Data: Source and Compilation

The One Thing

Geographical data become useful knowledge only when they are collected from suitable and reliable sources, carefully compiled, critically evaluated, and presented in an appropriate form. The choice between primary and secondary sources depends on the purpose, scale, time period, and type of information required.

Who and What

• Data: Facts, figures, measurements, or observations collected for analysis and interpretation.
• Geographical data: Information concerning the location, distribution, characteristics, and relationships of physical and human phenomena on the Earth’s surface.
• Primary data: Original information collected directly by an investigator for a specific study. Sources include field surveys, direct observation, interviews, questionnaires, measurements, experiments, GPS readings, and field photographs.
• Secondary data: Information already collected, processed, and published by another person or organisation. Sources include population censuses, government reports, administrative records, research studies, books, journals, maps, atlases, newspapers, websites, and digital databases.
• Census: Complete enumeration and collection of information about every unit in a population. A population census attempts to cover every member or unit of the population.
• Sample survey: Collection of information from a selected representative part of a population rather than from every unit. A representative sample should reflect important characteristics of the population; biased sampling can produce misleading conclusions.
• Questionnaire: A planned list of written questions used to collect information from respondents.
• Schedule: A list of questions filled in by an enumerator after asking the respondent.
• Observation: Collection of information by directly watching, measuring, or recording phenomena in the field.
• Interview: Collection of data through direct questioning and discussion with individuals or groups.
• Remote sensing: Collection of information about the Earth’s surface from a distance using sensors on satellites or aircraft.
• Qualitative data: Non-numerical information describing qualities, conditions, opinions, or characteristics.
• Quantitative data: Numerical information that can be measured, counted, and statistically analysed.
• Discrete data: Countable numerical data expressed in separate values, such as the number of households.
• Continuous data: Measurable data that can take any value within a range, such as temperature, rainfall, or height.
• Spatial data: Data connected with a specific location or geographic position.
• Attribute data: Descriptive information about a geographic feature, such as the population or land-use type of a place.
• Compilation: The process of arranging, classifying, checking, summarising, and presenting collected data.
• Classification: Grouping data into categories according to common characteristics. Data may be classified by time, place, quality, quantity, or other relevant characteristics.
• Tabulation: Systematic presentation of data in rows and columns.
• Frequency: The number of times a particular value or category occurs. A frequency distribution shows how often different values or class intervals occur.
• Data source: The person, institution, document, survey, instrument, or technology from which data are obtained.
• Reliability: The extent to which data are consistent, accurate, and dependable.
• Validity: The extent to which data actually measure the feature or condition intended to be studied.
• Metadata: Information describing the source, date, method, scale, units, and limitations of a dataset.
• Digital databases and Geographic Information Systems: Technologies that can store, combine, analyse, and display spatial and attribute data.

Causes and Consequences

• The purpose and scale of a study determine the appropriate data source. Primary data are usually more specific to a research problem, whereas secondary data are often quicker and less expensive to obtain. No single source is suitable for every geographical investigation.

• Different sources provide different forms of geographical evidence. Maps and remote-sensing images are especially useful for spatial data, while surveys and censuses are useful for demographic and socio-economic data. Combining census data, field observations, maps, and satellite images can provide a more complete understanding of a geographical issue.

• Collection must be followed by systematic compilation. The principal stages are editing, coding, classification, tabulation, calculation, interpretation, and presentation. Editing identifies incomplete, inconsistent, duplicated, or incorrect entries before analysis, while coding assigns symbols or numerical values to responses so they can be classified and processed easily.

• Classification and tabulation make scattered observations comparable. Data can be organised by time, place, quality, quantity, or other relevant characteristics and presented in rows and columns. Frequency distributions show how often different values or class intervals occur.

• Quantitative analysis depends on appropriate calculations and units. Class width in a grouped frequency distribution is calculated as:

$$\text{Class width}=\frac{\text{Upper class boundary}-\text{Lower class boundary}}{\text{Number of classes}}$$ Percentage is calculated as: $$\text{Percentage}=\frac{\text{Part}}{\text{Whole}}\times 100$$ The arithmetic mean is: $$\text{Mean}=\frac{\text{Sum of observations}}{\text{Number of observations}}$$ For a frequency distribution: $$\text{Mean}=\frac{\Sigma fx}{\Sigma f}$$ where $f$ is frequency and $x$ is the value or class midpoint. Data should include appropriate units, such as kilometres, hectares, degrees Celsius, millimetres, persons, or percentages.

• Critical evaluation is necessary because data quality affects geographical conclusions. A source should be checked for accuracy, authority, date of collection, method used, scale, coverage, and possible bias. Errors may result from poor measurement, incomplete coverage, inaccurate recording, outdated information, unsuitable classification, or personal bias.

• Reliable data improve the validity of interpretation. Reliability concerns consistency, accuracy, and dependability, while validity concerns whether the data measure the intended feature or condition. Metadata assist evaluation by recording the source, date, method, scale, units, and limitations of a dataset.

• Clear presentation improves communication and comparison. Data may be presented through tables, bar graphs, histograms, pie charts, line graphs, dot maps, choropleth maps, proportional symbols, and flow maps. Appropriate diagrams and maps make geographical data easier to understand, compare, and communicate.

• Ethical procedures protect participants and the integrity of research. Confidentiality, informed participation, honest reporting, and proper acknowledgement of sources are important requirements in data collection and use.

What Gets Asked

• Compare primary data and secondary data in terms of specificity, cost, speed of collection, and examples of sources.
• Distinguish between a population census and a sample survey, and explain why a representative sample is necessary.
• Explain how editing, coding, classification, tabulation, calculation, interpretation, and presentation convert scattered observations into organised geographical information.
• Evaluate the reliability and validity of a data source using its accuracy, authority, date, method, scale, coverage, metadata, and possible bias.
• Explain the differences between qualitative, quantitative, discrete, continuous, spatial, and attribute data, using the named examples provided.
• Apply the equations for class width, percentage, arithmetic mean, and frequency-distribution mean, and select suitable tables, graphs, maps, or GIS-based presentations for particular data.