JE Bowles' book on Foundation Analysis and Design publishes typical values for some soil types.  By cross checking the soil you got against the published values you will have an idea where you stand.

Failing that do a load test on soil.

For Mat foundations the Subgrade Modulus (k) can vary depending on the duration of the load.  To be conservative, as far as moments are concerned, you can look up the typical value for k based on your soil conditions and divide it in half.  For bearing pressure use the higher k, resulting in higher pressures.  Basically, you will have an upper and a lowere bound for your k.  If you don't have a soils report, this is about all you can do.

Firstly, I use the same approach to compute my spring value. Secondly subgrade modulus (k in pci) should really come from the geotechnical engineer. It is a tricky number and lots of people assume it. I also encountered geotechnical engineers who would give different values for (k). I would suggest to use a seasoned soils engineer not one who copies numbers from previous reports (oops..  I hope this does not upset any one).

Good luck.

Not me!

Lufti is on the mark - the geotechnical engineer should be providing those values - and if the structure is "significant", you should work together to iteratively improve them.  You see, there is no such thing as "the" k-value - it is a numerical convenience, not an intrinsic property.

Please see FAQ731-376  by VPL for tips on how to make the best use of Eng-Tips Fora.