DMS 411:

BUSINESS
RESEARCH
METHODS

Lecture VIII
1
2
 Scientific research process
Desktop
Field work & Define the problem research
report writing
=
Research
Review literature: proposal
Report writing Theory Vs practice
(findings)?

Formulate Hypothesis
Data analysis

Design research

Collect data

Tuesday, May 12, 2015 3

 Key questions addressed in methodology
• What design strategy will be used?

• Who/what is the study unit?

• Who/what forms the population of the study?

• Will sampling be done, and if so, what sampling design

will be used? What sample size?

• What kind of data will be used? Where /from whom

will it be collected? Which/what instrument will be
used and how will the researcher reach the
respondents/data source?

• How will the collected data be analyzed and

presented?
4
 Chapter Three: Research Methodology
• 3.1 Research Design

• 3.2 Population

• 3.3 Sampling and Sample Size

• 3.4 Data Collection

– Data type
– Data source
– Collection method
– Research Instrument
– Mechanism for reaching target
audience/object/subject

• 3.5 Data analysis

5
 Definition
• Data analysis is a body of methods that
help to describe facts, detect patterns,
develop explanations, and test hypotheses

• Data analysis is a process of gathering, modeling,

and transforming data with the goal of
highlighting useful information, suggesting
conclusions, and supporting decision making.

• Data analysis has multiple facets and approaches,

encompassing diverse techniques under a variety
of names, in different business, science, and
social science domains.

6
Data analysis … introduction
• But data analysis is not about numbers — it
uses them. Data analysis is about the world,
asking, always asking, “How does it work?”
And that’s where data analysis gets tricky.

• The data obtained from a study may or may

not be in numerical or quantitative form, or
both.

• Whichever form, one typically starts by

working out some descriptives to summarize
the pattern of findings.
7
INTERACTIVE PROCESS OF DATA ANALYSIS

Data Collection

Data display

Reflection on Data

ITERATIVE SIMULTANEOUS
Data Coding

Data distillation (reduction)

Generation of Themes

Story interpretation

Research Conclusions 8
 Steps to data analysis
• Once data is collected, some tasks / steps are
performed before data analysis is done.

• These tasks are done to ensure that the data

is appropriate, accurate and reliable (good
quality) for further analysis.

• These steps include:-

– Data editing/cleaning
– Data coding
– Data entry

9
 Data cleaning / editing
• During this stage, erroneous entries are
inspected and corrected.

• In some cases, it is easy to substitute suspect

data with the correct values. However, when it
is unclear what caused the erroneous data or
what should be used to replace it, subjective
decisions should NEVER be made. This may
undermine the quality of the data

• Data editing should be done as soon as data is

collected so that respondents may be
contacted for further information /
clarification.

10
• Common Data editing/cleaning activities
includes:-
– Checking for the accuracy of the data
– checking for the data consistency with other
information
– checking for uniformity of data
recording/capturing
– checking for completeness of the information
– checking for arrangement (to simplify coding
and tabulation)
11
 Data coding
• Coding involves assigning numbers or other
symbols to the various answers so that the
various responses obtained can be grouped
into a limited number of
classes/categories.

• The coding method used should be:-

– Appropriate to the research problem/and
purpose
– Exhaustive (cover all necessary aspects)
– Mutually exclusive (distinct and precise)

12
What is the purpose of measuring performance in Very
High Neutral Low Very low
your organization/business? high

Planning the future

Translating strategy into action plans

Ensuring the institution satisfies the needs of the

stakeholders
Increasing awareness of the organization for
advertising/marketing purposes.
Satisfying legal requirement regarding operating,

reporting and accounting for resources

Has your organization /business embraced management by performance

measurement? Yes [ ] No [
]

13
What is the purpose of measuring performance in 5 4 3 2 1
your organization/business?
Planning the future

Translating strategy into action plans

Ensuring the institution satisfies the needs of the

stakeholders
Increasing awareness of the organization for
advertising/marketing purposes.
Satisfying legal requirement regarding operating,

reporting and accounting for resources

Has your organization /business embraced management by performance

measurement? [1 ] No [ 0]

14
 Data Entry
• Data entry involves converting
information gathered (either from
secondary or primary sources) to a
medium for viewing and manipulation
e.g. entry/input into computer
programs
• Data entry option
– Optical scanning
– Voice recognition
– Keyboard
• Common errors in data entry?
15
 Data Analysis Methods
• The method of data analysis depends
mainly on
– Purpose of the study (e.g. descriptive,
analytical, explanatory, correlation, etc)

– Type of data/information collected

(qualitative and quantitative).

16
Qualitative
Data analysis

17
 Analyzing Qualitative data
• Qualitative data is analyzed using content analysis
method

• Applies statistical, linguistic, and structural

techniques to extract and classify information from
textual sources, a species of unstructured data.

• These may include:-

– Counting words/sentences
– Categorizing phrases
– Describing logical structures of expressions
– Ascertaining associations connotations
– Deducing the sociological or political interpretations etc.

• As such, a message can convey a multitude of count

outs

18
 The Qualitative Analytical Process
(Analytical thinking)

Components Procedures Outcomes

Data Reductions Description

Coding
Categorisation
Data Display Abstraction
Comparison
Dimensionalisation
Integration
Conclusions & Interpretation
Verification Explanation/
Interpretation

19
 Analyzing Quantitative data

Analytical Thinking/analysis processes:

– Standing back form the information given
– Examining it in detail from many angles
– Checking closely whether each statement follows logically from
what went before
– Looking for possible flaws in the reasoning, the evidence, or the
way that conclusions are drawn
– Comparing the same issues from the point of view of other writers
– Being able to see and explain why different people arrived at
different conclusions
– Being able to argue why one set of opinions, results or conclusions
is preferable to another
– Being on guard for literary or statistical devices that encourage
the reader to take questionable statements at face value
– Checking for hidden assumptions
– Checking for attempts to lure the reader into agreements

20
• Examples

21
Tuesday, May 12, 2015 22
23
 Primary school net enrolment rate or net attendance rate (2003–2008)

24
Quantitative
Data analysis

25
 Analyzing Quantitative data

• Quantitative data is mainly analyzed using

statistical data analysis.
• Analysis can be categorized by purpose. E.g.
– Descriptive statistics - Focuses on providing basic information
about the sample /population of interest

– Exploratory data analysis - Focuses on discovering new

features in the data. E.g. relationships, trends, etc

– Confirmatory data analysis - Focuses on confirming or

falsifying existing hypotheses (inferential statistics).

– Predictive analytics: - Focuses on application of statistical or

structural models for predictive/ forecasting or classification
26
 Analyzing Quantitative data

• Statistical data analysis is broadly

divided into two categories:-

– Descriptive statistics

– Inferential statistics

27
 Descriptive statistics
• Descriptive Statistics involves the development of certain indices
from the raw data for the purpose of description i.e. describing
the attributes/variables of the study unit.

• Descriptive statistics are also used to provide a feel of the data,

hence a preliminary step to inferential statistics and/or advanced
statistical analysis

• Common descriptive measures are clarified into four classes:-

• Measures of location (central tendency)

• Measures of spread/variability (measure of dispersion).

• Measures of normality

• Measures of relationship (correlation measures)

28
 Advertising
budget Sales
Advertising
Company budget Sales (000) (000) (millions)

1 26 73 Mean 26.75 76.42

2 22 69 Standard Error 0.95 2.94

Median 26.5 75.5
3 28 86
Mode 26 69
4 29 62
Standard Deviation 3.28 10.19
5 26 78
Sample Variance 10.75 103.90
6 25 69
Kurtosis -0.25 -1.19
7 22 67
Skewness 0.19 0.27
8 30 88
Range 11 32
9 27 81
Minimum 22 62
10 29 84 Maximum 33 94
11 33 94 Sum 321 917
12 24 66 Count 12 12

29
 Inferential statistics
• Also known as sampling statistics.
• Used to draw conclusions about a
population from sample information

• Divided into two:

– Estimation of population measures
(parameter)
– Testing of statistical hypothesis (e.g.
Z test, t test, X2 test, F-test etc)

30
 Hypothesis Testing
• What is a Hypothesis?

• Why is it necessary to have a

hypothesis?

• Hypothesis testing procedure

31
Hypothesis testing procedure
• State hypothesis.
– Both Ho and Hi

• State significance level (Confidence level)

– What is the allowable margin of error?

• State test statistic

– Which statistical test? Z test, chi square, F test, etc
– Identify relevant statistical table values for the test.
E.g. significance Z value, F value, etc

• State decision criterion

– When do you accept or reject the hypothesis?

32
Hypothesis testing procedure
• Compute statistical test
– Compute from available data

• Make statistical decision

– Invoke the decision making criterion, based on
comparison of table value and computed value
for the test statistic

• Make managerial decision

– What are the implication of the results? (i.e. so
what? – what does it mean to a layman)

33
Example of Hypothesis testing
• State hypothesis
Advertising Sales (000) – Ho: Advertising does not influence sales
– H1: Advertising affects sales significantly
26 73
22 69 • Test statistic R2
28 86 • Significance level = 0.05 (5% error
chance)
29 62
26 78 • Decision criterion
– If computed value is greater that
25 69 expected table value, reject Null
22 67 hypothesis
– Else, fail to reject
30 88 • Compute (perform test statistic)
27 81
29 84 • Compare & make statistical decision
33 94 • Managerial implication
24 66
34
 Regression Statistics
Multiple R 0.729679
R Square 0.532432
Adjusted R Square 0.485675
Standard Error 7.310211
Observations 12

ANOVA
df SS MS F Significance F
Regression 1 608.5248 608.5248 11.38724 0.007068
Residual 10 534.3918 53.43918
Total 11 1142.917

Standard
Coefficients Error t Stat P-value Lower 95% Upper 95%

Intercept 15.73432 18.10602 0.86901 0.405216 -24.6084 56.07706

Advertisement 2.268499 0.672248 3.374499 0.007068 0.770638 3.76636

35
 Types of statistical tests
• Regression Analysis
• ANOVA (F test)
• Chi-square (X2 test)
• Discriminant Analysis
• Factor Analysis
• Time series
• Run tests
• Correlation analysis
• Cross tabulations
36
 Pitfalls in data analysis
• Sample not been a representative of the
population in which you're interested.

• Lack of clear understand of the underlying

assumptions to method of analysis.

• Capacity /technical support - not too little,

not too much.

• Poor / weak measurement tools / approach

• Existence of multiple comparisons as well

as multiple interpretations

37
 Pitfalls in data analysis
• Objectivity – sticking to the study objectives;

• Use of numerical notation in a rational way--don't

confuse precision with accuracy (and don't let the
consumers of your work do so, either).

• Understanding the conditions for causal inference, If

there is need to make causal inference.

• Graphs should be accurate, well labelled and reflect

the data variation clearly.

38
• General questions and
answers?

• Lessons learnt?

• General comments?
39
Thank you

40