Job Analysis Chapter 3 PDF

CHAPTER Job Analysis 3 CHAPTER OUTLINE Some Terminology Approaches to Job Analysis Job-Analytic Methods Task-Oriented Techniques Worker-Oriented Techniques PRACTITIONER FORUM: John F. Binning Advances and Issues in Job Analysis Practice and Research Defining the Job: Descriptions and Specificat...

CHAPTER Job Analysis 3 CHAPTER OUTLINE Some Terminology Approaches to Job Analysis Job-Analytic Methods Task-Oriented Techniques Worker-Oriented Techniques PRACTITIONER FORUM: John F. Binning Advances and Issues in Job Analysis Practice and Research Defining the Job: Descriptions and Specifications Competency Modeling I/O TODAY The Future of Job Analysis The Many Purposes of Job Analysis Job Classification Criterion Development and Performance Appraisal Selection and Placement Job Design and Redesign Training Job Evaluation Point System Comparable Worth Summary LEARNING OBJECTIVES This chapter should help you understand: How important job analysis is to HR functioning The common terminology used in the area of job analysis How to differentiate between task-oriented and worker-oriented job analysis techniques How to conduct a job analysis using the Task Inventory Approach, Functional Job Analysis, Job Element Method, Position Analysis Questionnaire, and Common-Metric Questionnaire What’s included in the Dictionary of Occupational Titles (DOT) and how it has been improved through the development of the Occupational Information Network (O*NET) What a job description is and how it is used in human resource practices What job specifications are and how they are used in human resource practices The variety of human resource functions for which job analysis is of great importance The newly developing role of technology in the analysis of jobs The role of job evaluation in setting compensation levels The doctrine of comparable worth and the wage gap The second part of this book centers on industrial psychology, or what has been traditionally called personnel psychology. Before discussing the major applications of industrial psychology, you must first have an understanding of the foundation for virtually all of this work—job analysis, the focus of this chapter. Employees get evaluated at work all the time. Think about the person who serves you in your favorite restaurant; she gets evaluated by her boss in various ways, but that evaluation can’t be effectively done without the boss and the server knowing the job. That’s where the job analysis comes in. Imagine trying to evaluate someone’s performance without knowing what the job or what the outcomes of the work are intended to be. Now think about your first job, even as a babysitter or fast-food employee—picture someone appraising your performance without knowing the details of the job for which you were hired. Unfortunately, job analysis is often taken for granted by I/O researchers and practitioners, but in this chapter I argue that it is the most important building block for industrial psychology. In order to develop predictors to select and place employees in appropriate jobs, to identify criteria used to evaluate employees’ effectiveness, and to identify remedial training needs of particular employees, we must first understand the job. We gain this thorough and comprehensive understanding of jobs through job analysis. Doing performance appraisals without such knowledge would be like cooking a dish without knowing the ingredients, how to prepare them, or how the dish is supposed to taste. The job analysis provides the recipe— information that is integral to everything else that follows in industrial psychology. Figure 3.1 illustrates how job analysis relates to the major human resource (HR) functions that will be covered in this section of the text. Throughout this chapter we will talk about the three major products of the job analysis: the job evaluation that is used in setting pay levels in organizations, the job description that relates to what is actually done on the job, and the job specifications that identify the skills that folks need to be able to do the job effectively. FIGURE 3.1 Framework for Industrial Psychology SOME TERMINOLOGY Job analysis is the process of defining a job in terms of its component tasks or duties and the knowledge or skills required to perform them. Job analysis is the basis for the solution to any human resource problem (Brannick, Levine, & Morgeson, 2007). Before discussing approaches to job analysis, we must familiarize ourselves with some terms that have very specific meanings within this domain. First in the list are elements. An element is the smallest unit of work activity, such as pressing a button to start a machine or entering a line of code for a computer program. Multiple elements join together to form a task, commonly defined as the activity of work that is performed to achieve a specific objective. For instance, among the necessary elements involved in a cab driver taking his passenger to her destination are turning the key, shifting the transmission into “drive,” and pressing the accelerator. Collectively, these elements form a task that is the operation of the vehicle. The tasks performed by an individual in an organization define that person’s position. A job is a collection of positions similar enough to one another to share a common job title. KSAOs are the knowledge, skills, abilities, and other characteristics that are required for successful job performance (Primoff & Eyde, 1988). They are very important in I/O psychology because they are the basis for much of the work done in employee selection (see Chapters 6 and 7). In short, organizations prefer to hire those employees who possess KSAOs that are likely to make them successful performers. Indeed, this is largely what selection and placement in organizations are all about. job analysis The process of defining a job in terms of its component tasks or duties and the knowledge or skills required to perform them. element In job analysis, the smallest unit of work activity. task A work activity that is performed to achieve a specific objective. position An individual’s place in the organization defined by the tasks performed. job A collection of positions similar enough to one another to share a common job title. KSAOs The knowledge, skills, abilities, and other characteristics that are required for successful job performance. To recap here: The most basic level of work activity is made up of the elements of work, which combine to form tasks, which in turn constitute one’s position in the organization; similarly defined positions are what we term a job (see Figure 3.2). Of course, we should be careful to avoid confusing the job and the position with the individuals who occupy them (Harvey, 1991). In other words, job analysis should be focused not on the person who holds the job or occupies the position but, rather, on the job or position itself. FIGURE 3.2 Levels of Work Activity APPROACHES TO JOB ANALYSIS Most procedures in I/O psychology can be conducted in various ways. Job analysis is no exception. Whatever the approach, however, the job analyst needs to consider the purpose of the job analysis when deciding on a strategy. One approach may be more beneficial for a given purpose than a different approach. In this section and the one to follow, we will examine a couple of general perspectives taken in the area of job analysis as well as some specific ways in which job analysis data can be collected. The traditional classification espoused by most job analysis experts over the years categorizes job analysis methods as either task-oriented or workeroriented (McCormick, 1976). Task-oriented techniques focus on describing the various tasks that are performed on the job. Worker-oriented techniques examine broad human behaviors involved in work activities (Gatewood & Feild, 2001). Unlike task-oriented approaches, which focus on very specific levels of tasks, worker-oriented job analyses focus broadly on general aspects of the job, such as the physical, interpersonal, and mental factors necessary for completion of the job and the related worker attributes. In fact, a review and analysis in this area also suggests a third category—hybrid methods. These approaches attempt to gather information about the work and the worker at the same time (Brannick et al., 2007). The distinction between the two traditional categories is not always clear; some job-analytic techniques cross over into the other category (Sackett & Laczo, 2003). Furthermore, one approach is not necessarily better than the other; as a job analyst, you can choose a hybrid approach or any combination of pieces from different job-analytic techniques (Brannick et al., 2007) for whatever purpose is being served by your job analysis. task-oriented Referring to approaches to job analysis that focus on describing the various tasks that are performed on the job. worker-oriented Referring to approaches to job analysis that examine broad human behaviors involved in work activities. JOB-ANALYTIC METHODS For simplicity’s sake, this chapter largely employs the traditional taskoriented and worker-oriented distinction, but it also presents the important results of a hybrid approach employed in the development of the Occupational Information Network (O*NET). There are many job analytic techniques, but we will only cover a sampling of the most used approaches. Certain techniques might fit better with particular jobs. Task-Oriented Techniques Task-oriented techniques for conducting job analyses tend to be focused on tasks. (Recall our previous definition of a task as the activity of work that is performed to achieve a specific objective.) One task-oriented approach is called the Task Inventory Approach, in which task statements are generated by experts who are familiar with the jobs—subject matter experts (SMEs). These experts may be incumbents—that is, people who are currently occupying the job of interest. Or they may be individuals who have expertise or knowledge about the job for some other reason, such as being a supervisor of individuals in that job or an I/O psychologist who has gathered expertise through his or her work. Once a list of task statements has been generated, this list, which can include hundreds of task statements, is usually administered to incumbents, who put a check next to those statements that describe a task that they do on their job. A sample task inventory for the job of college professor is provided in Figure 3.3. The task statements that are checked as being done on the job are also often rated as to the importance or criticality of the task to the job and the relative time spent on the job doing each task. Together, these pieces of information can provide a thorough analysis, or picture, of the job. Task Inventory Approach Task-oriented approach which uses task statements generated by experts familiar with the job in question. subject matter experts (SMEs) Individuals who participate in job analyses as a result of their expertise. incumbents Employees who are currently occupying the job of interest. FIGURE 3.3 Task Inventory for College Professor A CLOSER LOOK From a Functional Job Analysis perspective, how would a computer technician score on data, people, and things? What would the scores look like for sales jobs? In Functional Job Analysis (FJA), a highly structured task-oriented approach developed by Sidney Fine many years ago, data are obtained about what tasks a worker does and how those tasks are performed. (For a summary of this technique, see Cronshaw, 2012.) Specifically, a series of task statements are developed that are believed to be relevant for the job in question. Incumbents or SMEs rate each of these statements (which are similar to those shown in Figure 3.3) on a series of dimensions. These dimensions are what set this approach apart from the standard Task Inventory Approach. Functional Job Analysis (FJA) A highly structured task-oriented approach developed by Sidney Fine in which data are obtained about what tasks a worker does and how those tasks are performed. A picture of the job is generated when these task statements have been rated with regard to data, people, and things—three factors with which employees in all jobs are assumed to interact. One way to understand a job is to know the extent to which these interactions take place and the complexity of those interactions. This is the focus of FJA. A task focusing largely on data requires the employee to use cognitive resources in handling information, ideas, and facts. Tasks that are coded along this dimension range from simple comparisons to more complex synthesizing. The people dimension refers to the extent to which the job requires employees to use interpersonal resources such as understanding, courtesy, and mentoring. Tasks coded along this dimension range from simply taking instructions to more sophisticated behaviors such as mentoring. Finally, a task that is linked to the things scale requires the use of physical resources and includes the use of strength, speed, and coordination. Tasks coded along this dimension range from the simple handling of things to more intricate and precise operations. In the 1930s, the Department of Labor used FJA to develop the Dictionary of Occupational Titles (DOT), a tool that matches people with jobs. Consisting of narrative descriptions of tasks, duties, and working conditions of about 12,000 jobs, the DOT codes each of these jobs according to the data, people, and things dimensions developed by Fine. The information presented in the DOT is based on years and years of data collected through job analyses. The DOT presents a hierarchical organization of jobs, with nine major occupational clusters at the uppermost level of the hierarchy. Figure 3.4 shows the DOT entry for a faculty member, including labels for the particular parts of the entry. The nine-digit occupational code is important in that it uniquely identifies this particular job. The first three digits classify the job with respect to its occupational category. The next three digits identify where the job is classified with respect to data, people, and things; in this example, the digits 227 indicate that the job is high on both data (analyze) and people (instruct) but low on things (handle). This probably fits your preconceived notion of what faculty members do, which is to spend a great deal of time working with data and people but less time manipulating tools, equipment, and so on. The industry designation in this case is “education.” Dictionary of Occupational Titles (DOT) A tool developed by the Department of Labor in the 1930s that has been used to classify occupations and jobs, consisting of narrative descriptions of tasks, duties, and working conditions of about 12,000 jobs. FIGURE 3.4 DOT Description for Faculty Member The lead statement is always the first sentence of the description itself, followed by a colon. This statement provides the best simple description of the job. The task element statements, which come next, describe with more precision what is done on the job to meet the objectives presented in the lead statement. In other words, college professors teach subjects, prepare lectures, stimulate class discussion, and so on, in order to conduct college courses. Finally, the “may” items indicate activities that may be done in some organizations by individuals who occupy this job but are not routinely part of this job in most organizations. The Occupational Information Network (O*NET) About 20 years ago the Department of Labor (DOL) undertook an initiative to replace the DOT, which it viewed as incompatible with the current workplace. This initiative, known as the Occupational Information Network (O*NET), identifies and describes the key components of modern occupations (Occupational Information Network, 2009, January 13) and has replaced the DOT, which has become obsolete as a resource. In contrast to the DOT, the O*NET is not based on FJA; rather, it is based on data gathered in various ways, including an initial review of existing literature in the domains of human performance and occupational analysis. Thus, it is categorized as a hybrid approach because it focuses simultaneously on both the work and the worker. This database of over 950 occupations is now available online (http://onetcenter.org), making it much more accessible than the older DOT. In addition to having data on thousands of jobs, it allows individuals to search for occupations through key words relating to skill requirements, and it is connected to America’s Job Bank and school-to-work initiatives in which schools train students to be better prepared for the workplace and to successfully find employment. The greater ease of access (through the Internet) and its focus on both job and worker characteristics make the O*NET a very useful and attractive tool for organizations and I/O psychologists. Indeed, the O*NET has widespread implications for the world of work and the practice of I/O psychology. The O*NET database of occupational information is based on the O*NET Content Model, which identifies the structure of information stored in the database. The structure, which is presented in Figure 3.5, involves six classes of information: (1) worker characteristics, such as abilities and interests; (2) worker requirements, such as general knowledge and skills; (3) experience requirements, such as past employment history, training, or licensure; (4) occupational requirements, such as generalized work activities; (5) occupation-specific information, such as technical information particular to a certain job family; and (6) occupational characteristics, such as labor market and social information. The database is quite useful for career management; for organizational functions like selection, training, and development; and for researchers trying to better understand organizational performance. Counselors, students, human resource professionals, and researchers are among those who benefit from this DOL initiative. Research on the O*NET has started to emerge over the past 10 years. For instance, researchers at the University of Nebraska, Omaha, have developed and tested a web-based job-analytic technique using the O*NET (ReiterPalmon, Brown, Sandall, Buboltz, & Nimps, 2006). More recent work has been focused on matching Occupational Interest Profiles (OIPs) from the O*NET with interest-based assessments often used in career counseling (Rounds, Armstrong, Liao, Lewis, & Rivkin, 2008). An interesting new study has used information from the O*NET to identify the likelihood of experiencing carpal tunnel syndrome based on various hand exposures such as repetitive movements, movement of objects, reliance on hand/finger speed, and so forth (Evanoff, Zeringue, Franzblau, & Dale, 2014). FIGURE 3.5 O*NET Content Model (Information from http://onetcenter.org) Task-oriented approaches to job analysis have been criticized as being potentially too narrowly focused on the tasks for a particular job (Harvey, 1991). Critics say it follows that by focusing microscopically on the tasks of a particular job, we rule out the possibility of making a real comparison across jobs because no job is likely to emerge as similar at such a finegrained level. In other words, we may miss similarities in jobs because of the high level of technological and behavioral specificity of the tasks and thereby fail to detect more abstract similarities across jobs. Worker-oriented approaches to job analysis address this very problem. Worker-Oriented Techniques Compared to task-oriented techniques, worker-oriented techniques focus less on the specific tasks done on a job and more on the human characteristics that contribute to successful job performance (Sackett & Laczo, 2003). E. J. McCormick (McCormick, Jeanneret, & Mecham, 1972) provided a very clear distinction between the two types: “The task-oriented concept typically would be reflected by the use of specific task statements. . . . In turn, the worker-oriented concept typically would be reflected by the use of descriptions of reasonably definitive human behaviors of many kinds” (p. 348). Worker-oriented methods are particularly effective for analysts who are comparing multiple jobs because they allow for comparisons to include a focus on KSAOs. The Job Element Method (JEM) was designed to identify the characteristics of superior workers in a particular job (Primoff & Eyde, 1988). Note that, in this context, the word element refers to general work behaviors, not to the smallest unit of work activity as defined earlier. More specifically, job elements in the JEM refer to the KSAOs we talked about earlier. The JEM’s aim is to directly connect job analysis to the selection context by identifying the KSAOs that are necessary for successful job performance. The procedure is rather simple. First, SMEs are asked to develop a comprehensive list of job elements, as well as subelements or employee characteristics that are more specific than the job elements. For instance, one element for a college professor might be “ability to communicate clearly,” and subelements within this element might include “can express ideas in an interesting manner to large groups of people” and “able to explain complex issues to students in a one-on-one environment.” Once the final list of elements and subelements has been generated, the SMEs are instructed to provide work examples of each. Together, the elements, subelements, and work examples provide a complete picture of the job. Job Element Method (JEM) A worker-oriented approach to job analysis that was designed to identify the characteristics of superior workers in a particular job. A common criticism of this approach is that the JEM ignores the specific job tasks, making it difficult to demonstrate that a particular element is related to the job (Gatewood & Feild, 2001)—in other words, a JEM approach may identify “ability to communicate clearly” as an important element, but this isn’t a specific job task and may leave us wondering what particular task is important and how the element is related to it. This criticism is consistent with other researchers’ (e.g., Harvey, 1991) concerns that job analysis needs to remain focused on observables instead of inferring hypothetical worker traits or abilities. However, much of what we do in psychology involves inferences of some sort, so excluding them seems unreasonable. We can resolve this paradox by realizing that an important focus of job analysis is to ensure that when inferences are drawn, they are based on observable work-related data. The job analyst can choose to focus only on work activities or only on worker attributes or on both, which is consistent with the hybrid approach discussed earlier. Perhaps the best-known job analysis method is the Position Analysis Questionnaire (PAQ), a standardized instrument that focuses on general work behaviors (McCormick et al., 1972). The PAQ consists of 195 items or elements, 187 of which describe general work behaviors, work conditions, and job characteristics. The incumbent employee or SME decides whether each item pertains to the job in question; each one that does is evaluated in terms of various dimensions, such as the extent to which it is used on the job, its importance to the job, and the amount of time on the job that is spent on this behavior. These 187 job elements are organized along six dimensions: information input, mental processes, work output, relationships with other persons, job context, and other job characteristics. Table 3.1 lists these six dimensions, along with two examples of each. A profile of the job emerges as a result of this rather large data collection effort. Position Analysis Questionnaire (PAQ) A widely used job analysis instrument that focuses on general work behaviors. PRACTITIONER FORUM John F. Binning PhD, 1985, I/O Psychology, The University of Akron Director, Research and Development The DeGarmo Group, Inc. The DeGarmo Group specializes in designing and administering preemployment selection tests. One of our assessment tools is the DeGarmo Personality Inventory (DPI). The DPI is delivered over the Internet, and job applicant responses are scored and available to client human resource (HR) representatives. What happens between the time an applicant presses the “Finish” button and the HR representative reviews the results is where the real psychological action is! You will read in later chapters about the Big Five personality factors and how different jobs require different profiles of personality traits. A challenge for HR and talent acquisition professionals is how to determine which personality traits are critical for a particular job. We decided to leverage the vast information available in O*NET to guide the process of determining personality requirements for different jobs. Our approach unfolds something like this: When an organization wants to hire for a particular job, we find that job in O*NET, and among other information, there are importance ratings for a variety of work styles. The work style ratings were determined by having job-knowledgeable people complete job analysis questionnaires about how important the different styles are for different jobs. The work styles can be interpreted in terms of the Big Five personality factors. For example, one of the O*NET work styles is attention to detail, and this is relevant to the Big Five factor of conscientiousness. We believe that each work style can be meaningfully mapped onto Big Five factors, and we have automated this process. For any of the thousands of jobs described in O*NET, we can automatically translate the work style ratings into importance scores on the Big Five factors (or facet subfactors) measured by the DPI. Once we know which DPI scores are important for a job, we have our IT staff write the computational code to score job applicants’ responses to determine how well their personalities “fit” the demands of that particular job. To summarize, when a job applicant presses the “Finish” button, his or her responses are mathematically “crunched” into various personality scores and then given different mathematical weights based on the O*NET information, and a final score and interpretive report are generated indicating how well the applicant’s personality matches the specific demands of the job. Client HR representatives then use this information to aid in making hiring decisions. This way we apply the science of job analysis and personality to the needs of various organizations. Applying Your Knowledge 1. How might the O*NET system be more helpful in job analysis to the DeGarmo Group than the DOT? 2. Given the changing nature of work, why is job analysis so important? How is the DeGarmo Group demonstrating this fact? 3. The DeGarmo Group specializes in trying to match the best applicants to its clients’ needs. But what other job-analytic steps are critical for each client organization to take in order to maximize productivity in the hiring process? TABLE 3.1 PAQ Dimensions and Elements Dimension Description Examples Information input Where and how does the worker get the information he uses in performing his job? Use of written materials Near-visual differentiation Mental processes What reasoning, decision-making, planning, and information-processing activities are involved in performing the job? Level of reasoning in problem solving Coding/decoding Work output What physical activities does the worker perform, and what tools does she use? Use of keyboard devices Assembling/disassembling Relationships with other persons What relationships with other people are required in performing the job? Instructing Contacts with public, customers Job context In what physical or social contexts is the work performed? High temperature Interpersonal conflict situations Other job What activities, conditions, or characteristics characteristics other than those described above are relevant to the job? Specified work pace Amount of job structure Source: McCormick (1979) Despite its historically widespread use, the PAQ has been the target of at least three major criticisms. First, some data indicate that the reading level of the PAQ is at least college level, if not graduate level (Ash & Edgell, 1975). This is a problem because it limits who can competently do the job analysis and rules out the less educated incumbents in many jobs. Second, the PAQ does not seem well suited for managerial jobs; and, third, the items themselves are too abstract (Harvey, 1993). These criticisms have led other job analysis researchers and practitioners to develop alternative workeroriented instruments. The Common-Metric Questionnaire (CMQ), developed by Harvey (1993), is one such instrument. Two basic beliefs about worker-oriented approaches to job analysis are reflected in the CMQ: (1) The instrument must comprehensively describe work activities using a common set of items written at a more behaviorally abstract level than the typical task statements, and (2) the rating scale must have the same meaning across all jobs. The focus here is on describing jobs at a level that allows the descriptions to be compared across jobs, as opposed to describing them so abstractly that differences across jobs are lost. Common-Metric Questionnaire (CMQ) A worker-oriented job analysis instrument that attempts to improve the generalizability of worker-oriented approaches through the use of items focused on slightly less general work behaviors. Think back to such task-oriented approaches as the Task Inventory Approach and Functional Job Analysis. In these cases, the focus is on extremely specific tasks conducted on the job, making it very difficult to uncover any similarities across jobs. But the worker-oriented approaches, best exemplified by the PAQ, may be so abstract as to result in the emergence of dissimilar jobs with very similar profiles. For instance, what jobs do you think might emerge with a profile similar to that of police officer? You probably listed things like firefighter, prison guard, and private detective. However, research indicates that one answer is the job of homemaker (Arvey & Begalla, 1975). This is probably not one of the jobs you had on your list, but the profiles tend to be similar because both police officers and homemakers are often involved in handling emergencies and troubleshooting. However, I would argue that if the PAQ elements were a little less abstract, these jobs would not emerge as so similar. A CLOSER LOOK Why are the job profiles of police officer and homemaker so similar according to the PAQ? How would the CMQ address this? The CMQ is computer-based, consisting of 2,077 items that are organized along 80 dimensions. Obviously, one issue with respect to this instrument is its length, but the author reports that over 90% of those who have used the CMQ completed it in less than three hours (Harvey, 1993). A sample question asks whether a particular job involves a skill that requires instruction at a college or university, along with the response scale for that item; the response scale displays the criticality of this particular skill. The items for all 80 dimensions are set up in a similar way. Its items are more behaviorally specific than those included on the PAQ, its reading level is much lower, and it is relevant for both managerial and nonmanagerial jobs. Advances and Issues in Job Analysis Practice and Research Technology has played a role in the development of job analyses. For instance, researchers have demonstrated how to use the Internet to generate job-analytic information via the O*NET (Reiter-Palmon et al., 2006). In their work, they emphasized the use of the web to gather and organize job-analytic information with minimal intervention and guidance from a consultant. Other researchers have worked to develop metrics, or quantitative rules of thumb, to help with analyzing and using the mass of job analysis data resulting from technological innovations like the O*NET (McEntire, Dailey, Osburn, & Mumford, 2006). They have developed metrics that would allow organizations to make better use of the existing O*NET data. These metrics, or mathematical formulae, can be used with the O*NET database to answer specific questions in the areas of selection, training, and development, for example, who is the best employee for a particular promotion. Given the ever-increasing use of the O*NET database in organizational analysis, the development and use of related metrics hold even greater potential for organizations. There has also been considerable advancement regarding issues of SME ratings. For instance, two studies have examined the role of personality and motivation in SME job analysis ratings. One investigation examined the potential role of the self-serving bias in affecting job analysis ratings (Cucina, Vasilopoulos, & Sehgal, 2005). The authors argued that SMEs may engage in self-serving behavior when asked to identify which traits are important for the jobs. This would suggest that an SME might weight intelligence, flexibility, and creative thinking as important characteristics for successful employees in their job even though others might more objectively not view these characteristics as terribly important. In a more sophisticated study testing self-presentation in SME ratings, it was found that SME incumbents tended to endorse ability statements as important for job effectiveness even when those abilities were really nonessential (Morgeson, Delaney-Klinger, Mayfield, Ferrara, & Campion, 2004). Individuals were presenting themselves in a positive light to suggest that, in order to do the job that they currently did, one needed to have a plethora of abilities. Individuals may do this unconsciously and without any real plan to alter the objective job analysis, but the result is job analysis data that are inaccurate—and the implications and potential detrimental effects to the organization are very real. For instance, requiring an ability for a particular job because the job analysis identified it as important makes good sense; however, if the ability really isn’t essential for the job, it may lead to many competent individuals not being hired for lack of this ability. Other researchers found that one’s attitude toward one’s job may bias one’s job analysis ratings; this was especially the case for tasks that were objectively coded as discretionary (Conte, Dean, Ringenbach, Moran, & Landy, 2005). By “discretionary,” the authors mean tasks like “join and participate in community and social organizations,” which may not be required as often as other tasks. It appears that SMEs (in this case, incumbents) allow their attitudes about their own jobs as well as the extent to which the task is required to impact job analysis ratings. Further, a recent examination of O*NET data that focused on the source of the job analysis ratings concludes that using both job analytic experts and incumbents to provide job data is the best approach because each may bring meaningful and useful information to the situation (Wamslet, Natali, & Campbell, 2012). Another recent study argues that the appearance of unreliable competency ratings made by incumbents is due to “job crafting” (Lievens, Sanchez, Bartram, & Brown 2010). For instance, in complex jobs that allow incumbents to make choices that allow them more creativity and flexibility, there are more opportunities for these incumbents to exercise discretion and “craft” their jobs in particular ways. Because employees can tweak their jobs, resulting in something that looks quite different from the way the jobs were originally defined, incumbent ratings of competencies may not agree with the competencies from the original job analysis. Results found that 25% of the differences in the two sets of competency ratings appear to be meaningful differences that emerged because of the job crafting (Lievens et al., 2010). That is, incumbents don’t see their jobs as the same—even though the jobs were originally defined as the same—because they have crafted their job to be different based on their skills and weaknesses. Variance in incumbent ratings may not be error at all, but rather a reflection of how incumbents view and craft their own jobs. Therefore, organizations might be wise to encourage employees to craft jobs in ways that are consistent with the strategic plan of the organization. A newer perspective on job analysis is taken by some recent researchers who argue that we need new types of job analyses. Traditional job analyses assume stable jobs, but jobs have a tendency to change and morph into different-looking jobs, leading some to argue for a novel approach to job analysis (Singh, 2008). With the changing of environments in North American organizations, we need more strategic job-analytic approaches in which experts consider not just the current status of jobs but also how jobs are likely to change and evolve in the future. Strategic job analysis is an intentional, systematic process of collecting current and future work-related aspects of a job within the organization’s strategic context (Singh, 2008). Although not much has been written about this approach yet, it still represents a potential trend in the discipline. For more information on the concept of strategic job analysis, see J. S. Shippmann’s 2013 book Strategic Job Modeling: Working at the Core of Integrated Human Resources (Psychology Press). strategic job analysis An approach to job analysis that considers the status of jobs as they currently exist but also factors in how jobs are likely to change in the future as a result of anticipated organizational or industry changes. TECHNICAL TIP In Chapter 2 we talked about variance as a measure of dispersion, or the “spreadoutedness” of the scores. Here we are talking about job crafting being responsible for those differences, or variance, in job ratings. DEFINING THE JOB: DESCRIPTIONS AND SPECIFICATIONS The basic goal of job analysis is to define the job in terms of its component parts. As noted earlier, job analysis yields three major outcomes: job description, job specifications, and job evaluation. The first of these will be discussed in a separate section. The second outcome, job description, is a written statement of what the jobholders do, how they do it, and why they do it (Cascio & Aguinis, 2010). It presents the task requirements of the job. The job description typically includes the job title and descriptions of the tasks and machinery involved, and it sometimes includes information about the working conditions and physical environment, social environment, and conditions of employment. Think about the job of college professor and ask yourself what college professors do, how they do it, and why they do it. Upon answering those questions, you might arrive at the following job description, which looks very similar to the DOT entry for faculty member, presented earlier in Figure 3.4: job description As an outcome of job analysis, a written statement of what jobholders actually do, how they do it, and why they do it. College Professor: Teaches college or university courses within a particular curriculum (e.g., psychology, mathematics, history) to classes of undergraduate and graduate students. Prepares and delivers lectures that conform to a predetermined course outline. Stimulates class discussions in smaller informal groups. Evaluates students’ knowledge and achievement through reading, writing, and other assignments as well as through examinations that are graded so as to provide feedback. Conducts research in his or her area of expertise and disseminates the findings of that research in appropriate outlets (e.g., scientific journals, books) to further the knowledge in the field. Provides mentoring and advising to students who are seeking academic degrees. Serves on university committees that serve the interests of students, the university, the government, and the community. The third outcome derived from the job analysis is a description of the people requirements that are reflected in the job specifications (often called job specs). The job specifications delineate the KSAOs deemed necessary to perform the job. Let’s return to our college professor example. We’ve generated a job description, but what would comprise the job specifications or KSAOs necessary to be a successful college professor? Here is one such list: Reading and writing at the post-college level Advanced knowledge verified by an advanced degree (master’s or doctorate) Willingness or desire to work with others Ability to communicate clearly Ability to communicate in an interesting manner Methodological and statistical skills Mentoring and advising skills A deep knowledge/understanding of one’s own discipline job specifications An outcome of job analysis delineating the KSAOs deemed necessary to perform a job; often called job specs. One way to view the relationships among job analysis, job description, and job specifications is to view the job description as stemming directly from the job analysis and the job specifications as being inferred indirectly from the job analysis through the job description. (Look back at Figure 3.1 for this organizational scheme.) For instance, we might use the JEM to develop a picture of the job and to write a job description (like the preceding one) that includes mentoring and advising students who are seeking academic degrees. From the job analysis and this job description, we might infer that successful college professors must have mentoring and advising skills. In short, the job analysis, job description, and job specifications are necessarily intertwined. As you may recall, many experts argue that job analysis should be focused only on observables; inferences about KSAOs from the job analysis should be avoided (Harvey, 1991). Of course, one of the major purposes of job analysis is to identify KSAOs that can be used in the selection of employees. This process is covered in Chapter 6, but it’s worth noting here that even if one limits job analysis to observables that result in a job description, the job analysis and job description must be used to infer job specifications that are of paramount importance for selection and placement. The bottom line is that, as I/O psychologists, we are not simply focused on describing the job; indeed, our longer-term goal is the use of this job analysis information for selection, placement, performance appraisal, and other HR functions. Competency Modeling In Chapter 1, I defined competencies as the skills, behaviors, and capabilities that allow employees to perform specific functions. Obviously, this definition is very similar to the one I have specified for KSAOs in this chapter. In fact, many experts consider competencies and KSAOs to be the same thing (e.g., Schippmann et al., 2000). Another way to think about this relationship, though, is to view competencies as sets or groupings of specific KSAOs that allow employees to perform specific organizational functions (Reilly & McGourty, 1998). For example, interpersonal communication may be considered a competency. That competency, in turn, may consist of several KSAOs such as (1) the knowledge of different communication styles, (2) the skill to communicate positive and negative feedback to individuals, and (3) the ability to understand what individuals need to hear from their supervisor to increase their motivation. Competency modeling may parallel job analysis in that both involve describing jobs in a careful and methodical way so that HR practices such as selection and placement can be based on this information (Schippmann et al., 2000). Differences have been observed as well. For example, competency modeling is believed to be more worker-oriented and job analysis more taskoriented (Schippmann et al., 2000). Competency modeling is usually much broader in focus because it tries to build a full framework for HR functioning by defining the responsibilities, skills, knowledge, abilities, personal attributes, and business challenges that are relevant for a particular organization (see Sanchez & Levine, 2009 and Stevens, 2012 for a thorough comparison of the two approaches). Although distinctions exist between the job-analytic approach to identifying KSAOs and the competency modeling approach to building competencies, the two terms are similar enough for our purposes that I will use them interchangeably throughout the text. Please understand, however, that job analysis tends to be a bit more methodologically rigorous, and competency modeling tends to consider organizational strategy a bit more and takes a broader look at the organization. I/O TODAY The Future of Job Analysis Although job analysis has historically been a key step in developing solid I/O practices, many researchers and practitioners have suggested that this approach is outdated given the ways that jobs have changed over the past few decades. Technological innovations lead to products that need rapid updates, globalization has increased the pace of competition, and organizations need to be flexible to meet customer needs. As a result, workers are less likely to be hired to do a specific job for the rest of their career; instead, a valuable worker will be able to grow and change with the needs of the company. This, in turn, means that a job analysis can quickly become obsolete. One potential solution to this problem is competency modeling. In practical terms, competency modeling focuses more on identifying people who have a broad set of capabilities that will help them to adapt as the job changes. In other words, for the role of a retail manager, a job analysis might indicate that the individual must know general management principles and have good communication skills. Competency modeling, meanwhile, might indicate that the manager has a competency in “engaging leadership”—a general competency that might include specific skills such as communication, critical thinking, and charisma. Often, organizations will use competencies such as “mentoring and teaching” or “strategic thinking” at all levels of the organization. A sales representative’s version of “mentoring and teaching” might look different from that of a manager’s —she might help train new employees or provide in-the-moment feedback to her peers—but arguably, a sales representative who can demonstrate these skills has the potential to one day be an effective manager. This example helps to illustrate both the benefits and difficulties of competency models. Having broad competencies such as “mentoring and teaching” can provide an organization with a consistent language that can be used with all employees to discuss expectations and performance. Competencies also make it easier to adapt these expectations as the organization changes. However, because competencies tend to be umbrella terms that encompass many different KSAOs, they are not as specific and detailed as a traditional job analysis would be. So while a competency might be more flexible and parsimonious, it also lacks specificity. Despite these shortcomings, competency modeling is becoming more common among organizations. Researchers have also found that consumers such as HR managers and CEOs are much more interested in adopting competency models rather than conducting numerous job analyses. The challenge for I/O psychologists will be to ensure that these competency models are developed with just as much care and rigor as a job analysis would be. Discussion Questions 1. What situations might lead you to recommend that an organization adopt a competency model approach over a more traditional job analysis approach? 2. How could you use principles for job analysis to help ensure you have created a high-quality competency model? 3. In his 2012 article in Human Resource Development Review, Gregory Stevens suggests that using solid research methodology and avoiding “buzzwords” are two possible best practices that practitioners can follow. What other “best practices” might be important to keep in mind if an organization chooses to use competency modeling? THE MANY PURPOSES OF JOB ANALYSIS As I suggested earlier, job analysis is the foundation for much of industrial psychology. There are two reasons for this. First, a job analysis is necessary to the success of human resource functions. In other words, without a job analysis, many of the HR functions in organizations would be carried out ineffectively. Second, an increased emphasis on laws associated with the workplace has made it more desirable for companies to use job analyses as their first line of defense when accused of an unfair practice. Table 3.2 presents two cases that demonstrate the need for adequate job analysis to prevent unfair or discriminatory hiring and promotion practices. We will discuss legal issues in more depth in the next few chapters; for now, let’s consider job classification, an important purpose served by job analysis. TABLE 3.2 Legal Importance of Job Analysis Case Description Legal Ruling Griggs v. Duke Power (1971): Among other tests/assessments, Duke Power was using measures of cognitive ability (Bennett and Wonderlic) to make hiring decisions into low-level laborer positions. This resulted in adverse impact against blacks who were hired at a much lower rate than whites. Though the defendant had no job analysis data linking cognitive ability to job performance, they argued that they used a professionally developed test and didn’t mean to discriminate against blacks. The court found the lack of intent to discriminate irrelevant and ruled that tests, even if professionally developed, must be job-related. The burden of proof of demonstrating the job relatedness (business necessity) of the tests/assessments falls on the defendant and the court strongly supported the use of job analysis in this regard. Albermarle Paper Company v. Moody (1975): The company was using the same two cognitive ability measures as Griggs, but had not done a validation study until just prior to trial and then did it poorly without a job analysis. The court ruled that a test validated in one job may be used in another, but only if it is clearly demonstrated to be job-related for the current job. Further, they ruled for the plaintiff because Albermarle had done no analysis of the attributes of the jobs or the skills needed to perform the job. Job Classification Jobs are often categorized in terms of job families, each of which can then be used as a level of analysis for various personnel decisions. For instance, Word Processing Specialist I, Word Processing Specialist II, and Data Technician I may all be declared parts of a job family called “computer support staff” if the job analyses conducted on each job demonstrated enough overlap to warrant such a designation. If they did, all jobs within this job family would operate under the same benefits package, including number of allowable sick days, number of vacation days, and amount of health benefits. In addition, if a company’s 2,000 jobs can be grouped into 100 job families, many of the company’s personnel practices will be simplified. TECHNICAL TIP Remember that we talked about the importance of test validity in Chapter 2. In particular, criterion-related validity focuses on whether the test is a good predictor of performance. With the legal cases described here, you can see the practical importance of test validity. Criterion Development and Performance Appraisal Performance appraisal is certainly one of the most frequent uses of job analysis data. As we will discuss in Chapters 4 and 5, performance appraisal is one of the most important functions performed by human resource departments. Of course, to conduct a useful and fair performance appraisal, organizations must have clearly stipulated goals, objectives, and criteria on which employees can be evaluated. These criteria should come directly from the job analysis. For instance, if college professors are going to be evaluated annually with respect to their success in teaching, the college must be able to demonstrate that teaching is an important part of a professor’s job. (This should be an easy thing to do, but it still needs to be verified by the job analysis.) Suppose that I am evaluated based on the reactions of my students to my classes (and, in fact, I am evaluated in this way every semester), but the job analysis done by my school does not demonstrate that teaching is an integral part of my job. This would be neither fair nor appropriate. Here, of course, I would argue that the problem is not with the criteria but with the job analysis, which did not identify an important element of the job. This example, too, demonstrates the importance of doing a thorough and accurate job analysis because the job analysis is used in important ways. Selection and Placement There is perhaps no area of industrial psychology for which job analysis is more important than selection and placement (see Chapters 6 and 7). Indeed, a major purpose of job analysis is to help industrial psychologists identify the KSAOs that are necessary for successful performance on the job. Let’s turn to the world of construction for an example. In doing a job analysis for the job of an architect, we conclude that spatial skills and knowledge of some basic engineering principles are required. Therefore, when screening applicants, we look for individuals who seem to be knowledgeable and skilled in these areas and hire them. This scenario makes sense, but it may also be too simplistic. For instance, we might ignore the part of our job analysis indicating that architects also need to have some drafting (or drawing) skills because they need to be able to develop plans for new buildings. Given this flaw in our process, we may hire knowledgeable and spatially skilled individuals who do not have the drafting skills to put their ideas on paper. In short, because we weren’t careful in conducting and interpreting our job analysis, we may have made costly personnel decisions. Does this cost our company money? It sure does, and it also may cost individuals their jobs. Job Design and Redesign Job analysis may sometimes uncover problems with a particular job that the organization may want to address. For instance, interviews with incumbents may reveal that some of the current equipment used on the job needs to be updated because it cannot keep pace with other, newer equipment during the production process. A company may use this information to purchase newer equipment or to redesign the job so that the current equipment will work more efficiently within the process. Alternatively, PAQ or CMQ data may indicate that employees spend far too much time on a particular facet of the job, information that the company may use to initiate a job redesign intervention to make the work process more efficient. We can see that in this area of I/O psychology, job analyses are often used to make alterations in the work process with the intent of making the company run more efficiently. Morgeson and Humphrey (2006) developed the Work Design Questionnaire (WDQ) as a comprehensive measure of job design to replace the incomplete and narrow existing measures. The WDQ consists of 77 items along 21 dimensions of work characteristics. The 21 work characteristics can be organized into: (1) task characteristics such as autonomy and task variety; (2) knowledge characteristics that reflect traditional KSAOs; (3) social characteristics that include social support and feedback from others; and (4) contextual characteristics, composed of physical demands and work conditions. Results suggest that the WDQ is a solid measure of work design that relates appropriately to various elements of the O*NET database—it has become a very popular measure for those doing research on job design. Training In many instances, employees are hired into an organization without all the KSAOs necessary to do every facet of their jobs. Sometimes even highly skilled new employees require training to learn the procedures used in a particular organization (see Chapter 8). Job analysis can be used to identify areas in which training programs need to be developed. Ideally, job analysis should identify what work behaviors occur on the job so that the human resource department can develop training programs around these general and specific work behaviors. In addition, given that jobs are dynamic—in that they change considerably with changes in technology, in the workforce, and in economic demands—job analysis can be used to identify the extent to which job behaviors have changed to match the times. This information can then be used to develop and target training programs for existing employees. JOB EVALUATION Although we like to think that our job is as important to our particular organization as anyone else’s, the reality is that some jobs are more important than others. Job evaluation is a technique based on job analysis that attempts to determine the value or worth of particular jobs to organizations so that salaries can be set accordingly. Most of us would agree that the clerk who works in the hospital restaurant makes a less important contribution to the organization than the chief surgeon. Job evaluation is about trying to quantify differences of this sort so that salaries can be set in a fair manner based on the value or contribution of jobs. Because compensation is often the largest budget item in many companies, it has an important effect on profits and, therefore, attracts a good deal of attention from organizational administrators and HR practitioners. job evaluation As an outcome of job analysis, a technique that attempts to determine the value or worth of particular jobs to organizations so that salaries can be set accordingly. Point System There are various approaches to job evaluation, but I will focus on the one most frequently used: the point system. This approach involves estimating the value of jobs based on points assigned to various predetermined dimensions (Milkovich & Newman, 1984). Among the most common of these dimensions—referred to as compensable factors to indicate that employees are compensated based on them—are effort, skill, responsibility, and working conditions. The idea here is that jobs requiring more effort and skill, involving more responsibility, and taking place in less desirable working conditions (e.g., outdoors in subzero temperatures versus climate-controlled office buildings) should be more highly compensated than jobs that don’t have these characteristics. point system The most common approach to job evaluation, which involves estimating the value of jobs based on points assigned to various predetermined dimensions. compensable factors Dimensions or factors that are used to rate jobs, indicating that employees are compensated based on these factors. Examples include effort, skill, responsibility, and working conditions. The steps involved in conducting a point-system job evaluation are rather simple. First, managers or some other group of SMEs use a job analysis to identify the compensable factors. Second, each job is assigned points by a compensation committee with respect to the degree that each has these compensable factors. Third, the points assigned to each job are summed across the factors to arrive at a total score. Fourth, these points are then used to assign wages, with the goal being to maintain equity so that jobs of greater value are better compensated than those of lesser value. Finally, the scores can be plotted against the current wage for existing jobs, providing a clear picture of whether a job is currently undercompensa