Attitude measures frequently produce distributions of item scores that attenuate interitem correlations and thus also distort findings regarding the factor structure underlying the items. An actual data set involving 260 adult subjects' responses to 55 items on the Love Relationships Scale is employed to illustrate empirical methods for identifying such items during test construction or test validation. Two methods for identifying these items are proposed. The first method involves the use of "best fit" factor rotation; the second method involves the comparison of "reproduced" correlation matrices computed from two structure matrices. It is noted that examination of item descriptive statistics may not successfully substitute for an empirical method for identifying potentially troublesome items. An appendix contains tabulated data. (Author/JAZ)