Identifying appropriate sources and methods

Overview

The difference between a source and a method, the secondary sources that support a human and a physical enquiry, and how group-collected data is kept comparable.

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 is the difference between a source and a method?

   Model answer

   A source is where the information comes from - the river channel itself, the shoppers in a street, the census, an old map, a gauging station record. A method is how you extract information from it - measuring depth at fixed intervals with a metre rule, administering a structured questionnaire to every fifth passer-by, extracting age structure by ward from a census table. A plan must state both, because naming a source without a method says nothing about how the data will be obtained or how reliable they will be.

2. 2

   Which secondary sources would support a human geography enquiry, and what does each provide?

   Model answer

   Census tables, giving population totals, age structure, housing tenure, employment and travel-to-work data by area; local authority databases and planning documents, giving development proposals, land use and deprivation measures; maps of successive dates, showing the growth and changing layout of settlement; trade directories and historic photographs, showing how the character of an area has changed; and published retail, tourism or transport statistics giving figures no fieldwork day could generate. Newspaper and council records add the sequence of events and the arguments around an issue.

3. 3

   Which secondary sources would support a physical geography enquiry, and what does each provide?

   Model answer

   River gauging station records of discharge and level over years, which set a day's measurements against the long-term regime; rainfall and temperature records from weather stations; geology and soil maps, which explain resistance, permeability and vegetation; Ordnance Survey maps at several scales, giving relief, drainage and gradient; historic maps and aerial photographs, from which coastal retreat, channel change or glacier retreat can be measured; and satellite imagery for land cover and flood extent.

4. 4

   What does census data contain, and how is it used in an investigation?

   Model answer

   It records the population of small areas at a fixed date, giving numbers, age and sex structure, household composition, housing type and tenure, employment by sector, car ownership, health and travel to work. It is used to describe the social and economic character of the areas being studied, to select contrasting areas for comparison, to work out the strata for a stratified sample, and to provide the second variable in a correlation with fieldwork measurements. Its figures can be reworked into rates and quotients rather than being quoted as published. Census material is secondary evidence, because it was collected by the state for another purpose; individual census returns are closed for 100 years, so investigations work from the published small-area tables.

5. 5

   What are the advantages and the risks of collecting data as a group rather than individually?

   Model answer

   The advantage is quantity and coverage: a class can survey far more sites, take more repeats and cover a much wider area in one day than an individual, which makes the sample big enough for statistical testing. The risk is inconsistency - different people measure, estimate and score differently, so the variation between sites may partly be variation between surveyors - along with the possibility that one person's careless or invented data contaminates the whole set. Group collection is also safer than working alone, which is itself a reason for it.

6. 6

   How do you keep group-collected data comparable between the people who collected it?

   Model answer

   Agree and write down the exact procedure before anyone starts: the equipment, the units, the number of repeats, the duration of counts, the rules for edge cases, and the wording of any questionnaire. Use a single identical recording sheet for everyone. Where judgement is involved, such as environmental quality scoring, either train everyone against agreed descriptors and worked examples or keep the same person on that task at every site. Check a sample of sites with a second observer, and record who collected each set so that any inconsistency can be traced.

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.