|
|
||||||||
Critical Issues in Dental Education |
Key words: National Board Dental Examination, dental education, dental curricula, dental licensure, competency, assessment
Submitted for publication 12/06/06; accepted 06/18/07
| Abstract |
|---|
|
|
|---|
Are these reasonable statements—or merely provocative assertions to enliven cocktail receptions and hallway conversation? Recent interest in curriculum reform has stimulated discussion on these and other issues related to the purpose and use of the National Board Dental Examination (NBDE). The diversity of thought and commentary on the National Board exams reveals a mixture of interest, misinformation, misunderstanding, misperception, and skepticism tempered by respect for this rite of passage from dental school to professional practice. Whether for student or dean, the results can be a welcomed affirmation of success or a nagging source of concern.
The purpose of this article is to establish a common understanding of the purpose of the NBDE along with the principles, policies, and procedures that guide development and administration of the examinations. In this context, we will explore the uses and misuses of the National Board exams and address timely questions relevant to curriculum reform. Should the dental curriculum define the content of the National Boards, or should the National Boards define the curriculum? If reform is needed, which comes first—a new curriculum or a new exam?
| Purpose and History of the National Board Dental Examination |
|---|
|
|
|---|
Part I of the NBDE was first administered in 1933, followed by Part II in 1934. State recognition of the National Board certificate was slow at first, with participation remaining low until the mid-1950s. By 1990, all U.S. licensing jurisdictions accepted the NBDE as fulfillment of the written examination requirement for licensure.
Both the structure and name of the body overseeing the examination have changed several times over the years.4 The dental and dental hygiene examination programs are currently administered under the direction of the Joint Commission on National Dental Examinations (JCNDE), which succeeded the Council on National Board Examinations in the early 1980s. Designation of the Joint Commission as the governing body for the National Board Dental Examination by agreement between the ADA and the American Association of Dental Examiners (AADE) resolved differing perspectives on a number of issues relating to the characteristics and control of the examination.5,6
Provisions of the 1980 agreement ultimately resulted in a governance structure and principles for operation of the exam program that remain in place today. These principles include the following: 1) because the National Board examination is intended for use in licensure and because licensure is solely for the protection of the public, operation of National Board programs constitutes a public trust; 2) the Joint Commission functions as an agency of the ADA for administrative purposes only; and 3) restrictions placed on the Joint Commission should be limited to those that serve to enhance public trust or are necessary to protect the ADA in its administrative role.5
Membership of the Joint Commission is made up of representatives from the ADA (three), AADE (six), American Dental Education Association (ADEA) (three), American Dental Hygienists Association (one), American Student Dental Association (one), and the public (one). The ADA provides staff support to the Joint Commission through its Department of Testing Services, as well as resources and administrative services that support administration of the National Board examination. In accordance with the 1980 agreement, the Joint Commission annually conducts a meeting for representatives of the state boards, the National Dental Examiners Advisory Forum, to provide an opportunity for dialogue and sharing of information about the examination. The Joint Commission remains accountable to its member organizations with respect to its budget and overall performance and must submit a report annually to the participating organizations.
The National Board examination, in its two parts, was originally characterized as "written examinations." While the designation and intent to provide an examination on the cognitive abilities or theoretical basis for dental practice remain, the format and mode of administration are no longer limited to the paper-and-pencil format. The Joint Commission currently uses the following statement to inform candidates and others of the purpose and general content of the exams: the National Board Dental Examination assesses the ability to understand important information from basic biomedical and dental clinical sciences and the ability to apply such information in a problem-solving context.7
Three documents guide the operations of the Joint Commission and the NBDE: the Bylaws, the Standing Rules, and the Examination Regulations. The Bylaws specify the membership, organization, and governance of the JCNDE. The Standing Rules provide the structure and function of the committees of the Joint Commission and the process for handling examination irregularities and candidate appeals.8 The standing committees of the Joint Commission include the Committee on Examination Development, the Committee on Administration, the Committee on Dental Hygiene, and the Committee on Research and Development. The Examination Regulations specify the details of examination operations, including fees, mechanisms for exam administration, and provisions for ensuring the validity and security of examination content.9 Examination regulations are updated annually and published in a candidate guide or examination manual.
In addition to its own guiding documents, the Standards for Educational and Psychological Testing promulgated by the American Educational Research Association, American Psychological Association, and National Council on Measurement in Education provide guidance to the Joint Commission in defining exam content and in conducting its testing programs. These standards are considered an authoritative source of professional technical guidance for testing agencies.10 They are intended to promote the sound and ethical use of tests and to provide a basis for evaluating the quality of testing practices.11 The Joint Commission periodically publishes a technical report to demonstrate the validity evidence supporting the examination programs and its compliance with accepted standards.12
| Theoretical Basis for the National Board Exams |
|---|
|
|
|---|
Scientific test theory is based on a link between the purpose of an assessment, conceptual perspectives on the nature of the knowledge or skills of interest, and observable manifestations of student competence, i.e., performance on an exam in the form of right and wrong answers. For the National Boards, the purpose, as previously described, is to determine whether students are qualified for entry into practice. Beliefs about their qualifications are based on statements of entry-level competence, as characterized by the ADEA Competencies for the New Dentist.14 Probability models allow us to sample domains of students knowledge related to these competencies, collect data, and draw conclusions about their level of competence and readiness for independent, entry-level dental practice. Matching the purpose of a test to the context in which it will be used is essential to achieving validity, i.e., confidence that the results provide relevant and appropriate information for decision making.
Validity is a fundamental consideration in the development and evaluation of examinations; it refers to the degree to which evidence and theory support the interpretations of test scores for the proposed purpose of the test. According to Fabrey, three assumptions are important for documenting the validity of a credentialing examination: 1) that there are certain critical abilities necessary for effective performance and that individuals who lack these abilities will not be able to adequately practice; 2) that individuals scoring low on the examination lack knowledge underlying these critical abilities and will not be able to practice in a safe and effective manner; and 3) that the examination can be designed to accurately identify the point at which the knowledge, skills, and abilities demonstrated on the examination are most indicative of the candidates ability to practice in a safe and effective manner.15 This point is most commonly known as the pass-fail point or the cut score. The relevant decision is this: does the candidate have the requisite knowledge to be granted a dental license?
| Development of the National Board Examinations |
|---|
|
|
|---|
Content Specifications and Topics Addressed or Emphasized
The NBDE Part II provides a good framework to illustrate how a national examination is constructed. This examination consists of 500 questions organized around nine major subject areas: operative dentistry, pharmacology, prosthodontics, oral and maxillofacial surgery and pain control, orthodontics and pediatric dentistry, endodontics, periodontics, oral diagnosis, and patient management (including behavioral science, dental public health, and occupational safety).16 The content specifications for the examination are determined using a recognized and structured psychometric process. The starting guide for selection of topic areas to be addressed in the examination is the 1997 ADEA Competencies for the New Dentist document.14 To confirm that these competencies continue to be important and relevant, the JCNDE conducts a practice analysis survey approximately every five years. The survey asks practitioners to rate the importance of sixty-five competencies (sixty-three ADEA Competencies and two competencies taken from the Accreditation Standards for Dental Education Programs)17 to their practice of general dentistry.
The practice analysis was last conducted in 2005 and included a randomized sample of 7,000 general dentists, stratified by licensure jurisdiction, who had been in practice for one to five years following graduation; 2,597 or 41 percent of those surveyed responded. Data from the survey are first used to confirm that each competency is still important and relevant to practice and the national examination and, second, to ultimately determine the number of test items that will be devoted to each competency and content area. The decision to retain a competency area in the examination and the number of exam items to be devoted to that competency are also influenced by comment from the profession and public as obtained through various forums, such as the JCNDE forum at the ADEA Annual Session. Using these data and input, JCNDE staff use the Rasch rating scale analysis as described by Kramer and Neumann18 to determine the number of questions that will be devoted to discrete topic areas in the examination. For example, in the 2006 Part II examination, forty-five test items were allotted to operative dentistry, thirty test items to endodontics, forty-four items to oral diagnosis, etc. Subsequently, a panel of experts reviews and cross-links the calculated allocation of items among competencies and content areas, and may make minor adjustments in the distribution of items among content areas.
Piloting New Test Items and Role of TCCs in Exam Construction
The Joint Commission uses a specialized computer software application to manage and store National Board questions. This item bank includes questions currently in use, draft questions, and questions that have been retired from use. Each examination includes a mix of previously used items and a number of new questions created by Test Construction Committees (TCCs) in an effort to expand the item bank and to provide a supply of fresh, contemporary questions for the examination. New questions are piloted before they are used as scored items on subsequent exams.12 Piloted questions are analyzed statistically for difficulty and discrimination and are not included in scoring unless they fulfill JCNDE criteria. For an item to be considered effective, it must produce a difficulty index between 40 percent and 80 percent, and a corresponding discrimination index of 0.15 or higher for Part I or 0.08 or higher for Part II.12 Questions that satisfy these criteria are retained in the test item bank and are then available for use in future examinations.
There are sixteen TCCs encompassing Part I (n=5) and Part II (n=11) related to the following subject areas: (Part I) Anatomical Sciences, Biochemistry/Physiology, Microbiology/Pathology, Dental Anatomy and Occlusion, and a multidisciplinary Testlet Development Committee; (Part II) Endodontics, Operative Dentistry, Oral and Maxillofacial Surgery-Pain Control, Oral Diagnosis, Orthodontics/Pediatric Dentistry, Patient Management, Periodontics, Pharmacology, Prosthodontics, Component B-Case Composition and Case Selection Committee, and a Consultant Review Committee.8 The size of each TCC varies but typically consists of four to six members with approximately 80 percent of members affiliated with dental schools (full-time or part-time appointments) and approximately 20 percent from the private practice sector. Tables 1
–4![]()
![]()
list the membership for each TCC. Each year the Joint Commission asks its member organizations (AADE, ADEA, and ADA), along with other dental groups, to nominate new candidates to fill vacant positions on the TCCs. The qualifications of the candidates are reviewed by the Joint Commission before they are approved by majority vote. The standard term of a TCC member is one year, renewable for a total of five consecutive terms.
|
|
|
|
Exam Construction
The role of each TCC ends when test items have been selected and assembled into a draft section for the examination. As a final step for the Part II examination, the Consultant Review Committee comprised of two experienced test constructors—a practitioner and a discipline-based dental expert—works with staff to review and refine the examination, with particular attention to case-based items. Consultant reviewers are responsible for test coherence and cohesion and for reviewing case materials for proper orientation, labeling, and linkage to test questions. At this stage, Joint Commission staff assumes primary responsibility for technical formatting, editing, and preparing the examinations for publication and administration. The Part I examination is compiled independently by Commission staff from items developed and selected by the various TCCs, while the Part II examination is compiled by staff from items produced in both the discipline-based and case-based TCCs with assistance from two dentists/consultant reviewers. At any one time, as many as six comparable versions of each exam (NBDE I, NBDE II) may be active in the system.
| Common Perceptions and Misperceptions About Testing and the NBDE |
|---|
|
|
|---|
"National Boards should test whats in the dental school curriculum."
The content of the National Board examination is outlined in a set of examination specifications that are reviewed and updated annually. The Part I and Part II exam specifications provide a blueprint for the respective examinations, delineating the major subject headings and number of items devoted to the content area, as well as the subtopics to be covered in each subject area.16 These blueprints guide test constructors in selecting and developing test items to sample candidate knowledge across the subject areas that have been identified as important to the safe and effective practice of dentistry.
Good testing practice, as defined by the Standards for Educational and Psychological Testing, requires that the blueprint or content domain of a credentialing exam be based on a job analysis or practice analysis derived from information concerning the actual behavior of competent practitioners.11 Joint Commission policy directs that a practice analysis be conducted every five years and that the exam specifications be based on the results of the practice analysis. The process for conducting a practice analysis has been described earlier. The Standards further advise that, in tests used for licensure, skills that may be important to success in the field but that are not directly related to the purpose of licensure should not be included.11 As argued by Chambers in his 2004 article on initial licensure testing, content validity of an exam depends equally on knowing that tasks on the test are part of dental practice and that the tasks are critical to the safe and effective practice of dentistry.23 Because the goals of most dental education programs emphasize the preparation of dental practitioners, there will be a substantial overlap between the content of dental curricula and the National Board exams. However, not all topics deemed appropriate for each schools mission and curriculum will be important for the purposes of demonstrating competence for safe, entry-level dental practice.
"Dental educators should determine National Board content."
Historically, the Joint Commission relied on the opinions and recommendations of subject matter experts serving on a number of discipline-based Test Construction Committees to formulate and update examination content specifications. For the most part, these individuals were dental educators nominated by their peers or personally volunteering their expertise. These test constructors were fully responsible for specifying examination content and for developing the test questions in each content area.
While the input and judgment of subject matter experts continue to play an important role in the development of individual test items, the current process for determining the overall examination blueprint, as previously explained, is confirmed by the results of the practice analysis. The Standing Rules of the Joint Commission delineate the criteria and professional credentials for selection and appointment of the subject matter experts, or consultants, to serve on Test Construction Committees.8 Because the purpose of the exam is to determine competence for practice, each committee includes at least one full-time dental practitioner, i.e., a dentist who has practiced thirty to forty hours per week for at least ten years.
Reliance on the results of a practice analysis to provide a broad outline of the knowledge underlying contemporary, entry-level practice combined with the expertise of educators and experienced practitioners serving on Test Construction Committees offers the potential for an exam that will be valid for the intended purpose and independent of the bias of any stakeholder group. The exam blueprint and the substance of individual items must broadly sample the biomedical, behavioral, and clinical sciences underlying the new dentist competencies in a manner proportionate to the relative importance of that content to current dental practice. Joint Commission policies regarding Test Construction Committees and determination of test content are intended to guard against turf battles and self-protectionism among committees and disciplines, as well as individual consultants who may attempt to dominate a committee or a component of an exam. Because of the significant overlap of dental curricula and exam content, it is important to have an objective process that is free of the territorialism and entrenched turfdom that have stymied curricular innovation.24,25 Reliance on exam specifications determined by a practice analysis is intended to guard against the tendency of dental faculty to write questions on what they are teaching instead of what they should be teaching.1 In this regard, the NBDE exam specifications set a baseline standard of the knowledge required for entry-level dental practice and can serve as an external standard or benchmark against which curriculum content and areas of emphasis can be compared.
"National Boards test minutiae and force students to memorize useless facts."
As suggested by the preceding discussion, the National Board exams sample from the vast areas of knowledge in the dental curriculum that define the competent dental practitioner. Through this sampling process, the exams will cover both critical principles and information that may be deemed by some to be of lesser importance. This approach satisfies state dental boards charged with protection of the public who believe that candidates for licensure must demonstrate that they have knowledge of basic dental terminology and sufficient foundational knowledge for use in patient care and for reading and evaluating dental research, literature, and manufacturers product and marketing materials.
Based on a standard-setting process described below, the JCNDE identifies important criterion items and develops a pass-fail rule that relates these criterion items to a standard score scale. This process allows the National Board exams to fully sample from the breadth and depth of their content specifications while preventing individual examinees from being disadvantaged by items that may be perceived as obscure or unimportant. Also, for this reason, the raw score percentage of correct items required for a passing score may seem relatively low.
The Joint Commission agrees that the exams should stress understanding and ability to apply information in a problem-solving context and has taken a number of steps to increase the proportion of exam items that require application and analysis of information rather than rote recall. Test Construction Committees identify the cognitive level of each item, and the Joint Commission tracks the relative proportion of items at each cognitive level in each exam.12 Table 5
provides a sample summary of the distribution of Part I items from a 2006 exam by clinical applicability and cognitive level, and Table 6
provides a sample summary of the distribution of Part II items from a 2006 exam by content category and cognitive level. Classification of items according to discipline and cognitive level is done by committees during the final phase of exam construction. All National Board items are evaluated for quality, either through pre-testing or through post-exam analyses of item statistics, such as indices of difficulty and discrimination. Items that are too easy, too difficult, or not discriminatory are eliminated from scoring and/or revised for future use.
|
|
"National Boards are graded on a curve."
The perception that National Board exams are graded on a curve, and as a corollary that they are competitive, is both fallacious and unhealthy for candidates, education programs, and the profession. National Board exams are criterion-referenced tests.28 That is, candidates are evaluated in relation to a standard criterion of performance. Exam results are intended to show if the candidate has met the standard, i.e., that the candidate has the minimum level of knowledge for safe, entry-level dental practice. National Board exam results are not intended to compare individuals or groups of candidates (a norm-referenced exam) or to show gradients of achievement or ability.
Terminology relating to scoring and exam results can be confusing. Official National Board scores are reported as standard scores. A raw score is simply the number or percentage of items answered correctly. A standard score is the result of the conversion of a raw score by small mathematical adjustments that make the score comparable to scores from all other test forms, past, present, and future. The conversion process equates raw scores on a common scale across multiple forms or versions of the test with a constant passing score point. This process ensures fairness and a valid interpretation of scores.
The pass-fail point is determined by a committee of distinguished content experts who make judgments to define a minimum level of acceptable performance using a formal standard-setting process.28 Setting the performance standard involves several steps. First, the content of each item is reviewed by the content experts to determine its importance to the practice of dentistry. Items essential to success in practice are known as criterion items. Next, the content experts estimate the minimum level of knowledge of the subject matter represented in the criterion items that would be required for licensure. By means of a statistical process, candidates performance on the criterion items is used as an index to calculate the minimum performance standard on all 400 or 500 items in a complete exam. The Joint Commission assigns the standardized score of 75 to signify the minimum performance for passing the examination and eligibility for National Board certification. The goal of the National Board scoring process is to answer this question: has the candidate passed the exam and met the standard?
"I got a 74. There was an unfair question (technical error, inaccuracy, etc.) or I would have passed. I want one point added to my score."
This plea reflects a common misperception of candidates whose scores are close, but short of the standard for passing. They perceive that the reported score is a raw score or percentage score rather than a standard score. In fact, one right or wrong answer in an exam of several hundred items generally does not equate to one point on the standard score scale.
As previously described, the Joint Commission uses a post-exam analysis process to evaluate the performance of every item on the exam with respect to difficulty, discrimination, and other characteristics. Items that do not perform appropriately or that include technical errors impacting the item may be removed from scoring. Joint Commission policy provides that up to 15 percent of items may be removed from scoring, although the number and percentage of items removed are usually quite small.
From another perspective, it is worthwhile to consider the quality of the National Board exams with respect to reliability. Over the history of the examination program, the reliability (KR20) of the National Boards has consistently exceeded acceptable levels.12,29 Accordingly, a strong likelihood exists that a candidate with a score of 74 truly lacks the minimum knowledge for safe practice and will fail to achieve a passing score on a subsequent examination without remediation or other intervention. The relatively high failure rate of repeating candidates affirms this assumption: the failure rate for repeating candidates from accredited dental schools was 26.5 percent for Part I and 25.1 percent for Part II, while the failure rate for first-time test takers was 16.3 percent for Part I and 7.7 percent for Part II in 2005.30
"How can I get my school into the top quintile?"
This has been a common question of dental school administrators in search of objective metrics that will serve as benchmarks to demonstrate program outcomes and satisfy university administrators requests for accountability. The question references an annual report produced for a number of years by the Joint Commission itemizing the coded results for each school, with a breakdown of mean score in quintiles from highest to lowest mean score. The format of the report was originally intended to respond to dental schools desire for an objective outcome measure for self-evaluation, accreditation, and reporting to higher level administration. While appearing to meet these needs, the ranking of schools by mean scores emphasized differences that were, in fact, quite small and, for the most part, practically insignificant. For example, for the reporting period May 1, 2004 through April 30, 2005, 78 percent of Part II mean scores for all schools were between 79.5 and 83.9.31 The difference between school #6 (83.9) in the top quintile and school #14 (83.0) in the next quintile was less than one point. The difference between schools #4 and #5 (84.1) and school #34 (81.1) was only three points. The difference in mean score between the top school in the third quintile and the next highest ranked school at the bottom of the fourth quintile was one tenth of a point; the difference between the top school in the third quintile and the top school in the fourth quintile was nine-tenths of a point. The separation of schools by quintile was the result of an arbitrary mathematical computation, rather than a meaningful distinction. While it might be argued that the difference between school #1 and school #54 was significant, the more meaningful information is reflected in data relevant to the purpose of the exam—the proportion of students failing the examination.
With respect to accreditation, the outcome of interest to the Commission on Dental Accreditation (CODA) and the U.S. Department of Education (USDE) is the first-time pass rate on licensure exams.32–34 Joint Commission scoring procedures are designed to ensure a high degree of confidence about pass-fail decisions, not about relative achievement of individual students or the effectiveness of individual faculty or educational programs based on fine distinctions among closely clustered scores at other points along the standard score continuum. In fact, the Standards for Educational and Psychological Testing indicate that tests used in credentialing may be designed to be precise only in the vicinity of the cut score; they may not be precise for those who clearly pass or clearly fail.11
"Competition for postgraduate programs hinges on National Board scores."
The discussion above applies equally to this statement. While National Board scores may represent objective measures of student knowledge and competence for practice, they were never intended to distinguish students by gradients of ability or achievement. Because U.S. dental schools are accredited and select students from a highly qualified and motivated pool of applicants who are cognizant of the expectations for graduation, licensure, and the demands of professional practice or advanced education, the range of scores for National Boards for students in accredited schools is narrowly clustered and skewed to the high end of the score scale.
The Joint Commission has a high level of confidence about score differences at the pass-fail point, but does not have evidence to support assumptions about comparisons at higher or lower ends of the score scale. Some studies of predictors of success in medical residencies have shown that applicants scores on the National Board of Medical Examiners (NBME) examinations have little or no value in predicting success of residents.35–37 A study of measures for the selection of international dental students to a U.S. D.D.S. program showed that Part II was the most significant predictor of academic performance and clinical competency in the admission process, but this use of the exam as described by the authors "to measure real-life dentistry knowledge" is more consistent with the intended purpose.38
| Summary and Conclusions |
|---|
|
|
|---|
As a first step, educators and practitioners should jointly review and revise the Competencies for the New Dentist. The existing competencies have served the profession well and, to a certain extent, have been validated through the Joint Commissions practice analyses. Because the competencies focus primarily on areas of knowledge and technical skill, it will be important that revisions or additions focus on competencies identified by some as lacking in both the dental curriculum and the National Board exams—skills for critical thinking and information evaluation and management.25,26
A companion article (following in this issue) will consider the framework for test construction activities and offer suggestions for meeting these needs. However, it is important to remember that the content and quality of items resulting from test construction activities are a reflection of the input and perspectives of individual test constructors. Reforming the exams will require the commitment and collective efforts of all who participate, the support of a dental education system that values the desired knowledge and skills, and acceptance by the state boards of dentistry.
| Footnotes |
|---|
A version of this article and a companion piece were presented at the American Dental Education Association (ADEA) 47th Deans Conference in November 2005.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. L. MacNeil and L. M. Neumann Realigning the National Board Dental Examination with Contemporary Dental Education and Practice J Dent Educ., October 1, 2007; 71(10): 1293 - 1298. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |