Introduction and Study Purpose
It has been noted that the onset of the psychiatric nursing clinical experience is a particularly anxiety-laden one for nursing students (Williams, 1979), and that many students feel inadequate in their knowledge and application of therapeutic communication skills in relating to patients (Schoffstall, 1983). In addition, it has been suggested that simulated computer learning regarding various aspects of psychiatric care may help ease students' transition into the psychiatric care arena (Kamp & Burnside, 1974).
These observations generated our interest in creating a computer- video interactive system (CVIS) program regarding therapeutic communication skills, and designing a research project to test whether or not students who received this supplemental program prior to the onset of their clinical psychiatric nursing experience would evidence greater learning and decreased state anxiety than would a group of students who received a traditional lecture and supplemental videotape. This article chronicles the research design and test of the CVIS program with baccalaureate nursing students. The findings raise several intriguing questions regarding psychiatric-mental health nursing education.
Literature: Learning and Computer-Based Instruction
When computer-assisted instruction (CAI) is used as a supplement to ongoing instruction, it generally increases learning. However, if these instructional systems are to be widely accepted and implemented in nursing, details regarding their development and effectiveness in enhancing learning in all specialty areas of nursing need to be further documented.
A few research studies in nursing have been designed to evaluate the relative effectiveness of CAI and CVIS versus more traditional modes of instruction in enhancing learning outcomes (Halverson & Bellinger, 1978; Bitzer, Boudreaux, & Avner, 1973; Huckabay, Anderson, Holm, & Lee, 1979; Kirschoff & Hölzerner, 1979; Boettcher, Alderson, & Saccucci, 1981; Conklin, 1983; Edwards & Hannah, 1985; Allen, Devney, & Sharpe, 1986). As noted, findings from these studies have been mixed. For instance, Boettcher and her colleagues (1981) studied nursing students' learning outcomes in the cognitive categories of knowledge and application regarding psychopharmacological content via either a CAI program or a printed programmed instruction and found no significant difference between the groups.
Similarly, in a study to test whether a computer-based interactive video program would be more effective than a traditional linear video computer program in enhancing students' abilities to detect errors regarding the aseptic technique of gloving and degloving, Allen and his associates (1986) found that there were no significant differences between the groups.
More positive findings in favor of CAI are reported by Huckabay and her associates (1979), who found that graduate nursing students, given a CAI lesson, scored significantly higher on a posttest regarding the nursing management of hypertensive patients in the cognitive categories of knowledge and application of knowledge than did a traditional lecture group.
In an experimental pretest-posttest design to compare the effects of teaching cardiopulmonary resuscitation (CPR) with either computer interactive videodisc or the traditional "live" instructor method to 65 employees of a major oil company, Edwards and Hannah (1985) found that there were no significant differences between the two groups. The authors concluded that the computer interactive videodisc was at least as effective a method as "live" instruction to teach CPR to lay persons.
The general consensus of the few studies designed to determine the effectiveness of CAI and CVIS in the health professions is that they are at least as effective as most other means of teaching. The present study was designed to determine the relative effectiveness of a CVIS instructional program versus a traditional video viewing program in enhancing learning outcomes regarding therapeutic communication. For the purposes of this study, learning was defined as the evidence of increased application of knowledge on a posttest regarding therapeutic communication.
Literature: Anxiety and Computer-Based Instruction
Research that specifically investigated the impact of computer-video interactive system instructional programs on student anxiety related to actual or anticipated clinical nursing experiences was not found in the literature. However, it has been noted that nursing students experience an increase in anxiety in adjusting to all types of clinical experiences (Mauksch, 1963; Stein, 1969; Lewis, Gadd, & O'Conner, 1987) and especially so in relation to the initial clinical experience in psychiatric nursing (Williams, 1979; Schoffstall, 1981). In fact, in surveying nursing students' concerns about beginning their psychiatric nursing clinical rotation, Schoffstall (1981) found that the largest group of concerns was related to the students' perception of lacking sufficient therapeutic skills to interact effectively with patients.
Williams (1979) investigated the relationship among psychiatric clinical settings and level of anxiety in associate degree nursing students. Findings indicated that for all groups of students (those in different psychiatric clinical settings), anxiety level was increased at the onset of the psychiatric nursing experience.
It should be noted that, in measuring the students' anxiety, Williams used the IPAT Anxiety Scale (Cattell & Scheier, 1961) and limited the measurement of anxiety to "trait anxiety," or the more stable individual differences occurring with students, and omitted the measure of situational "state anxiety." Although Williams concluded that the initial experience in psychiatric nursing should be considered a potentially anxiety producing one, she did not fully evaluate students' state anxiety pertinent to the perceived stress of the psychiatric clinical experience.
In an earlier study, Kamp and Burnside (1974) studied the use of an interactive computer learning system in a graduate psychiatric nursing class and observed that this system allowed students to move with ease from the simulated computer situation of learning about various aspects of care related to psychiatric patients to the more stressful real clinical situation with patients. However, these investigators made no systematic attempt to study the students' state or trait anxiety profiles prior to and following the computer instructional experience.
The present study investigated whether or not a CVIS instructional program, which presented information regarding therapeutic communication skills in interacting with psychiatric patients, would help mediate students' state anxiety related to their anticipated interactions with patients in the clinical setting. Both state and trait anxiety of subjects were measured. For this study, anxiety was defined as a state of uneasiness, apprehension, or tension based on an anticipated stressor.
This study tested the following hypotheses:
1. Subjects who receive a lecture-discussion plus supplementary CVIS instruction (LD + CVIS) regarding therapeutic communication will evidence significantly higher scores on a Therapeutic Communication Posttest than will subjects who receive the same lecture-discussion plus a supplementary videotape (LD + VT).
2. Subjects who receive LD + CVIS instruction regarding therapeutic communication will evidence significantly lower state anxiety scores following this instruction than will subjects who receive LD + VT.
SAMPLEAND SETTING- Subjects for this study were an entire class (N = 75) of first semester senior students enrolled at The University of Texas at Austin School of Nursing for their psychiatric-mental health nursing experience during the Spring semester, 1984. Descriptive demographic data revealed that ages reported by this sample ranged from 19 to 40 years, with the majority of subjects (65%) falling into the age group of 19 to 24 years, 20% in the age group of 25 to 30, and 15% in the age group of 31 to 40. A clear majority (84%) of the total sample had never had any previous work experience with psychiatric patients, nor had they worked in psychiatric settings, while 16% of the sample indicated that they had had previous experience with psychiatric patients and had worked in psychiatric settings. Regarding previous experience with computers, 70% of the total sample had had some experience with computers, mostly from playing computer games or using computers in class assignments.
PROCEDURES - Because a genuine control group was missing, a quasiexperimental pretest-posttest design guided this study. Participation in this study was optional and informed consent was gained from each subject. The subjects were randomly divided into two groups: Group I (LD + CVIS; N = 36), and Group II (LD + VT; N = 39).
Pretest measures. On the first class day, prior to any instruction, all subjects were gathered into a classroom and asked to complete the Demographic Data Questionnaire, a knowledge Pretest on Therapeutic Communication, and the STAI-S and STAI-T anxiety scales (Spielberger, Gorsuch, & Lushene, 1983).
Treatment. Within the next week, prior to entering the psychiatric clinical area, all subjects received the basic content regarding therapeutic communication in a lecturediscussion. Following this basic content lecture, Group I received the additional CVIS instruction treatment and Group II subjects were required to view a videotape. Both the CVIS program and the videotape were approximately 30 minutes in length.
Posttest Measures. Following the instructional treatments, subjects were asked to complete a knowledge Posttest on Therapeutic Communication, and the STAI-S scale. After all testing was completed, subjects were told that they could view whichever supplementary instruction program they did not initially receive, i.e., Group I subjects could view the videotape, and Group II subjects could run the CVIS program.
INSTRUMENTS- In addition to the CVIS program and the videotape, four instruments were used in this study. They were: 1) the Demographic Data Questionnaire, 2) the Pretest on Therapeutic Communication, 3) the Posttest on Therapeutic Communication, and 4) the STAI-State and STAI-Trait Anxiety Scales (Spielberger, Gorsuch, & Lushene, 1983).
The CVIS Instruction Program. Computer-interactive video systems are proving to be meaningful contributions to education technology. The integration of a computer with the audio- visual properties of video (videotape, videodisc, or television) offers a variety of applications in nursing education (Redland & Kilmon, 1986; Schwartz, 1984).
The CVIS program on therapeutic communication techniques was developed by four of the investigators (Rickelman, Payne, Taylor-Fox, & Reisch) during a sixweek period in the summer of 1983 and refined and formatted during the following Fall semester. Dr. Betty Skaggs, director of The University of Texas at Austin School of Nursing Learning Center, and several of her staff utilized the Whitney Authoring System (1981) to format the CVIS program. Currently, the CVIS program is being reformatted for use on a wider variety of computers.
The CVIS program was used in the present study to present simulated nurse-patient counseling situations to the Group I subjects and to provide the subjects an opportunity to augment and apply knowledge of therapeutic communication skills. Using the CVIS program, Group I subjects interacted with the computer by observing on the video monitor five different simulated nurse-patient situations. After observing each situation, seeing and hearing both patient comments and nurse-counselor responses, the subject was requested to apply knowledge of therapeutic communication by selecting (via the typewriter keyboard) his or her choice of the "best" or "most therapeutic" nurse reply to the patient's comment. This selection was made from a multiple-choice format of responses presented to the student on the computer monitor. Five areas of therapeutic communication content were tested: 1) clarifying indefinite pronouns, 2) clarifying general terms, 3) facilitating further description of an event or observation, 4) focusing on pertinent versus extraneous content, and 5) avoiding automatic knowing.
Following each situation, the subject was given the option of reviewing any of the nurse-patient interactions and deciding when he/she was ready to make a choice of the "best" nurse reply. Feedback assessing the correctness of each selection was provided by an audio recording and a written summary that was presented on the computer monitor.
The Videotape. The videotape depicted a simulated home visit by a public health nurse who used therapeutic communication techniques in interviewing a family about a medical health problem and psychological responses. Although this videotape did not emphasize a nurse-patient interaction in a psychiatric setting, similar therapeutic communication techniques as presented in the CVIS were illustrated. However, the videotape required only passive observation on the part of the student and did not provide any interactive processing of information.
The Other Instruments. The Demographic Data Questionnaire, the Pretest on Therapeutic Communication, and the Posttest on Therapeutic Communication were developed by four of the investigators (Rickelman, Payne, Taylor-Fox, and Reisch). The Demographic Data Questionnaire sought information regarding subjects' age group, sex, previous experience with psychiatric patients, and previous experience with computers.
The Pretest on Therapeutic Communication evaluated both Group I and Group II subjects' knowledge regarding therapeutic communication prior to any instructional strategies. This test consisted of 24 items of different statements made by hypothetical patients. There was a blank space below the patient's statement indicating "Nurse Response," and subjects were required to write in a therapeutic reply to the patient's comment. The 24 items included equal numbers of statements designed to elicit responses reflective of the same therapeutic communication content as described in the CVIS program (i.e., clarifying indefinite pronouns and general terms, facilitating further description of an event or observation, focusing on pertinent versus extraneous content, and avoiding automatic knowing). Subjects' responses were evaluated and coded as either correct or incorrect. Interrater reliability among four psychiatric-mental health nursing faculty who evaluated the subjects' responses on the Pretest was .96. In addition, reliability of the pretest was calculated by the Cronbach Alpha procedure and results indicated a coefficient alpha of .61.
The Posttest on Therapeutic Communication was also designed to measure subjects' knowledge of therapeutic communication. The posttest consisted of 24 items, matched with the pretest for content categories and format, but with different statements by hypothetical patients. Interrater reliability among the four faculty who evaluated the subjects' responses on the posttest was .96. The coefficient alpha reliability for the posttest was .85.
The State-Trait Anxiety Inventory (STAI) consists of separate self-report scales for measuring state and trait anxiety. The STAI state scale is composed of 20 items that evaluate how respondents are feeling at the moment or how they anticipate they will feel either in a specific situation that is likely to be encountered in the future or in a variety of hypothetical situations, depending on the interests of the researcher or clinician administering the instrument (Spielberger, Gorsuch, & Lushene, 1983). For the purpose of the current investigation, the STAI state scale (Form Y-I) was administered to all subjects with the following special instructions: "Imagine that you are in the psychiatric clinical setting, and that you are about to meet your one-toone patient for the first scheduled therapeutic communication session. Imagine how you feel in anticipating this interviewing session with the patient. Keep these feelings in mind as you respond to the items on the scale."
The trait-anxiety scale (Form Y-2) consists of 20 statements that assess how people "generally feel." The STAI scale was administered prior to the instructional methods and was used to assess group similarity only.
Reliability of the STAI is well established. Reported alpha coefficients for the state anxiety scale were above .90 for samples of working adults, college students, and military recruits (Speilberger, Gorsuch, & Lushene, 1983). Calculation of reliability for the data collected during the STAI state scale in this CVIS study resulted in a coefficient alpha of .95.
Data Analysis and Results
ANALYSIS OF PRETEST MEASURES: SUBJECTS' KNOWLEDGE OF THERAPEUTIC COMMUNICATION PRIOR TO INSTRUCTION- Although the subjects had already been randomly assigned to either Group I or Group II when pretest data were collected, the investigators wished to test whether or not there were any significant differences between the groups regarding their pretest knowledge of therapeutic communication. An analysis of variance was used to determine if there was a significant difference between Group I and Group II on the Knowledge Pretest on Therapeutic Communication. The knowledge pretest total score was the dependent variable, and the group category was the independent variable. The results revealed that there was no significant difference between the groups regarding their pretest knowledge of therapeutic communication, F(I) = 1.08, p<.316 (Group I X=5.58, SD=2.78 and Group II X=4.94, SD=2.66).
ANALYSIS OF POSTTEST MEASURES: SUBJECTS' KNOWLEDGE OF THERAPEUTIC COMMUNICATION FOLLOWING INSTRUCTION- HYPOTHESIS 1- To determine whether Group I scored significantly higher than Group II on the Cognitive Posttest on Therapeutic Communication, an analysis of covariance using the knowledge pretest total score as the covariate was performed. The main effect results indicated that a significant difference existed between the two groups, F(D= 37.69, p<.001.
Table 1 presents the analysis of covariance data. Group I subjects who had participated in the supplementary CVIS instructional program regarding communicating therapeutically with psychiatric patients evidenced a significantly higher posttest score than did Group II, who received the supplementary videotape. This finding is also revealed in the significantly higher adjusted mean score on the knowledge posttest for Group I subjects (X =21. 18, DS=3.97) over the Group II subjects (X=8.52, DS=4.80).
ANALYSIS OF PRETEST MEASURES: SUBJECTS' STATE -TRAIT ANXIETY PROFILES PRIOR TO INSTRUCTION- In an effort to gather baseline data and to assess similarity of the subjects regarding both their general anxiety proneness (trait anxiety) and their situational anxiety related to anticipating their first psychiatric nursing clinical experience (state anxiety), prior to any type of instruction regarding therapeutic communication, two analyses of variance (ANOVA) were done. The state anxiety and the trait anxiety scores were the dependent variables and the independent variable was the group category, Group I or Group II. Results indicated no significant difference between groups on either state anxiety, F(l)=.46, p<.49, or trait anxiety, F(l)=3.08, p<.08.
ANALYSIS OF POSTTEST MEASURES: SUBJECTS' ANXIETY STATE PROFILE FOLLOWING INSTRUCTION-HYPOTHESIS 2- To test the hypothesis that Group I subjects would evidence significantly less state anxiety at posttest (following lecture discussion plus CVIS instruction regarding therapeutic communication) than would Group II (following lecture discussion plus videotape), an analysis of covariance, using the pretest state anxiety total score as the covariate, was performed. The main effect results revealed no significant difference in state anxiety total scores between the two groups, F(I)=. 618, p<.434. Table 2 presents the adjusted means, standard deviations, and results of the analysis of covariance regarding posttest state anxiety total scores.
MEANS, STANDARD DEVIATIONS, AND RESULTS OF AN ANCOVA FOR KNOWLEDGE POSTTEST SCORES ON THERAPEUTIC COMMUNICATION
An additional observation that may be seen in Table 2 is that the state anxiety mean scores following instruction regarding therapeutic communication for both Group I and Group II subjects are significantly related to their state anxiety mean scores prior to instruction, as indicated by the significant covariate effect, F(ll)=65.12, p<.001. Thus, subjects' experience of situational state anxiety related to anticipating their therapeutic interactions with psychiatric patients remained similar from pretest to posttest.
Additional Data Analysis
SUBJECTS' PRETEST ANXIETY STATE MEAN SCORES COMPARED TO NORMATIVE MEAN SCORES FOR COLLEGE STUDENTS- -An additional comparison was made of the pretest state anxiety mean scores for males and females in both Group I and Group II and the normative mean scores for college students as reported by Spielberger, Gorsuch, and Lushene (1983). Both male and female subjects in this study evidenced higher mean scores on the state anxiety scale than did the normative group. The mean scores for the three males in Group I (X=47.53) and the three males in Group II (X=45.66) correspond to the 84th percentile of the normative mean scores for male college students (X=36.47). The mean scores for the 33 females in Group I (X= 46.21) and the 36 females in Group II (X=47.53) correspond to the 78th percentile of the normative mean scores for female college students (X=38.76). This finding is most likely due to the specific mind set suggested to subjects in this study prior to their completing the state anxiety scale, i.e., that subjects try to imagine and anticipate their first clinical therapeutic interviewing session with a psychiatric patient. More pertinently, however, it does appear that the situation of anticipating interviewing psychiatric patients in a clinical setting is anxiety inducing.
RELATIONSHIP OF DEMOGRAPHIC VARIABLES TO POSTTEST KNOWLEDGE SCORES REGARDING THERAPEUTIC COMMUNICATION- A multiple regression analysis was performed to determine which linear combination of the following demographic variables best predict the knowledge posttest scores on therapeutic communication, after adjusting for the effects of the pretest scores: age group, sex, previous work experience in psychiatry, previous computer experience, and membership in either the experimental or control group. A stepwise regression procedure was used, after the covariate (pretest total score) was entered. The results indicated that after the pretest covariate was accounted for, the next variable to enter the equation was the Group I or Group II membership, which accounted for 38% of the variance in predicting the knowledge posttest scores (r=.61, F=21.86, p<.001). Previous work experience in psychiatry was the only other variable in the demographic set, which added significantly to the prediction of therapeutic communication knowledge posttest scores. Interestingly enough, after group status was accounted for, findings indicated that those subjects without previous work experience with psychiatric patients tended to score higher on the knowledge posttest than did those with such previous work experience. The simple bivariate correlation between previous work experience in psychiatry and posttest scores was significant (r=.23, p<.05). None of the other demographic variables met the SPSS default minimum criteria for inclusion in the equation.
Limitations and Conclusions
Several limitations regarding this study are acknowledged. First of all, the sample groups represented only one baccalaureate school of nursing and findings are not generalizaba beyond the sample setting. Secondly, this study made no attempt to test the subjects' application of learned therapeutic communication skills in the clinical setting, nor the point at which the subjects may have experienced a decrease in their state anxiety related to interacting with psychiatric patients.
MEANS, STANDARD DEVIATIONS, AND RESULTS OF ANCOVA FOR POSTTEST ANXIETY-STATE SCORES
Based on the findings of this study, two conclusions were drawn. First, a more powerful and significant (p<.001) learning effect was observed for Group I subjects than for Group II subjects. This finding lends support to the conclusion of Huckabay and her associates (1979) that computer-assisted instructional programs do make significant contributions in enhancing knowledge. The CVIS instruction program used in this study was a very useful strategy for enhancing students' knowledge of therapeutic communication. This finding provides a basis for future studies using computer-video interactive programs to teach therapeutic communication and nurse-patient interaction skills to baccalaureate nursing students. Replication of this study is needed to further validate the finding that learning is increased when the CVIS instructional program is used to supplement lecture-discussion content regarding therapeutic communication. Specifically, such study might increase nurse educators' awareness of CVIS strategies in teaching nurse-patient interaction skills and in determining whether such CVIS instructional methods should be incorporated on a regular basis as an adjunct to other teaching strategies, such as lecture-discussion. Further, a wider variety of simulated nurse-patient interaction situations, including problematic ones, need to be incorporated into CVIS instructional programs and tested for learning effectiveness with students prior to, during, and after their clinical psychiatric nursing experience.
The second conclusion that may be drawn from the findings in this study is that the lecture-discussion and the CVIS instructional program on therapeutic communication techniques appeared to have little effect on either Group I or Group II subjects' state anxiety scores following the instruction. The state anxiety scores of Group I subjects did not differ significantly from those of Group II subjects at either pretest or posttest. The fact that the state anxiety scores of both Group I and Group II subjects remained relatively high at pretest and posttest lends further credence to the notion that the anticipated psychiatric nursing experience, including anticipated interactions with patients, is extremely anxiety producing for students, despite any cognitive learning regarding therapeutic communication techniques before the students enter the clinical experience. Additional questions arise, such as when, during the course of clinical psychiatric nursing, do students experience a decrease in their state anxiety levels? What instructional strategies could be utilized to help alleviate students' apprehension about the clinical psychiatric nursing experience? Is the anxiety experienced by students related to their perceived lack of sufficient therapeutic skills in interacting with patients?
Further issues evolved from the finding that those subjects who had previous work experience in psychiatry tended to score lower on the Therapeutic Communication Posttest than did those who had not previously worked in psychiatric settings. Does previous work experience in psychiatric settings interfere with new learning regarding therapeutic interactions with patients? Do students who have had previous work experience in psychiatry believe that they already know how to interact with patients, and thus, have nothing new to learn? What type of role models are encountered in psychiatric settings in terms of interacting with patients? Are role models in psychiatric settings seen as primarily socializing with patients rather than using deliberative therapeutic communication strategies in their interactions?
In addition, further studies are needed to investigate whether the CVIS program enhances the application of therapeutic communication skills by students and graduate nurses during their actual interactions with patients in psychiatric settings. Such educational technology may impact directly on the application of therapeutic communication skills by care providers, and thus, foster more positive patient care outcomes.
- Allen, B.S., Devney, A.M., & Sharpe, D.M. (1986, January-February). The effect of practice in detecting technical errors on performance of a simple medical procedure: Current research on interactive video at San Diego State University. Computers in Nursing, 11-16.
- Bitzer, M.D., Boudreaux, M., & Avner, R.A. (1973). Computer-based instruction of basic nursing utilizing inquiry approach (CERL Report X-40). Urbana, IL: Computer-Based Education Research Laboratory.
- Boettcher, E., Alderson, S., & Saccucci, M. (1981). A comparison of the effects of computer-assisted instruction versus printed instruction on student learning in the cognitive categories of knowledge and application. Journal of Computer-Based Instruction, 8, 13-17.
- Cattell, R., & Scheier, I. (1961). The meaning and measurement of neuroticism and anxiety. New York: Ronald Press.
- Conklin, D. (1983). A study of computer-assisted instruction in nursing education. Journal of Computer-Based Instruction, 9, 98-106.
- Edwards, M., & Hannah, K. (1985). An examination of the use of interactive videodisc cardiopulmonary resuscitation instruction for the lay community. Computers in Nursing, 3(6), 250-252.
- Green, C, & Mink, W. (1973). Evaluation of computer simulation of experiments in teaching scientific methodology (DHE WTDE No. OEG-5-72-0046-509). Washington, DC: National Center for Educational Research and Development.
- Halverson, J., & Bellinger, M. (1978). Computer-assisted instruction in surgery. Surgery, 6, 633-640.
- Huckabay, L., Anderson, N., Holm, D., & Lere, J. (1979). Cognitive, affective, and transfer of learning consequences of computerassisted instruction. Nursing Research, 26, 380-385.
- Kamp, M., & Burnside, I. (1974). Computer-assisted learning in graduate psychiatric nursing. The Journal of Nursing Education, 11, 18-25.
- Kirschoff, K., & Hölzerner, W. (1979). Student learning and a computer-assisted instructional program. Journal of Nursing Education, 3, 22-30.
- Lewis, L., Gadd, H., & O'Conner, K. (1987). Relationship of anxiety level and recall of information to the interval between orientation and first patient care day. Journal of Nursing Education, 26(3), 14-98.
- Mauksch, H. (1963). Becoming a nurse: A selective view. Annual American Political Science, 364, 88-89.
- Redland, A., & Kilmon, C. (1986). Interactive video: Rationale and practicalities of one experience. Computers in Nursing, 4(2), 68-72.
- Schoffstall, C. (1981). Concerns of student nurses prior to psychiatric nursing experience: An assessment and intervention technique. Journal of Psychosocial Nursing and Mental Health Services, 19, 11-14.
- Schwartz, M. (1984). An introduction to interactive video systems. Computers in Nursing, 2(2), 8-15.
- Spielberger, C, Gorsuch, R., & Lushene, R. (1983). Manual for the state-trait anxiety inventory, STAI (Form Y). Palo Alto, CA: Consulting Psychologist Press.
- Stein, R. (1969). The student nurse: A study of needs, roles and conflicts. Nursing Research, 18, 308-315.
- Thomas, D. (1979). The effectiveness of computer-assisted instruction in secondary schools. Association for Educational Data Systems Journal, 12, 103-116.
- Williams, M. (1979). Effects of clinical setting on anxiety and achievement in psychiatric nursing education. Journal of Nursing Education, 18, 4-14.
- Whitney, D. (1981). Insight 1000 plus documentation. San Mateo, CA: Whitney Educational Services.
MEANS, STANDARD DEVIATIONS, AND RESULTS OF AN ANCOVA FOR KNOWLEDGE POSTTEST SCORES ON THERAPEUTIC COMMUNICATION
MEANS, STANDARD DEVIATIONS, AND RESULTS OF ANCOVA FOR POSTTEST ANXIETY-STATE SCORES