Skip to main content

Introduction to Research Methods

📹 Video Overview​

🎯 Why Study Statistics?​

The Big Question: Can you succeed without academic education?

Why anecdotal evidence ("I know a guy who...") fails:

  1. Small sample - "My uncle dropped out and is rich" = 1 person vs. millions

  2. Selection bias - You only hear success stories, not failures

  3. Confirmation bias - You remember stories that confirm what you believe

  4. Inaccuracy - Stories get exaggerated over time

💡 Memory hack: Think SSCI = Small Sample, Selection, Confirmation, Inaccuracy

📊 What is Statistics?​

Simple definition: Statistics = Using data to answer questions about large groups when you can't check everyone

Three main uses:

  • Collect data

  • Describe data (make it understandable)

  • Analyze data (draw conclusions)

Real-world example: Netflix can't ask all 200M users what they want to watch, so they use data from what people actually watch to recommend shows.

🍞 The Bread Example - Critical Thinking​

The slides show scary "facts" about bread:

  • 98% of criminals eat bread

  • Bread is "addictive"

  • Life expectancy was shorter when everyone ate homemade bread

What's wrong? These are correlations, not causations!

💡 Memory hack: Just because two things happen together doesn't mean one causes the other. Ice cream sales and drowning both increase in summer, but ice cream doesn't cause drowning!

Common statistical biases to watch for:

  • Unrepresentative sample

  • Poor quality data

  • Only people who care respond to surveys

  • Sample too small

  • Misleading graphs

  • Who paid for the research? (conflict of interest)

🔬 The 6 Stages of Statistical Research​

When studying the ENTIRE population:​

Example: "What's the average height of ALL students in our university?"

  • Population = All students in the university

  • If we can measure everyone → Use this process

When studying a SAMPLE (most common):​

Why use a sample?

  • Can't ask ALL customers what they think

  • Too expensive to test every product

  • Takes too long to survey everyone

Key terms:

  • Random sample: Everyone has equal chance of being picked (like lottery)

  • Representative sample: Sample looks like the population (same age mix, gender mix, etc.)

💡 Memory hack: A sample is like tasting soup - you don't eat the whole pot to know if it needs salt!

📈 Statistical Inference​

What is it? Making predictions about the WHOLE population based on your sample

Two main methods:

  1. Estimation - "Based on our sample, we estimate the average salary is $50,000"

  2. Hypothesis Testing - "We test if men and women have different average salaries"

Example:

  • Sample: Survey 1,000 voters

  • Inference: "Candidate A will get 55% of ALL votes"

🏷️ Variables - The Heart of Research​

Variable = Something you measure that changes from person to person

📋 Sorting Variables by Nature​

Main Division:​

TypeWhat it meansExamplesHow to remember
QualitativeCategories/names (can't do math)Gender, eye color, city, yes/noQUALITY = description
QuantitativeNumbers (can do math)Height, salary, age, test scoreQUANTITY = amount

Sub-division of Quantitative:​

TypeWhat it meansExamplesTest
DiscreteCountable numbers (1, 2, 3...)Kids in family, rooms in apartmentCan you have 2.5 of it? If NO → Discrete
ContinuousCan have decimalsHeight (175.3 cm), weight (68.7 kg)Can you have 2.5 of it? If YES → Continuous

💡 Memory hack:

  • Discrete = can count on fingers → separate values (2 kids, 3 kids, never 2.7 kids)

  • Continuous = flows like water → any value possible (can be 175.1 or 175.11 or 175.111 cm)

🎯 Sorting by Direction (Dependent vs Independent)​

Why this matters: When studying relationships between variables

Variable TypeAlso calledQuestion to askExample
DependentExplained variable"What am I trying to understand/predict?"Monthly salary
IndependentExplaining variable"What might affect/cause the dependent variable?"Years of education, hours worked

Real Examples:​

Study 1: "Does studying more hours increase test scores?"

  • Dependent (effect): Test score

  • Independent (cause): Study hours

Study 2: "Do students who submit exercises get higher scores?"

  • Dependent: Test score

  • Independent: Submitted all exercises (yes/no)

Study 3: "What affects monthly salary?"

  • Dependent: Salary

  • Independent variables: Education, experience, industry, city, gender

💡 Memory hack:

  • Dependent = DEPENDS on other things (like a baby depends on parents)

  • Independent = INDEPENDENT, does its own thing

⚠️ Important: The same variable can be dependent in one study and independent in another!

🧩 Practice Exercise​

Classify these variables:

VariableQualitativeDiscrete QuantitativeContinuous Quantitative
Car driving speed✓
Football player number✓
IDF rank (private, corporal...)✓
IDF rank (Private=1, Corporal=2)✓
Amount of precipitation✓
Country of birth✓
Temperature✓
Calendar year✓

Tricky ones explained:

  1. Football player number - Even though it's a number, you can't do math with it. Player #10 isn't "twice" player #5. It's just a label → Qualitative

  2. IDF rank vs coded rank:

    • Words (private, corporal) → Qualitative

    • Coded as numbers (1, 2, 3) → Discrete quantitative (now you CAN compare them mathematically)

  3. Calendar year - Discrete because you can't have year 2023.5. It jumps from 2023 to 2024.

🎓 Key Takeaways​

  1. Statistics helps us avoid being fooled by stories and small samples

  2. Samples must be random AND representative

  3. Critical thinking > just knowing formulas

  4. Variables are qualitative (categories) or quantitative (numbers)

  5. Quantitative splits into discrete (countable) and continuous (measurable)

  6. Always ask: What's dependent (effect) and what's independent (cause)?

📝 Quick Reference Chart​

Pro tip for exams: When in doubt about variable type, ask yourself:

  1. Is it a word/category? → Qualitative

  2. Can I add/subtract/average it meaningfully? → Quantitative

  3. Can it have decimals in reality? → Continuous, otherwise Discrete