- Data points on the map are joined up with data points of equal value.
- Temperature/atmospheric pressure/gradient e.g. contour lines.
- Shows gradual changes - avoids abrupt change.
- Can clearly see boundaries.
- Can see areas of equal value.
- Assumes a gradual change exists.
- Small numbers/units may be difficult to read.
- Only works with a large quantity of data.
- Independent variable goes on the x-axis e.g. distance downstream.
- Dependent variable goes on the y-axis e.g. discharge.
- Any two variables with a relationship.
- Anomalies are easily identifiable.
- Uses bivariate data which enables you to see whether there is a relationship between the variables - aids interpretation.
- Strength of correlation can be confirmed using a statistical test e.g. spearman's rank.
- Line of best fit - predict future data sets.
- Doesn't show cause and effect.
- 'Overplotting' can be an issue with lots of similar results.
- Have to have continuous data.
- Full log/log-log - both axis are logarithmic scales.
- Semi-log - one axis is linear and the other is logarithmic.
- Population graphs - used with very large ranges.
- Allows you to work with/plot a large range of numbers.
- Shows overall trend/previously unseen patterns that normal graphs do not show.
- Smaller values are given greater priority due to the nature of the logarithmic scale 1-10.
- Postitive and negative values can't be plotted on the same graph.
- Zero can't be plotted
- May show info like a bar chart.
- May show orientation e.g. compass points.
- May show continuous cycles e.g. time.
- Environmental quality survey (if using as a radial bar chart).
- Visual representation of data.
- Displays multiple variables.
- Suitable for only continuous data - limited use.
- May only show general trends if based on averages.
- Can be difficult to read/interpret.
- Employment structures (tertiary, secondary, primary)/ethnicity.
- Visual representation of the relationship between three variables.
- Percentages are plotted - especially easy to compare/contrast.
- Shows clusters of data.
- Raw data must be converted into %.
- Can be difficult to interpret, especially if there is a lot of data plotted.
- Thickness shows number/percentage of each special at a point in time.
- Thickness is balanced equally below and above the line.
- Distribution of plant species along a transect of a sand dune.
- Visual representation of change and progress over a specific distance.
- Uses raw data and percentages.
- Comparisons can be made between different species - can identify zones.
- Limited to the transect lines.
- Only suitable for specific data with a specific purpose.
- Visually subjective - the scale used can affect the diagram.
Proportional pie charts
- x/y x 360 = number of degrees, when x = variable and y = total.
- Proportions = square root total.
- Use of services in a town/amount of crops grown in a certain area.
- Clear, visual representation of data.
- Able to compare easily.
- Relatively easy to construct.
- May not show numerical data.
- May get crowded if there are too many divisions.
- Categoric data only.
- Population of a city/country/region.
- Shows spacial distribution and density.
- Anomalies shown if there is a lot of data.
- Clustering and patterns identifiable.
- Large amount of data may lead to overcrowding.
- Areas may seem empty if the data is lower than the scale.
- Large dot values may be inaccurate.
Map showing movement - trip, flow and desire lines
- Flow - width of arrow represents the flow rate and direction the flow is moving in e.g. migration.
- Desire - where a quantity moves from orgin to destination e.g. migration
- Trip - shows regular trips e.g. where people shop.
- Strong visual impression of movement.
- Clear sense of direction.
- Can be hard to interpret if map becomes obscured.
- Can be difficult to draw.
- Difficult to show the meeting point of wide bands.
- GNP per country/levels of extreme poverty per country/region.
- Visual representation of data.
- Can easily identify patterns/clusters.
- Anomalies identified if cells are an adequate size.
- Assumes abrupt change at boundaries - no gradient shown.
- Can hide anomalies within an area.
- Shows one variable only.
- Simplistic view of sample site and main features.
- Can pick out features to annotate/comment on.
- Qualitative - based on obervation and personal perspective - may be bias.
- May be hard to interpret if skill of drawer isn't good.
Cross sectional diagrams
- Shows what the area looks like before you have been.
- Only show a snapshot in time.
- Suseptable to external influences e.g. weather.
- Only show a small section of the area.
- Identifies the most important features.
- You can add as much/as little data as you'd like.
- Shows your interpretation of the area.
- Qualitative - may be bias.
- Only shows one view at one point in time.
- May lack detail.
- May be difficult to interpret depending on drawer.
- Cartographic modelling/determining land use, soil, vegetation, elevation, land ownership, characteristics.
- Bypass the mechanical processes of mapping.
- Higher quality.
- Expensive - can't be used all over the world.
- Time consuming
- Needs regularly updating.
- Databases/census data/capture/store/analyse data.
- Easy to make comparisons over time.
- Saves time.
- Can be converted into graphs - visual representation.
- Can be expensive.
- Not available everywhere.
Measures of central tendancy - mean, median, mode.
Mean - Average of all the values. Total number of data sets/(divided by) the number in the sample.
Median - Middle value. If there are two middle values, add them together and divide by 2.
Mode - Most common value.
- Clear and simple.
- Mode can me used with non-numerical data.
- Median - very large and small numbers do not affect result.
- Mean - useful in making measurements more accurate.
- Can't use continuous data.
- Median and mode do not account for whole spread of data.
- Mean is easily distorted by very large/small values and anomalies.
- Count the number of values.
- Work out the LQ ranking: 3(n+1)/4
- Work out the UQ ranking: (n+1)/4
- Find out the data set/value that matches the ranking for each quartile.
- Minus the LQ value from the UQ value to find the interquartile range.
For box and whisker, the mediam - (n+1)/2
- Shows spread of data around the mean.
- Not influenced by extreme/outlying data sets.
- Not all data is considered
- Complicated to calculate
- Calculate the mean and minus it from the value in the column.
- Square each answer and add up the totals.
- Submit into equarion - this will give you your standard deviation number.
- + and - the standard deviation number from the mean, this shows you the range of data around the mean.
- Work out how many of the data sets are within the range.
- Number of data sets in the range/total number of values x100 = %
- If answer is above 68% then the data is close to the mean. Lower than 68% - 2nd SD needs to be taken by doubling the SD score and + and - that from the mean to find the new range - the answer needs to be 95% confident.
- More accurate than the range as it uses all of the data - more accurate.
- Low S.D score means small range so the mean is more reliable as there is little variation.
- Can be affected by anomalied/outliers.
- Display the main pattern in the distribution of data.
- Visually effective - full range of data is seen together.
- Useful for making comparisons.
- Data must be in a form that can be placed along a number line.
- Lots of values may lead to clustering - difficult to interpret.
- Identify null hypothesis - no significant difference between observed an expected.
- Subtract observed frequencies from expected and square the result.
- Divide this by the expected value for that group.
- Compare with degrees of freedom: on the critical values chart, the degree will be one less than the total number of observed values.
- To assess the degree of difference between observed and theoretical data e.g. number of pebbles along a river.
- Statistical significance of results can be tested.
- Doesn't explain why there is a pattern.
- Does not give the strenght of the relationship.
- Percentages cannot be used.
Spearman's rank correlation co-efficient
- Formulate a null hypothesis.
- Individually rank the values of each variable. 1 = highest value.
- Find the difference between the two.
- Square the differences and sum the values.
- Input into the formula.
- Appropriate for data with 10-30 values with 2 variables that are believed to be related.
- Indicates the statistical significance of a result - rules out chance.
- Gives numerical value to the strength and direction of a correlation.
- Does not show if there is a casual link
- Too many tied ranks affect the validity of the test.
- Subject to human error.
- Select null hypothesis.
- Rank the data sets across the two columns. 1 = lowest value.
- Treat as two seperate columns. Add ranks in first column to get your R1 value then add ranks in the second column to get your R2 value.
- Input int the formula.
- Choose the smaller U value of either U1 or U2.
- Compare to the critical values table: less than the critical value means you should reject the null hypothesis at 95% confident. Greater than the critical value - accpet the nul.
- Used to show if there is a statistical difference between two sets of data e.g. size of rocks in upper course and lower course.
- Does not explain cause and effect.