G.5 Synthesis, Evaluation, and Skills

Source reliability is evaluated against credibility (who produced it and do they have a vested interest?), data collection methods (was the sample adequate and the methodology sound?), and potential biases (does the producer's purpose distort the data?). Each fact in the question should map onto one of these criteria.

The commissioning authority has a vested interest in a positive finding → credibility/bias. A sample of 40 out of 4,000 is only 1% of the population → inadequate sample size. No explanation of how the 40 were chosen → undisclosed sampling method, a key methodology concern.

Option A correctly names all three concerns directly arising from the stated facts. Options B, C, and D introduce concerns — temporal bias, comparison groups, disaggregation, leading questions — that are plausible source limitations in general but are not directly evidenced by the specific facts given in the scenario.

Option A is the only response whose three components (credibility bias, inadequate sample size, undisclosed sampling method) are each directly supported by a specific stated fact in the scenario, making it the most precise and evidentially grounded evaluation.

The question asks which combination of reliability concerns is most directly raised by the specific facts given — not which concerns could theoretically apply to any survey. Each concern must be traceable to a specific detail in the scenario.

Temporal bias and spatial restriction are not evidenced by any fact stated in the scenario. The scenario does not mention when the survey was conducted or restrict it geographically in a way that raises spatial bias — these are plausible concerns but are not 'directly raised' by the given information.

Low reproducibility (survey instrument not shared) and insufficient qualitative methods are reasonable general critiques but are not supported by any specific fact in the scenario. The scenario says nothing about whether the instrument was published or whether qualitative data was sought.

Lack of a comparison group, failure to disaggregate, and leading question wording are all valid general source limitations but none of the specific facts in the scenario (commissioning authority, sample of 40, no sampling explanation) directly evidence these particular concerns.

Each element of Option A maps precisely onto a stated fact: the commissioning authority → credibility bias; 40 out of 4,000 → inadequate sample size; no explanation of sample selection → undisclosed methodology. All three are directly raised by the evidence provided.

Data limitations in fieldwork fall into three categories: sample size (how many observations), spatial restrictions (where data was collected), and temporal constraints (when data was collected and over what duration). The question specifically asks for the temporal limitation.

A temporal limitation arises when the observation window is too narrow to capture the true pattern — for example, a count taken only at lunchtime misses morning rush hour, evening leisure, weekend variation, and seasonal tourist peaks, all of which affect any claim about 'general' footfall.

Option A describes a spatial limitation (one intersection). Option C describes an instrument reliability / observer accuracy issue. Option D describes a sampling representativeness issue related to demographic composition. Only Option B specifically addresses the time of data collection and the duration of the observation window.

Option B names the specific temporal restriction (single weekday lunchtime), identifies the consequence (results may reflect a market-day peak rather than general activity), and links this directly to why the conclusion ('general pedestrian activity') is unwarranted — this is a precise, mechanistic temporal limitation statement.

The question explicitly asks for the temporal limitation — not spatial, not instrument-related, not demographic. The correct answer must involve the timing or duration of data collection and explain why this undermines generalizability across time.

Option A identifies a genuine limitation — only one intersection surveyed — but this is a spatial restriction, not a temporal one. It addresses where data was collected, not when.

Option C raises observer/counting error, which is an instrument reliability concern. It does not relate to the timing of the survey or the duration of the observation window, so it is not a temporal limitation.

Option D raises demographic representativeness, which is partly temporal in implication but is framed as a sample composition issue (different demographic groups), not as a direct critique of the time period chosen for observation. It is less precise than Option B.

Option B precisely names the temporal constraint (single lunchtime on one Tuesday), explains the specific mechanism of distortion (market-day footfall peak), and links it to the invalid generalization ('general level of pedestrian activity'). This is a complete and precise temporal limitation.