Developing investigative skills, and the conventions of mapping and graphical work

Overview

Sensible order of operations when processing a data set and point of putting a frame round a map or diagram.

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

   At A2 you are expected to develop your investigative skills further by collecting and processing data yourself from primary and secondary sources. What does handling raw data add that reading someone else's finished results does not?

   Model answer

   Working with raw data forces the decisions that a published summary has already made invisible: which cases to include, how to group them, which average and which measure of spread describe them fairly, which technique to test them with, and what to do with anomalies. Because you made those decisions you know the assumptions behind the figures and can judge how much weight the conclusion will bear. A finished chart or headline statistic conceals all of that, and can only be accepted or rejected, not interrogated. This progression in investigative skill is an A2 expectation, building on the collection and processing skills established at AS.

2. 2

   Which secondary sources can be processed as though they were raw data, and what can be done with them?

   Model answer

   Census tables, government and agency databases, river gauge and rainfall records, house price and land registry data, traffic and air quality monitoring records, and historic maps and photographs. Rather than quoting the published summary, the figures can be recalculated at a scale that suits the question - combining wards into districts, working out rates per thousand rather than totals, deriving change between two dates, ranking places for a correlation test, or computing a location quotient. Processing at this depth is expected at A2.

3. 3

   How can processing raw figures yourself reveal something a published summary hides?

   Model answer

   A summary usually reports a single average for a whole area, which conceals the distribution behind it: the spread, the skew, any anomalous cases, and any internal contrast between parts of the area. Recalculating from the raw figures exposes those, so a district reported as having average deprivation may turn out to contain both the most and the least deprived wards. The published boundaries and time periods may also be the wrong ones for your question, and only the raw data allow you to regroup them. Working at this level of detail is what the A2 requirement to improve your investigative skills through collecting and processing data amounts to in practice.

4. 4

   What is the sensible order of operations when processing a data set?

   Model answer

   Check and clean the data first, looking for missing values, wrong units and obvious transcription errors, and note rather than delete anything unusual. Then summarise, with an appropriate average and a measure of spread. Then display, choosing the graph or map that suits the type of data and the question. Then test, choosing the statistical technique that matches the question and the data type, and reading the result against the provided table. Only then interpret, relating the figures back to the hypothesis. Displaying before checking, or testing before deciding what the question is, is how misleading results get published. Carrying out this whole routine on data you have collected yourself is an A2 expectation.

5. 5

   How would you decide whether to process data by hand or in a spreadsheet?

   Model answer

   By the size of the data set and the number of times the operation must be repeated. A handful of readings is quicker and safer done by hand, and doing at least one calculation by hand demonstrates that you understand what the technique is doing. Large data sets, repeated calculations across many sites, and anything requiring the same graph to be drawn several times are far better in a spreadsheet, which removes arithmetic slips and allows the effect of changing an assumption to be seen instantly. The risk of the spreadsheet is that a wrongly entered formula produces a confident wrong answer, so results should always be checked for plausibility against the raw figures. Handling data sets at this scale belongs to the A2 development of investigative skills.

6. 6

   Which conventions must be followed in every mapping, drawing and graphical technique?

   Model answer

   Five: a title, a key, a scale, a frame, and a direction arrow showing north. They apply to all the mapping, drawing and graphical techniques used in geography, not only to the ones that look most like maps, and marks are lost for each that is missing. In practice a graph also needs both axes labelled with their units, which is how scale is expressed on a graph.

Other ways to work this

The same content, four other ways.

What is inside Levo

Six ways to work the same material.