Journal of Nursing Education

The articles prior to January 2012 are part of the back file collection and are not available with a current paid subscription. To access the article, you may purchase it or purchase the complete back file collection here

Performance Predictors for Nursing Courses and NCLEX-RN

Joan Felts, PhD, RN

Abstract

Introduction

Accountability is a pervasive and controversial issue in education. Questions regarding educational outcomes have been asked persistently, if intermittently, for several decades. Nursing education has a twofold accountability, for it must meet the needs of students for a quality education and society's need for professionals capable of delivering nursing care to meet its health needs (Rothman & Rothman, 1977).

Curriculum requirements in broad categories of science and nursing are set forth for schools of nursing by state boards of nursing. Programs are developed by faculty within the academic guidelines of the parent institution. Nursing educators have a duty to examine their curricula periodically in order to identify those components most predictive of their students' ultimate success in the profession (Melcolm, Venn, & Bausel, 1981).

The predictive power of the traditional cognitive measures as grades, college admission scores, nursing achievement tests, and college GPA is generally supported in the literature (Beale & McCutcheon, 1980; de Tornyay & Russell, 1978; Halpin, Halpin & Häuf, 1976; Kissinger & Munjas, 1982; Owen & Feldhusen, 1970; Sharpe, 1984; Wolfe & Bryant, 1978).

The use of noncognitive variables such as age, psychological measures, personality traits, and previous nursing experience to predict success has produced inconclusive results (Aldag & Rose, 1983; Donsky & Judge, 1981; Frerichs, 1973; Hayes, 1981; Hutcheson, Garland & Lowe, 1979; Lunneborg, Olch & deWolf, 1974; Yess, 1980).

The changing demographic pattern impacts nursing education since 20% of new college enrollments are not coming directly from high school (Hodgkinson, 1983). Declining enrollments mandate that institutions of higher learning place emphasis on quality in academic programming and instruction (The Carnegie Foundation for the Advancement of Teaching, 1978).

There is a paucity of literature relative to the contribution of the sciences and humanities to nursing. Educators are beginning to assess the role of a liberal education in curriculum (Donaldson, 1983; Fields, 1984; Riesman, 1980; Winter & McClelland, 1981; Zwerling, 1984).

The curriculum provides the mechanism by which a graduate is legally eligible to write the licensure examination. These examinations serve to establish limits of accountability by defining the parameters of activity for a given nurse (Hull, 1981). The national licensure examination, NCLEX-RN, in place since 1982 under the aegis of the National Council of State Boards, is criterion referenced and is based on behaviors identified in a validity study (Holmes, Showalter & Heidorn, 1983). No studies have been found which use the NCLEX-RN as the dependent variable in studying predictors of success for nursing students.

An assessment of nursing education is essential for diverse reasons including: financial exiguity in higher education in both the public and private sector; the rights of students as consumers; decreasing enrollment in the traditional age group with an increase in attendance of the nontraditional and minority student; the role of liberal arts in professional curricula; faculty responsibility germane to curriculum development; the controversy regarding accountability in education; career mobility and the changing role of the health care provider.

The purpose of this study was twofold: (1) to determine which selected cognitive variables most effectively predicted successful performance in nursing courses, and (2) to examine the relationship between selected cognitive and demographic variables and performance on the NCLEXRN.

Hypotheses

Ho1 There are no significant admission variables useful in predicting the grade point average in the nursing courses.

Ho2 There are no significant cognitive variables (grades in support courses, grade point averages in the humanities, the biological, physical and social sciences and the support courses GPA) useful in predicting the grade point average in the nursing courses.

Ho3 There are no cognitive variables (high school GPA, ACT scores, grades in support and…

Introduction

Accountability is a pervasive and controversial issue in education. Questions regarding educational outcomes have been asked persistently, if intermittently, for several decades. Nursing education has a twofold accountability, for it must meet the needs of students for a quality education and society's need for professionals capable of delivering nursing care to meet its health needs (Rothman & Rothman, 1977).

Curriculum requirements in broad categories of science and nursing are set forth for schools of nursing by state boards of nursing. Programs are developed by faculty within the academic guidelines of the parent institution. Nursing educators have a duty to examine their curricula periodically in order to identify those components most predictive of their students' ultimate success in the profession (Melcolm, Venn, & Bausel, 1981).

The predictive power of the traditional cognitive measures as grades, college admission scores, nursing achievement tests, and college GPA is generally supported in the literature (Beale & McCutcheon, 1980; de Tornyay & Russell, 1978; Halpin, Halpin & Häuf, 1976; Kissinger & Munjas, 1982; Owen & Feldhusen, 1970; Sharpe, 1984; Wolfe & Bryant, 1978).

The use of noncognitive variables such as age, psychological measures, personality traits, and previous nursing experience to predict success has produced inconclusive results (Aldag & Rose, 1983; Donsky & Judge, 1981; Frerichs, 1973; Hayes, 1981; Hutcheson, Garland & Lowe, 1979; Lunneborg, Olch & deWolf, 1974; Yess, 1980).

The changing demographic pattern impacts nursing education since 20% of new college enrollments are not coming directly from high school (Hodgkinson, 1983). Declining enrollments mandate that institutions of higher learning place emphasis on quality in academic programming and instruction (The Carnegie Foundation for the Advancement of Teaching, 1978).

There is a paucity of literature relative to the contribution of the sciences and humanities to nursing. Educators are beginning to assess the role of a liberal education in curriculum (Donaldson, 1983; Fields, 1984; Riesman, 1980; Winter & McClelland, 1981; Zwerling, 1984).

The curriculum provides the mechanism by which a graduate is legally eligible to write the licensure examination. These examinations serve to establish limits of accountability by defining the parameters of activity for a given nurse (Hull, 1981). The national licensure examination, NCLEX-RN, in place since 1982 under the aegis of the National Council of State Boards, is criterion referenced and is based on behaviors identified in a validity study (Holmes, Showalter & Heidorn, 1983). No studies have been found which use the NCLEX-RN as the dependent variable in studying predictors of success for nursing students.

An assessment of nursing education is essential for diverse reasons including: financial exiguity in higher education in both the public and private sector; the rights of students as consumers; decreasing enrollment in the traditional age group with an increase in attendance of the nontraditional and minority student; the role of liberal arts in professional curricula; faculty responsibility germane to curriculum development; the controversy regarding accountability in education; career mobility and the changing role of the health care provider.

The purpose of this study was twofold: (1) to determine which selected cognitive variables most effectively predicted successful performance in nursing courses, and (2) to examine the relationship between selected cognitive and demographic variables and performance on the NCLEXRN.

Hypotheses

Ho1 There are no significant admission variables useful in predicting the grade point average in the nursing courses.

Ho2 There are no significant cognitive variables (grades in support courses, grade point averages in the humanities, the biological, physical and social sciences and the support courses GPA) useful in predicting the grade point average in the nursing courses.

Ho3 There are no cognitive variables (high school GPA, ACT scores, grades in support and nursing courses and CGPAl that significantly discriminate those students who pass the NCLEX-RN from those students who do not pass.

Ho4 There is no difference between age groups in the proportion of students who pass the NCLEXRN.

Ho5 There is no difference between unlicensed students and licensed practical nurse students in the proportion that pass the NCLEX-RN.

Assumptions and Limitations

The study assumed that differences in incentive to successfully write the licensure examination would occur randomly among the subjects and that the disparity in the number of support courses and nursing courses required in the different programs would not bias the results. It was also assumed that the NCLEX-RN is a reliable and valid measurement instrument and that it remains stable across the years.

The study was limited in its investigation because data from one of the six unilevel associate degree programs in the state were not available. A second limitation was that the differences in grading policies among the high schools and colleges germane to work and GPA was not adjusted or standardized.

Definition of Terms

For the purpose of this study, the following terms were operationally defined.

Admission Criteria: High School GPA and American College Testing (ACT) Scores.

Age Groups: Age is divided into three groups as delineated by the National Center for Education Statistics - 18-24; 25-34 and 35 and older.

American College Jesting (ACT) Scores: Four Academic Tests, English Usage, Mathematics Usage, Social Studies, Reading, and Natural Sciences assess reasoning abilities and knowledge. Standard scores are reported in each area, with a composite which is the mean score.

Behavioral or Social Sciences Courses: Those courses dealing with the functioning of society and the interpersonal relationships of individuals, as psychology, sociology and human development.

Biological Science Courses: The knowledge of objects or processes observed in nature as anatomy, physiology, microbiology, and nutrition.

Humanities Courses: The study of human values, traditions, ideals, thoughts and actions as philosophy, language, theology, and literature.

Nursing Courses: Those courses that comprise the nursing component of the program.

Physical Science Courses: The study of natural laws as chemistry and physics.

Support Courses: All of the non-nursing courses that are required in the nursing program.

Cumulative Grade Point Average lCGPA>: The summation of grade points (A - 4 points, B - 3 points, C - 2 points, D=I point) in the designated number of courses and obtained by calculating the statistical mean.

Dependent or Criterion Variables: The nursing grade point average and the NCLEX-RN.

Independent or Predictor Variables: Admission criteria, age, course grades, GPA on designated academic areas, CGPA, and previous licensure as a practical nurse.

Unilevel Associate Degree Nursing Programs: A program that provides the curriculum for only one entry level to practice.

Research Methodology

An ex post facto design was used to test the hypotheses. The setting was five unilevel National League for Nursing accredited associate degree nursing programs in a midwestern state. The subjects were 297 first time writers of the NCLEX-RN between July 1982 and February 1984.

Information on each variable was not available for all students because of lack of data on high school transcripts, ACT scores not required of the non-traditional age students, and because of the differences in course requirements in the five programs. The number of subjects included in the statistical procedure to test each hypothesis is listed in the analysis of data.

Courses were categorized in the areas of humanities, and biological, behavioral and physical sciences. No differentiation was made between courses transferred into a program and courses taken at the parent institution. If a student repeated a college course, the highest grade earned was used. A GPA was calculated for the defined courses in the sciences and the humanities, and for all the nursing courses. All of the collected data were treated as confidential and coded for analysis.

The distributional characteristics of the variables were examined by measures of central tendency and dispersion. Regression analyses were computed to examine the relationship between the GPA in the nursing courses and sets of predictor variables.

Discriminate analysis and chi-square tests were used to statistically distinguish between those who passed the NCLEX-RN and those who did not pass utilizing the hypothesized variables (Huberty, 1975 ). Statistical analysis was accomplished through the use of the system of computer programs, SPSSX (Norusis, 1983).

Results

The age of the subjects ranged from 17 to 57 with 49% in the 18-24 group and 16% in the 35 and older. The mean age was 27. Thirty-five subjects were licensed practical nurses. Table 1 illustrates the dependent variables.

Table

TABLE 1SUMMARY OF DEPENDENT VARIABLES

TABLE 1

SUMMARY OF DEPENDENT VARIABLES

Table

TABLE 2MULTIPLE REGRESSION: ADMISSION CRITERIA VARIABLES MEANS AND STANDARD DEVIATIONS N =99

TABLE 2

MULTIPLE REGRESSION: ADMISSION CRITERIA VARIABLES MEANS AND STANDARD DEVIATIONS N =99

Table

TABLE 3MULTIPLE REGRESSION; COLLEGE COURSE VARIABLES MEANS AND STANDARD DEVIATIONS N = 166

TABLE 3

MULTIPLE REGRESSION; COLLEGE COURSE VARIABLES MEANS AND STANDARD DEVIATIONS N = 166

Testing of Hypothesis

H] There are no significant admission variables useful in predicting the GPA in the nursing courses.

Table 2 illustrates the variables used in the forward stepwise multiple regression procedure to select the minimum number of variables which provide the best prediction. The ACT composite score was a significant predictor (p<.001), thus H1 was rejected. The F was 35.867 and accounted for 27% of the variance in the nursing courses GPA.

Table

TABLE 4REGRESSION SUMMARY TABLE FOR COGNITIVE VARIABLES AND NURSING COURSES GPA N = 166

TABLE 4

REGRESSION SUMMARY TABLE FOR COGNITIVE VARIABLES AND NURSING COURSES GPA N = 166

Table

TABLE 5DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN ADMISSION CRITERIA VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 99

TABLE 5

DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN ADMISSION CRITERIA VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 99

H2 There are no significant cognitive variables (grades in support courses, GPA in the humanities, biological, physical and social sciences, and the support courses) useful in predicting the GPA in the nursing courses.

Variables used in testing this hypothesis are found in Table 3. As shown in Table 4, the support courses GPA and microbiology are the significant predictors (p<.001), thus H2 was rejected. The support courses GPA accounted for 46% of the variance in the nursing courses GPA and the inclusion of microbiology only accounted for one additional percent.

Table

TABLE 6DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN COURSE GRADE VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 121

TABLE 6

DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN COURSE GRADE VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 121

H3 There are no cognitive variables (high school GPA, ACT scores grades in support and nursing courses and CGPA) that significantly discriminate those students who pass the NCLEX-RN from those students who do not pass.

Three discriminant analyses using admission criteria (Table 5), college course grades (Table 6), and college GPA's (Table 7) were used to test this hypothesis.

As illustrated in Table 5, the admission variables of the high school GPA and ACT Social Studies score yielded significant discrimination (p<.001) between the pass/fail groups on the NCLEX-RN. The canonical correlation coefficient was .403 which indicates that 16% of the variance in the discriminant function can be explained by group membership. The discriminating functions of these variables correctly classified the subjects in the group they most likely belonged to 69.7% of the time and incorrectly for 30.3% of the time. Of the 60 cases predicted to pass, three failed; of the 39 cases predicted to fail, 27 passed.

The next set of variables used in the discriminant analysis was the grades obtained in the college courses. The initial variables list with all of the identified course grades required by each college produced no cases. The courses of nutrition, theology and philosophy were deleted because they were required by only two schools in the sample. Table 6 lists the courses used in calculating the discriminant function of course grades. Variables which yielded significant discriminations between the pass/fail groups were microbiology, anatomy/physiology, sociology, child psychology and English I. The canonical correlation coefficient obtained was .541 which indicates that 29% of the variance in the discriminant function can be explained by group membership. The discriminating functions of the five significant variables that entered the stepwise procedure correctly classified the 172 subjects in the group they most likely belonged to 73.26% of the time and incorrectly for 26.74% of the time. Of the 116 cases predicted to pass, six failed; and of the 56 cases predicted to fail, 40 passed.

Table

TABLE 7DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN GPA VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 261

TABLE 7

DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN GPA VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 261

The last variables used in the discriminant analysis were the average grades in the sciences and the humanities, the support and nursing courses GPA, and the CGPA which is shown in Table 7. The variable which yielded significant discrimination between the pass/fail groups was the cumulative GPA. The equivalent F was 77.109, p<.001. The canonical correlation coefficient was .478 which indicates that 23% of the variance in the discriminant function can be explained by group membership. The discriminating function of the CGPA correctly classified the 297 subjects in the group they most likely belonged to 77.44% of the time. Of the 191 cases predicted to pass, two failed, and of the 106 cases predicted to fail, 65 passed. The data analysis did not support H3.

H4 There is no difference between age groups in the proportion of students who pass the NCLEX-RN.

A chi-square for independence was computed for graduates on the relationship between age (18-24, 25-34, and 35 and older) and performance on the NCLEX-RN. The chisquare was non-significant, X2 ^) = 4.51008, p<.05. Thus, H4 failed to be rejected.

H5 There is no difference between unlicensed students and licensed practical nurse students in the proportion that pass the NCLEXRN.

A chi-square was computed for nursing graduates on the relationship between licensure as a practical nurse and performance on the NCLEX-RN. The chi-square was nonsignificant X2 (1) = 0.65515, p<.05. Thus, H5 failed to be rejected.

Conclusions and Recommendations

Based on the findings, the researcher concludes: (1) The ACT composite score is the best admission criteria predictor for success in nursing courses, while the support courses GPA and microbiology are the best college variable predictors. This is consistent with the review of over a decade of research by Grant (1983), who concluded that future success is best predicted by past success. The identification of microbiology as a predictor may be due to the integration of the scientific principles in clinical practice.

(2) Performance in college courses predicts pass/fail status on the NCLEX-RN with greater accuracy than does performance in high school. While there were significant predictors from both high school and college, performance in college carries a greater ability to differentiate pass/fail group membership. In the validity study a significant correlation was found with the licensure examination of July 1977 and the students' GPA's (Holmes, Showalter & Heidorn, 1983). The relationship of the CGPA and the NCLEX-RN is supported in this study.

(3) Grades in the courses in the biological sciences, social sciences, and the humanities differentiate those students who pass and those who fail the NCLEX-RN. Nursing behaviors derived from the nursing process are assessed by the NCLEX-RN and the integration of the problem solving process and the relevance of this content to the nursing practice of the graduate may be a contributing factor to the discrimination effectiveness of these courses.

(4) The role of the nursing courses in relation to the NCLEX-RN was not identified. Nursing courses contain content identified as imperative for nursing practice by educators and regulatory agencies. This supports the need for the validation of the professional licensure examination, a study which is currently under (American Journal of Nursing, 1984).

(5) Age and licensure as a practical nurse does not differentiate the group who passed from the group who failed the NCLEX-RN. Age does not affect an individual's ability to perform, which is of particular importance because of the enrollment patterns showing an increase in age.

The recommendations for further research based on the results of this study are:

1 . Nursing educators must further identify the knowledge and skills essential for basic nursing practice.

2. The process validating the identified knowledge and nursing skills should be completed to assist nursing curriculum developers.

3. Nurse educators should conduct further analysis of the contributions of the humanities, the biological, physical and social sciences to nursing.

Particular curriculum content extends beyond entry level job performance. Attention must be paid to the role of the humanities in nursing education, especially in view of the exodus of nurses from the profession. If, as Robert Bellah (1981) states, moral inaneness generates cognitively superficial work, then it is imperative to strengthen the humanities dimension of nursing education with the intent of assisting the nurse in developing his or her human potential, both in and out of the workplace.

References

  • Aldag, J. & Rose, S. (1983). Relationship of age, American college testing scores, grade point average, and state board examination scores. Research in Nursing and Health, 6, 69-73.
  • Beale, A. V. & McCutcheon, A.F. (1980). On becoming a nurse. Journal of Nursing Education, 19(4), 28-32.
  • Bellah, RN. (1981). Ethical aims of social inquiry, Teach Call Ree, 83, 1-18.
  • Carnegie foundation for the advancement of teaching fair practices in higher education. San Francisco: Jossey-Bass Publishers, 1979.
  • de Tornyay, R. & Russell, M.L. (1978). Helping the high risk student achieve. Nursing Outlook, 26(9), 576-580.
  • Donaldson, S.K. (1983). Let us not abandon the humanities. Nursing Outlook, 31, 40-43.
  • Donsky, A.P. & Judge, A.J. Jr. (1981). Academic and non-academic characteristics as predictors of persistence in an associate degree nursing program, Paper presented at the Twenty First Annual Forum of the Association for Institutional Research.
  • Fields, C.M. (1984). Medical schools urged to stress critical thought. The Chronical of Higher Education, 29, 1.
  • Frerichs, M. (1973). Relationship of self-esteem and internalexternal control to selected characteristics of associate degree nursing students. Nursing Research, 22, 350-352.
  • Grant, R.E. (1983). Predicting academic success. In WL. Hölzerner (Ed.), Review of research in nursing education. Thorofare, New Jersey: SLACK Inc.
  • Halpin, G., Halpin, G., & Häuf, B. (1976). Incremental validity of the ACT test battery for predicting success in a school of nursing over a ten year period. Educational and Psychological Measurement, 36(1-2), 433-437.
  • Hayes, E. R. (1981). Prediction of academic success in a baccalaureate nursing education program. Journal of Nursing Education, 20, 4-8.
  • Hodgkinson, H.L.I 1983 ). Guess who's coming to college. Academic, 69, 13-20.
  • Holmes, S.E., Showalter, R.E., & Heidorn, D. L. (1983). Developing, constructing and scoring the national council licensure examination. Issues, 4(1), 1-8.
  • Huberty, C. J. (1975). Discriminant analysis. Review of Educational Research, 45, 543-597.
  • Hull, R. T. (1981). Responsibility and accountability analyzed. Nursing Outlook, 29, 707-712.
  • Hutcheson, J.D. Jr., Garland, L.M., & Löwe, L.S. (1979). Antecedents of nursing school attrition: Attitudinal dimensions. Nursing Research, 25(1), 57-63.
  • Kissinger, J.F. &Munjas, B.A. (1982). Predictors of student success. Nursing Outlook, 30(1), 53-54.
  • Lunneborg, P.W., Olch, D.R., & deWolf, V. (1974). Prediction of college performance in older students. Journal of Counseling Psychology, 21(3), 215-221.
  • Melcolm, N., Venn, R. & Bauseil, R.B. U98U The prediction of state board test pool examination scores within an integrated curriculum. Journal of Nursing Education, 20(5), 24-28.
  • NCSBN slates job analysis study of RN licensure exam. (1984). American Journal of Nursing, 667-676.
  • Norusis, M.J. (1983). Spss* advanced statistical guide. New York: McGraw-Hill.
  • Owen, S.V. & Feldhusen, JF. (1970). Effectiveness of three models of multivariate prediction of academic success in nursing education. Nursing Research, 19(6), 517-525.
  • Riesman, D. (1980). On higher education. San Francisco: JosseyBass Publishers.
  • Rothman, D.A. & Rothman, N.L. (1977). The professional nurse and the law. Boston: Little, Brown and Co.
  • Sharp, T.G. (1984). An analysis of the relationship of seven selected variables to state board test pool examination performances of the University of Tennessee, Knoxville, college of nursing. Journal of Nursing Education, 57-63.
  • Spss". ? complete guide to Spss1 language and operations. New York: McGraw-HUl, 1983.
  • Winter, D.G. & McClelland, D. C. (1981). A new case for the liberal arts. San Francisco: Jossey-Bass Publishers.
  • Wolfe, L.M. & Bryant, L. W. (1978). A causal model of nursing education and state board examination scores. Nursing Research, 27(5), 311-315.
  • Yess, J-P (1980). Predictors of success in community college nursing education. Journal of Nursing Education* 19, 19-24.
  • Zwerling, L.S. (1984). A new mission for continuing education: Teaching the skills of the liberal arts. The Chronical of Higher Education, 28, 80.

TABLE 1

SUMMARY OF DEPENDENT VARIABLES

TABLE 2

MULTIPLE REGRESSION: ADMISSION CRITERIA VARIABLES MEANS AND STANDARD DEVIATIONS N =99

TABLE 3

MULTIPLE REGRESSION; COLLEGE COURSE VARIABLES MEANS AND STANDARD DEVIATIONS N = 166

TABLE 4

REGRESSION SUMMARY TABLE FOR COGNITIVE VARIABLES AND NURSING COURSES GPA N = 166

TABLE 5

DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN ADMISSION CRITERIA VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 99

TABLE 6

DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN COURSE GRADE VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 121

TABLE 7

DISCRIMINANT ANALYSIS FOR PASS/FAIL GROUPS ON NCLEX-RN GPA VARIABLE MEANS AND STANDARD DEVIATIONS FOR PASS GROUP AND FAIL GROUP N = 261

10.3928/0148-4834-19861101-06

Sign up to receive

Journal E-contents