Procedures for Administration and Scoring

Procedures for Administration and Scoring Obtaining valid information from the MMPI-3 requires that standard procedures be employed in the administration and scoring of the instrument. Users are encouraged to follow the procedures described in this chapter so that they can rely optimally on the normative and validation data for the MMPI-3, which were collected following these procedures. Deviation from these standard procedures jeopardizes the integrity and interpretability of the resulting test scores. The greater the deviation from standard procedures for administration and scoring, the less confidence can be placed in the validity of the test results. Administration Proper administration of the MMPI-3 requires that the test taker be able to respond meaningfully to the items, that standard administration modalities be used, and that testing be adequately supervised in an appropriate environment. The individual administering the test should be either a qualified test user (described in Chapter 2 of this manual) or a technician trained by and working under the direct supervision of a qualified user. Following are the standard administration procedures for the MMPI-3. Assess the Testability of the Test Taker The MMPI-3 is typically administered through a visual medium—a test booklet or a computer screen. For a test taker to provide meaningful results he or she must be able to see, read, and comprehend the test items. Conduct a preliminary assessment of the testability of the test taker by determining whether he or she has any limitations or conditions that might impair his or her ability to respond meaningfully to the test items, and, if so, whether proper corrective steps can be taken. Ask the test taker whether she or he has any visual impairment. If she or he requires reading glasses, these should be available. If the test taker’s visual impairment is not correctable with reading glasses, use a standard audio administration of the MMPI-3 (see description below) and make arrangements for the test taker to record her or his responses privately. Forbey et al. (2009) demonstrated the comparability of MMPI results obtained with standard audio and computer administration of the MMPI (recall that the MMPI-3 norms were collected by computer administration of the instrument). Do not ask the test taker to respond to the test items orally and do not read the test items to the test taker, as both of these practices would constitute significant deviations from standard procedures. You can implement a private recording of the test taker’s responses by providing him or her an appropriate means for recording his or her responses (e.g., a notepad or a computer), which can later be transcribed onto an answer sheet or entered into the computerized scoring and reporting systems. Standard computerized administration of the MMPI-3 with Pearson scoring and reporting systems (described in the MMPI-3 User’s Guide for the Score and Clinical Interpretive Reports [Ben-Porath & Tellegen, 2020b]) includes audio and visual administration of the test items and ensures privacy as well. After confirming that the test taker can see the test items (or making appropriate accommodations), consider his or her reading ability. Some researchers and test developers have established a reading level for selfreport inventories by calculating a single aggregate indicator based on the entire item pool of the test. Following this procedure, Schinka and Borum (1993) estimated the reading level of the entire MMPI-2 item pool to fall between fourth and fifth grade. Morey (2007) reported that the Personality Assessment Inventory™ requires a fourth-grade reading level based on a similar analysis of its entire item pool using the FleschKincaid reading level index available in Microsoft® Word. Applying the same procedure to the MMPI-3 item pool yields a Flesch-Kincaid index of 4.5. Appendix D contains the Flesch-Kincaid reading levels for each MMPI-3 item. Establishing and comparing reading levels for self-report inventories is further complicated by the use of different response formats. The True/False response format of the MMPI-3 is less complex than Likert scale response formats, and, all things being equal, is less challenging for the test taker. This is particularly true of items that include quantifiers (e.g., “I sometimes . . .” or “I frequently . . .”) or are worded negatively (e.g., “I don’t . . .” or “I rarely . . .”), which gives rise to ambiguities when used with a Likert response format that includes its own quantifiers (such as “very inaccurate” or “slightly true”). Therefore, reading estimates based solely on item content that ignore response format differences do not offer a complete comparison of the cognitive resources needed to meaningfully complete self-report measures such as the MMPI-3. If a test taker’s ability to adequately read the MMPI-3 items is in doubt (e.g., because of possible reading limitations or because the test taker is not a native English speaker), first administer a standard test of reading ability. If the test taker’s reading level is less than sixth grade, or if it is not feasible to administer a reading test when reading ability is in doubt, administer the MMPI-3 using the standard audio recording with paper and pencil or the audio within the computer scoring and reporting system. As just noted, reading the items to the test taker represents a deviation from standard administration procedures and introduces an uncontrolled interpersonal element to the testing and is not recommended. Himsl el al. (2017) demonstrated that language comprehension deficits can reduce the validity of MMPI-2-RF responses independently of reading level. Audio presentation of the MMPI-3 items would not be an adequate solution for an individual with substantial language comprehension deficits. If the test is administered to an individual with reading or language comprehension deficits, the findings of Himsl et al. (2017) also indicate that this validity threat is likely to be reflected in the MMPI-3 Combined Response Inconsistency (CRIN) and Variable Response Inconsistency (VRIN) scale scores (see the interpretive guidelines for these scales in Chapter 5 of this manual). On the other hand, test takers with adequate reading, language comprehension, and cognitive abilities are likely to produce CRIN and VRIN T scores below 60 (assuming that scorable responses were given to all the scale items). Scores below 60 on these scales can, therefore, be interpreted as ruling out language comprehension problems. Use Standard MMPI-3 Administration and Response-Recording Modalities Standard MMPI-3 administration modalities include a booklet, computerized administration using Pearson’s scoring and reporting systems, and audio recordings of the test items. Each of these includes standard instructions for completing the test, which should not be altered or augmented. If a test taker requests clarification or interpretation of an item, refrain from providing guidance other than supportively restating the standard instructions. When the booklet or audio recording is used to administer the test, an answer sheet is the recording modality. In addition to the Spanish-language MMPI-3 (Ben-Porath et al., 2020), other approved translations of the test are in development. Up-to-date information on approved translations of the MMPI-3 can be obtained at www.upress.umn.edu/test-division. Administration time for the MMPI-3 depends on the administration modality. Computerized administration is appreciably faster than use of a booklet and answer sheet. Test takers with normal-range cognitive functioning and reading skills can typically complete a computer administration of the MMPI-3 in 25–35 minutes. A booklet and answer sheet administration typically requires 35–50 minutes. Supervise Testing and Maintain Test Security The Standards for Educational and Psychological Testing (Standard 9.21) require that test users make reasonable efforts to protect the integrity of test scores by eliminating opportunities for test takers to obtain scores fraudulently. Although the MMPI-3 is a self-administered test, a qualified user or a technician working under the supervision of a qualified user must supervise completion of the inventory. Adequate supervision ensures that the test taker completes the inventory on his or her own, that any unusual events that may occur during testing are recorded and can be considered in the interpretation of the results, and that conditions conducive to obtaining optimally valid information are maintained. In addition, to protect test security, test takers should not be allowed to take photographs of MMPI-3 test materials. Supervision does not require that the individual administering the test be in the same room as the test taker throughout the session. However, the test taker should be within the supervisor’s line of sight. For the same reasons, MMPI-3 materials should not be sent home with test takers nor, in institutional settings, should test takers be allowed to complete the instrument in their rooms or anywhere else where supervision is not possible. The standards (9.21) also state that test users are responsible for protecting the security of test materials at all times. Maintain a Quiet, Comfortable Environment MMPI-3 administration should occur in a quiet, comfortable environment in which the test taker is free of distraction. Testing can be conducted individually or in groups, but when more than one individual is tested the supervisor must remain in the room (or in visual line of sight) to ensure that test takers do not bother or distract each other. Remote administration of the MMPI-3 is possible using Pearson’s scoring and reporting systems. Corey and Ben-Porath (2020b) provide guidance for this procedure. A key requirement is that remote visual and audio supervision be incorporated in a remote administration of the test by means of a telehealth service that complies with federal and state mandates. Remote supervision is necessary to maintain test security and the integrity of test scores under Standard 9.21, as just discussed. Scoring MMPI-3 protocols can be scored by computer or by hand. Computerized scoring offers several advantages, including speed, accuracy, and completeness. It may also be cost-effective if the clerical expenditures associated with hand scoring are considered. The most efficient way to implement computerized scoring is to administer the MMPI-3 by computer (including tablet) using Pearson’s scoring and reporting systems. It eliminates the need to enter the test taker’s responses into the system and shortens administration time considerably. If administering the MMPI-3 by booklet and answer sheet, users can use a scanner to enter the item responses into the Pearson scoring and reporting systems or users can enter the test taker’s responses by keyboard and score using the scoring and reporting systems. If keying in the responses, the scoring and reporting systems provide the option of re-entering a protocol, which is strongly recommended to reduce the likelihood of entry error. Users can also mail answer sheets to Pearson for computerized scoring. Computerized scoring produces the MMPI-3 Score Report or one of the available interpretive reports. In addition to raw scores and T scores on all the scales of the instrument, these reports indicate the percentage of items answered on each scale, provide a list of unscorable items (either unanswered or answered both True and False) and the scales on which they appear, and include a list of items answered in the keyed direction on elevated “critical scales” (a subset of scales designated by the authors as having critical item content that may require immediate attention and follow-up). Users have the option of including additional item-level information when printing the test results. The MMPI-3 reports also provide users the option to plot a test taker’s results together with those of one of the comparison groups described in Appendix F of the MMPI-3 Technical Manual. This option reports the percentage of comparison group members scoring at or below the test taker for each scale. The interpretive reports also include a narrative interpretation of the test results. The MMPI-3 User’s Guide for the Score and Clinical Interpretive Reports (Ben-Porath & Tellegen, 2020b) provides more detailed information about the MMPI-3 Score Report and the MMPI-3 Interpretive Report for Clinical Settings. Separate user’s guides are available for setting-specific interpretive reports (e.g., the MMPI-3 User’s Guide for the Police Candidate Interpretive Report [Corey & Ben-Porath, 2020a]). Hand scoring requires the use of answer keys (plastic templates) available from Pearson. Appendix F contains the hand-scoring profiles. Given the number of scales on the MMPI-3 (52) and the complexity of scoring some of these scales (particularly VRIN and TRIN), hand scorers may be tempted not to score all the scales. Because scores on all the MMPI-3 scales need to be considered for a comprehensive and complete interpretation of a protocol, users must not omit scoring any scales. Moreover, the additional information provided in the MMPI-3 reports is not readily available when the instrument is hand scored. Computerized scoring, therefore, offers several advantages over hand scoring of the MMPI-3.