You’re at a maquis with friends, debating who makes the best attieké. How do you prove it?
Do you: A) Ask your 3 drunkest friends and call it a day? B) Taste-test every attieké stall in Abidjan (good luck)? C) Pick 50 random stalls, rate them 1-10, and crunch the numbers?
If you picked C, congrats—you’ve just designed a quantitative study. And no, it’s not just for lab-coat nerds. Whether you’re comparing phone prices at Plateau markets or measuring how many woro-woro drivers honk per hour, numbers tell stories. This quiz will test if you can spot the good, the bad, and the ugly in research.
1. The Backbone: What Even Is Quantitative Research?
Quantitative research is not about drowning in Excel sheets. It’s about answering questions with numbers you can analyze. Think of it like cooking kedjenou: you need the right ingredients (data), a solid recipe (method), and a way to taste-test (analysis).
Definition: > Quantitative research = Systematic collection of numerical data to test hypotheses, spot patterns, or predict outcomes. Key tools: surveys, experiments, stats.
Why bother? Because "most people prefer X" is a weak argument. "68% of 500 surveyed students prefer X (p < 0.05)"? That’s power.
2. Variables: The Cast of Your Research Story
Every study has characters—aka variables. Mix them up, and your research flops like a sauce graine without palm oil.
| Variable Type | Example | Role in Study |
|---|---|---|
| Independent | Price of brochette (500 vs 1000 FCFA) | What you change to test effect |
| Dependent | Customer satisfaction (1-10 scale) | What you measure for results |
| Confounding | Stall’s hygiene rating | Sneaky factor that messes up results |
Key point: > A good study controls confounding variables. Example: If testing *brochette* prices, ensure all stalls use the same meat quality—otherwise, you’re measuring *hygiène*, not *price*.
3. Sampling: How to Avoid Asking Only Your Cousins
Bad sample = Asking 10 rich kids at Lycée Blaise Pascal if school fees are "too high." Good sample = Randomly surveying 300 students across 5 public and private schools.
Sampling Methods: Pick Your Fighter
- Simple random: Names in a hat. Fair but tedious.
- Stratified: Divide by groups (e.g., 50% public school, 50% private).
- Cluster: Pick whole classes (not individual students).
- Convenience: Ask whoever’s near (aka lazy and risky).
Warning: > Small or biased samples = garbage results. Example: Surveying *only* *Yopougon* residents about Abidjan’s traffic tells you nothing about *Cocody*.
4. Data Collection: Surveys vs. Experiments (Fight!)
Surveys: The Gossip Method
- Pros: Cheap, fast. "Rate your allooco on a scale of 1-10."
- Cons: People lie ("I floss daily!"), or misread questions.
Example: > Bad question: *"Do you like bad food?"* (Leading + vague)
Good question: *"How often do you eat at *maquis* per week? (0/1-2/3+)"*
Experiments: The Mad Scientist Route
- Pros: Control variables. Example: Test if red vs. blue packaging sells more sachet water.
- Cons: Expensive. Hard to do for big topics ("Does corruption lower GDP?").
5. Stats 101: Mean, Median, and Why Your Prof Hates "Average"
You collected data. Now what? Descriptive stats summarize it; inferential stats test theories.
| Stat | What It Tells You | Example |
|---|---|---|
| Mean | "Average" (add all ÷ number) | Avg mobile money fee = 250 FCFA |
| Median | Middle value (50% above/below) | Half of taxi fares cost ≤1000 FCFA |
| Mode | Most frequent value | Most brochettes sold at 800 FCFA |
| Std. Dev. | How spread out the data is | Gbaka wait times vary by ±15 mins |
Formula: > Mean = $$ \bar{x} = \frac{\sum_{i=1}^{n} x_i}{n} $$
(Sum of all values ÷ number of values)
6. The "Oh No" Moment: 3 Mistakes That Ruin Studies
Warning: Common Pitfalls
1. Correlation ≠ Causation**: *"Ice cream sales rise with drowning deaths"* → Both happen in summer. One doesn’t cause the other!
2. Non-response bias: Only angry customers fill your survey. Now your data says *"90% hate us!"*
3. P-hacking: Tweaking stats until you get "significant" results. (Cheating.)
7. Quiz Time: Spot the Flaws!
Scenario: A student claims "80% of Ivorians prefer kedjenou over sauce claire" based on:
- Surveying 20 people outside a kedjenou restaurant in Bingerville.
- Asking: "Don’t you agree kedjenou is the best?"
What’s wrong? (Check all that apply)
- Sample size too small
- Sample location biased
- Leading question
- No control group
(Answer: All of the above. Ouch.)
8. Key Takeaways: Your Quantitative Cheat Sheet
Summary:
✅ Start with a clear question (e.g., *"Does X cause Y?"*).
✅ Pick the right variables—control the sneaky ones!
✅ Sample smart: Big enough + random = reliable.
✅ Use stats to tell the story, not just decorate it.
✅ Avoid bias like you avoid *zémidjan* scams.
9. Explore More on ORBITECH
Want to design flawless studies (or just ace your next assignment)? ORBITECH’s free research toolkit includes:
- Template surveys for Ivorian case studies
- Video breakdowns of stats (with attieké analogies, obviously)
- A sampling calculator to avoid tiny, biased samples
No textbook jargon—just tools that work. Check it out.