Perhitungan Uji Kruskal Wallis dengan Excel, STATCAL dan SPSS

Summary

In this video, the presenter explains the manual calculation of the Kruskal-Wallis test using Excel, demonstrating how to compare the results with software like SPSS and Starcall. The example focuses on testing whether there's a significant difference in math exam scores among students with different hobbies: reading, sports, and music. The video covers the process step by step, from organizing data and ranking it to performing calculations and comparing results with statistical software. It also highlights important concepts like the Kruskal-Wallis test, ranking, and statistical adjustments, making it a helpful guide for those new to the test.

Takeaways

• 😀 The video introduces the Kruskal-Wallis test and its manual calculation using Excel, with a comparison to results from SPSS and StarCall software.
• 😀 The example case involves a teacher analyzing whether there is a significant difference in math exam scores between students with different hobbies: reading, sports, and music.
• 😀 The data for the analysis includes 20 students, with 7 students in each of the reading and sports groups, and 6 in the music group.
• 😀 The Kruskal-Wallis test is used when there are more than two groups, which is the case here, as there are three hobby groups being analyzed.
• 😀 The video explains how to organize and input the data into SPSS for the Kruskal-Wallis test, where 'hobby' is the categorical variable and 'math exam score' is the numeric variable.
• 😀 The process involves ranking the data from lowest to highest, assigning ranks to ties by averaging the ranks of tied values.
• 😀 Excel is used to manually calculate the Kruskal-Wallis test statistic (H), with steps detailed for sorting, ranking, and calculating the sums of ranks for each group.
• 😀 The formula for calculating the Kruskal-Wallis statistic is provided, including adjustments for tied ranks and sample sizes for each group.
• 😀 The results of the manual calculation are compared to the SPSS and StarCall software outputs, and the conclusion is that the Kruskal-Wallis test statistic is 13.318, confirming a significant difference in scores across the hobby groups.
• 😀 The final conclusion is that there is a significant difference in math scores between students with different hobbies, as the Kruskal-Wallis statistic exceeds the critical chi-square value at the 5% significance level.

Q & A

What is the Kruskal-Wallis test and how is it used in the script?

The Kruskal-Wallis test is a non-parametric method used to determine if there are significant differences between three or more independent groups. In the script, it is used to compare the test scores of students with different hobbies (reading, sports, music) to see if there are significant differences in their math exam scores.

Why is the Kruskal-Wallis test chosen in this case?

The Kruskal-Wallis test is chosen because the data involves more than two groups (reading, sports, and music) and the test is specifically designed for such cases where the groups are independent and the data is non-normally distributed.

What role does Excel play in the manual calculation of the Kruskal-Wallis test?

Excel is used to manually rank the exam scores and calculate the Kruskal-Wallis statistic. The script walks through the process of organizing data, ranking values, and performing necessary calculations to find the Kruskal-Wallis statistic (H).

How does the script handle tied ranks in the dataset?

When there are tied ranks, the script uses the average of the ranks for the tied values. For example, if two students have the same score, their ranks are averaged to ensure fairness in the ranking process.

What are the key steps involved in performing the Kruskal-Wallis test manually in Excel?

The key steps include organizing the data by group, sorting the exam scores, assigning ranks to the scores, calculating the total rank for each group, and then using a specific formula to calculate the Kruskal-Wallis statistic (H).

How does the script compare the manual calculation with results from SPSS and StarCall?

After performing the manual calculation in Excel, the script compares the calculated Kruskal-Wallis statistic with the results obtained from SPSS and StarCall software to ensure consistency and validate the manual method.

What does the value of H (13.318) indicate in the Kruskal-Wallis test?

The value of H (13.318) is the Kruskal-Wallis statistic. This value is then compared to the critical value from the Chi-Square distribution to determine whether there is a statistically significant difference between the groups.

What is the critical value used for comparison in the Kruskal-Wallis test, and how is it determined?

The critical value for the Kruskal-Wallis test is derived from the Chi-Square distribution table. It depends on the degrees of freedom (df), which in this case is the number of groups minus one. The critical value is used to determine whether the Kruskal-Wallis statistic is in the rejection region of the test.

What does the result of H = 13.318 being greater than the critical value (5.99) imply?

Since the calculated H value (13.318) is greater than the critical value (5.99), we reject the null hypothesis (H0). This implies that there is a statistically significant difference in the math test scores between the three hobby groups (reading, sports, and music).

How does the manual Kruskal-Wallis test result compare with the output from the StarCall software?

The manual calculation result (H = 13.318) matches the result from StarCall software, confirming the accuracy of the manual process. This consistency demonstrates that the manual and software methods yield the same outcome for this analysis.