Magazine article National Association of School Psychologists. Communique

Flawed Analyses and Old Data Are Misleading/Response to Erchul, Schulte, and Sabourin Ward regarding EPPP Scores

Magazine article National Association of School Psychologists. Communique

Flawed Analyses and Old Data Are Misleading/Response to Erchul, Schulte, and Sabourin Ward regarding EPPP Scores

Article excerpt

We are writing to register our concern over the article entitled, "School Psychologists' Performance on the Examination for the Professional Practice in Psychology (EPPP)"by Joel Kupfersmid [Communiqué, 39(8)]. We maintain the reported data analysis is (a) inappropriate and (b) based on outdated EPPP passing score data. Consequently, the article provides inaccurate information about doctoral training programs and has the potential to mislead applicants who may use this information to select a program.

There are several problems with the data analysis. First, Kupfersmid (2011) did not use actual passingrates from programs, but rather estimated them using each program's reported mean and standard deviation and then calculating the area under the normal curve that fell at or above a universal passing score (selected based on its wide use across states) for each program. In addition to the concern that Table 1 portrays estimated passingrates as actual passingrates, the analysis provides correct data only if one presumes that each of the small samples from individual programs was normally distributed. This assumption is unlikely, and does not appear to have been directly tested. Score outliers or any type of nonnormal distribution would distort the estimated passing rates, a situation that is quite likely with small sample sizes.

Kupfersmid (2011) also used EPPP score data compiled from January 1997 to July 2006. Among other things, this time frame suggests that data were included from graduates who began training 20 or more years ago. Because individual training programs change significantly over time in terms of faculty, emphases, resources, etc., this outdated information could be misleading to potential doctoral program applicants.

The question then becomes, "Are there more recent data available?" Yes. The Association of State and Provincial Psychology Boards (ASPBB) issued a report in 2010 that covers the period from January 2005 to December 2009 (downloadable at: .cfm?pageid=357i). Moreover, this report lists actual (i.e., not estimated) passing rates by training program; there is no reason to transform scores as was done in the original article.

An updated analysis is therefore in order. Using the same doctoral training programs that Kupfersmid (2011) analyzed, Table 1 below is the alphabetical listing within groupings of actual EPPP passing rates for more recent graduates using data from the 2010 ASPPB report. (Note that the University of Massachusetts [Amherst and Boston], included in the original analysis, is not mentioned in the 2010 ASPBB report and therefore is excluded here.)

Again, these results reflect actual passing rates for individual programs based on data compiled since 2005. They are not estimates based on older data or disputable data transformations/analyses.

In comparing this table to Kupfersmid's Table 1, several results standout. First, a majority (77%) of the programs demonstrated higher EPPP passing rates than those described originally. As one illustration, Ball State and Indiana State went from the 50%-59% group to the 90% or higher group. Second, none of the 39 programs fell into the less than 50% group, a position held by five programs in the original analysis.

A major implication of the current analysis is that doctoral training programs in school psychology are in general doing better in preparing graduates to pass the EPPP than indicated by Kupfersmid (2011). Some programs are doing considerably better.

In conclusion, about half of school psychologists with doctoral degrees choose to take the EPPP. It is not for everyone. Passingrates on this exam, however, have the power to reflect well (or poorly) on a training program and its graduates. It is, therefore, critical to disseminate current and accurate information on EPPP passing rates to the professional community and public at large. It is in that spirit this response was written. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.