Individual assignment for lab class; (20 %)
Analyze the following data using SPSS software and prepare your report on the Microsoft word.
Instruction to prepare your report prepares your report in Microsoft word and follows the following specific
formatting guidelines,
Cover page: font size 14, font: Times New roman, bold, uppercase, center text
Body text: font size 12, Times New Roman font, Alignment: use justify option
Spacing: 1.5 spaces between lines, but no indentation between paragraphs
Number: number Tables and provide explanatory title for the tables and interpretation below the table.
Number: number Figures and provide explanatory title for the Figures and interpretation below the Figures.
Deadline for submission your report: Dec 15/2025 submit your softcopy of report via tenati773@[Link].
1. Consider the [Link] data from SPSS.
a. Create the stacked bar chart for the variable level of education stacked by marital status and comment.
b. Investigate the association between marital status and level of education using Pearson chi-square. Your
answer must include: The hypothesis, value of the test statistics, p-value and Conclusion
c. Of those unmarried, what percent are graduate of high school degree?
d. Is there significant difference between average household income (in thousands) of married and
unmarried?(assume the distribution of household income is normal)
e. At α=0.05, test the claim that there is no difference in the average household income (in thousands) of
different level of education. (assume the distribution of household income is normal)
f. Test the claim that there is no difference among the average household income (in thousands) of five regions.
(assume the distribution of household income is normal)
g. Use the appropriate non parametric test for C, D, and E if the distribution of household income is not normal
and compare the result.
h. Develop a regression equation (written out) which explains the amount of the variation in Household income
in thousands based on the variables: Age in years, Level of education, Years with current employer,
Retired, Gender, Number of people in household.
State the estimated regression function and Interpret the results, Interpret the value of R and R-
square, Comment on the significance of the parameter estimate and list the predictors in order of
their better explanation of the dependent variable.
2. Refer the following data which represent persons in a class and consist of their name, sex,
weight, height, and age.
a. Construct scatter plot of height versus age and weight versus age;
b. Construct scatter plot of height versus age but differentiated (grouped) by sex;
c. Construct a bar graph of sex group;
d. Construct a box-and-whisker graph of weight across the sex group;
e. Construct a histogram to demonstrate the distribution of weight.
f. Compute appropriate summary statistics for SEX of students
g. Compute appropriate summary statistics for height of students
h. Compute
i. appropriate summary statistics for weight of students by gender
3. The following data are on the length of time (in months) between the onset of a particular illness
and its recurrence recorded for random sample of 50 patients at Arba Minch health center:
2.1,14.7,4.1,14.1,1.6,4.4,9.6,18.4,1.0,3.5,2.7,16.7,0.2,2.4,11.4,18.0,9.9,8.2,13.5,18,26.7,2.0,6.9,0.2,2
4,12.6,6.6,4.3,8.3,1.4,23.1,5.6,8.2,0.3,3.3,3.9,1.6,1.2,1.3,5.8, 4. 4
a. Summarize the data by using appropriate summary statistics
b. Construct box-plot and describe the distribution of data
c. Construct histogram with normal curve
4. The following data is on the amount of nutrients in a food (in mg per 100 gram)
before and after exposure to heat for a sample of n=10 food samples.
Amt before 81.2 66.7 68.6 71.6 67.9 65.9 48 51.5 68 65.7
Amt after 85.5 76.7 72.3 79.9 74 70.7 51.4 57 77.7 74.7
Test the hypothesis of no heat effect on the amount of nutrient in the food at 5% level of
significance.
5. In an experiment to compare rice varieties, six plots of each of four varieties
were grown, the plots being allocated to varieties in a completely random
manner and the results on the yield of rice crop are given below:
A B C D
25.12 40.25 18.30 28.05
17.25 35.26 22.60 28.05
26.42 31.98 25.90 33.20
16.80 36.52 15.05 31.68
22.15 43.32 11.42 30.32
15.92 37.10 23.68 27.58
a. Construct comparative box plot and compare variability of the yield
of rice crop by varieties.
b. Test of whether there is statistically significant difference on the
average the yield of rice crop by varieties.
c. If there is statistically significant difference on the average yield of
rice crop by varieties, then conduct post ANOVA test to identify
where the difference lies.
6. Consider the following data from n = 20 random sample of employees from
MOHA Soft drink manufacturing company collected to study the prediction of
monthly salaries (in 100 birr) of employees from the variables Experience( in
years) and age of employees. The variables are y =salaries of employee (in
100 birr), x1= Experience of employees (in years) and x2 = age of employees.
Subject Salary(y) Experience(x1) Age(x2)
1 38 1.47 40
2 58 8.00 48
3 80 9.00 53
4 30 0.00 30
5 50 0.00 50
6 49 1.00 49
7 45 4.00 45
8 42 0.00 42
9 59 3.00 38
10 47 0.00 47
11 34 3.00 34
12 53 0.00 53
13 35 1.00 35
14 42 2.00 42
15 42 2.00 42
16 51 7.00 51
17 51 8.00 51
18 40 3.00 30
19 48 1.00 48
20 34 7.00 34
21 46 2.00 50
Then,
a. Compute coefficients of correlation matrix between salaries,
experience and age of employees.
b. Fit multiple linear regression model that relates salaries of employees to the
predictors
c. Construct a normal probability plot of the residuals
d. Plot the residual versus fitted values
e. test for the significance of overall regression model
f. Give the highest variable correlation.
g. Give the value of the highest VIF.
h. In your opinion, is multicollinearity a possible problem in this
regression? Explain why it was/was not a problem, mentioning all
relevant tests.
i. In your opinion, is non-linearity a possible issue in this regression?
Explain briefly why you thought non-linearity was/was not an issue,
giving all relevant tests. If you think non- linearity might be an issue,
suggest possible solutions without going into too much detail.
j. Do you believe that there may be heteroskedasticity in the regression?
Explain your answer (why you do/do not believe that
heteroskedasticity exists in the regression or why you're not sure)?
k. Are the residuals normally distributed? Explain briefly why you do/do
not believe the residuals to be normally distributed.