We describe the development of the Mental Health Inventory (MHI), a new 38item measure of psychological distress and well-being, developed for use in general populations. The MHI was fielded in four large samples having quite different characteristics (N = 5,089). One data set was used to explore the MHI's factor structure, and confirmatory factor analyses were used for cross-validation. Results support a hierarchical factor model composed of a general underlying psychological distress versus well-being factor; a higher order structure defined by two correlated factors-Psychological Distress and Well-Being; and five correlated lower order factors-Anxiety, Depression, Emotional Ties, General Positive Affect, and Loss of Behavioral Emotional Control. Summated rating scales produced high internalconsistency estimates and substantial stability over a 1-year interval. Our results provide strong psychometric support for a hierarchical model and scoring options ranging from five distinct constructs to reliance on one summary index. This tradeoff, which is between the unique information contained in the subscales versus the simplicity of a single score, should be evaluated further.A review of general population mentalhealth-survey instruments pointed out important trends in questionnaire content and conceptual issues that deserve empirical attention (Ware, Johnston, Davies-Avery, & Brook, 1979). Early instruments were very heterogeneous in content (Gurin, 1960;Langner, 1962;Macmillan, 1957). They included measures of physical and psychosomatic symptoms, functional status, other health problems or worries, and health habits, in addition to measures of more straightforward psychological constructs (e.g., symptoms of anxiety and depression). More recent instruments seem to focus almost exclusively on the more straightforward psychological constructs (Bradburn, 1969;Cleary, Goldberg, & Kessler, 1982;Dupuy, 1972). Whereas the more heterogeneous measures may be satisfactory for testing hypotheses about health status in general, they do not seem to do well in distinguishing changes in mental health from changes in physical health (Ware et al., 1979; Ware, Brook, Davies-Avery, et al., 1980b).Another characteristic of more recently developed mental health surveys is their focus Requests for reprints should be sent to

A basis for distinguishing between subtractive and ratio models of perceptual judgment is presented.Based on direct-scaling data, Torgerson speculated that judges may perceive only a single relation between a pair of stimuli regardless of instructions to judge differences or ratios. Illustrations demonstrate the inadequacy of direct-scaling methods and the advantages of applying algebraic models and factorial designs. However, additional criteria are required to determine appropriate models of stimulus comparison.Birnbaum and Veit, in a two-task (difference and ratio) lifted-weight experiment, imposed the criterion of scale invariance as one such additional constraint. Scales obtained using difference and ratio instructions were linearly related only when both of the resulting data matrices were fit to the same model-either subtractive or ratio.The first experiment of the present research found the same ratio-difference indeterminacy for another stimulus dimension, grayness. This indeterminacy was also replicated for both category ratings of differences and magnitude estimations of differences. The raw data appeared to more closely approximate the predictions of a ratio model when magnitude estimations were used and a subtractive model when category ratings were used. However, both response procedures led to the same ordering of the stimulus pairs for all three sets of data.Contrary to the notion that it would be impossible to distinguish between these models, the appropriateness of the subtractive model was demonstrated by using a new task in which judges estimated the ratio of differences. The data conformed to the qualitative predictions of the ratio-of-differences model and allowed rejection of the ratio-of-ratios, difference-of-differences, and differenceof-ratios models. Scales derived from the ratio-of-differences model agreed with those derived from the fit of the subtractive model to data obtained from difference and ratio tasks, suggesting that experimental observers estimate intervals when instructed to judge either differences or simple ratios. The data also implied that magnitude estimations induce a nonlinear response transformation that can make subtractive processes appear multiplicative.The procedure developed for the present research provides a general methodological tool for studying alternative models of perceptual judgment.

Ss lifted pairs of weights simultaneously. one in each hand, and judged either the difference, ratio. or average heaviness of the two weights. Data for the difference and ratio tasks were in general agreement with subtractive and ratio models. but the averaging data showed discrepancies from the constant-weight averaging model similar to those reported in previous psychophysical research. Rescaling was ruled out for the averaging data, because responses to pairs of equal weight were a linear function of subtractive model scale values derived from the difference task data. Scale values for the ratio and difference task data were related exponentially. as were the responses to the pairs, consistent with Torgerson's conjecture that Ss do not distinguish "differences" from "ratios." They appear to use the same composition rule but different output functions, depending on the procedures for responding. The scale convergence criterion can thus prevent inappropriate rescaling when a model fails and can dictate rescaling even when a model fits.

In two experiments, Ss rated the difference in heaviness between two objects varying in both size and weight. Assumption of the subtractive model and the use of factorial designs allow separation of judgmental effects from psychophysical processes. Difference ratings were rescaled by monotone transformation to fit the subtractive model, yielding scale-free values for the size-weight combinations. The subtractive model provided a good description of the difference ratings, but critical violations of the additive model for the size-weight illusion were obtained. The experiments illustrate how ordinal information can be used to differentiate additive from multiplicative processes.where h is the heaviness of the object, s reflects the subjective heaviness due to physical weight apart from the effect of size, and s* represents the effect of size on "Size-weight illusion" refers to the fact that the subjective heaviness of an object depends upon its size as well as its weight. The smaller the object, holding physical weight constant, the heavier it feels when lifted. Although both weigh the same, a pound of lead does feel heavier than a pound of feathers.

Few eligible postmenopausal women participate in clinical trial research to prevent breast cancer or coronary heart disease, making it impossible to adequately assess the efficacy of tested interventions for this vulnerable group. To elucidate the causal factors and decision model underlying participation behavior, 180 white, African American, and Hispanic postmenopausal women judged their likelihood of participation in a breast cancer or coronary heart disease prevention clinical trial in scenarios with varied cost/remuneration, perceived risk, doctor's recommendation, and expected toxicity. In addition, 293 white, African American, and Hispanic male and female physicians judged the strength of their participation recommendation in scenarios with varied cost/remuneration, expected toxicity, patient's age, and the source of the information about the clinical trial. An additive and constant-weight-averaging model were rejected. The same configural-weight-range model accounted for judgments in both breast cancer and coronary heart disease scenarios, with different parameter values for each group. According to this model, white and Hispanic women under 70 years of age are most likely to participate, even under somewhat adverse conditions; costs and high toxicity levels act as severe barriers to physicians' positive recommendations and women's participation. Perceived risk was the most important factor for women, yet only 8% and 15% reported ever having received risk information from their doctor for breast cancer and coronary heart disease, respectively. For these two diseases, respectively, 75% and 48% of women rated their risk of the disease as low and 76% and 88% reported they had never heard of a randomized clinical trial or of a prevention clinical trial being conducted. These results have implications for education, information dissemination, and prevention clinical-trial planners.