## Hypothesis test Independent sample T test, F test with Excel

## 1. Data Collection

Living expenses for college students

**2. Data Preprocessing**

Use only the necessary gender and monthly living expenses columns.

Since there are empty cells (nulls), they are removed first.

Sort males and females and check the number using the COUNTIF function.

Male =COUNTIF(A2:A100,A2)

Female =COUNTIF(A2:A100,A50)

**3. Data visualization and analysis**

Establish a null hypothesis to conduct the F test and T test.

Null hypothesis: There is no difference in monthly living expenses between male and female college students. (The variance of the two groups is the same.)

Alternative hypothesis: There is a difference in monthly living expenses between male and female college students. (The variances of the two groups are not the same.)

*** F test**

First, find the variance of the two groups

The size of variance in the male group is larger than that of the female group.

In other words, it is interpreted that men’s living expenses are more spread out in both directions from the men’s average living expenses than women’s.

Since Male dispersion is large, set it as Variable1 (

P-value without using the test statistic (F ratio = F) in step 4) If you only use it, you do not need to consider the size of the variance of the sample group)

Significance level (Alpha): Tested at 95% confidence level (0.05)

P-value 0.44617 > Since the significance level (Alpha) is higher than 0.05, the null hypothesis fails to be rejected (null hypothesis accepted).

Test statistic (F=F ratio) 1.038957534 < Since it is lower than the rejection value (F Critical on-tail) 1.6083, the null hypothesis is rejected. Failed to reject (null hypothesis accepted)

**Conclusion: The population variance of the two groups is the same. (= null hypothesis)**

**In other words, since the two groups have equal variances, run the “t-test: two groups assuming equal variances” to run the t-test.**

***T black**

Since it is not a comparison of the size of means, the P value of the two-tailed test is used.

P(T=t) tw-tail 0.7611 > Since it is greater than the significance level (Alpha) 0.05, the null hypothesis is rejected (null hypothesis is adopted).

**Conclusion: There is no difference in the average monthly living expenses of male and female college students.**

**The population variance of the two groups is the same. (Adopt null hypothesis)**

*** T test**

**determines whether there is a difference in means between two groups**

Independent sample t-test: Comparison of means of two different groups (feet sizes for men and women)

Paired-sample t-test: Comparison of pre-post means of the same group (hair thickness before and after taking hair loss treatment – same subjects)

**Independent samples t test**

Assumption of equal variance: Two groups: When equal variance can be assumed (null hypothesis)

Assumption of equal variance: Two groups: When equal variance cannot be assumed (alternative hypothesis)

**Before performing the T test, determine whether the variance is equal or heterogeneous using the F test P value.**

If P value > greater than significance level 0.05, equal variance (null hypothesis). If

P value < less than significance level 0.05, heteroscedasticity (alternative hypothesis).

**Resources **

University Students Monthly Expenses Monthly living expenses for male and female university students

Removing empty cell (null) values in Excel

Distribution concept and how to use Excel

Hypothesis test, F test concept and how to use Excel

Independent samples T test, F test concepts and how to use Excel

- Sales data analysis_Supermarket
- Standard deviation, skewness, kurtosis, outlier analysis_salary data
- Correlation Analysis, Countifs, Date_Bakery Sales
- Correlation, linear regression line, slope_base interest rate and housing sales volume
- Hypothesis test Independent sample T test, F test Analysis of cost of living for male and female college students
- Hypothesis test, paired-sample T test, analysis before and after taking hair loss medication
- Parametric Pearson Correlation Consumer Price Index Sentiment Index Analysis
- Spearman correlation coefficient ranking analysis of credit card usage amount by region
- Korean single-person household income consumption pattern