inclusion in the final version of a test.

This
Item Analysis process ensures that each item contributes
I. Understanding Item Analysis meaningfully to the assessment's overall goals.

2. Detection of Item Defects. Through careful

A. Definition and Purpose analysis, educators can identify structural or
Item analysis is a systematic method of content-related problems in test items. These
examining student responses to individual test might include:
questions (items) to evaluate the quality and - Ambiguous wording
effectiveness of those items and the test as a - Multiple correct answers
whole. This process is crucial in educational - Misleading distractors
assessment as it provides teachers with - Unintentional clues to the correct answer
valuable information about both the test items 3. Identification of Class-Wide Learning Gaps.
and student learning (Brown & Abeywickrama, By analyzing patterns in student responses,
2019). teachers can detect areas where the entire
The practice of item analysis emerged from class may be struggling, potentially indicating:
classical test theory in the early 20th century, - Insufficient instructional coverage
as educators sought more scientific approaches - Unclear presentation of concepts
to assessment development. According to - Need for additional teaching strategies
Miller et al. (2013), item analysis serves as a 4. Recognition of Individual Learning Needs.
crucial bridge between test development and Item analysis helps identify specific areas where
instructional improvement, offering insights individual students or groups of students may
that can enhance both assessment quality and need additional support or remediation.
teaching effectiveness.

When properly conducted, item analysis helps

educators: B. Types of Item Analysis
- Identify problematic test items that may need Modern item analysis typically encompasses
revision three main components, each providing unique
- Understand patterns in student learning and insights into test quality and student
misconceptions performance:
- Make informed decisions about test
construction 1. Difficulty Index (p)
- Improve the overall quality of assessments The difficulty index, denoted as 'p', is a
fundamental measure that indicates the
proportion of students who answered an item
Primary Purposes of Item Analysis correctly. As Reynolds and Livingston (2018)
Drawing from Payne's (1992) foundational explain, this index provides a straightforward
work in educational assessment, item analysis way to assess whether test items are
serves four essential purposes: appropriately challenging for the intended
audience.
1. Selection of Optimal Items. Item analysis
helps identify the most effective questions for
The difficulty index ranges from 0.0 to 1.0, - Draw responses from the lower-performing
where: group more than the higher-performing group
- 0.0 indicates that no students answered the
item correctly
- 1.0 indicates that all students answered the Practice Activity
item correctly
Before moving on, try this quick self-check
activity:

2. Discrimination Index (D) Consider a test item that:

- Was answered correctly by 85% of students
The discrimination index measures how
- Was answered correctly by 90% of high-
effectively an item distinguishes between high-
performing students and 80% of low-
performing and low-performing students. A
performing students
good test item should be answered correctly
- Had one distractor that no student chose
more often by students who perform well on
the overall test than by those who perform
poorly.
II. Computing the Difficulty Index (p)
This index helps ensure that test items are valid
A. Understanding Difficulty Index
indicators of student achievement and
understanding. According to Haladyna and The difficulty index, commonly denoted as 'p', is
Rodriguez (2013), effective discrimination is a fundamental statistical measure in test
crucial for maintaining test validity and analysis that represents the proportion of
reliability. students who correctly answered a particular
test item. As noted by Oosterhof (2009), this
metric is somewhat counterintuitive: higher p-
3. Analysis of Response Options/Distractor values indicate easier items, while lower p-
Analysis values indicate more difficult ones.

For multiple-choice items, distractor analysis

examines how well the incorrect options
Interpretation Scale
(distractors) function. This analysis reveals:
- Which distractors are effectively drawing The difficulty index typically ranges from 0.0 to
students with incomplete understanding 1.0, and can be interpreted using the following
- Which distractors might be too obvious or too scale:
confusing
- Whether all options are plausible and serving Difficulty Index (p) | Interpretation -
their intended purpose Implications for Test Item
Below 0.25 | Very Difficult - May need revision
A well-functioning distractor should: unless intended to identify top performers |
- Appear plausible to students who don't fully 0.25 – 0.75 | Average Difficulty - Optimal
understand the concept range for most classroom assessments |
- Be clearly incorrect to knowledgeable Above 0.75 | Easy - May need revision unless
students intended for mastery testing |
According to McCowan and McCowan (2016), lower 27% (low group) of test-takers
items falling in the average difficulty range - The use of 27% was established by Kelley
(0.25 - 0.75) are generally most useful for (1939) as optimal for creating sufficiently
classroom assessment as they provide optimal different groups while maintaining adequate
discrimination among students and yield more sample sizes
reliable test scores.
3. Count Correct Responses
- Tally the number of correct responses for
each item in both groups
B. Calculation Methods
- Create a systematic recording system to
The method for calculating the difficulty index avoid errors
varies depending on the type of test item being
4. Apply the Formula
analyzed. Let's examine both major approaches:
- Insert the values into the formula
1. For Items Scored with 0 and 1 - Calculate the difficulty index for each item
This method applies to objective-type
questions such as:
- Multiple-choice items Example:
- True/False questions 1. Let's analyze a test item with the following
- Matching-type items data:
- Fill-in-the-blank items with single correct - High group (n=15): 12 correct responses
answers - Low group (n=15): 6 correct responses

Formula: p = (12 + 6) / (2 × 15)

p = 18 / 30
p = (Hc + Lc) / 2n p = 0.60
Interpretation: With p = 0.60, this item falls
Where:
within the average difficulty range and is
p = difficulty index
appropriate for classroom assessment.
Hc = number of correct responses in high group
Lc = number of correct responses in low group 2.
n = number of examinees in each criterion
group
2. For Items Scored Other Than 0 and 1
This method applies to:
Step-by-Step Calculation Process: - Essay questions
- Problem-solving items
1. Arrange Test Scores
- Performance tasks
- Order all test papers from highest to lowest
- Projects with scoring rubrics
scores
- This ordering is crucial for identifying Formula:
criterion groups
p = (ΣXi + ΣXj) / (2n(Xmax - Xmin))
2. Select Criterion Groups
- Identify the upper 27% (high group) and
Where: IV. Computing the Discrimination Index (D)
p = difficulty index
A. Understanding Discrimination Index
ΣXi = sum of scores from high group
ΣXj = sum of scores from low group The discrimination index (D) is a critical
n = number of examinees in criterion group measure that indicates how effectively a test
Xmax = maximum possible score item distinguishes between high-performing
Xmin = minimum possible score and low-performing students. According to
Ebel and Frisbie (2019), an item with good
discrimination power will be answered correctly
Implementation Guidelines:
more often by students who performed well on
1. Use consistent scoring criteria
the entire test than by those who performed
2. Apply rubrics uniformly
poorly.
3. Consider inter-rater reliability for subjective
items Theoretical Foundation

The discrimination index is based on the

assumption that a valid test item should
C. Practical Applications
effectively differentiate between students who
Calculating Test-Wide Difficulty have mastered the content and those who
The overall difficulty index of a test can be haven't. As explained by Reynolds et al. (2017),
calculated using: this measure helps ensure that test items are
functioning as intended and contributing
p̄ = Σpi / k meaningfully to the assessment's overall
validity.
Where:
p̄ = mean difficulty index
pi = difficulty index of item i
Interpretation Scale
k = number of items
The discrimination index typically ranges from -
1.0 to +1.0, with the following interpretations:

Using Results for Test Improvement Discrimination Index (D) | Interpretation -

Based on Haladyna's (2004) recommendations: Recommended Action
Above 0.40 | Very Good Item - Retain item
1. For Very Difficult Items (p < 0.25)
0.30 to 0.40 | Good Item - Minor improvements
- Review for unclear wording
possible
- Check for curriculum alignment
0.20 to 0.29 | Fair Item - Needs improvement
- Consider moving to advanced assessments
Below 0.20 | Poor Item - Major revision or
2. For Very Easy Items (p > 0.75) rejection
- Use as warm-up questions Negative Values | Problematic Item -
- Consider for mastery testing Immediate revision or rejection
- May need replacement for discriminating
As noted by Nitko and Brookhart (2021),
among students
negative discrimination values are particularly
concerning as they indicate that lower-
performing students are more likely to answer
correctly than higher-performing students—
D = (18 - 8) / 20
suggesting a fundamental problem with the
D = 10 / 20
item.
D = 0.50

Interpretation: With D = 0.50, this item is

B. Calculation Methods functioning very well at discriminating between
high and low performers.
1. For Items Scored with 0 and 1
2.
This method applies to objective-type
questions where answers are either correct or 2. For Items Scored Other Than 0 and 1
incorrect.
This calculation method applies to subjective
Formula: items like essays and problem-solving
questions.
D = (Hc - Lc) / n
Formula:
Where:
D = discrimination index D = (ΣXi - ΣXj) / [n(Xmax - Xmin)]
Hc = number of correct responses in high group
Where:
Lc = number of correct responses in low group
D = discrimination index
n = number of examinees in criterion group
ΣXi = sum of scores from high group
ΣXj = sum of scores from low group
n = number of examinees in criterion group
Step-by-Step Calculation Process: Xmax = maximum possible score
Xmin = minimum possible score
1. Prepare the Data
- Use the same 27% criterion groups
established for difficulty index
- Organize data systematically for analysis C. Special Considerations

2. Count Response Patterns 1. Relationship with Difficulty Index

- Tally correct responses in high group (Hc)
According to DiBattista and Kurzawa (2011),
- Tally correct responses in low group (Lc)
the relationship between item difficulty and
3. Apply the Formula discrimination is curvilinear:
- Calculate the difference between high and - Very easy items (p > 0.90) typically show poor
low group performance discrimination
- Divide by the number of students in each - Very difficult items (p < 0.20) also tend to
group discriminate poorly
- Items of moderate difficulty (0.40 - 0.60) have
Example:
the greatest potential for discrimination
1. Consider a test item with the following data:
- High group (n=20): 18 correct responses 2. Impact of Guessing
- Low group (n=20): 8 correct responses
For multiple-choice items, random guessing can A. Purpose and Importance
affect discrimination. Haladyna (2015) suggests:
Distracter analysis, also known as response
- Using quality distractors to minimize
option analysis, is a crucial component of item
successful guessing
analysis for multiple-choice tests. According to
- Considering the number of options when
Haladyna and Rodriguez (2013), effective
interpreting results
distractors—the incorrect options in multiple-
- Adjusting expectations based on item format
choice items—play a vital role in measuring
3. Sample Size Considerations student understanding and preventing
successful guessing.
The stability of discrimination indices depends
on sample size: Significance in Test Development
- Minimum recommended sample: 50 students
Distracter analysis serves several critical
- Ideal sample size: 100 or more students
purposes:
- Smaller samples require cautious
1. Validates the effectiveness of incorrect
interpretation
options
2. Identifies misleading or problematic answer
choices
D. Practical Applications
3. Helps distinguish between student
1. Item Analysis Matrix knowledge and test-taking strategies
4. Contributes to overall test reliability and
Create a decision matrix based on both
validity
difficulty and discrimination:
As noted by Osterlind (2018), well-functioning
Difficulty | Discrimination - Decision
distractors should be:
- Plausible but unquestionably incorrect
Average | High - Ideal item - retain
- Based on common student misconceptions
Average | Low - Revise distractors
- Similar in length and complexity to the correct
Extreme | Low - Consider replacement
answer
Any | Negative - Investigate and revise
- Free from obvious clues or patterns
2. Common Issues and Solutions

1. Low Discrimination with Average Difficulty

B. Conducting Distracter Analysis
- Check for ambiguous wording
- Review distractors 1. Components of Analysis
- Examine content alignment
A comprehensive distracter analysis examines
2. Negative Discrimination three key elements:
- Check for scoring errors
a) Response Frequency
- Review for misleading clues
For each option in a multiple-choice item:
- Assess content accuracy
- Total number of students selecting each
option
- Percentage of students selecting each option
V. Distracter Analysis
- Distribution pattern across all options
b) High-Low Group Selection Patterns option
For each option: - Suggests possible alternative correct answer
- Number of high-performing students selecting or ambiguous wording
the option
Example:
- Number of low-performing students selecting
the option
Option B: Hg = 8, Lg = 2
- Comparison of selection patterns between
IE = (8 - 2) / 20 = 0.30
groups
Interpretation: Distracter needs revision
c) Index of Effectiveness (IE)
Calculate using the formula:

c) For Zero IE Values

IE = (Hg - Lg) / n
Two possible scenarios:
Where: 1. No students selected the option
IE = Index of Effectiveness - Non-functional distracter
Hg = Number of high group students selecting - Needs replacement
the option 2. Equal numbers from both groups selected
Lg = Number of low group students selecting the option
the option - May be functioning randomly
n = Number of students in each criterion group - Review for clarity and plausibility
```

2. Interpretation Guidelines
C. Practical Application
According to Burton et al. (2021), distracter
1. Sample Distracter Analysis Table
effectiveness can be interpreted as follows:

a) For Negative IE Values

- Indicates a functional distracter Item 1: What is the primary function of
- More low-performing students selected the mitochondria?
option Options High Group Low Group IE
- Suggests the distracter is attracting students Interpretation
with incomplete understanding A (correct) 18 8 0.50
Good discrimination
Example:
B 2 6 -0.20
Functional distracter
Option A: Hg = 3, Lg = 9
C 0 4 -0.20
IE = (3 - 9) / 20 = -0.30
Functional but weak
Interpretation: Good distracter functioning
D 0 2 -0.10
Review for improvement

b) For Positive IE Values

- Indicates a problematic distracter
2. Decision-Making Guidelines
- More high-performing students selected the
Based on research by King and Singh (2019): analysis can significantly reduce computational
errors and save valuable time.
IE Value Range | Recommended Action
< -0.15 | Retain distracter 1. Computer Programs and Scanning Devices
-0.15 to 0.15 | Review and possibly revise
a) Automated Scoring Systems
> 0.15 | Major revision needed
- Optical Mark Recognition (OMR) systems
0 (unused) | Replace distracter
- Digital assessment platforms
- Learning Management System (LMS) analytics

D. Common Issues and Solutions Benefits:

- Rapid processing of large datasets
1. Non-Functioning Distractors
- Reduced computational errors
Problems: - Immediate feedback availability
- No students select the option - Standardized reporting formats
- Only random selection patterns
b) Popular Software Solutions
- No discrimination between ability levels
1. Spreadsheet Applications
Solutions: - Microsoft Excel
- Base distractors on common misconceptions - Google Sheets
- Use student errors from free-response - Numbers (for Mac)
questions
2. Specialized Testing Software
- Test items with target audience before final
- ExamSoft
use
- Questionmark
2. Too-Attractive Distractors - TestAnalysis

Problems: 2. Online Tools and Resources

- Draws more high-performing students than
a) Free Online Calculators
intended
1. Website Reactions
- May be partially correct or ambiguous
(www.surveyreaction.com/itemanalysis.asp)
- Could indicate multiple acceptable answers
- Basic item analysis statistics
Solutions: - User-friendly interface
- Review for technical accuracy - Free access
- Clarify wording
2. DepEd Tambayan Calculator
- Ensure single best answer
- Specifically designed for educators
- Comprehensive analysis reports
- Excel-based calculations
VI. Practical Applications and Limitations
b) Implementation Guidelines
A. Using Technology for Item Analysis
According to Chase and Brown (2020):
In today's digital age, educators have access to 1. Verify data entry accuracy
various technological tools that can streamline 2. Cross-check results with manual calculations
the item analysis process. According to Qualls 3. Maintain backup copies of raw data
(2019), the effective use of technology in item 4. Document analysis procedures
B. Creating an Item Data File 2. Practical Limitations

1. Advantages of Item Banking a) Impact on Table of Specifications

Challenges identified by Wauters et al. (2018):
Research by McCallin (2020) identifies several
- Maintaining content coverage
benefits:
- Balancing statistical and content validity
a) Quality Improvement - Addressing curriculum requirements
- Tracks item performance over time
b) Resource Constraints
- Facilitates systematic revision
1. Time requirements
- Enables trend analysis
- Data collection
b) Efficiency - Analysis procedures
- Reduces test development time - Implementation of revisions
- Streamlines test assembly
2. Technical expertise needed
- Facilitates item sharing
- Statistical understanding
c) Documentation - Software proficiency
- Maintains item history - Interpretation skills
- Records revision details
- Tracks usage statistics
D. Best Practices for Implementation

1. Item Revision Guidelines

C. Limitations of Item Analysis
Based on Miller's (2022) recommendations:
1. Statistical Constraints
Decision Matrix for Item Revision
According to Thompson and Levine (2021),
several factors can affect item analysis Statistical Indicators | Action Required
accuracy: p < 0.30, D > 0.30 | Review difficulty level
p > 0.90, D < 0.20 | Consider replacement
a) Sample Size Effects
Negative D value | Immediate revision
- Minimum recommended sample: 50 students
Poor distractors | Revise options
- Optimal sample size: 100+ students
- Impact on reliability of indices

b) Group Homogeneity
2. Documentation Requirements
- Restricted range effects
- Impact on discrimination values Essential records to maintain:
- Need for representative samples 1. Item performance history
2. Revision decisions and rationale
c) Test Length Considerations
3. Student performance patterns
- Minimum items for reliable analysis
4. Analysis procedures used
- Impact of test length on indices
- Balance between comprehensiveness and
practicality