IDRISI did not assign values to polygons correctly when importing from shapefiles.  Instead, it assigned the polygon (or line, or point - depending on what coverage was used) ID's to the polygons in IDRISI vector layers.  This required many additional steps in reclassing the data.

In general, I found that all coverages that are imported from shapefiles have to go through the following process:

1.    Import shapefile into IDRISI vector layer
2.    Open database and export attribute value file for the value represented by lines, polygons, or points in that coverage
3.    Rasterize the vector file into raster image
4.    Use the attribute value file to assign proper values to raster cells of a given ID

shapefile ----- (IMPORT) ----> filename.vct ---- (POLYRAS) ----> filename.rst ---- (ASSIGN) ----> filename2.rst
                                                            |
                                                             ------------------------------------------------( filename.avl )
 

The data with contours of mean annual precipitation were not complete so the interpolation was not exact.  In some instances, interpolations was not correct because of lack of data.  The contours were only accurate over short distances away from the point feature.
 

I could not overlay the interpolated beach attribute images with the cost-distance analysis of the road network because the two "streams" of analysis used different basemaps for raster images.  All images were 1000 x 1000 rasters, but in visualization of beach attributes I had to extend the image bounds to extrapolate a few more kilometers beyond the coastlines (to arrive at the result of narrow buffer around the coastlines for showing results of analyses).  I should have calculated a different size for resulting raster image (not 1000 x 1000 but slightly larger).
 

IDRISI requires many conversions when performing various analyses, sometimes unexpectedly.