Choosing an issue, hypothesis or question

Overview

What makes a topic workable, why it should grow out of work already done, how a broad interest is narrowed into a testable hypothesis, and the signs that one is still too vague.

What you need to know

Answer each one before you reveal it.

Questions from this subtopic

Each one with the model answer underneath.

1. 1

   What makes a topic suitable for a geographical investigation?

   Model answer

   It must rest on a geographical idea rather than simply on curiosity, so that there is a process or a relationship to test; it must vary across space or over time, since a topic that is the same everywhere gives nothing to measure; the data must be collectable safely and legally within the area and time available; and it should be narrow enough that the evidence can settle it. A topic that meets all four gives an enquiry with a real answer at the end of it.

2. 2

   Why should the investigation grow out of work already studied rather than being chosen from nowhere?

   Model answer

   Because the theory studied supplies the expected pattern, the vocabulary and the explanations that turn measurements into geography. Knowing how a channel is predicted to change downstream, or how land use is predicted to vary with distance from a centre, tells you what to measure, where to measure it and what a departure from the prediction would mean. An investigation with no theoretical background collects numbers with nothing to compare them against, and the analysis has nowhere to go.

3. 3

   How would you turn a broad interest into a focused hypothesis?

   Model answer

   By narrowing each vague element in turn until every word implies a measurement. An interest in coastal erosion becomes a question about one stretch of coast, then about one measurable variable, and finally a statement such as beach material becomes smaller and rounder with distance eastwards along the spit. Ask of each draft: what exactly would I measure, where, and how would I know if it were false. If any of those cannot be answered, the wording is still too broad.

4. 4

   Give an example of a workable hypothesis for a physical geography investigation and explain why it works.

   Model answer

   Pebble roundness increases with distance along the beach in the direction of longshore drift. It works because it names one measurable variable, roundness scored on a standard index; one controlling variable, distance measured along the shore; a defined study area; and a predicted direction of change that the data could contradict. The paired data it generates can be plotted on a scatter graph and tested with Spearman's rank correlation, and the outcome connects directly to the process of attrition during transport.

5. 5

   Give an example of a workable hypothesis for a human geography investigation and explain why it works.

   Model answer

   Pedestrian density decreases with distance from the peak land value intersection of the town centre. It works because pedestrian density is countable at fixed sites for a fixed time, distance from the centre is measurable on a map, both give figures that can be ranked and correlated, and the prediction is directional so the data can refute it. It also links to established ideas about the bid-rent and land value gradient within an urban area, which gives the analysis something to argue with.

6. 6

   Why might an investigation use more than one hypothesis?

   Model answer

   Because a single measurement rarely settles a geographical question on its own, and several related hypotheses tested together give a much stronger conclusion. Testing that channel width increases downstream, that velocity increases downstream and that bedload size decreases downstream builds a coherent picture that any one of them alone would not, and where two agree and one does not, the disagreement is informative. The limit is practical: each hypothesis needs its own data, so too many produce shallow evidence for all of them.

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.