Item Generation Using Constraint Coding
For our medical example, we defined the constraints in the cognitive model. While this format is convenient for reading, it is not convenient for machine processing. As a result, the constraints must be converted into bitmasks. While this process sounds complicated, it can be simplified and expedited with the use of checkboxes. An example of a checkbox representation is the element cough type using the correct options given as follows:
|
COUGH_TYPE |
Common Cold |
Seasonal Flu |
|
mild |
0 |
□ |
|
hacking |
0 |
□ |
|
severe |
□ |
0 |
This checkbox summary produces a relation matrix of:
When we consider all elements in the medical model, we get the full constraint matrix, which was presented in Chapter 4 as follows:
|
Element |
Scenario |
Cough Type |
Body Aches |
Onset |
Results |
Temperature |
|||||||||||||
|
Element |
Values |
Common Cold |
Seasonal Flu |
mild |
hacking |
severe |
slight body aches |
slight body pains |
severe body aches |
severe body pains |
over a few days |
within 3-6 hours |
suddenly |
Positive |
Negative |
CO VI |
37.8 |
CO CD |
39.5 |
|
Scenario |
Common Cold |
1 |
0 |
1 |
1 |
0 |
1 |
1 |
0 |
0 |
1 |
0 |
0 |
0 |
1 |
1 |
1 |
0 |
0 |
|
Seasonal Flu |
0 |
1 |
0 |
0 |
1 |
0 |
0 |
1 |
1 |
0 |
1 |
1 |
1 |
0 |
0 |
0 |
1 |
1 |
|
|
Cough Type |
mild |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
hacking |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
severe |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
Body Aches |
slight body aches |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
slight body pains |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
severe body aches |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
severe body pains |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
Onset |
over a few days |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
within 3-6 hours |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
suddenly |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
Results |
Positive |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
Negative |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
Temperature |
37 |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
37.8 |
1 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
39 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
|
39.5 |
0 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
1 |
|
Systematic Distractor Generation
Our second example uses the key features model. To specify the distrac- tors for this model, we need to conduct three steps: identify the rationales, group distractors by rationales, and ensure that the model can generate enough incorrect options for each correct option. The features for the distractors, as presented in Chapter 5, can be summarized as follows:
|
|
||
|
1 . |
Bronchitis |
(CC, SF) |
|
2 . |
Streptococcal infection |
(CC, SF) |
|
3 . |
Hay Fever |
(CC, SF) |
|
4 . |
Otitis Media |
(CC, SF) |
|
5 . |
Acute Sinusitis |
(CC) |
|
6 . |
Bacterial Pneumonia |
(SF) |
From this summary, we can then specifically identify two rationales: rationales related to the correct option and rationales not related to the correct option.
|
Related |
Not related |
|
Acute Sinusitis, Bacterial Pneumonia |
Bronchitis, Streptococcal Infection, Hay Fever, Otitis Media |
The total number of options in the parent item is five. To simplify our presentation, we generated items with four options in total, meaning we need three distractors for each generated item. The first distractor is easy to identify: flu for cold and cold for flu.
Distractorl: (1) Common Cold, (2) Seasonal Flu
The second distractor can be located in the list of distractors related to key features:
Distractor2: (1) Acute Sinusitis, (2) Bacterial Pneumonia
The remaining distractors can be randomly selected from the list of the values not related to the key features. We just need to ensure that the generation process does not produce duplicates among the item options:
Distractor3: (1) Bronchitis, (2) Streptococcal Infection (3) Hay Fever, (4) Otitis Media
Distractor4: (1) Bronchitis, (2) Streptococcal Infection (3) Hay Fever, (4) Otitis Media
Once we have defined all four distractor values, we can produce the item model in its final form:
|
Stem |
[AGE] female sees her doctor and reports that she's been experiencing a [COUGH TYPE] cough and [BODY_ACHES] that have developed [ONSET]. Upon examination, she presents with an oral temperature of [TEMPERATURE]°C. What is the most likely diagnosis? |
|
Elements |
AGE: (1) An 18, (2) A [AGEJRANGE] AGE RANGE: 19 to 30, by COUGH_TYPE: (1) mild, (2) hacking, (3) severe BODY ACHES: (1) slight body aches, (2) slight body pains, (3) severe body aches, (4) severe body pains ONSET: (1) over a few days, (2) within 3-6 hours, (3) suddenly TEMPERATURE: (1) 37, (2) 37.8, (3) 39, (4) 39.5 |
|
Key |
Common Cold, Seasonal Flu |
|
Distractors |
Distractorl: (1) Common Cold, (2) Seasonal Flu Distractor2: (1) Acute Sinusitis, (2) Bacterial Pneumonia Distractor3: (1) Bronchitis, (2) Streptococcal Infection, (3) Hay Fever, (4) Otitis Media Distractor4: (1) Bronchitis, (2) Streptococcal Infection, (3) Hay Fever, (4) Otitis Media |
A Sample of Generated Medical Items
Using the method presented in this chapter, a total of 1,248 medical items were generated. When the second layer was added to the medical example (see Chapter 5, Table 5.7), 1 1,232 items were generated. A sample of the generated content from the 1-layer medical model created using the example in this chapter includes the following items:
1 An 18-year-old female sees her doctor and reports that she's been experiencing a mild cough and slight body aches that have developed over a few days. Upon examination, she presents with an oral temperature of 37°C. What is the most likely diagnosis?
A. Bronchitis
B. Seasonal flu
C. Common cold *
D. D. Bacterial pneumonia
E. E. Streptococcal infection
2 A 19-year-old female sees her doctor and reports that she's been experiencing a mild cough and slight body aches that have developed over a few days. Upon examination, she presents with an oral temperature of 37.8°C. What is the most likely diagnosis?
A. Seasonal flu
B. Otitis media
C. Common cold *
D. Bacterial pneumonia
E. Streptococcal infection
3 A 29-year-old female sees her doctor and reports that she's been experiencing a hacking cough and slight body aches that have developed over a few days. Upon examination, she presents with an oral temperature of 37°C. What is the most likely diagnosis?
A. Bronchitis
B. Hay fever
C. Seasonal flu
D. Common cold *
E. Bacterial pneumonia
4 А 21 -year-old female sees her doctor and reports that she's been experiencing a severe cough and severe body aches that have developed within three to six hours. Upon examination, she presents with an oral temperature of 39°C. What is the most likely diagnosis?
A. Bronchitis
B. Seasonal flu *
C. Common cold
D. Acute sinusitis
E. Streptococcal infection
5 A 24-year-old female sees her doctor and reports that she's been experiencing a severe cough and severe body aches that have developed suddenly. Upon examination, she presents with an oral temperature of 39.5°C. What is the most likely diagnosis?
A. Seasonal flu *
B. Otitis media
C. Common cold
D. Acute sinusitis
E. Streptococcal infection
6 A 23-year-old female sees her doctor and reports that she's been experiencing a severe cough and severe body pains that have developed suddenly. Upon examination, she presents with an oral temperature of 39°C. What is the most likely diagnosis?
A. Hay fever
B. Seasonal flu *
C. Common cold
D. Acute sinusitis
E. Streptococcal infection
Methods for Validating Generated Items
