Survey | The use of core concepts during geospatial analysis

04-08-2023 07:22 AM
New Contributor

For my research, I am looking for participants with general knowledge of GIS.

The research tries to find behavioral evidence on whether GIS users make use of core concepts as proposed by Werner Kuhn (2012) during geospatial analysis. For anyone with the time to fill in the survey, your response is much appreciated! It is estimated to take around 7 - 12 minutes.

0 Kudos
2 Replies
MVP Frequent Contributor

Hey Casper,

just did the survey. A few pointers:

  • I missed a "back" button. You present a definition at the start of each section. I can't go back to check that definition if I have a question in the middle of a section.
  • The use of ArcGIS documentation is problematic.
    • It makes it hard to include GIS users who are accustomed to other GIS software.
    • Sometimes you link to documentation of a concept (like areal interpolation), but often you link the tools directly. Again, bad for non-ArcGIS users and for users not experienced with a certain tool
    • For example, I wondered why you often asked for Euclidian Distance (a tool I'm not familiar with at all). The tool summary says this: "Calculates, for each cell, the Euclidean distance to the closest source.", and there is an image with raster cells. Alright, a tool that works with rasters, not applicable for a point dataset. Except that it is, because you can also use point features as input.
      A user that used that tool in the past might know that. Users like myself and non-ArcGIS users probably don't. A link that explains the concept of Euclidian distance as opposed to the way it is done in ArcGIS would be way more helpful. Also for you, because you would get more meaningful and fewer wrong answers.
  • "meaningfully applicable" can be tricky.
    • In one question, you ask if you could meaningfully apply a Dissolve on a point layer with a unique field. Yes, you can do that, but the output will be the same as the input. The output is conceptually sound (I think that was part of the definition, again, a back button would be nice), but you didn't do anything meaningful.
    • In some other question, you couldn't apply a tool to a dataset, but you could apply it to a result of an intermediate step. For example, you asked if you can apply areal interpolation to a point dataset. Not directly, but you can apply it to Thiessen polygons you create from the points. Maybe you should clarify that you don't (or do) want to consider intermediate steps.

Have a great day!
0 Kudos
MVP Esteemed Contributor

dissolve a point layer based on an attribute will produce a multipoint

... sort of retired...
0 Kudos