Please I would like a summary of this article not less than 200 words. Thank you.
390 Journal of Nursing Education
This study evaluated students’ demographic and nurs- ing program variables and standardized test scores to determine whether significant differences existed between students who successfully completed the National Council Licensure Examination for Registered Nurses (NCLEX-RN®) and those who were unsuccessful. In addition, the predictive accuracy of two standardized examinations, the Mosby AssessTest and the Health Education Systems, Incorporated (HESI) Exit Examination were compared.
Two cohorts of graduating senior nursing students were studied (1999 cohort N = 121; 2000 cohort N = 103). Demographic and nursing program variables were obtained from student records. The Undergraduate Studies Committee provided standardized test scores (Mosby AssessTest in 1999; HESI Exit Examination in 2000). Only two program variables were consistently associated with success on the NCLEX-RN—final course grade for a didactic, senior-level medical-surgical nursing course and cumulative program grade point average. Scores on both standardized tests were significantly dif- ferent in students who were successful on the NCLEX-RN
and those who were not. The HESI Exit Examination demonstrated greater sensitivity, specificity, positive and negative predictive value, and test efficiency, compared with the Mosby AssessTest.
Use of program variables and students’ standardized test scores may allow faculty to identify students at risk for failing the NCLEX-RN and to provide structured remediation so these students may be successful on the licensing examination and begin their nursing careers.
S uccessful completion of national licensure examina- tions is a key outcome for nursing graduates, as well as an important and highly visible indication of
effective nursing programs. Fewer nursing students across the United States are taking the National Council Licensure Examination for Registered Nurses (NCLEX- RN®), and those graduates who fail exhibit decreasing likelihood of successful completion with successive attempts (Dennis et al., 1990; Zuzelo, 1999). Recent NCLEX-RN data demonstrate a steady downward trend in successful completion for first-time candidates (National Council of State Boards of Nursing, 2000). In the United States, in 1997, 87.7% of graduate candidates for licensure successfully completed the NCLEX-RN. In 1998, 85.0% of candidates were successful, and in 1999 and 2000, licensure was obtained by 84.5% of first-time candidates. Only 82% of first-time candidates obtained licensure after their first NCLEX-RN attempt in 2001 (National Council of State Boards of Nursing, 2000).
The decline in successful completion of the NCLEX-RN is of concern not only for nursing graduates, faculty, and administration, but also for those who employ nurses and society in general. Nursing graduates must attain licen- sure prior to beginning their nursing career, and failure of the NCLEX-RN may produce both emotional and finan- cial hardship for graduates (Poorman & Webb, 2000). Although nursing programs evaluate the effectiveness of
Predictors of NCLEX-RN Success in a Baccalaureate Nursing Program as a Foundation for Remediation Linda K. Daley, PhD, RN; Bonnie L. Kirkpatrick, MS, RN; Susan K. Frazier, PhD, RN; Misook L. Chung, PhD, RN; and Debra K. Moser, DNSc, RN
Received: April 28, 2002 Accepted: January 30, 2003 Dr. Daley is Assistant Professor, Clinical, Ms. Kirkpatrick is Clinical
Instructor, and Dr. Frazier is Associate Professor, Ohio State University College of Nursing, Columbus, Ohio; and Dr. Chung is Postdoctoral Fellow, and Dr. Moser is Professor and Gill Chair of Cardiovascular Nursing, University of Kentucky, College of Nursing, Lexington, Kentucky.
Address correspondence to Linda K. Daley, PhD, RN, Assistant Professor, Clinical, Ohio State University College of Nursing, 1585 Neil Avenue, Columbus, OH 43210; e-mail: [email protected]
DALEY ET AL.
September 2003, Vol. 42, No. 9 391
their programs with many criteria, licensure examination success rates often are used as a broad indicator of pro- gram quality. In addition, accreditation by the American Association of Colleges of Nursing is based in part on NCLEX-RN success rates of programs’ graduates. Given the current and projected nursing shortage and its effect on health care, it is imperative that educators identify students at risk for failure to effectively intervene and increase the likelihood of their success.
Prior studies identified a variety of predictors of NCLEX-RN success for graduates of all types of nursing programs, including associate degree, diploma, and bac- calaureate degree programs (Arathuzik & Aber, 1998; Barkley, Rhodes, & DuFour, 1998; Roncoli, Lisanti, & Falcone, 2000). Campbell and Dickson (1996) found that cumulative nursing grade point average (GPA) was the best predictor of NCLEX-RN success. Parents’ level of edu- cation and student’s age were the best demographic pre- dictors of NCLEX-RN success in their meta-analysis of 47 studies of NCLEX-RN predictors (Campbell & Dickson, 1996). During a 9-year period, Mills, Sampel, Pohlman, and Becker (1992) identified cumulative nursing GPA as the best predictor of NCLEX-RN success in their sample of generic baccalaureate students (N = 534) and described admission criteria, pre-nursing GPA, and American College Test (ACT) scores as the least helpful determi- nants of NCLEX-RN success. Unfortunately, cumulative GPA is calculated after completion of the program and does not provide an opportunity for remediation.
Other investigators specifically examined variables related to the nursing curriculum. McClelland, Yang, and Glick (1992) found that in addition to pre-nursing GPA and ACT scores, grades in the biological and social sciences and chemistry were predictive of NCLEX-RN success in their baccalaureate students (N= 1,070). Endress (1997) com- pared predictors of NCLEX-RN success with White, Black, and foreign-born baccalaureate graduates (N = 150). There were no significant differences in NCLEX-RN pass or fail rate among the three groups. In general, graduates who were successful on the NCLEX-RN were more likely to have a higher admission GPA, higher cumulative nursing program GPA, and higher college final cumulative GPA than graduates who were unsuccessful.
Alexander and Brophy (1997) reviewed 5 years of NCLEX-RN performance and determined that Scholastic Aptitude Test (SAT) verbal scores, nursing cumulative GPA, and National League for Nursing comprehensive achievement scores were the most significant predictors of NCLEX-RN success in their sample of associate and baccalaureate degree graduates (N = 188). More recently, Roncoli et al. (2000) found that students with higher grades in science and nursing courses were more likely to be successful on state licensure examinations. Although GPA and science grades are identified to differing degrees as predictive of NCLEX-RN success in many studies,
other factors may significantly influence the success or failure of graduates.
Inclusion of nonacademic factors, such as family demands, finances, living arrangements, disorganization, role strain, and lack of confidence, may be important con- siderations in licensing examination success. Arathuzik and Aber (1998) studied academic, as well as nonacadem- ic, factors associated with NCLEX-RN performance among baccalaureate students with diverse backgrounds (N = 79). Of the academic variables studied, only cumula- tive undergraduate GPA was significantly related to NCLEX-RN success (p < .05). Among the nonacademic variables studied, English as the student’s primary lan- guage, family demands, emotional status, and a sense of test-taking competency were found to be significantly associated with NCLEX-RN success (p < .05).
A number of studies have examined the use of stan- dardized achievement examinations to predict NCLEX- RN readiness and likelihood of examination success. Among these, the Mosby AssessTest was found to be a moderate to strong predictor of NCLEX-RN success (Ashley & O’Neil, 1991; Fowles, 1992; Jenks, Selekman, Bross, & Paquet, 1989). Endress (1997) determined that students with Mosby AssessTest percentile ranks below 21 were more likely to fail the NCLEX-RN than students with ranks above this percentile rank. However, signifi- cant disadvantages of the Mosby AssessTest include its paper-and-pencil format, its length, and the time span between test administration and the ability to review the results. Recently, investigators have developed computer- ized examinations that may be effective predictors of NCLEX-RN success.
Lauchner, Newman, and Britt (1999) reported that use of the Health Education Systems, Incorporated (HESI) Exit Examination was a predictive indicator of student success on the NCLEX-RN. This examination provided immediate feedback to students in the form of a probabil- ity score (i.e., probability of success on the NCLEX-RN) and rationale for the correct answers for items missed. The HESI Exit Examination was administered to gradu- ates from associate degree (N = 1,976), diploma (N = 59), and baccalaureate (N = 520) programs. The HESI Exit Examination was determined to be predictive of success- ful completion of the NCLEX-RN in 99.5% of cases when the examination was monitored and 96.82% of cases when not monitored. In a follow-up study (Newman, Britt, & Lauchner, 2000), the investigators again determined the HESI Exit Examination to be highly predictive of NCLEX-RN success (98.3%). The investigators further examined the NCLEX-RN success rate for students clas- sified as low scoring on the HESI Exit Examination (i.e., probability score < 69%) and found that significantly fewer (p = .05) of these students failed when the HESI Exit Examination was used as a guide for remediation. Therefore, the HESI Exit Examination may be an impor- tant component of a remediation program.
The initial component of a comprehensive, structured approach to remediation requires determination of valid
PREDICTORS OF NCLEX-RN SUCCESS
392 Journal of Nursing Education
and reliable indicators of the need for help. Therefore, this study evaluated demographic and nursing program vari- ables to determine whether differences existed between students who were successful and unsuccessful on the NCLEX-RN and compared the predictive ability of the Mosby AssessTest and HESI Exit Examination in bac- calaureate graduates who took the NCLEX-RN in 1999 and 2000. The research questions that guided this study were:
• Are there demographic variables (i.e., gender, age, ethnicity, prerequisite GPA, ACT scores) associated with successful completion of the NCLEX-RN?
• Are there nursing program variables (i.e., final grades for chemistry, anatomy, sociology, zoology, patho- physiology, and didactic and clinical senior medical-surgi- cal nursing courses; final cumulative program GPA) asso- ciated with successful completion of the NCLEX-RN?
• What is the predictive ability of standardized tests (i.e., Mosby AssessTest, HESI Exit Examination) to iden- tify students in need of remediation prior to administra- tion of the NCLEX-RN?
Design In this ex post facto study, the demographic and acade-
mic characteristics of students who took the NCLEX-RN in 1999 and 2000 and were successful or unsuccessful in their initial attempt at the examination were examined and compared. In 1999, the Mosby AssessTest was admin- istered to graduating students, and in 2000, the HESI Exit Examination was provided to graduating seniors in the final quarter of the nursing program. The predictive ability of these two standardized tests for NCLEX-RN success also were examined and compared.
Sample Data from two cohorts of generic baccalaureate stu-
dents were analyzed. The first cohort consisted of 121 graduating undergraduate seniors in 1999. The second consisted of 103 graduating undergraduate seniors in 2000. In 1999, all graduating senior students were required to take the Mosby AssessTest. In 2000, graduat- ing senior students were offered the opportunity and strongly encouraged to take the HESI Exit Examination.
Measures Demographic and nursing program variables were col-
lected from student records. Variables were selected based on the literature review and prior research find- ings. Standardized test scores were obtained from the Undergraduate Studies Committee. All data were placed in a statistical spreadsheet for later analysis (i.e., Statistical Package for the Social Sciences 11.0).
Demographic Variables Demographic variables included student age, gender,
and ethnic background, as identified by the student; pre- requisite GPA; and ACT scores.
Program Variables Program variables included in this analysis were stu-
dents’ earned grades for several prerequisite courses and three nursing program courses. All courses were 10 weeks in length. The students were admitted to the undergrad- uate nursing program as sophomores, after completion of prerequisite courses as freshmen. The prerequisite cours- es included in this analysis were:
• One organic and one inorganic chemistry course (5 credit hours each).
• The human anatomy course (5 credit hours). • The introduction to sociology course (5 credit hours). • A zoology course that provided an introduction to
normal human physiology (5 credit hours). Course grades were based on a 0 to 4.0 scale and are equivalent to letter grades as follows: 3.8 to 4.0 = A; 3.4 to 3.7 = A–; 3.1 to 3.3 = B+; 2.8 to 3.0 = B; 2.4 to 2.7 = B–; 2.1 to 2.3 = C+; 1.8 to 2.0 = C; 1.4 to 1.7 = C–; 1.1 to 1.3 = D+; .8 to 1.0 = D; and < .7 = E.
The nursing program course grades included in this analysis were:
• The pathophysiology course that provided basic con- tent about human pathophysiology (5 credit hours).
• The didactic, senior medical-surgical nursing course that provided content about the care of adults with acute and chronic multiple system physical illnesses, with con- sideration of family, social issues, environmental systems, and continuity of care (4 credit hours).
• The clinical senior medical-surgical course in which students were provided with an intensive, precepted clin- ical experience caring for adults with acute and chronic multiple system physical illnesses (6 credit hours). The final cumulative program GPA also was considered a program variable in this analysis.
Standardized Test Scores In 1999, students were required to take the Mosby
AssessTest, and in 2000, students were offered the oppor- tunity and strongly encouraged to take the HESI Exit Examination. Both standardized tests were administered at the beginning of the students’ final quarter of the nurs- ing program. The Mosby AssessTest is a multiple-item, multiple-choice examination. This standardized, norma- tive examination was administered as a paper-and-pencil examination for 5 hours in a controlled environment. Mosby scored the examinations, and after a period of weeks, the results were reported to the faculty, who noti- fied students of their earned scores. Results were report- ed as raw scores, percentage of items correct, and per- centiles. There were no established cutoffs for prediction of success. However, the Professional Testing Corporation (1999) since reported that 87% of individuals who scored between 51% and 56% were successful on the NCLEX- RN.
The HESI Exit Examination is a multiple-item, multi- ple-choice examination. This examination used a blue- print similar to that of the NCLEX-RN and was adminis- tered in a controlled environment as a computerized
DALEY ET AL.
September 2003, Vol. 42, No. 9 393
examination lasting 4 hours. At the completion of the examination, the computer scored the HESI Exit Examination using the HESI Predictability Model (Lauchner et al., 1999). Students received immediate feedback and were able to meet with a faculty member to discuss their individualized results at the completion of the examination. Results were reported as raw scores and probability of success on the NCLEX-RN.
Data Analysis Means plus or minus standard deviations and/or fre-
quencies were used to describe sample characteristics. To examine differences in demographic variables, program variables, and standardized test scores between students who successfully completed NCLEX-RN and those who did not, independent t tests or chi-square tests were used, depending on the level of measurement of the dependent variable. The investigators were unable to perform a regression analysis because of the number of variables and the small number of students who were unsuccessful in their initial NCLEX-RN attempt. The significance level for this investigation was set a priori at p < .05. Bonferroni corrections were used where appropriate for multiple tests.
The sensitivity, specificity, positive and negative pre- dictive values, and test efficiency of the Mosby AssessTest and the HESI Exit Examination were calculated using standard formulas (Figure). Sensitivity evaluated the ability of the measure to predict failure on the NCLEX- RN in the entire group, and the specificity evaluated the
ability of the measure to predict success on the NCLEX- RN in the entire group. The positive predictive value indi- cated the percentage of students predicted to fail the NCLEX-RN who actually did fail, and the negative pre- dictive value indicated the percentage of students who were predicted to be successful on the NCLEX-RN who actually did pass. The efficiency of the test was calculat- ed as the percentage of time that the test provided the correct prediction, either pass or fail. The cutoff used for prediction of NCLEX-RN success was a rank at the 50th percentile or greater on the Mosby AssessTest. For the HESI Exit Examination, a cutoff for prediction of NCLEX-RN success was a probability score of � 69% (Newman et al., 2000).
Sample Characteristics A total of 224 undergraduate seniors in two cohorts
participated in this study. No significant differences exist- ed between the two cohorts (p > .05 for all variables). In 1999, the first cohort consisted of 121 students, who were all required to take the Mosby AssessTest. In the Mosby AssessTest cohort, 13 (10.7%) students failed the NCLEX- RN on their first attempt. In 2000, the second cohort con- sisted of 103 students. Eighty students voluntarily took the HESI Exit Examination. In the HESI Exit Examination cohort, 7 (6.8%) students failed the NCLEX- RN on the initial attempt. However, only 2 students who failed the NCLEX-RN also took the HESI Exit
Sensitivity of test = a/(a + c)
Specificity of test = d/(b + d)
Positive predictive value = a/(a + b)
Negative predictive value = d/(c + d)
Test efficiency = (a + d)/(a + b + c + d)
Figure. Calculation of the predictive value of the Mosby AssessTest and the HESI Exit Examination.
cell a cell b
cell c cell d
Predicted to Fail
Predicted to Pass
PREDICTORS OF NCLEX-RN SUCCESS
394 Journal of Nursing Education
Examination. Demographic characteristics of these two cohorts are presented in Table 1. Most students were White women in their early 20s, who graduated with a cumulative B+ average.
Demographic Variables In the Mosby AssessTest cohort, several demographic
characteristics were significantly different between stu- dents who were successful on the NCLEX-RN and those who were not (Table 2). Students who were successful on the NCLEX-RN were older (22.9 � 5.6 versus 20.4 � .9, p < .001), had earned a higher prerequisite GPA prior to program admission (3.3 � .4 versus 3.1 � .3, p = .005), and scored significantly higher on the ACT (23.6 � 2.9 versus 19.6 � 3.4, p < .005). In the HESI Exit Examination cohort, ethnicity was the only statistically different variable between students who were successful on the NCLEX-RN and those who were not (Table 3). Thirty-three percent of the non-White students were unsuccessful on the NCLEX-RN, compared to 4% of the White students. However, the validity of this finding is questionable because so few non-White students were in the group.
Program Variables In the Mosby AssessTest cohort, a number of program
variables demonstrated significant differences between students who were successful on the NCLEX-RN and those who were not (Table 2). Students who were success- ful on the NCLEX-RN achieved higher final grades for anatomy (3.2 � .6 versus 2.7 � .4, p = .009), pathophysi- ology (3.3 � .5 versus 2.5 � .5, p < .001), and both the
didactic and clinical senior medical-surgical nursing courses (didactic: 3.4 � .4 versus 2.8 � .6, p < .001; clini- cal: 3.5 � .5 versus 2.9 � .5, p < .001). Cumulative GPA also was significantly higher for students who were suc- cessful (3.4 � .2 versus 3.0 � .1, p < .001). Only two pro- gram variables demonstrated significant differences in the HESI Exit Examination cohort (Table 3). Students who were successful on the NCLEX-RN earned a higher final grade in the didactic portion of the senior medical- surgical course (3.1 � .6 versus 2.4 � .5, p = .004) and a higher final cumulative GPA (3.3 � .3 versus 3.0 � .02, p = .04).
Standardized Test Scores In the Mosby AssessTest cohort, significant differences
existed in Mosby AssessTest scores (Table 2). Students who were successful on the NCLEX-RN earned signifi- cantly higher raw scores (93.8 � 10.7 versus 77.5 � 11.6, p < .001), had a higher percentage of items correct (62.5 � 7.1 versus 51.5 � 7.7, p < .001), and a higher percentile rank (55.2 � 28.2 versus 19.7 � 19.3, p < .001). In the HESI Exit Examination cohort, students who were suc- cessful on the NCLEX-RN scored significantly higher on the HESI Exit Examination (67.6 � .2 versus 48.0 � .9, p = .002) (Table 3). Interestingly, students who voluntarily took the HESI Exit Examination were more likely to be successful on the NCLEX-RN than those who did not (p = .001).
The HESI Exit Examination demonstrated greater sensitivity, specificity, positive and negative predictive value, and test efficiency, compared to the Mosby AssessTest (Table 4).
TABLE 1 Student Characteristics
Mosby AssessTest Cohort HESI Exit Examination Cohort (N = 121) (N = 103)
Characteristic n (%) n (%)
Gender Men 21 (17.4) 10 (9.7)
Women 100 (82.6) 93 (90.3)
Ethnicity White 93 (76.9) 94 (91.3)
Black 8 (6.6) 4 (3.9)
American Indian 0 1 (1.0)
Asian 1 (.8) 4 (3.9)
Hispanic 1 (.8) 0
Missing data 18 (14.9) 0
Age* 22.7 � 5.7 22.5 � 5.1
Prerequisite GPA* 3.3 � .4 3.3 � .4
Final GPA* 3.4 � .3 3.3 � .3
* Values are reported as mean � standard deviation.
DALEY ET AL.
September 2003, Vol. 42, No. 9 395
The purposes of this investigation were to evaluate selected demographic and nursing program variables to determine whether differences existed between students who successfully completed the NCLEX-RN and those who were unsuccessful, to evaluate the usefulness of stan- dardized examinations in the prediction of NCLEX-RN success, and to compare the predictive ability of the Mosby AssessTest and the HESI Exit Examination with baccalaureate graduates who took the NCLEX-RN in 1999 and 2000. The results of this study suggest that few demographic or program variables consistently were
associated with success on the NCLEX-RN. Both stan- dardized examination scores were significantly different in students who were successful on the NCLEX-RN and those who were unsuccessful. When the standardized tests were compared, the HESI Exit Examination provid- ed greater sensitivity, specificity, positive and negative predictive value, and test efficiency than the Mosby AssessTest.
Many studies have evaluated predictors of NCLEX-RN success in associate degree, diploma, and baccalaureate program graduates (Alexander & Brophy, 1997; Arathuzik & Aber, 1998; Barkley et al., 1998; Endress, 1997; Fowles, 1992; McClelland et al., 1992), but few pre-
TABLE 2 Comparison of Demographic and Program Variables and Mosby AssessTest Scores
by NCLEX-RN Success for all Students (N = 121)
Unsuccessful on NCLEX-RN Successful on NCLEX-RN (n = 13) (n = 108)
Demographic Variables Mean �� SD Mean �� SD p Value†
Age 20.4 � .9 22.9 � 5.6 < .001
Ethnicity* .13 White 6 (75.0) 87 (91.6)
Non-White 2 (25.0) 8 (8.4)
Gender* .56 Men 3 (23.1) 18 (16.7)
Women 10 (76.9) 90 (83.3)
Prerequisite GPA 3.1 � .3 3.3 � .4 .005
ACT scores 19.6 � 3.4 23.6 � 2.9 .005
Chemistry 1 final course grade 2.9 � .7 3.2 � .6 .17
Chemistry 2 final course grade 3.1 � .6 3.1 � .6 1.0
Anatomy final course grade 2.7 � .4 3.2 � .6 .009
Social science final course grade 3.2 � .6 3.4 � .6 .2
Zoology final course grade 2.5 � .7 2.8 � .7 .13
Pathophysiology final course grade 2.5 � .5 3.3 � .5 < .001
Didactic senior medical-surgical final course grade 2.8 � .6 3.4 � .4 < .001
Clinical senior medical-surgical final course grade 2.9 � .5 3.5 � .5 < .001
Final cumulative program GPA 3.0 � .1 3.4 � .2 < .001
Mosby AssessTest Values
Mosby AssessTest raw score 77.5 � 11.6 93.8 � 10.7 < .001
Mosby AssessTest percentage of items correct 51.5 � 7.7 62.5 � 7.1 < .001
Mosby AssessTest percentile 19.7 � 19.3 55.2 � 28.2 < .001
* Values are reported as frequency and percentage. † p values for nominal variables were derived from chi-square tests, interval or ratio level variables from t tests. Note: Percentages for Ethnicity do not total 100% due to missing data.
PREDICTORS OF NCLEX-RN SUCCESS
396 Journal of Nursing Education
dictors have been identified consistently across studies. In this respect, data from this study were consistent with the literature. For both cohorts of students, only two pro- gram variables were found to be associated consistently with NCLEX-RN success. In the Mosby AssessTest cohort, several demographic predictors (i.e., age, higher prerequi- site GPA, higher ACT scores) and several program predic- tors (i.e., final grades for anatomy, pathophysiology, and the didactic and clinical medical-surgical nursing course; cumulative program GPA) were found to be significantly different in students who were successful on the NCLEX-
RN and those who were not. In the HESI Exit Examination cohort, the only significantly different demographic variable was ethnicity. Horns, O’Sullivan, and Goodman (1991) found ethnicity to be predictive of NCLEX-RN success. However, Endress (1997) compared Black, foreign-born, and White baccalaureate graduates and found no differences in NCLEX-RN success among the groups. Because of the small numbers of non-White students in this study’s sample, this finding is question- able and requires further study with a larger sample of non-White students.
TABLE 3 Comparison of Demographic and Program Variables and HESI Exit Examination Scores
by NCLEX-RN Success for all Students (N = 103)
Unsuccessful on NCLEX-RN Successful on NCLEX-RN Demographic (n = 7) (n = 96) Variables Mean �� SD Mean �� SD p Value†
Age 20.7 � 1.3 22.7 � 5.2 .33
Ethnicity* < .001 White 4 (57.1) 90 (93.8)
Non-White 3 (42.9) 6 (6.3)
Gender* .23 Men 0 10 (10.4)
Women 7 (100.0) 86 (89.6)
Prerequisite GPA 3.1 � .2 3.3 � .4 .3
ACT scores 12.0 � 10.0 16.1 � 10.1 .41
Chemistry 1 final course grade 2.6 � .3 3.0 � .6 .08
Chemistry 2 final course grade 2.8 � .5 3.0 � .7 .48
Anatomy final course grade 2.9 � .4 3.0 � .6 .56
Social science final course grade 3.2 � .4 3.4 � .6 .54
Zoology final course grade 2.2 � .9 2.7 � .7 .13
Pathophysiology final course grade 2.8 � .7 3.1 � .7 .32
Didactic senior medical-surgical final course grade 2.4 � .5 3.1 � .6 .004
Clinical senior medical-surgical final course grade 3.2 � .6 3.5 � .5 .09
Final cumulative program GPA 3.0 � .2 3.3 � .3 .04
HESI Exit Examination§ (n = 2) (n = 78)
HESI Exit Examination score 48.0 � .9 67.6 � .2 .002
HESI Exit Examination probability score 60.3 � 1.2 84.6 � 10.1 .001
Took HESI Exit Examination§ .001 No* 5 (71.4) 18 (18.8)
Yes* 2 (28.6) 78 (81.3)
* Values are reported as frequency and percentage. † p values for nominal variables were derived from chi-square tests, interval or ratio level variables from t tests. § Of the 103 students in this cohort, only 80 chose to take the HESI Exit Examination.
DALEY ET AL.
September 2003, Vol. 42, No. 9 397
Significantly different program variables in the HESI Exit Examination cohort were the didactic medical-surgi- cal nursing course grade and the cumulative program GPA. These findings are similar to many prior studies that found higher science grades and higher grades in nursing courses to be predictive of NCLEX-RN success (Griffiths, Bevil, O’Connor, & Wieland, 1995; McClelland et al., 1992; Waterhouse, Bucher, & Beeman, 1994). Similar to earlier studies (Alexander & Brophy, 1997; Campbell & Dickson, 1996), cumulative program GPA was found to be predictive in both cohorts of students. However, this variable does not provide a foundation for remediation. Although admission criteria and nursing program content were consistent for both groups, differ- ences existed in demographic and program predictors of NCLEX-RN success between the two cohorts. The major- ity of variables used to predict NCLEX-RN success appear to be inconsistent both in the literature and these two cohorts. Continued investigation of these variables as predictors is unlikely to be productive. It is probable that other variables, such as test-taking ability, academic and mental preparation, motivation, and response to stress, are important to students’ success. These factors require identification and investigation to determine their impor- tance.
The Mosby AssessTest and the HESI Exit Examination are standardized tests that have been used to predict NCLEX-RN success in other studies (Arathuzik & Aber, 1998; Endress, 1997; Newman et al., 2000; Roncoli et al., 2000). This investigation found that scores on both exam- inations were significantly different in students who were successful on the NCLEX-RN and those who were not. There were differences in the numbers of students in each cohort who took the standardized examinations. Students who finished the program in 1999 were required to take the Mosby AssessTest, and the college was able to provide financial support for this endeavor, so all students partic- ipated. However, in 2000, students who finished the pro- gram were strongly encouraged to take the HESI Exit
Examination, but they were required to pay for the exam- ination themselves (approximately $30). Therefore, only 80 students in this cohort took the HESI Exit Examination. Surprisingly, the choice to take the HESI Exit Examination was found to be significantly different in students who were successful on the NCLEX-RN and those who were not (p = .001). Students who chose not to take the HESI Exit Examination (n = 23) may have felt unprepared to take a comprehensive examination that provided a probability of being successful on the NCLEX- RN. The results would have provided concrete evidence of their lack of preparation for the licensing examination. Lack of financial support and time (4 hours) also may have prevented some students in this cohort from taking the HESI Exit Examination.
When the accuracy of the two examinations was com- pared, the HESI Exit Examination demonstrated greater sensitivity, specificity, positive and negative predictive values, and test efficiency. This is the first study to com- pare these two standardized tests in this way. The posi- tive predictive value of both examinations was low (Mosby: 19%, HESI: 22%). However, students who took these examinations and were predicted to fail most likely studied, took an NCLEX-RN review course, or prepared in some way for the NCLEX-RN. Unfortunately, data were not collected that describe this preparation. Test efficiency (i.e., the percentage of time the test correctly predicted success or failure on the NCLEX-RN) also was greater for the HESI Exit Examination (Mosby: 60%, HESI: 91%).
Based on the findings of this study, taking the HESI Exit Examination currently is required of all graduating senior nursing students at the college of nursing. Students meet with a faculty member after completion of the HESI Exit Examination to discuss their performance and plan for remediation. In addition, the investigators are in the process of developing a structured approach to ensure students are prepared to take the NCLEX-RN after graduation. Currently, students who receive less
TABLE 4 Measures of Accuracy for the Mosby AssessTest and the HESI Exit Examination
Mosby AssessTest HESI Exit Examination Accuracy Measure (N = 121) (n = 80)
Sensitivity (i.e., percentage predicted to fail in entire group who actually failed) 85% 100%
Specificity (i.e., percentage predicted to pass in entire group who actually passed) 57% 91%
Positive predictive value (i.e., percentage of students predicted to fail NCLEX-RN who actually failed) 19% 22%
Negative predictive value (i.e., percentage of students predicted to pass NCLEX-RN who actually passed) 97% 100%
Test efficiency (i.e., percentage of time the test provided the correct prediction) 60% 91%
PREDICTORS OF NCLEX-RN SUCCESS
398 Journal of Nursing Education
than a B– in any nursing course receive an invitation to participate in a structured NCLEX-RN preparation inde- pendent study course. The HESI Exit Examination is administered halfway through this review. Based on the probability score on the HESI Exit Examination, students prepare a written plan of study, identify materials they will use for preparation, and describe their strengths and weaknesses. Data from this cohort of students will be used to further develop this structured approach to stu- dent preparation.
This investigation has several limitations. The two cohorts contained unequal numbers of students, and there were fewer students in the HESI Exit Examination cohort. Not all students in the second cohort (2000) took the HESI Exit Examination. In addition, not all data were available, in particular, data related to student age and ACT scores. For example, older students and stu- dents who transferred from other schools did not have ACT scores in their student records.
The prevalence of NCLEX-RN failure is a concern with the accuracy evaluation of these standardized tests (i.e., the Mosby AssessTest and HESI Exit Examination). Prevalence in the population influences the positive and negative predictive values calculated. If prevalence is low, as with NCLEX-RN failures, the positive predictive value tends to exhibit a number of false positives, and the neg- ative predictive value tends to be close to 1.0. The preva- lence of NCLEX-RN failure in the Mosby AssessTest cohort was 10.7%, and in the HESI cohort, the failure prevalence was 6.8%. Unfortunately, only 2 students who failed the NCLEX-RN chose to take the HESI Exit Examination, which lowered the prevalence in the group analyzed to 3%. These findings must be considered with regard to the prevalence of NCLEX-RN failure.
Considering the current and projected nursing short- age and its effect on health care, nurse educators must identify students at risk for NCLEX-RN failure to effec- tively intervene and increase the likelihood of success. Use of predictive program variables and scores from stan- dardized tests, such as the HESI Exit Examination, allow faculty to identify vulnerable students and provide reme- diation so they may be successful on the NCLEX-RN and begin their nursing careers.
REFERENCES Alexander, J., & Brophy, G. (1997). A five-year study of gradu-
ates’ performance on NCLEX-RN. Journal of Nursing Education, 36, 443-445.
Arathuzik, D., & Aber, C. (1998). Factors associated with National Council Licensure Examination-Registered Nurse success. Journal of Professional Nursing, 14, 119-126.
Ashley, J., & O’Neil, J. (1991). The effectiveness of an interven- tion to promote successful performance on NCLEX-RN for baccalaureate students at risk for failure. Journal of Nursing Education, 30, 360-366.
Barkley, T., Rhodes, R., & DuFour, C. (1998). Licensure exami- nation. Predictors of success on the NCLEX-RN among bac- calaureate nursing students. Nursing and Health Care Perspectives, 19, 132-137.
Campbell, A., & Dickson, C. (1996). Predicting student success: A 10-year review using integrative review and meta-analysis. Journal of Professional Nursing, 12, 47-59.
Dennis, K., Edwards, S., Grau, W., Henning, E., Lee, B., & Moses, D. (1990). Enhancing the success of NCLEX-RN retesters. NursingConnections, 3(3), 43-50.
Endress, D. (1997). A comparison of predictors of success on NCLEX-RN for African American, foreign-born, and white baccalaureate graduates. Journal of Nursing Education, 36, 365-371.
Fowles, E. (1992). Predictors of success on NCLEX-RN and with- in the nursing curriculum: Implications for early interven- tion. Journal of Nursing Education, 31, 53-57.
Griffiths, M., Bevil, C., O’Connor, P., & Wieland, D. (1995). Anatomy and physiology as a predictor of success in bac- calaureate nursing students. Journal of Nursing Education, 34, 61-66.
Horns, P., O’Sullivan, P., & Goodman, R. (1991). The use of pro- gressive indicators as predictors of NCLEX-RN success and performance of BSN graduates. Journal of Nursing Education, 30, 9-14.
Jenks, J., Selekman, J., Bross, T., & Paquet, M. (1989). Success in NCLEX-RN: Identifying predictors and optimal timing for intervention. Journal of Nursing Education, 28, 112-117.
Lauchner, K., Newman, M., & Britt, R. (1999). Predicting licen- sure success with a computerized comprehensive nursing exam. The HESI Exit Exam. Computers in Nursing, 17, 120- 125.
McClelland, E., Yang, J., & Glick, O. (1992). A statewide study of academic variables affecting performance of baccalaureate nursing graduates on licensure examination. Journal of Professional Nursing, 8, 342-350.
Mills, A., Sampel, M., Pohlman, V., & Becker, A. (1992). The odds for success on NCLEX-RN by nurse candidates from a four- year baccalaureate nursing program. Journal of Nursing Education, 31, 403-408.
National Council of State Boards of Nursing. (2000). NCLEX-RN examination test plan for the National Council Licensure Examination for Registered Nurses, effective date April, 2001. Chicago: Author.
Newman, M., Britt, R., & Lauchner, K. (2000). Predictive accu- racy of the HESI Exit Exam. A followup study. Computers in Nursing, 18, 132-136.
Poorman, S., & Webb, C. (2000). Preparing to retake the NCLEX- RN: The experience of graduates who fail. Nurse Educator, 25(4), 175-180.
Professional Testing Corporation. (1999). Reliability and validity of Mosby’s RN AssessTest: Secured test 99. New York: Author.
Roncoli, M., Lisanti, P., & Falcone, A. (2000). Characteristics of baccalaureate graduates and NCLEX-RN performance. Journal of the New York State Nurses Association, 31(1), 17- 19.
Waterhouse, J., Bucher, L., & Beeman, P. (1994). Predicting NCLEX-RN performance: Cross-validating an identified clas- sification procedure. Journal of Professional Nursing, 10, 255- 260.
Zuzelo, P. (1999). Professional practice and the NCLEX exami- nation: A bottom-line approach. Nurse Educator, 24(3), 11-12.