This study reports an attempt to improve the administration of the Peabody Picture Vocabulary Test (PPVT) by identifying and modifying aspects of the test which adversely affect disadvantaged preschoolers' performance. The resultant test was called the Modified Peabody Picture Vocabulary Test (M-PPVT). Two samples from the same lower class population were compared on the standard and modified test versions. The M-PPVT required the examiner to provide encouragement and approval, to follow a schedule of high reinforcement, and to give specific instructions before each stimulus word. A modified record sheet facilitated administration and recording of responses. In Study I both test forms were administered twice to 36 Negro children. There was slightly increased score stability under the M-PPVT, but practice effects were not significant. Boys scored significantly higher than girls. In Study II, 19 black children took the PPVT and 19 took the M-PPVT. Results suggested that the M-PPVT offered a slight advantage to low-scoring subjects possibly due to examiner controls. Selection and evaluation of test instruments for disadvantaged preschoolers should consider psychometric properties, and the influences of examiner styles and test administration controls on performance. (DR)