Have a personal or library account? Click to login
Ensuring the quality of multiple-choice exams administered to small cohorts: A cautionary tale Cover

Ensuring the quality of multiple-choice exams administered to small cohorts: A cautionary tale

Open Access
|Jan 2017

Figures & Tables

Fig. 1

Variance in difficulty coefficients (Panel A) and discrimination coefficients (Panel B) across repeat use of 32 MCQ items across 22 small student clerkship cohorts (n = 10–15 students). Error bars represent range

40037_2016_322_Fig1_HTML.gif

Fig. 2

Variability in difficulty and discrimination indices for a single MCQ item (Item 6) graphed across cohort

40037_2016_322_Fig2_HTML.gif

Table 1

Frequency of items being categorized in each of Ebel and Frisbie’s [13] categories, displayed for each item

Discrimination categorization

Item

Poor

Low

Acceptable

Good

Excellent

1

12

0

2

2

6

2

19

1

0

0

2

3

20

1

0

1

0

4

12

4

3

2

1

5

14

0

0

5

3

6

10

1

4

2

5

7

11

2

1

2

6

8

14

2

2

2

2

9

12

1

3

3

3

10

16

3

1

1

1

11

14

1

3

1

3

12

7

1

4

2

8

13

5

4

2

5

6

14

12

2

2

1

5

15

8

4

2

1

7

16

19

0

1

0

2

17

18

1

1

2

0

18

15

0

2

0

5

19

13

3

0

0

6

20

11

3

1

1

6

21

19

0

0

1

2

22

5

1

3

3

10

23

16

0

1

2

3

24

19

0

1

1

1

25

20

0

1

1

0

26

9

2

0

5

6

27

12

1

1

1

7

28

8

3

5

2

4

29

9

6

2

2

3

30

9

3

1

1

8

31

9

1

2

2

8

32

13

2

2

1

4

Table 2

Frequency of items either removed or kept for the total exam score, presented by item, and for each quality monitoring guidelinea

Item

Number of times item is removed from the total score applying Guideline 1

Number of times item is removed from the total score applying Guideline 2

Number of times item is removed from the total score applying Guideline 3

1

0

7

18

2

0

18

21

3

0

17

22

4

9

1

1

5

0

11

21

6

17

3

2

7

0

1

14

8

0

12

22

9

0

10

21

10

0

14

22

11

0

3

11

12

2

0

7

13

0

2

10

14

0

6

20

15

0

1

11

16

0

15

21

17

0

15

22

18

0

14

22

19

1

6

18

20

2

1

6

21

0

15

22

22

2

1

6

23

0

14

22

24

0

17

21

25

0

15

22

26

3

1

3

27

0

8

19

28

1

1

4

29

0

4

20

30

6

1

2

31

0

1

11

32

0

12

21

aGuideline 1: an item should be excluded from the examination if the item difficulty was ± two standard deviations from the average difficulty of the examination. Guideline 2: an item should be excluded if it fell ± two standard deviations from the passing score. Guideline 3: an item should be excluded if difficulty was less than 0.2 or more than 0.8 [15].

Language: English
Published on: Jan 3, 2017
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2017 Meredith Young, Beth-Ann Cummings, Christina St-Onge, published by Bohn Stafleu van Loghum
This work is licensed under the Creative Commons Attribution 4.0 License.