First, welcome to the 3D Printing Stack Exchange!
On bed leveling
Bed leveling, or more accurately bed tramming, adjusts the bed so that it is even in relation to the print head. Typically it is done by sliding a paper between the nozzle and the bed when Z = 0, or the print head is at its lowest. The amount of friction should be similar to the amount of friction of a sheet of paper between two magazines (recommended by Tom Salander on YouTube I believe). One sheet of paper is about 0.1 mm, which is a common gap and allows for good adhesion of the deposited plastic, and typically the gap is measured at the four corners and the center of the bed.
But wait, there's more!
Mesh bed leveling can account for a physically uneven print bed (like a bowing glass sheet) and can be done manually like above or one can mount a sensor so that everything is automatic. The print bed is broken down into a grid (typically 3x3 or 5x5) and the Z height adjusts so that the gap is 0.1 mm at each point in the grid. A mesh is then made and stored in the printer. As the printer prints it makes the necessary variations in height so that the print is flat. Please note that the bed should be leveled as best as possible before setting up the mesh as it will vastly improve the resulting mesh. Mesh leveling is primarily software. The actual level of the bed remains unchanged, thus needing to level it prior to setting up the mesh.
Bed leveling (including mesh bed leveling) is one aspect of calibration. Other important factors to tune in are temperatures of the nozzle and the bed. For example, PLA has a temperature range of 180-230 °C but each printer is different and may need to print at a higher or lower temperature for best results. Or the printer reports a temperature of 200 °C but the thermistor is off and is actually printing at 190 °C. It's up to the user then to compensate if they want/need to print at an actual 200 °C.
Other settings to tune are print speeds, acceleration and jerk, retraction, and extrusion. Faster print speeds can result in poorer quality prints, namely ghosting and under extrusion. Acceleration/jerk works in the same vein since it directly affects speed. Retraction helps with stringing. If the retraction settings are too low then there will be stringing on the model, if it's too high then there can be under extrusion and heat creep. Adjusting the extrusion factor will tell the printer to push more plastic out or push less. The higher the factor the more plastic is pushed. IMO best practice is to leave the factor at 1 in the slicer and calibrate the stepper motor itself (as can be seen here)
There are several calibration models on thingiverse that can show common print issues. Some of the popular ones are:
- The calibration cube which one can ensure that the motors are moving accurately to print out a 20 mm cube. This also shows ghosting along any of the three axes. This cube may also point out if any of the axes aren't 90 degrees perpendicular to each other as the cube should be, well, a cube and not an amorphous shape (thanks to Joel Coehoorn for pointing this out).
- 3DBenchy which shows a myriad of potential print quality issues including ghosting, poor overhangs/bridging, and stringing.
- A Temperature Tower uses post scripting in the slicer to change the temperatures so one can determine the different print qualities across a spectrum of temperature. There are a lot of different models online if this one does not work for you.
- An all-in-one test can do a lot of tests in one model and they are all labeled within the model. (The linked one is the mini version, there are several similar tests online.)
- This list is by no means inclusive, there are several other models that can test other problems. These are just what's usually printed (especially the cube and Benchy).