ࡱ>  bjbjVV >R<<377777KKK8oK-%8(111   &$($($($($($($e&)($7     ($7711=$@ 7171&$ &$1KV$}$-%)))7   ($($   -%    )          : Prediction of MMPI-2 Clinical Scales for Incomplete Protocols: Comprehensive Short-Form Analysis F. Scott Winstanley Department of Veterans Affairs Medical Center, Detroit, Michigan Bradley N. Axelrod Department of Veterans Affairs Medical Center, Detroit, Michigan Leif V. LaLone Anishnabek Community and Family Services, Sault St. Marie, Michigan John A. Schinka Department of Veterans Affairs Medical Center, Tampa, Florida Running Head: MMPI-2 Short Forms Address correspondence to:  HYPERLINK "mailto:Bradley" Bradley N. Axelrod; Psychology Section (11-MH-PS); John D. Dingell DVAMC; 4646 John R, Detroit, MI 48201-1916 Electronic communication may be sent to:  HYPERLINK "mailto:Bradley.axelrod@va.gov" Bradley.axelrod@va.gov The MMPI-2 is a thorough tool for personality assessment that has substantial importance in the field of neuropsychology. However, there have been reported problems for many neurologically impaired individuals who do not complete the test because of the demands of its length. Incomplete protocols are of little value with no formal way of scoring and interpreting the completed items. The following study examined the clinical utility of short-form versions of the MMPI-2 validity and clinical scales. Raw score correlations between various short-form and full-form tests on all validity and clinical scales, as well as mean raw score differences between short-form and full-forms were examined. These mean raw scores were converted into T-scores to determine how accurately short-form versions can predict T-scores within 5 and 10 points. The following provides lookup tables which can provide useful scoring and interpretation guidelines for incomplete protocols with varying number of items completed (e.g. 180, 200, 250, and 300 items completed). The Minnesota Multiphasic Personality Inventory-2 (MMPI-2) is one of the more popular instruments in psychological assessment (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989). However, many neurologically impaired patients cannot handle the demands of an arduous 567 questionnaire, or even the required 370 items necessary for validity and basic clinical scale interpretation. In these special circumstances when a patient discontinues prior to completing 370 questions, there exists a need to salvage some clinical interpretation of their emotional status from the objective test data. For this reason, various short-forms of the MMPI-2 continue to be developed and tested against the full-item version. A variety of interpretation strategies have been evaluated regarding the utility of these shortened versions. Interpretive approaches such as raw score correlations, two point and high point code type congruence, classification of pathological (T > 65) versus non-pathological, and classification accuracy of T-scores within 5 and 10 T-score points, have all been evaluated and have met with fair amounts of criticism. It should be stated at the outset that literature pertaining to the original MMPI clearly come out opposed to shortened versions of the test. Hathaway (1975; as presented in Butcher & Williams, 2009) clearly pointed to concern regarding any version of the MMPI that would be abbreviated, fearing that the loss of data points would result in loss of accuracy. At the dawn of the MMPI-2 introduction, Butcher and Hostetler (1990) presented a detailed article on MMPI short forms, and their apparent failures. Primary concerns raised include decreased reliability, internal consistency, and subsequent diminished validity. Consequently, they adamantly opposed the introduction, study, or use of any short form aside from using only the first 370 items. The exception mentioned was to use a data from a single scale, for research purposes, and explicitly state that it was not the full MMPI-2. Despite the admonitions by Hathaway, Butcher, and Hostetler, a few studied examining short forms have been published explicitly to evaluate if full validity and clinical scales might be estimated if an examinee failed to complete the entire task. One of the more prominent short-forms of the MMPI-2 was the 180-item short-form developed by Dahlstrom and Archer (2000). The protocols from the re-standardization sample of the MMPI-2, consisting of 2,600 men and women, were used in their analysis. Their results showed correlations between 180-item short-form and full-form scores ranging from .78 (scale 6) to .94 (L scale). Dahlstrom and Archer cross-validated these findings on a psychiatric sample of 632 inpatients, and yielded correlations between .82 (scale 5) to .99 (scale 1). Their findings also revealed very small raw score mean differences between prorated and full scale scores in both the validity and basic clinical scales (between 3 raw score points). These results were produced in both their validation and cross-validation sample. Though this article offered evidence toward utilization of this 180-item short-form version, critics point out that linear relationships and accurate mean score predictions may not be sufficient when examining other areas of interpretation with the individual protocols (Gass & Gonzalez, 2003). Other interpretive strategies where shortened versions of the MMPI-2 have received criticism are in the code type interpretations and peak score interpretation (Gass & Gonzalez, 2003). Both the original Dahlstrom and Archer study (2000), as well as a similar study by Gass and Luis (2001), revealed persistent lack of code-type and peak score congruence when examining individual protocols. At times, as low as one-third accurate prediction in two-point codes and only one-half peak score congruence (Dahlstrom & Archer, 2000; Gass & Luis, 2001). Another approach of incomplete protocol interpretation is determining whether or not full scale scores are pathological (T> 65) or normal (T< 65). In the Gass & Luis study (2001) the shortened form version appeared to be a reliable predictor, with average classification accuracy of 88.5%. The highest accuracy scores were on scale 5 (98%), K scale (97%), scale 3 (94%), and the L scale (93%). The poorest accuracy rate was on scale 0 (77%). Clearly, the 180-item short-form can be seen reliable when trying to determine if there is some sort of pathology, or not, on any given scale of the MMPI-2. However, the authors note two major problems with using the test in this manner. The first weakness is that even with low error rate per scale, the probability for error in the overall interpretation increases as multiple scales, each with individual error rates, are examined (Gass & Luis, 2001). Another limitation of this approach is that there is limited information provided regarding number of symptoms, symptom type, and symptom severity that can be associated with any one scale that has been classified as pathological. Gass & Luis (2001) give an example of the interpretation one would make if there were a T of 70 versus T of 90 on the scale 2. Clearly, there is much more to symptom interpretation to be explored with the latter T-score of 90. With regard to supplemental interpretation, the very nature of the task prohibits any utilization of the Harris Lingoes subscales. Though not without its merits in certain emergencies, the authors warn heavily against the utilization of the short-form version in MMPI-2 interpretation. It should be noted that the Gass and Luis (2001) sample was heterogeneous in its make-up (e.g., including stroke, TBI, other neurodegenerative disorders). We contend that of the multiple approaches mentioned above, raw score and T-score correlation and regression analysis, along with classification rates within 5 T-score points between the short-form and full-item versions, are the best way to extract the more salient information from these incomplete protocols. This particular interpretation has also been met with criticism in the literature. Upon reviewing the frequency of accurate score prediction using the MMPI-2 in their sample of 205 brain-injured patients, Gass and Luis (2001) found <60% accuracy of prediction (+/- 5T-score points), on scales F, 3, 4, 5, 6, 7, 8, 9, and 0, with the 180-item short-form version. When the margin of error was raised to +/-10T, there was still an error rate of over one-third of the cases with scales 6, 7, and 8 (Gass & Luis, 2001). One positive aspect of the Gass & Luis (2001) study, regarding utilization of the short-form, was that there was an over 80% classification accuracy rate within 5T with the L scale and scale 1. Another overlooked area is the validity of short-forms with varying number of items. Dahlstrom and Archer (2000) reportedly ran analyses for 150, 180, 200, 250, and 300 item short-forms. By their judgment, the 180-item short-form appeared to provide a maximum of valid variance with a minimum of time to administer the various test segments (2000, p. 133). They failed to report the results for the other short-form comparisons. Re-evaluation of those additional short-forms is an area that all subsequent research has neglected and will be addressed in this paper. The purpose of this study was to further preliminary research on the utility of short-form versions of the MMPI-2. These short-forms are not intended to replace administration of the full MMPI-2, but serve as a basis for interpretation in special situations when a full MMPI-2 is not completed by a patient. Short-form versions consisting of 180, 200, 250, and 300 items were examined in both male and female populations. As stated above, others have taken the 180-item short-form developed by Dahlstrom and Archer (2000) as the benchmark for short-form analysis and interpretation. This study will attempt to examine if adding a few more items to the set (e.g. 200, 250, or 300 item versions) will make the use of short-forms more valid, and offer clinicians their own choice with regard to what is the most parsimonious short-form measure. METHODS Participants The original database consisted of 2468 records. This sample was taken from the general psychological testing service at a large metropolitan Department of Veterans Affairs Medical Center (VAMC). After screening for invalid profiles (F with T> 100, raw > 20) there were 1938 cases remaining. Of those cases, 1747 are male and 191 female. Procedure The following procedures were performed on both the male only sample (n= 1747) and female only sample (n=191). In order to establish regression equations, 70% of each sample was selected as an origination group. The origination group for the male only sample was n= 1257, and for the female only, n= 142. Those cases were used in establishing the predictive regression equation. Raw scores from the shorter version of the tests were regressed onto the full versions. This was done instead of a simple arithmetic prorating. Reasons for this are that currently there is no information that indicates whether all items within each scale are responded to in the same manner. Individuals who endorse depression items may respond more during the latter half of the scale as opposed to earlier, and vice versa. Therefore, eliminating one half of the scale and prorating would fail to address the true nature of item responses within the scale. The regression analysis controls for these potential differences of response rates. Estimated raw scores were computed by using the raw scores for the cross validation sample and applying the obtained regression equations. Cross-validation of this equation was used on the remaining 489 cases in the male sample, and on the 49 cases in the female sample. The following are reported analyses for the MMPI full item set (Full) versus 180, 200, 250, and 300-item short-form version of the test, for both male and female populations. RESULTS 300-item MMPI-2 for Males The observed and estimated raw scores for all validity and clinical scales are presented in Table 1. Results of paired sample t-tests with Bonferroni correction (p = .05/13= p< .004) between raw scores of predictive 300-item short-form and the full set revealed significant differences only on scale 7, t (1, 488) = 2.858, p= .004. Pearsons correlations for the F and K scales, as well as scales (2, 6, 7, 8, and 0) were all .95 and above. The expected scores accounted for at least 90% of the overall variance of the full item scores. Correlations and t-tests were not performed for the L scale and scales 1, 3, 4, 5, and 9 because all items for these scales are included in the first 300 items. When raw scores were converted to T-scores, within subjects ANOVA between means of full item and short-form tests revealed significant differences only on scale 7, F (1, 488) = 8.166, p = .004, eta2 = .016 and scale 0, F (1, 488) = 6.749, p = .010, eta2 = .014. The effect size on both scales is very small indicating these differences may have little to do with the variance in number of items. With regard to classification accuracy, the 300-item short-form version yielded perfect classification rates (100%) within 5 T on the L scale, scales 1-5, and scale 9. Classification rates within 5 T as high as 90% and above were observed on the K scale and scale 8. The lowest rates within 5 T were observed on scales 6 (85%), 7 (87%), and 0 (84%). 250-item MMPI-2 for Males The observed and estimated raw scores for all validity and clinical scales are presented in Table 2. Results of paired sample t-tests with Bonferroni correction between mean raw scores of predictive 250-item short-form and the full item set revealed no significant differences on any validity or clinical scale. All Pearsons correlations were .90 or above between short-form and full-form mean raw scores, except for scale 6 (r =. 88). Short-form scores on almost all scales accounted for at least 85% to 98% of the overall variance. When raw scores were converted to T-scores, within subjects ANOVA between full item and 250-item short-form tests revealed no significant differences on any clinical or validity scale. Analysis of percentage of correctly classified individual cases on the 250-item short-form version within 5T-score points revealed perfect 100% classification accuracy rates for the L scale and scales (1, 2, 3, and 9). Scales 4 and 5 had accuracy rates of 95% and 96% respectively. The F and K scale revealed 89% and 93% accuracy rates respectively. Classification rates for scales (6, 7, 8, and 0) were 52%, 60%, 71%, and 71% respectively. As is represented in the literature, these scales have poorer classification rates between +/- 5 T-score points. With regard to classification rates within 10 T-score points, scales L, 1-5, and 9, all had perfect 100% classification rates. Scales F, K, 8, and 0 all had 95% and above classification rates within 10 T-score points. The lowest classification rates were again found on scale 6 (81%) and scale 7 (90%). 200-item MMPI-2 for Males The observed and estimated raw scores for all validity and clinical scales are presented in Table 3. Results of paired sample t-tests with Bonferroni correction between predictive 200-item short-form from the validation sample and the full set are reported. Paired samples t-test between raw scores of 200-item short-form and full item scores revealed no significant differences in raw scores on any of the validity or clinical scales. When raw scores were converted to T-scores, within subjects ANOVA revealed no significant differences on any scale. Pearson correlations revealed all validity scales and seven of the 10 clinical scales to have correlations of .90 and above (scales 1-4, 7-9). Only five of the ten clinical scales and none of the validity scales accounted for 85% or more of the variance. With regard to classification accuracy, the highest rates within 5 T were observed on scale 1 (96%), the L scale (93%), and scale 3 (90%). Scales (6, 7, 8, 0) and the F scale had classifications rates lower than 65% accuracy within 5 T. Consistent with all other tests, when the margin is widened to 10 T, all classification rates were elevated. 180-item MMPI-2 for Males The observed and estimated raw scores for all validity and clinical scales are presented in Table 4. Results of paired sample t-tests with Bonferroni correction between raw scores of predictive 180 item short-form from and the full item set revealed no significant differences on any validity or basic clinical scale. The high raw score correlations between short-form and full-form tests on each scale are also consistent with those reported by other studies (Dahlstrom & Archer, 2000; Gass & Luis, 2001; Gass & Gonzalez, 2003). When raw scores were converted to T-scores, within subjects ANOVA between 180-item short-form and full item tests revealed no significant differences on any clinical or validity scale. These analyses were not reported in the previous studies. With regard to classification accuracy rates within 5 T, the highest scale was scale 1 (96%). Only two other scales had accuracy rates above 80% (scales 1 and L scale). Scales (5, 6, 7, 8, and 0) had a classification rate of 60% or below. Overall poor classification accuracy rates within 5 T-score points are consistent with what is reported elsewhere regarding 180-item short-form versions, continuing to question the purpose of setting the short-form benchmark at 180 (Dahlstrom & Archer, 2000; Gass & Luis, 2001; Gass & Gonzalez, 2003). The results of our 180-item classification rates were slightly higher on most scales, but clearly they remain poor predictors of full-item T-scores within a reasonable range. When the range was moved more liberally to within 10 T-score points, roughly 70% of the scales were 90% and above. This is also slightly higher than what was found in the Gass & Gonzalez (2003) and Gass & Luis (2001) studies, but it is agreed that 10 T-score points in either direction is too wide of a range for valid interpretation. Classification accuracy rates were never performed in the original Dalhstrom and Archer (2000) study. 300-item MMPI-2 for Females The observed and estimated raw scores for all validity and clinical scales are presented in Table 5. Paired sample t-tests with Bonferroni correction between observed and estimated raw scores for 300-itme short-form for the female sample revealed no significant differences on any validity or clinical scale. When raw scores were converted to T-scores, within subjects ANOVA between means of full item and short-form tests revealed no significant differences only on any clinical or validity scale. Classification rates of the 300 item female only short-form revealed 100% accuracy within 5 T on scales L, 1-5, and 9. The lowest scale was scale 7 (76% accuracy within 5 T). The remaining scales had accuracy rates of 85% and above. 250-item MMPI-2 for Females The observed and estimated raw scores for all validity and clinical scales are presented in Table 6. Results of paired sample t-tests with Bonferroni correction between raw score female only predictive 250-item short-form and the full item set revealed no significant differences on any clinical or validity scale. As the table indicates, Pearson correlations were .90 or higher on all scales except for scales 6 and 0. This indicates predicted scores accounting for 85% or more of the variance of the observed items on all but two scales. Within subjects ANOVA of converted T-scores between 250 item short-form and full item tests revealed significant differences on scale 1, F (1, 48) = 5.258, p= .026, eta2= .099 and scale 8, F (1, 48) = 4.557, p= .038, eta2 = .087. Classification rates for the 250 item female only short-form revealed 100% accuracy within 5 T on the L scale and scales (1, 2, 3, and 9). Scales 4 and 5 yielded moderately high rates of 98% and 94% respectively. Classification rates within the 5 T range were poor (below 75%) for the F scale and scales (6, 7, 8, and 0). 200-item MMPI-2 for Females The observed and estimated raw scores for all validity and clinical scales are presented in Table 7. Results of paired sample t-tests with Bonferroni correction between raw score female only predictive 200 item short-form and observed full item raw scores revealed no significant differences detected on any validity or clinical scale. When mean raw scores were converted to T-scores, within subjects ANOVA of converted T-scores of full item and short-form tests revealed significant differences on scale 1, F (1, 48) = 7.855, p = .007, eta2 = .141 and scale 8, F (1, 48) = 5.880, p= .019, eta2 = .109. The effect size on both scales is very small indicating these differences may have little to do with the variance in number of items. Classification rates for the 200 item female only short-form revealed 96% accuracy within 5 T on scales 1 and 2. Scales 5-8 and scale 0 yielded rates of 65% and below. 180-item MMPI-2 for Females The observed and estimated raw scores for all validity and clinical scales for 180-item short-form, females, are presented in Table 8. Results of paired sample t-tests with Bonferroni correction between raw score female only predictive 180-item short-form and observed full-form revealed no significant differences on any validity or clinical scale. Within subjects ANOVA of converted T-scores between full-item and short-form tests revealed significant differences on scale 1, F (1, 48) = 7.855, p= .007, eta2 = .141; scale 4, F (1, 48) = 5.614, p= .022, eta2 = .105, and scale 8, F (1, 48) = 4.150, p= .047, eta2 = .080. In this case as well, the effect size on these scales is small indicating these differences may have little to do with the variance in number of items. With regard to classification accuracy rates within 5 T-score points, the highest accuracy rate was observed on scales 1 (96%) and 2 (90%). The F scale and scales (5, 6, 7, 8, 9, and 0) all had classification accuracy rates of 65% and below. These poor classification rates are similar to what was reported by Gass & Luis (2001), and Gass & Gonzalez (2003). DISCUSSION The major implication of this study is that there are now data available to interpret incomplete MMPI-2 protocols at a number of different cut-offs (e.g. 180, 200, 250, and 300). This is the first study to report such comprehensive analysis on all basic clinical and validity scales, for multiple short-form tests. The intent was to evaluate protocols that might result from an examinee discontinuing early. In other words, if one were to respond to only the last 300 items of the MMPI-2, the information from these tables would not obtain. To use these estimations of full MMPI-2 test scores, the clinician must first find the appropriate gender tables and number of items that had been completed. The raw scores for each scale should be entered as X in the regression equations, and then add the constant to the resulting product. The final total is the prorated raw score which can then be plotted on the MMPI-2 profile sheet. K corrections for each of the relevant scales must be computed based on the obtained prorated raw score for K. This paper provided information for both males and females separately. Historically, separate norms have been developed for males and females on all validity and clinical scales of the MMPI-2. Given the differences between male and female samples revealed in the study, particularly in paired samples t-tests, obtained T-scores on particular scales, within subjects contrasts of converted T-scores on certain scales, and T-score classification rates on multiple scales, this analysis seems warranted. Also, there are no means and standard deviations to create linear T-scores for analysis. The newest non-gendered norms are uniform T-scores on all scales (Ben-Porath & Forbey, 2003). Compared with all other short-forms in this analysis, the 300-item short-form had either equal to or higher classification rates within 5 and 10 T, equal to or higher correlations, and accounted for equal or more variance on all validity and clinical scales. However, the most parsimonious short-form based on our research (weighing clinical information gained versus time saved with item deletion) may be the 250-item short-form. The results showed this short-form had no significant differences from the full-form in raw score paired sample t-test or within subjects ANOVA for converted T-scores. In the context of administration time, the mean scores on this version showed the best raw score correlations and accounted for the most variance per scale of the full-form. The 250-item test also showed the high T-score classification rates within 5 T. Upon analyzing the data and weighing how much information is gained over the 180-item short-form versus the additional amount of time it takes to complete 70 extra items, the 250-item short-from is a much more useful and statistically valid instrument. The results of this study yield similar results regarding the 180-item short-form in its inability to correctly classify an adequate amount of cases between 5T and 10T-score points. Scales which consistently yielded the best classification accuracy rates within 5T, regardless of number of items completed, were scales 1-3 and the L scale. Perfect classification accuracy rates within 5 T were noted on scales 1-3, scale 9, and the L scale, with only 250 items completed, followed by 95% and above accuracy rates on scales 4 and 5. Scales which consistently did not yield high classification rates on short-form versions were scales 6-8, scale 0, and the F scale. These patterns were observed in both the male and female population. This paper would be remiss if the newest version of the MMPI were not mentioned. The Restructure Form of the MMPI (MMPI-2-RF) was published in 2008 using different statistical methodology from the MMPI-II (Ben-Porath & Tellegen, 2008). The intent was to retain the construct validity of the clinical scales which were to be more orthogonal than observed in the MMPI-II. The present study addressed the traditional validity and clinical scales of the MMPI-II, not the Restructured Clinical scales or the newer MMPI-2-RF version of the task. Critics state that short-forms utilizing correlations and regression equations is insufficient because it overlooks the degree of absolute score agreement (Gass & Gonzalez, 2003). However, there appears to be adequate amounts of score agreement between these forms, particularly in the area of classification rates per scale, particularly in the context of salvaging an incomplete protocol in special circumstances, and not entirely replacing a 567 or 370 item MMPI-2 for one of these shortened versions. Gass & Gonzalez (2003) argue that unless research has established extra test behavioral correlates for a short-form, the frequency of accurate individual full-form prediction is essential for determining short-form validity. (p. 526). The current research has demonstrated that on certain scales on certain short-forms, there is 100% agreement between full and short-form amongst individual protocols, and therefore, demonstrate its clinical validity for use in these emergency situations. The MMPI-2 remains an excellent tool for objective assessment of personality. Information obtained is too valuable to be discarded on the basis of a semi complete protocol, and the following research has yielded results which can salvage useful information from these protocols. MALE TABLES Table 1: Means, standard deviations of raw scores, correlations, and converted T-scores for males only cross-validation sample (N=489) for Full MMPI-2 (observed) and 300-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 4.75(2.27) 55.35(9.96) 100 100 L300 (X* 1.0) + .0 4.75(2.27) 55.35(9.96) F 9.51(5.17) .976 65.38(15.97) 89 99.4 F300 (X*1.166) + .244 9.54(5.03) 65.47(15.52) K 12.41(4.93) .957 43.92(10.36) 93 100 K300 (X*1.226) + .816 12.43(4.84) 43.96(10.17) SC1 16.58(6.68) 59.80(17.48) 100 100 SC1-300 (X*1.0) + .0 16.58(6.68) 59.80(17.48) SC2 29.63(7.32) .998 74.65(15.96) 100 100 SC2-300 (X*1.027) - .399 29.63(7.32) 74.64(15.94) SC3 30.21(6.61) 69.74(13.98) 100 100 SC3-300 (X* 1.0) + .0 30.21(6.61) 69.74(13.98) SC4 22.72(6.14) 50.06(13.26) 100 100 SC4-300 (X*1.0) + .0 22.72(6.14) 50.06(13.26) SC5 25.21(4.43) 48.42(8.71) 100 100 SC5-300 (X*1.0) + .0 25.21(4.43) 48.42(8.71) SC6 13.34(4.57) .967 61.29(15.91) 85 98 SC6-300 (X*1.097) + .885 13.36(4.37) 61.38(15.22) SC7 22.74(10.45) .973 42.13(21.67) 76 96 SC7-300 (X*1.454) - .937 22.43(10.32)* 41.48(21.42)** SC8 25.71(12.66) .993 48.63(22.09) 96 100 SC8-300 (X*1.142) - .847 25.58(12.46) 48.39(21.75) SC9 19.12(4.85) 48.06(11.34) 100 100 SC9-300 (X*1.0) + .0 19.12(4.85) 48.06(11.34) SC0 34.04(11.82) .955 60.53(13.37) 100 100 SC0-300 (X*1.724) + .539 33.63(11.06) 60.06(12.52)** *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" Table 2: Means, standard deviations of raw scores, correlations, and converted T-scores for males only cross-validation sample (N=489) for Full MMPI-2 (observed) and 250-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 4.75 (2.27) .994 55.35(9.96) 100 100 L250 (X*1.028) - .056 4.74(2.25) 55.29(9.89) F 9.51(5.17) .946 65.38(15.97) 73 95 F250 (X*1.306) + .880 9.60(4.95) 65.64(12.29) K 12.41(4.93) .927 43.92(10.36) 86 99 K250 (X*1.314) + 1.429 12.35(4.73) 43.80(9.94) SC1 16.58(6.68) .997 59.80(17.48) 100 100 SC1-250 (X*1.021) + .189 16.59(6.67) 59.81(17.47) SC2 29.63(7.33) .994 74.65(15.96) 100 100 SC2-250 (X*1.043) - .321 29.61(7.28) 74.60(15.86) SC3 30.21(6.61) .993 69.74(13.98) 100 100 SC3-250 (X*1.004) + 1.435 30.24(6.60) 69.81(13.95) SC4 22.72(6.14) .978 50.06(13.26) 95 100 SC4-250 (X*1.126) + .178 22.68(5.91) 49.98(12.76) SC5 25.21(4.43) .950 48.42(8.71) 96 100 SC5-250 (X*1.068) + 1.414 25.15(4.18) 48.31(8.22) SC6 13.34(4.57) .880 61.29(15.91) 52 100 SC6-250 (X*1.374) + 4.152 13.31(3.93) 61.20(13.70) SC7 22.74(10.45) .951 42.13(21.67) 60 90 SC7-250 (X*1.708) + .761 22.60(10.05) 41.83(20.86) SC8 25.71(12.66) .973 48.63(22.09) 71 95 SC8-250 (X*1.483) + 1.777 25.71(12.22) 48.63(21.33) SC9 19.12(4.85) .987 48.06(11.34) 100 100 SC9-250 (X*1.026) + .295 19.13(4.80) 48.09(11.22) SC0 34.04(11.82) .917 60.53(13.37) 71 95 SC0-250 (X*2.288) - .054 33.90(10.79) 60.36(12.19) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" Table 3: Means, standard deviations of raw scores, correlations, and converted T-scores for males only cross-validation sample (N=489) for Full MMPI-2 (observed) and 200-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 4.75 (2.27) .939 55.35(9.96) 93 99.8 L200 (X*1.153) + .380 4.77(2.22) 55.45(9.71) F 9.51(5.17) .920 65.38(15.97) 64 91 F200 (X*1.431) + 1.602 9.52(4.70) 65.41(14.52) K 12.41(4.93) .914 43.92(10.36) 80 99 K200 (X*1.477) + 1.431 12.32(4.69) 43.74(9.85) SC1 16.58(6.68) .987 59.80(17.48) 96 100 SC1-200 (X*1.150) + 1.192 16.63(6.62) 59.92(17.33) SC2 29.63(7.33) .971 74.65(15.96) 86 100 SC2-200 (X*1.132) + 3.182 29.62(7.16) 74.71(15.60) SC3 30.21(6.61) .969 69.74(13.98) 90 100 SC3-200 (X*1.081) + 4.376 30.23(6.49) 69.80(13.72) SC4 22.72(6.14) .953 50.06(13.26) 83 99.6 SC4-200 (X*1.200) + 3.547 22.63(5.71) 49.86(12.33) SC5 25.21(4.43) .850 48.42(8.71) 74 99 SC5-200(X*1.159) + 5.714 25.18(3.76) 48.37(7.41) SC6 13.34(4.57) .859 61.29(15.91) 45 80 SC6-200 (X*1.493) + 4.215 13.23(3.80) 60.92(13.23) SC7 22.74(10.45) .942 42.13(21.67) 56 86 SC7-200 (X*1.799) + 1.393 22.61(9.98) 41.85(20.71) SC8 25.71(12.66) .956 48.63(22.09) 59 90 SC8-200 (X*1.740) + 2.440 25.65(11.99) 48.52(20.92) SC9 19.12(4.85) .895 48.06(11.34) 73 96 SC9-200 (X*1.287) + 3.304 19.21(4.49) 48.26(10.49) SC0 34.04(11.82) .891 60.53(13.37) 64 93 SC0-200 (X*2.728) + 2.238 33.80(10.83) 60.25(11.57) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" Table 4: Means, standard deviations of raw scores, correlations, and converted T-scores for males only cross-validation sample (N=489) for Full MMPI-2 (observed) and 180-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 4.75 (2.27) .923 55.35(9.96) 88 99 L180 (X*1.214) + .387 4.79(2.17) 55.52(9.52) F 9.51(5.17) .912 65.38(15.97) 62 91 F180 (X*1.518) + 1.616 9.58(4.69) 65.60(14.47) K 12.41(4.93) .904 43.92(10.36) 79 99 K180 (X*1.565) + 1.178 12.34(4.68) 43.79(9.83) SC1 16.58(6.68) .987 59.80(17.48) 96 100 SC1-180(X*1.150) + 1.192 16.63(6.62) 59.92(17.33) SC2 29.63(7.33) .960 74.65(15.96) 79 99 SC2-180 (X*1.178) + 3.900 29.62(7.11) 74.63(15.49) SC3 30.21(6.61) .960 69.74(13.98) 84 99.8 SC3-180(X*1.076) + 5.934 30.16(6.42) 69.65(13.58) SC4 22.72(6.14) .943 50.06(13.26) 79 99.6 SC4-180(X*1.212) + 4.307 22.59(5.64) 49.79(12.18) SC5 25.21(4.43) .732 48.42(8.71) 60 94 SC5-180(X*1.164) +10.038 25.17(3.26) 48.35(6.42) SC6 13.34(4.57) .859 61.29(15.91) 45 80 SC6-180 (X*1.493) + 4.215 13.23(3.80) 60.92(13.23) SC7 22.74(10.45) .939 42.13(21.67) 53 86 SC7-180 (X*1.880) + 1.817 22.56(10.01) 41.74(20.76) SC8 25.71(12.66) .952 48.63(22.09) 60 87 SC8-180(X*1.798) + 3.827 25.63(11.95) 48.47(20.85) SC9 19.12(4.85) .869 48.06(11.34) 67 94 SC9-180-(X*1.331) + 3.827 19.14(4.25) 48.10(9.94) SC0 34.04(11.82) .877 60.53(13.37) 60 89 SC0-180 (X*2.994) + 2.769 33.76(9.97) 60.20(11.28) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" FEMALE TABLES Table 5: Means, standard deviations of raw scores, correlations, and converted T-scores for females only cross-validation sample (N=49) for Full MMPI-2 (observed) and 300-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 5.35(2.34) 58.54(11.26) 100 100 L300 (X*1.000) + .000 5.35(2.34) 58.54(11.26) F 7.90(4.45) .966 64.56(15.29) 90 98 F300 (X* 1.154) + .365 7.76(4.21) 64.09(14.46) K 14.02(5.32) .957 47.80(11.61) 90 100 K300 (X*1.249) + .623 13.80(4.89) 47.32(10.68) SC1 15.61(6.24) 54.72(15.41) 100 100 SC1-300 (X*1.000) + .000 15.61(6.24) 54.72(15.41) SC2 30.51(7.12) .998 70.87(14.33) 100 100 SC2-300 (X*1.018) - .150 30.49(7.03) 70.83(14.14) SC3 31.35(6.58) 69.59(13.90) 100 100 SC3-300 (X*1.000) + .000 31.35(6.58) 69.59(13.90) SC4 22.49(6.43) 50.60(14.22) 100 100 SC4-300 (X*1.000) + .000 22.49(6.43) 50.60(14.22) SC5 33.71(4.08) 44.54(9.99) 100 100 SC5-300 (X*1.000) + .000 33.71(4.08) 44.54(9.99) SC6 13.18(4.19) .955 59.95(14.10) 88 96 SC6-300 (X*1.115) + .519 13.08(4.26) 59.60(14.33) SC7 21.16(9.59) .973 37.07(18.91) 84 100 SC7-300 (X*1.471) - 1.216 21.27(9.81) 37.28(19.35) SC8 21.88(12.21) .993 42.59(20.58) 98 100 SC8-300 (X*1.168) - 1.381 21.81(12.34) 42.48(20.81) SC9 17.82(5.36) 47.01(12.58) 100 100 SC9-300 (X*1.000) + .000 17.82(5.36) 47.01(12.58) SC0 34.29(11.10) .938 57.81(11.73) 88 98 SC0-300 (X*1.769) - .733 33.60(10.38) 57.08(10.97) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" Table 6: Means, standard deviations of raw scores, correlations, and converted T-scores for females only cross-validation sample (N=49) for Full MMPI-2 (observed) and 250-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 5.35(2.34) .991 58.54(11.26) 100 100 L250 (X*1.022) + .033 5.33(2.15) 58.47(10.36) F 7.90(4.45) .931 64.56(15.29) 67 94 F250 (X*1.317) +.939 8.09(4.43) 65.22(15.24) K 14.02(5.32) .932 47.80(11.61) 80 98 K250 (X*1.328) + 1.333 13.66(4.73) 47.02(10.32) SC1 15.61(6.24) .997 54.72(15.41)** 100 100 SC1250 (X*1.027)+ .045 15.76(6.30) 55.10(15.56) SC2 30.51(7.12) .992 70.87(14.33) 100 100 SC2-250 (X*1.045) - .385 30.35(7.09) 70.54(14.27) SC3 31.35(6.58) .992 69.59(13.90) 100 100 SC3-250 (X*1.007)+ 1.349 31.29(6.60) 69.48(13.97) SC4 22.49(6.43) .984 50.60(14.22) 98 100 SC4-250 (X*1.146) - .654 22.50(6.48) 50.62(14.33) SC5 33.71(4.08) .951 44.54(9.99) 94 100 SC5-250 (X*1.063)+ 2.566 33.89(3.73) 44.98(9.15) SC6 13.18(4.19) .847 59.95(14.10) 51 86 SC6-250 (X*1.379)+ 3.567 13.36(3.67) 60.54(12.37) SC7 21.16(9.59) .954 37.07(18.91) 67 94 SC7-250 (X*1.721)+ .902 21.38(9.60) 37.49(18.93) SC8 21.88(12.21) .972 42.59(20.58)** 69 96 SC8-250 (X*1.540)+ .760 22.76(12.16) 44.08(20.51) SC9 17.82(5.36) .993 47.01(12.58) 100 100 SC9-250 (X*1.028)+ .062 17.75(5.17) 46.85(12.13) SC0 34.29(11.10) .878 57.81(11.73) 71 96 SC0-250 (X*2.355) - 1.722 34.04(10.33) 57.54(10.92) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" Table 7: Means, standard deviations of raw scores, correlations, and converted T-scores for females only cross-validation sample (N=49) for Full MMPI-2 (observed) and 200-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 5.35(2.34) .915 58.54(11.26) 82 100 L200 (X*1.087) + .447 5.24(2.02) 58.02(9.71) F 7.90(4.45) .923 64.56(15.29) 69 94 F200 (X*1.426) + 1.293 8.13(4.36) 65.37(14.97) K 14.02(5.32) .908 47.80(11.61) 76 100 K200 (X*1.488) + 1.544 13.87(4.54) 47.47(9.90) SC1 15.61(6.24) .990 54.72(15.41)** 96 100 SC1200 (X* 1.143) + .991 15.97(6.33) 55.60(15.63) SC2 30.51(7.12) .978 70.87(14.33) 96 100 SC2-200 (X*1.153) + 2.603 30.18(6.88) 70.20(13.84) SC3 31.35(6.58) .964 69.59(13.90) 88 100 SC3-200 (X*1.059) + 4.905 31.36(6.40) 69.62(13.52) SC4 22.49(6.43) .968 50.60(14.22) 88 98 SC4-200 (X*1.193) + 3.646 22.03(5.94) 49.58(13.15) SC5 33.71(4.08) .784 44.54(9.99) 59 90 SC5-200 (X*1.168) + 7.762 34.03(3.46) 45.32(8.47) SC6 13.18(4.19) .844 59.95(14.10) 55 82 SC6-200 (X*1.455) + 3.801 13.21(3.44) 60.05(11.58) SC7 21.16(9.59) .944 37.07(18.91) 59 92 SC7-200 (X*1.795) + 1.536 21.43(9.40) 37.59(18.55) SC8 21.88(12.21) .955 42.59(20.58)** 65 86 SC8-200 (X * 1.782) + 1.781 23.13(11.84) 44.70(19.97) SC9 17.82(5.36) .932 47.01(12.58) 84 98 SC9-200 (X* 1.198) + 3.673 18.00(4.70) 47.44(11.04) SC0 34.29(11.10) .848 57.81(11.73) 61 94 SC0-200 (X* 2.712) + 2.380 34.32(9.23) 57.84(9.76) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" Table 8: Means, standard deviations of raw scores, correlations, and converted T-scores for females only cross-validation sample (N=49) for Full MMPI-2 (observed) and 180-item short-form (predicted). Regression Mean (Std.Dev.) Pearsons Mean(Std.Dev.) Scale Equation Raw Scores r (raw) T-scores %5 T %10T L 5.35(2.34) .866 58.54(11.26) 78 98 L180 (X* 1.120) + .435 5.12(1.91) 57.46(9.17) F 7.90(4.45) .911 64.56(15.29) 65 94 F180 (X* 1.536) + 1.266 8.23(4.42) 65.69(15.20) K 14.02(5.32) .893 47.80(11.61) 74 98 K180 (X*1.577) + 1.326 13.91(4.41) 47.55(9.62) SC1 15.61(6.24) .990 54.72(15.41)** 96 100 SC1-180 (X*1.143) + .991 15.97(6.33) 55.60(15.63) SC2 30.51(7.12) .969 70.87(14.33) 90 100 SC2-180 (X*1.190) + 3.561 30.25(7.11) 70.34(14.31) SC3 31.35(6.58) .954 69.59(13.90) 84 98 SC3-180 (X*1.049) + 6.470 31.15(6.42) 69.18(13.57) SC4 22.49(6.43) .958 50.60(14.22)** 76 100 SC4-180 (X* 1.182) + 4.692 21.84(5.64) 49.16(12.47) SC5 33.71(4.08) .652 44.54(9.99) 57 84 SC5-180(X* 1.089) + 14.285 34.49(3.00) 46.44(7.35) SC6 13.18(4.19) .844 59.95(14.10) 55 82 SC6-180 (X* 1.455) + 3.801 13.21(3.44) 60.05(11.58) SC7 21.16(9.59) .936 37.07(18.91) 61 94 SC7-180 (X*1.874) + 1.918 21.38(9.39) 37.50(18.52) SC8 21.88(12.21) .941 42.59(20.58)** 57 82 SC8-180 (X *1.855) + 1.921 23.08(11.85) 44.63(19.98) SC9 17.82(5.36) .902 47.01(12.58) 65 98 SC9-180 (X* 1.256) + 3.925 17.89(4.43) 47.19(10.40) SC0 34.29(11.10) .822 57.81(11.73) 49 94 SC0-180 (X* 3.116) + 2.299 34.16(9.18) 57.67(9.70) *significance with Bonferroni correction. **significance within subjects ANOVA SC refers to "Scale" REFERENCES Ben-Porath, Y. S. & Forbey, J. D. (2003). Non-gendered norms for the MMPI-2. Minneapolis: University of Minnesota Press. Ben-Porath, Y.S., & Tellegen, A. (2008). MMPI-2-RF (Minnesota Multiphasic Personality Inventory-2 Restructured Form): Manual for administration, scoring, and interpretation. Minneapolis: University of Minnesota Press. Butcher, J. N., Dahlstrom, W. G., Graham, J. R., Tellegen, A., & Kaemmer, B. (1989). Minnesota Multiphasic Personality Inventory (MMPI-2): Manual for administration and scoring, Minneapolis, MN: University of Minnesota Press. Butcher, J. N., & Hostetler, K. (1990). Abbreviating MMPI item administration: What can be learned from the MMPI for the MMPI-2? Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2, 12-21. Butcher, J. N., & Williams, C. L. (2009). Personality assessment with the MMPI-2: Historical roots, international adaptations, and current challenges. Applied Psychology: Health and Well-Being, 2, 105-135. Dahlstrom, W. G., & Archer, R. P. (2000). A shortened version of the MMPI-2. Assessment, 7, 131-137. Friedman, A. F., Lewak, R., Nichols, D. S., & Webb, J. T. (2001). Psychological assessment with the MMPI-2. Lawrence Earlbaum Assoc.: NJ. Gass, C. S., & Gonzalez, C. (2003). MMPI-2 short-form proposal: CAUTION. Archives of Clinical Neuropsychology, 18, 521-527. Gass, C. S., & Luis, C. A. (2001). MMPI-2 short-form: Psychometric characteristics in a neuropsychological setting. Assessment, 8, 213-219. Levy, P. (1967). The correction of spurious correlation in evaluation of short-form tests. Journal of Clinical Psychology, 23, (1), 84-86.     MMPI-2 Short-Forms Page  PAGE 34 =>?`a E ` a b    ) U  + J [ q Ͻ}yyuh h!h VhOh h^5\ hb5\Hh{Fh)CmHnHuhGhxj0J\ hhxj hxj\jhxjU\hoh)C\ h)C\ hk8\h3^h8lhb\ hb\ hYhThn}GhT hYhn}G+?abcw   a b r { dgd^ $da$gd^dgd2dgd2 $da$gd2dgd2   3 ^ u  ;SY`'4*g!3:ABL_d%OJÿÿ黿ݳݳ hYh6ch {5huh6chn}GhsehNhihEhju hYhb hb5\h3^hvyhkhhg6h Vh/hbAh hYh3^<Sz> %)+&/./;/0024g6o667dgd2 $da$gd2 $dgd2 $$da$gd2 d`gd2dgd^dgd2J\hDN\^_e *fh 8?EWpRSp@A Xyz2  . : = > T ^ 2!7!8!A!b!m! ","Z"c"񻷻hYhn}G6]h$Chh|%hYhn}G>*haxh8w hYh8whjVhEhSbhvyhseh Vh/h{ hYh{h hYhn}Gh {5h"*\hTFh3^5\h,h* hoJD6]hQvhoJD6]h{hoJD hYhohlxho6>*]h)C6>*]hlxh3J6>*]hTFhaxhh'hFrhpA{ hYhoh?a%hQ~h|%h?a%6] hYh?a%h^hhh?a%6]-7J9:<<V<=,?\ABBDE7GQG\JNOPQR?S[VwVW Z&Z1]dgd2dgd2 d`gd2&:(:3:::<:G:H:O:P:]:s:::; ;;;;;;;;7<9<;<<<E<U<V<<<<<<<<==='=1=2=6=?=D=H==˿˻|݄xx݄݄hh{hoJD hYhn}Gh' hYh'h)Chr5\h)Ch)C56>*]h)Chr56>*\]hrhpA{h9S0hhE hYho hYh_Yh}Owh_Y h9WlH* h9Wl6]h9Wlh a h a6].====== >>>r>>>>>>>>?*?+?,?|??????????@@H@@@@@@@A AIAJAQA[A\AgAAAAA'B.B]BiBBBBBBѾѾѾѰh3Ihx 6>*]h8e]hh1Y-hYhl:6]hc hYhoJDh*]h)Chx 56>*\]h3Ihx 56>*\] h3Ih3I hUtdhx hA,Fhkh5ehhh{hoJD hoJDhoJDh=hx hYhx h3Ihx 6>*]hlxh)C6>*]h)C6>*]/GGG HH'H-H:H?HCHHHHHHHHdIfIIIIIIIIII4J@J[JaJgJJJJKK K!KKKKKSL}L MMdMkM+N2NNNNOOOEOeO~OOOOOûh)Ph{h|%hlxh)C6>*]h)C6>*]h3Ihx 6>*]hS?SqSSS T"T2T7TCTETʿʿݫݻh h9h~3h.PhtEh\hUtdhhsh{hoJDh|%h yhns{h y6>*]h)C6>*]hns{hUtd6>*]hns{hkhFr hYhFrhx hh)PhA6ETLT_TiTmTnTxTTTTTTTTTTTTTTTTTTTTTTUUUU U=UGUWUUUUUV V=VPVZV[VcVdVrVuVvVwVþh rh y6>*]hlxh)C6>*]h)Ch)C6>*]h)C6>*] hYh yh rhH}h y hA,FH*hA,FhA,FH*hQ~hA,F6]hA,FhUH*hU hU6]hA,Fh-0h~3hh hk3wVVVVW WW!W+W;WKWUWhWlWWWWWWWWWWWWWWWXX&X+X9XCXQXwXyXXXXXXXXXXXXXXXXX_YYYYY Z Z hoJDh=he~he~6]he~h=hA,Fh]H* h]6]h]h~3h D?h(8hshkhdinhUtdhQ~hhT h rh{hoJDh|%h y; ZZ!Z$Z%Z&Z~ZZZZZZZZZZ[[&[+[8[9[>[G[H[V[b[i[q[r[~[[[[[[[[[[[[[\\\\\\ϵϪϦϦ˖{ h/6]h1Zh/lhD:hkhObh~3h{h/h yhQ~ h1Zh/h h1ZhObh{hoJD h1Zh|%h h1Zh yh]h y6>*]hlxh)C6>*]h)C6>*]h]hD:6>*]0\\\%\&\'\6\7\8\:\K\L\W\X\e\l\m\n\p\v\\\\\\\\\\0]1]g]n]]]]]]]^^^^^^^^^_⫷ާ⣘{hTFh6 hYhn}G hTFhTF hTF5\ h $h yhkh yh~3h2hObh/lH*h/lh1Zh/l6] h/6]hObh8cH*h1Zh8c6]h1Zh8chObhObh/H*h1Zh/6]h/01]^^`bre|hjlnpssttXuuuuv@vsvvvw:wdgd2dgd2 $da$gd2 d`gd2______`bbb/ccccccccddGdNdgdnddddddeh6hk hYhdhBh\c5ttttttu*u+u,u1u2u9u:uGuTuXufusu~uuuuuuuuuuuuuvvv&v3v>v?vHvXvfvrvsvuvxphzCJaJh:sCJaJhnCJaJhKCJaJh~UCJaJh)GCJaJhzho5CJ\aJhn>*CJaJh)G>*CJaJhYho>*CJaJhkCJaJhA CJaJhYhA CJaJhYhA 5CJ\aJhYhoCJaJ,uvvvvvvvvvvvvvvvvvvvvwwwwww-w9w:w:wfwwwwxKxzxxxyRyyyyzOzPzQz{zz}{{$|p|||dgd2dgd2dgd2xxx!x%x1x?xJxKxMxNxaxnxyxzx|x~xxxxxxxxxxxxxxxxݼwl^Ph!h #5CJ\aJh!hz5CJ\aJh!hA CJaJh!h~UCJaJh!ho5CJ\aJh!hnCJaJh!h)GCJaJh!hoCJaJh!h #CJaJh!hzCJaJh~UCJaJhzho5CJ\aJhYhoCJaJh #CJaJhzCJaJh:sCJaJxx y yyyyyyy yAyFyQyRyTyVy\yly{yyyyyyyyyyyyyyyyyyy縭yqih #CJaJhzCJaJh:sCJaJh!hN67CJaJh!h<CJaJh!hnCJaJh!h)GCJaJh!h #CJaJh!hzCJaJhQ~5CJ\aJh!h<5CJ\aJh1Z5CJ\aJh!ho5CJ\aJh!hoCJaJ%yyzzzzzz-z2z@zLzMzNzOzPzdznz{z}zzzzzzzzzzzzzzz{ {\{e{~w~w~slhla hd5h:hk hYh:hz h^h^h^mH sH hwwhbmH sH  h06~hbhbmH sH h{hb hYhohQ~CJaJh1ZCJaJh<CJaJhNho5CJ\aJh #CJaJhzCJaJhnCJaJh)GCJaJhYhoCJaJ%e{o{p{}{~{{{{{{{{{{{{||| |$|@|A|E|Q|\|]|p|u|||||||||||}}+},}6}=}N}\}g}i}غخ~~h`\tCJaJh(SCJaJhYh:5CJ\aJh`\t>*CJaJhM>*CJaJhYh:>*CJaJhkCJaJhMCJaJhYhM5CJ\aJhYhMCJaJhYh:CJaJ hYh: hd5h:h0| }8}i}}} ~[~~~9oN4pqdgd2dgd2i}l}}}}}}}}}}}}} ~~ ~$~(~8~E~R~Y~[~^~j~~~~~~~~~~~~~~~~~  +79<P[gorswݿݿݷݯݷǤݷhdin5CJ\aJhl]CJaJhZCJaJh(SCJaJhYh:5CJ\aJhzCJaJhYh:CJaJh`\tCJaJhMCJaJhZGCJaJ>,9DNQRVfu߶ߠ߶|q|f[Ph!hMCJaJh!h`\tCJaJh!hl]CJaJh!hm}CJaJh!h:CJaJhPh(S5CJ\aJhz5CJ\aJhZ5CJ\aJhPh(SCJaJhPh:5CJ\aJhzCJaJhZCJaJhPh`\tCJaJhPh:CJaJhPhMCJaJhPhl]CJaJÀĀŀɀڀ +478>O\hqtƁՁƻwoaYh(SCJaJhYh:5CJ\aJhzCJaJh`\tCJaJhl]CJaJhYh:CJaJhm}CJaJh!h`\tCJaJh!hl]CJaJh!hG&CJaJh!h(SCJaJh!hMCJaJh!h:5CJ\aJh!hzCJaJh!hm}CJaJh!h:CJaJ!$%-48>CHIJPQ"T]eЃ߃ϱ|shj<>*CJaJhYh:>*CJaJhkCJaJhj<CJaJhYhj<5CJ\aJhYhj<CJaJh)Ch:mH sH h hd5h:hkhN hYh: h^h^h^mH sH  h06~hbhbmH sH h{hbhg-5V>p؄ DۅPņ=x [ֈdgd2dgd2dgd2߃ '28@R^iov˄Մքބ 378CDHM^aqƅDž҅؅ۅͿͷͿͯͷͿͯͯͧ͟͟Ϳͯ̈́ͷhm}CJaJhYh!CJaJhCJaJhzCJaJh=_CJaJh!CJaJhj<CJaJhYh:5CJ\aJhYh:CJaJhYh:>*CJaJhj<>*CJaJh=_>*CJaJh!>*CJaJ4ۅޅ߅.;<GMPSTYjxÆņɆΆ߆"-23:=@ADUWYfqx忷Ͽ׿Ͽ忷Ϥ寜hdinCJaJh CJaJhYh!CJaJhZCJaJh!CJaJhj<CJaJh:CJaJh=_CJaJhYh:5CJ\aJhYh:CJaJhzCJaJhm}CJaJ7x{}ȇև  #$):HT[]^arÈꢔ갆{pp{h!hCJaJh!hm}CJaJh!h=_5CJ\aJh!h 5CJ\aJh!hZ5CJ\aJh!h:5CJ\aJh!h CJaJh!h=_CJaJh!h!CJaJh!hj<CJaJh!h:CJaJh!hZCJaJ,Èψֈو '6CJLORcq{‰Éĉʼn߾ꨝ}ogc_hghNhdinCJaJhYh:5CJ\aJh CJaJh=_CJaJh!CJaJhj<CJaJhYh:CJaJhm}CJaJh!h:5CJ\aJh!h CJaJh!h!CJaJh!hj<CJaJh!hG&CJaJh!h:CJaJh!h=_CJaJ#ֈ KLÉĉʼn6X̌B2kݎdgd2dgd2ʼn؉ #()*1yي4=EijkpqxyϼwhYh:>*CJaJh|CJaJhkCJaJhUi:CJaJh 2+CJaJhYh:5CJ\aJhYh:CJaJh)Ch:mH sH hhk hYh:hN h^h^h^mH sH  h06~hbhbmH sH h{hb+ˋ̋͋7DOV]n{Ìʌь!.:BEFIZ_jxƻ}hm}CJaJh CJaJhzCJaJh|CJaJh=CJaJh 2+CJaJhYh:5CJ\aJhYh:CJaJhYh:>*CJaJhs_>*CJaJh|>*CJaJhUi:>*CJaJh=>*CJaJh 2+>*CJaJ1Íčȍٍڍ#$02569JXdjknŽЎ܎ݎ whNCJaJh 1CJaJhZCJaJhgkCJaJh`1CJaJhzCJaJhFuCJaJh_.CJaJh{CJaJhYh:5CJ\aJh CJaJhm}CJaJh|CJaJh=CJaJh 2+CJaJhYh:CJaJ. /=HMNORTlyȏˏ }odVh!hZ5CJ\aJh!hNCJaJh!h:5CJ\aJh!h CJaJh!h`1CJaJh!h 1CJaJh!h 2+CJaJh!h:CJaJh!hZCJaJhNCJaJh 1CJaJhYh:5CJ\aJh CJaJhZCJaJhYh:CJaJh`1CJaJ!OȏB~34rۑ$1Ó/iϔdgd2dgd2dgd2  /;@ABDEF_kqr|֐ +124ٲ٧ّٲٜ{ٜ{نٜph!hNCJaJh!hG&CJaJh!h CJaJh!h`1CJaJh!h 1CJaJh!h 2+CJaJh!hm}CJaJh!hN5CJ\aJh!h 15CJ\aJh!h:CJaJh!h:5CJ\aJh!h 5CJ\aJ*478P]horuvzđΑۑݑ #$+szyrnjfhkh9hs h!h KMh!h95\ h^h^h^mH sH hwwhbmH sH  h06~hbhbmH sH h{hbh} hYh`<hYh:5CJ\aJh CJaJh`1CJaJh 1CJaJh 2+CJaJhYh:CJaJhm}CJaJ(zԒޒ/8@defklstÓɓٓ'.5FVbh}ƽƢ|tl|thl*CJaJhDvCJaJh^(}CJaJh#CJaJh)h95CJ\aJh9CJaJh9>*CJaJh >*CJaJhl*>*CJaJhDv>*CJaJh#>*CJaJhkCJaJh~sCJaJhYh~s5CJ\aJhYh~sCJaJhh9*”ΔϔҔ '4;<?R_jqrv|͕ؕەߕ '5@HILMQam{ʺʺhZCJaJhm}CJaJh^CJaJhzCJaJh^(}CJaJh)h95CJ\aJh#CJaJh9CJaJhl*CJaJAϔ<rەI(bԗ L234^adgd2dgd2͖ٖ%(+,02@ALMZbeixʼܰܤܤ~rcܰrrh^h"!CJaJmH sH h"!CJaJmH sH h!h95CJ\aJhl*CJaJmH sH h^CJaJmH sH h#CJaJmH sH h^(}CJaJmH sH h)h95CJ\aJ"h)h95CJ\aJmH sH h9CJaJmH sH hm}CJaJmH sH hZCJaJmH sH #͗ԗ֗חܗ Ĺth\hPD8DPh\hl*CJaJmH sH h^CJaJmH sH h"!CJaJmH sH h3CJaJmH sH h9CJaJmH sH h^(}5CJ\aJmH sH (h"!h956CJ\]aJmH sH "h^h"!5CJ\aJmH sH h"!5CJ\aJmH sH h!h9CJaJh)h9CJaJmH sH h95CJ\aJmH sH hm}5CJ\aJmH sH hZ5CJ\aJmH sH ')+5CKLO]kvxӘ*12Ŷznnzn`h)h95CJ\aJh3CJaJmH sH hl*CJaJmH sH h^CJaJmH sH hG&CJaJmH sH h#CJaJmH sH h^(}CJaJmH sH h^h"!CJaJmH sH h^h9CJaJmH sH h!h95CJ\aJ"h)h95CJ\aJmH sH h9CJaJmH sH %23GQ^`lmst|ISacmrٚښۚ'2<ÿ|shU{>*CJaJh >*CJaJh9>*CJaJhkCJaJh yLCJaJhYh yL5CJ\aJhYh yLCJaJhhkh9hs h^h^h^mH sH hwwhbmH sH  h06~hbhbmH sH h{hbhbCJaJ,<uߛL0fםG%_˟56dgd2dgd2dgd2<CS^lʛכޛ *7DLOboq|Ü֜ '03FS^fjn~Ýϝ֝םڝ۝ߝ ²h3CJaJh CJaJhzCJaJh<)UCJaJhU{CJaJh CJaJhFCJaJh)h95CJ\aJh9CJaJD'3?GJKO_kxўݞ "#%()ȼԒȼԼzk\h35CJ\aJmH sH hZ5CJ\aJmH sH h CJaJmH sH hFCJaJmH sH h<)UCJaJmH sH hU{CJaJmH sH "h)h95CJ\aJmH sH h9CJaJmH sH h3CJaJmH sH hZCJaJmH sH hU{CJaJhFCJaJh9CJaJhZCJaJ")-<HV_abrvß˟Ο(469MZefmnǻǯmaamhG&CJaJmH sH "h)h95CJ\aJmH sH h9CJaJmH sH hU{CJaJmH sH h CJaJmH sH hFCJaJmH sH h CJaJmH sH h3CJaJmH sH h<)UCJaJmH sH hU{5CJ\aJmH sH h)h9CJaJmH sH h95CJ\aJmH sH &6n٠ܡ!I~"Y̤<pdgd2dgd2nrv ̠٠۠ cjġΡܡޡʾ趲}y}u}j\jhYh yL5CJ\aJhYh yLCJaJhhkh9hs h^h^h^mH sH hwwhbmH sH  h06~hbhbmH sH h{hbh9mH sH hU{CJaJmH sH h<)UCJaJmH sH "h)h95CJ\aJmH sH h9CJaJmH sH h3CJaJmH sH #(0TUV[\cdq~ˢ֢(4A]jṿ٣|tl|h"!CJaJhzCJaJh#CJaJhJZ;h95CJ\aJh_$CJaJh<*CJaJh<*CJaJh_$>*CJaJh9hkCJaJhYh yL5CJ\aJhYh yLCJaJh yLCJaJ*٣ۣ !"%8EQY]ars~Ĥ̤ФԤ +6<?@DUcopsʤଜشଜhZCJaJh%CJaJh yLCJaJh#CJaJh3CJaJh"!h"!6CJ]aJhJZ;h95CJ\aJhzCJaJh9CJaJh"!CJaJh<*CJaJh>*CJaJh<)U>*CJaJhK>*CJaJhkCJaJhKCJaJhYhK5CJ\aJhYhKCJaJhhkh KMhshbCJaJ h^h^h^mH sH hwwhbmH sH $2JbozĪͪڪ%&)<=IKW^_bcgwxǫϫӫث -8@DHYgsz}ͬڬºʲºʲhgCJaJh?CJaJhDIXCJaJhzCJaJh~sCJaJhJZ;h KM5CJ\aJhCJaJh KMCJaJhKCJaJFϫ @z$_ҭB{*+bȯ $da$gd^dgd^dgd2dgd2ڬ$'(+-?KW^_buʭѭҭխȼԼȼvk`kԼh!hDIXCJaJh!h KMCJaJh!h KM5CJ\aJhCJaJmH sH hDIXCJaJmH sH hJZ;h KM5CJ\aJ"hJZ;h KM5CJ\aJmH sH h KMCJaJmH sH hgCJaJmH sH hZCJaJmH sH hCJaJhZCJaJh KMCJaJhDIXCJaJ$  -:ABDEYZfhs}ĹӑymaUC"h"!h KM5CJ\aJmH sH hCJaJmH sH h?CJaJmH sH h yLCJaJmH sH hDIXCJaJmH sH hgCJaJmH sH h KMCJaJmH sH h!hDIX5CJ\aJh!h KM5CJ\aJh!h KMCJaJh"!h KMCJaJmH sH h KM5CJ\aJmH sH hg5CJ\aJmH sH hZ5CJ\aJmH sH ήۮ")+.BOZbfjl|ƺҮޜޮƺޮޜumemVRh KMh]h KMCJaJmH sH h yLCJaJh KMCJaJh"!h KM5CJ\aJh!h~sCJaJh!h KM5CJ\aJ"h"!h KM5CJ\aJmH sH hgCJaJmH sH hCJaJmH sH h yLCJaJmH sH hG&CJaJmH sH h KMCJaJmH sH h!h KMCJaJh!h yLCJaJȯʯ֯ׯݯޯ;]_z9fgJKβĺ{{wskhuhu6huh h{h h{hbhb hz hz huhz 6hz hx-hH hX 6]hX hThl:hghb5\ hb5\ h^h^h^mH sH hwwh?mH sH  h06~h?h?mH sH h{h?h9(gK#[gh &`gdk8 $&`a$gdk80d^`0gd2"#CPQ[ȴ24\]µߵKLTvxһݨ夠~~~~zh:shjhU hYhS hHhchHhc6hG&hc hHhl:hHh{h h}P1h}P1hHhH6hHh}P16h}P1 h{hl:hHhl:6 hYhl: hu6huhb huhu0 hYhShh:sh^0JmHnHu h:s0Jjh:s0JU0d^`0gd2<P1F:p2/ =!"#$% Dp^ 2 0@P`p2( 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p 0@P`p8XV~_HmH nH sH tH J`J Normal 7$8$H$CJ_HaJmH sH tH 8@8  Heading 1$@&>*:@:  Heading 2$@&5\DA`D Default Paragraph FontRiR  Table Normal4 l4a (k (No List 2>@2 Title$a$5\NZN o Plain Text 7$8$H$CJOJQJ^JaJ4@4 THeader  !.)@!. T Page Number4 24 TFooter  !ZAZ\)CInformation ServicesCJOJQJ^JaJphF[@RF )CE-mail Signature 7$8$H$6U@a6 )C Hyperlink >*B*phPK![Content_Types].xmlj0Eжr(΢Iw},-j4 wP-t#bΙ{UTU^hd}㨫)*1P' ^W0)T9<l#$yi};~@(Hu* Dנz/0ǰ $ X3aZ,D0j~3߶b~i>3\`?/[G\!-Rk.sԻ..a濭?PK!֧6 _rels/.relsj0 }Q%v/C/}(h"O = C?hv=Ʌ%[xp{۵_Pѣ<1H0ORBdJE4b$q_6LR7`0̞O,En7Lib/SeеPK!kytheme/theme/themeManager.xml M @}w7c(EbˮCAǠҟ7՛K Y, e.|,H,lxɴIsQ}#Ր ֵ+!,^$j=GW)E+& 8PK!Ptheme/theme/theme1.xmlYOo6w toc'vuر-MniP@I}úama[إ4:lЯGRX^6؊>$ !)O^rC$y@/yH*񄴽)޵߻UDb`}"qۋJחX^)I`nEp)liV[]1M<OP6r=zgbIguSebORD۫qu gZo~ٺlAplxpT0+[}`jzAV2Fi@qv֬5\|ʜ̭NleXdsjcs7f W+Ն7`g ȘJj|h(KD- dXiJ؇(x$( :;˹! I_TS 1?E??ZBΪmU/?~xY'y5g&΋/ɋ>GMGeD3Vq%'#q$8K)fw9:ĵ x}rxwr:\TZaG*y8IjbRc|XŻǿI u3KGnD1NIBs RuK>V.EL+M2#'fi ~V vl{u8zH *:(W☕ ~JTe\O*tHGHY}KNP*ݾ˦TѼ9/#A7qZ$*c?qUnwN%Oi4 =3ڗP 1Pm \\9Mؓ2aD];Yt\[x]}Wr|]g- eW )6-rCSj id DЇAΜIqbJ#x꺃 6k#ASh&ʌt(Q%p%m&]caSl=X\P1Mh9MVdDAaVB[݈fJíP|8 քAV^f Hn- "d>znNJ ة>b&2vKyϼD:,AGm\nziÙ.uχYC6OMf3or$5NHT[XF64T,ќM0E)`#5XY`פ;%1U٥m;R>QD DcpU'&LE/pm%]8firS4d 7y\`JnίI R3U~7+׸#m qBiDi*L69mY&iHE=(K&N!V.KeLDĕ{D vEꦚdeNƟe(MN9ߜR6&3(a/DUz<{ˊYȳV)9Z[4^n5!J?Q3eBoCM m<.vpIYfZY_p[=al-Y}Nc͙ŋ4vfavl'SA8|*u{-ߟ0%M07%<ҍPK! ѐ'theme/theme/_rels/themeManager.xml.relsM 0wooӺ&݈Э5 6?$Q ,.aic21h:qm@RN;d`o7gK(M&$R(.1r'JЊT8V"AȻHu}|$b{P8g/]QAsم(#L[PK-![Content_Types].xmlPK-!֧6 +_rels/.relsPK-!kytheme/theme/themeManager.xmlPK-!Ptheme/theme/theme1.xmlPK-! ѐ' theme/theme/_rels/themeManager.xml.relsPK] R  222225 Jc"}-57&:=BGOETwV Z\_ituvxxye{i}߃ۅxÈʼn 4z2<)n٣he2ڬ\^`abcdfghijklmnpqrsuvwxz{|}71]:w|ֈϔ6ϫ]_eoty~ XX$+.5!8@0(  B S  ?lqӬ۬ΎԎAGhnڏ)/RXȐWYɖ˖57ekїח<Bǝ39Z`̞Ҟ!<B?Eekإޥ!'HNĦ 17}33333333333333333333333333333333333333333333//rrrrrr4zIz)fz zz__ҥҥBBggKK##[[ @ @)P/lvlx{&/Tr2 #x  z V5 o X #r~ g*/knuYDdn rkD:p_cH}V3^OQ :WZGPux "!K"#Q#nj# $K%?a%G&';l'(a( 2+;=,-1Y-x-_.9S0 1}P1`12D2,K22x 3~3d5 {5167N67k8@g9t:Ui:3;JZ;wl;<%> D?s?sN@F*A<Exjnl*joRv,E2 aT 8c "S)9NAikIKv]DWblK89q`<seJR!bA()j<T]8(8z{sH!jVd}ByHr5|%g6:|)bS.Ui6d@Zki2) Pc-?^ Vb t yrB8yz b;IFrQ~;'^&1|ekR~"g38cDU{ q-^4 _$j=os@h@UnknownInformation Services&Information Services20051115T144018093{FInformation ServicesG* Times New Roman5Symbol3. * Arial?= * Courier NewA BCambria Math"AhFAF_DF.R2"R2"!xx23Q? 2!xx/The original database consisted of 2468 recordsJohn D. DingellAxelrodOh+'0( 8D d p | 0The original database consisted of 2468 recordsJohn D. Dingell Normal.dotmAxelrod5Microsoft Office Word@m@*!@+ @VBv"R2՜.+,D՜.+,l( hp  VA Medical Center 0The original database consisted of 2468 records Title 8@ _PID_HLINKSA *Qmailto:Bradley.axelrod@va.gov9tmailto:Bradley  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~Root Entry F҉1Table)WordDocument>RSummaryInformation(DocumentSummaryInformation8CompObjy  F'Microsoft Office Word 97-2003 Document MSWordDocWord.Document.89q