SECU0006 Probability, Statistics and Modelling I
SECU0006 Overview
- University College London
- T2 2026/27
- 6 chapters
- Statistics
SECU0006 Probability, Statistics and Modelling I introduces quantitative reasoning for real-world uncertainty. The module develops formal probability, common distributions, statistical descriptions of data, visual communication, association and the evaluation of evidence and interventions.
- Assessed by Class Test 50% · Assignment 50%
- Key terms Conditional probability, Complement, Random variable, Expected value
- Assessment move Divide study time between without-reference-materials recall and reproducible analysis.
- Most asked Are past papers included in the available materials?
What you get in the full SECU0006 Study & Assessment Guide
- 6 chapters, 21 pages — every assessed topic
- 16 original practice questions, each with a model answer
- 6 worked examples, solved the way SECU0006 marks
- 6 computed figures, redrawn from the course’s own numbers
- Before-you-submit checklist for the Class Test (50%) — what markers reward, what costs marks
Free on this page: assessment table, chapter map, one worked example, key terms, FAQ. The rest is in the full study & assessment guide.
How SECU0006 is assessed
| Component | Weight | Format |
|---|---|---|
| Class Test | 50% | 60-minute in-person, without-reference-materials computer-based test under exam conditions on 24 March 2027. |
| Assignment | 50% | 1,500-word statistical research report in Jupyter Notebook, due 15 April 2027 at 16:00 UK time. |
What other students are asking about SECU0006
- Where can I find lecture notes or study guides for SECU0006 at ucl?
- The SECU0006 study guide in AskSia Library covers Probability, Statistics and Modelling I in 6 chapters, with the assessment table, a chapter-by-chapter topic map, key terms and a worked example, free to read on this page.
- Is there a final exam in SECU0006?
- No final exam is listed in the offering this AskSia Library guide is built from; SECU0006 is assessed by Class Test 50%, Assignment 50%.
- What does SECU0006 cover on probability distributions?
- In AskSia Library's SECU0006 guide, the Probability Distributions chapter explains random variables, expectation, binomial models, normal models and tail areas.
- What does SECU0006 cover on describing and visualising data?
- In AskSia Library's SECU0006 guide, the Describing and Visualising Data chapter explains centre, spread, frequencies, histograms, box plots and honest comparison.
- What does SECU0006 cover on association and statistical evidence?
- In AskSia Library's SECU0006 guide, the Association and Statistical Evidence chapter explains correlation, confounding, competing explanations and limits on causal claims.
Current SECU0006 dates
| Date | Item | Control |
|---|---|---|
| 24 March 2027 | Class Test | The test contributes 50% of the module grade. |
| 15 April 2027, 16:00 UK time | Assignment deadline | The report contributes 50% of the module grade. |
Dates are as published in the current course materials. Confirm exact deadlines and submission settings in the live LMS.
What SECU0006 covers
Six chapters connect probability, distributions, data description, association, Python workflows and research communication.
Uncertainty and Probability
Events, complements, conditional probability and evidence under uncertainty.Uncertainty and Probability introduces events, complements, unions, intersections and conditioning as tools for reasoning about uncertain real-world phenomena.
- Events and sample spaces
- Complements and unions
- Conditional probability
- Independence
- Probability trees
- Evidence under alternatives
Probability Distributions
Random variables, expectation, binomial models, normal models and tail areas.Probability Distributions links random processes with patterns of possible values. The chapter covers expectation, variance, standard deviation, binomial counts, normal standardisation and tail areas.
- Random variables
- Expected value
- Variance and standard deviation
- Binomial conditions
- Normal standardisation
- Tail probability
Describing and Visualising Data
Centre, spread, frequencies, histograms, box plots and honest comparison.Describing and Visualising Data combines centre, spread, frequencies and graphics to communicate quantitative patterns. It contrasts mean with median, standard deviation with interquartile range, and bar charts with histograms.
- Frequency inspection
- Mean and median
- Range and quartiles
- Standard deviation
- Histograms and bars
- Honest axes and captions
Association and Statistical Evidence
Correlation, confounding, competing explanations and limits on causal claims.Association and Statistical Evidence examines relationships without overstating what they prove. Scatterplots and correlation describe linear co-movement, while confounding, reverse direction, selection and measurement can alter interpretation.
- Scatterplots
- Correlation
- Nonlinear patterns
- Confounding
- Competing hypotheses
- Causal boundaries
Reproducible Analysis in Python
Jupyter workflows, data checks, cleaning, validation and annotated code.Reproducible Analysis in Python organises Jupyter work into loading, inspection, cleaning, analysis, validation and communication. It focuses on visible dependencies, documented recodes, valid counts, codebook use and clean-kernel execution.
- Notebook order
- Variable inspection
- Missing-value recodes
- Sample validation
- Annotated code
- Clean-kernel execution
Communicating a Crime Data Study
Research questions, method justification, results, limitations and policy meaning.Communicating a Crime Data Study builds a coherent report from a focused research question through justified data choices, methods, results, limitations and policy meaning.
- Research questions
- Literature context
- Method justification
- Results presentation
- Limitations and bias
- Policy implications
Weekly practical work uses Python in Jupyter Notebook so that calculations, tables, figures and explanations can form one reproducible analysis.
The assessment has two equal-weight components. The Class Test contributes 50% and is a 60-minute, in-person, without-reference-materials computer test under exam conditions.
Published instructions describe four questions worth ten marks each, several parts and a mixture of multiple-choice, short-answer and single-answer formats. It covers Topics 1 to 9 and does not test Python programming.
The Assignment contributes the remaining 50% and is a 1,500-word statistical research report produced in Jupyter Notebook, with both the notebook and HTML export required.
The most useful preparation is therefore split. For the Class Test, practise recognising the correct event, denominator, distribution or summary before calculating, then interpret each result in the scenario.
For the Assignment, build one evidence chain from a research question through justified variables, cleaning, analysis, visualisation, limitation and policy meaning. The source does not publish a difficulty score or pass rate.
Its demands are visible instead in the breadth of test material, the without-reference-materials time limit and the requirement that the report go beyond basic summaries or one inferential test while remaining accurate, accessible and reproducible.
Use the published assessment format to shape weekly preparation.
After each topic, write one short explanation without notes, one numerical or graphical check, and one sentence limiting the conclusion. Then reproduce the practical result in Jupyter from a clean kernel. This paired routine prepares the theoretical reasoning required in the Class Test while also building the transparent workflow expected in the Assignment.
When reviewing an error, classify it as an event-definition, denominator, model-assumption, summary-choice, coding, interpretation or communication problem. Correcting the cause is more valuable than memorising the right answer.
In assignment planning, inspect the survey codebook and frequencies before choosing variables, and reserve enough time to justify transformations, review missing data, check figures and interpret limitations for a non-expert policy audience. The course rewards a connected argument in which calculations, code and prose support the same research question.
Worked example · free
Update a probability from observed frequencies
- 1Count all flagged cases: 90 + 88 = 178.
- 2Use the requested conditional denominator: 90/178 = 0.506.
- 1Interpret this as about half of flagged cases in the stated population.
Key terms
- Conditional probability
- The probability of an event after restricting attention to cases where another event is known.
- Complement
- The event that the named event does not occur.
- Random variable
- A numerical value assigned to each outcome of a random process.
- Expected value
- The probability-weighted long-run centre of a random variable.
- Standard deviation
- A measure of spread expressed in the same unit as the variable.
- Interquartile range
- The distance between the first and third quartiles.
- Distribution
- The pattern of possible values and their frequencies or probabilities.
- Correlation
- A standardised measure of linear association between two quantitative variables.
- Confounding
- Distortion of a relationship by another variable connected with both variables of interest.
- Sampling uncertainty
- Variation in an estimate arising because a sample rather than the whole population is observed.
- Reproducibility
- The ability to obtain reported outputs again from the stated data and code.
- Analytical sample
- The observations remaining after the documented inclusion, exclusion and missing-data rules.
SECU0006 FAQ
How difficult is the module?
The module does not publish a difficulty rating or pass rate, and it assumes no prior statistics or mathematics. Its challenge comes from combining theory, short timed responses, interpretation, Python analysis and a research report. Students need to choose methods accurately and explain what results mean, rather than memorising formulas without context.
What is the assessment structure?
The Class Test contributes 50% and the Assignment contributes 50%, so the numeric weights sum to 100%. The test is a without-reference-materials computer assessment under exam conditions. The assignment is a 1,500-word statistical report prepared in Jupyter Notebook, supported by reproducible code, figures, tables and plain-language interpretation.
What is the Class Test format?
The published instructions describe four questions worth ten marks each, with several parts and a mixture of multiple-choice, short-answer and single-answer items. It lasts 60 minutes, takes place in person and covers Topics 1 to 9 from lectures and tutorials. Python programming is not tested in the Class Test.
What does the Assignment require?
The Assignment asks for a focused crime-data research report using the supplied survey dataset. Students must formulate research questions, justify variables and methods, analyse multiple relevant relationships, communicate results to non-experts, discuss limitations and include reproducible annotated code. Both the Jupyter Notebook and an exported HTML version are required.
Are past papers included in the available materials?
No past Class Test paper or completed Practice Test appears in the available folder. The module says a Practice Test will be released to demonstrate format, not likely content. This guide therefore uses original questions shaped by the published question types and does not copy or lightly rewrite proprietary assessment stems.
How should I revise efficiently?
For the Class Test, use short mixed-format retrieval drills and check denominators, assumptions, units, rounding and causal language. For the Assignment, repeatedly run the notebook from a clean kernel and audit the chain from research question to variable, method, result, limitation and implication. Keep an error log organised by the reason a result failed.
Questions students ask in each SECU0006 chapter
- Why does conditioning change a probability?
- It restricts the denominator to cases where the stated condition is already known, so the comparison group changes. Recalculate after changing the conditioning set to see how strongly the result depends on that choice. Chapter 01 · Uncertainty and Probability
- Can an expected value be impossible to observe?
- Yes. It describes a long-run weighted centre and need not equal a possible single outcome. Interpret it across repeated processes rather than as a fractional outcome in one observed batch. Chapter 02 · Probability Distributions
- When is the median useful?
- It is useful when a skewed distribution or extreme values would pull the mean away from the typical observation. Pair it with the interquartile range and describe the skew rather than hiding the tail. Chapter 03 · Describing and Visualising Data
- Does zero correlation mean no relationship?
- No. A curved pattern or opposing subgroup patterns can produce little overall linear correlation. Inspect a scatterplot and subgroup structure before concluding that the variables are unrelated. Chapter 04 · Association and Statistical Evidence
- Why restart and run all?
- It exposes hidden dependencies on objects created by an earlier interactive session or out-of-order cells. A clean execution proves that every reported object is created visibly in the submitted notebook. Chapter 05 · Reproducible Analysis in Python
- How much analysis is enough?
- The instructions require more than basic summaries or a single inferential test, but accuracy and coherence matter more than output volume. Each additional method should answer a necessary part of the research question and be explained accurately. Chapter 06 · Communicating a Crime Data Study
How to prepare for the assessments
Divide study time between without-reference-materials recall and reproducible analysis. Begin each theory session by defining the event or variable in words, then solve an original problem without notes and inspect the denominator, assumptions, units and interpretation. Rotate probability, distributions, descriptive summaries and association so recognition does not depend on chapter order.
For practical work, inspect frequencies before cleaning, document every recode, print valid counts after filters and restart the Jupyter kernel before running all cells. Build the Assignment around one research question and remove outputs that do not serve it. Use figures to clarify a comparison, not decorate a page.
In the final week before the Class Test, complete timed multipart scenarios with a basic calculator and practise concise short answers. Before submitting the Assignment, compare the notebook with its HTML export, verify anonymous identifiers, recheck the word limit and ensure every reported value is generated by visible code.
Your AI Statistics tutor for SECU0006
Stuck on a hard SECU0006 question? Sia is AskSia’s AI Statistics tutor — ask any SECU0006 Probability, Statistics and Modelling I question and get a clear, step-by-step explanation grounded in how the course is actually taught and assessed. Read this whole study guide free, then take your hardest questions to Sia.