Assignment: Statistical Methods Data Analysis and Interpretation
Assignment using SPSS
Estimation and Hypothesis Testing for means
Part 1. Estimation (Point and Interval estimations)
Dataset: Use the built-in SPSS dataset Employee [Link] (Found in: File -> Open -> Data -> Samples -
> English -> Employee [Link]).
Exercise 1: Confidence Interval for the Mean
SPSS Steps: Analyze-- Descriptive Statistics – Explore-- Dependent List (salbegin).
Questions:
i. Calculate a 95% confidence interval for the mean Beginning Salary (salbegin) for the
entire company.
ii. What is the mean beginning salary?
iii. What are the lower and upper bounds of the 95% confidence interval?
iv. Interpret this interval in a sentence.
Exercise 2: Confidence Interval for Proportions
SPSS Steps: Analyze -- Descriptive Statistics -- Frequencies.
Questions:
i. Calculate the 95% confidence interval for the proportion of employees in the "Custodial"
category (jobcat).
ii. What is the percentage of custodial employees?
iii. Calculate the margin of error based on the frequency output.
Part 2: Hypothesis Testing
Exercise 1: One-Sample t-Test
The HR department claims that the average beginning salary (salbegin) for all employees is higher
than $15,000
SPSS Steps: Analyze -- Compare Means-- One-Sample T Test. (Test Value = 15000).
Questions
i. State the null and alternative hypothesis
ii. State the t-statistic and the p-value.
iii. Is the result significant at 5% level of significance?
iv. Do you reject or fail to reject the null hypothesis?
Exercise 2: Independent Samples t-Test
Determine if there is a significant difference in current salary (salary) based on gender (gender).
SPSS Steps: Analyze -- Compare Means-- Independent-Samples T Test. (Grouping variable: gender).
Questions
i. State the null and alternative hypothesis
ii. Report the means for both groups.
iii. Look at Levene’s Test. Do you assume equal variances?
iv. What is the p-value (Sig. 2-tailed)? Interpret the finding.
Part 3: Paired Samples tests
Create a new dataset with two variables Pre_Test (Score) and Post_Test (Score) using the following
data
ID Pre_Test (Score) Post_Test (Score)
1 65 70
2 70 75
3 55 60
4 80 78
5 60 68
6 72 74
7 50 60
8 68 72
Questions
i. What is the 95% Confidence Interval for the mean difference? Interpret this interval.
ii. State the Null and Alternative hypotheses.
iii. Locate the t-statistic, degrees of freedom and -value (Sig. 2-tailed).
iv. Based on a 5% significance level, do you reject or fail to reject the null hypothesis?
v. What is your conclusion in the context of the training program?
Part 3: One-Way ANOVA
Analyze if the current salary (salary) differs significantly based on employment category (jobcat).
SPSS Steps: Analyze -- Compare Means -- One-Way ANOVA. (Dependent: salary; Factor: jobcat). Use
Post Hoc (Tukey) if necessary.
Questions
i. State the null and alternative hypothesis
ii. Report the F-statistic and p-value
iii. Is there a significant difference?
iv. Based on the Post Hoc tests, which groups are significantly different from each other?
Part 4. Estimation and Hypothesis testing about Proportion
Download the dataset: Marketing_Survey.sav in SPSS software
This dataset contains responses from 150 customers regarding a new product launch.
Aware: 1 = Aware of product, 0 = Not aware
Purchase: 1 = Purchased, 0 = Did not purchase
Gender: 1 = Male, 0 = Female
Questions
1. The marketing team wants to estimate the true proportion of all customers who are aware of
the new product with 95% confidence.
I. What is the sample proportion of customers aware of the product?
II. Construct a 95% Confidence Interval for the population proportion.
III. Interpret the Confidence Interval in the context of the data.
2. Historically, 60% of customers are aware of new products. The marketing team believes the
awareness for this new product is different from the historical average.
i. State the null and alternative hypothesis
ii. What is the exact significance level (p-value) provided by SPSS?
iii. At the 5% level of significance level, do you reject or fail to reject the null hypothesis?
iv. Is there significant evidence that the awareness rate is different from
3. A company is testing a new website layout (Treatment) against the old layout (Control) to see if
it increases the click-through rate (Yes/No).
Group Outcome (Yes) Outcome (No) Total
Treatment (1) 45 55 100
Control (2) 30 70 100
Questions
i. Calculate proportions and the 95% Confidence Interval for the difference between two
proportions.
ii. Test if the new layout significantly improves the click-through rate at 5% level of
significance.
Part 5: Chi-Square Test of Independence
Examine if there is a relationship between gender (gender) and job category (jobcat).
SPSS Steps: Analyze -- Descriptive Statistics-- Crosstabs. (Rows: gender, Columns: jobcat). Click
'Statistics' and check 'Chi-square'.
Questions
i. State the null and alternative hypothesis
ii. What is the Pearson Chi-Square value and the p-value (Asymp. Sig)?
iii. What is your conclusion?
Part 6. Correlation and Simple linear regression
Open SPSS data called [Link]
i. Investigate the relationship between MPG, Horsepower, Weight, and Acceleration.
ii. What is the Pearson correlation coefficient (r) between Weight and MPG? Is it positive
or negative?
iii. Is the relationship between Horsepower and MPG statistically significant at 5% level of
significance?
iv. Which variable has the strongest relationship with MPG?
Use the Analyze > Regression > Linear path in SPSS to solve the following questions
Predict MPG based on the weight of car
i. Set variables : Dependent : MPG, Independent : weight
ii. In the "Model Summary" table, what is the R Square value? Explain what this value
means in the context of this study.
iii. In the "ANOVA" table, is the regression model statistically significant?
iv. In the "Coefficients" table, look at the unstandardized β coefficient for Weight. Write the
regression equation (Y =a + bx)
v. Interpret the slope β of weight on MPG