The six-stage enquiry sequence
A geographical enquiry starts from a clear, testable question or hypothesis (true-or-false by evidence) and moves through a fixed sequence: (1) form the question/hypothesis; (2) plan what data to collect and how; (3) collect data in the field; (4) present it (maps, graphs, tables); (5) analyse and interpret it; (6) conclude and evaluate reliability and limitations. The sequence is not one-way: evaluation can send it back to refine the question or method.
Primary vs secondary data and methods
Primary data is collected first-hand in fieldwork -- counting pedestrians, measuring velocity, or scoring an EQS. Secondary data already exists, gathered by someone else, from a census, council report or map. Methods: questionnaires (questions asked of other people); an environmental quality survey (a -5 to +5 sheet the STUDENT scores from their own observation); field sketches; measurements/counts.
Sampling methods and risk assessment
A sample selects part of a population or area. Random: random numbers give every point an equal, unpredictable chance, no pattern. Systematic: points at a fixed regular interval (every 10th person, or every 20 m). Stratified: divide into sub-groups (strata), then sample randomly WITHIN each in proportion to size, so each is fairly represented. A risk assessment pairs each hazard with a specific control (hi-vis, working in pairs), done BEFORE fieldwork.
Drawn from real examiner reports.
No explicit Yes/No/Partially decision
The most repeated Paper 4 message. Asked whether results support a hypothesis, OPEN with an explicit decision -- "Yes", "No" or "Partially" -- BEFORE the evidence. A vague verdict ("it might be true") does not earn the decision mark. If the question tells candidates to accept a decision, disagreeing also forfeits credit: explain the GIVEN decision, do not re-argue it.
Digest (## Skills, Paper 4 investigation skills): hypothesis-decision technique is the single most repeated Paper 4 message -- open with an explicit Yes/No/Partially decision before evidence; vague verdicts or disagreeing with a given decision both forfeit credit (s23 Paper 41/42/43 key messages).
Systematic mislabelled as random
The topic's key definition trap. The commonest error calls systematic sampling ("every 10th person") random -- but systematic has a FIXED interval, the opposite of random's equal, unpredictable chance. And stratified sampling is given as "different groups" without WHY: it represents sub-groups FAIRLY, in proportion to size.
Digest (## Skills, Paper 4 investigation skills): sampling-technique knowledge is consistently weak -- systematic confused with random ("every 10th person" mislabelled random), and stratified described without reference to representing population sub-groups (w22 Paper 41 Q1b(i)).
EQS is not a questionnaire
An environmental quality survey (EQS) is widely confused with a questionnaire. An EQS is a bi-polar sheet (-5 to +5) the STUDENT scores from their OWN observation -- no one else involved. A questionnaire is questions ASKED of other people. Confusing them gives an EQS's limit (one subjective observer) as a questionnaire's (non-response, dishonest answers).
Digest (## Skills, Paper 4 investigation skills): bi-polar environmental quality surveys are widely confused with questionnaires -- it is the STUDENT's own on-the-spot scored observation, not something administered to other people (s23 Paper 41 Q1d).
A vague or untestable hypothesis
A hypothesis must be one testable statement with a clear variable and direction -- e.g. "environmental quality improves with distance from the CBD." Weak enquiries give a vague statement ("the town has different areas") or set out 4-5 statements or open guiding questions instead of 2-3 focused ones. Fewer, sharper hypotheses let the data reach a decision.
Digest (## Skills, Paper 4 / June 2023 Paper 3 coursework report): hypotheses work best as 2-3 focused statements, not 4-5 or guiding questions (s23 Paper 3 coursework report).
Naming a sample method, not justifying it
A sampling method scores little unless JUSTIFIED for the enquiry. Candidates name a method but not why it suits the context -- systematic spreads counts evenly so a lunchtime peak is not missed; random avoids deliberate bias in a uniform area; stratified fairly represents different-sized zones. Link the method to why it fits this population or area.
Digest (## Skills, June 2023 Paper 3 coursework report): sampling justification is consistently thin (s23 Paper 3 coursework report).
Vague hazard or vague control measure
A risk assessment pairs a SPECIFIC hazard with a SPECIFIC control. Vague answers -- "it could be dangerous", "be careful" -- earn no credit, nor does a hazard with no matching control. Each control must fit the site: fast water -> stay in shallow, slow sections with suitable footwear; traffic -> hi-vis and a safe crossing point; weather -> check the forecast.
"Every 10th" is systematic, not random
A testable wrong belief. Some assume any method avoiding hand-picking -- "every 10th person", or a reading every 20 m -- must be RANDOM. It is not: systematic sampling has a FIXED, regular interval, a predictable pattern. Random needs every point to have an EQUAL, unpredictable chance, from random number tables, not a fixed rule.
Digest (## Skills, Paper 4 investigation skills): sampling-technique knowledge is consistently weak -- systematic is confused with random, e.g. "every 10th person" mislabelled random (w22 Paper 41 Q1b(i)).
An EQS is not filled in by others
A second testable wrong belief. Some assume an EQS works like a questionnaire -- filled in by residents or passers-by. In fact the STUDENT completes it: at a site, they score bi-polar criteria from their OWN judgement, no one else involved. Confusing them makes a candidate blame "dishonest answers" (a questionnaire limit), not the EQS's real limit: one subjective scorer.
Digest (## Skills, Paper 4 investigation skills): bi-polar environmental quality surveys are widely confused with questionnaires -- it is the STUDENT's own on-the-spot scored observation, not something administered to other people (s23 Paper 41 Q1d).
Structure a hypothesis-support answer
For "state whether the hypothesis is supported", answer in order: (1) an explicit decision -- Yes, No or Partially -- FIRST; (2) evidence using SPECIFIC figures, not a general description; (3) if asked, a limitation of the method (small sample, one observer).
Never leave a plotting task blank
Practical completion tasks -- plotting a point, completing a graph, adding a best-fit line -- are accessible marks, yet often left blank. Attempt every one: a plotted point or line earns credit with no written explanation, a blank cannot. Use the given scale and axes.
Name the exact method, not "a survey"
Specify the exact method, not "a survey": a questionnaire (other people's answers), an EQS (your own scored observation), a measurement/count, or a field sketch. To measure environmental quality itself, use an EQS, not a questionnaire. Classify sources as primary or secondary.
Full notes, flashcards, Q&A and the topic quiz for every premium subject.
Premium plans are US$8.99/month or US$49.99/year — first month free.
Studying with a parent's blessing? Show them this.
