Examination for Professional Practice in Psychology

Examination for Professional Practice in Psychology (EPPP) taken by professional (Vail Model) graduates had a failure rate of 30.82% in contrast to 7.60% for EPPP licensing exams taken by traditional (Boulder Model) graduates of clinical psychology programs. Thus, exams taken by professional graduates were 4.06 more likely to result in failure. It was acknowledged that it is not known whether EPPP performance is related to psychotherapy outcome.

It has now been more than a decade since the correlates of performance on the national licensing exam, the Examination for Professional Practice in Psychology (EPPP) of clinical psychology graduates was reported (Templer & Tomeo, 1998b;Yu et al., 1997). In that research with 45 professional school (Vail Model) programs and 83 traditional (Boulder Model) programs, EPPP scores correlated .78 with 1988-1995 median Verbal + Quantitative + Analytical Graduate Record Examination (GRE) scores, .65 with minimum Verbal + Quantitative + Analytical GRE, .56 with American Psychological Association approval, −.49 with ratio of fulltime students to core faculty, −.46 with number of doctorates awarded, .46 with minimum Analytical GRE, and −.44 with number of full-time students enrolled. The findings that were most controversial and most professionally relevant involved the high EPPP scores of the traditional program graduates, which were especially obvious at both ends of the distribution. The top of the distribution was dominated by traditional programs and the bottom of the distribution was dominated by professional programs. The traditional programs were superior not only on the EPPP total score and the Research subtests but also on the Diagnostic, Intervention, and Professional/Ethics subtests . Subsequent research demonstrated that a salient difference still existed for EPPP scores 1997-2005 (Templer, Stroup, Mancuso, & Tangen, 2008).
The findings are especially noteworthy to professional schools because they purport to produce a superior professional product. Independent professional practice is not possible without passing the EPPP. Unfortunately for the image of professional schools, their graduates tend not to do well on other indicators of professional functioning. They are less likely to be a diplomate in the American Board of Professional Psychology (ABPP), less likely to be president of state psychological associations, and less likely to be directors of the Association of Psychology Postdoctoral and Internship Centers (APPIC; . The generally less satisfactory performance of professional school graduates has led to the inference that there are two tiers of clinical training in the United States (Templer, 2005b).
The present study went beyond the previous EPPP research comparing professional (Vail Model) and traditional (Boulder Model) clinical psychology graduates in two ways. (a) The Educational and Reporting Service had reported mean EPPP scores of programs and such was used in the previous research. The Educational and Reporting Service currently provides number and percentage of exams taken and passed for each program. Kenkel, DeLeon, Albino, and Porter (2003) suggested that pass rate may be more important than mean score. This appears to be a reasonable perspective. The psychologist with a perfect EPPP score receives the same license as the one who barely passes. (b) The previous research on the EPPP compared mean scores of professional and traditional graduates has actually attenuated the true differences because professional school means are based on a much larger N. The larger N is a function of both professional schools graduating more students per program and because of having to take the EPPP again. In the present study, we not only report pass/fail mean program rates but also consider the number of exam taken and passed/failed by the composite of professional programs and traditional programs.

Method
The EPPP pass rate information was obtained from the Association of State and Professional Provincial Boards (ASPPB; 2010) for 2005-2009 for 217 clinical psychology programs: 154 traditional programs (140 American and 14 Canadian) and 63 professional (all American) programs. ASPPB only included programs that had three or more exams. An additional caveat is that pass is based on a score of 500, which is the passing score of the preponderance of states and provinces. If one wishes to determine the exact pass rate of a jurisdiction, one should communicate with that jurisdiction. It should be borne in mind that N pertains to number of exams taken and not necessarily the number of different graduates having taken the exam. That is, the number of exams taken includes an unknown number of retakes. The Alliant/ California School of Professional Psychology (CSPP)-Berkeley 11 exams were combined with the 268 Alliant/ CSPP-Alameda graduates because there was a change in location but not program. Only straight clinical psychology programs were included. Clinical-neuropsychology, clinicalhealth, clinical-ethnic minority, clinical-family child, clinical psychodynamic, combined clinical/counseling/school, clinicalcommunity, child-clinical, clinical school, combined professional scientific, clinical psychology-health and medical, clinical psychology-law, physiological-developmental clinical, and adult-clinical were excluded. There were 31 programs excluded due to being combined with other areas in contrast to 217 solely clinical programs whose data were analyzed. It should be further noted that the American Psychological Association accredits not only clinical psychology programs but also counseling and school programs. The latter two categories were not used because virtually all of the professional schools have clinical psychology programs, although some may have counseling and school elements.

Results
The readers should bear I mind that the past pass/failure rates pertain to the exams taken and not number of graduates. It is not known whether a failure is for the 1st time the graduate took the test or the 20th time. It is conceivable that a small percentage of professional graduates experience multiple failures but the overwhelming preponderance of professional graduates pass the EPPP on the first attempt. Table 1 contains the number of exams taken and pass rates for the 63 professional (Vail Model) and 154 traditional (Boulder Model) programs. Tied programs are listed in alphabetical order. Professional programs have a "P" in front of them and are capitalized. Traditional programs have a "T" in front of them.
Nine (14.30%) of the 63 professional programs and 128 (83.02%) of the 154 traditional programs had mean pass rates higher than the overall mean pass rate of 86.97%, χ 2 (1, N = 217) = 163.42, p < .001. The 63 professional schools had a mean pass rate of 73.41% (SD = 14.02) and the traditional programs had a mean pass rate of 92.77 (SD = 8.49), t(df = 84.77) = 10.35, p < .001. As stated above, however, such analyses mask the true differences between professional and traditional clinical psychology programs because most professional schools have many more EPPP exams taken and because these data include retakes. In the present study, in one state, there were six professional programs that each had more than 200 exams taken for a total of 1,775 exams, and 690 exams failed. This is over twice that of the 154 traditional programs combined.
Of the 3,447 EPPP exams taken by traditional clinical psychology graduates, 3,185 (92.20%) were passed and of the 7,202 exams taken by professional clinical psychology graduates, 4,946 (69.20%) were passed, χ 2 (1, df = 10,647) = 5,321.53, p < .001. Failure rate on the EPPP of 30.82% was 4.06 greater than that of the traditional graduates' failure rate of 7.60%.

Discussion
The vastly superior quality of the students of the traditional programs must certainly contribute to the higher EPPP pass rate of the traditional programs. Traditional programs accept 10% and professional schools 44% of applicants (Templer, 2005a(Templer, , 2005bTempler & Bedi, 2003). The mean grade point average of traditional programs is two standard deviations higher (Templer, 2005b). Their mean GRE scores are more than two standard deviations higher (Templer & Arikawa, 2004). The GRE difference may be especially important because it correlated substantially with Wechsler Adult Intelligence Scale-Revised IQ (Carvajal & Pauls, 1995). This is especially unfortunate in view of the current demand for evidence-based assessment and treatment (e.g., Levant, 2004). Less talented people are less capable of examining scientific evidence.
It should be noted that not all professional schools have high failure rates. Five professional programs have a pass rate of more than 95%. For example, the Rutgers professional program mean pass rate of 95.80% is almost identical to that of the Rutgers traditional program rate of 95.50%. The  Rutgers professional school mean pass rate of 95.80% is almost identical to that of the Rutgers traditional program rate of 95.50%. The Rutgers professional program acceptance rate of 6% compares most favorably with the general professional school rate of 44% and is even lower than the traditional acceptance rate of 10%. Templer and Tomeo (1998a) warned Canadian psychologists, "Don't let it happen in Canada." They said that if Canada starts professional schools, they should "be located in very strong universities and they should have the most elite admission standards possible" (p. 4). Rutgers appears to fit such a description better than most professional schools. Unfortunately, most professional programs are free-standing or in smaller universities. They must accept less than optimal applicants for financial survival. Forty-five percent (N = 1,009) of the total of 2,220 professional schools' failed exams can be accounted for by 8 of the 65 professional programs. A fundamental problem, in view of the huge differences in selection ratio and GRE score, is that there is simply insufficient, very high-level talent to go around. It would appear to be difficult to find creative solutions for this deficiency without reducing the number of students admitted or the phasing out of some of the weaker programs.
A troubling situation is that the National Council of Schools of Professional Psychology (NCSPP) lists 19 associate members who do not meet the criteria for full membership. The EPPP information on these programs is not provided by the ASPPB and therefore is not contained in Table 1. The NCSPP also lists four programs that are in the process of developing doctoral programs in clinical psychology. All 23 of these programs appear to be free-standing or in nonprominent universities. The prognosis for psychology's professional school problem does not appear to be remarkably good.
A final caveat is that the EPPP assesses mastery of academic knowledge but may or may not be related to the very important therapeutic ingredient of client-clinician relationship. An even more fundamental gap in our knowledge is that we do not know if there is a relationship between EPPP performance and client outcome. Research on this issue is recommended.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research and/or authorship of this article.