Welcome to

PSY 107 Introduction to Quantitative Methods II

First, let’s go over the Course Outline…
Quizzes
• The quizzes will be conducted on HLMS.
• You will be given 48 hours to complete the quiz (this will also serve as
your attendance for the class)
• You can access the quiz through “Quiz” section in HLMS.
Quiz 1 (cont)
5. Choose / State the description that fits you best regarding your
expectations of PSY 107.
A. I expect to fail, so why bother?
B. I just wanna pass. I can’t really be bothered to put in much effort, so I’ll just do the
minimum.
C. I want to do well without having to put in the effort. You see, I’m entitled and
expect things to come easy. When I don’t do well, it’s really my lecturers’ fault –
not mine. And because I’m smart, I would also get good group mates who will pick
up my slack for me. Cheating is totally acceptable cos it fulfils my agenda.
D. I want to do well and I recognise that it requires hard work on my part. I am
committed to fulfilling my responsibilities for the class and I’ll take the initiative to
seek help when I do not understand the materials. Also, because I am a person of
integrity, I will not engage in dishonest conduct regardless of the consequences
that I would have to face.
Recap of the Research Process, Research Questions,
Hypotheses, Scales of Measurement, and Operational
Definitions
Lecture Learning Outcomes
By the end of this lecture, students should be able to:
1. Describe the research-related activities and decisions associated to each step
in the research process.
2. Explain the principle of science and the approaches to take for a study to be
“scientific”.
3. Explain the characteristics of a good research question.
4. Explain the purpose of forming research question.
5. Explain the meaning and functions of hypothesis, variables, scales of
measurement, conceptual definition, and operation definition.
6. State the type of information an informative operational definition should
contain.
7. Explain the benefits and restrictions of objectively operationalizing the
variables in research.
 Steps in the Research Process
Formulate
research question

Literature search
Formulate
testable
hypothesis Proposal
Determine the
research design
Determine
appropriate
measures Conduct the research
Ethical Review
Analyze the data
Determine the
appropriate
population
Report the
findings
Identify a
problem/idea
 Steps in the Research Process
**Repeat process if necessary**
Literature search
Implication

Identify a Formulate Formulate

testable Determine the
problem/idea research question research design
hypothesis

Determine Determine the

Analyze the data Conduct the research appropriate appropriate
measures population

Report the Proposal Ethical Review

findings
 Principles of Science

Empirical Objective, systematic observation

Rational
Testable

Falsifiable
Parsimonious
Tentative

Rigorously
Evaluated
Note on Objective (Systematic) Observation
• Variables can be operationally defined based on facts and logic (supported by
existing, updated scientific studies)
• Measurements are reliable and valid (Re: Lecture on Reliability and Validity)
• Procedures can be reproduced
• Results are replicable provided that the exact procedures are rerun, and the
characteristics of the samples are comparable
 Principles of Science
Empirical Objective, systematic observation
Rational Logical, consistent with known facts
Testable Can be systematically measured, scrutinised, and
replicated
Falsifiable Possibility for claim(s) to be unsupported by evidence
Parsimonious Explains behavior with fewest number of assumptions
Tentative Open to modification (even abandonment) as new
evidence arises
Rigorously Constantly evaluated for consistency with the above
Evaluated principles
Note on Rigorous Evaluation
• i.e. Replicable
• Results from rigorously evaluated studies are informative only when:
• The assessment(s) are appropriate (i.e. reliable and valid for the population and context being
studied)
• The assessment(s) are comprehensive
The Scientific Method
1. Objectivity
2. Replication
3. Self-correction
4. Control
Forming Research Questions
 Research Question

Qualities of a Good Research Question

• Empirical
• Answerable with the scientific method
• i.e. Enables structured, objective observations under specifically defined conditions
• Specific
• Variables are clearly stated, topic is focused
 Research Question

Qualities of a Good Research Question

• Ask important questions
• Justify the time and expenses involved
• Provide answers that serves practical implications
• Supports one of the many competing theoretical views

NOTE: Good research are not defined by novelty alone. Good

research are those that adds value.

Value = Novel + Rigorous

 Research Question

Phrasings (word choices) are important too

Are people who refuse to kill an innocent even when it is for the greater
good better social partners compared to people who are willing to
sacrifice one to save many?

Are people who refuse to kill an

innocent even when it is for the There’s a looot of
greater good perceived as being grey area…
better social partners compared
to people who are willing to
sacrifice one to save many?
Formulate Hypothesis
= provides tentative answers to the research question
 Hypothesis

Hypothesis / Hypotheses
• Tentative answer to the research question (an educated guess)
• Based on prior knowledge acquired from past research (rational; objective)
• Must be testable and falsifiable
• Direction of hypothesis must be coherent with the theoretical
framework
• Note: Hypotheses are conceptually different from “predictions”
Variables
 Variables

Variables
• Anything that varies/changes
• E.g. event, situation, behaviour, or individual characteristics
• An item, factor, or condition that can be subject to change and can
take on different values
• Should be phrased as the constructs themselves; nouns (different from
OD, different from predictions)
• E.g. cognitive performance vs. performance on the Stroop task
• E.g. cognitive performance vs. total number of items remembered
Scales of Measurement and
Operational Definitions
 Scales of Measurement

Scales of Measurement of Variables

Scales of measurement: categorizing variables according to
how they are measured
• 4 major types:
• Nominal (= name)
• Ordinal
• Interval
• Ratio
 Scales of Measurement

Nominal Ordinal
• No meaningful numeric attached • Variables with ≥ 2 categories, ranked in a
• Variables that have 2 or more categories, meaningful order
but cannot be arranged in any particular • No value attached to intervals between
Categorical order ranks

Examples: Gender, color, type of food Examples: Socioeconomic status, level of

study, preferences
Interval Ratio
• Variables that have no true zero • Variables that have a true zero (zero =
• Similar to ordinal, but values fall on a absence of quantity)
continuum and are equally spaced • Ratio of values on the scale are
Continuous meaningful
Examples: Temperature in Celsius /
Fahrenheit, Likert scales Examples: Height, weight, time
 Scales of Measurement

Importance of Identifying Scales of

Measurement
1. Determines display of frequency distribution (e.g. histogram, scatter plots).
2. Guides the statistical test to use (e.g. interaction effect using ANOVA vs. PROCESS
Macro; Pearson’s r vs. Chi-Square; more in future semesters).
3. Increase accuracy (i.e. better representation of real-life situation of the data)
4. SoM of a variable depends on how it is operationally defined. BUT,
• nature of the constructs themselves MUST be taken into consideration.
• suitability of a SoM has to be verified by valid theory (literature support at this stage).
 Scales of Measurement

Additional Notes
• Ideally = continuous
• More accurate (statistically)
• E.g. low and high self-esteem vs. self-esteem along a continuum
• Consider the nature of your data in the real world
• E.g. extroverts and non-extroverts vs. extraversion along a continuum
• Is extraversion an all-or-none trait?
• E.g. Gender; Biological sex (not practical to be continuous)
• Sometimes it works both ways (continuous and categorical), this is
when you need to consider the research question as a whole (take
into account the DV and purpose of study)
Always back up your decision with existing (credible) theories
Conceptual and Operational Definitions
 Conceptual Definition

• Ideal = supported by peer-reviewed scientific articles

• Avoid definitions from dictionaries (e.g. Cambridge, Merriam
Webster, etc.).
• Make sure conceptual definitions reflect the arguments in the
theoretical framework

Variables Biological Sex Gender

Social constructs / influence;
Basis of conceptual definition Genes / Heredity
VS. self-perception
Basis for theoretical
? ?
framework
 Operational Definition

• All variables need to be operationally defined (shows objectivity;

measureable)
• Operational definition
• Defining a variable in concrete, measurable terms
• Done when designing the research
• Indicate also what a high/low score/number means
• e.g. Total time taken (in seconds) to complete the Stroop Task, where a longer time
taken indicates weaker cognitive inhibition.

Type of score Context/Materials

Meaning
 Operational Definition

1. Type of score
• Total?
• Average?
• NOTE: NOT total average / average total !!!
2. Context / Materials
• Who is the person administering (scoring) the behaviour?
• Self-report?
• Confederate?
• E.g. Total number racist comments uttered by participants during the
experiment as recorded by the confederate, where more racist comments
uttered indicates higher level of racism.
• E.g. Total score on the Patient Health Questionnaire (PHQ-9) as rated by
the clinician, where a higher score indicates more severe level of
depression.
 Operational Definition

3. Meaning
• Indications in terms of the psychological construct being measured (i.e. the
variable name)
• E.g. “… higher score indicates higher level of cognitive performance.”
• NOT “… higher score indicates more items remembered from the memory
task.”

IMPORTANT NOTE: Think from the perspective of a reader who has no clue of the experimental
process. if you’re the reader, having no context of the details of the experiment, would you be
able to score the construct / behaviour accurately with the help of the operational definition? If
not, you might need to rephrase your OD.
 Operational Definition

Restrictions
• Only observable and measureable behaviours are included
• Only scientifically verified behaviours can be used as “objective
measurements”; or
• Measurements are limited to what the field of science knows up
until that point
• E.g. How to measure creativity?

NOTE: When operationally defining a variable, consider ALL the variables being
studied, and the conceptual definitions for all of them. Not just that one variable
itself.
Final Note:
Despite the restrictions, we still need to
operationally define the variables. that is
how we get to derive meaningful results
out of our studies. The fact that the
assessments for certain under-studied
constructs might not be comprehensive
further justify the NEED for rigorous
evaluation.
To Do List
• Quiz: due 25th August, 3PM
• Report Statistical Anxiety Scale result
• Form a group of 4-5 members → Register yourself in LMS
• Use Group Members Recruitment Center to look for group members.
• Design your ex post facto research!