Table 4. Previous studies from the literature investigating correlations between OSCE and other academic assessment methods.
Reference | Country | Students, No. | Academic assessment method compared to OSCE | Statistical evaluation of the inter-relationship | Conclusions |
---|---|---|---|---|---|
Smith et al (1984) | United Kingdom | 229 | Viva voce examination, in-case assessment (clinical aptitude and written project), MCQ examination, comparable traditional clinical examinations. | Significant correlation between OSCE and marks from MCQ (r = 0.34, P<0.001), comparable clinical examination (r = 0.26, P<0.001), and previous in-case assessment (r = 0.24, P<0.001). | In contrast to viva-voce examination, OSCE results correlated well with an overall assessment of the student’s ability. |
No correlation between OSCE and viva voce examination (r = 0.08, P>0.05). | The clinical component of OSCE did not correlate well with MCQ. | ||||
Probert et al (2003) | United Kingdom | 30 | Long and short case-based viva voce examinations. | Overall performance at traditional finals was correlated with the total OSCE mark (r = 0.5, P = 0.005). Dichotomizing traditional examinations into surgery and medicine assessment resulted in significant correlations between OSCE and surgery marks (r = 0.71, P<0.001) but not between OSCE and medicine marks (r = 0.15, P = 0.4). | This was a pilot study for OSCE implementation, and the analyzed sample of students who performed both examination methods was representative of the whole population. |
The authors added independent consultant evaluations to assess clinical performance by students. | OSCE assesses different clinical domains than do traditional finals and improved the prediction of future clinical performance. | ||||
Dennehy et al (2008) | USA | 62 | National Board Dental Examination (NBDE, computerized assessments of theoretical knowledge in part I, and clinical knowledge in part II), and MCQ examinations. | NBDE score was statistically associated with OSCE score (P ranging from <0.001 to 0.04). | Didactic predictors (NBDE, MCQ examinations) explained around 20% of the variability in OSCE scores. |
There was no significant association between OSCE and MCQ scores. | OSCE may be a tool that allows educators to assess student capabilities that are not evaluated in typical standardized written examinations. | ||||
In multiple regression models none of the didactic predictors were significantly associated with overall OSCE performance. | |||||
Sandoval et al (2010) | Chile | 697 | Written examination and daily clinical practice observation guidelines. | Positive correlation between percentages of success for all three evaluation methods with OSCE (P<0.001). | Pearson’s correlation co-efficient was higher between assessment methods after seven years of OSCE implementation. |
These evaluations are complementary. | |||||
Kirton et al (2011) | United Kingdom | 39 per year during a 3-year long evaluation | Medicine and pharmacy practice (MPP) written examination combining MCQ and essays expected to relate to clinical practice. | Moderate positive correlation between OSCE and MPP (r = 0.6, P<0.001). | For 20% of students, experience in OSCE did not increase marks or individual performance. |
These two examinations assess different areas of expertise according to Miller’s Pyramid of Competence and both should be performed. | |||||
Kamarudin et al (2012) | Malaysia | 152 | Student’s clinical competence component evaluated during the final professional long-case examination. | Positive correlation between OSCE and long case for the diagnostic ability (r = 0.165, P = 0.042) and total exam score (r = 0.168, P = 0.039). | There is a weak correlation between OSCE and long-case evaluation format. These two assessment methods test different clinical competences. |
Tijani et al (2017) | Nigeria | 612 | Long-case examination (end of posting in the 4th and 6th years of medical school), final MCQ, and total written papers (TWP): sum of MCQ examinations and essays. | Positive correlation between OSCE and MCQ (r = 0.408), TWP (r = 0.523), and long case (r = 0.374), P<0.001. | The total clinical score combining OSCE and long-case marks was a better predictor of student clinical performance than each assessment method analyzed separately. |
These two evaluations could be complementary. | |||||
Previous experience with OSCE was not taken into account in the analysis. | |||||
Lacy et al (2019) | Mexico | 83 | Communication skills evaluated during direct observation of a clinical encounter (DOCE) using the New Mexico Clinical Communication Scale. | Students’ matched scores on OSCE and DOCE were not correlated. | The discordance between OSCE and DOCE suggests that OSCE may not be an optimal method to identify students requiring additional communication training. |
Mean scores were not statistically different between faculty evaluators for individual communication skills (P = 0.2). |
No. = number; OSCE = objective structured clinical examination; MCQ = multiple-choice question; NBD = national board dental examination; MPP = medicine and pharmacy practice; TWP = total written papers; DOCE = direct observation of a clinical encounter.