# Different length measures in QGIS vs. ArcMap

I noticed that QGIS and ArcMap return slightly different length measures when calculating geometries (length) on one and the same shapefile (see image).

In QGIS I used the \$length command, in ArcMap the "Calculate Geometry" function from the attribute table.

Is there an explanation for the difference observed?

The data are in the Lambert Conical Conformal projection, with parameters used by the Mexican authorities (INEGI). Units are meters and I use the data’s coordinate system (the LCC)

And generally speaking: How exact can I expect the length-calculus to be? Do I have to consider a confidence interval? • What coordinate system is the data using? Did you use the data source or data frame's coordinate system in Calculate Geometry? What are the units? – mkennedy Apr 29 '19 at 23:14
• It´s the Lambert Conical Conformal projection, with parameters used by the Mexican authorities (INEGI). Units are meters and I use the data’s coordinate system (the LCC). – Flo Apr 30 '19 at 4:02
• Please Edit the question in response to requests for clarification. It's not fair to those who would want to answer to need to mine comments for critical information. Further details may be necessary as to the exact Calculate Geometry parameters. – Vince Apr 30 '19 at 12:27
• Seems to be an issues of QGIS 3.4. I repeated the calculation in QGIS 2.18 and obtained exactly the same values as with ArcMap. Will report this issue rather to the QGIS project. – Flo Apr 30 '19 at 18:12
• It's possible QGIS 3.4 is giving the geodesic-based distance rather than the planar distance. You can check this in ArcMap by adding the data to a data frame and setting the measure tool to return geodesic distances. – mkennedy Apr 30 '19 at 18:47

As described in QGIS documentation, "\$length" gives us length based on ellipsoid, while "length(\$geometry)" calculated based on CRS. If using QGIS AddGeometryAttributes and choose By Layer_CRS, it gives the same values as "length(\$geometry)". It appears ArcGIS computed length values (by its shape_length in geodatabase) are closer to the values computed by "length(\$geometry)" in QGIS. Here is a quick stats showing calculation comparison: 