STATISTIKA - Cara Manual Uji Validitas Pearson Product Moment Correlation

Summary

This tutorial explains how to perform a validity test for individual test items using the Product Moment Correlation technique manually. It walks through calculating the correlation coefficient (r-hitung) for each test item and compares it with a critical value (r-tabel) to determine validity. The process involves summing scores, calculating necessary values (e.g., X, Y, Sigma), and referencing the Product Moment correlation table for validation. The tutorial also covers how to apply this method for multiple test items, providing a clear step-by-step guide for testing item validity in an educational context.

Takeaways

• 😀 Validity of test items is determined using Pearson's Product Moment correlation.
• 😀 A test item is considered valid if the calculated correlation coefficient (r-hit) is greater than the critical value (r-table).
• 😀 The formula for calculating Pearson's Product Moment correlation involves comparing the sum of products of individual scores with the sum of squared scores for each test item.
• 😀 The significance level (alpha) is set at 5% (0.05) for this validation test.
• 😀 To perform the validity test, first, sum the scores of each student for all test items.
• 😀 For each test item, calculate the Pearson correlation by considering the individual item score as 'x' and the total score as 'y'.
• 😀 After calculating the correlation, compare the result (r-hit) to the critical value (r-table) based on the sample size (n) and significance level.
• 😀 If the calculated r-hit value is greater than the r-table value, the test item is considered valid.
• 😀 An example is provided using data from 10 students, each with scores for four test items ranging from 1 to 5.
• 😀 If any test item is invalid (r-hit less than r-table), it should be discarded as it does not measure what it is supposed to.
• 😀 The process can be repeated for each test item individually to assess their validity.

Q & A

What is the main purpose of testing item validity using the Pearson Product Moment Correlation?

The main purpose is to assess whether each individual test item (question) accurately measures the concept it is intended to evaluate. This is done by calculating the correlation between the score of each individual item (X) and the total score (Y) to check if the item is valid.

What does the formula for Pearson Product Moment Correlation involve?

The formula involves summing up the product of the scores for each item and the total scores, and using those sums to calculate a correlation coefficient (r). This is done by considering variables X (individual item scores) and Y (total scores), and applying a formula that incorporates sums of X, Y, X², Y², and XY values.

What does 'r-hitung' represent in this context?

'r-hitung' represents the calculated correlation coefficient between each item score and the total score. It is compared to a critical value from a correlation table ('r-tabel') to determine whether the item is valid or not.

How is the critical value ('r-tabel') determined?

'r-tabel' is determined by the number of subjects (n) and the significance level (alpha). For example, if alpha = 0.05 and n = 10, the critical value can be found in a Pearson Product Moment correlation table, which provides the threshold value to compare against 'r-hitung'.

What is the significance of comparing 'r-hitung' with 'r-tabel'?

The comparison determines whether the item is valid. If 'r-hitung' is greater than 'r-tabel', the item is considered valid because it shows a strong correlation with the total score. If 'r-hitung' is lower, the item is deemed invalid.

How is the total score (Y) for each subject calculated?

The total score (Y) is the sum of the individual scores for each item (X) for a subject. For example, if a subject scores 4, 3, 5, and 3 on four items, their total score would be the sum of these values.

What happens if an item is deemed invalid?

If an item is deemed invalid, it is discarded because it does not measure the intended concept effectively. Only valid items are used in further analysis, such as reliability testing.

What does the process of calculating 'r-hitung' involve in practice?

In practice, the process involves adding the individual scores for each subject, computing the necessary sums (such as X, Y, X², Y², and XY), applying the Pearson Product Moment formula, and comparing the resulting 'r-hitung' with 'r-tabel'.

How is the value of 'r-tabel' related to the sample size?

'r-tabel' is influenced by the sample size (n) and the significance level (alpha). As the sample size increases, the critical value from the table typically decreases, making it easier to achieve a statistically significant result.

What is the role of reliability testing after determining item validity?

Once the validity of each item is confirmed, reliability testing is conducted to assess whether the items consistently measure the intended concept across multiple trials or subjects. If the items are reliable, they can be used in further research or assessments.