Introduction to Research - State Question & Hypothesis

Logic of Reasoning
Deductive Inductive

General → Specific / Top-Down Approach Specific → General / Bottom-Up Approach

Starting with a general rule or fact and using Look at specific examples or patterns and
it to make a specific conclusion make a general rule based on them

Guaranteed conclusion Likely Conclusion

The Research Journey

The Research Question

Identify a research question that interests you personally with reliable sources
1.​ Derived from a research problem
2.​ Relevant & supported by previous findings and scholarly review
3.​ Guided by a theoretical framework
4.​ Follows the SMART principles

Research Methods - Determine Design Structure

Types of data (by source)

1.​ Primary Sources - Collected by the researcher
2.​ Secondary Sources - Collected by others

Research Designs & Research Framework

RD - The overall strategy or plan for conducting a study, detailing the methods and procedures
for collecting, analyzing, and interpreting data to answer a specific research question
RF - The conceptual structure that outlines the key variables, concepts, and their relationships,
providing a foundation for understanding the study and guiding its analysis.

Interviews
Interviews are a method of data collection that involves two or more people exchanging
information through a series of questions and answers
●​ Structured - Every interviewee gets the same questions
●​ Unstructured - Informal & Free flowing
●​ Semi-Structured
Utilize interviews when:
1.​ Topic is complex
2.​ Topic requires a lengthy explanation
3.​ A hypothesis is not yet available and exploration needs to be done

Experiments
A study in which an intervention is deliberately introduced to observe its effects

Randomized experiment - An experiment in which units are assigned to receive the treatment
or an alternative condition by a random process…
Quasi-experiment - An experiment in which units are not assigned to conditions randomly
Random Assignment - Any procedure that assigns units to conditions based only on chance ~
this heavily reduces the plausibility of alternative explanations by distributing such threats
randomly over conditions

Focus Groups
Structured discussion with a small group of people, run by a facilitator or using a moderating
team, to generate qualitative data on a precise topic of interest, using a set of open-ended
questions
●​ Discussion Panel of 6-10 people
●​ Participants are chosen based on demographics

Surveys & Experiments

Survey - A research design
 ●​ Cross sectional - data collected at one point in time
●​ Longitudinal - Data collected at different time points
Questionnaire - A data collection tool

Sampling & Questionnaire Design - Identify Population & Sample

Sampling
The process of identifying individuals for your study to generalize the population
!! If a sampling plan is flawed, increasing the sample size won't make the sample more
representative !!

Sampling Methods
1.​ Probability Sampling - A sampling method where every individual in the population
has a known and equal chance of being selected, ensuring representativeness.
a.​ Simple random sampling - Every individual in the population has an equal chance
of being selected
b.​ Stratified random sampling - The population is divided into groups (strata) based
on shared characteristics (e.g., age, gender), and a random sample is taken from
each group to ensure representation.
c.​ Cluster random sampling - The population is divided into clusters (e.g., cities,
schools), a few clusters are randomly selected, and everyone within the chosen
clusters is included in the sample.

2.​ Nonprobability Sampling - A sampling method where individuals are selected based
on non-random criteria, making it less representative but often easier and quicker to
conduct.
a.​ Convenience sampling - Selecting participants who are easiest to reach or
available
b.​ Purposive sampling - Choosing participants based on specific characteristics or
qualities relevant to the study, like selecting experts in a field.
c.​ Quota sampling - Setting quotas for specific groups (e.g., 50 males and 50
females) and selecting participants until each quota is filled, without
randomization.
d.​ Snowball sampling - Recruiting participants through referrals, where initial
participants suggest others, often used for hard-to-reach populations.

Questionnaire Design
!! Survey is NOT a Questionnaire !!
Questionnaire - A formal framework consisting of a set of questions/scales designed to generate
primary data
●​ Unstructured - open-ended, no predetermined answers offered → difficult to code
●​ Structured - closed-ended, predetermined responses or scale point → easier to code

Measurement - Design Instruments

Validity - Accuracy ~ do they really measure perceived usefulness?
Reliability - Consistency ~ Do repeated applications under similar conditions yield consistent
results?

Scale Types

1.​ Nominal - Data is divided into distinct categories without any order.
2.​ Ordinal - Data is categorized, and the categories have a meaningful order, but the
intervals between them are not equal.
a.​ Satisfaction levels, class + educational level & rankings etc.
3.​ Interval - Data has meaningful order, and the intervals between values are equal, but
there’s no true zero point.
4.​ Ratio - Similar to an interval scale, but it has a true zero point, which means you can
meaningfully multiply and divide values.

Descriptive & Inferential Statistics - Select Statistical Tests

Descriptive Statistics - Summarizes and describes data (e.g., averages, percentages) with
proper methods (graphs, tables, statistical values etc.)
Inferential Statistics - Using data to make predictions or finalize conclusions about a larger
group ~ stating conclusions with a certain level of probability

Quantitative Methods - Software

1.​ Excel (not the best due to formatting and problematic for data analysis)
2.​ SPSS
3.​ R/Python

Descriptive Statistics
In the context of business research, descriptive statistics are tools used to summarize and
understand the data collected from a study, making it easier to spot trends, patterns, and
insights.
 1.​ Frequency Tables
a.​ The frequency of a particular data value is the number of times the data value
occurs. A frequency table shows how often any value occurs
i.​ We use categorical (non-metric) data eg. Nominal & Ordinal
2.​ Measures of Central Tendency ~ Calculating Averages / Location Parameters
a.​ Mean - Determining the average
i.​ Best for metric data
ii.​ Sensitive to outliers
b.​ Median - Determining the middle value
i.​ Best for ordinal data
ii.​ Robust against outliers
c.​ Mode - Determining the most frequent
i.​ Best for nominal data

3.​ Measures of Dispersion (Data Spread)

a.​ Range - Difference between MAX and MIN
b.​ Variance & Standard Deviation - Tells us how spread out or scattered the data is
from the average (mean)
c.​ Interquartile Range - The middle 50% of all values ~Q3-Q1

Hypotheses
An unproven statement about a fact or phenomenon that is of interest to the researcher, more
specific than research question
●​ Often gives a possible answer to the research question
●​ Statement/proposition of relations
●​ Declarative
●​ Empirically testable
●​ Potentially falsifiable

Hypotheses Development
A Linear Process: Research Question → Review Literature → Formulate Hypotheses → Extract
Variables & Test Hypotheses

Types of Hypotheses - Null Hypotheses

The Null Hypothesis (H₀) is a statement used in statistics that assumes there is no effect, no
difference, or no relationship between certain variables in a study - H₀ is ALWAYS NULL
1.​ Relational - There is NO relationship…
2.​ Differential - There is NO significant difference…
Hypotheses comes in Pairs - Null & Alternative
1.​ Null
a.​ Comes Relational or Differential
2.​ Alternative
a.​ Comes Relational or Differential (Non-Directional / Two-Tailed)
i.​ You are not focusing on a direction, only for a difference
b.​ Comes Positive or Negative (Directional - One-Tailed)
i.​ You are focusing for a change in a certain direction eg. Positively or
Negatively correlated / higher or lower / more or less

Correlations
Shows a visual relationship of two variables
1.​ Non-Metric (Ordinal & Nominal) - Use Spearman
2.​ Metric (Interval & Ratio) - Use Pearson

Correlation Coefficient (r) - Determines both the intensity and direction of the relationship

Inferential Statistics
In business research, inferential statistics go beyond summarizing data; they help make
predictions or generalizations about a larger group based on a sample.

The P-Value
A number that helps you determine whether the results of your test are statistically significant.
●​ A small P-value (usually less than 0.05) suggests that the results you found are
unlikely to have occurred by chance, so you REJECT the null hypothesis and accept
the alternative hypothesis - The observed results is REAL and NOT by chance
●​ A large P-value suggests that the results could easily be due to chance, and you
RETAIN null hypothesis - The observed result can be attributed to Chance
Statistical Tests
1.​ Non-Parametric Tests - Uses data that is Non-Metric (Nominal & Ordinal)
a.​ Data is therefore NOT normally distributed
i.​ Eg. Mann Whitney U Test, Kruskal-Wallis Test, Chi-Square Test
2.​ Parametric Tests - Uses data that is Metric scale
a.​ Data IS normally distributed
i.​ Eg. T-Test, ANOVA etc.

Type 1 and Type 2 Errors:

If we are investigating merely a sample of the target population, it is possible to make one of
the following two errors
1.​ Type 1
a.​ The null hypothesis is rejected and the alternative hypothesis is accepted,
although in the population the null hypothesis is true
2.​ Type 2
a.​ The null hypothesis is retained, although the alternative hypothesis is true in the
population.

Publication Bias
The tendency on the parts of investigators, reviewers, and editors to submit or accept
manuscripts for publication based on the direction or strength of the study findings.
●​ Publishing studies with significant results over those without any
●​ Certain types of research findings are more likely to be published than others, leading to
an incomplete or distorted view of the evidence

Qualitative Approaches
Qualitative Data: Non-numerical information that describes qualities or characteristics
●​ Data Collection Methods
○​ Interviews, Observations, Focus Groups, Document Analysis etc.
●​ Data Types:
○​ Observed Behaviour Data
○​ Interviews
○​ Text Documents
○​ Digital & Audiovisual Materials

Qualitative Content Analysis

Deductive Inductive

Develop Categories based on Pre-Existing Develop categories based on your own data
Theory

Useful when there is a lot of host studies Useful when there is little to no previous
available studies available
Objectivity
The practice of ensuring that the findings, interpretations, and conclusions are free from
personal biases, emotions, or subjective influences
●​ Data is undistorted by the biases of researched
○​ Personal viewpoints should be controlled and irrelevant to the findings
●​ Relying on evidence, facts, and impartial methodologies to produce credible and reliable
results

Mixed Methods
The use of both Qualitative and Quantitative approaches

Designs:
1.​ Convergent Design - One Phase

2.​ Explanatory Sequential - Two Phase ~ QUANT!!!

3.​ Exploratory Sequential - Three Phase ~ QUAL!!!