Cara / Tutorial Menghitung Validitas Butir Soal manual dengan mudah dan praktis

Summary

This educational video focuses on the process of validating individual test items rather than the entire test. It demonstrates how to calculate item validity using the Point Biserial correlation formula, ensuring each question is empirically validated. The script walks through an example involving multiple-choice questions, explaining how to analyze data from a test tryout to determine which items are valid. The video emphasizes the importance of calculating various metrics such as MP3, MT, and standard deviation, and compares results with a threshold for validity. Viewers are encouraged to apply these methods to assess the quality of their own test items.

Takeaways

• 😀 The focus of the session is on how to validate individual test items, not the entire test. This approach is important for assessing each question's validity.
• 😀 The previous lesson covered validating the entire test using the product-moment correlation, comparing valid items with new ones to assess their validity.
• 😀 The goal of the session is to determine which individual test items are valid and which are not, using point-biserial correlation.
• 😀 This validation method is typically applied to objective-type questions such as multiple-choice questions, true/false questions, and short-answer questions.
• 😀 The process for calculating point-biserial correlation involves several steps, including calculating the mean score of the test, the total score, the standard deviation, and the proportion of students answering correctly.
• 😀 Example case: A teacher conducting a daily quiz needs to ensure the test items are valid by performing a tryout with other students before administering it.
• 😀 In the example, 15 students were tested, with answers indicated by a correct or incorrect response for each item.
• 😀 For each test item, the proportion of students answering correctly is calculated, which is used in the point-biserial correlation formula to assess validity.
• 😀 After calculating various parameters such as XT (total correct responses by students), the teacher uses these values in the formula to determine item validity for each question.
• 😀 Once all calculations are complete, the teacher can assess the validity of each test item using the threshold of a correlation coefficient (0.3). Items with a correlation greater than 0.3 are considered valid, while those below this threshold are not.

Q & A

What is the focus of the video lesson?

The lesson focuses on how to validate each individual question in a test, rather than validating the entire test at once.

What method was used to validate the entire test in a previous lesson?

The previous lesson used the Product-Moment correlation method to compare valid questions with new ones to determine their validity.

How does the validation of individual questions differ from validating the entire test?

The validation of individual questions involves assessing each question separately to determine whether it is valid or not, while validating the entire test involves evaluating the overall structure and consistency of the test as a whole.

What type of questions is the validation method applicable to?

The validation method is suitable for questions with clear right or wrong answers, such as multiple-choice questions, true/false questions, and short-answer questions.

What is the purpose of using the Point Biserial formula in question validation?

The Point Biserial formula is used to calculate the correlation between the responses to each question and the overall test score, helping to determine the validity of each individual question.

What are the key elements needed to use the Point Biserial formula?

The key elements required are the average total score, standard deviation of the total score, the proportion of students answering correctly, and the proportion answering incorrectly for each question.

In the example, how many students are being tested?

There are 15 students in the example, each answering 12 multiple-choice questions.

What does the table of responses show in the example?

The table of responses shows how each student answered each question, with '1' representing a correct answer and '0' representing an incorrect answer.

What is the next step after calculating the necessary elements (P, Q, MP, SD) for each question?

After calculating P, Q, MP, and SD, the next step is to use these values to determine the validity of each question by calculating the Point Biserial correlation coefficient for each item.

How do you determine if a question is valid or not?

A question is considered valid if its Point Biserial correlation coefficient is greater than 0.3. If the value is less than 0.3, the question is considered invalid.