteristics of the job.

6.1.1. Task-KSA Analysis

Task-KSA analysis: Involves generating a list of critical job tasks, and the KSAOs needed to do
them, through observations and SME interviews.

Task-KSA analysis involves generating the list of critical job tasks involved in a given job.
The list of tasks can be used to generate the KSAOs needed to do the job, and which are
clearly linked to the job. Task analysis in its own right has been around for many years.
Although KSAOs can be deduced from the list of tasks, the explicit integration of KSAs
into the task analysis process is also common, as we will describe here.

In its simplest form, task-KSA analysis would involve generating a list of critical tasks
and the KSAOs needed to do them through observations by the job analyst and SME
interviews. Tasks are generally expressed in this form:

ACTION VERB_OBJECT_HOW; USING WHAT EQUIPMENT_PURPOSE​

For example, for the job of carpenter, a task might be:

“Measures (action verb) doorway (object) using tape measure (how) to properly install
door (purpose).”
For task-KSA analysis, throughout the job analysis interview, the focus would be on what
the critical tasks are, and what KSAOs are needed to do these tasks, documenting in the
interviews that the KSAOs really are linked to the tasks. The interview material is then
documented by the job analyst into a list of tasks and KSAOs for the job. If there are
only a few SMEs available, the process often stops here).

However, if there are larger numbers of SMEs, it can become cumbersome to interview
all of them. At the same time, these larger numbers provide the opportunity to further
examine and document that the list of tasks and KSAOs really are critical to the job and
that the tasks and KSAs are linked to each other. We will describe here an example of a
task-KSA analysis survey process for the job of professor.

6.1.2. Task Criticality Survey

Criticality: How important a task is to job performance, typically in terms of importance to the
job or relative time spent on the job.

In this survey, a sample of SMEs is asked to rate each task identified by the job analysts
from the interviews. The SMEs rate the tasks in terms of criticality, or how critical the
tasks are to job performance, most typically in terms of importance to the job or relative
time spent on the job. (There are other criticality measures sometimes used in task
analysis, such as whether the task needs to be performed the first day on the job.) Note
that these task surveys can be quite long, commonly over 100 tasks. That means that
the SMEs would have to make a lot of ratings, sometimes leading to fatigue and
decreased motivation on the part of SMEs. Figure 3.5 provides a hypothetical criticality
rating survey for the tasks associated with the job of college professor.

At this point you may be wondering just what you do with this criticality data that has
been collected from the SMEs. The answer is that you want to be sure that all of the
tasks you have identified (1) really are critical to the job and (2) that the SMEs generally
are in agreement on their ratings. To do this, you perform some simple statistics on the
ratings. First, you calculate a mean SME rating for each task to see if the ratings are high
enough, and you calculate a standard deviation (SD) to see if the SMEs are in
agreement. If the SMEs are in disagreement about the ratings of a task, it will show up
in a high standard deviation. The job analyst usually determines in advance the mean
and SD that will be required to retain a task. Any task with a low mean – say, below 4.00
on a 5-point scale – is tossed out. Similarly, any task where there is a high SD – say,
above 1.00 – is tossed out.

Let’s look at some hypothetical data collected from 50 professors, shown in Figure 3.6.
Figure 3.6 shows the mean importance ratings for each task as well as the standard
deviation. Keeping in mind we want the mean importance rating to be greater than 4 and
the SD to be less than 1.00, two tasks stand out as problematic. Task 7, “Completes
paperwork regarding research grants,” has a high mean, but the SD is high (over 1.00),
showing that the SMEs are in disagreement about its importance. Similarly, Task 88,
“Keeps desk and office tidy to facilitate work with students,” has a low mean (less than
4.00), showing that the professors don’t think it is important. For these reasons, we will
drop these two tasks from the job analysis, and you won’t be seeing them in later
discussions. What you are left with, then, is a list of tasks that a large number of SMEs
have rated as critical to the job.

6.1.3. KSA Criticality

The KSAOs that have been identified in earlier stages of the job analysis are the
employee characteristics that are needed to do the tasks. How do you then assess the
criticality of KSAs that have been identified in earlier stages of the job analysis? It is
really quite similar to the way we treat the tasks – through another criticality survey.
Figure 3.7 shows an example of a hypothetical KSA survey for the job of college
professor. In this case, we have given three ratings of KSA criticality. Note that the data
from the KSA criticality survey would be analyzed in the same way as the task survey.
Again, you would want to retain the KSAs that have been given high mean criticality
ratings by SMEs and low SDs. What you are left with, then, is a list of KSAs that a large
number of SMEs have rated as critical to the job.

MIRAR TABLAS LIBRO

6.1.4. Task-KSA LinkageSurvey

Now that you have your list of critical tasks and critical KSAOs, you must be done, right?
Well, almost. One last step that is typically carried out is what is called a task-KSA
linkage survey, in which SMEs document the degree to which the KSAOs really are
needed to do the tasks. For each task-critical job task, a new group of SMEs is asked to
rate how important each KSAO is for each task. In the end, you want all KSAs to be
linked to at least one critical job task. This documents yet further that the KSAOs really
are needed to do the critical job task, and you can be confident that these are the key
KSAOs. Any KSAOs that are not linked by the SMEs to a job task are discarded.

Task-KSA linkage survey: A step in the task-KSA analysis in which SMEs document the degree
to which the KSAOs really are needed to do the tasks.
6.1.5. Summary of Task-KSA Analysis
As you can see, task-KSA analysis provides rich job analysis data, including a description
of hundreds of tasks on the job and the KSAOs needed to do them, including robust
documentation for the criticality of the tasks and KSAOs and how they link up. This sort
of job analysis is great if you are developing a personnel selection test based on content
validity where you need to carefully sample what the job is. This can also provide a rich
source of job analysis information for developing training. On the downside, task-KSA
analysis is very time-consuming, requiring that the job analyst develop a custom set of
tasks and KSAOs for each job. One way to think of task-KSA analysis is to say that it
provides in-depth information about a single job. For this reason, it is more commonly
used when it is worthwhile, that is, when you need detailed information about a single
job that might have hundreds of positions in it, as in when analyzing jobs such as for a
police officer or firefighter. On the other hand, there are simpler ways to approach job
analysis, which are less time-consuming and allow you to compare across jobs. We will
describe these in the coming sections.

6.1.6. Other Work-Oriented Job Analysis Methods

Critical incidents technique: A worker-oriented method of job analysis focused on documenting

examples of critical situations faced by job incumbents, such as examples of good and poor ways
to handle them, and the results.

Another commonly used work-oriented method of job analysis is the critical incidents
technique, which is focused on documenting examples of critical situations faced by job
incumbents, examples of good and poor ways to handle them, and the results. The
critical incidents technique can generate rich information that can be used in a number of
ways, for instance, to develop performance appraisal forms such as behaviorally
anchored rating scales or BARS, developing work-related scenarios for training, or
writing interview and test questions for selection. During the job analysis interview,
SMEs are asked to think about some critical situations faced on the job, and to note
examples of poor and good responses that they have seen employees take to those
incidents.

6.2. Worker-Oriented Methods

In contrast to work-oriented methods of job analysis, worker-oriented job analysis

approaches focus on characteristics of the employee. We will discuss two examples of
these methods.

Worker-oriented job analysis approaches: Methods of job analysis that focus on the
characteristics of the employee; examples of this are the position analysis questionnaire (PAQ) and
competency modeling.

6.2.1. Position Analysis Questionnaire

Position Analysis Questionnaire (PAQ): A standardized, pre-written job analysis questionnaire

containing 195 items. The items describe a broad range of jobs, and are an alternative to the more
time-consuming method of developing a task-KSA analysis.
The Position Analysis Questionnaire (PAQ) is an off-the-shelf (pre-written, standardized)
job analysis method based on a survey of 195 items. These items are written to describe
a broad range of jobs. As opposed to the lengthy development process involved in
task-KSA analysis, the PAQ process involves the completion of a survey which is then
scored online. Because the PAQ has been around since the 1960s, there is a large
accompanying database including hundreds of thousands of entries. This means that
when the survey is scored, the I/O psychologist can receive not only the profile of the
job, but also recommendations for the types of assessments that might be used to hire
people into this job. For example, the PAQ database might recommend that for a
particular job you use a test of clerical ability for hiring employees. This does not mean
that you can simply start using a clerical ability test to hire workers, but it does mean
you might research such tests for your job to see if they actually are valid predictors of
job performance. In addition, the PAQ database can provide guidance for the point
values of jobs relative to one another to help in developing a pay plan in the
organization.

What are the advantages and disadvantages of the PAQ? If one needs a fairly simple job
analysis completed, the PAQ is much easier than other methods in that it is pre-made
(“off-the-shelf”) and does not require creating extensive lists of tasks or KSAOs from
scratch. Further, if you want to compare jobs, the PAQ is a good choice: Whereas
task-KSA analysis allows for detailed analysis of only one or two jobs, the PAQ is good
for comparing jobs in order to look for differences. Moreover, the PAQ database can even
provide some information about the relative value of jobs for making pay decisions. One
should be aware of the potential disadvantages, though. The PAQ does not provide great
detail about jobs, such as you might get with other job analysis methods. Rather, it uses
the 195 generic descriptors to describe the job. Second, the PAQ requires a fairly high
reading level (college level), such that you can’t simply hand it to SMEs to be completed,
but that the job analyst must complete the PAQ survey after interviewing SMEs. But if
the need is to get a general idea about a job’s requirements or to compare jobs, the PAQ
can be a practical choice.

6.2.2. Competency Modeling

Competency modeling: A method of job analysis that involves describing the general
characteristics needed in jobs in a company, especially within a series of jobs or across a range of
jobs. This often includes the company values or mission statement.

Over the last couple of decades, the popularity of competency modeling has grown in
many organizations. Related to job analysis, competency modeling involves describing
the general characteristics needed in jobs in an organization, especially within a series of
jobs or across a range of jobs. What is interesting about competency modeling, though,
is that it often includes some information about the values of the organization and its
mission. This is quite different from standard job analysis methods: Whereas job analysis
usually examines individual jobs and the human requirements for them, competency
modeling also includes the flavor of the goals and values of the organization. For
example, if an organization values its customers, these sorts of values would be
expressed in the competency model, across all jobs in the organization. In short,
competency models provide similar information to job analyses regarding the human
requirements for a job, but they also include more about the organizational context and
culture. Recent discussions of competency modeling and job analysis note that although
competency modeling and job analysis are quite similar, the two processes may
complement each other when used together in organizational practice. For example,
Campion et al. (2011) point out that, compared to standard job analysis, competency
models tend to focus on top performance rather than just average performance; focus
on how the competencies vary across different organizational levels; and tend to be
directly linked to business strategies.

6.3. Other Job Analysis Methods

Functional job analysis: Focuses on the purpose or functions of the job as opposed to the actual
tasks being performed.

There are a number of other systems for conducting job analyses. Although we only
touch on these briefly, they have a long history of application, and we point the
interested reader to other sources to learn more about them. Functional job analysis
focuses on the fact that all jobs require that workers need to deal with people, data, and
things, and that the job tasks can be specified to include who performs the task, what
action is performed, what the result is, what tools or equipment are used, and what
procedures and instructions are followed. Although FJA provides some excellent detail
about the job, it does require a good bit of work on the part of a well-trained job analyst.
Another established job analysis method is the job element method. The JEM is a
worker-oriented job analysis method designed to specify the worker characteristics
needed to do a job. After these elements are identified, they are then specified further
into subelements. For example, for the job of barista, an element might be “ability to
make coffee drinks”, while the subelements might be “finding out what the customer
wants”, “estimating the heat of coffee and milk”, and “understanding the equipment”.
Note that JEM is much more focused on describing worker characteristics but to a large
degree skips over the explicit definition of job tasks. For that reason, it may be less
suitable in situations where you want to know a lot of detail about what is actually done
on the job.

7. ASSESSING AND ASSURING THE QUALITY OF JOB ANALYSIS DATA

One question that you might have is how accurate job analysis data are. The good news
is that because job analyses are based on the judgments of SMEs – experts on the job in
question – one can argue that job analysis data do, at least to some extent, have
content validity. However, the quality of job analysis data should be of some concern
because job analysis data come from the subjective judgments of SMEs. Moreover, we
don’t have any objective measures to tie job analysis data to. So how do we know
whether job analysis data are accurate? There are several ways to try to estimate how
accurate job analysis data are. There are also some studies that can give us hope that
job analysis is fairly accurate.

7.1. Reliability of Job Analysis Data

As we’ve said, there is no standard to compare job analysis data to, so trying to assess
their validity against some criterion or standard is difficult. However, reliability is a
necessary condition to show validity. This being the case, how can we show the reliability
of job analysis data? Typically, this is done through some demonstration of interrater
reliability or agreement. In other words, if the SMEs are generally in agreement in their
ratings on a job analysis survey, this is a good sign that the data are reliable, and at
least holds promise that they are valid as well.

7.2. Detecting and Removing SMEs Who Are Providing Poor Ratings

Carelessness index: A method for detecting whether SMEs are paying attention. These consist of
bogus or nonsensical questions throughout a survey, which if endorsed, indicate that the SME is
being careless.

Remember when we discussed task/KSA analysis, that we noted how long some task
surveys can be, and how tiring they can be for SMEs? One concern with job analysis
surveys, then, is to see whether SMEs are paying attention, or are showing signs of
boredom, fatigue, or carelessness. This has led to the development of carelessness
indices to detect whether SMEs are paying attention. The carelessness index consists of
bogus job analysis items, scattered throughout the survey, that are either nonsensical or
have nothing to do with the job. If an SME endorses too many of these items as being a
part of his job, it suggests that he/she isn’t paying attention, and that his/her data
should be removed from further analysis.

7.3. Who Are the Best SMEs?

A good bit of thought should go into how to choose SMEs for participation in the job
analysis. In other words, how do you choose SMEs in order to end up with the most
successful job analysis process? There are a number of issues to consider here, such as
SME demographics in terms of gender and ethnicity, the level of experience of the SMEs,
and whether to choose some of the top performers in the organization. Let’s go through
each of those issues.

With regard to gender and ethnicity, it is important that the SME sample is
representative of the demographic makeup in the organization and applicant pool. This is
for at least two reasons. First, the US courts have indicated that SME samples used for
job analysis should represent the makeup of the organization. Second, it is important to
be inclusive of all groups in the organization. For example, let’s say you were doing a job
analysis for police sergeants, but asked mostly the male sergeants to participate in the
job analysis. Such an approach could lead to the female sergeants being shut out of the
process and having no voice. Further, if the stated purpose of the job analysis was to
develop tests for promoting officers to sergeant and training them, it could cause
concerns that the promotional test and training was “biased in favor of men” and lead to
significant feelings of resentment. One final issue regarding gender and ethnicity is
whether men and women and people of different ethnicities actually provide different
types of job analysis data. The answer is “generally no.” Research shows that people give
relatively similar job analysis data regardless of their gender or ethnic background, or
that the differences are fairly small. In addition, recent reviews of the job analysis
literature suggest that any such demographic differences may be besides the point, and
that instead of examining the effects of SMEs’ demographic characteristics, we should be
looking at more substantive factors like the effects of SMEs’ social or professional
identities.

What about the experience levels of SMEs? In other words, will a person on the job for
one year provide different job analysis data than a person with 10 years of experience?
Let’s consider an example of a firefighter. Would you be concerned that a firefighter who
has been on the job for 10 years provides different types of job analysis information
from one who has been in the job for two years? This is a fairly “thorny” question. One
could argue that a newer firefighter would have the standard tasks uppermost in their
mind, but on the other hand they might not have learned all of the tricks to be most
effective on the job. In contrast, a more experienced firefighter could know all of the
“tricks of the trade”, but might also be less able to articulate all aspects of their work
because they are “second nature”. This is an interesting phenomenon that we have found
in doing job analysis interviews. Sometimes when experienced employees are asked to
describe the main tasks they do on their job, they will not mention some fairly obvious
tasks. When they are questioned about why they have skipped these important tasks,
they’ll reply, “Well of course I do that – it’s so obvious I forgot to mention it.” In any
case, the research shows that people of different experience levels may give different
types of job analysis data, and thus sampling SMEs of differing experience levels can
affect the job analysis results.

With regard to choosing your top performers in the organization to participate in the job
analysis, it is obvious that these need to be considered to provide legitimacy for the job
analysis. Going back to our example of choosing police sergeants as SMEs for a job
analysis, what if by chance you chose only those sergeants who were known not to be
good supervisors, or who had had some disciplinary problems? How would this affect the
credibility of your job analysis process? As you can imagine, under those conditions the
job analysis might not have much credibility among the police officers, and anything you
developed from it might be considered flawed. That in itself suggests that you should be
cognizant of who the most respected SMEs are in an organization and be sure to include
them in your job analysis. On the other hand, does the level of SME skill or performance
have much effect on job analysis results? While there is some research that shows that it
might, SMEs’ job performance has not been consistently shown to affect their job
analysis results.

Finally, given that the SMEs are often job incumbents, just how good are the data
obtained? Dierdorff and Wilson (2003) found in a meta-analysis that the job analysis
information from incumbents may be less reliable than the information obtained from job
analysts or technical experts. Although this may be concerning, other authors have
pointed out that lower reliability in job analysis results (i.e., disagreement among raters)
may merely indicate that the job is done differently by different employees with the
same job title.

7.4. What about Tasks versus KSAs?

One question that has come up in the job analysis literature is what types of job analysis
data are more susceptible to judgment errors. The way that this has played out in the
literature is comparing data collected from tasks versus KSAs in terms of their quality.
Here’s the reason that researchers suspect that there may be some differences between
the quality of data obtained from tasks ratings versus KSA ratings. For one thing, tasks
are more concrete in the mind of the SMEs, whereas KSAs are more abstract. Let’s
explain this a little bit more. Tasks are definite behaviors that SMEs may do all day long,
and watch their colleagues’ performance many times. On the other hand, KSAs are not
actual behaviors of people, but are the characteristics a person needs to do the job. They
are not easily observable, but require inferences on the part of SMEs. As an actual
example, let’s get back to our police sergeant example. When comparing the task of
“Completes required reports and paperwork regarding police officers” with the associated
KSA of “Knowledge of government personnel policies” the task is something the sergeant
does quite frequently and observes her colleagues doing quite frequently. On the other
hand, the KSA is not something that can be easily observed but must be inferred by an
SME. This notion has in fact been supported by some research. Dierdorff and Morgeson
(2009) found that KSA ratings showed lower reliability than task ratings. Similarly, x
found that KSA ratings were more likely to be inflated than tasks ratings. In considering
these findings, we would not suggest that analysts should focus on task statements
rather than KSAOs in doing job analyses. Indeed, KSAOs are very important to
understanding what the personal characteristics are that a person must possess to do
the job, and thus are very important to I/O psychologists. Rather, these findings just
suggest that we should be especially careful in developing KSAO statements and in
interpreting data based on them.

7.5. Training SMEs

So far in this section we have talked about ways to assess whether job analysis data are
good, how to identify careless SMEs, and how to choose SMEs so that you have a
representative sample. We’ve also talked about the fact that different types of job
analysis data (e.g., task ratings) may lead to more reliable data than others (e.g., KSA
ratings). One last issue is how we might help SMEs to make good ratings. Some of these
are obvious and are standard practice in job analysis: For example, we should do all that
we can to remove the cognitive load from SMEs and give them sufficient time to
complete their job analysis forms; this might involve asking SMEs to attend a job
analysis session during work hours rather than just sending them a job analysis form to
complete. (Just sending SMEs job analysis questionnaires to complete often leads to low
response rates because the job analysis questionnaires may not be a lot of fun to
complete and so the SME never gets around to completing it.) We might also provide
some incentive for SMEs who complete their surveys. In addition, some research has
examined the possibility of training SMEs to improve job analysis quality. For example,
Aguinis, Mazurkiewicz, and Heggestad (2009) examined the possibility that SMEs could
be trained to improve the quality of their job analysis data. They found that providing
SMEs with training reduced the correlation between SMEs’ personality characteristics and
the ratings SMEs gave for a job’s personality requirements.
8. JOB EVALUATION: JOB ANALYSIS INFORMING PAY STRUCTURES

Compensation: Involves setting pay levels within an organization.

Internal equity: A key goal of job evaluation, the intention of internal equity is to assure fairness
of relative pay values within the organization.

External equity: Assessment of fair compensation in relation to market conditions for a particular
job. This is important for attracting and retaining the best talent for each position.

These job analyses are typically used for developing criteria and other measures to
assess performance, performance appraisal systems, personnel selection procedures,
and training programs – the types of issues that most I/O psychologists are involved in.
However, there is another HR function that is important in most organizations:
compensation and pay. Generally, the issue of compensation involves setting the pay
levels within an organization. We present here a particular type of job analysis that is
used in developing compensation systems, namely, job evaluation, or determining the
relative value of jobs within the organization (“who makes more than whom”).

When considering pay, a key point is to make sure that the pay system is fair to
employees. Fairness is an important factor in the way that employees interpret their
relationship with their organization. For this reason, compensation systems need to
assure that pay systems are administered in a way that is fair to employees, such that
an employee’s pay is fair relative to other employees in the organization (internal equity)
as well as to others in the external labor market (external equity). In fact, pay fairness
was one of the original applications of a major motivation theory, equity theory.

Thus, a key goal of job evaluation is to assure internal equity, that is, fairness of relative
pay values within the organization. The actual pay levels of jobs are then largely
determined by outside market forces, or external equity. Here we present an overview of
how job evaluation works. Note that it is just an overview, and interested readers should
know that many business schools devote entire courses to the issue of determining pay.

8.1. Internal Equity

One of the key issues involved in doing job evaluation is assessing the relative value of
jobs in the organization. For example, we know that if salespersons receive a certain
level of pay, it is assumed that their supervisors will make more given their supervisory
responsibilities. But how much more? Thinking more broadly to other jobs, how much
relative value would customer service specialists, truck drivers, and secretaries bring to
the company relative to the sales people?

This is where job evaluation can come in. Although there are many ways to conduct a
job evaluation, a common approach is to assess the relative value of each job in the
organization in terms of points. The number of points given to each job in the
organization is determined based on pre-determined criteria, often referred to as
compensable factors. One of these compensable factors might include the working
conditions. Tough working conditions normally command more pay. For example, a
firefighter’s job is paid more relative to some others because of the dangerous and rough
working conditions. A second compensable factor often used is the experience or
knowledge required of the job, with greater experience and knowledge requirements
commanding more pay. For example, a research scientist would get more pay relative to
some other employees because of the technical knowledge they are required to have.
And a third compensable factor is the level of responsibility: Jobs with a lot of
responsibility get paid more than those with less responsibility. For example, a manager
supervising a dozen employees would get paid more than a worker with no supervisory
responsibilities.

Consider Figure 3.11, which shows the relative number of points that has been given to
a number of jobs in a fictitious organization. Based on the compensable factors, the
managers have 952 points, the engineers 750 points, the secretaries 350 points, and the
truck drivers 325 points. Thus we can assume that the managers should make
approximately three times as much as the secretaries and truck drivers, and the
secretaries and truck drivers should make about half as much as the engineers.

8.2. External Equity

At this point, we only know the relative values of the jobs in terms of points, not in
terms of what they should be paid. A next step would be to assess external equity
relative to the market conditions for those particular jobs. This can be done through
salary surveys of the market. These salary surveys might be conducted by the
organization, obtained from government data, or purchased from a consulting firm. The
salary survey allows you to see what the value of a job is on the market. Maintaining this
external equity with the job market is important for attracting the best talent, as well as
for retaining the best employees. Also, it may be that the company does not have the
resources to do salary surveys for all of the different types of jobs in its organizational
structure. Instead, organizations often only do surveys for their most essential jobs
(often called benchmark jobs), and the pay for the other jobs is extrapolated. In the
example above, if the salary survey showed that engineers were paid about $150,000
per year, we could extrapolate that managers should make approximately $200,000.

8.3. Other Considerations

Keep in mind, though, that although this all sounds quite logical, sometimes it’s not as
simple as calculating the point values of jobs relative to each other and plugging in the
dollar values from the salary surveys. A number of other issues may need to be
considered. Perhaps the most important of these is that the market rates for jobs may
not match the relative values of the jobs as determined through the job evaluation. For
example, because of the market value that engineers might have, you could end up with
a situation where the salary surveys would indicate that the engineers are paid more
than a supervisor at the next level above them. As another example, you might find that
the external market is paying secretaries less than truck drivers. This means that the
organization will have to find some way to reconcile the value of the jobs as determined
by the job evaluation with the external market forces.
8.4. Comparable Worth

Comparable worth: Commonly discussed in terms of gender differences. Points to differences in

pay for typically male versus typically female jobs based on job evaluation and market value.

The earlier example of secretaries and truck drivers also raises the issue of comparable
worth, most commonly discussed in terms of gender differences in jobs. Comparable
worth as a legal issue points out that many traditionally female jobs (in this case,
secretaries) are often paid less than traditionally male jobs (in this case, truck drivers)
that have been shown through the job evaluation to be worth the same to the
organization. In this example, even though secretaries were found to have slightly
greater value to the job, the market might suggest that they should get paid less. Note
that the comparable worth issue is different from the fact that women and men within
the same profession (e.g., male and female college professors) might get paid
differently. This would be a simple matter of equal pay for the same work, an issue for
many professions, from stockbrokers to movie stars. Rather, comparable worth focuses
on pay differences for different jobs that are traditionally associated with men (truck
driver) versus those associated with women (secretary). Although the issue of
comparable worth gained some traction in the 1980s, it has to some extent died down
because the courts have generally come down on the side of organizations needing to
use market forces to set pay. Moreover, there is today greater fluidity in the
opportunities open to men and women in different jobs: Women and men have far
greater opportunity in terms of which jobs are available to them than at any other point
in history.

9. GLOBAL IMPLICATIONS

Given the recent growth in global organizations, we see a number of implications for job
analysis. First is the issue of what a particular job title means in one country versus
another. Although job titles may be the same in different countries, the responsibilities
may be quite different: This may be due to cultural differences, or due to the fact that
global organizations may be the result of the merger of different organizations. Whatever
the reason, one challenge, then, is to assure that a company’s job titles mean the same
thing in different parts of the world. Second, the use of competency modeling, which is
so highly dependent on the culture and values of the organization, seems particularly
susceptible to cultural differences from one part of a company to another, creating some
headaches in its implementation. On the other hand, because competency modeling
focuses on showing how the organization’s goals and values are reflected in the job,
competency modeling may be especially useful in helping to transmit the organization’s
goals and values across different parts of the company, including in different parts of the
world. Although the job analysis literature has done little to tackle issues related to
multinational organizations, the time seems ripe to do so.

10. CURRENT ISSUES AND CONTROVERSIES

Although job analysis has been an integral part of I/O psychology for many years, there
are some new issues beginning to emerge within the field. Many of these new issues
have to do with the fact that job analysis practices have not changed over the years,
while workplace practices and the field of I/O have changed quite a bit. While we don’t
see this as an end to job analysis, it does mean that a number of approaches to job
analysis may change in the coming decades. We provide several such examples below.

First, as we describe throughout this book, the ability of large employers to collect data
in the workplace has changed through the years. This has included​
a detailed monitoring of what employees do at work. While this raises a number of
issues, both positive (e.g., ability to know what workers are actually doing) and negative
(e.g., decreased privacy), it also means that job analysis data can be collected in new
ways.

Job analysis data collection has generally been considered a subjective process. But with
the ability to know what employees are actually doing at work, job analysis may be able
to include more objective information, not just the subjective judgments of SMEs. This is
a fairly large shift from the way that organizations have traditionally done job analysis.

Second, many of our approaches to job analysis have been around for decades, when
jobs were more finite and changed very slowly. Today, jobs​
can change quickly and be much more fluid. For that reason, there has been a
movement towards using the term “work analysis”, which implies the fact that clear-cut,
specific “jobs” are not always easily analyzed in today’s workforce. As noted by some
authors, rather than just thinking about single jobs with tasks that do not change much
over time, we should be thinking more broadly of work roles. This acknowledges that the
work that individuals do in organizations changes over time and that workers do not
work in isolation but are integrated into large teams and contexts.

Work analysis: The term acknowledging that jobs are quickly changing and more fluid in today’s
market.

Moreover, interest in job analysis research seems to be falling off in many of the top I/O
psychology journals. However, in their review of the job analysis literature, Sanchez and
Levine (2012) point out that this decreased interest is not because job analysis is
unimportant, but because job analysis is not usually an end itself; rather, job analyses
are usually done in support of other more “important” I/O functions such as personnel
selection or training. In other words, some I/O psychologists may not see job analysis as
that important because it is only one part of a larger HR process. However, Sanchez and
Levine (2012) argue that doing a job analysis is clearly a psychological process,
requiring judgments by SMEs about what they do on the job and what characteristics a
person needs to carry out the job. In short, job analysis is still​
a very important part of any organizational human resources system, although its focus
and approach may change over the coming years as job environments become more
dynamic and fluid.

Cognitive task analysis: A newer approach to job analysis, this goes beyond traditional task
analysis by focusing specifically on the cognitive processes involved in doing the job.

Third, there are already a number of new approaches to job analysis on the scene. One
of these is cognitive task analysis, which goes beyond traditional task analysis by
focusing specifically on the cognitive processes involved in doing the job. In other words,
rather than just focusing on what a person does (e.g., “prepare coffee drinks”), this
would be a focus on the cognitive processes involved (e.g., “calculate the heat of
steamed milk,” “remain aware of own emotions under stress.”) Such cognitive task
analysis is time-consuming​
and expensive, but may be worthwhile for understanding high-risk, critical jobs. Further,
there is an increased interest in explicitly assessing elements of jobs and additional
worker characteristics, such as personality traits required for the job, perhaps because of
increased interest in the effects of personality on worker performance over the last two
decades. Although we only provide these two examples, we anticipate changes in the
types of information that can be collected through job analyses in the coming years.

11. CONCLUSION

We have illustrated the importance of job analysis to a number of issues of interest to

I/O psychologists such as selection, performance appraisal, and training. There are a
number of ways to collect job analysis data, and a number of systems for framing the
issue of job analysis, including more work-oriented and more worker-oriented
approaches. Each of these approaches has a number of advantages and disadvantages
depending on what it is that you will use the job analysis data for. At the same time, the
research and practice on job analysis have identified a number of ways to improve job
analysis data. The future of job analysis (or “work analysis”) is expected to evolve in the
coming years as the nature of work continues to change in radical ways.