JOMO KENYATTA UNIVERSITY OF AGRICULTURE AND TECHNOLOGY

BACHELOR OF BUSINESS INFORMATION TECHNOLOGY

RESEARCH METHODOLOGY

ASSIGNMENT

VICTOR AMUTALLA MACHIO

HDB 212-0335/2019
1. Briefly explain the meaning and relationships among the following pairs of concepts.
a) Reliability and validity

Reliability is referred to as the constancy of a metric (whether the results can be reproduced
under the same conditions).

Validity is the term for a measure's precision (whether the results really do represent what they
are supposed to measure).

Concepts like validity and reliability are used to assess the caliber of research. They demonstrate
how accurately a methodology, method, or test measure something.

b) Significance level, Type I error, Type II error

Significance level is referred to as the chance of rejecting a null hypothesis that is actually true.
The Greek letter alpha (a) is commonly used to represent this value, which spans from zero (0.0)
to one (1.0). Type I error, often known as a false-positive, happens when a researcher rejects a
null hypothesis even when it holds true in the population. If a researcher accepts a null
hypothesis that is actually wrong in the population, they have made a type II error.

They are connected in a way that the Type II error rate is inversely proportional to statistical
power, which is affected by the significance level. This implies that Type I and Type II errors
involve a significant trade-off. Setting a lower significance level decreases Type I error risk, but
increases a Type II error risk.

2. Random sampling and random assignment are important concepts in empirical

research. Define each of these concepts and briefly explain their importance in
research.
Random sampling refers the technique you employ to choose participants from the population
for your study. To put it another way, random sampling refers to the process of selecting
participants for a study at random from the population. The technique you use to divide
individuals into groups in an experimental study is known as random assignment.

Because random sampling helps ensure that your sample is objective and representative of the
entire population, it improves the external validity or generalizability of your conclusions. You
can draw more reliable statistical conclusions as a result.
Random assignment improves the study's internal validity because it assures that there are no
obvious discrepancies between the participants in each group. You can draw the conclusion that
the independent variable is to blame for the results as a result of this..

3. One-way analysis of variance (ANOVA) is a common data analytic technique used in

quantitative research. Briefly describe a research problem where the one-way ANOVA
model could be appropriate to analyze data. Be sure to identify the variables in the
study and their scale of measurement.

A large-scale farm is curious to know which of three fertilizers results in the greatest crop output.
At the end of the growing season, they measure the overall yield after applying each fertilizer to
ten different fields.

Researchers can use a one-way ANOVA with "kind of fertilizer" as the factor and "crop yield" as
the answer to determine whether there is a statistically significant difference in the mean yield
that results from these three fertilizers.
We can conclude that there is a statistically significant difference in mean crop production
between the three fertilizers if the total p-value of the ANOVA is less than our significance level
(usually set to be 0.10, 0.05, or 0.01). We can then conduct post hoc tests to determine exactly
which fertilizer lead to the highest mean yield.
4. Explain the difference between “qualitative”, “quantitative”, and “mixed methods”
research and indicate the potential challenge each will present in a research project.

While qualitative research focuses on description and experiences, quantitative research deals
with figures and statistics. While qualitative approaches let you delve deeper into ideas and
experiences, quantitative methods let you measure variables and test hypotheses in a methodical
way. In a study using a mixed methods approach, both quantitative and qualitative data are
gathered and used.

Potential challenge of quantitative research

Data analysis might be challenging for researchers without statistical backgrounds because the
quantitative study calls for a thorough statistical examination. Since statistical analysis is based
on scientific methodology, it is challenging for non-mathematicians to complete.
 Potential challenge of qualitative research

Quantifying qualitative ordinal values into meaningful quantitative interval data is the primary
problem in qualitative research. This is crucial since it can often be challenging to draw
statistical inferences from non-numeric data when using statistical software.

Potential challenge of mixed method research

Mixed methods research is labor-intensive and time-consuming because it calls for gathering
and evaluating two different types of data. If distinct results are obtained from the two separate
data sources, the results of mixed methods study can also be challenging to comprehend .

5. What is commonly referred to as “negative findings” is not desirable in most research

situations. What is the meaning of the term “negative findings” and how can the
chances of its occurring be minimized in a study?

Negative findings are those that invalidate the hypothesis and negate the purpose of the study.
Bad outcomes frequently leave researchers unsatisfied, and there is a general bias against
disclosing negative results.

You can lessen it by developing a detailed research plan. When organizing a research study, keep
in mind that bias may exist in any step of the procedure. It could be beneficial to evaluate your
interview or survey questions with your team, as their diverse viewpoints can help you choose
the best path of action. Use parameters appropriate for your sort of research to reduce bias if you
use sampling to find participants.

6. Analysis of variance (ANOVA) is a common data analytic technique used in

quantitative research.
i) Briefly describe a research problem where the use of ANOVA model could be
appropriate to analyze the data.

A grocery chain is curious as to whether the three styles of ads have varied effects on mean sales.
The total sales for each store are measured at the end of the month after using each style of
advertisement at ten separate stores.
Researchers can use a one-way ANOVA with "kind of advertisement" as the factor and "sales"
as the response variable to determine whether there is a statistically significant difference in
mean sales between these three forms of adverts.
We can say that there is a statistically significant difference in mean sales between the three
types of marketing if the total p-value of the ANOVA is less than our criterion of significance.
The precise advertising that produce noticeably different outcomes can subsequently be
identified using post hoc tests.
ii) Explain how quantitative and qualitative research differ in terms of the process
of hypothesis testing.

Qualitative research uses observation and interviews to understand a phenomenon in its context
whereas quantitative research uses testing hypotheses to understand the causal or correlational
link between variables. Both research approaches have merit, albeit some study topics are more
suited to one than the other.

7. Many students and beginning researchers view the research proposal as unnecessary
and a waste of time. This is however not the case, as a research specialist. Discuss
the relevance of the research proposal before embarking on actual research.
i. It is done by the researcher outlining the purpose of his or her study, which communicates
one's intentions to participate in research.
ii. The researcher can prepare and evaluate the steps that will be taken in the study while
composing the proposal. It allows the researcher the chance to identify logical flaws, false
assumptions, and even issues that the study's aims and design are not sufficiently
addressing.
iii. A proposal gives the researcher the chance to talk about the research projects of other
people who have worked on comparable topics.
iv. A well-designed research proposal aids the researcher in avoiding time-consuming and
exhausting changes after the project begins.
v. A proposal can also offer the information needed to solve the issue, as well as how that
information would be obtained, handled, and analyzed.
 8. The Ugandan Oil Company is attempting to develop a reasonably priced gasoline
that will deliver improved gasoline mileage. As part of this development process, the
company would like to compare the effect of three types of gasoline (A, B, and C) on
gasoline mileage. For testing purposes, the company will compare the effects gasoline
types have on the mileage obtained by a Toyota Prado TX. The descriptive statistics
below are based on data from 30 vehicles that were randomly assigned to each of the
three gasolines types:

Descriptive Statistics

Dependent Variable: MILEAGE

GTYPE Mean N Std. Deviation
A 35.00 10 1.000
B 36.60 10 1.140
C 3500 10 1.100
Total 36.20 30 1.574

a) State a null hypothesis appropriate for data analysis in this study

The null hypothesis is that there is no difference in in the effects of gasoline A,B and C on
gasoline mileage.
b) Construct and complete an ANOVA Table for this analysis
GTYPE MEAN N STD VARIANCE
DEVIATION
A 35.00 10 1.000 0.100
B 36.60 10 1.140 0.114
C 3500 10 1.100 0.110
Total 36.20 30 1.574 0.078
c) Test the null hypothesis you stated above at 0.05 level of significance.

The null hypothesis is that there is no difference in gasoline mileage between the three types of
gasoline. The alternative hypothesis is that there is a difference in gasoline mileage between the
three types of gasoline.
The test statistic is calculated as follows:
F = (0.100 + 0.114 + 0.110) / 0.078 = 16.667
The critical value of F at 0.05 level of significance is 3.89.

d) Provide a brief interpretation of the findings of the analysis above.

Since the test statistic is greater than the critical value, we reject the null hypothesis. There is a
difference in gasoline mileage between the three types of gasoline.

e) Find the Coefficient of Determination in this analysis.

R2 = 1 - (0.078 / 0.100) = 0.221
This means that 22.1% of the variation in gasoline mileage is explained by the type of gasoline
used.