Designing a pay structure case study solution

when designing a pay structure the first step involved is designing a base pay structure pdf free download
Dr.AshleyBurciaga Profile Pic
Published Date:05-07-2017
Your Website URL(Optional)
INSTRUCTOR’S MANUAL Designing A Pay Structure CASe STUDy AND INTegRATeD APPLICATION exeRCISeS Designing A Pay Structure By Lisa A. Burke, Ph.D., SPHR Instructor’s Manual TOTAL REWARDS ©2008 SHRM Lisa Burke, Ph.D., SPHR 1INSTRUCTOR’S MANUAL Designing A Pay Structure 2 ©2008 SHRM Lisa Burke, Ph.D., SPHRCASe STUDy AND INTegRATeD APPLICATION exeRCISeS Designing A Pay Structure Designing a Pay Structure About this C Ase Learning Objectives In this case, students will learn to In this case, upper-level undergraduate or master’s level HR students design a pay structure. To do so, will learn how to design a pay structure using a case scenario and they will: integrated application exercises. • Write a job description using the ONET website. • Use the point method to conduct This case is rated as slightly challenging and requires familiarity with a job evaluation. and use of the Internet and Microsoft Excel. Instructors can make the • Analyze pay survey data for benchmark jobs. case and associated exercises less challenging by eliminating certain • Create a market pay line in Excel. tasks assigned in the case, or may increase the difficulty by adding • Create a pay policy line based other relevant tasks and questions. Teaching notes accompany the on a stated pay-level strategy. case. Instructors who have previously taught compensation courses, • Create pay grades. • Establish pay ranges. are familiar with the Internet and Excel, have work experience with pay systems, or who conduct research in compensation area may find the case easier to facilitate. Recommended Reading Milkovich, G., and Newman, J. (2008). Compensation. McGraw- Hill Irwin. Chapters 1-8. This case complements the first 40 percent or so of chapters in most compensation textbooks. The amount of time the case takes for students to complete will depend on students’ skills and education level. Time can be allotted during class in a computer lab so the instructor can facilitate students’ work on the associated exercises, but some outside-of-class work is also necessary. Table of Contents CAse with teAChing no s .................................................................................................... 2 Psibe soll ution for eACh CAse exeC r is e ....................................................29 snt version of the CAs e ..........................................................................................45 ©2008 SHRM Lisa Burke, Ph.D., SPHR 1 detu os teINSTRUCTOR’S MANUAL Designing A Pay Structure Case with Teaching Notes Introduction to Compensation and Designing a Pay Structure Compensation is a critical area of human resource (HR) management, and one that can greatly affect employee behavior. To be effective, compensation must be perceived by employees as fair, competitive in the market, accurately based, motivating and easy to understand. HR professionals might create the pay structure for their organization, or they might work with an external compensation consultant. There are several steps to designing a pay structure: job analysis; job evaluation; pay survey analysis; pay policy development; and pay structure formation. Each step is briefly explained below. For a more extensive discussion, please review Milkovich & Newman (2008). Step 1: Job Analysis Job analysis is the process of studying jobs in an organization. The outcome of this process is a job description that includes the job title, a summary of the job tasks, a list of the essential tasks and responsibilities, and a description of the work context. Also included are the knowledge, skills and abilities needed to perform the job. Step 2: Job Evaluation Job evaluation is the process of judging the relative worth of jobs in an organization. The outcome of job evaluation is the development of an internal structure or hierarchical ranking of jobs. Job-based evaluation is used more often than person-based evaluation, and so the former will be the focus in this case. There are three methods of job-based evaluation: the point method (which is the most commonly used); ranking; and classification. Job evaluation helps to ensure that pay is internally aligned and perceived to be fair by employees. Step 3: Pay Policy Identification Pay policy identification is the process of determining whether the organization wants to lead, lag or meet the market in compensation. The pay policy or strategy will likely influence employee attraction and retention. Pay policies can vary across job families (i.e., groups of similar jobs) and job levels if the top management feels that different strategies can be effective in different areas of the organization. Step 4: Pay Survey Analysis Pay survey analysis is the process of analyzing compensation data gathered from other employers in a survey of the relevant labor market. Gathering external pay data (e.g., base pay, bonuses, stock options and benefits) is essential to keep the organization’s compensation externally competitive within its industry. Employee attraction and retention can be improved by maintaining externally aligned pay structures. Step 5: Pay Structure Creation Pay structure creation is the final step, in which the internal structure (Step 2) is merged with the external market pay rates (Step 4) in a simple regression to develop a market pay line. Depending on whether the organization wants to lead, lag or meet the market, the market pay line can be adjusted up or down. To complete the pay structure, pay grades and pay ranges are developed. In this case, upper-level undergraduate or graduate HR students will design a pay structure using a case scenario and integrated application exercises. 2 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure CAse Learning Objectives You are the newly hired human resource (HR) director for an engineering consulting In this case, students will learn to firm that is expanding its operations to Chattanooga, Tenn. The organization is design a pay structure. To do so, headquartered in Indianapolis, Ind. Based on the organization’s mission statement, they will: you know the firm strives to create customized and technically proficient electrical engineering plans for regional clients. The following personnel are required to • Write a job description, using the start the Chattanooga operation (the numbers in parentheses indicate the number ONET website. of positions): • Use the point method to conduct a job evaluation. • Director of regional operations• Lead engineer (3) • Analyze pay survey data for • Assistant to the director of • Engineer (6) benchmark jobs. operations • Engineering associate for special • Create a market pay line using • Operations analyst (2) projects Excel. • Operations trainee• Manager of information systems • Establish a pay policy line based • HR director (this is you)• Senior information systems analyst on a pay level strategy. • Administrative assistant in HR• Information systems analyst • Create pay grades. • Benefits manager• Security guard • Establish pay ranges. • Benefits counselor • Front desk receptionist Recommended Reading • Payroll assistant Milkovich, G., and Newman, J. (2008). Compensation. McGraw- You can see from the list that there are several job families, including operations, HR, Hill Irwin. Chapters 1-8. engineering, information systems and office support. You can now begin the process of designing a pay structure for the organization. Job analysis is central to many HR functions, including compensation, recruiting and training. You need to understand what tasks, duties and responsibilities various jobs will entail before you can assign fair and competitive pay rates. Begin the process by gathering the needed job description information. To do so, you will combine information from ONET (, an online job analysis resource developed by the Department of Labor, and existing internal corporate HR documents (such as previous job descriptions). Each job description includes the job title; a job summary; essential job tasks; the job’s work context; and job-relevant knowledge and skills that an incumbent must possess. Benchmark jobs (jobs that are common and consistent across a wide range of employers) will be the focus of this exercise because they will be used to design the pay structure. Appendix A contains the job descriptions of the benchmark jobs. You have one description left to complete; your first task is to create a job description for the benefits manager position. ©2008 SHRM Lisa Burke, Ph.D., SPHR 3INSTRUCTOR’S MANUAL Designing A Pay Structure » Task A: Create a complete job description for the Benefits Manager position using ONET. Teaching Note Students’ answers for this first task may slightly vary; a possible job description is presented in the solution set. ONET was developed for the U.S. Department of Labor by the National Center for ONET Development. The website is free, easy to navigate, and has a wealth of information. As organizations become increasingly market-driven, job analysis is becoming less commonly performed. Instructors should note to students that it is extremely difficult to design a valid pay structure without accurate job descriptions. If job descriptions are not created, or are created incompletely, it is nearly impossible to create a sound and fair internal ranking or hierarchy of jobs. Also, because the term “benchmark job” may be new to some students, instructors can provide examples not necessarily found in this case, such as a r fi st-line supervisor, accountant, marketing analyst, recruiter and n fi ancial analyst. Benchmark jobs are used to design pay structures because external market data about them is readily available. See Milkovich & Newman (2008) or any other basic compensation text for further discussion. To design a pay structure, there must be a formal way to value the work inside the organization so that pay is awarded fairly. The job evaluation process will help develop this internal work hierarchy. Different evaluation methods, pay strategies, and pay structures will be used for different job families in the organization. You decide to use a job-based evaluation approach for the operations, office support, and HR job families. A skills-based approach will be used for information systems and engineering job families, although it is not included as a task in this case. The security guard and director of regional operations jobs will be assigned pay rates primarily using market pricing and slotted later into the pay structure. Teaching Note To ensure students understand and remember the big picture for designing a pay structure, frequently remind them of the steps outlined at the beginning of the case. They will need to go through each step to complete their pay structure. To keep this case to a reasonable length yet expose students to the reality of many decisions involved in designing compensation systems, the present case has students execute a job evaluation, not a skills-based evaluation. Job evaluation is more common in the workplace because most pay structures are job-based. To illustrate that skills-based evaluation may be recommended or preferred in certain situations, the case discusses that certain job families, such as engineering and information systems (which are more skills- driven), could be evaluated with a skills-based approach. See Milkovich and Newman (2008) or any other basic compensation text for further discussion of skills-based approaches. External market pricing can be used to establish pay rates for uncommon jobs in a specific organization (such as the security guard job in this case) and for jobs where competitive pay is critical to attracting and retaining talent (such as the director of regional operations job). After the pay structure is developed, these jobs would be slotted in later based mostly on relevant external competitors’ pay rates. 4 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure Company representatives from various job levels and families will periodically provide you with input during the job evaluation process. This will help you gain acceptance of the established job structure. You ask this job evaluation committee whether they agree with the specific benchmark jobs identified in the job analysis step (see below). Office Support Operations HR HR Director Assistant to the director of operations Director of regional operations Benefits manager Admin assistant (HR) Operations analyst Benefits counselor Front desk receptionist Operations trainee Payroll assistant Benchmark job. The committee studies the various job titles and asks why the administrative assistant in HR is not included in the HR job family. You explain that administrative assistants perform similar tasks across departments and do not handle function-specific tasks (e.g., HR). You suggest grouping the front-line administrative jobs in a separate job family called office support. The other job families that will be evaluated are operations and HR. Teaching Note There are a number of ways to design a pay structure. The organization should determine what approach is best for them based on business strategy, culture and work content. For example, students may recommend that if the administrative assistant in operations and the administrative assistant in HR perform functional tasks (e.g., operations tasks, HR tasks), that they be included in the respective job family. That would be n fi e. The front desk receptionist job would then likely be slotted later into the pay structure. Benchmark jobs should be identified throughout the internal structure as much as possible, not just at low or high levels of the organization, or else the pay structure may be biased toward the lower or higher job levels. It is important for students to understand that a pay structure should not be developed by one person (e.g., the HR manager) in isolation; input from internal stakeholders is useful in creating buy-in of the n fi al pay structure. That is why a job evaluation committee is created in this case. You decide to use the point method for job evaluation for operations, HR, and office support job families because it is the most commonly used job evaluation method. Next, the compensable factors, degrees and weights of each factor must be determined. With input from the job evaluation committee and your knowledge of the organization’s mission and work content, three common compensable factors are selected: skill, responsibility and effort, each having two specific sub-factors. For example, the compensable factor of skill is comprised of education level and the degree of technical skills. You recommend weighting the skill compensable factor at 50 percent because the organization is very knowledge-intensive and depends heavily on its human capital. Responsibility is weighted 30 percent because each job has the potential to affect other jobs, and effort is assigned 20 percent because problem solving and task complexity are integral across jobs in the organization. ©2008 SHRM Lisa Burke, Ph.D., SPHR 5INSTRUCTOR’S MANUAL Designing A Pay Structure Four degrees should be sufficient for rating the various jobs. For example, the four degrees for education level are identified as: 1=High School/GED 2=Associates 3=Bachelors 4=Masters/Graduate Points are then calculated by multiplying the degrees by the weights. You present an example of how this point scheme is applied to the front desk receptionist benchmark job (see below). The committee agrees with the approach. Compensable Factor Job evaluation for front desk receptionist Degree (1, 2, 3, 4) Weight Points Skill (50%) -Education Level 1 25% 25 -Degree of Technical Skills 1 25% 25 Responsibility (30%) -Scope of Control 1 10% 10 -Impact of Job 2 20% 40 Effort (20%) -Degree of Problem Solving 1 10% 10 -Task Complexity 1 10% 10 120 points The next task is to calculate the job evaluation points for the remaining benchmark jobs using the established compensable factors and specified weights above. In other words, the degrees of each remaining benchmark job must be determined based on a logical rationale and then the total job evaluation points for each benchmark job can be calculated. To do so, consult the job descriptions in Appendix A. 6 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure » Task B: Calculate the job evaluation points for the administrative assistant, payroll assistant, operational analyst, and benefits manager jobs. Provide a rationale for assigning specific degrees to the various jobs. Teaching Note Students’ answers will vary based on the degrees (1 – 4) they assign to various factors for each benchmark job. See the proposed solution for possible answers for Task B. Instructors should ensure that students’ rationales for assigning points are logical and reasonable and that their basic math calculations are correct. Points are determined by multiplying the weights by the degrees. So in the above example, under the compensable factor of education level, a degree of 1 for the receptionist job multiplied by a weight of 25 percent leads to 25 points (1 25). A few important notes to highlight to students about the specific point scheme created above include: The high weight for the skills factor: Skills account for much of the variance in job content and should probably be assigned a significant weight in a point method. The total number of compensable factors: Three factors should be sufficient for capturing variance in most job content. Three are used here, but each is broken down into 2 sub-factors to capture the nuances of the various benchmark jobs. The number of degrees used: The number of degrees can vary for each compensable factor, as needed (e.g., 4 to 8 degrees or levels) based on the extent to which the respective factor varies across jobs. For simplicity, four degrees are used for each factor in this job evaluation plan. To add a level of complexity to the case, instructors could require that students produce a degree key so that the assignment of degrees to various jobs is clearly thought-out and applied consistently (see the degree key for education presented in the case). For general guidelines or further information regarding the point method, see Milkovich, G., & Newman, J. (2008) Compensation, McGraw-Hill Irwin, or any other basic compensation text. After determining the job evaluation points for the remaining benchmark positions, you meet with the president, the head of corporate HR in Indianapolis and the director of regional operations in Chattanooga to discuss a pay level strategy for each job family. One decision resulting from these meetings is that your organization will pay 3 percent above the market in base pay for the HR, operations and office support job families. The group realizes that this lead pay policy will help meet the firm’s customer-focus business strategy by attracting and retaining high-potential employees without incurring labor costs too far above their competitors. Top management also decides to match the market in benefits to contain benefit costs (e.g., health care costs). After analyzing web-based data about benefits offered in your industry by smaller organizations (retrieved from BenefitsLink, SHRM, and Employee Benefits Research Institute) you discern that on average, employee benefits costs are approximately 25 percent of total compensation. Once the pay structure is finalized, you will set benefits at a similar ratio of total compensation to achieve a matching benefits policy. ©2008 SHRM Lisa Burke, Ph.D., SPHR 7INSTRUCTOR’S MANUAL Designing A Pay Structure Teaching Note This case mentions various compensation and benefits decisions that must be made, even if students do not have to perform a relevant task in the designed application exercises. Some pieces of information (like pay level strategy) will inform other tasks they will use in the case, thus providing necessary “context” for the case. Others represent related decisions that organizations make (e.g., benefit strategy). By including these decisions, students will begin to understand the complexity of compensation and benefit design. To ensure that the pay structure is externally competitive, a pay survey will be conducted. For the results of a survey to be valid, the market pay data must be from the relevant labor market for each benchmark job. That is, regional pay data should be gathered because most of the office support, HR and operations jobs will be filled by regional candidates (i.e., within a 90-mile radius of Chattanooga). You develop a streamlined pay survey and administer it to industry competitors. Descriptive organization data (e.g., size, industry, annual revenue) is gathered as well as compensation data for each of the benchmark jobs, including base pay, bonuses, stock options and benefits. Note: All participating organizations will receive the survey results. Surveys are completed and returned by six organizations (referred to as companies A, B, C, D, E, and F) who recruit and hire similar benchmark jobs in the surrounding region. Base pay salary data from the responding organizations are reflected in the following table (pp. 10–14). You have already checked to ensure that summary job descriptions for the benchmark jobs (in the sample data) are appropriately similar to those in your organization (to ensure you are comparing “apples to apples”). The next step is to analyze the pay data and generate weighted means for each benchmark job to use in future parts of the case. 8 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure » Task C: If there were any outliers (i.e., extreme data points) in these data, what would you recommend doing with them? From this point forward, assume no extreme data points exist in the dataset. Second, calculate the weighted means (for base pay) for each benchmark job. Teaching Note Students may be surprised that organizations would respond to a survey of this nature, so it can be helpful to remind them that participating organizations will get useful market data to use for their internal pay structure design and development. Pay data can also be purchased from HR consulting r fi ms and other vendors. Purchasing the data could help avoid “price x fi ing” lawsuits in which organizations are alleged to have worked together to set labor prices. Survey base pay data from the responding companies for each benchmark job are located in the next table you should choose just enough organizations to meet your goals; too much data will be overwhelming. For Task C, students should first identify what they would do with any extreme data points (outliers) in the salary data. Outliers are addressed by examining the number of job incumbents and the min/max pay data. For example, if one of the employers had 240 administrative assistants, that would not be representative of other organizations’ numbers of job incumbents in this job category. Or if one employer had salaries for operations analysts that ranged from a minimum of 35,000 to a maximum of 129,000, that would not be representative of other pay ranges for this job in the pay data. Outliers should be eliminated because they would not be representative of the sample; i.e., they would be extreme data points to be considered separately, if at all. In this set of data, there are no extreme outliers to be concerned about. If instructors would like to add another layer of complexity to this case, they could add some extreme outliers to test students’ ability to identify and eliminate them. Weighted means of base pay should be calculated for each benchmark job from the survey data. Weighted means, as compared to simple means, are calculated to better represent the market data (Milkovich & Newman, 2008). A simple mean would be calculated by adding up the average base pay rates and dividing by the number of organizations (six in this case); but small and large companies would both be given the same weight if using a simple mean. A weighted mean gives equal weight to each job incumbent’s wage and thus is more representative of the data. For example: Mean of employees Co. A 30,000 2 Co. B 15,000 10 The simple mean salary is 22,500. (30000 + 15000) / 2 = 22500 But the weighted mean salary is 17,500. (2/12 30000) + (10/12 15000) = 17500 If we examine the raw data (i.e., the number of employees making various salaries), we can see that the weighted mean is more representative of the data presented. See the proposed solution for the weighted mean answers in this case. ©2008 SHRM Lisa Burke, Ph.D., SPHR 9INSTRUCTOR’S MANUAL Designing A Pay Structure Company of Job Incumbents Base Pay A Front Desk Receptionist 1 Average 21,000 Minimum Maximum B Front Desk Receptionist 2 Average 22,000 Minimum 21,000 Maximum 23,000 C Front Desk Receptionist 1 Average 18,000 Minimum Maximum D Front Desk Receptionist 2 Average 21,000 Minimum 20,000 Maximum 22,000 E Front Desk Receptionist 2 Average 18,500 Minimum 18,000 Maximum 19,000 F Front Desk Receptionist 1 Average 17,500 Minimum Maximum 10 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure Company of Job Incumbents Base Pay A Administrative Assistant 4 Average 25,000 Minimum 21,000 Maximum 28,000 B Administrative Assistant 4 Average 31,000 Minimum 27,000 Maximum 34,500 C Administrative Assistant 3 Average 30,000 Minimum 29,000 Maximum 32,000 D Administrative Assistant 5 Average 33,000 Minimum 28,000 Maximum 34,000 E Administrative Assistant 4 Average 29,000 Minimum 27,000 Maximum 30,000 F Administrative Assistant 4 Average 28,000 Minimum 27,000 Maximum 30,000 ©2008 SHRM Lisa Burke, Ph.D., SPHR 11INSTRUCTOR’S MANUAL INSTRUCTOR’S MANUAL Designing A Pay Structure Designing A Pay Structure Company of Job Incumbents Base Pay A Operations Analyst 2 Average 55,000 Minimum 50,000 Maximum 60,000 B Operations Analyst 4 Average 57,000 Minimum 54,000 Maximum 59,000 C Operations Analyst 3 Average 56,000 Minimum 54,000 Maximum 58,000 D Operations Analyst 5 Average 58,500 Minimum 52,000 Maximum 61,000 E Operations Analyst 3 Average 59,000 Minimum 57,000 Maximum 61,000 F Operations Analyst 3 Average 54,000 Minimum 53,000 Maximum 55,000 12 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure Company of Job Incumbents Base Pay A Payroll Assistant 2 Average 35,000 Minimum 34,000 Maximum 36,000 B Payroll Assistant 3 Average 34,000 Minimum 32,000 Maximum 35,000 C Payroll Assistant 1 Average 35,000 Minimum Maximum D Payroll Assistant 3 Average 35,000 Minimum 33,000 Maximum 37,000 E Payroll Assistant 2 Average 36,000 Minimum 35,000 Maximum 37,000 F Payroll Assistant 2 Average 29,000 Minimum 27,000 Maximum 31,000 ©2008 SHRM Lisa Burke, Ph.D., SPHR 13INSTRUCTOR’S MANUAL Designing A Pay Structure Company of Job Incumbents Base Pay A Benefits Manager 1 Average 62,000 Minimum Maximum B Benefits Manager 2 Average 61,500 Minimum 61,000 Maximum 62,000 C Benefits Manager 1 Average 60,000 Minimum Maximum D Benefits Manager 3 Average 64,000 Minimum 62,000 Maximum 65,000 E Benefits Manager 2 Average 63,000 Minimum 62,000 Maximum 64,000 F Benefits Manager 1 Average 66,000 Minimum Maximum 14 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure Teaching Note For this case, six organizations’ base pay data are used. The number of organizations providing data in a pay survey will vary, but should be manageable as well as representative (Milkovich & Newman, 2008). Although various forms of compensation were collected in the pay survey described in the case, students will be asked to analyze only the data for base pay (as illustrated in the table) for simplicity. The number of job incumbents is reported for each benchmark job so that the average, minimum and maximum pay rates can be analyzed more meaningfully. Employee names should not be reported or collected in pay surveys, but you also don’t want data that is so aggregated it is useless. No minimum or maximum base pay rates are reported in the table for jobs with only one incumbent, since only one salary is relevant. Also, the average salary for jobs with two incumbents will automatically be the average of the minimum and maximum. Students should find the same weighted averages for each benchmark job, because they have been given the same data and there are no outliers. If the instructor is concerned about students exchanging answers, multiple sets of salary survey data can be created. Weighted averages for this dataset are located in the case solution (rounded to the nearest cent). ©2008 SHRM Lisa Burke, Ph.D., SPHR 15INSTRUCTOR’S MANUAL Designing A Pay Structure At this point in the case study, we have used job analysis to create job descriptions for our benchmark jobs. We created an internal hierarchy of jobs (i.e., an internal job structure) by using the point method to evaluate each of our benchmark jobs. After our pay strategy was determined for the three job families (HR, office support, and operations), we analyzed the pay survey information gathered from competitors in our relevant labor market. We now have internal job point data and externally gathered weighted average base pay for each benchmark job. We can now proceed to merge the internal and external data to create our market pay line. The next task is to conduct a simple regression using Microsoft Excel to create a market pay line. Enter the job evaluation points (as X) and weighted average base pay rates (as Y) for each benchmark job and generate the regression results. » Task D: Conduct a simple regression in Excel to create a market pay line by entering the job evaluation points (on the X axis) and the respective weighted average market base pay (on the Y axis) for each benchmark job. Identify the slope and y-intercept and write the equation for the market pay line. Teaching Note Answers to Task D will vary depending on the assigned job evaluation points; a potential answer is presented in the solution set. This exercise assumes that students are familiar with using Excel to run a simple regression (although instructors may need to remind students how to interpret Excel’s regression results). Regression creates a “line of best fit” by merging the job evaluation points (X) and the external salary data (Y). The resulting regression line is used to predict the base pay (Y) for a specific number of job evaluation points (X). The equation for the simple regression line (as it is for any line) can be represented as: y=mx+b; in which: y =the predicted base pay m =the slope of the line x =the job evaluation points b =the y-intercept So, for example, if the regression results show that m = 400 and b is -20000, then the equation is y=400(x) – 20000 and the predicted pay rate for a job assigned 100 points would be y= 400(100)-20000, or 20,000. The regression output will also show information about how good the regression line fits the data. Specifically, look at the “R squared” in the regression output. Generally, the R squared, referred to as variance explained, should be .95 or higher. If R squared is significantly lower than this, there may be problems stemming from the job evaluation step. For example, the points assigned to certain benchmark jobs may be off – i.e., not make sense given the level of tasks, duties and responsibilities required for the job and the knowledge, skills and abilities needed by the job incumbent. If this is the case, re-examine the job descriptions and reconsider the points assigned to the benchmark jobs. Alternatively, there may be errors in the weighted average calculations. After conducting the regression again, examine the new R squared. 16 ©2008 SHRM Lisa Burke, Ph.D., SPHRINSTRUCTOR’S MANUAL Designing A Pay Structure » Task E: What is your R squared (variance explained)? Is it sufficient to proceed? Teaching Note The multiple R squared will indicate the amount of variance explained in Y (pay) by X (job evaluation points). The variance explained should be sufficiently high to ensure that the regression line fits the input data fairly well. A multiple R of .98 would produce an R squared of .96 .98.98 = .96., indicating that X (i.e., your job evaluation points) explains about 96% of the variance in Y (i.e., the market pay). The .95 guidepost for variance explained is a general guideline (see Milkovich, G., & Newman, J. (2008) Compensation, McGraw- Hill Irwin). It is useful to note that the multiple R in a simple regression will be equal to the correlation coefficient between X and Y. Thus, if students want to double-check their regression output, ask them to run a simple correlation between X and Y in Excel; their correlation coefficient should be the same number as the Multiple R. In Task E, students who haphazardly assigned points to the benchmark jobs (in the job evaluation step) may have a low R squared in their regression output. They may ask the instructor for help. Typically, a review of their point method will reveal problems and a lack of integrity (e.g., they may have the payroll assistant assigned more job points than any other benchmark job, or the operations assistant job may have significantly more points than the benefits manager job, which requires supervisory responsibilities). Discuss any questions you have about their assigned points and let them repeat that part to ensure learning. ©2008 SHRM Lisa Burke, Ph.D., SPHR 17INSTRUCTOR’S MANUAL Designing A Pay Structure Using the regression output (the slope and y-intercept), calculate the predicted market pay rate (using Excel) for each benchmark job. » Task F: Calculate the predicted base pay for each benchmark job. Teaching Note These calculations are pretty straightforward. Students should use their regression output, which will include a slope and y-intercept, and insert their job evaluation points for each benchmark job to determine the job’s predicted base pay. Students’ answers to this task will vary; the case solution set presents a possible response based on the inputs to earlier tasks. For example, with a slope of 250 and y-intercept of -1200, the 120 points assigned to the Front Desk Receptionist job would translate into a predicted base pay rate of 28,800. y=(250120) –1200 = 28800. Students must accurately calculate the base pay rate for the remaining benchmark jobs using their regression output. Excel automates this type of simple math with the use of formulas. Ask students to submit their Excel files electronically so you can see the equations used to generate their answers. 18 ©2008 SHRM Lisa Burke, Ph.D., SPHR