POLI 397 QUIZ 1

What are the three types of research methods?

quantitative, qualitative, mixed methods

Quantitative

• Uses numerical data and statistical analysis, usually with many observations (large n).

  • Example: Surveying 1,000 students and analyzing their responses with statistics.

Qualitative

• Uses words, meanings, and detailed descriptions, usually with fewer observations.

  • Example: Interviewing 20 students about their experiences.

Mixed-Methods

• Combines quantitative and qualitative methods in one study.

  • Example: Surveying 500 students and then interviewing 20 of them to understand their answers better.

What Is a Research Question?

• a specific question that your research aims to answer.

• Should not be too general.

• Must be researchable and testable.

• Narrows your research and keeps it focused.

• Determines your research design and how you collect data.

• Helps explain what you are trying to find out.

What makes a research question good?

1. Answerable (you specifically)

2. Hasn't already been answered

3. Not based on assumptions

4. Not tautological (circular/redundant)

5. Ellicits a puzzle

What makes it difficult to make a RQ?

• regardless of the answer, it makes a contribution

• needs to be unknown (not defitively answered)

• requires familiarity with preexisting literature

Purpose of Literature Review

• demonstrates the contribution of your research question

• grounds your research by supporting any required premise (starting point for a rq)

  • e.g.: What has caused democratic backsliding in Indonesia?

    • prove that democratic backsliding has occurred in Indonesia

Types of RQ

Descriptive, Explanatory, Prescriptive, Normative, Predictive

Descriptive RQ

• “What is happening?” (correlation)

  • Example: Is there a relationship between social media use and students’ grades?

Explanatory RQ

• “Why did something happen?” (causation)

  • Example: Does increased social media use cause students’ grades to decrease?

Prescriptive RQ

• “What should we do?” (policy)

  • Example: Should schools limit students’ social media use during school hours?

Normative RQ

• “What is best?” (theory)

  • Example: Is limiting social media use the best way to improve students’ academic performance?

Predictive RQ

• “What will happen?” (AVOID, cannot be falsified)

  • Example: Will students’ grades decrease in the future because of increased social media use?

What is a hypothesis?

• An assumption, suspicion, assertion, or idea about a phenomenon, relationship, or situation

• Statement (not a question) driven by your research question

• Basis of inquiry (how research starts)

• Evidence will either support or refute your hypothesis (meaning it’s falsifiable)

• Basis of your argument

MISCONCEPTIONS ABOUT HYPOTHESES

• Not every research question needs a hypothesis.

• Only some RQs have hypotheses → Usually used when testing relationships, effects, or differences.

  • Only quantitative research uses hypotheses → Myth. Qualitative research can also use hypotheses.

WHAT IS A THEORY?

• A hypothesis that has been elaborated and/or has withstood repeated testing

• Simplifies reality to a few relevant factors

• Generalizable

Inductive approach

• Starts with data

• Observations > identify patterns > develop theory

• Exploratory research (hypothesis development)

Deductive approach

• Starts with theory

• Tested and known theory > hypothesis > data

• Confirmatory research (hypothesis testing)

WHAT MAKES A HYPOTHESIS GOOD?

1. It answers the research question

2. It can be tested with evidence (measurable)

3. It can be falsified (there is a null hypothesis)

4. It makes a contribution

5. It’s specific

• If the RQ asks to explain something, but the hypothesis describes a relationship, the hypothesis is not doing its job

Descriptive Hypothesis

X has A, B, C characteristics. X behaves in D, E, F ways. If X increases, it is likely that Y will increase.

Explanatory Hypothesis

X causes Y. X exists because of A, B, C.

Predictive Hypothesis

If X happens, Y will happen.

Prescriptive Hypothesis

In order for X, we need to do A, B, C.

Normative Hypothesis

X is better than Y.

Variables

• A concept or factor that can change (i.e., vary) – it can assume different values

  • E.g., economic development – low, medium, high

  • E.g., form of government – democracy, autocracy, monarchy

Binary variables

• can take only one of two distinct values

  • E.g., landlocked countries – yes or no

Dependent Variables (DVs)

The outcome we wish to understand (depends on the IV)

Independent Variables (IVs)

• BUT not all research questions will have an IV and a DV

• Influences/affects the DV (what changes)

  • Needs to come before the DV in time and space

What do descriptive hypotheses often focus on?

The relationship between the IV and DV.

Positive Relationship between IV and DV

• both variables move in the same direction

  • IV increases, DV increases | IV decreases, DV decreases

Negative Relationship between IV and DV

• variables move in opposite directions

  • IV increases, DV decreases (or vice versa)

Positivism

• identifies patterns, relationships, explanations, and causal effects

  • Investigated systematically using observable evidence

  • Lends itself to variables, measurement, statistics

  • Example: Do territorial disputes increase the likelihood that two states will go to war?

Interpretivism

• seeks to grasp the meaning behind actions

  • Contextual understanding

  • Human-centric, placing experiences, emotions, and cultural context at its center

  • Example: How do states construct and interpret territorial disputes as threats to national identity?

What type of questions does positivism focus on?

Questions that measure and test relationships between variables.

What type of questions does interpretivism focus on?

Questions that understand meanings, interpretations, and experiences.

What are Concepts

• Conceptual definitions → Clearly define what important terms in your research question or hypothesis mean.

  • Ask: Does the term mean what we intend it to mean?

  • Ask: Will other people understand the term the same way?

• Example: “Colonialism’s impact on racialized identity” → You need to clearly define what colonialism and racialized identity mean.

• Conceptual stretching → Using a concept too broadly so that it loses its original meaning.

Example of Conceptual Stretching

• Concept: Democracy

• Original meaning: A system where people have a role in choosing their government.

• Conceptual stretching: Calling any country with an election a “democracy,” even if people have very limited political participation or choice.

• This stretches the concept so broadly that it becomes less useful.

Easy way to remember: Conceptual stretching = making a concept so broad that it starts losing its meaning.

OPERATIONALIZATION

• Operational definitions → Explain how you will measure a concept in the real world.

  • Example: Define war as a conflict with a certain number of battle deaths.

  • You must decide what counts as “significant” (e.g., a specific number of deaths).

Proxy Measurements

• A measurable variable used to represent something that is difficult or impossible to measure directly.

• Example: Intelligence is difficult to measure directly, so IQ test scores can be used as a proxy for intelligence.

• Easy way to remember: Proxy = something you use to stand in for something else.

Conceptualization vs. Operationalization

• Conceptualization = Deciding what a concept means.

• Operationalization = Deciding how to measure that concept.

• Example: Intelligence

  • Conceptualization: What do we mean by “intelligence”? → The ability to learn, reason, and solve problems.

  • Operationalization: How will we measure intelligence? → Use an IQ test score.

Necessary components of a research design

Part	Meaning	Example (idea)
Research question	What you want to find out	“Does study time affect grades?”
Hypothesis	Your predicted answer	“More study time increases grades.”
IV (Independent Variable)	What you change/manipulate	Study time
DV (Dependent Variable)	What you measure as the outcome	Grades/test scores
Conceptualization	The “meaning” of the variables (concepts)	“Study time = hours studying” (in concept)
Operationalization	How you measure/define variables (data/instruments)	“Study time measured by a survey/log of hours”

Internal Validity

• Internal validity = Can we be confident that the IV caused the DV?

• We need to rule out alternative explanations for the result.

• Example: Education increases the likelihood of voting.

• We need to make sure education caused the difference in voting, rather than something else such as income, age, or political interest.

🧠 Remember: Internal validity = Did the IV really cause the DV?

External Validity

• External validity = Can the results of a study be generalized to other people, places, or situations?

• Asks: Can your conclusions apply more widely?

• Example: Can a lab experiment tell us something about the real world?

• Example: Can interviews in Calgary tell us something about Canada as a whole?

🧠 Remember: External validity = Can the results apply outside the study?

Descriptive Statistics

• Descriptive Statistics describe (or summarize) data - offer generalizations

  • help you see the general pattern or characteristics of the data without trying to explain why something happened.

  • Example:

    You survey 100 students about how many hours they study per week.

    • Average study time = 10 hours

    • Most students study between 7–12 hours

    • 60% study more than 8 hours

Mode

the value that occurs most often

Example: 5', 6', 6', 4' → mode = 6'

Median

the middle value when the data is arranged from smallest to largest.

Example: 3', 4', 4', 5', 6' → median = 4'

Mean

the mathematical average. Add all the values and divide by the number of values.

Example: 5, 6, 7 → (5 + 6 + 7) ÷ 3 = 6.

EAMPLE: List of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10

• Outlier = a value that is extremely high or low compared to the rest of the data.

• Q1 (first quartile) = median of the lower half of the data.

  • Lower half: 1, 2, 3, 4, 5

  • Q1 = 3

• Q3 (third quartile) = median of the upper half of the data.

  • Upper half: 6, 7, 8, 9, 10

  • Q3 = 8

• IQR (Interquartile Range) = Q3 − Q1

  • 8 − 3 = 5

• Lower bound = Q1 − 1.5(IQR)

  • 3 − 1.5(5) = −4.5

  • Values below −4.5 are outliers.

• Upper bound = Q3 + 1.5(IQR)

  • 8 + 1.5(5) = 15.5

  • Values above 15.5 are outliers.

• Statistical outliers = values below −4.5 or above 15.5.

• In your list 1–10, there are 0 outliers because every value falls between −4.5 and 15.5.

Variance

• = measures how spread out the data is from the mean.

  • It gives us a numerical summary of how much the values differ from the average.

  • Variance = spread

Standard deviation

• = the square root of the variance.

  • It tells you approximately how far the data points are from the mean.

  • It uses the same units as the original data, making it easier to understand.

  • typical distance from the mean

• Categorical variables

• = puts data into different groups or categories.

• examples: nominal, ordinal, binary

Categorical Variables: Nominal

• categories that have no meaningful order or ranking.

• The categories are simply different from each other.

• You cannot rank them from least to greatest.

Examples:

• Religion: Muslim, Christian, Hindu, Non-Religious

Categorical Variables: Ordinal

• categories that have a meaningful rank or order.

• However, the distance between categories is not necessarily equal.

• You can say one category is higher or lower than another, but you cannot say the difference between them is exactly the same.

Example: Likert scale:

• Strongly agree

• Somewhat agree

• Neither agree nor disagree

• Somewhat disagree

• Strongly disagree

Categorical Variables: Binary

• Binary = a categorical variable with only two possible categories/options.

• The two options are usually opposites or two distinct choices.

Examples:

• Yes / No

VARIABLE TYPES - NUMERICAL

• Numerical variables = variables that have equal unit differences.

• A 1-unit increase always means the same amount.

  • Example: The difference between 20 and 21 is the same as between 50 and 51.

VARIABLE TYPES - NUMERICAL: Continuous

• Can take an infinite number of possible values within a range.

• Can include decimals and fractions.

• Examples:

  • Age: 20, 20.5, 20.75 years

  • Temperature: 20°C, 20.5°C, 20.55°C

  • Time: 5 minutes, 5.5 minutes, 5.55 minutes

  • Income: $20,000, $20,000.50, etc.

👉 Continuous = can be measured and can have decimals.

VARIABLE TYPES - NUMERICAL: Discrete

• Consists of whole, countable numbers.

• You cannot have fractions/decimals of the thing being counted.

• Examples:

  • Number of people: 1, 2, 3, 4...

  • Number of pets: 0, 1, 2, 3...

  • Number of countries visited: 1, 2, 3...

👉 Discrete = countable whole numbers.

VISUALIZING DATA

• Categorical data

  • Use bar graphs

  • Avoid pie charts

  • Don’t manipulate the Y-axis

  • Use logical, equally spaced scales

  • Use complementary colours

  • Use real data

Visualizing Data — Continuous Data

• Histogram: Shows the frequency of numerical data across different ranges.

• Kernel density plot: Shows a smooth estimate of how data is distributed across a range of values.

• Box-and-whisker plot (box plot): Shows the distribution of numerical data, including the middle, spread, and possible outliers.

intervening variable

A mediating (intervening) variable acts as the middle link or mechanism through which the independent variable causes a change in the dependent variable. In contrast, a moderating variable does not transmit the effect itself, but instead changes the strength, direction, or intensity of that relationship.