# MDA 5201 – MATHEMATICAL STATISTICS AND DATA ANALYSIS – THE KCA Masters Past Papers

UNIVERSITY EXAMINATIONS: 2019/2020
EXAMINATION FOR THE DEGREE OF MASTER OF SCIENCE IN
DATA ANALYICS
MDA 5201: MATHEMATICAL STATISTICS AND DATA ANALYSIS
DATE: MAY 2020 TIME: 6 HOURS

Take-Home Examination (THE)
QUESTION ONE: (15 MARKS)
A study to compare the cost of land per acre (in 1,000 Shilling units) between two suburban
regions of Nairobi revealed the following descriptive statistics:

a) State the hypothesis for this analysis.
b) Using a t-test, determine whether there are statistically significant differences in land cost
between Northside and Southside at the 0.05 alpha level. Include the critical value, the
observed test statistic, decision rule, and your decision.
c) Repeat the analysis in b) above using one-way analysis of variance (ANOVA)
d) Name at least two methodological challenges a research would encounter in conducting such
a study?
e) Describe a research design that can minimize the negative effects of these challenges.
QUESTION TWO: (15 MARKS)
For a study designed to investigate the relative effectiveness of three marketing strategies on
overall sales of an online product, stores were randomly assigned to three strategies. The 27
stores were assigned to Strategy A, 30 to Strategy B, and 30 to Strategy C. Overall sales of the
product were observed at the 87 stores at the end of the month. In a report comparing
differences in marketing effectiveness of each of these online strategies, the researcher writes
that a p-value of 0.041 (p= 0.041) was obtained in the test of the null hypothesis:

(a) Define precisely the meaning of the statement p = 0.041 in the context of this research.
(b) State an appropriate alternative hypothesis for the study
(c) Briefly describe a statistical model you could use to analyze data in order to test the
hypothesis in this study
(d) What other variable do you think should have been considered in this investigation? How
is it measured?
(e) Describe an appropriate statistical procedure you could use when the variable you have
names in (d) is considered in the data analysis.

(Visited 141 times, 1 visits today)