Statistics & Probability - Misconception buster
Statistics and Probability questions in UPCAT and other college entrance exams are designed to test your understanding of fundamental concepts, not just formula memorization. Many students lose valuable marks due to common misconceptions that seem logical but are mathematically incorrect. This guide identifies the most critical mistakes that cost students exam points and shows you how to recognize and avoid these traps. Understanding these misconceptions is crucial because Statistics & Probability questions often appear in both Mathematics and Quantitative Reasoning sections, making up 10-15% of total exam questions.
Common misconceptions
Mean, median, and mode are always close to each other in value
Exam impact
Students choose wrong answers when comparing measures of central tendency, especially in word problems about income distributions, test scores, or real estate prices.
The reality
Mean, median, and mode can be very different, especially in skewed distributions. In right-skewed data, mean > median > mode. In left-skewed data, mode > median > mean. Only in perfectly symmetric distributions are they equal.
Trap question
Question
The salaries of 7 employees in a small company are: ₱15,000, ₱16,000, ₱17,000, ₱18,000, ₱19,000, ₱20,000, ₱85,000. Which statement is correct?
The one very high salary (₱85,000) pulls the mean up significantly, but the median stays at the middle value. This is right-skewed data where mean > median. No value repeats, so there's no mode.
Wrong answer
Mean ≈ Median ≈ Mode (around ₱27,000)
Correct answer
Mean (₱27,143) > Median (₱18,000), Mode doesn't exist
Correct vs incorrect approach
Correct approach
First identify if data is skewed. For income data (usually right-skewed): median < mean. For test scores with a few very low scores (left-skewed): mean < median.
Incorrect approach
Student thinks: 'The mean is 50, so the median must also be around 50.' They don't consider the distribution shape.
Why students believe it
Students often see examples with normal distributions where these measures are similar, leading them to think this is always the case. They assume that all data sets follow a bell curve pattern.
Probability can be greater than 1 if an event is very likely to happen
Exam impact
Students make arithmetic errors in probability calculations and don't catch them because they don't know the basic constraint that P ≤ 1.
The reality
Probability is ALWAYS between 0 and 1 (inclusive). P(E) = 0 means impossible, P(E) = 1 means certain. If you get a value > 1, you made a calculation error.
Trap question
Question
A bag contains 3 red balls and 2 blue balls. What's the probability of drawing a red ball OR a blue ball?
Since every ball is either red or blue, you're certain to draw one or the other. P(red or blue) = 1. Any answer > 1 indicates a calculation mistake.
Wrong answer
P(red) + P(blue) = 3/5 + 2/5 = 5/5 = 1, but they write 1.2 due to calculation error
Correct answer
1 (or 100%)
Correct vs incorrect approach
Correct approach
Use complement: P(at least one H) = 1 - P(no heads) = 1 - P(TT) = 1 - (1/2)(1/2) = 1 - 1/4 = 3/4 = 0.75
Incorrect approach
Student calculates P(getting at least one head in 2 coin tosses) = P(H on first) + P(H on second) = 1/2 + 1/2 = 1. Thinks this is correct.
Why students believe it
Students confuse probability with odds or percentages. They might think 150% chance means probability = 1.5, or they add probabilities incorrectly.
In permutations and combinations, order never matters
Exam impact
Students consistently choose the wrong formula, leading to answers that are off by factors of r! (which can be very large).
The reality
Permutations (nPr) count arrangements where order matters (like ranking 1st, 2nd, 3rd). Combinations (nCr) count selections where order doesn't matter (like choosing team members). Use the context to decide which applies.
Trap question
Question
A committee must elect a President, Vice President, and Secretary from 8 candidates. How many ways can this be done?
These are different positions (President ≠ Vice President), so order matters. Use permutations: 8P3 = 8!/(8-3)! = 8×7×6 = 336.
Wrong answer
8C3 = 56 ways
Correct answer
8P3 = 336 ways
Correct vs incorrect approach
Correct approach
The word 'arrange' means order matters. Book A-B-C-D-E is different from E-D-C-B-A. Use 5P5 = 5! = 120 ways.
Incorrect approach
Problem: 'How many ways to arrange 5 books on a shelf?' Student thinks: 'Just selecting 5 books, so use 5C5 = 1'
Why students believe it
Students memorize 'combinations are when order doesn't matter' but forget that permutations are specifically when order DOES matter. They use combinations formula for all counting problems.
The range includes all the numbers between the minimum and maximum
Exam impact
Students give intervals as answers instead of single numbers, or they identify the maximum and minimum correctly but fail to subtract them.
The reality
Range is ONE number: the difference between the highest and lowest values. Range = Max - Min. It measures the spread of data, not the interval itself.
Trap question
Question
The test scores in a class are: 65, 78, 82, 89, 95. What is the range of these scores?
Range = Maximum - Minimum = 95 - 65 = 30. The range is always a single number representing the spread, not an interval.
Wrong answer
65 to 95
Correct answer
30
Correct vs incorrect approach
Correct approach
Range = 20 - 3 = 17. The range is the single number 17, representing how spread out the data is.
Incorrect approach
Data: 3, 7, 12, 15, 20. Student answers: 'Range is from 3 to 20' or 'Range is 3, 4, 5, ..., 20'
Why students believe it
Students confuse 'range' (a single number) with 'span' or 'interval'. They think range means 'from 5 to 15' instead of '10'.
If two events are independent, they cannot happen at the same time
Exam impact
Students incorrectly set P(A and B) = 0 for independent events, or they use wrong formulas for calculating joint probabilities.
The reality
Independent events can absolutely happen together. Independence means P(A|B) = P(A) - one event doesn't change the probability of the other. Mutually exclusive events cannot happen together: P(A and B) = 0.
Trap question
Question
The probability of rain tomorrow is 0.3. The probability of having a quiz tomorrow is 0.4. These events are independent. What's the probability of both rain AND a quiz tomorrow?
Independent means the rain doesn't affect the quiz probability and vice versa. They can definitely both happen. P(rain and quiz) = P(rain) × P(quiz) = 0.3 × 0.4 = 0.12.
Wrong answer
0 (because they're independent, so they can't both happen)
Correct answer
0.3 × 0.4 = 0.12
Correct vs incorrect approach
Correct approach
If A and B are independent, then P(A and B) = P(A) × P(B). They CAN happen together; their occurrence just doesn't influence each other.
Incorrect approach
If A and B are independent, student thinks P(A and B) = 0 because 'they can't happen together'
Why students believe it
Students confuse 'independent' with 'mutually exclusive'. They think independence means the events don't affect each other AND cannot occur together.
Mean = Mode = Median for all data sets
Exam impact
Students assume they can find one measure and automatically know the others, leading to wrong answers in problems asking for specific measures of central tendency.
The reality
Mean = Median = Mode only in perfectly symmetric distributions (like normal distribution). In most real-world data, these three measures are different and provide different insights about the data.
Trap question
Question
A data set has mean = 40. If the data is right-skewed, which statement is most likely true?
In right-skewed data, the tail extends to the right, pulling the mean up. The order is: Mode < Median < Mean. So if Mean = 40, then Median < 40 and Mode < Median.
Wrong answer
Median = 40 and Mode = 40
Correct answer
Median < 40 and Mode < Median
Correct vs incorrect approach
Correct approach
Calculate each measure separately. Only conclude they're equal after verifying the data is perfectly symmetric.
Incorrect approach
Student calculates mean = 25, then assumes median = 25 and mode = 25 without checking the actual data distribution.
Why students believe it
Students see this relationship in symmetric distributions and generalize incorrectly. They memorize this as a universal rule rather than understanding it applies only to specific distribution shapes.
Sample size doesn't matter for accuracy - a small sample is just as good as a large one
Exam impact
Students make wrong choices about which survey or study is more reliable, or they don't understand why larger samples are preferred in statistical inference questions.
The reality
Larger samples generally provide more accurate estimates of population parameters. The margin of error decreases as sample size increases. Small samples have higher variability and less reliability.
Trap question
Question
Two polls about election preferences: Poll A surveyed 100 randomly selected voters, Poll B surveyed 2000 randomly selected voters. Both show Candidate X leading by 5%. Which is more reliable?
While both use proper random sampling, Poll B's larger sample size means smaller margin of error and more reliable results. The 5% lead is more meaningful with n=2000 than n=100.
Wrong answer
Both equally reliable since both used random sampling
Correct answer
Poll B is more reliable due to larger sample size
Correct vs incorrect approach
Correct approach
Larger samples reduce sampling error. The study with 5000 people will have a much smaller margin of error and more reliable results than the 50-person study.
Incorrect approach
Student thinks: 'Both studies used random sampling, so the study with 50 people is just as reliable as the one with 5000 people.'
Why students believe it
Students think that as long as sampling is random, size is irrelevant. They don't understand the concept of sampling error or margin of error.
Correlation always implies causation
Exam impact
Students make incorrect conclusions about cause-and-effect relationships in data interpretation questions or research scenario problems.
The reality
Correlation only shows that two variables tend to change together. Causation requires proof that changes in one variable directly cause changes in the other. There could be hidden variables, reverse causation, or pure coincidence.
Trap question
Question
A study shows that students who eat breakfast regularly have higher test scores than those who don't. What can we conclude?
While there's a correlation, other factors could explain this: family income (affecting both nutrition and educational resources), health consciousness, or morning routine discipline. Correlation doesn't prove causation.
Wrong answer
Eating breakfast causes higher test scores
Correct answer
There's a correlation between breakfast eating and test scores, but causation isn't proven
Correct vs incorrect approach
Correct approach
Both are caused by a third variable (hot weather). Hot weather increases swimming (more drowning risk) and ice cream consumption. No direct causal relationship between ice cream and drowning.
Incorrect approach
Student sees: 'Ice cream sales and drowning incidents both increase in summer' and concludes: 'Ice cream causes drowning'
Why students believe it
Students see strong correlations and immediately assume one variable causes the other. This seems logical - if two things happen together, one must cause the other.
Standard deviation is always less than the mean
Exam impact
Students eliminate correct answer choices or doubt their calculations when standard deviation exceeds the mean, especially in problems with small means or high variability.
The reality
Standard deviation can be larger than, smaller than, or equal to the mean. It depends on the data's variability and scale. There's no mathematical constraint relating standard deviation to the mean.
Trap question
Question
A data set has mean = 3 and standard deviation = 5. This means:
This is perfectly possible. A standard deviation larger than the mean indicates high variability. For example, data like 0, 0, 1, 2, 15 has mean ≈ 3.6 and standard deviation ≈ 6.1.
Wrong answer
There's an error in calculation because standard deviation cannot exceed the mean
Correct answer
The data is highly variable with some values far from the mean
Correct vs incorrect approach
Correct approach
Standard deviation measures spread around the mean. If data is highly variable, standard deviation can exceed the mean. Both calculations can be correct.
Incorrect approach
Student calculates standard deviation = 15 and mean = 12, then thinks 'This can't be right because standard deviation should be smaller than mean'
Why students believe it
Students work with data sets where this happens to be true and think it's a mathematical rule, similar to how probability is always ≤ 1.
In sampling, bigger populations always need bigger samples
Exam impact
Students make wrong choices about appropriate sample sizes or think surveys of large populations are automatically less reliable if the sample size seems 'small' relative to population.
The reality
Sample size depends more on desired precision and confidence level than population size. A sample of 1000 can represent 10,000 people just as well as 10 million people, assuming proper random sampling.
Trap question
Question
Which survey is more reliable: 1000 random people from Metro Manila (pop. 13M) or 1000 random people from Quezon City (pop. 3M)?
Sample size, not sample proportion, determines reliability. Both samples are large enough to provide similar margins of error regardless of population size.
Wrong answer
Quezon City survey because the sample is a higher percentage of the population
Correct answer
Both surveys have similar reliability if properly randomized
Correct vs incorrect approach
Correct approach
Sample size depends on desired margin of error. A random sample of 1000 from either city can provide similar reliability if properly conducted.
Incorrect approach
Student thinks: 'To survey Metro Manila (13 million people), we need a much larger sample than to survey Baguio (350,000 people)'
Why students believe it
Students think sample size should be proportional to population size, like taking 10% of any population regardless of its size.
Quick self-check
If the mean of a data set is 50, then the median must also be approximately 50
Answer
False
Mean and median are only equal in symmetric distributions. In skewed distributions, they can be very different.
A probability of 1.2 is impossible because probabilities cannot exceed 1
Answer
True
Probabilities are always between 0 and 1. Any calculation giving P > 1 contains an error.
If two events are independent, they cannot occur at the same time
Answer
False
Independent events can occur together. You're thinking of mutually exclusive events.
The range of a data set is the difference between the maximum and minimum values
Answer
True
Range = Maximum - Minimum. It's a single number measuring data spread.
For any counting problem, if order doesn't matter, use combinations; if order matters, use permutations
Answer
True
This is the key distinction: combinations for selection, permutations for arrangement.
Correlation between two variables proves that one causes the other
Answer
False
Correlation shows variables change together, but doesn't prove causation. Other factors may be involved.
Larger samples are generally more reliable than smaller samples
Answer
True
Larger samples reduce sampling error and provide more accurate estimates.
Standard deviation must always be less than the mean
Answer
False
Standard deviation can be larger than the mean if data is highly variable.
Diagrams
Probability Calculation Decision Tree
Decision tree showing how to approach different types of probability problems and avoid the common mistake of getting probabilities greater than 1
Central Tendency Misconceptions Map
Mind map showing the relationships between measures of central tendency and how distribution shape affects their relative positions
Permutation vs Combination Choice Guide
Decision flowchart to help students choose between permutations and combinations based on whether order matters in the problem context
Common Statistics Exam Question Types
Distribution of different types of statistics questions commonly found in UPCAT and other entrance exams
In summary
The biggest trap in Statistics & Probability is assuming patterns that work in simple examples apply everywhere. Remember: (1) Probability NEVER exceeds 1 - if you get >1, you made an error, (2) Mean, median, mode are only equal in symmetric data - check the distribution shape, (3) Use permutations when order matters (arrangements), combinations when it doesn't (selections), (4) Range is ONE number (max-min), not an interval, (5) Independent events CAN happen together - don't confuse with mutually exclusive, (6) Correlation doesn't prove causation - look for other explanations. Always verify your answers make mathematical sense and match the problem context. These misconceptions cost students 15-20 marks on average in entrance exams.