The Proposal With a Borrowed Hypothesis
A student submits a research proposal investigating whether a local creek's phosphate levels affect algal bloom frequency. The proposal states: 'Hypothesis: phosphate causes algal blooms.' The methodology section says samples will be taken 'weekly from the creek' with no specified sites, volumes, or preservation method. The timeline lists data collection in weeks 1–2 and analysis in weeks 3–4, but the student also plans to run a 6-week laboratory mesocosm experiment. The budget requests $400 for a spectrophotometer but no reagents. The ethics section states 'no ethics approval needed because we are only testing water.' The teacher notes that the creek runs through private farmland and a public park.
1. Identify the three most serious flaws in this proposal and explain why each undermines the research.
Read the worked answer
Flaw 1 - The hypothesis is not testable as stated. 'Phosphate causes algal blooms' is a causal claim that a correlational field survey cannot support. A testable version would specify a predicted relationship (e.g. 'sites with higher phosphate will show higher chlorophyll-a, controlling for light and flow'). As written, no observation could count against the hypothesis, so it fails the falsifiability criterion central to scientific thinking.
Flaw 2 - The methods are underspecified and internally inconsistent. 'Weekly from the creek' gives no sites, replication, sample volume, filtration, or preservation, so the data would not be reproducible or comparable over time. The timeline is also impossible: a 6-week mesocosm cannot fit into a 4-week schedule, so the proposal promises data it cannot collect. This signals the design was not thought through.
Flaw 3 - Ethics and access are dismissed without analysis. Sampling on private farmland requires landowner permission, and public park sampling may need council approval. 'Only testing water' does not exempt the student from consent, safety, or environmental-impact considerations (e.g. removing water, disturbing habitat). The ethics section should identify stakeholders, permissions, and risk controls.
A fourth, lesser flaw is the budget: $400 for an instrument with no reagents means the assay cannot actually be run.
2. Rewrite the hypothesis and outline what the methods section must specify to make the study reproducible.
Read the worked answer
Revised hypothesis (testable and directional): 'In [named creek], sampling sites downstream of agricultural runoff will have higher dissolved phosphate concentrations and higher chlorophyll-a concentrations than upstream reference sites, and phosphate will remain a significant predictor of chlorophyll-a after accounting for light exposure and flow rate.'
This is testable because it predicts specific, measurable differences and a specific statistical relationship, and it could be falsified if downstream sites show no difference or if phosphate drops out as a predictor.
Methods must specify:
- Sites: exact coordinates, number of replicate sites per condition (upstream vs downstream), and justification for their selection.
- Sampling: volume per sample, number of replicates, depth, time of day, frequency, and total duration consistent with the timeline.
- Preservation and analysis: filtration, storage temperature, holding time, and the specific phosphate assay (e.g. colorimetric method) with reagents costed in the budget.
- Confounders measured: light, temperature, flow, and any known nutrient inputs.
- Permits and ethics: landowner consent, council approval, and safety plan.
- Analysis plan: the statistical test and how confounders will be controlled.