Creating feature data with R and QGIS

Key questions

Data does not exist
Data exists but…
- you can’t get it
- you aren’t allowed to use it
Data exists and you can get it but…
- The format can’t be accessed effectively
- The geometry is inaccurate
- The geometry too general

Collect non-spatial information (e.g. a set of addresses or a survey that asks respondents to provide spatial information) then add spatial data to it
Collect spatial information using a device/sensor with a built-in GPS or using a smartphone application (e.g. Strava or StreetComplete)
Collect spatial information by drawing features using a web application or desktop GIS

Different situations require different approaches. Privacy, cost, accuracy, ease of use, or training requirements are all relevant factors.

Do you need points, lines, polygons, or something else?
Do you need a consistent representation of the spatial characteristics of “real world” conditions or phenomena?
What scale or resolution is appropriate for your data?
Should your data be “snapped” (or related) to another existing feature?

Again, your approach depends on what you plan to do with the data in the future. Is your primary goal visualization? Analysis? Data sharing?

There are several popular web applications that support the creation of spatial data:

Google MyMaps: Supports import of CSV, TSV, KML/KMZ, GPX, XLSX, Google Sheet, and photos with EXIF location data. Few recent updates.
geojson.io: Supports GeoJSON import/export along with KML, GPX, CSV, GTFS, TopoJSON, and other formats, raster tile layers. Recently updated by Mapbox.
Felt: Supports collaborative editing and a wide range of embedded rich media.

How to decide? Look at pricing, data import and export options, collaboration needs.

Open geojson.io
Optional: Load a xyz tile layer using a rectified map from Map Warper
Using the basemap (or your tile layer) as a reference zoom and pan to an area of interest
Select the point, line, polygon, or rectangle tool from the options on the right side of the map pane
Add a few features to the map
Select a feature from the map and use the pop-up table to enter an attribute name (left side) and value (right side)
Switch to the table view on the right to see the attribute data
Select the Edit map tool from the options on the right side of the map pane. Try editing a feature (make sure to save your edits).
Select Save > GeoJSON from the menu to export data to a GeoJSON file

Are there any existing data standards you can use or adapt?
Is there an existing unique identifier you can use or adapt? Or do you need to assign an arbitrary unique identifier? Is your unique identifier stable or subject to change?
If you are basing your data on an existing source, do you transcribe all attributes in the original source? Or are you only transcribing some of those attributes?
Can you use a “flat” table structure? Or do you need multiple tables or a nested format?
Can you validate your attributes?

Google Forms offers built-in validation by limiting inputs, requiring a set length, or requiring input text match a specific pattern
Google Sheets and Excel both provide

Create a Google Sheet (I’ll provide it this time)
Set up columns for longitude and latitude (or for a partial or full address). Note: Separate street address, city, and state columns can be combined using formulas.
Set up additional columns as needed. Make sure to avoid merging column ranges and keep all column names in the first row.
Consider creating a separate “data dictionary” sheet where you can use the TRANSPOSE() formula to switch column names into rows.
Consider using the “Data validation” menu to add validation to any columns where appropriate.

The mapedit package offers multiple editors: