Perceptions of Nurse Caring

Three problems guided the study. The first was to determine differences in perceptions of nurse caring behavior by physicians and graduate nurse students. Second, further reliability and validity testing of the investigator-developed instrument was conducted. Third, comparison of findings generated by a previous investigator-performed study with results of the present research was carried out. The theory of Nurse Caring (Weiss, in press) formed the framework. Four theoretical propositions were supported.

An experimental 2 x 2 x 2 x 2 factorial design utilizing eight investigator-developed videotape segments which portrayed a nurse interacting with a patient was used. Four main effects, 6 two-way, 4 three-way, and 1 four-way interaction effect hypotheses were tested. The independent variables were verbal and nonverbal caring and uncaring and technical competent and incompetent nurse behaviors. Health professional role was also considered. Subjects (N = 112) were randomly assigned to 16 cells containing 7 physicians and 7 graduate nurse student subjects. The eight videotape segments were randomly assigned to the cells for subject viewing.

The dependent variable consisted of one investigator-developed 10-item, 7-point Likert-type Caring Scale which allowed the subject to measure nurse behavior in terms of work capability and level of interpersonal awareness capacity. Four factor fixed effect analysis of variance was performed on the data generated by the scale.

Statistically, the findings overall indicated both physician subjects and graduate nurse subjects preferred the verbal and nonverbal caring and technical competent nurse behavior and rejected the opposite combination of behaviors. Four two-way interaction effects: verbal by nonverbal, verbal by technical, nonverbal by technical, and technical by health professional role emerged as the principle findings. Verbal behavior occupied the position of prime importance followed in turn by nonverbal behavior and technical competent behavior.

The instrument achieved an alpha coefficient of .95 which shows a high level of consistency and dependability. Factor analysis of the scale resulted in 1 factor accounting for 69% of the accountable variance indicating a high degree of construct validity. Previous research results of the study problem when compared with the present research findings demonstrated like results.