Development of Knowledge Test

A knowledge test is a tool used to measure the factual information, understanding, and awareness of respondents about a particular subject, innovation, or practice (e.g., farmers’ knowledge about Integrated Pest Management, soil conservation, or dairy practices).

It must be valid, reliable, and objective.

Steps in Developing a Knowledge Test

1. Planning the Test

• Define the objectives: What do you want to measure? (e.g., knowledge of farmers on organic farming).
• Specify the content area: List the important units/topics/subtopics.
• Decide the type of knowledge:
  • Factual knowledge (facts, terms, definitions).
  • Comprehension (understanding of principles, concepts).
  • Application knowledge (ability to apply knowledge in practical situations).

2. Collection of Items (Question Pool)

• Prepare a large pool of items (statements/questions) covering the full content area.
• Items can be:
  • Multiple-choice questions (MCQs)
  • True/False statements
  • Fill-in-the-blanks
  • Matching type
• Ensure items are simple, unambiguous, and relevant.

Example:

• IPM includes both chemical and non-chemical methods. (True/False)
• Which of the following is a bio-control agent? (Options…)

3. Editing of Items

• Remove ambiguous, vague, or overlapping items.
• Ensure language is simple and free from technical jargon.
• Each item should test one aspect of knowledge only.

4. Pre-Testing / Item Analysis

• Administer the draft test to a small representative sample of respondents (pilot testing).
• Analyze each item using:
  • Difficulty Index (P): % of respondents answering correctly. Ideal range: 20–80% (neither too easy nor too difficult).
  • Discrimination Index (D): Ability of an item to discriminate between high and low scorers. Good items: D ≥ 0.30.
  • Point-biserial correlation (rpbis): Correlation of item score with total test score. Accept items with rpbis ≥ 0.20.

5. Selection of Final Items

• Retain items that fall within acceptable difficulty and discrimination ranges.
• Discard weak items.
• Ensure content coverage and balance.

6. Administration of Final Test

• Prepare instructions for respondents.
• Ensure standardized administration (same procedure for all).
• Decide scoring pattern (1 mark for correct, 0 for wrong/blank).

7. Establishing Reliability

• Reliability = Consistency of the test results.
• Methods:
  • Split-half reliability (test divided into two halves, scores correlated).
  • Test-retest method (administer twice, correlate scores).
  • Kuder-Richardson formula (KR-20/21) for dichotomous items.
  • Acceptable reliability: ≥ 0.70.

8. Establishing Validity

• Validity = Does the test measure what it is supposed to measure?
• Types:
  • Content validity (adequate coverage of subject matter).
  • Construct validity (logical structure of the test).
  • Criterion-related validity (correlation with external criterion, e.g., experts’ ratings).

9. Standardization of the Test

• Prepare the final test with instructions, scoring key, and interpretation guidelines.
• Document procedure for administration, scoring, and interpretation.

Characteristics of a Good Knowledge Test

• Validity – measures intended knowledge accurately.
• Reliability – gives consistent results.
• Objectivity – free from personal bias.
• Usability – easy to administer and score.
• Discrimination – differentiates between knowledgeable and less knowledgeable respondents.

Example Application in Extension Education

• Measuring farmers’ knowledge on organic farming.
• Measuring students’ knowledge of ICT tools in extension.
• Assessing knowledge gain before and after training programs.