The Influence of Item Wording on Scale Scores

Authorities on scale construction recommend that negatively worded items be used along with positively worded items to reduce careless responding. Negatively worded items are assumed to force respondents to consider item content more diligently. The present study investigates the influence of item wording in three versions (i.e., 33% positive, 67% positive, and 100% positive) of the Hardiness scale (Bartone, Ursano, Wright, &. Ingraham, 1989). These versions were administered to undergraduate psychology students (N = 325). Multi-sample, confirmatory factor analyses is used to compare the measurement models obtained with each version of the hardiness scale. The results of the study show that the mixture of negatively and positively worded items have undesirable effects on the measurement properties of a scale. Future research needs to include a questionnaire with no positively worded items, and to compare motivational and cognitive explanations for the influence of negatively worded items.