Statistics basics for analysts
You compute means, medians, spread, percentiles and rates on real event data and learn when numbers mislead.
Level 101 · 17 lessons · 15 practice labs
Lessons
Centre and spread
- What statistics gives an analyst
- Mean, median and mode
- Range, variance and standard deviation
- Percentiles and the interquartile range
- Quick task: summarise server response times
Distributions
- Frequency tables and histograms
- Normal and skewed distributions
- Outliers: error or incident
- Standard scores (z-scores)
- Quick task: hosts with outlying traffic
Rates and comparison
- Rates and percentages: why the denominator matters
- Comparing two groups
- Correlation is not causation
- Sampling: deciding without seeing everything
- Quick task: the false-positive rate of an alert rule
Telling numbers honestly
- Misleading statistics and how to spot them
- Capstone: a weekly alert statistics report
Practice labs
- Summarising delivery times: Compute the mean, median and mode of Atlas Cargo's delivery times and write which one better describes the typical customer.
- Same mean, different spread: Compute the standard deviation and interquartile range for two servers with the same mean response time and show which one is less predictable.
- A percentile-based service target: Compute the 90th and 99th percentiles of Novaro Telecom's call wait times and prove whether the target “95 per cent under 30 seconds” is met.
- Build a frequency table: Bin a week of login attempts into intervals of ten and produce the frequency table and the busiest interval.
- Recognise a skewed distribution: Draw a text histogram of file-size data and explain its right skew using the gap between mean and median.
- The cause of a two-peaked distribution: Split a two-peaked session-length distribution by shift and report each group's summary and the reason for the peaks.
- Outlier or typo: Find the extreme values in a payroll file and mark which ones are data-entry errors such as an extra zero.
- Outliers with the IQR rule: Use the 1.5 IQR rule to find hosts with outlying night-time traffic and write them to outliers.csv.
- Compare z-scores with a median-based method: Apply z-scores and median absolute deviation to a set where extremes distort the mean and show which method catches the real attacker.
- The denominator matters: Divide two branches' incident counts by their headcount and show with rates which one is really riskier.
- Compare two groups fairly: Compare the phishing click rates of trained and untrained staff while accounting for the different group sizes.
- Estimating from a sample: Estimate the misclassification rate from 200 random Tessera Insurance tickets and write the expected number of wrong ones among all 12,000 tickets with its limits.
- Catch a misleading percentage: Find the raw numbers behind the headline “attacks up 200 per cent” and rewrite it as a more honest sentence.
- Simpson's paradox: Show with data two detection rules where one looks better overall but is worse in every subgroup and write the right decision.
- Truncated axes and cherry-picked periods: Test four statistical claims from a slide deck against the data and list, with reasons, which ones use a cherry-picked period or the wrong denominator.
About this course
You compute means, medians, spread, percentiles and rates on real event data and learn when numbers mislead.
What you will learn
- Centre and spread
- Distributions
- Rates and comparison
- Telling numbers honestly
Prerequisites
No prior knowledge is needed; the course starts from zero.
Lab environment
Lessons run in your own Kali Linux terminal that opens in the browser. Nothing to install; the lab starts clean in every lesson.
Frequently asked questions
- Is this course right for my level?
- The course is at beginner level; every term is explained where it first appears.
- Is there anything to install?
- No. Every lesson runs in your own lab environment that opens in the browser.
- How do lessons work?
- Each lesson has 8 to 10 steps and a short quiz. Early steps show the command to type, later steps give less help; if two attempts fail, the solution opens.
- Can I try it for free?
- Yes. The first three lessons of every course are free.