GIS / Mapping Working Group

Public Group active 18 hours, 35 minutes ago

Top Mapping Mistakes

Tagged: ,

Viewing 13 posts - 1 through 13 (of 13 total)
  • Author
    Posts
  • #114227

    Hi all,

    I’m writing an article now that I hope will be helpful for mapping beginners. It’s about the top mistakes that folks make when they are first learning GIS. Are there any that come to mind for you that you remember from your mapping journey or from teaching others? It would be things like forgetting to make sure the field types match when you are doing a spatial join. Any examples that come to mind for you?

    Thanks so much!

    Olivia

    #114246
    Juliana Maantay
    Participant

    One of my pet peeves is novice GIS-ers making choropleth maps to display absolute numbers (counts), rather than rates, percentages, proportions, per capitas, medians, or averages, etc. Another problem people often get into is relying on default options in GIS software packages, without really thinking about what they are doing and whether or not in the specific analysis or map they are working on makes sense with the defaults.
    Using bi-chromatic color ramps to symbolize a low to high continuum, instead of single color in graduated hues. The bi-chromatic color ramps should really be used only for situations where a classification range goes above and below a standard deviation mid-point, or a threshold value, for instance. Even worse is when people use a multi-colored circus jumble ramp for quantitative data, which negates the intuitive purpose of a single color ramp with gradation, letting the map viewer understand at a glance without much reference to the legend, what the colors on the map indicate.
    I realize that all of these are more or less cartographic issues, not GIS specifically, but because GIS now allows map making almost indiscriminately, it’s important to understand all the ways that maps can be cartographically inaccurate and misleading if one isn’t well-versed in standard cartographic best practices. Some very incorrect maps can be produced which have all the appearances of a professionally created map.

    Juliana Maantay, Ph.D., M.U.P., F.R.G.S.
    Professor of Urban Environmental Geography
    Director: Geographic Information Science (GISc) Program;
    MS-GISc Program; and the Urban GISc Lab
    NOAA-CREST Research Scientist
    Department of Earth, Environmental, and Geospatial Sciences
    Lehman College, City University of New York
    250 Bedford Park Blvd. West
    Bronx, NY 10468
    TEL: (718) 960-8574; FAX: (718) 960-8584
    juliana.maantay@lehman.cuny.edu
    Faculty Profile: http://www.lehman.edu/academics/eggs/fac-maantay.php(https://mail.lehman.edu/owa/redir.aspx?C=JyvINcu5vqHrfPIz4NvBpSPK-oFcbs1Ep7J2CAaJl7JtUdYs4oXVCA..&URL=http%3a%2f%2fwww.lehman.edu%2facademics%2feggs%2ffac-maantay.php)

    “Geography, sir, is ruinous in its effects on the lower classes. Reading, writing, and arithmetic are comparatively safe, but geography invariably leads to revolution.” (1879 testimony before a Select Committee of the House of Commons, London, England, regarding expenditures of the London School Board)

    “The scientist is not a person who gives the right answers, [s]he’s one who asks the right questions.” (Claude Lévi-Strauss, 1964, Le Cru et le Cuit [The Raw and the Cooked])

    “Those who plan do better than those who do not plan, even though they rarely stick to their plan.” Winston Churchill

    “Be sure that no good may be expected of anyone who is satisfied with what he knows. To investigate is the task of the knowledgeable.” (Book of the Sea, 1513, by Piri Re’is, Ottoman cartographer, geographer, and navigator to Suleiman I)

    #114253

    Juliana raises some very important points. I would just add that people new to map-making need to remember that it’s “choropleth” not “chLoropleth”! (A common and annoying mistake!)

    Also, even though the book is a bit dated, everyone in this field should read Monmonier’s “How to Lie with Maps”, because it explains so clearly and effectively why you shouldn’t make the kinds of mistakes that Juliana notes below.

    Steve

    =======================================
    Steven Romalewski
    Director, CUNY Mapping Service
    Center for Urban Research at The Graduate Center / CUNY
    365 Fifth Ave., Room 6202
    New York, NY 10016
    TEMPORARY PHONE: 917-975-3794
    Office phone: 212-817-2033
    sromalewski@gc.cuny.edu<mailto:sromalewski@gc.cuny.edu>
    http://www.urbanresearch.org(http://www.urbanresearch.org&d=DwMFAg&c=8v77JlHZOYsReeOxyYXDU39VUUzHxyfBUh7fw)
    http://www.gc.cuny.edu/urbanresearchmaps
    @SR_spatial
    =======================================

    From: Juliana Maantay (GIS / Mapping Working Group) <noreply@gc.cuny.edu>
    Sent: Wednesday, April 21, 2021 1:54 PM
    To: Romalewski, Steven <SRomalewski@gc.cuny.edu>
    Subject: Re: Top Mapping Mistakes

    #114254
    Ellynne Dec
    Participant

    I am a recent learner, and here are a few things that I find challenging:

    • GIS-specific software (such as QGIS) is great, but there are so many package features and several ways to access these capabilities, some of which can be somewhat redundant. Figuring out how to do any one specific thing – even the simplest things – can be a fairly tedious exercise in trial and error. If you are working in GIS software, definitely make careful notes for yourself from the very beginning, so that you can replicate what you did or retrace your steps if you need to troubleshoot later. It’s a huge headache to get several steps into a mapping project and then realize you made a mistake somewhere but can’t remember the exact steps/assumptions/settings you used along the way.
    • Be prepared to redo everything from the beginning at least three times.
    • A lot of choropleth maps kind of tell the same story. Even when values are normalized in some way, it often looks like the map title could be “Where are the people?”. So one trick I use is to just make sure my choropleth map isn’t trivial in that way; if that generic title could apply, then I think about whether I am showing the right variables in the right way.
    • I totally agree with Prof Maantay’s point about bi-chromatic color ramps! These can be very powerful tools for visualizing difference relative to some intermediate value, but shouldn’t be used arbitrarily to add visual interest/appeal to a map.
    • I was so excited to make maps that made sense, I wanted to include tons of details. But they get messy really fast. When in doubt, try leaving it out. Think Tufte. Every choice you make should be part of your story.
    • Learning mapping visualization is really forcing me to think deeply about the way visualizations are organized in “layers”. If you are familiar with R and ggplot, this may make sense. If you are not familiar with the concept of visualization layers, definitely pay attention to how maps are done in layers, and think about how this generalizes to other visualizations.
    #114266

    It’s great to hear that you like Monmonier’s _How to Lie with Maps_, Steve,
    as I teach it every year in my intro to digital humanities course!

    #114273
    Juliana Maantay
    Participant

    Matthew – Another oldie but goodie along similar lines is the book “Mapping: Ways of Representing the World,” by Danny Dorling (of cartogram fame) and Dave Fairbairn, from about 1997. Not sure if it is still in print, but it’s a slim little book that does a great job of explaining some of these issues, and a nice overview of the art and science of mapping.

    Juliana Maantay, Ph.D., M.U.P., F.R.G.S.
    Professor of Urban Environmental Geography
    Director: Geographic Information Science (GISc) Program;
    MS-GISc Program; and the Urban GISc Lab
    NOAA-CREST Research Scientist
    Department of Earth, Environmental, and Geospatial Sciences
    Lehman College, City University of New York
    250 Bedford Park Blvd. West
    Bronx, NY 10468
    TEL: (718) 960-8574; FAX: (718) 960-8584
    juliana.maantay@lehman.cuny.edu
    Faculty Profile: http://www.lehman.edu/academics/eggs/fac-maantay.php(https://mail.lehman.edu/owa/redir.aspx?C=JyvINcu5vqHrfPIz4NvBpSPK-oFcbs1Ep7J2CAaJl7JtUdYs4oXVCA..&URL=http%3a%2f%2fwww.lehman.edu%2facademics%2feggs%2ffac-maantay.php)

    “Geography, sir, is ruinous in its effects on the lower classes. Reading, writing, and arithmetic are comparatively safe, but geography invariably leads to revolution.” (1879 testimony before a Select Committee of the House of Commons, London, England, regarding expenditures of the London School Board)

    “The scientist is not a person who gives the right answers, [s]he’s one who asks the right questions.” (Claude Lévi-Strauss, 1964, Le Cru et le Cuit [The Raw and the Cooked])

    “Those who plan do better than those who do not plan, even though they rarely stick to their plan.” Winston Churchill

    “Be sure that no good may be expected of anyone who is satisfied with what he knows. To investigate is the task of the knowledgeable.” (Book of the Sea, 1513, by Piri Re’is, Ottoman cartographer, geographer, and navigator to Suleiman I)

    #114422
    Javier Otero Peña
    Participant

    There are many great comments here!

    Here are a few other contributions, though I’m not sure if they pertain
    specifically to beginners:

    – Although they are still the most widely distributed GIS format, take
    into account the limitations of shapefiles, such as:
    – The limit of 10-character-length for attribute names. This can be
    particularly problematic after a spatial join in which you kept the join
    layer filename as a prefix in the new dataset… once you save it as a
    shapefile, attribute names might all turn into EXAMPLE~1,
    EXAMPLE~2… and
    so on, making it impossible to know what each attribute is.
    – That you need to keep all files that come with your .shp together
    in the same folder and with the same filename, especially the
    .shx and the
    .dbf, which are mandatory files to keep.
    – Related to the previous one, that spatial index files (.sbn, .sbx)
    can sometimes corrupt the spatial representation of the layer and affect
    any geospatial analyses you run on the layer. To solve this issue, just
    delete those files. This has happened to me with MapPLUTO layers on
    QGIS, and I am not sure whether this is an issue in other
    software or with
    other data.
    – For the reasons above, consider using other file formats, such as
    Geopackage, Spatialite, Geojson, though each of these may have their own
    pros and cons, the biggest con likely being limited/no compatibility with
    certain GIS software such as ArcGIS.
    – And here are some newbie issues that you could run into when using CSV
    files:
    – Be careful if you edit a CSV dataset with Excel prior to importing
    to your GIS software, as Excel might automatically change long
    numbers into
    scientific notation. Once you save the CSV, those long numbers will be
    truncated. This is particularly problematic for US Census numeric ID
    numbers. Excel is also known for messing up date fields of CSV files.
    – If the software keeps misinterpreting an attribute type (e.g.,
    reads an ID as an integer instead of a string), and this limits your
    ability to join data to a layer, you can create a .csvt file
    with the same
    name and folder as the .csv file, in which you can specify the data type
    for each column in your dataset. For example: a .csvt which
    contains solely
    the text:* “String”, “Integer”, “Real”, “Date”, “Time”, “DateTime”*,
    will automatically interpret the first data column as a string,
    the second
    column as an integer, and so on, and will be opened as such by
    Excel, QGIS
    and many other CSV-reading apps.

    Hope this helps!

    Best,

    Javier

    #114426
    Elia Machado
    Member

    Thanks Olivia for starting this great discussion!

    A few thoughts:

    – Not setting up the “environmental settings”, particularly when working with raster data. The “mask” , “cell size”, and “snapping to” settings can make it or break it.
    Also, env. settings can be set at the general level and tool level. I recommend to always double check before running any tool.

    – This is not specific to GIS, but “being too enthusiastic” to start the analysis without cleaning the data first is obviously an issue, as well as
    using one output to as the next input without keeping track of what you are doing and most importantly, without checking that the result is as expected.

    -Long path names or file names containing special characters or that are not meaningful (e.g. temp3).

    -Not having a proper file management structure can lead to many headaches as the number of files can grow very quickly.

    – Following up with Juliana on the defaults, the same applies to the tools and modules. Assuming that the default parameters for a tool
    are appropriate for all data types or analyses can be very problematic.

    – In response to Javier regarding the field limitations of shp. I would add the geodatabase format to your list of suggested formats, it is proprietary
    and there is still a limit of 64 characters for the file geodatabases, but it is usually enough. I insist on using ArcCatalog for file management to keep the files together in every course I teach with ArcGIS. .

    – A few cartographic issues: 1) the first maps of some students include scale bars and north arrows that are too big and/or too ornamented, which often clashes
    with the style and/or its purpose. Along the same lines, I have noticed the use of effects in the background or colors that are distracting.
    2) using decimal degrees on a scale bar, it is rare, but definitely has an impact when seen and not a good one.

    I am not sure if this is what you are looking for, happy to talk more.

    Best,
    Elia

    #114427
    Elia Machado
    Member

    Thanks Olivia for starting this great discussion!

    A few thoughts:

    – Not setting up the “environmental settings”, particularly when working with raster data. The “mask” , “cell size”, and “snapping to” settings can make it or break it.
    Also, env. settings can be set at the general level and tool level. I recommend to always double check before running any tool.

    – This is not specific to GIS, but “being too enthusiastic” to start the analysis without cleaning the data first is obviously an issue, as well as
    using one output to as the next input without keeping track of what you are doing and most importantly, without checking that the result is as expected.

    -Long path names or file names containing special characters or that are not meaningful (e.g. temp3).

    -Not having a proper file management structure can lead to many headaches as the number of files can grow very quickly.

    – Following up with Juliana on the defaults, the same applies to the tools and modules. Assuming that the default parameters for a tool
    are appropriate for all data types or analyses can be very problematic.

    – In response to Javier regarding the field limitations of shp. I would add the geodatabase format to your list of suggested formats, it is proprietary
    and there is still a limit of 64 characters for the file geodatabases, but it is usually enough. I insist on using ArcCatalog for file management to keep the files together in every course I teach with ArcGIS. .

    – A few cartographic issues: 1) the first maps of some students include scale bars and north arrows that are too big and/or too ornated, which often clashes
    with the style of the map. Along the same lines, I have noticed the use of effects in the background or colors that are distracting.
    2) using decimal degrees on a scale bar, it is rare, but definitely has an impact when seen and not a good one.

    I am not sure if this is what you are looking for, happy to talk more.

    Best,
    Elia

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Elia Axinia Machado
    Associate Professor
    Earth, Environmental, and Geospatial Sciences,
    Lehman College, City University of New York
    Office: Gillet Hall, room 323
    Tel:718-960-1103 – Fax: 718-960-8584

    #115924

    Thank you so much to everyone who contributed to this discussion! Your insights were super helpful. Here the article that I wrote that incorporates a lot of your feedback: Top Mapping Mistakes

    #115926
    Juliana Maantay
    Participant

    Excellent synopsis! I will share it with my students. Thanks. JAM

    Juliana Maantay, Ph.D., M.U.P., F.R.G.S.
    Professor of Urban Environmental Geography
    Director: Geographic Information Science (GISc) Program;
    MS-GISc Program; and the Urban GISc Lab
    NOAA-CREST Research Scientist
    Department of Earth, Environmental, and Geospatial Sciences
    Lehman College, City University of New York
    250 Bedford Park Blvd. West
    Bronx, NY 10468
    TEL: (718) 960-8574; FAX: (718) 960-8584
    juliana.maantay@lehman.cuny.edu
    Faculty Profile: https://www.lehman.edu/academics/eggs/fac-maantay.php
    Urban GISc Lab Research Work: https://www.lehman.edu/academics/eggs/maantay-research.php

    “Geography, sir, is ruinous in its effects on the lower classes. Reading, writing, and arithmetic are comparatively safe, but geography invariably leads to revolution.” (1879 testimony before a Select Committee of the House of Commons, London, England, regarding expenditures of the London School Board)

    “The scientist is not a person who gives the right answers, [s]he’s one who asks the right questions.” (Claude Lévi-Strauss, 1964, Le Cru et le Cuit [The Raw and the Cooked])

    “Those who plan do better than those who do not plan, even though they rarely stick to their plan.” Winston Churchill

    “Be sure that no good may be expected of anyone who is satisfied with what he knows. To investigate is the task of the knowledgeable.” (Book of the Sea, 1513, by Piri Re’is, Ottoman cartographer, geographer, and navigator to Suleiman I)

    #115932

    Olivia, this is a great overview. Helpful article!

    =======================================
    Steven Romalewski
    Director, CUNY Mapping Service
    Center for Urban Research at The Graduate Center / CUNY
    [CUNY-GC-logo_whitetextblueback60hi]
    365 Fifth Ave., Room 6202
    New York, NY 10016
    Phone: 212-817-2033
    sromalewski@gc.cuny.edu<mailto:sromalewski@gc.cuny.edu>
    http://www.urbanresearch.org(http://www.urbanresearch.org)
    http://www.gc.cuny.edu/urbanresearchmaps
    @SR_spatial
    =======================================

    From: Olivia Ildefonso (GIS / Mapping Working Group) <noreply@gc.cuny.edu>
    Sent: Wednesday, May 12, 2021 10:41 AM
    To: Romalewski, Steven <SRomalewski@gc.cuny.edu>
    Subject: Re: Top Mapping Mistakes

    #116113
    Elia Machado
    Member

    Thanks for sharing Olivia! great article.
    Best,
    Elia

    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    Elia Axinia Machado
    Associate Professor
    Earth, Environmental, and Geospatial Sciences,
    Lehman College, City University of New York
    Office: Gillet Hall, room 323
    Tel:718-960-1103 – Fax: 718-960-8584

Viewing 13 posts - 1 through 13 (of 13 total)

You must be logged in to reply to this topic.