OxfordAQA International A-level Geography (9635): 3.5 Geographical and fieldwork skills – Practice Questions

These are original questions written for Marlbridge, for revision and practice on this content. They are not reproduced past-paper questions, and they do not replicate the exam’s exact structure, question count or mark tariffs – examination boards hold copyright in their own papers. Use these alongside the official past papers from your board or school.

These questions cover section 3.5, Unit 5: Geographical and Fieldwork skills, of the OxfordAQA International AS and A-level Geography (9635) specification, Version 3.2: the geographical skills checklist (3.5.1) and fieldwork skills (3.5.2). Unit 5 is a 1 hour 30 minute written exam with 60 marks, all questions compulsory, worth 20% of the International A-level; 15 of those marks assess your own investigation. All places and data below are fictional, except where a question asks about your own fieldwork.

Revise first with the study guide and revision notes. The exam preparation guide covers paper technique. Track your progress on the OxfordAQA A-level Geography hub and checklist.

Questions

1. Outline two limitations of choropleth maps. [4]

2. Explain why a logarithmic scale might be used to plot the populations of all the settlements in a region, from hamlets to a large city. [3]

3. A student measured the long axis (mm) of 11 pebbles on a fictional beach, Skelby:

34, 41, 28, 55, 47, 39, 62, 30, 44, 51, 36

(a) Calculate the median. [1] (b) Calculate the inter-quartile range. Show your working. [3] (c) Suggest why the inter-quartile range is a better measure than the range for comparing this sample with a sample from another beach. [2]

4. Infiltration times (seconds) for a fixed volume of water were recorded at six points on two fictional fields.

Site	Times (s)	Mean (s)
A	14, 18, 22, 26, 30, 34	24.0
B	8, 15, 21, 27, 33, 40	24.0

(a) Calculate the standard deviation for Site A, using the formula: square root of [sum of (x - mean)^2 / n]. Show your working. [3] (b) The standard deviation for Site B is 10.7 s. Compare the two sites. [2]

5. A student measured channel cross-sectional area at 10 sites along a fictional river, Tarrow Water.

Site	Distance downstream (km)	Cross-sectional area (m^2)
1	0.5	0.42
2	1.2	0.61
3	2.0	0.38
4	2.8	0.90
5	3.5	0.55
6	4.4	1.12
7	5.1	0.74
8	6.0	0.66
9	6.8	1.30
10	7.5	1.05

Two-tailed critical values for n = 10 are 0.648 (0.05 level) and 0.794 (0.01 level).

(a) State a suitable null hypothesis. [1] (b) Calculate Spearman’s rank correlation coefficient. Use rs = 1 - 6(sum of d^2) / (n^3 - n). Show your working. [4] (c) Interpret your result using the critical values. [3]

6. In a fictional coastal town, Port Ellery, 100 people were asked their view of a new waterfront scheme.

Group	Positive	Neutral	Negative	Total
Residents	18	14	18	50
Visitors	30	12	8	50
Total	48	26	26	100

Critical values: df = 1: 3.84 (0.05), 6.63 (0.01); df = 2: 5.99 (0.05), 9.21 (0.01).

(a) Show how the expected frequency for residents with a positive view is calculated. [2] (b) Calculate the chi-square value. Show your working. [3] (c) State the degrees of freedom and interpret the result. [3]

7. On a proportional circle map of bus trips, a stop with 400 trips per day is drawn with a radius of 10 mm. Calculate the radii for stops with 900 and 1,600 trips, and explain your method. [3]

8. Explain why a triangular graph is suitable for showing the percentage of workers in primary, secondary and tertiary employment in different countries, and how one country is plotted. [4]

9. Students plan to investigate how residents of a fictional riverside town, Lanmore, perceive flood risk, using a questionnaire.

(a) Suggest and justify a sampling strategy. [4] (b) Outline two ethical issues and how the students should deal with them. [4] (c) Outline how the students should carry out a risk assessment for this fieldwork. [3]

10. With reference to your own fieldwork investigation, evaluate the methods you used to present your data. [9]

11. Assess the value of remotely sensed and crowd-sourced data compared with primary data collected in the field. [9]

12. “Statistical tests are the most useful way to analyse fieldwork data.” To what extent do you agree? [12]

Answers

Questions 10 to 12 are extended responses. The real exam marks these with levels of response, not point by point; the [1] points below show the kind of content that earns credit.

1. Choropleth maps imply sudden change at boundaries [1], when in reality values often change gradually across them [1]. They show one value for each whole area [1], hiding variation within it, such as a wealthy street in a poor district [1]. [4] Examiner insight: each limitation needs a reason or effect after it; two bare labels earn only half the marks.

2. Settlement populations cover a very wide range, from tens to millions [1]. On a linear scale, small settlements would be squashed near zero and impossible to tell apart [1]. A log scale gives each tenfold increase equal space, so small and large settlements can be shown and compared clearly on one axis [1]. [3] Examiner insight: say what a log scale actually does (equal space per tenfold step), not just that it “fits the data”.

3. (a) Ordered: 28, 30, 34, 36, 39, 41, 44, 47, 51, 55, 62. Median = 6th value = 41 mm [1]. (b) Lower quartile at position (11 + 1) / 4 = 3rd value = 34 mm [1]. Upper quartile at position 3(11 + 1) / 4 = 9th value = 51 mm [1]. IQR = 51 - 34 = 17 mm [1]. (c) The range uses only the two extreme values, so one unusually large or small pebble would distort it [1]. The IQR uses the middle half of the data, so it gives a fairer comparison of typical spread between beaches [1]. Examiner insight: the data must be put in order before finding positions; a common slip is counting along the unsorted list.

4. (a) Deviations from 24: -10, -6, -2, 2, 6, 10; squared: 100, 36, 4, 4, 36, 100; sum = 280 [1]. 280 / 6 = 46.7 [1]. Square root = 6.8 s [1]. (b) Both sites have the same mean (24.0 s) [1], but Site B’s standard deviation (10.7 s) is higher, so its infiltration times are more spread out and less consistent than Site A’s (6.8 s) [1]. Examiner insight: a comparison needs both figures and a word such as “more varied”; repeating the numbers alone is not a comparison.

5. (a) There is no relationship between distance downstream and channel cross-sectional area on Tarrow Water [1]. (b) Ranks for distance (largest = 1): 10, 9, 8, 7, 6, 5, 4, 3, 2, 1. Ranks for area: 9, 7, 10, 4, 8, 2, 5, 6, 1, 3 [1]. d: 1, 2, -2, 3, -2, 3, -1, -3, 1, -2; d^2: 1, 4, 4, 9, 4, 9, 1, 9, 1, 4; sum of d^2 = 46 [1]. rs = 1 - (6 x 46) / (1,000 - 10) = 1 - 276 / 990 [1] = 0.72 [1]. (c) 0.72 is greater than 0.648, so the result is significant at the 0.05 level and the null hypothesis is rejected [1]. It is below 0.794, so it is not significant at the 0.01 level [1]. There is a moderately strong positive relationship: cross-sectional area tends to increase downstream, though sites 3, 5 and 8 are smaller than expected and need explaining, for example by local bedrock or a measurement error [1]. Examiner insight: quote both critical values and say which level the result passes; “significant” alone is incomplete.

6. (a) Expected = (row total x column total) / grand total [1] = (50 x 48) / 100 = 24 [1]. (b) Expected values: 24, 13, 13 in each row [1]. (O - E)^2 / E for residents: 1.5, 0.08, 1.92; for visitors: 1.5, 0.08, 1.92 [1]. Chi-square = 7.0 [1]. (c) df = (2 - 1) x (3 - 1) = 2 [1]. 7.0 exceeds 5.99, so there is a significant difference between residents’ and visitors’ views at the 0.05 level [1]. It is below 9.21, so not significant at the 0.01 level; visitors are more positive and residents more negative than expected [1]. Examiner insight: the interpretation must say which group differs and in what direction, not only that the result is significant.

7. Circle area is proportional to the value, so radius is proportional to the square root of the value [1]. 900 trips: 10 x square root of (900 / 400) = 10 x 1.5 = 15 mm [1]. 1,600 trips: 10 x square root of (1,600 / 400) = 10 x 2 = 20 mm [1]. Examiner insight: scaling the radius directly (to 22.5 mm and 40 mm) makes large values look far too big and scores no method mark.

8. The three sectors are parts of one whole that add to 100% [1], which is exactly what a triangular graph needs, and it lets many countries be compared on one chart and grouped into clusters [1]. To plot a country, find its primary percentage on one axis, its secondary on the second and its tertiary on the third [1], and follow each value along that axis’s own gridlines until the three lines meet at a single point [1]. Examiner insight: reading along the wrong set of gridlines is the usual error, so say which lines you follow.

9. (a) Stratified sampling [1], dividing the town into zones by distance from the river or by flood-risk zone and sampling in proportion to the number of households in each [1]. This ensures residents in both high-risk and low-risk areas are represented [1], because perception is likely to vary with exposure, which a purely random sample might miss [1]. (b) Informed consent: explain the purpose and let people decline or stop at any time [1] [1]. Anonymity and sensitivity: do not record names or addresses, and word questions carefully because some residents may have been flooded before [1] [1]. (c) Identify hazards, such as traffic on busy streets, riverbanks and approaching strangers [1]. Judge how likely and how serious each one is [1]. Set control measures, such as working in groups, avoiding the riverbank, wearing visible clothing and agreeing a meeting time [1]. Examiner insight: justification must link the sampling type to the question (perception varies with risk), not just define it.

10. Indicative content:

• Name the presentation methods used, with place and topic (for example, located proportional circles of pedestrian counts on a base map) [1].
• Justify each choice in relation to the data type: continuous, categorical or spatial [1].
• Strength: a map showed the spatial pattern clearly, linking results to location [1].
• Strength: a scatter graph made the relationship between two variables visible before the statistical test [1].
• Limitation: a specific graph that hid variation, overlapped or was hard to read, with its effect [1].
• Limitation: qualitative data, such as interview responses, was summarised too much and lost detail [1].
• Improvement tied to a named weakness, such as a dispersion diagram to show spread or quotations to keep detail [1].
• Judgement on how well the presentation supported the analysis and conclusion [1].
• Reference to real values or results from the enquiry throughout [1]. [9]

Examiner insight: familiar fieldwork answers that could describe anyone’s investigation stay in the lower levels; use your own place, numbers and findings.

11. Indicative content:

• Remotely sensed data covers large or inaccessible areas and shows change over time, such as land-use change [1].
• It is consistent and repeatable, but limited by cloud cover and resolution [1].
• It needs ground checking to confirm what features are [1].
• Crowd-sourced and big data are very large and close to real time [1].
• They can show patterns of movement or opinion that a student could never collect alone [1].
• But they are unrepresentative, missing people without technology, and raise privacy concerns [1].
• Primary data is designed for the exact question, with known methods and sampling [1].
• But primary data is small in scale and limited to one time and place [1].
• Judgement: the sources are most valuable when combined, with secondary data giving context and scale and primary data testing a specific hypothesis [1]. [9]

Examiner insight: “assess” needs a weighed judgement at the end; a list of strengths and weaknesses with no verdict caps the mark.

12. Indicative content:

• Statistical tests give an objective measure of whether a relationship or difference is likely to be chance [1].
• They let you accept or reject a hypothesis at a stated significance level, such as 0.05 [1].
• They allow comparison between sites or with other studies [1].
• Spearman’s rank suits paired numerical data, such as distance and an environmental score [1].
• Chi-square suits frequency counts in categories, such as responses by group [1].
• But tests need suitable data: enough pairs or counts, and raw frequencies for chi-square [1].
• A significant result does not explain why, and does not show causation [1].
• Small fieldwork samples can produce misleading results [1].
• The specification itself says to use statistical techniques “where appropriate” and other techniques where not [1].
• Qualitative data, such as interviews, is better analysed by coding themes and using quotations [1].
• Maps and graphs often reveal patterns and anomalies that a single coefficient hides [1].
• Judgement: statistical tests are most useful for testing a clear quantitative hypothesis, but the most useful analysis combines them with description, mapping and qualitative evidence [1]. [12]

Examiner insight: top answers use examples of specific tests and data, and reach a judgement that depends on the type of data, not a flat yes or no.

Where marks are usually lost

• Not ordering data before finding the median and quartiles.
• Ranking the two Spearman variables in opposite directions.
• Missing the square root at the end of a standard deviation calculation.
• Calculating chi-square from percentages instead of raw counts.
• Interpreting a test without stating the critical value and significance level.
• Treating correlation as proof of cause.
• Scaling proportional circle radius directly with the value.
• Defining a sampling method without justifying it for the scenario.
• Ethics answers that say “be ethical” without a specific issue and action.
• Writing about your own fieldwork with no place, data or result.

Next steps

• Re-read the study guide sections on any question you got wrong.
• Practise using these skills on content from other units with the physical geography practice set and globalisation practice set.
• Find your gaps with our diagnostics.
• Book a free trial class with a Marlbridge geography teacher.

Official syllabus

These questions are written against the OxfordAQA International AS and A-level Geography (9635) specification, Version 3.2, section 3.5 (Unit 5: Geographical and Fieldwork skills) and the scheme of assessment in section 4.