The effects of viewing a single model (one female model) as compared to viewing multiple models (two male and two female) were tested with 27 beginning master's degree students enrolled in a prepracticum course. Four measures of the dependent variable, open-ended exploratory leads (OEEL's) were taken to allow for examination of both quantitative and qualitative differences due to the modeling conditions. Viewing multiple models, although not statistically superior to viewing a single model, did result in the highest means and smallest within treatment variation on all dependent measures. Only the quantity of produced OEEL's was significantly greater for those participants who viewed models as compared to those who did not. As a result of both the descriptive and inferential data, implications for the use of multiple and single models for counselor training are presented and discussed.The theoretical framework for many skills training programs in counselor education (e.g., Ivey's 197 1 microcounseling; parent education programs by Nay, 1975; semi selfinstructional programs) is found in Bandura's (1969) social learning theory. Use of observational learning, a basic component of social learning theory, is well-documented in the literature (e.g., Bailey, Deardorff, & Nay, 1977;Fridman & Stone, 1978;Gulanick & Schmeck, 1977;OToole, 1979; Dalton & Sundblad, 1976; Robinson, Froehle, & Kurpius, 1979a). There has been little research however, on how models can best be presented to enhance learning.
29Some counselor educators assume that multiple models will "increase the effectiveness of the modeling condition" (Uhlemann, Lea & Stone, 1976, p. 510). For this reason they often present a variety of models rather than relying on one model.A review of the modeling literature reveals that Bandura's early work with children favored the use of multiple symbolic models as compared to a single model (Bandura, 1969; Bandura 8c Menlove, 1968), and heterogeneous models over single models (Bandura, Ross & Ross, 1963). A more recent study of assertiveness training had participants imagine a single model or multiple models behaving assertively (Kazdin, 1976). In none of these studies however, were the behaviors being modeled counselor interviewing skills. The training research literature fails to compare the effects of different types of modeling on learner performance (Bourdin, 1970;Eskedal, 1975).The purpose of this study was to determine the effect of modeling variations through the presentation of a series of videotaped vignettes of the same model counselor (single model) versus presenting the same vignettes with the two males and two females functioning as model counselors (multiple models). I t was hypothesized that trainees who observed multiple models would perform significantly better than trainees who observed the single model. It was also hypothesized that trainees who observed models would perform significantly better than trainees who did not observe models.The dependent variable chosen to assess the effects of multiple,...