I choose the following 3 hypothesis to test:
- Marriages where the man is older than the woman will be perceived as being more satisfying than marriages where the woman is older than the man.
- There will be a positive association between perceived level of relationship satisfaction and optimism.
- There will be a positive association between perceived level of relationship satisfaction and relationship longevity
The data columns will be the following: Age, Gender and MSATISFA (marriage satisfaction).
- Determine which a) graphs you will use to display the data and b) table(s) you will use to display the data
- Determine the descriptive statistics you will use to summarize the data
The descriptive statistics of my data is size of two samples, sample means and sample standard deviations of MSATISFA.
- Determine whether you will calculate the strength of the association between variables or whether there is a significant difference between group means.
There is no need to calculate the strength of the association, because we assume, that these samples are independent and what to compare their means at a 5% level of significance.
- Create the appropriate graphs and tables
- Calculate the appropriate descriptive statistics
- For males:
x=59.697N=33s=31.965
- For females:
x=61.579N=57s=25.269
7. Calculate the r, t, or F
Here we perform two-sampled Student’s t-test to compare means (I perform it in Excel):
Two-sample T for Males vs Females
N Mean StDev SE Mean
Males 33 59,7 32,0 5,6
Females 57 61,6 25,3 3,3
Difference = mu (Males) - mu (Females)
Estimate for difference: -1,88
95% CI for difference: (-14,90; 11,13)
T-Test of difference = 0 (vs not =): T-Value = -0,29 P-Value = 0,773 DF = 55
8.Refer to the appropriate tables to determine whether the results are significant
According to the Excel output, we’ve got the following conclusion:
Since p-value of test is higher than 0.05, we fail to reject the null hypothesis and have no evidence to say, that Marriages where the man is older than the woman will be perceived as being more satisfying than marriages where the woman is older than the man at 5% level of significance.
Start working with a second hypothesis.
1. Select the appropriate data column(s)
2. Determine which a) graphs you will use to display the data and b) table(s) you will use to display the data
3. Determine the descriptive statistics you will use to summarize the data
4. Determine whether you will calculate the strength of the association between variables or whether there is a significant difference between group means.
5. Create the appropriate graphs and tables
6. Calculate the appropriate descriptive statistics
7. Calculate the r, t, or F
8. Refer to the appropriate tables to determine whether the results are significant
Start working with a third hypothesis:
1. Select the appropriate data column(s)
The following columns: MLENGTH and MSATISFA
2. Determine which a) graphs you will use to display the data and b) table(s) you will use to display the data
All we need here is a Pearson’s correlation coefficient. It shows us the result from -1 to 1. The values closed to 1 are an evidence of strong positive association between the variables.
Also we need the sample size to check the significance of Pearson’s r.
- Determine whether you will calculate the strength of the association between variables or whether there is a significant difference between group means.
I will calculate the strength of the association between variables.
- Create the appropriate graphs and tables
- Calculate the appropriate descriptive statistics
The sample size N=90
We will calculate Pearson’s r below.
- Calculate the r, t, or F
According to the Excel output, the value of Pearson’s r is the following:
r=0.65831
This value is quite close to 1, and it is an evidence of a moderate positive association between the level of satisfaction and marriage length.
8. Refer to the appropriate tables to determine whether the results are significant
We are using the following formula to check the significance:
t=rN-21-r2=0.65831*90-21-0.658312=8.20395
tcritical=t89,0.05=1.987
Since t-observed is higher than t-critical, we can state, that Pearson’s r is significant at 5% level of significance.
