The Effectiveness of Sentence Cues in Measuring the Big Three Motives

Abstract

Despite the popularity of free response measures in the motivation literature, research geared toward the development of a standard battery of cues for measuring the Big Three motives (achievement, affiliation, power) has been lacking. The current research examined the effectiveness of sentence cues in eliciting motive imagery in two studies (students, entrepreneurs) comprising 242 men and women. Results indicated that sentence cues were effective in eliciting achievement and affiliation imagery, but not power imagery. In addition, an examination of the subcategories underlying each motive scoring system indicated that there were several infrequently scored subcategories in the achievement and power motive scoring systems that could be considered for removal.

ACKNOWLEDGMENT

We thank Susannah Roth for her assistance with data collection in Study 2.

Notes

^a See McClelland (1985) for a review.

¹Of course, analysis of scoring subcategories alone is not sufficient to argue for or against the elimination of particular subcategories, and evidence of differential validity is needed to determine the validity of revised motive scoring systems before abridged systems can be adopted.

²It should be noted that the “ethnicity” of the characters' names (e.g., Carol/David) also may need to be altered to ensure that the names do not introduce a cultural basis. The names used in the current research (see Method) were conventional names at the time of the research and were appropriate for participants given the ethnicity of the sample (e.g., Australian born and educated). We advise, however, that researchers may need to substitute alternative, culturally appropriate names into the sentence cues for different samples.

³CA represents the percentage of agreement between two scorers on the presence of a category and Rho represents the agreement between two scorers on the total scores (Smith, Feld, & Franz, 1992). CA is computed by counting the number of times the category was scored as present by each scorer and the number of times they agreed on its presence and is equal to

Agreement about the absence of a category could be misleading because some subcategories are scored infrequently (Smith et al., 1992). In the current research, CA is reported for primary categories only, that is, agreement between scorers about whether achievement, affiliation, or power imagery is present.

^a For nAch, all differences between the means for different cues were significant (Cohen's d = −.62 to .87) except Cue 2 versus Cues 4 and 5, Cue 3 versus Cue 6, and Cue 4 versus Cues 5 and 6. For nAff, all differences were significant (Cohen's d = −.76 to −.91) except Cue 3 versus Cues 1, 5, and 6; Cue 2 versus Cue 4; and Cue 5 versus Cue 6. For nPow, Cue 5 was significantly higher than Cues 1, 3, and 6 (Cohen's d = −.36 to .25). For activity inhibition, Cue 2 was significantly higher than Cue 5 (Cohen's d = .29).

^b n = 89.

^c n = 101.

^a For nAch, Cues 1 and 3 were significantly higher than Cue 2 (Cohen's d = .59 and .81, respectively). For nAff, Cues 2 and 4 were significantly higher than Cue 1 (Cohen's d = 1.83 and .98, respectively). Cue 2 was also significantly higher than Cue 3 for nAff (Cohen's d = .48).

^a Bold indicates moderate to high cue strength.