Guidelines and Ethical Considerations for Assessment Center Operations

International Journal of Selection and Assessment Volume 17 Number 3 September 2009 Guidelines and Ethical Considerations for Assessment Center Operations1 International Task Force on Assessment Center Guidelines2 This document is an update of several prior editions of guidelines and ethical considerations for assessment center operations dating back to 1975. Each set of guidelines was developed and endorsed by specialists in the research, development, and implementation of assessment centers. The guidelines are a statement of the considerations believed to be most important for all users of the assessment center method. For instance, the use of job-related simulations is a core concept when using the method. Job simulation exercises allow individuals to demonstrate their abilities in situations that are important on the job. As stressed in these guidelines, a procedure should not be represented as an assessment center unless it includes at least one, and usually several, job-related simulations that require the assessee to demonstrate a constructed behavioral response. Other important areas include assessor selection and training, using ‘competencies’ as dimensions to be assessed, validation, participants’ rights, and the incorporation of technology into assessment center programs. The current guidelines discuss a number of considerations in developing and using assessment centers in diverse cultural settings. 1. Purpose staff of an assessment center; and (4) guidance on the use of technology in assessments. T his document’s intended purpose is to establish professional guidelines and ethical considerations for users of the assessment center method. These guidelines are designed to cover both existing and future applications. The title ‘assessment center’ is restricted to those methods that follow these guidelines. These guidelines will provide: (1) guidance to industrial/organizational psychologists, organizational consultants, human resource management (HRM) specialists and generalists, and others who design and conduct assessment centers; (2) information to managers deciding whether or not to institute assessment center methods; (3) instruction to assessors serving on the 1 Endorsed by the 34th International Congress on Assessment Center Methods, 24 September 2008, Washington, DC, USA. 2 Task Force Members are listed in Appendix A. Inquiries may be sent to Taskforce Co-Chairs Deborah E. Rupp ([email protected]) and Doug Reynolds ([email protected]), or any Taskforce member. 2. History of guidelines The growth in the use of the assessment center method over the last several decades has resulted in a proliferation of applications in a variety of organizations. Assessment centers currently are being used in industrial, educational, military, government, law enforcement, and other organizational settings. From the beginning, practitioners began to raise concerns that reflected a need for standards or guidelines for users of the method. The 3rd International Congress on the Assessment Center Method, which met in Quebec (May 1975), endorsed the first set of guidelines. These were based on the observations and experience of a group of professionals representing many of the largest users of the method. & 2009 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main St., Malden, MA 02148, USA 244 Developments in the period 1975–1979 concerning federal guidelines related to testing, as well as professional experience with the original guidelines, suggested that the guidelines should be evaluated and revised. Therefore, the 1979 guidelines included essential items from the original guidelines, but also addressed the recognized need for: (1) further definitions, (2) clarification of impact on organizations and participants, (3) expanded guidelines on training, and (4) additional information on validation. Since 1979, the use of assessment centers has spread to many different organizations that are assessing individuals representing diverse types of jobs. During this period, pressures to modify the assessment center method came from three different sources. First, there had been attempts to streamline the procedures to make them less time consuming and expensive. Second, new theoretical arguments and evidence from empirical research had been interpreted to mean that the assessment center method does not work exactly as its proponents originally had believed, suggesting that the method should be modified. Third, many procedures purporting to be assessment centers had not complied with previous guidelines, because the guidelines may have been too ambiguous. Revisions in the 1989 third edition were designed to incorporate needed changes and to respond to some of the concerns raised since 1979. The 1989 revision of these guidelines was begun at the 15th International Congress on the Assessment Center Method in Boston (April 1987) when Dr Douglas Bray held discussions with many attendees. Subsequently, Dr Bray and Dr George Thornton solicited additional comments from a group of assessment center practitioners. The 1989 Task Force provided comments on drafts of a revision prepared by Bray and Thornton. A later draft was circulated and discussed at the 16th International Congress held in May 1988 in Tampa. The 1989 guidelines were written in response to comments obtained at the 1988 Congress and from members of the Task Force. The 1989 guidelines were endorsed by a majority of the Task Force and by participants at the 17th International Congress held in May 1989 in Pittsburgh. Changes in the 1989 guidelines from prior editions included: (1) specification of the role of job analysis; (2) clarification of the types of attributes/dimensions to be assessed and whether or not attributes/dimensions must be used; (3) delineation of the processes of observing, recording, evaluating, and aggregating information; and (4) further specification of assessor training. The 2000 revision of these guidelines was initiated at the 27th International Congress on Assessment Center Methods in Orlando (June 1999) when Dr David R. MacDonald conducted discussions with a number of assessment center experts in attendance and also International Journal of Selection and Assessment Volume 17 Number 3 September 2009 International Task Force on Assessment Center Guidelines solicited input at a general session regarding aspects of the guidelines that needed to be (re)addressed. A primary factor driving the revision was the passage of a full decade since the 1989 edition. Other factors included a desire to integrate technology into assessment center methods and recognition of the need for more specific definitions of several concepts and terms. Input from members of the Task Force for the 2000 edition was synthesized into a final draft that was presented and endorsed at the 28th International Congress held in May 2000 in San Francisco, attended by 150 participants representing Australia, Belgium, Brazil, Canada, Columbia, Germany, India, Indonesia, Italy, Japan, Mexico, the Netherlands, the Philippines, Singapore, Sweden, Switzerland, Taiwan, the United Arab Emirates, the United Kingdom, and the United States of America. The current revision of these guidelines was initiated at the 32nd International Congress on Assessment Center Methods, which was held in Las Vegas in October 2004. A roundtable discussion addressed contemporary assessment center issues on which there had been little previous guidance. Subsequently, this Congress decided that additions and revisions were needed in two areas: (1) Because of the proliferation of multinational organizations using assessment centers across geographic regions, more guidance was needed on global assessment center practices. The 32nd Congress established a separate task force to examine the issue. A report from this task force served as the foundation for a new section of the guidelines. (2) Given recent research on the effectiveness of various assessor training components, the Congress suggested an expansion of guidelines in this area as well. A second round of discussions on these issues was held in 2006 at the 33rd International Congress in London. These discussions suggested additional guidance in two areas: (1) the use of technology in assessment center practices, and (2) recognition of methodological differences among assessment centers used for different purposes. This revision, co-chaired by Deborah Rupp and Doug Reynolds, was unanimously endorsed by the 34th International Congress (2008, Washington, DC), which was attended by delegates representing Austria, Belgium, Canada, China, Germany, India, Indonesia, Mexico, the Netherlands, Romania, Russia, Singapore, South Africa, South Korea, Spain, Sweden, the United Arab Emirates, the United Kingdom, and the United States of America. 3. Assessment center defined An assessment center consists of a standardized evaluation of behavior based on multiple inputs. Several & 2009 Blackwell Publishing Ltd. Guidelines and Ethical Considerations for Assessment Center Operations trained observers and techniques are used. Judgments about behavior are made, in major part, from specifically developed assessment simulations. These judgments are pooled in a meeting among the assessors or by a statistical integration process. In an integration discussion, comprehensive accounts of behavior – and often ratings of it – are pooled. The discussion results in evaluations of the assessees’ performance on the dimensions or other variables that the assessment center is designed to measure. Statistical combination methods should be validated in accordance with professionally accepted standards. Technology may be used to facilitate the writing of reports, presentation of exercise materials, scoring of dimensions, classification of behavior, etc., as long as the essential elements described below are not compromised and validation standards are upheld. There is a difference between an assessment center and assessment center methodology. Various features of the assessment center methodology are used in procedures that do not meet all the guidelines set forth herein, such as when a psychologist or human resource professional, acting alone, uses a simulation as part of an individual’s evaluation. Such personnel assessment procedures are not covered by these guidelines; each should be judged on its own merits. Procedures that do not conform to all the guidelines herein should not be represented as assessment centers or imply that they are assessment centers by using the term ‘assessment center’ as part of the title. The following are the essential elements for a process to be considered an assessment center: 1. Job analysis/competency modeling – A job analysis of relevant behaviors must be conducted to determine the dimensions or competencies important to job success in order to identify what should be evaluated by the assessment center. Throughout this document the terms ‘dimensions’ and ‘competencies’ are used interchangeably. The type and extent of the job analysis depend on the purpose of the assessment, the complexity of the job, the adequacy and appropriateness of prior information about the job, and the similarity of the new job to jobs that have been studied previously. If past job analyses and research are used to select dimensions and exercises for a new job, evidence of the comparability or generalizability of the jobs must be provided. When the job does not currently exist, analyses can be done of actual or projected tasks or roles that will comprise the new job, position, job level, or job family. Target dimensions can also be identified from an analysis of the organization’s vision, values, strategies, or key objectives. Competency-modeling procedures may be used to determine the dimensions to be assessed by the & 2009 Blackwell Publishing Ltd. 245 assessment center, if such procedures are conducted with the same rigor as traditional job analysis methods. Rigor in this regard is defined as the involvement of subject matter experts who are knowledgeable about job requirements, the collection and quantitative evaluation of essential job elements, and the production of evidence of reliable results. Any job analysis or competency-modeling process must result in clearly specified categories of behavior that can be observed in assessment procedures. A ‘competency’ may or may not be amenable to behavioral assessment as defined herein. A competency, as used in various contemporary sources, refers to an organizational strength, an organizational goal, a valued objective, a construct, or a grouping of related behaviors or attributes. A competency may be considered a behavioral dimension for the purposes of assessment in an assessment center provided it can be defined precisely and expressed in terms of behaviors observable on the job or in a job family and in simulation exercises. A competency also must be shown to be related to success in the target job or position or job family. 2. Behavioral classification – Behaviors displayed by participants must be classified into meaningful and relevant categories such as behavioral dimensions, attributes, characteristics, aptitudes, qualities, skills, abilities, competencies, or knowledge. In these guidelines, the term ‘dimension’ is used as a general descriptor for each type of behavior category. Note that other classification schemes also may be used. For example, categories may reflect components of the target jobs or the assessment itself. 3. Assessment techniques – The techniques used in the assessment center must be designed to provide information for evaluating the dimensions previously determined by the job analysis. Assessment center developers should establish a link from behaviors to dimensions to assessment techniques. This linkage should be documented in a dimension  assessment technique matrix. 4. Multiple assessments – Multiple assessment techniques must be used. These can include tests, interviews, questionnaires, and simulations. The assessment techniques are developed or chosen to elicit a variety of behaviors and information relevant to the selected dimensions. Self-assessment and 3601 assessment data may be gathered as assessment information. The assessment techniques will be pretested to ensure that they provide reliable, objective, and relevant behavioral information for the organization in question. Pretesting might entail trial administration with participants similar to assessment center candidates, thorough review by subject matter experts as to the accuracy and representativeness of behavioral sampling, and/or evidence from the use of these techniques for similar jobs in similar organizations. International Journal of Selection and Assessment Volume 17 Number 3 September 2009 246 5. Simulations – The assessment techniques must include a sufficient number of job-related simulations to allow opportunities to observe the candidate’s behavior related to each dimension/competency being assessed. At least one – and usually several – job-related simulations must be included in each assessment center. A simulation is an exercise or technique designed to elicit behaviors related to dimensions of performance on the job requiring the participants to respond behaviorally to situational stimuli. Examples of simulations include, but are not limited to, group exercises, inbasket exercises, interaction (interview) simulations, presentations, and fact-finding exercises. Stimuli also may be presented through video, audio, or virtual simulations delivered via computer, telephone, video, the Internet, or an intranet. For simple jobs, one or two job-related simulations may be used if the job analysis clearly indicates that one or two simulations alone sufficiently simulate a substantial portion of the job being evaluated. If a single comprehensive assessment technique is used, then it must include distinct job-related segments. Assessment center designers should also be careful to design exercises that reliably elicit a large number of dimension-related behaviors. This should provide assessors with sufficient opportunities to observe dimension-related behavior. The stimuli contained in a simulation should parallel or resemble stimuli in the work situation, although they may be in different settings. The desirable degree of fidelity is a function of the assessment center’s purpose. Fidelity may be relatively low for early identification and selection programs for non-managerial personnel and may be relatively high for programs designed to diagnose the training needs of experienced managers. Assessment center designers should be careful that the exercise content does not favor certain assessees (e.g., those in certain racial, ethnic, age, or sex groups) for irrelevant reasons. To qualify as a behavioral simulation for an assessment center as defined herein, the assessment method must require the assessee to overtly display certain behaviors. The assessee must be required to demonstrate a constructed response. Assessment procedures that require the assessee to select only among provided alternative responses, such as seen in multiple-choice tests or multiple-choice computerized in-baskets, do not conform to this requirement. Likewise, a situational interview that calls for only an expression of behavioral intentions does not conform. Neither do ‘low-fidelity’ simulations and situational interviews. Although they may yield highly reliable and valid assessment ratings, they do not constitute the behavioral assessment required in assessment centers. Assessment center materials (e.g., simulations and other exercises, rating scales, assessor training materi- International Journal of Selection and Assessment Volume 17 Number 3 September 2009 International Task Force on Assessment Center Guidelines als) are often intellectual property protected by international copyright laws. Respect for copyrights and the intellectual property of others must be maintained under all circumstances. 6. Assessors – Multiple assessors must be used to observe and evaluate each assessee. When selecting assessors, where appropriate, the program should strive to have diverse assessors, both in terms of demographics (e.g., race, ethnicity, age, sex) and experience (e.g., organizational level, functional work area, managers, psychologists, etc.). The maximum ratio of assessees to assessors is a function of several variables, including the type of exercises used, the dimensions to be evaluated, the roles of the assessors, the type of integration carried out, the amount of assessor training, the experience of the assessors, and the purpose of the assessment center. A participant’s current supervisor should not be involved in the assessment of a direct subordinate when the resulting data will be used for selection or promotional purposes. 7. Assessor training – Assessors must receive thorough training and demonstrate performance that meets the guidelines in the ‘Assessor Training’ section of this document before participating in an assessment center. 8. Recording behavior and scoring – A systematic procedure must be used by assessors to record specific behavioral observations accurately at the time of observation. This procedure might include techniques such as handwritten notes, behavioral observation scales, or behavioral checklists. Audio and video recordings of behavior may be made and analyzed at a later date. When technology-based tools are used during the scoring process, these approaches should be evaluated for reliability and accuracy. Assessors must prepare a record of the observations made during each exercise before the integration discussion or statistical integration. 9. Data integration – The integration of each individual’s behaviors (individual dimension scores aggregated across exercises; exercise-specific scores; or potentially, depending on the purpose of the assessment center, across-exercise scores aggregated into an overall assessment rating) must be based on pooled information from assessors or through a statistical integration process. The process used should be carried out in accordance with professionally accepted standards. If an integration discussion is used, assessors should consider information derived from the assessment techniques for each dimension, but should not consider information irrelevant to the dimensions or the purpose of the assessment process. & 2009 Blackwell Publishing Ltd. Guidelines and Ethical Considerations for Assessment Center Operations Methods of combining assessors’ evaluations of information discussed in their integration sessions must be based on reliable individual assessors’ ratings. Computer technology may also be used to support the data integration process, provided the conditions of this section are met. 4. Non-assessment center activities The following kinds of activities do not constitute an assessment center: 1. Assessment procedures that do not require the assessee to demonstrate overt behavioral responses are not behavioral simulations; thus, any assessment program that consists solely of such procedures is not an assessment center as defined herein. Examples of these are computerized in-baskets calling only for multiplechoice responses, situation interviews calling only for behavioral intentions, and written competency tests. Procedures not requiring an assessee to demonstrate overt behavioral responses may be used within an assessment center, but must be coupled with at least one simulation requiring the overt display of behaviors. 2. Panel interviews or a series of sequential interviews as the sole technique. 3. Reliance on a single technique (regardless of whether it is a simulation) as the sole basis for evaluation. However, a single comprehensive assessment technique that includes distinct job-related segments (e.g., large, complex simulations or virtual assessment centers with several definable components and with multiple opportunities for observation in different situations) is not precluded by this restriction. 4. Using only a test battery composed of a number of paper-and-pencil measures, regardless of whether the judgments are made by a statistical or judgmental pooling of scores. 5. Single-assessor evaluation (i.e., measurement by one individual using a variety of techniques, such as paper-and-pencil tests, interviews, personality measures, or simulations). 6. The use of several simulations with more than one assessor but with no pooling of data (i.e., each assessor prepares a report on performance in an exercise, and then the individual, un-integrated reports are used as the final product of the center). 7. A physical location labeled as an ‘assessment center’ that does not conform to the methodological requirements noted above. 5. Assessment center policy statement Assessment centers operate more effectively as part of an integrated human resource system. Before the & 2009 Blackwell Publishing Ltd. 247 introduction of an assessment center into an organization, a policy statement should be prepared and approved by the organization. This policy statement should address the following areas: 1. Objective – An assessment center may be used for a variety of purposes. Falling into the broad categories of selection vs diagnosis vs development, such purposes might include prescreening, hiring, early identification and evaluation of potential, performance appraisal, succession planning, and professional development. An assessment center participant should be told, before the assessment, what decision(s) will or might be made with assessment center data. If the organization desires to make decisions with the data other than those previously communicated to the participant, the decision(s) should be clearly described to the participant and consent obtained. In addition, the assessment center developer must remain cognizant of the assessment center’s purpose when designing, developing, implementing, and validating the program. Decisions about the choice of dimensions, content of simulations, selection and training of assessors, scoring, feedback, and evaluation all should be made with the objective in mind. 2. Assessees –The population to be assessed, the method for selecting assessees from this population, the procedure for notification, and the assessment process should be specified. 3. Assessors – The assessor population (including sex, age, race, and ethnic mix), limitations on the use of assessors, assessor experience, and evaluation of assessor performance and certification requirements, where applicable, should be specified. 4. Use of data –The process flow of assessment records within the organization, individuals who receive reports, restrictions on access to information, procedures and controls for research and program evaluation purposes, feedback procedures to management and employees, and the length of time data will be maintained in files should all be specified. Particularly for a selection application, it is recommended that the data be used within 2 years of the date of administration because of the likelihood of change in the participant or the organization. 5. Qualifications of consultant(s) or assessment center developer(s) – The internal or external consultant(s) responsible for the development of the center or of the exercises/simulations for the center should be identified and his or her professional qualifications and related training specified. 6. Validation – The statement should specify the validation model being used. If a content-oriented validation strategy is used, the documentation of the relationship of the job/job family content to the dimensions and exercises should be presented along with evidence of the reliability of the observations and rating International Journal of Selection and Assessment Volume 17 Number 3 September 2009 248 of behavior. If evidence is being taken from prior validation research, which may have been summarized in meta-analyses, the organization must document that the current job/job family and assessment center are comparable and generalized to the jobs and assessment centers studied elsewhere. If local criterion-related validation has been carried out, full documentation of the study should be provided. If validation studies are under way, there should be a schedule indicating when a validation report will be available. Finally, the validation technique employed should be appropriate given the purpose/objective of the assessment center program (e.g., selection vs development). 7. Legal context – Laws existing both in an organization’s/agency’s home state, province, or nation – as well as in the state, province, or nation where the assessment center program is being carried out – may have implications for program design, validation, implementation, and documentation. Most nations have disadvantaged and protected groups (such as native/ aboriginal people, recent immigrants, racial groups, religious groups, and groups defined by age, political opinion, sexual orientation, etc.) with accompanying regulations for legal protections for their members. The policy statement should articulate the particular laws and policies that are relevant for the assessment center program and state how legal compliance will be ensured. 6. Assessor training Assessor training is an integral part of the assessment center program. Assessor training should have clearly stated training objectives, performance guidelines, and quality standards. The following issues related to training must be considered: 1. Training content – Whatever the approach to assessor training, the objective is to obtain reliable and accurate assessor judgments. A variety of training approaches may be used (e.g., lectures, discussion, observation of practice candidates, video demonstrations, observation of other assessors) as long as it can be shown that reliable, accurate assessor judgments are obtained. At a general level, all assessor training programs should include training on:  The dimensions to be assessed, including their behavioral definitions.  The observation, recording, classification, and evaluation of behaviors relevant to the dimensions to be assessed.  The content of the exercises as well as which dimensions are targeted in which exercises. International Journal of Selection and Assessment Volume 17 Number 3 September 2009 International Task Force on Assessment Center Guidelines  The avoidance of common observational and rating errors (including how to distinguish behaviors from inferences). Depending on the purpose of the assessment center, the training might include additional components such as professionalism, knowledge of the organization, knowledge of the target job, the ability to give accurate oral or written feedback, and consistency in role playing. The following minimum training goals are required: a. Knowledge of the organization and job/job family or normative group being assessed to provide an effective context for assessor judgments where appropriate. b. Thorough knowledge and understanding of the assessment dimensions, their definitions, their relationship to job performance, and examples of effective and ineffective performance. c. Thorough knowledge and understanding of the assessment techniques, exercise content, relevant dimensions to be observed in each portion of the assessment center, expected or typical behavior, examples or samples of actual behaviors, etc. d. Demonstrated ability to observe, record, and classify behavior in dimensions, including knowledge of the protocol for documenting behavior. e. Thorough knowledge and understanding of evaluation and rating procedures, including how data are integrated. f. Thorough knowledge and understanding of assessment policies and practices of the organization, including restrictions on how assessment data are to be used, when this is a requirement of assessors. g. Thorough knowledge and understanding of feedback procedures and strategies, where appropriate, to maximize assessees’ acceptance of feedback and behavior change. h. Demonstrated ability to give accurate oral and written feedback, when feedback is given by the assessors. i. Demonstrated knowledge and ability to play objectively and consistently the role called for in interactive exercises (e.g., one-on-one simulations or factfinding exercises) when role playing is required of assessors. Non-assessor role players also may be used if their training results in their ability to play the role objectively and consistently. 2. Training length – The length of assessor training may vary due to a variety of considerations that can be categorized into three major areas: a. Trainer and instructional design considerations:  The instructional mode(s) utilized.  The qualifications and expertise of the trainer.  The training and instructional sequence. & 2009 Blackwell Publishing Ltd. Guidelines and Ethical Considerations for Assessment Center Operations b. Assessor considerations:  Previous knowledge and experience with similar assessment techniques.  Type of assessors used (e.g., professional psychologists vs managers).  Experience and familiarity with the organization and the target position(s)/job(s)/job families/target level.  The frequency of assessor participation.  Other related qualifications and expertise (e.g., testing and assessment, executive coaching). c. Assessment program considerations:  The target position’s level of difficulty.  The number of dimensions/competencies to be rated.  The anticipated use of the assessment information (e.g., immediate selection, broad placement considerations, diagnosis, development).  The number and complexity of the exercises.  The division of roles and responsibilities between assessors and others on the assessment staff (e.g., administrator and other support staff).  The degree of support provided to assessors in the form of observation and evaluation guides. It should be noted that length and quality of training are not synonymous. Precise guidelines for the minimum number of hours or days required for assessor training are difficult to specify. One day of training may be sufficient for a well-structured assessment center using a small number of exercises, a qualified trainer, and carefully selected assessors. However, extensive experience has shown that, for the initial training of assessors who have no experience in an assessment center that conforms to the guidelines in this document, more training may be needed (e.g., 2 days of assessor training for each day of assessment center exercises). Assessors who have experience with similar assessment techniques in other programs may require less training. More complex assessment centers with varied formats of simulation exercises may require additional training; simple assessment centers may require less. In any event, assessor training is an essential aspect of an assessment program. The true test of training quality should be assessor competence as described below. 3. Performance guidelines and certification – Each assessment center should have clearly stated performance guidelines for assessors contingent on the purpose of the assessment center and the various assessor roles. These performance guidelines should include, at a minimum, the ability to: a. Rate behavior in a standardized fashion. b. Recognize, observe, and report behavior into the appropriate dimensions, etc. & 2009 Blackwell Publishing Ltd. 249 c. Administer an exercise if the assessor serves as an exercise administrator. d. Play objectively and consistently the role called for in those interactive exercises where role playing is required of assessors. e. If applicable, deliver positive and negative behavioral feedback, if applicable, with supporting evidence in a manner that conveys concern/empathy and maintains or enhances the assessee’s self-esteem. f. Motivate assessees and engage in coaching, action planning, and goal setting. Some measurement is needed to indicate that the individual being trained is capable of functioning as an assessor. This measurement may vary and could include data in terms of (1) accuracy and reliability of rating performance (defined with regard to either an ‘expert’ standard or convergence with other assessors), (2) critiques of assessor reports, and (3) observation or shadowing of assessors in training by the assessment center staff. It is important that, before carrying out their actual duties, assessors’ performance is evaluated to ensure that they are sufficiently trained to function as assessors and that such performance is periodically monitored to ensure that the skills learned in training are applied. Each organization must be able to demonstrate that its assessors can meet minimum performance standards. This may require the development of additional training or other prescribed actions for assessors not meeting these performance guidelines. The trainer of assessors should be competent to enable individuals to develop the assessor skills stated above and to evaluate the acquisition of these skills. 4. Currency of training and experience – The time between assessor training and initial service as an assessor should be short (e.g., not to exceed 6 months). If a longer period has elapsed, prospective assessors should attend a refresher course or receive special coaching from a trained assessment center administrator. Assessors who do not have recent experience as an assessor (i.e., fewer than two assessment centers over 2 consecutive years) should attend a refresher course before they serve again or receive special coaching from a trained assessment center administrator. 7. Informed participation The organization is obligated to make an announcement before the assessment so that participants will be fully informed about the program. Ideally, this information should be made available in writing before the center. A second option is to present the information at the start International Journal of Selection and Assessment Volume 17 Number 3 September 2009 250 of the program. While the information provided will vary across organizations, the following basic information should be given to all participants: 1. Objective(s) – The objective(s) of the program and the purpose of the assessment center. The organization may choose to disclose the dimensions measured and the general nature of the exercises before the assessment. 2. Selection – How individuals are selected to participate in the center. 3. Choice(s) – Any options the individual has regarding the choice of participating in the assessment center as a condition of employment, advancement, development, etc. 4. Staff – General information on the assessor staff and the role of the observer, including composition, relevant experience, and assessor training. 5. Materials – What assessment center materials completed by the individual are collected and maintained by the organization. 6. Results – How the assessment center results will be used, what recommendations will be made, and how long the assessment results will be maintained on file. 7. Feedback – When, how (e.g., written, face-to-face, technology-aided), and what kind of feedback (e.g., by dimension, by exercise, by a combination) will be given to the participants. 8. Development – Mechanisms for follow-up support and monitoring, if any (e.g., external coaching, training, mentoring; top management/supervisory support). 9. Alignment – How the assessment center results will be aligned with organizational strategy and culture, and how the results will be integrated with other HRM systems. 10. Reassessment – The procedure for reassessment (if any). 11. Access – Who will have access to the assessment center reports (and audio and/or video files, if applicable) and under what conditions. 12. Contact – Who will be the contact person responsible for the records and where the results will be stored or archived. 8. Validation issues A major factor in the widespread acceptance and use of assessment centers is related directly to an emphasis on sound validation research. Numerous studies demonstrating the predictive validity of individual assessment center programs have been conducted in a variety of organizational settings and reported in the professional literature. However, the historical record of this process’s validity cannot be taken as a guarantee that a given assessment program (or new application of an existing program) will or will not be valid. International Journal of Selection and Assessment Volume 17 Number 3 September 2009 International Task Force on Assessment Center Guidelines Ascertaining the validity of an assessment center program is a complicated technical process, and it is important that validation research meets both professional and legal standards. Research should be conducted by individuals knowledgeable in the technical and legal issues pertinent to validation procedures. In evaluating the validity of assessment center programs, it is particularly important to document the selection of the dimensions assessed in the center. The relationship of assessment exercises to the dimensions assessed should be documented as well. Validity generalization studies of assessment center research suggest that overall assessment ratings derived in a manner conforming to these guidelines show considerable predictive validity. Such findings support the use of a new assessment center in a different setting if the job, exercises, assessors, and assessees in the new situation are similar to those studied in the validation research and if similar procedures are used to observe, report, and integrate the information. The validity generalization studies of the predictive validity of the overall assessment rating do not necessarily establish the validity of the procedure for other purposes such as diagnosis of training needs, accurate assessment of skill level in separate dimensions, or the developmental influence of participation in an assessment center. The standards and principles for validation appear in Principles for the Validation and Use of Personnel Selection Procedures (Society for Industrial and Organizational Psychology Inc., 2003) and Standards for Educational and Psychological Testing (American Educational Research Association, American Psychological Association, & National Council on Measurements in Education, 1999). In principle, technology can be used for writing reports, presenting exercises, scoring dimensions, classifying behavior, or creating behavioral checklists, as long as the assessment process remains valid. 9. Assessment centers for different purposes Assessment centers are generally used for three major purposes: (1) to predict future behavior for decision making, (2) to diagnose development needs, and (3) to develop candidates on dimensions of interest. However, additional purposes for the assessment center method currently exist and will continue to evolve with further use. The design and operation of an assessment center may vary, depending on its intended purpose. For example, with assessments designed purely to support a personnel decision (e.g., promotion), the emphasis may be on a reliable and valid overall assessment rating. There may be little reason to generate individual dimension scores in this context. Alternatively, diagnostic assessment centers may require the generation of reliable and valid dimension scores only. & 2009 Blackwell Publishing Ltd. Guidelines and Ethical Considerations for Assessment Center Operations Developmental assessment centers (DACs) seek to both assess and develop dimensions that may be learned (i.e., skills and competencies that can be improved upon in a reasonable amount of time). DACs involve multiple points of feedback and repeated practice, and may repeat exercises of the same type(s) as a way to track improvement on the dimensions over time. As a result, such programs may be longer than assessment centers for prediction and diagnosis. Feedback is an essential component of a DAC program, and in order to foster learning, this feedback needs to be immediate. Often, the role of assessors in DACs is not only to observe and evaluate behavior, but also to facilitate learning and development by delivering feedback and facilitating development planning during the assessment process. If the focus is purely on learning, DACs may be less standardized than assessment centers used for personnel decision making, and they may be customized to meet participants’ developmental needs. When validating or otherwise evaluating DACs, the appropriate criterion is change in participants’ understanding, behavior, and proficiency on targeted dimensions. Because of the differences outlined above, assessment centers must be designed and implemented appropriately for their intended purpose. 10. Rights of the participant Assessment center activities typically generate a large volume of data on an individual who has gone through a center. These assessment data come in many forms and may include observer notes, reports on performance in the exercises, assessor ratings, peer ratings, paper-andpencil or computerized tests, video files, and final assessment center reports. This list, while not exhaustive, does indicate the extent of information about an individual that may be collected. The following guidelines for use of these data are suggested: 1. Assessees should receive feedback on their assessment center performance and should be informed of any recommendations made. 2. Assessees who are members of the organization have a right to read any formal summary written reports concerning their own performance and recommendations that are prepared and made available to management. Applicants to an organization should be provided with, at a minimum, what the final recommendation is and, if possible and if requested by the applicant, the reason for the recommendation. 3. To ensure test security, assessment center exercises and assessor reports on performance in particular exercises are exempted from disclosure, but the rationale and validity data concerning ratings of dimen- & 2009 Blackwell Publishing Ltd. 251 sions and the resulting recommendations should be made available on request of the individual assessee. 4. The organization should inform the assessee what records and data are being collected, maintained, used, and disseminated. Assessees should be informed if their activities in the assessment center are being recorded as well as if such recordings or other personal data will be transferred across national borders or over the Internet. The organization must take precautions to ensure the security of data transferred over the Internet. Organizations that collect personal data during the assessment process should comply with applicable data protection regulations, such as the European Union Directive on Data Protection and the US Safe Harbor Privacy Principles. 5. If the organization decides to use assessment results for purposes other than those originally announced and that can have an impact on the assessee, then the assessee must be informed and consent obtained. 11. Conducting assessment centers across cultural contexts It is common for single assessment center programs to cross both cultural and national boundaries. In some situations it may be necessary to adapt many assessment center practices to the local culture in which an assessment center is deployed. In other situations it may be decided that an assessment center program should be standardized across all regions to which it is being deployed. Practitioners using assessment center methods beyond the boundaries of the country/region from which the assessment center program originated should determine the extent to which cultural accommodations may be necessary. The analysis should include evidence that the validity and applicability of the assessment center have not been compromised at either the design or implementation phase. A range of contextual factors should be considered during such a process, including:  Extent of commonality in the cultural, business, legal, and socio-political environments between countries (e.g., cultural beliefs and behaviors, local business laws).  Differences in national guidelines set by local professional associations.  Commonality of skills/dimensions critical for job success.  Commonality in performance standards/behaviors required for job success.  Extent of commonality of the business models between the organizations across which the assessment center/method is being adapted (i.e., overall business strategy, vision, values, and practices). International Journal of Selection and Assessment Volume 17 Number 3 September 2009 252 International Task Force on Assessment Center Guidelines  Degree of centralized vs decentralized (i.e., local) control across branches of the organization.  Whether comparison statistics (e.g., normative comparisons) are required to interpret the results across locations.  Extent to which personnel need to be transferred across countries/cultures. When assessment center programs are designed to be culturally specific, the following aspects should be considered for modification based on each culture in which the assessment center is used:      Selection of performance criteria. Criteria for occupational success. Selection of exercises. Assessor training. Feedback process. In contrast, several aspects of the assessment center process should remain standardized, even when the process has been culturally adapted. Features that should remain the same across cultures include:  Inclusion of behavioral observation.  Training of assessors in the process of behavioral observation.  Classification and rating of behavior.  A systematic process of integrating evaluations across exercises, dimensions, and assessors. 12. National assessment center guidelines tion. Available at http://www.apa.org/science/standards.html (accessed 1 August 2008). American Psychological Association Council of Representatives. (1990) APA Guidelines for Providers of Psychological Services to Ethnic, Linguistic, and Culturally Diverse Populations. Boston: American Psychological Association. Available at http://www.apa.org/pi/guide.html (accessed 1 August 2008). American Psychological Association Public Interest Directorate and Council of Representatives. (2002) Guidelines on Multicultural Education, Training, Research, Practice, and Organizational Change for Psychologists. Washington, DC: American Psychological Association. Available at http://www.apa.org/pi/multiculturalguidelines.pdf (accessed 1 August 2008). International Test Commission. (2000) International Test Commission Test Adaptation Guidelines. International Test Commission. Available at http://www.intestcom.org/test_ adaptation.htm (accessed 1 August 2008). Society for Industrial and Organizational Psychology & American Psychological Association. (2003) Principles for the Validation and Use of Personnel Selection Procedures (4th edn). Bowling Green, OH: Society for Industrial and Organizational Psychology. Available at http://www.siop. org/_Principles/principles.pdf (accessed 1 August 2008). US Department of Commerce. (2000). Safe harbor privacy principles. Available at http://www.export.gov/safeharbor (accessed 1 August 2008). Appendix A Table A1. Task force members 1975 Edition Albert Alon Douglas W. Bray, PhD William C. Byham, PhD In addition to the international guidelines presented herein, some countries have developed local guidelines to govern (in parallel with the international guidelines) assessment center practices in their specific national contexts. Examples of national standards include: Lois A. Crooks Donald L. Grant, PhD  South Africa: Assessment Centre Study Group. (2007). Guidelines for assessment and development centres in South Africa (4th ed.).  Germany: Standards der assessment-center-technik. (2004). Überblick und Hintergrundinformationen Hamburg, Germany: Arbeitskreis Assessment Center e.V.  Indonesia: Indonesian Task Force on Assessment Center Guidelines. (2002). Daya Dimensi Indonesia. Ethical guidelines for assessment center operations. Cabot L. Jaffee, PhD Alan I. Kraut, PhD References American Educational Research Association, American Psychological Association, & National Council on Measurements in Education. (1999) Standards for Educational and Psychological Testing. Washington, DC: American Psychological Associa- International Journal of Selection and Assessment Volume 17 Number 3 September 2009 Lowell W. Hellervik, PhD James R. Huck, PhD John H. McConnell Leonard W. Slivinski, PhD Thomas E. Standing, PhD Edwin Yager 1979 Edition Albert Alon Dale Baker Douglas W. Bray, PhD William C. Byham, PhD Steven L. Cohen, PhD Lois A. Crooks Donald L. Grant, PhD Milton D. Hakel, PhD Miracle Food Mart (Canada) AT&T Development Dimensions International Inc. Educational Testing Service AT&T and University of Georgia University of Minnesota AT&T – Michigan Bell Telephone Company Assessment Designs Inc. International Business Machines American Management Association Public Service Commission (Canada) The Standard Oil Company – Ohio Consulting Associates Miracle Food Mart (Canada) US Civil Service Commission AT&T Development Dimensions International Inc. Assessment Designs Inc. Educational Testing Service University of Georgia Ohio State University & 2009 Blackwell Publishing Ltd. Guidelines and Ethical Considerations for Assessment Center Operations Table A1. (Contd.) Lowell W. Hellervik, PhD James R. Huck, PhD Cabot L. Jaffee, PhD Frank M. McIntyre, PhD Joseph L. Moses, PhD (Chair) Nicky B. Schnarr Leonard W. Slivinski, PhD Thomas E. Standing, PhD Edwin Yager 1989 Edition Virginia R. Boehm, PhD Douglas W. Bray, PhD (Co-Chair) William C. Byham, PhD Anne Marie Carlisi, PhD John J. Clancy Reginald Ellis Joep Esser Fred Frank, PhD Ann C. Gowdey Dennis A. Joiner Rhonda Miller Marilyn Quaintance-Gowing, PhD Robert F. Silzer, PhD George C. Thornton III, PhD (Co-Chair) 2000 Edition William C. Byham, PhD Richard Flanary Marilyn K. Gowing, PhD James R. Huck, PhD Jeffrey D. Kudisch, PhD David R. MacDonald, PhD (Chair) Patrick T. Maher, DCrim Jeroen J. J. L. Seegers George C. Thornton III, PhD 2008 Edition William C. Byham, PhD Anuradha Chawla, PhD Alyssa Mitchell Gibbons, PhD Sebastien Houde, MSc Dennis Joiner, MS & 2009 Blackwell Publishing Ltd. 253 Table A1. (Contd.) University of Minnesota Human Resources International Assessment Designs Inc. Consulting Associates AT&T International Business Machines Public Service Commission (Canada) Standard Oil of Ohio Consulting Associates Assessment & Development Associates Development Dimensions International Inc. Development Dimensions International Inc. BellSouth Clancy & Associates Canadian National Railway Mars B.V. (the Netherlands) Electronic Selection Systems Corporation Connecticut Mutual Joiner & Associates New York Power Authority US Office of Personnel Management Personnel Decisions Inc. Colorado State University Development Dimensions International Inc. National Association of Secondary School Principals US Office of Personnel Management Human Resources International University of Southern Mississippi Steelcase Inc. Personnel & Organization Development Consultants Inc. Assessment & Development Consult (the Netherlands) Colorado State University Development Dimensions International Inc. RHR International Colorado State University University of Guelph & Royal Military College of Canada Dennis A. Joiner & Associates Myungjoon Kim, PhD Diana Krause, PhD Jeffrey D. Kudisch, PhD Cara Lundquist, MA David R. MacDonald, PhD Patrick T. Maher, DCrim Doug Reynolds, PhD (CoChair) Deborah E. Rupp, PhD (CoChair) Deidra J. Schleicher, PhD Jeroen J. J. L. Seegers, PhD George C. Thornton III, PhD Korean Psychological Testing Institute DHV Speyer University of Maryland Southern California Edison Steelcase Inc. Personnel & Organization Development Consultants Inc. Development Dimensions International Inc. University of Illinois at Urbana-Champaign Purdue University Right Management Benelux (the Netherlands) Colorado State University Table A2. Glossary Assessee: An individual whose competencies are measured by an assessment center. Assessment center: A process employing multiple techniques and multiple assessors to produce judgments regarding the extent to which a participant displays selected behavioral dimensions. Assessor: An individual trained to observe, record, classify, and make reliable judgments about the behaviors of assessees. Competency: See Dimension. Developmental assessment center: An assessment center designed for the purpose of directly developing/improving assessees on the dimensions of interest. Dimension: Also called competency. A constellation or group of behaviors that are specific, observable, and verifiable and that can be reliably and logically classified together and that relate to job success. Feedback: Information comparing actual performance to a standard or desired level of performance. High (or low) fidelity: The extent to which an assessment center simulation requires the assessee to actually display jobrelevant behaviors related to one or more select dimensions. Fidelity is related to the realism of the simulation as compared with an actual job situation, task, etc. Job analysis: The process used to determine the behavioral dimensions linked to success or failure in a job, job role, or job grouping. The process typically consists of a combination of techniques to collect job information, such as interviews with and observations of incumbents, job checklists, interviews with upper-level managers/executives, and review of existing job documentation (job descriptions, training manuals, etc.). Reliability: The extent to which a measurement process yields the same results (given identical conditions) across repeated measurements. Simulation: An exercise or technique designed to elicit behaviors related to dimensions of performance on the job requiring the participants to respond behaviorally to situational stimuli. Validity: The extent to which a measurement tool or process, such as an assessment center, yields useful results. Multiple validities might be measured (e.g., ‘construct,’ ‘content,’ ‘face,’ ‘predictive,’ ‘social’) depending on the questions being explored and the tool or process being investigated. International Journal of Selection and Assessment Volume 17 Number 3 September 2009