The development of an instrument for the measurement of psychological comfort

The purpose of the study was to determine the reliability and validity of the researcher-developed Comfort Scale measuring psychological comfort among pre and postoperative patients. The Comfort Scale, based on Davitz' The Language of Emotion (1969), consisted of 60 items which had been approved for content validity by three groups, each of nine professional raters in psychology and nursing. The 60 items manifested an apriori Alpha Coefficient reliability of .96. Four subscales of 15 items each were: Sensations, Attitude to Self, Control, and Attitude to Life and Others, manifesting apriori reliabilities of .93, .93, .84, and .82, respectively.

The Comfort Scale, Spielberger's State Anxiety Inventory, and the McMaster Health Index: Emotional Function were administered preoperatively and postoperatively to 116 patients, hospitalized for uncomplicated surgeries. Internal reliabilities of the Comfort Scale manifested .95 preoperatively and .96 postoperatively. The four subscales of Sensations, Attitude to Self, Control, and Attitude to Life and Others manifested reliabilities of .91, .90, .81, .70 preoperatively and .92, .88, .77, .79 postoperatively, respectively. A factor analysis of each subscale manifested two factors within each, except Attitude to Self, which manifested one. The seven factors of five items each were entitled: Sensations. (1) General Positive Emotion, (2) General Negative Emotion; Attitude to Self. (1) Self Satisfaction; Control. (1) Self Integration, (2) Coping Ability; Attitude to Life and Others. (1) Trust; (2) Loss of Meaning, comprising a modified Comfort Scale of 35 items.

A Hotelling T Squared manifested a significant difference between preoperative and postoperative Comfort Scale scores, contributing to construct validity, as hypothesized. Only the subscale of Sensations and General Negative Emotion factor manifested a significant difference between preoperative and postoperative scores. The Comfort Scale scores correlated negatively both preoperatively and postoperatively with the State Anxiety Inventory scores, a measure of Discomfort, contributing to construct validity, as hypothesized. The Comfort Scale correlated positively with the McMaster Emotional Function, both preoperatively and postoperatively, contributing to concurrent validity, as hypothesized. The process of how persons comfort themselves was also described identifying five styles: Rationalization, Substitution of Ideation, Replacement, Serial Equalization, and Goal Setting. In conclusion, reliability, content, concurrent, and construct validity were established for the Comfort Scale.