Fifteen-mark essays describe a generic place, for example 'a country' or 'a river', rather than a named, specific case study, which caps the answer at Level 2, a ceiling of around 10 marks.

Before you write, note down the place name, one statistic, and one dated fact for the case study you are using, then build the essay around them from the first sentence.

Data from a resource, such as a river level in metres or a temperature change in degrees, gets described in vague terms instead of quoted as an actual figure.

If the resource gives a number, put that number in your sentence. A trend described without a figure attached reads as an impression rather than evidence.

Bedload erosion, meaning attrition between particles, gets confused with erosion of the channel boundary itself, which covers hydraulic action, abrasion and corrosion.

Keep the two separate: attrition is particles wearing each other down as they collide, while hydraulic action, abrasion and corrosion describe the river wearing away its own bed and banks.

Choropleth map keys are built with overlapping category intervals, for example 10 to 20 and 20 to 30 sharing the boundary value, or with gaps left between bands.

Build the key before you shade anything. Check every data value falls into exactly one band, with no overlaps and no gaps, then shade from dark for the highest values to light for the lowest.

Graphs constructed in AS 3 are missing a full title naming both variables, axis units, or a key, even when the plotting itself is accurate.

Title every graph with both variables named, label both axes with units, keep the scale regular, and add a key whenever more than one data set is plotted.

Statistical results are stated as a bare number without saying what they mean geographically, missing the link back to the theory or model the fieldwork was testing.

After any calculation, add one sentence connecting the result to the geographical theory it supports or challenges. A correlation coefficient on its own earns fewer marks than one that is explained.

Risk assessment answers confuse avoidance strategies, the preventative steps taken before fieldwork, with contingency planning, the actions taken if an accident happens anyway.

Keep the two in separate sentences: avoidance comes before the incident, contingency comes after it. Name at least one example of each when a question asks for both.

CCEA AS-Level · Exam Tips

Geography 3910 Exam Tips

AS Level Geography (CCEA) exam tips built from real examiner reports: why named case studies and quoted figures separate Level 3 from Level 2, how the Section A and Section B timing works, and the statistics and graph conventions that lose marks in Fieldwork.

4 min readUpdated: 3 Sept 2026

Exam at a Glance

Papers

Total Marks

210

Time Limit

3h 30min

Question Types

Paper	Duration	Marks	Questions	Weighting	Question Types
AS 1: Physical Geography	1h 15min	75	5	36%	Section A: Compulsory Short Structured Questions, Section B: Extended Writing (choose 2 of 3)
AS 2: Human Geography	1h 15min	75	5	36%	Section A: Compulsory Short Structured Questions, Section B: Extended Writing (choose 2 of 3)
AS 3: Fieldwork	1h	60	13	28%	Short Answer (Planning and Risk Assessment), Extended Response (Primary Data Collection), Quantitative Analysis and Significance Interpretation, Extended Response (Geographical Reasoning and Theory Link), Short Answer (Fieldwork Extension and Evaluation), Apparatus and Methodology Analysis, Spatial and Statistical Calculation and Hypothesis Testing, Applied Geographical Management Explanation, Sampling Theory Definition, Descriptive Statistics Calculation, Cartographical Key and Choropleth Map Completion, Evaluation of Cartographic or Presentation Technique

Grade Scale

ABCDEU

Calculator Policy

Calculators are allowed in every paper under the general JCQ and CCEA rules: no stored retrievable information such as formulae, text or programs, no computer algebra function, no internet or translator function. There is no Geography-specific calculator ban in the specification. You will need one in AS 3 for statistical tests such as Spearman's rank correlation and for interpreting your own fieldwork data, and for any calculation elsewhere in AS 1 or AS 2. If your calculator has an exam mode, switch it on rather than relying on the reset button, which does not clear stored memory.

AO1: AO1: demonstrate knowledge and understanding of places, environments, concepts, processes, interactions and change at a variety of scales (36%)
AO2: AO2: apply knowledge and understanding in different contexts to analyse, interpret and evaluate key concepts, information and issues (40%)
AO3: AO3: use a variety of relevant methods and techniques to investigate geographical questions and issues, analyse, interpret and evaluate data and resources, and construct arguments and draw conclusions (24%)

Built from real past papers and marking schemes (2022–2025).

Tips & Strategies

Where the marks actually sit

AS Geography carries 210 marks across three papers, each running at almost exactly one mark per minute: AS 1, Physical Geography (75 marks, 75 minutes), AS 2, Human Geography (75 marks, 75 minutes), and AS 3, Fieldwork Skills and Techniques in Geography (60 marks, 60 minutes). AS 1 and AS 2 share an identical structure: Section A has three compulsory, short, structured questions worth 15 marks each (45 marks), and Section B offers three extended writing questions, of which you answer two, again worth 15 marks each (30 marks). AS 3 tests fieldwork you have carried out yourself, marked through your own submitted Table of Data.

Paper by paper timing

Because AS 1 and AS 2 run at one mark per minute, the arithmetic is simple: budget 15 minutes for each of the three Section A questions (45 minutes total) and 15 minutes for each of your two chosen Section B essays (30 minutes total), leaving no slack, so choose your two Section B questions quickly and do not start a third. AS 3 also runs at one mark per minute across its two questions, Question 1 on your own Fieldwork Statement and Table of Data, Question 2 on unseen fieldwork scenarios, so treat every mark as worth roughly one minute of writing and no more.

What actually separates a Level 3 essay from a Level 2 one

The single most common reason a 15-mark essay gets capped at Level 2, a ceiling of around 10 marks, is a generic, textbook-style account with no named place. Examiners want a specific location, specific figures, and a specific outcome, not 'a country' or 'a river'. Before you write a Section B essay, force yourself to write down the place name, one statistic, and one dated fact, then build the essay around them.

The second habit that keeps essays out of the top level is quoting data from a resource in vague terms rather than with the actual figures attached. If a resource shows a river level in metres, or a temperature change in degrees, use that number in your sentence. A described trend with no figure attached reads as an impression, not evidence, and is marked that way.

AS 3: the statistics questions are where marks leak

Fieldwork data analysis in AS 3 regularly asks for a statistical test, most often Spearman's rank correlation. Two errors recur: stating the correlation coefficient without saying whether it is positive or negative, and confusing the degrees of freedom, which is the sample size minus 2, with the sample size itself when reading the critical value from the significance table. Show your working step by step, state the direction of the correlation in words, and use \( n - 2 \) when you look up the critical value, not \( n \).

Choropleth maps and other cartographic work in AS 3 lose marks for two specific reasons: category intervals in the key that overlap, for example 10 to 20 and 20 to 30 sharing the boundary value with no rule for which band it belongs to, and intervals that leave gaps between them. Build the key first, check every value in your data falls into exactly one band, and shade from dark for the highest values to light for the lowest, consistently.

Any graph you construct, whether in AS 3 or supporting an answer elsewhere, needs a title naming both variables, axis labels with units, a regular scale, and a key if more than one data set appears. Missing any one of these costs a mark even when the plotting itself is accurate.

CCEA conventions to know

Extended writing in Section B of AS 1 and AS 2 is marked against quality of written communication as part of the same mark, not separately: legible writing, accurate spelling and correct specialist vocabulary all affect where your answer sits within its level. Answers are placed in a level first, then a specific mark within it, based on how well they meet that level's descriptors for detail, balance and judgement.

Risk assessment questions in AS 3 test a distinction that is easy to blur: avoidance strategies are the preventative measures you put in place before fieldwork to stop an accident happening, while contingency planning covers what you actually do if one happens anyway. Keep the two in separate sentences when you answer.

Exam day plan

In AS 1 and AS 2, read all three Section B options before choosing your two, since a case study you can name with real figures beats a topic you understand only in general terms. Answer Section A first while the paper is fresh, then give your two Section B essays 15 minutes each, opening each one with the place name you are using. In AS 3, keep your Table of Data open beside every question, show full working on any statistical calculation, and build your key or graph before you write the surrounding analysis so the figures are in front of you as you argue from them.

Ready to test yourself?

Turn these notes into exam-style practice. Get unlimited AI questions on this topic with instant marking and explanations.

Practise This Topic

Common Mistakes

1highMarks at stake: 5Section B extended writing (AS 1 and AS 2)
Fifteen-mark essays describe a generic place, for example 'a country' or 'a river', rather than a named, specific case study, which caps the answer at Level 2, a ceiling of around 10 marks.
How to avoid it: Before you write, note down the place name, one statistic, and one dated fact for the case study you are using, then build the essay around them from the first sentence.
2highMarks at stake: 3Data response questions across AS 1, AS 2 and AS 3
Data from a resource, such as a river level in metres or a temperature change in degrees, gets described in vague terms instead of quoted as an actual figure.
How to avoid it: If the resource gives a number, put that number in your sentence. A trend described without a figure attached reads as an impression rather than evidence.
3mediumMarks at stake: 2Fluvial processes
Bedload erosion, meaning attrition between particles, gets confused with erosion of the channel boundary itself, which covers hydraulic action, abrasion and corrosion.
How to avoid it: Keep the two separate: attrition is particles wearing each other down as they collide, while hydraulic action, abrasion and corrosion describe the river wearing away its own bed and banks.
4mediumMarks at stake: 3Statistical testing in AS 3
In Spearman's rank correlation questions, candidates confuse the degrees of freedom, which is the sample size minus 2, with the sample size itself, and read the wrong row of the significance table.
How to avoid it: Write \( n - 2 \) explicitly before you look up the critical value, and state whether your calculated correlation is positive or negative in words, not just as a number.
5mediumMarks at stake: 2Cartographic and choropleth map skills
Choropleth map keys are built with overlapping category intervals, for example 10 to 20 and 20 to 30 sharing the boundary value, or with gaps left between bands.
How to avoid it: Build the key before you shade anything. Check every data value falls into exactly one band, with no overlaps and no gaps, then shade from dark for the highest values to light for the lowest.
6highMarks at stake: 2Graphical construction in AS 3
Graphs constructed in AS 3 are missing a full title naming both variables, axis units, or a key, even when the plotting itself is accurate.
How to avoid it: Title every graph with both variables named, label both axes with units, keep the scale regular, and add a key whenever more than one data set is plotted.
7mediumMarks at stake: 3Question 1(c)(ii): geographical reasoning
Statistical results are stated as a bare number without saying what they mean geographically, missing the link back to the theory or model the fieldwork was testing.
How to avoid it: After any calculation, add one sentence connecting the result to the geographical theory it supports or challenges. A correlation coefficient on its own earns fewer marks than one that is explained.
8mediumMarks at stake: 2Planning and risk assessment
Risk assessment answers confuse avoidance strategies, the preventative steps taken before fieldwork, with contingency planning, the actions taken if an accident happens anyway.
How to avoid it: Keep the two in separate sentences: avoidance comes before the incident, contingency comes after it. Name at least one example of each when a question asks for both.

Turn these tips into top grades

thinka turns your weak spots into targeted practice, with instant marking and exam-style feedback. Study smarter, not longer.

Start Practising Free See pricing

Practise real exam questions with instant AI feedback and marking.

Start Practising Free