Accurately predicting student success and passage of state licensure examinations has become a very critical issue for nursing schools. Up until approximately five years ago, nursing educators were able to enjoy the benefits of selecting top students for nursing schools from a large applicant pool. However, times have changed and there has been a sharp decline in the number of students applying for admission into baccalaureate nursing programs. In fact, in order to fill classes, nursing schools have had to accept students with less than ideal academic backgrounds (i.e., students they would have previously not admitted because of low grade point averages, class standing, etc.).
At the same time, a cry is heard from the State of Florida that the products of colleges and universities are not of the caliber that they should be, particularly in the public university system.
In response to this dilemma and to upgrade the educational standards of public university students in Florida, as well as to ensure that students have the necessary skills to be successful in a university setting, the State instituted the College Level Academic Skills Test (CLAST) in 1984 as a mandatory requirement for admission to the junior year in a state university. This test examined student achievement in the areas of computation (math), reading, writing, and essay writing. In order to enter the junior year at any Florida public university, the test had to be passed with a minimum score of 260 out of a maximum score of approximately 400 (this total varies somewhat each time the examinations are given) in each of the areas of computation, reading, and writing, and a score of four (out of a possible eight) in essay writing. These scores will be raised by a phase-in process, to 295 and five respectively in 1989 (Florida Department of Education, 1984).
The completion of the CLAST exam is a new requirement. Until this past year, students entering the junior level at Florida International University were not required to take the CLAST examination. They had been requested to do so on a voluntary basis, but many students chose not to take the exam. Others who did take the exam did so with very little preparation beforehand. Thus, admission criteria for the first nursing class in 1982 did not include the CLAST scores. Students were admitted if they had an earned grade point average (GPA) of 2.5 on a 4.0 scale and had completed the necessary prerequisite science courses with a grade of C or better.
Currently, to be admitted to the university and subsequently to the school of nursing, prospective students have to meet the following criteria: 1) a grade point average (GPA) of 2.5 on a 4.0 scale; 2) completion of all prerequisite science courses with a grade of C or better; and 3) earn a satisfactory score on the CLAST. A satisfactory score is one that is determined by the Board of Regents of the Florida State University System.
OVERVIEW OF ADMITTED NURSING STUDENTS' ETHNICITY, ATTRITION AND PASSING RATES FOR 1984 AND 1985
The CLAST Examination tests the ability of the student to read comprehensively and to write a sentence clearly and concisely. The ability to conceptualize and analyze at a basic level appears to be tested by computation and essay portions of the examination.
Sixty-five generic students, i.e., those who were not registered nurses, were admitted in August 1982. Of these 65 students admitted, 49 graduated in April 1984. This demonstrated an attrition rate of 25%. Of the 49 graduating and sitting for the nursing licensing examination, 69% passed and became registered nurses. Sixty-nine percent represented a less than desirable passing rate, as a school of nursing in the State of Florida must have at least 80% passage to remain in good standing with the board of nursing.
After receiving the results, the school of nursing analyzed the graduates' records and other data to determine where the problem lay. The major factor found in the data of those students who had failed appeared to be the ethnicity of the student. Upon further analysis, those individuals with foreign backgrounds had the most failures in the program, as well as the most failures on the licensing examination (Table 1).
In 1983, the school of nursing admitted a second class of 58 generic students. Thirty-four of these students graduated in April 1985. This class had a much higher attrition rate than the previous class. The attrition rate was 41%. However, the passing rate for this group taking the licensing examination was 88%. To date, the four graduates failing the exam have not contacted the school, so their characteristics cannot be used to determine whether there is any similarity with the first class.
In reviewing the data collected on the most recent graduating class, the school noted that ethnicity was still a major indicator of the ability to successfully complete the program and pass the licensing exam. However, in further pursuing the data, CLAST scores appear to be the most predictive of the ability to graduate from the program. Of interest is the fact that the foreign-born student, regardless of ethnicity, had the lowest CLAST scores, even where the primary language of the student was English.
The 45 students in the class of 1985, with a combined average score of 290 or above in reading, writing, and computation, and a minimum of four on the essay, had no difficulty earning a passing grade in nursing courses and graduating with the bachelor of science in nursing. The classes in the program have shown that the failure rate has been exclusive with those having earned CLAST scores below the 290 composite average and less than four on the essay.
The CLAST exam is now fully implemented. This implementation means that new or prospective students earning less than the required entering score on the CLAST exam are prevented from enrolling in the university. Prospective students are allowed to repeat the examination and are admitted if they successfully pass each section with at least 260 in reading, writing, and computation and a score of four on the essay portion of the exam.
The school of nursing studied the possibility of an entrance requirement of a composite average of 290 for scores earned in computation, reading, and writing. The rationale for this action was that students admitted with scores below 290 were being set up for failure. It would seem that students scoring below 290 would not have the necessary skills to be successful. The school of nursing does not have the resources to assist these students to attain the necessary academic skills. However, the university and the community colleges do offer such classes, and with the nursing school's recruiter/advisor's assistance, students are referred to the necessary remedial classes.
A major question still remaining was whether CLAST scores have any correlation with scores received by graduate nurses on the licensure examination. The most recent graduates sat for the licensure exam in July 1985. The overall result of the school's graduates was 88%. Thus, this study addresses the significance of the admission criteria used by the school of nursing at Florida International University and how it relates to the success of students in completing the nursing program, passing the state licensure examination, and becoming registered nurses (RNs).
Review of Literature
A review of literature revealed limited information on CLAST-type exams and was void of data that related CLAST to the nursing licensure exam. Further investigation did demonstrate that the variables of ethnicity, age, capacity to abstract, Scholastic Aptitude Scores, GPA, and National League for Nursing scores on achievement tests were used in predicting success on the nursing licensure exam.
Halprin (1984) studied a predominantly Black baccalaureate nursing school in a private college. She investigated the relationship between the students' high school grade point averages, Scholastic Aptitude Test's (SAT) verbal and quantitative scores, and National League for Nursing pre-nursing examination scores. Discriminate analysis of these predictors, plus college grade point averages, demonstrated a correlation between those who passed and those who failed the nursing licensure exam.
A recent study by Sharp (1984 ) at a Tennessee university, was conducted to determine whether any one of seven selected variables, or a combination of the variables, were predictive of performance on the nursing licensure exam. The selected variables were; high school grade point average, the college of nursing's grade point average, the American College Test Assessment standard, scores in English, mathematics, social studies, natural sciences, and composite. Data utilized were from graduates of the baccalaureate program from 1977 through 1979. Selected variables found to be predictive of the nursing licensure exam were grade point averages and standard achievement scores in mathematics and natural sciences.
A similar study by Rose (1983) looked at the relationships between age, American College Testing (ACT) scores, and college grade point averages to the results on the nursing licensure exam. She found there was a significant negative correlation between the ACT scores and age. The proportion of older students who graduated and passed the licensure examination was significantly higher than the proportion of younger students. Two admission procedures used, ACT and GPA, resulted in similar rates for graduation and passage on the licensure exam. Age and ACT scores were positively correlated with the licensure exam scores.
Data on all graduates from 1970 through 1976 from a nursing school in southwest Virginia were analyzed by Wolf and Bryant (1983). The school achievement test and the Scholastic Aptitude Test were used to determine if any relationship existed between performance on the two exams and the nursing licensure examination scores. Wolfe and Bryant concluded that licensure exam results on medical, surgical, and obstetrical portions depended on mastery of the subject matter and correlated positively with National League for Nursing Achievement Test results. However, performance on the psychiatric and pediatrie portions, which measured deductive ability as well as content, correlated with the SAT results.
The ability of a student to use abstract and deductive thinking plays a major role in student success. Jean Piaget, a cognitive psychologist, studied the ability to deduct and use abstract thinking. According to Piaget, an individual moves through several developmental stages before reaching adulthood and full cognitive ability in Western cultures (Adams and Macione, 1983, p. 65).
In Western society, adolescents from age 12 to 15 years begin to demonstrate the intellectual capacity for symbolic logic in thinking about concepts and ideas. However, in more primitive cultures, this ability to abstract is less necessary and, therefore, may not be evident, but it can be developed. Unlike its concrete predecessor, this stage involves a manipulation of principles and the construction and formulation of theoretical and conceptual abstractions, such as mathematical ability, particularly geometry, and creative writing, such as essay, and utilized abstract thinking (Adas and Macione, 1983).
Abstract and creative thinking were used in the curriculum design of one nursing school (Koehler, 1984). The department of nursing at California State University revised its curriculum to allow students to become more independent learners. Courses were designed in a modular format which allowed students to move from concrete to abstract material as they achieved mastery. Faculty expectations were that students would display creativity and abstract ability once they felt secure with the assigned learning. Faculty also believed that students would learn flexibility and independence, skills needed to pass the nursing licensure exam. One third of the students completed the old curriculum (standard didactic instruction) and two thirds the new curriculum. Results showed that faculty expectations were not met. Failure on the licensing examination was higher for students completing the new curriculum.
A literature review on CLAST produced the following information about the examination: The 1979 Florida Legislature Chapter 79-222 charged the State Department of Education with the responsibility for 1) compiling a list of communication and computation skills associated with successful student performance in baccalaureate programs at public universities; 2) approving tests to measure achievement of these skills; and 3) setting performance standards. The 1982 Florida Legislature mandated the use of scores on the test as a prerequisite for an associate of arts degree, or admission to upper division status in a state university (Walsh, 1985).
Validity of CLAST, as a measure of achievement of academic skills, was established by following a plan for developing and selecting items. Content and testing specialists judged the adequacy of the items for measuring the skills, and the plan for selecting items ensured that each form of CLAST was representative of the domain of skills tested. Given the specifications and conditions of administration, interpretation, scoring, calibration methods, score reliability, item bias prevention, and item analysis, the test has, from its inception, met the requisites for a valid and reliable standardized test (Walsh, 1985).
In validating the CLAST for use as an admission tool, Miami-Dade Community College investigated whether a combination of freshman and sophomore grades and student scores on the CLAST would be a better predictor of student academic success at the four-year level than CLAST scores alone. In the absence of data to determine predictive validity of CLAST, college performance of students with a high grade point average from high school and scores on the Comparative Guidance Placement test were analyzed. The study concluded that achievement test results were not better predictors of overall success than previous academic performance, and that the use of CLAST results alone would probably lead to erroneous admission decisions (Walsh, 1985).
Statement of Hypotheses
Based upon the problem statement and the review of the literature, the following hypotheses were formed:
1. Nursing students with a score of less than 290 on reading, writing, and computation, and less than four on the essay will fail to complete the nursing program more frequently than those students with scores of 290 or above and four or more on the essay.
2. Graduate nurses with CLAST scores below a composite average of 290 in reading, writing, and/or computation, and less than four on the essay will be less less likely to pass the nursing licensure exam than those with 290 and above and four or more on the essay.
3. The ethnicity of the nursing students at Florida International University will be a factor in successfully completing the nursing program.
4. The ethnicity of the graduate nurse will be a factor in the ability to pass the state examination for registered nurse licensure at Florida International University.
5. Nursing graduates who were born and received their basic education in a foreign country will be more likely to fail the nursing licensure examination than nursing graduates born and receiving basic education in the United States.
1. CLAST Scores: Scores earned by individual students on the standardized tests administered by the Florida public colleges and universities to all students earning an associate in arts degree and/or entering the junior year at a public university.
2. Examination for Registered Nurse Licensure: A nationally prepared examination administered by the Florida State Board of Nursing to graduates from approved schools of nursing. The examination tests minimum nursing competencies of graduates and must be passed before the graduate nurse can practice as a registered nurse.
3. Passing Rate for Registered Nurse Licensure Examination: Passing rate has been established to be 1600 or above in the State of Florida.
Significance of the Study
This study addressed the issue of whether CLAST scores were predictors of academic success for students seeking a baccalaureate degree in nursing and for their ability to pass the examination for registered nurse licensure on their first attempt.
Students' GPA, ethnicity, and place of birth (in or outside of the United States) were addressed to determine whether these factors had any bearing in the success rate of nursing students and graduates.
The study was conducted to assist the school of nursing in determining the type of students who would be successful in pursuing an academic path to professional nursing, i.e., at the baccalaureate level. The study further identified the level of academic skills such as reading, writing, and math that may be necessary for students prior to their entrance into the nursing program in order for them to successfully complete the nursing program and, subsequently, to pass the necessary examination for registered nurse licensure.
The sample population consisted of 55 students admitted to the program in 1983 and graduating in 1985. The sample population was ethnically mixed, with minorities making up approximately 50% of the group. Minorities included were identified as Black, Hispanic, and Asian.
Independent: The independent variable was the scores earned by subjects on the CLAST examination.
Moderator: Moderator variables included the ethnicity of the student, whether the student was born in and completed basic education in another country, and the student's grade point average at the time of being admitted to the nursing program.
Dependent: Dependent variables consisted of the success of students in pursuing the baccalaureate degree and the ability of the graduate nurse to pass the registered nurse licensure examination on the first attempt.
Student files maintained in the school of nursing for individuals entering the nursing program in 1983 were reviewed. To aid in controlling other variables that might bias the study, the following types of students were excluded from the study:
1. Students transferring from another upper division university with more than 90 credits.
2. Students having an earned baccalaureate or a higher degree at time of admission.
3. Students not completing the CLAST examination because of an earlier admission to the university when the examination was not required.
Information from student nies included each student's learned GPA earned at the time of admission to the nursing program, the student's ethnicity and age as identified by the student on the admission form, and the scores earned on the CLAST as indicated on the individual's university transcript.
Scores from the nursing licensure exam were obtained by direct contact with 1985 graduates from the school of nursing. Individual graduate scores were not available from the State Board of Nursing. The State Board of Nursing would not release them due to the students' right to privacy. Therefore, each graduate was individually contacted by phone and asked for their score on the licensure exam.
Collected data were analyzed using the Pearson Product Moment Correlation Coefficient Index (Pearson r). The Pearson r was used to summarize the magnitude and direction of the relationships between the identified variables.
When determining the overall correlation of the variables of the licensure examination, (N-CLEX), CLAST, and GPA, all three variables were found to be significantly correleted, i.e., .6811 for N-CLEX and CLAST (p<.005); .7270 for N-CLEX and GPA (p<.005 ); and .4954 for CLAST and GPA (p<.01) (Table 2).
CORRELATIONS BY COUNTRY
CORRELATIONS BY ETHNICITY
The correlations for students born and completing high school in the United States versus the foreign-born and educated students could not be shown to differ. (GPA and CLAST were not correlated because of the small sample size) (Table 3).
In correlations by ethnicity, the CLAST and N-CLEX, and GPA and N-CLEX were not correlated for foreign-born due to the small sample size. The correlations for whites and minorities did not differ. However, the mean CLAST for white was higher than that for the foreign born at .0031 level of significance (Table 4). In analyzing correlations by sex, the number of males (N = 4) was too small to show anything. However, all pairs were significant for females at the .005 level of significance (Table 5).
CORRELATIONS BY SEX
PARTIAL CORRELATIONS - EFFECT OF AGE
Age does help to predict succès on the N-CLEX at .00368 level of significance, on the CLAST at .00263 level of significance, and the GPA at .0040 level of significance. Even after adjusting for age, N-CLEX and CLAST, and NCLEX and GPA were significantly correlated. However, after adjusting for age, GPA and CLAST were no longer correlated (Table 6).
When determining the relationship between nursing licensure (N-CLEX) and CLAST scores, it was found that there was sufficient evidence to show they were directly correlated, r = .69 (p<.01). The relationship between NCLEX and GPA was also significant, r = .73 (p<.01). Further, the relationship between CLAST and GPA was r = .50 (p=. 01) (Table 7).
Discussion and Conclusion
The issue of predicting success of nursing students on the nursing licensure exam has been a critical, timely, and continuing dilemma. In the past, schools of nursing have used criteria such as GPA, age of student, SAT scores, various achievement tests, and the teaching of critical thinking as variables to determine whether there was any correlation with success in nursing school and with the licensure exam scores.
The CLAST exam has been added as a possible predictor of nursing student success. This exam is new and is required of all students before they are accepted into the junior year in Florida state universities. Its use in identifying whether students can conceptualize and analyze data makes it an excellent tool for predicting academic success in baccalaureate nursing schools and state licensure exams, since both require the ability to think abstractly.
It is important to remember, however, that academic predictor tests do not always measure success in nursing school, as cognitive acumen does not necessarily correlate with psychological maturity and discipline. Although chronological age does positively correlate with higher CLAST scores, it does not mean that chronological age measures emotional age, which can be a deciding factor of whether a student remains in school or is able to function under the pressure of a national examination.
Although the pool of qualified nursing students is diminishing, the nursing profession cannot afford to admit unqualified students. Thus a possible solution could be that nurse educators use exams similar to the CLAST exam as a predictive variable in nursing success and passage of state licensure exams. At the same time, they must also look at the student as a person, i.e., motivation, reason for wanting to be a nurse.
The literature on the CLAST exam and its ability to predict that which it was intended to has thus far been scarce. In searching for predictive measures of the licensure exam success, the question was asked whether CLAST scores were predictors of academic success for students seeking a baccalaureate degree in nursing, and whether CLAST scores could predict a student's ability to pass the licensure exam on the first attempt.
The Pearson r showed that CLAST scores do correlate positively with licensure exam scores. That is, the greater the CLAST score, the greater the licensure exam score.
Further analysis using the Pearson r demonstrated that GPA and licensure exam scores were positively correlated, which gives further support to Halprin's (1984) study, where there was a direct correlation between SAT scores, NLN prenursing examination scores, the student GPA, and those individuals who failed or passed the nursing licensure exam. This study further supported studies conducted by Sharp (1984) and Rose (1983). Both Sharp and Rose found that grade point average, and ACT scores were most predictive of nursing licensure success. Rose further found that age was positively correlated with the licensure exam success. That is, the older the student, the more often they passed the licensure examination on the first attempt.
Correlation coefficient analysis for ethnicity, sex, and age revealed age to be highly correlated with CLAST, GPA, and licensure exam scores. That is, the older the student, the higher the CLAST score, the licensure exam score, and the GPA. Ethnicity and sex did not reveal any correlations, but this is likely to be due to the small sample size, i.e., only four males in the sample population.
It was intended that the licensure scores of the graduates would be obtained through the Florida State Board of Nursing, which would have resulted in a more complete and accurate presentation. However, the scores could not be released. The Florida State Board of Nursing has a policy that releasing the scores of individual students constitutes breaching of confidentiality and the individual's Constitutional right to privacy. Therefore, in order to obtain the scores for this study, each individual graduate was telephoned and asked for the score they received on the licensure exam. Seven of the students were unable to remember their scores and gave approximate guesses. Therefore, findings presented in this paper must be used as approximations, not as hard data.
Another limitation of this study was the small sample size, N = 55 of students entering the program and N = 34 for those graduating from the program. The limited sample size was due to the newness of the CLAST exam, and the use of only one school of nursing for the study.
Future recommendations include: 1) continue research in this area; 2) enlarge the study to include more schools of nursing in Florida; 3) compare the results found in Florida with other states requiring similar testing for admission to the junior year; and 4 ) conduct investigations to determine which of the variables isolated in this and other studies were most predictive of nursing student success.
- Adams, C. G., & Macione, A. (1983). Handbook of Psychiatric Nursing. New York: Wiley Medical.
- Questions and Answers about CLAST. (1984-85). Tallahassee, FL: State Department of Education.
- Halprin, G. (1984). Predictors of success in nursing school and on state board examinations in a predominantly black baccalaureate nursing program. Journal of Nursing Education, 23, 147-150.
- Koehler, M. (1984). Evaluating a curriculum. Journal of Nursing Education, 23, 1-39.
- Magarrel, J. (1981). Fall enrollment put as 12.3 million: A new record; up 1.6% in a year. Chronicle of Higher Education, 23(4), 1.
- Ostar, A.W. (1981). Part-time students: The new majority for the 1980s. Chronicle of Higher Education, 23(4), 56.
- Rose, S. (1983). Relationship of age, American College Testing scores, grade point average and state board examination results. Research in Nursing Health, 6(2), 69-73.
- Sharp, T.G. (1984). An analysis of the relationship of seven selected variables to state board test pool examination performance of the University of Tennessee, Knoxville, College of Nursing. Journal of Nursing Education, 23(1), 57-63.
- Walsh, S.L. (1985). CLAST study results. In Planning and Analysis, Miami, FL: Florida International University, Division of Planning and Analysis.
- Wolfe, L.M., & Bryant, L.W. (1983). A causal mode of nursing education and board examination scores. Nursing Research, 27(5), 311-316.
- Wright, T. (1983). Should one variable (CLAST) be used to determine entrance to upper division at the State University of Florida? (Report No. 83-23). Research Report, 143, 110.
OVERVIEW OF ADMITTED NURSING STUDENTS' ETHNICITY, ATTRITION AND PASSING RATES FOR 1984 AND 1985
CORRELATIONS BY COUNTRY
CORRELATIONS BY ETHNICITY
CORRELATIONS BY SEX
PARTIAL CORRELATIONS - EFFECT OF AGE