- Created by: zelahl32
- Created on: 18-11-16 16:46
Located Symbol Maps
- The size of simple symbols is proportional to the data found at that location.
- The bigger the symbol, the larger the data.
- They are flexible --> numerical, ordered categorical data.
- They can be used for data where location is needed,
- It is easy to extract the number from the symbol by estimating the area.
- There is a lot of overlapping of symbols, making it harder to read.
- Most map readers cannot accurately read off the values, which can cause bias.
- It is difficult to differentiate between similar data values.
- Litres of water consumed per capita per country in 2015.
- Location and magnitude of earthquakes in California from 1900-2010.
- Population totals of the largest cities in the world.
- They represent data as a column, either as separate columns or on top of each other to equal a certain value e.g. 100% (composite bar graph).
- The x-axis width usually stays the same while the y-axis length changes.
- They can summarise a large data set in visual form easily.
- You can compare several variables at the same time.
- It can be easily understood by anyone.
- It can be hard to read off exact data values.
- You cannot necessarily use bar graphs for continuous data e.g. length, weight, height.
- They cannot show key assumptions and causes.
- Number of visitors to a park every year from 2000 - 2010.
- The genre of books read by 100 people.
- The number of days of snow from December - February.
- Is used to represent continuous data using dots where the raw data is and a line of best fit is drawn through the dots.
- You can estimate other data values using the line of best fit.
- They clearly show any patterns in the data.
- They can compare multiple data sets easily.
- Comparing data sets is only useful if the data uses the same scales on their axes.
- You can only use numerical continuous data for the y-axis.
- Line graphs can easily be manipulated to show something.
- The temperatures in New York City across a week.
- The discharge rate of the Red River at the mouth across a year.
- A map with continuous lines that joins points of the same values.
- It shows gradual change and patterns over a large area.
- It uses fixed intervals, so changes can be easily identified.
- You can add colour to emphasise the trends.
- It can only be used with located data.
- It usually requires getting a huge amount of data, which can be costly to get.
- It is implied that the values between the lines don't change.
- In an OS map showing contour lines of equal elevation.
- On a world map showing where the temperatures differ.
- Showing isobars (barometric pressure) on a regional map.
Flow and Trip Maps
- Flow map: Represents movement along a given route. The lines have variable width.
- Trip map: Represents movement from one area to another using a straight line with no specific route.
- Allows you to clearly see movement.
- The data is clearly located.
- You can use almost any scale map from regional to local.
- Maps lack precise data needed, so any numbers that are similar yet different e.g. 4600 and 5200, can;t be told apart.
- The route of this movement is not taken into account.
- The meeting of several thick lines can overwhelm the map.
- To show migration in an area.
- The imports and exports of goods.
- Graphs that have three axis instead of two, forming an equilateral triangle.
- Each axis is split into 100 parts, representing percentage, allowing 3 variables to be plotted against each other.
- You can show the relationship between three variables.
- There is no issue with scaling because it is always percentage.
- You can easily see any dominant variables.
- Can only be used when a whole figure can be broken down into 3 components expressed as a percentage.
- It can be hard to interpret.
- Showing the percentage of elements in a chemical composition.
- Used in plotting employment structures (primary, secondary and tertiary) in companies.
- Can be used to show soil structure.
- Used to represent both categorical and numerical data on two axis.
- The x-axis has one or more categories and the y- axis has the numerical scale. A line is drawn through the centre to represent zero.
- They are easy to interpret.
- They are good for displaying changes over distance.
- There are visual and you can easily distinguish between categories.
- Comparisons can be made.
- They can be manipulated by changing the scale.
- It only works with a specific type of data.
- It could be hard to identify anomalies without background knowledge.
- Showing the numbers of specified plants along a transect line or a distance away from the starting point.
- A type of graph where values extend out from a central point, which show the relationship of each variable to the central point.
- Concentric circles are used, where each circle has a value assigned to it.
- More than one axis can be used, allowing you to plot several variables simultaneously.
- You can clearly see trends in the data.
- They are useful for data with directions or data that is time-based.
- Using over 5 variables can make interpreting the graph complicated.
- It can be difficult to find a suitable scale to use.
- It can be difficult to extract precise data from the graph.
- Wind rose diagrams --> used by meteorologists to show the speed and direction of wind in a given location.
- Showing the number of people in a shopping centre at certain hours throughout the day.
Logarithmic Scale Graphs
- Graphs that use a non-linear scale when there is a large range of quantities. Usually in the form of line graphs or bar charts.
- It allows you to work with a large range of numbers, which allows you to see a better overall trend.
- It avoids certain ranges of data, i.e. really small or really large, getting compressed.
- Zero can't be plotted.
- Positive and negative values can't be plotted at the same time.
- It can be difficult to interpret data directly from the graph because the axis are distorted.
- Studying population data.
- The Hjulstrom curve.
- The Richter magnitude scale for earthquakes.
- Magnitude frequency flood risk analysis.
- Where areas are shaded according to a key made previously; each colour or shade represents a range of values.
- There are good for indicating differences in land use.
- It can be effectively used to report data scales to a local scale.
- You can easily find trends and anomalies.
- They provide a way of visualising how measurements vary.
- They give a false impression of abrupt change at the boundaries.
- It can be difficult to distinguish between shades on the map.
- Boundaries of unit areas are sometimes vague e.g. the North.
- The population density of a country by region.
- The land use in the centre of a large city such as London.
- Showing what the electorate voted during an election.
- When dots are used to show density differences in geographic distributions across a location.
- There are two types; one-to-one dot density maps, and one-to-many dot density maps.
- You can map raw data/rates/ratios e.g. a number of farms per square kilometre.
- They do not require colour to be interpreted.
- They can be used to represent a wide range of data.
- It is time consuming to draw.
- They must be drawn on an equal area map projection otherwise the perceived density of the dots will be distorted.
- They're terrible for retrieving rates or numbers from the map.
- The distribution of car dealerships in Belgium (1 dot = 1 dealership).
- Earthquake epicentres across the Pacific for the past 10 years.
- Number of people in the UK by county (1 dot = 10,000 people).
- A quantity expressing by how much the members of a group differ from the mean value of the group.
- Shows how much data is clustered around a mean value.
- It is not as affected by extreme values.
- It gives a more accurate idea of how the data is distributed.
- It assumes a normal distribution pattern.
- It doesn't give you the full range of data.
- It can be difficult and time-consuming to calculate.
- Finding the standard deviation of the world population.
- Calculating the spread of test scores within a class or school.
- Spearman's Rank is a test which produces a value telling you how strong the relationship between two sets of data is.
- It shows the significance of the data.
- Proves/disproves correlation using levels of significance tables.
- It doesn't assume normal distribution.
- It has quite a complicated formula and a large number of steps.
- It can be misinterpreted.
- Need 2 sets of variable data so that the test can be performed.
- The correlation between the temperature and the number of ice creams sold.
- A test used to determine whether there is a significant association between two categorical variables.
- It is used to look at differences between what you found and what you expected.
- Chi square makes no assumption on the distribution of the sample.
- It is easier to work out than some other statistics.
- The Chi-square test is sensitive to sample size.
- It does not give us much information about the strength of the relationship.
- All participants measured must be independent.
- Used when in each category, the data is displayed as frequencies.
- Seeing the relationship between the type of subject people study and the style of lecture they prefer.
Mann Whitney U
- A non-parametric test that is used to compare whether two population means are equal or not.
- It states whether the difference is significant or occurred by chance.
- Shows the median between 2 sets of data.
- You can use data sets of different sizes.
- Good with dealing with skewed data.
- It is a lengthy calculation so it is prone to human error.
- Does not explain why there is a difference.
- Becomes less accurate when the sample size is below 5 or above 20.
- This test is used for ordinal data (rankings), which requires a nonparametric test.
- Comparing the effectiveness of branding for two rival companies.