MEASUREMENT RESEARCH ASSOCIATES
TEST INSIGHTS
April 2009
Greetings
 
The re-use of protocols in oral examinations seems to be a constant concern.  However, the data presented in this brief article suggests that repeating candidates do not appear to have an advantage or to perform better if they challenge a protocol a second time.
 
Mary E. Lunz, Ph.D.

Executive Director

Repeating Candidate Performance on Re-Used Protocols

Oral certification examinations are generally more stable if protocols are repeated at logical intervals.  However, there is a concern that candidates will share the content of the protocols with their colleagues, thus giving subsequent candidates an advantage. The data have shown that the difficulty of the protocols does not change if the same protocol is used in more than one examination administration.  This report discusses candidate performance on overlapping protocols that appear on more than one examination administration

The repeating candidates are most likely to have the opportunity to challenge the same protocols if the protocols are used on more than one exam.  Therefore, one might predict that repeating candidates should have an advantage when challenging the previously used protocols.  For purposes of this study, the candidates were divided into two groups, first time candidates (n = 387) and repeating candidates (n = 63).  The performance of the two candidate groups on the overlapping protocols was compared using ANOVA and statistically significant differences were found in the performance of the repeating candidates and the first time candidates on all previously used protocols.  However, the descriptive statistics show clearly that the repeating candidates performed significantly less well on the overlapping protocols. This information appears in the Table.


Overlapping Protocols

Candidate Group

Mean Score

Std. Deviation

Minimum Score

Maximum Score

Statistical Significance

 

 

 

 

 

 

 

Protocol 1

First

6.19

1.27

2.52

10.39

 

 

Repeat

4.88

1.59

.25

8.27

 

 

Total

6.00

1.39

.25

10.39

P< .01

 

 

 

 

 

 

 

Protocol 2

First

6.29

1.27

2.51

11.13

 

 

Repeat

5.12

1.25

2.27

7.91

 

 

Total

6.12

1.33

2.27

11.13

P <.01

 

 

 

 

 

 

 

Protocol 3

First

6.32

1.54

1.21

10.81

 

 

Repeat

5.49

1.62

2.66

11.04

 

 

Total

6.20

1.58

1.21

11.04

P< .01


These findings suggest that even if candidates do share information with their colleagues, or if repeating candidates challenge the same protocols more than once, repeating candidates do not seem to benefit from the exposure.  All candidates must be able to problem solve and demonstrate appropriate clinical judgment, as well as, clinical knowledge and skill to perform at satisfactory levels on the protocol.  Thus re-using protocols does not appear to provide repeating candidates an advantage.  It appears that similar judgmental errors may have occurred more than once.











Measurement Research Associates, Inc.
505 North Lake Shore Dr., Suite 1304
Chicago, IL  60611
Phone: (312) 822-9648     Fax: (312) 822-9650



Coming Rasch-related Events
June 23 - July 21, 2023, Fri.-Fri. On-line workshop: Practical Rasch Measurement - Further Topics (E. Smith, Winsteps), www.statistics.com
Aug. 11 - Sept. 8, 2023, Fri.-Fri. On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets), www.statistics.com