BAB V STATISTIKA NON PARAMETRIK|RELATED SAMPLE|STATISTIKA TERAPAN

Summary

This video tutorial covers non-parametric hypothesis testing using SPSS, specifically focusing on related samples. The presenter demonstrates three practical examples: testing the impact of Idul Fitri on gamis production, evaluating consumer satisfaction with new product packaging, and assessing the effect of a new flavor on consumer purchase behavior. Through step-by-step instructions, viewers learn how to apply tests like the Wilcoxon signed-rank test and the Sign Test, interpret SPSS outputs, and draw conclusions based on hypothesis results. The video offers clear insights into statistical analysis for business decisions, with practical applications for real-world data.

Takeaways

• 😀 Non-parametric hypothesis testing is used for analyzing data where assumptions of parametric tests are not met, such as paired samples.
• 😀 The first case study tests whether production at a boutique increases during Ramadan, using the Wilcoxon Signed Rank Test with SPSS.
• 😀 The hypothesis in the first case study is that production after Ramadan will be higher than before, with a significance level of 0.10.
• 😀 The SPSS setup in case study 1 involves entering data for 'before' and 'after' Ramadan production and selecting the Wilcoxon Signed Rank Test.
• 😀 In case study 1, the p-value was 0.0000, which is below the significance level of 0.10, leading to the rejection of the null hypothesis.
• 😀 Case study 2 examines whether a new packaging design increases customer satisfaction, using the Sign Test with a significance level of 0.06.
• 😀 The hypothesis in case study 2 is that customer satisfaction after the redesign will be greater than before, indicating improvement.
• 😀 The SPSS setup for case study 2 involves entering satisfaction scores for before and after the redesign and performing a Sign Test.
• 😀 In case study 2, the p-value was 0.302, which is greater than 0.06, so the null hypothesis is accepted, indicating no significant increase in satisfaction.
• 😀 Case study 3 tests whether a new fruit drink flavor increases consumer purchases, using the Wilcoxon Signed Rank Test with a significance level of 0.05.
• 😀 In case study 3, the p-value was 0.427, which is higher than 0.05, leading to the acceptance of the null hypothesis, meaning the new flavor did not increase purchases.

Q & A

What is the main focus of the video?

The video focuses on explaining applied statistics, specifically non-parametric hypothesis testing for two related samples using SPSS, with three business case studies.

What is the null hypothesis (H0) in the first case study about butik Lestari?

In the first case study, the null hypothesis (H0) is that the proportion of models produced before and after Idul Fitri is the same.

What was the outcome of the first case study after hypothesis testing?

The outcome of the first case study was that the null hypothesis (H0) was rejected because the significance value (Sig = 0.0000) was less than the significance level (0.10), indicating that production increased during the holiday period.

Which statistical test was used in the first case study, and why?

The Wilcoxon Signed-Rank Test was used in the first case study because it is suitable for comparing two related samples, in this case, production data before and after Idul Fitri.

What is the significance of the result in the second case study about PT Rama?

In the second case study, the result of the Sign Test (Sig = 0.302) was greater than the significance level (0.06), meaning the null hypothesis (H0) could not be rejected, and the new packaging did not significantly affect consumer satisfaction.

Why was the Sign Test used in the second case study on consumer satisfaction?

The Sign Test was used in the second case study because it is a non-parametric test suitable for analyzing paired data, specifically consumer satisfaction before and after a change in product packaging.

What does a Sig value greater than the significance level indicate in hypothesis testing?

A Sig value greater than the significance level indicates that there is no significant difference between the two groups being compared, meaning the null hypothesis (H0) is not rejected.

In the third case study about PT Gemilang Sosro, what is the hypothesis being tested?

In the third case study, the null hypothesis (H0) is that the number of bottles purchased before and after the introduction of a new flavor is the same, while the alternative hypothesis (H1) suggests that the purchases increase after the new flavor introduction.

What statistical test was used in the third case study, and why?

The Wilcoxon Signed-Rank Test was used in the third case study, similar to the first, because it is appropriate for testing differences between two related samples, such as purchases before and after the new flavor introduction.

What conclusion was drawn from the third case study regarding the new flavor's impact on sales?

The conclusion from the third case study was that the new flavor did not significantly affect sales, as the Sig value (0.427) was greater than the significance level (0.05), meaning the null hypothesis (H0) was not rejected.

How do the results of the three case studies reflect the importance of statistical hypothesis testing in business decision-making?

The results show that statistical hypothesis testing can help businesses assess the effectiveness of changes, such as production adjustments, packaging alterations, or product innovations, and make informed decisions based on the evidence from data analysis.