Source: Fundamental Statistics Concepts for Data Analysis
Difficulty: Introductory | Prerequisites: None, though basic algebra helps.
Tags: cumulative relative frequency, standard deviation, variance, descriptive statistics, mean, median, dispersion, frequency distribution, data summary, STAT 101
This material covers the tools you use to describe and summarise a dataset before you start making predictions. Cumulative relative frequency tells you what proportion of observations fall below a given value. Standard deviation tells you how spread out those observations are. Descriptive statistics, as a category, is every technique that summarises data without attempting to generalise beyond it. If you are coming into this cold, these are the building blocks for everything else in the course, from probability through to hypothesis testing.
Descriptive statistics are how you summarise and organise raw data so it makes sense at a glance. Cumulative relative frequency shows you what fraction of your data sits at or below a value. Standard deviation measures how far, on average, individual data points fall from the mean.
Cumulative relative frequency
The sum of the relative frequencies for all values that are less than or equal to a given value. Think of it as a running total that tells you what percentage of your data sits at or below a particular point.
Standard deviation (s for sample, σ for population)
A measure of how spread out the values in a dataset are around the mean. In simple terms, it tells you whether your data points cluster tightly around the average or scatter widely.
Variance
The average of the squared differences between each data point and the mean. It is the square of the standard deviation. Think of it as the "raw" measure of spread before you take the square root to get back to the original units.
Descriptive statistics
Techniques that summarise and organise data so it is easier to understand, without making predictions about a larger group. In simple terms, this is everything you do to describe what the data looks like: averages, percentages, charts, tables.
Mean (average)
The sum of all values divided by the number of values. Think of it as the balance point of your dataset.
Dispersion
A general term for how spread out data values are. Standard deviation and variance are both measures of dispersion.
Relative frequency
The proportion of times a particular value or category appears, calculated as the frequency of that value divided by the total number of observations. Think of it as "what fraction of the whole does this value represent?"
Cumulative relative frequency builds on relative frequency by adding up each value's proportion as you move through an ordered dataset.
How it works: order the data from smallest to largest. For each value, sum the relative frequencies of that value and every value below it. The result is a running total that climbs from 0 toward 1 (or 0% toward 100%).
What it tells you: the proportion of observations that fall at or below a given value. If the cumulative relative frequency at a test score of 75 is 0.60, that means 60% of all scores are 75 or lower.
Where you see it: cumulative frequency tables, ogive charts (cumulative frequency polygons), and percentile calculations all rely on this concept.
Connection to percentiles: the cumulative relative frequency at a data point is, in effect, that point's percentile rank divided by 100.
Standard deviation is the most commonly used measure of spread. It tells you, in the same units as your data, how much individual observations typically differ from the mean.
Sample vs. population notation:
s is used for the sample standard deviation (divides by n − 1).
σ (sigma) is used for the population standard deviation (divides by N).
Relationship to variance: standard deviation is the square root of the variance. Variance is useful mathematically, but standard deviation is easier to interpret because it shares the same units as the original data.
Why n − 1 for samples: dividing by n − 1 (called Bessel's correction) compensates for the fact that a sample tends to underestimate the true population variability. This gives a less biased estimate.
Interpreting the number: a small standard deviation means data points cluster near the mean; a large one means they are spread out. In a roughly normal distribution, about 68% of values fall within one standard deviation of the mean, and about 95% within two.
Descriptive statistics is the branch of statistics concerned with summarising and presenting data. It does not attempt to draw conclusions beyond the data you have.
Measures of central tendency: mean, median, and mode. These answer the question "where is the centre of the data?"
Measures of spread (dispersion): range, interquartile range (IQR), variance, and standard deviation. These answer the question "how spread out is the data?"
Visual tools: histograms, bar charts, box plots, pie charts, and frequency tables. These let you see patterns, outliers, and the shape of the distribution at a glance.
What it does not do: descriptive statistics describes the dataset in front of you. It does not make predictions, test hypotheses, or generalise to a larger population. That is the job of inferential statistics.
\text{Population variance: } \sigma^2 = \frac{\sum (x_i - \mu)^2}{N}\text{Sample variance: } s^2 = \frac{\sum (x_i - \bar{x})^2}{n - 1}\text{Standard deviation: } \sigma = \sqrt{\sigma^2} \quad \text{or} \quad s = \sqrt{s^2}\text{Cumulative relative frequency at value } x_k = \sum_{i=1}^{k} \frac{f_i}{n}where f_i is the frequency of the i-th value and n is the total number of observations.
Quality control in manufacturing: factories track the standard deviation of product measurements (length, weight, voltage) to decide whether a production line is running within acceptable tolerances. A rising standard deviation signals that something in the process is drifting.
Grading on a curve: cumulative relative frequency is what lets an instructor say "you scored in the 85th percentile," meaning 85% of the class scored at or below your mark.
Finance and investment: standard deviation of returns is one of the most common measures of investment risk. A fund with a high standard deviation has more volatile, less predictable returns.
"Standard deviation and variance are the same thing." They are not. Variance is in squared units; standard deviation is its square root and shares units with the original data. You can compare a standard deviation directly to the mean, but you cannot do that with variance.
"You always divide by n when calculating standard deviation." Only for a population. For a sample, you divide by n − 1 (Bessel's correction). Using n for a sample systematically underestimates the true spread.
"A small standard deviation means the data is good." Standard deviation measures spread, not quality. A small standard deviation simply means the values are clustered. Whether that is desirable depends entirely on context.
"Descriptive statistics can tell you whether a result is statistically significant." It cannot. Significance testing belongs to inferential statistics. Descriptive statistics only summarises what you already have.
⚠️ Know the difference between sample (s, n − 1) and population (σ, N) standard deviation. Exams frequently test whether you use the correct denominator.
⚠️ Be able to interpret a cumulative relative frequency value in a sentence: "X% of observations fall at or below this value."
⚠️ Understand that descriptive statistics describes, inferential statistics predicts. This distinction appears in multiple-choice questions regularly.
⚠️ You may be asked to calculate standard deviation step by step: find the mean, subtract the mean from each value, square the differences, sum them, divide by n − 1 (sample) or N (population), then take the square root.
True or False: Standard deviation can be negative.
Answer: False. Standard deviation is always zero or positive.
Fill in the blank: The cumulative relative frequency of the largest value in a dataset is always ___.
Answer: 1 (or 100%).
True or False: Descriptive statistics can be used to test a hypothesis about a population.
Answer: False. That is inferential statistics.
Fill in the blank: Variance is measured in ___ units, while standard deviation is in ___ units.
Answer: Squared units; original units.
True or False: When calculating sample standard deviation, you divide by n.
Answer: False. You divide by n − 1.
Q: A dataset has values 2, 4, 4, 4, 5, 5, 7, 9. What is the mean?
A: (2 + 4 + 4 + 4 + 5 + 5 + 7 + 9) / 8 = 40 / 8 = 5.
Q: Using the dataset above, what is the cumulative relative frequency at the value 5?
A: Values at or below 5: 2, 4, 4, 4, 5, 5 = 6 values. Cumulative relative frequency = 6/8 = 0.75 (75%).
Q: Explain why sample standard deviation uses n − 1 rather than n in the denominator.
A: Using n tends to underestimate the population variance because the sample mean is closer to the sample data points than the true population mean is. Dividing by n − 1 (Bessel's correction) adjusts for this bias and produces a better estimate of the population standard deviation.
Q: A set of exam scores has a mean of 72 and a standard deviation of 8. Assuming a roughly normal distribution, approximately what percentage of scores fall between 64 and 80?
A: 64 and 80 are each one standard deviation from the mean (72 − 8 = 64, 72 + 8 = 80). By the empirical rule, about 68% of scores fall in this range.
Q: What is the key difference between descriptive and inferential statistics?
A: Descriptive statistics summarises the data you have collected. Inferential statistics uses sample data to make predictions or draw conclusions about a larger population.
Standard deviation feeds directly into the standard normal distribution and z-scores. When you standardise a value (subtract the mean, divide by the standard deviation), you are converting it to units of standard deviation, which is the basis for hypothesis testing and confidence intervals.
Cumulative relative frequency connects to probability distributions. The cumulative distribution function (CDF) in probability is the continuous equivalent of what cumulative relative frequency does for observed data.
Descriptive statistics is the prerequisite for inferential statistics. You need to summarise and understand your sample before you can make claims about the population it came from.
Descriptive statistics, cumulative relative frequency, cumulative frequency, ogive, percentile, standard deviation, sample standard deviation, population standard deviation, sigma, variance, Bessel's correction, mean, average, dispersion, spread, range, IQR, interquartile range, frequency distribution, relative frequency, histogram, box plot, measures of central tendency, measures of variability, STAT 101, intro stats, data summary