Classical measurement theory was used to investigate measurement characteristics of both parts of the Finding Embedded Figures Test (FEFT) when the test was: administered in either a "no guessing" supply format or a multiple-choice selection format; administered to either undergraduate college students or middle school students; and completed in a timed ("speed") format as opposed to untimed ("power") format. Analysis is based on data from several studies conducted by Melancon and Thompson in 1989: (1) 69 undergraduates in the "power" supply format study; (2) 155 undergraduate students completing the FEFT in a "power" multiple-choice format study; (3) 1,528 middle school students completing the FEFT in a "power" multiple-choice format; and (4) 45 undergraduate students completing the FEFT in a "speed" selection administration format in the present study. Coefficient alpha for the FEFT ranged between 0.86 and 0.90 across samples and administration formats. Items generally had desirable psychometric characteristics across the studies. Construct validity analyses support a conclusion that the measure is reasonably valid. A 37-item list of references is included. Five tables present results from the studies. (SLD)