Section A:

1. Explain the difference between statistics and probability in your own

words.
Statistics is more about looking at data that already exists, collecting it, organizing it,
summarizing it and trying to make sense of it. So if I am given a list of exam scores and I find
the average, that's statistics. Probability however, it's about predicting what might happen before
it actually happens. It's the chance or likelihood of an event occurring. So for example, before the
exam results come out, calculating the chance that a student passes based on past exam results
would be probability.
2. Why is the median considered insensitive to outliers compared to the
mean? Provide an example.
The median is simply the middle value once the data has been arranged in order, so it doesn't
really matter how large or small some of the numbers are, the median just shows the value sitting
in the centre position. While the mean involves adding up every single number and dividing by
how many there are, so one unusually large number can make the whole result look inconsistent
For example, five labourer earn ₦50,000, ₦55,000, ₦60,000, ₦65,000 and one person earns
₦1,000,000. If you calculate the mean, you would get about ₦246,000, which doesn't represent
what a typical labourer earns at all. But the median would be ₦60,000, which gives a much more
realistic picture of the wages. That's why the median is preferred whenever there's an outlier in a
data set
3. Identify one real-life scenario where the mode is the most appropriate
measure of central tendency.
I think a good example would be a shoe company trying to decide which shoe size to produce in
the largest quantity. The mean wouldn't really help here because you could end up with
something like size 8.7, which doesn't exist. The median wouldn't be too useful either. What the
company actually needs is the size that occurs most often among customers, and that's exactly
what the mode gives you.
4. Describe the role of standard deviation in understanding data
variability.
Standard deviation basically tells you how spread out the values in a dataset are compared to the
mean. If the standard deviation is small, it means most of the data points are close to the mean
(so there's not much variation), and if it's large, it means the values are scattered further away
from the mean (more variation).
This is useful because two datasets can have exactly the same mean but behave very differently.
For instance, two students could both average 70 in their tests, but one student might score 68,
70, 71, 72 (very consistent), while the other scores 30, 60, 90, 100. The mean alone wouldn't
show this difference, but the standard deviation would, since the second student's standard
deviation would be much higher. So basically, it helps us understand consistency.
5. Differentiate between descriptive statistics and inferential statistics with
examples.
Descriptive statistics is mainly about summarizing and describing the data that you already have
in front of you, things like the mean, median, mode, range, charts and graphs. It doesn't try to go
beyond the data to make any predictions. For example, if I take the heights of all 30 students in
my class and calculate the average height, that's descriptive statistics.
While Inferential statistics, uses information from a sample to make conclusions or predictions
about a much larger population. For example, instead of asking all 5,000 students in a university,
a researcher might survey 200 of them about their opinion on a topic and then use that result to
estimate what the opinion of the whole student population is likely to be. So descriptive statistics
describes what's already there, while inferential statistics tries to make an educated guess about
something bigger based on a smaller sample.
Section B:
1. Calculate the mean, median, and mode of the dataset: 12, 15, 20, 20, 25,
30, 35
12 + 15 + 20 + 20 + 25 + 30 + 35 = 157
Mean = 157 ÷ 7 = 22.43
Median = 20
Mode = 20
2. For the dataset: 5, 7, 9, 12, 15, 18, 20, 22, 25
• Find the five-number summary (minimum, Q1, median, Q3, maximum).
 • Construct a simple box plot sketch based on your results.
There are 9 values here, and they're already sorted, so this makes things a bit easier.
Minimum = 5
Maximum = 25
Median = since there are 9 numbers, the middle one is the 5th value, which is 15.
For Q1: 5, 7, 9, 12. Since there are 4 values here, Q1 is the average of the two middle ones,
which are 7 and 9.
Q1 = (7 + 9) ÷ 2 = 8
For Q3: 18, 20, 22, 25. The two middle values are 20 and 22.
Q3 = (20 + 22) ÷ 2 = 21
So the five-number summary is:
Minimum = 5, Q1 = 8, Median = 15, Q3 = 21, Maximum = 25
Box plot sketch
Box Plot (Min=5, Q1=8, Median=15, Q3=21, Max=25)

5 8 15 21 25
|--------[==========|===========]--------|
Min Q1 Median Q3 Max
3. A dataset has values: 10, 12, 15, 18, 20
• Compute the variance and standard deviation step by step.
Step 1: Find the mean.
Mean = (10 + 12 + 15 + 18 + 20) ÷ 5 = 75 ÷ 5 = 15
Step 2: Subtract the mean from each value to get the deviations.
10 - 15 = -5
12 - 15 = -3
15 - 15 = 0
18 - 15 = 3
20 - 15 = 5
Step 3: Square each of these deviations.
(-5)² = 25
(-3)² = 9
(0)² = 0
(3)² = 9
(5)² = 25
Step 4: Add up all the squared deviations.
25 + 9 + 0 + 9 + 25 = 68
Step 5: Divide by the number of values to get the variance.
Variance = 68 ÷ 5 = 13.6
Step 6: To get the standard deviation, take the square root of the variance.
Standard deviation = √13.6 ≈ 3.69
Section C:
1. Two datasets have the same mean and median:
• Dataset A: 199, 200, 201
• Dataset B: 0, 200, 400
Discuss why measures of dispersion are necessary to compare these datasets.
If I just calculate the mean and median for both datasets, I would get the same answer for each,
mean = 200 and median = 200 for both A and B. So on the surface, it looks like these two
datasets are basically the same. But clearly, they are not. Dataset A has values that are all tightly
packed around 200 (199, 200, 201), while Dataset B has values that are wildly spread apart (0,
200, 400).
This is exactly why measures of central tendency alone aren't enough to fully describe a dataset.
We also need measures of dispersion, like range, variance, and standard deviation, to know how
spread out or how consistent the data actually is. For Dataset A, the range is only 2 (201 - 199),
but for Dataset B, the range is 400 (400 - 0). If these represented something like daily sales
figures or temperatures, knowing the spread would matter a lot, because Dataset A would
suggest something very stable and predictable, while Dataset B would suggest something
extremely volatile, even though "on average" they look identical.
 2. Provide a real-world example of a left-skewed distribution and a right-
skewed distribution. Explain how the skewness affects the mean and
median in each case.
A real-world example of a right-skewed distribution would be income or salary data in most
countries. Most people earn somewhere within a moderate range, but a small number of very
wealthy individuals earn extremely high amounts. These few very high values stretch the tail of
the distribution out to the right. Because of this, the mean ends up being pulled higher than the
median, since the mean is sensitive to those extreme high earners, while the median stays closer
to where most people actually fall. So in a right-skewed distribution, mean > median.
A good example of a left-skewed distribution would be something like the age at which people
retire, or scores on a very easy exam where most students score quite high (say in the 80s and
90s) but a few students score really low, maybe due to absenteeism or not preparing at all. These
low scores stretch the tail out to the left. In this case, the mean is pulled down below the median,
because those few low scores drag the average down even though most of the data is bunched up
at the higher end. So for a left-skewed distribution, mean < median.
3. A researcher wants to study the effect of study hours on exam scores.
• Write the simple linear regression equation that could be used.
• Interpret what the slope and intercept would represent in this context.
The simple linear regression equation for this would be written as:
Y = a + bX
Where Y represents the exam score (the dependent variable, i.e. what we're trying to predict),
and X represents the number of study hours (the independent variable, i.e. what we're using to
make the prediction). "a" is the intercept and "b" is the slope.
Interpretation: The slope (b) tells us how much the exam score is expected to change for every
additional hour of study. So if b = 5, it would mean that, on average, each extra hour of studying
increases the exam score by about 5 points (assuming everything else stays the same).
The intercept (a) represents the predicted exam score when study hours = 0, that is, the score
someone might be expected to get even if they didn't study at all. In real life this number might
not always make perfect sense (it could even be a value that seems unrealistic), but
mathematically it's just the point where the regression line crosses the Y-axis, and it's needed to
complete the equation.
Section D: Research & Reflection
1. Choose one sampling method (e.g., stratified, cluster, convenience).
Explain how it works and give a practical example of when it would be
used.
Stratified sampling.
The way it works is that the entire population is first divided into smaller groups, called "strata,"
based on some shared characteristic, and then a random sample is picked from each of these
groups proportionally. The idea behind it is to make sure that every important subgroup is
properly represented in the final sample, rather than relying on pure chance which might leave
some groups out completely.
A practical example would be a university wanting to find out students' opinions about the
quality of hostel accommodation. Instead of just picking 500 students randomly from the whole
school (which might end up being mostly final year students, or mostly one faculty), the
university could divide the students into strata based on their level of study (100 level, 200 level,
300 level, 400 level), and then randomly select a proportional number of students from each
level. This way, the opinions of every level are properly captured, and the results would be more
representative of the whole student body than a purely random or convenience sample would be.
2. Reflect on why outliers should be carefully handled in statistical analysis. Provide at
least two strategies for dealing with them.
Outliers can be really tricky because they have the power to distort the overall picture of a
dataset, especially when it comes to measures like the mean, variance, and standard deviation,
which we already saw in earlier questions are quite sensitive to extreme values. If outliers are
just ignored or left in without thinking about where they came from, conclusions drawn from the
data could end up being misleading. On the flip side, if they're removed carelessly without
proper justification, you could end up throwing away genuinely important information, like a
fraud case in financial data, or an unusually high reading that's actually pointing to a real
problem.
One strategy for dealing with outliers is to first investigate where they came from before doing
anything. Sometimes an outlier is simply a data entry error (like someone typing 1000 instead of
100), and if that's the case, it can either be corrected or removed because it's clearly a mistake,
not a real value.
Another strategy is to use measures that are naturally resistant to outliers when summarizing the
data, such as using the median instead of the mean, or using the interquartile range (IQR) instead
of the standard deviation. This way, even if the outlier stays in the dataset, it won't distort the
summary statistics as badly.