Sampling
BENEFITS
Disadvantages
This is called
• They study consumer behavior, sustainability, and branding. Surely
they’ll give me the most useful feedback.
• So I decide that instead of asking random people, I’ll only collect data
from marketing students.
Disadvantage
• “It’s biased.”
• “You’re missing out on real consumers who aren’t marketing
students.”
• “The results won’t represent the whole market.”
Advantage
• “They’ll understand the questions better.”
• “They may give deeper insights about eco-friendly products.”
• “It saves you time because you don’t ask people who don’t know
much.”
‘Can you please share this survey with a few of
your friends who also care about eco-friendly
products?’
Advantages
• It’s useful when studying special groups that are hard to reach.
• Saves time and effort — one person leads you to many.
• You get respondents who are relevant to your topic.
Disadvantages
• It’s likely to be biased because people tend to refer friends who
think like them.
• The sample may lack diversity — mostly eco-minded people, not
the broader market.
• It won’t represent the whole population of young consumers.”
Non-Probability Sampling
Simple Random Sampling
• 200 JGU STUDENTS
• 40 at random
Advantages
• “It’s fair.”
• “There’s no bias.”
• “Every student had the same chance of being chosen.”
Disadvantage
• “It takes more effort to collect all the names.”
• “It can be time-consuming.”
• “You might still get an uneven mix by chance.”
Systematic Sampling
• Alright, let’s imagine this. I want to collect feedback on my green
cosmetics idea from JGU students. I know the total number of
students is 200, but I only need a sample of 50.
• So, how do I do this in a systematic way?
• First, I divide the total population by the sample size:
200 ÷ 50 = 4.
• That means I will select every 4th student from the list
Advantages
• “It’s easy to calculate and follow.”
• “You don’t need to draw names from a bowl every time.”
• “It ensures you spread across the entire list.”
Disadvantage
Stratified Sampling
Advantages
• It makes the sample more representative.”
• “You can compare groups more fairly.”
Disadvantages
• Selecting a wrong Strata
Cluster Sampling
Disadvantage
• “If one section has mostly eco-conscious students, the results will be
skewed.”
• “You might miss diversity.”
Start with your Population
• Population: People familiar with eco-friendly products
• Sampling Frame: students with age less than 25 years
• Sampling frame splits your population
Type 1 and Type 2 errors
Meet Skyview Towers
• - A high-rise apartment complex with a fancy new fire alarm system.
• - The alarm is supposed to ring only when there’s a fire.
• Hypotheses:
• - Null Hypothesis (H₀): There is no fire.
• - Alternative Hypothesis (H₁): There is a fire.
Type 1 Error (False Positive): The Midnight
Panic!
• - One night, at 2 AM, the fire alarm suddenly blared loudly!
• - Residents ran outside in their pajamas! Firefighters rushed to the scene!
• - But… there was NO fire!
• - The culprit? A faulty sensor mistook steam from a shower for smoke!
• Conclusion:
• - The system wrongly rejected the null hypothesis (assumed there was a
fire when there wasn’t).
• - This is a Type 1 Error (False Positive).
Type 2 Error (False Negative): The Silent
Danger!
• - One evening, Mrs. Patel left a towel near her stove while cooking.
• - It caught fire, and smoke began filling her kitchen!
• - The problem? The fire alarm ring!
• Conclusion:
• - This is a Type 2 Error (False Negative).
A Dating App Disaster
• super-advanced AI to match people based on personality, interests,
and past relationship trauma
• Null Hypothesis (H₀): Two people are NOT a good match.
• Alternative Hypothesis (H₁): Two people are a good match.
• Emma, a book-loving, introverted cat mom
• "Adventurer. Skydiver. Lives on the edge. Loves loud parties and
extreme sports. If you're into quiet nights, I'm not your guy!"
• But then... a glitch happened. The App mistakenly matched them
anyway!
• Hey babe, wanna go bungee jumping on our first date?"
What type of Error?
• Null Hypothesis (H₀): Two people are NOT a good match.
• Alternative Hypothesis (H₁): Two people are a good match.
• Type 1 Error (False Positive): The app incorrectly matches two people
who are NOT compatible (Emma and Jake, the extreme sports vs.
book-lover disaster).
Meet Bob and Sarah
• - Bob, a curious guy, decides to take a pregnancy test for fun.
• - Sarah, who has been feeling weird lately, takes a test too.
• - But what happens if the tests give wrong results?
• Hypotheses:
• - **Null Hypothesis (H₀):** The person is **not pregnant.**
• - **Alternative Hypothesis (H₁):** The person **is pregnant.**
Bob’s Shock!
• - Bob’s pregnancy test shows **positive!**
• - He panics and tells his friends he’s going to be a dad!
• - But wait… Bob **can’t even get pregnant!**
Type 2 Error (False Negative): Sarah’s
Surprise!
• - Sarah’s test says **negative**, so she goes on with life as usual.
• - A month later, she visits the doctor… **She’s actually pregnant!**
• - The test **failed to reject the null hypothesis** (assumed she
wasn’t pregnant when she was).
• **Conclusion:** This is a **Type 2 Error (False Negative).**