Geographical data, surveying and GIS
A geographical question needs suitable evidence
Geography investigates where things are, how places differ and how physical and human processes connect. A useful enquiry begins with a focused question, such as whether access to a market varies with distance from a main road. The question determines which locations, measurements and time period are relevant. Primary data are collected directly for the enquiry, while secondary data were collected or published by others. Neither type is automatically reliable. Check definitions, methods, dates, coverage and whether the evidence actually answers the question being asked.
Quantitative data describe amounts, counts or measured values, while qualitative data describe characteristics, experiences or meanings. A traffic count and an interview about transport difficulties can complement each other. A sample surveys part of a population because a complete census may be impractical. Random sampling gives units a defined chance of selection; systematic sampling uses a regular interval; stratified sampling represents meaningful subgroups. Convenience sampling is easier but can introduce bias if accessible locations differ from those omitted. State the target population before claiming that a sample represents it.
Field observation and ethical practice
Standardise what is counted or measured. For a traffic survey, define the counting point, vehicle categories, direction, observation duration and treatment of unusual events. Counts during one quiet hour cannot automatically represent an entire week. For an environmental-quality survey, clear scoring criteria improve consistency between observers. Photographing a site can record context, but a selected image shows only one viewpoint and moment. Do not treat a dramatic photograph as a complete account of an area.
Human fieldwork requires consent, respectful questions and protection of personal information. Avoid collecting names or precise home locations unless necessary and authorised. Do not enter private land, dangerous roads, floodwater or unstable slopes for a school task. Prepared maps and datasets can support the same analytical skills when field access is unsuitable. A limitation should explain its likely effect, for example that surveying only roadside shops may miss informal sellers away from the road. Merely writing “human error” does not identify a useful source of uncertainty.
Presenting numerical evidence
A bar chart suits separate categories, while a line graph can show change across a continuous sequence such as time. A histogram represents grouped continuous data, with adjoining bars and appropriate treatment of unequal class widths. A pie chart shows shares of a whole and requires all categories to use the same total. For a category containing 30 of 120 observations, the share is 25% and its pie angle is 90°. Do not compare pie slices from different totals as though they necessarily represent equal absolute numbers.
The summarises a total divided by the number of values, the is the middle value after sorting and the mode is the most frequent value. Extreme values can affect the strongly. Range is maximum minus minimum, giving a simple measure of spread. A scatter graph can reveal association between two variables, but correlation is not proof of causation. An apparent relationship between shop count and road access may also reflect population or planning decisions. Explanations require consideration of alternatives and the scale of analysis.
Basic surveying principles
Surveying measures positions, distances and directions to represent features accurately. Chain or tape surveying can divide an area into triangles and use measured baselines and offsets. Triangles provide a stable geometric framework when measured carefully. A prismatic compass can help measure bearings for a traverse. An open traverse ends at a different location, while a closed traverse returns to its start or another known point, allowing a check on closure. A failure to close indicates accumulated errors or inconsistencies that need examination rather than concealment.
Measurements require suitable instrument alignment, consistent units and attention to slope. A distance along sloping ground differs from its horizontal equivalent. Magnetic bearings also differ from true or grid bearings where declination applies. Record the north reference and date when that distinction matters. Obstacles, magnetic interference and reading errors can affect compass work. School surveying should be supervised and use safe open areas. A sketched map can communicate a pattern without pretending to have the accuracy of a professional legal boundary survey.
What a GIS contains
A Geographic Information System combines spatially referenced data, software, hardware, methods and people to store, analyse and display geographical information. Vector data use points, lines and polygons, such as a borehole, a road and a district boundary. Raster data divide space into cells, useful for satellite imagery, elevation or rainfall surfaces. Attribute tables connect features with properties such as name, population or road type. A location without an appropriate coordinate reference can be plotted in the wrong place even if its numerical coordinates appear precise.
Layers allow related information to be compared. A flood-risk enquiry might combine elevation, drainage, land cover and settlement locations. An overlay identifies spatial relationships, while a buffer defines a zone within a stated distance of a feature. A 500 m road buffer shows straight-line proximity under the chosen method, not necessarily walking distance or travel time. Network analysis follows routes and their connections. The analytical operation must match the question; choosing a colourful map does not make the result meaningful.
Remote sensing and data quality
Remote sensing collects information from a distance, commonly through sensors on satellites or aircraft. Different bands can help distinguish vegetation, water and built surfaces, but interpretation depends on sensor properties and processing. Spatial resolution describes the ground detail represented by pixels, while temporal resolution concerns how often observations recur. Cloud cover can limit some optical images. A land-cover classification contains uncertainty, especially where categories mix within pixels or seasonal conditions alter appearance. Ground verification can help assess accuracy where safe and appropriate.
GIS challenges can include outdated data, incompatible formats, limited equipment, training needs, incomplete coverage and privacy concerns. A dataset from a credible institution can still be unsuitable if its scale or date does not match the enquiry. Aggregated district averages may conceal differences among neighbourhoods. When mapping rates, use the relevant denominator rather than raw counts if the question concerns relative risk or prevalence. State source, date, units and classification choices so another person can assess the map.
Self-check and interpretation
Choose a choropleth map for comparable area-based rates, a proportional-symbol map for quantities at locations and a flow map for movement between places. Explain why population counts alone are not the same as population . For a GIS result, state what operation was performed and what its output can and cannot establish. A good geographical conclusion connects the pattern with the original question, cites specific evidence and identifies a meaningful limitation. The aim is a defensible explanation, not the most complicated display.