Bilingual nurse education program: applicant characteristics that predict success.ABSTRACT Nurses are in great demand across the United States, but those fluent in both Spanish and English are in particularly short supply.This study examined three cohorts of students that entered a Spanish-English Nursing Education program to determine characteristics of applicants that produced student success. Unlike many nursing programs, entrance requirements for this bilingual program did not include a minimal grade point average (GPA) or previous course completions. Logistic regression was used to analyze the relationship between five different characteristics of entering students and their later success in the program. Success was measured in terms of program persistence and performance on the NCLEX-PN and NCLEX-RN exams. Incoming students with relatively high GPAs (M = 3.2) were significantly more likely to persist through the entire nursing program and pass the NCLEX-RN exam (p < .05) than those with lower GPAs (M = 2.5).
Key Words Academic Persistence--Bilingual Nurses--Nursing Education--Hispanic Grade Point Average- NCLEX Examination
WITH THE RAPID RISE IN THE SPANISH-SPEAKING PATIENT POPULATION ACROSS THE UNITED STATES, NURSES ARE NEEDED WHO ARE FLUENT IN BOTH SPANISH AND ENGLISH (Bathurst, 2007; Glazer & Alexandre, 2009; Kelley, 2007; Moore, 2005; Passel & Cohn, 2008; Rivera-Goba & Wallen, 2008; Smith, 2006; US Census Bureau, 2007, 2008; Vogt & Taningco, 2008). Hispanic students capable of meeting the demand for bilingual nurses are underrepresented in college science and nursing programs and face a number of important challenges to their success (Fry, 2002; Gasbarra & Johnson, 2008; Velez-McEvoy, 2010). For some potential nursing students, these challenges include weak academic backgrounds, a history of poor academic performance, and/or a lack of experience with rigorous science and mathematics classes (Anderson & Kim, 2006; Lopez, 2009). Nurse education programs that can address these challenges have the potential for graduating Hispanic students with the knowledge, skills, and language abilities needed to meet the strong demand for bilingual nurses.
Thacker (2005) has called for a "comprehensive, long-term approach for the recruitment and retention of Hispanic nurses" (p. 57), which can be answered with college-level nurse education programs capable of identifying and recruiting appropriate students. Such programs are most likely to succeed if academic strengths of Hispanic students are recognized and supported (e.g., encouraging cooperative and student-centered activities), while challenges such as low science and math skills are addressed by program components (e.g., specialized workshops or classes) (Drane, Smith, Light, Pinto, & Swarat, 2005). Many community colleges are meeting the urgent need for nurses by creating and expanding nursing programs, sometimes with financial, instructional, or logistical help from local health care organizations (Curtis & West, 1992; Thacker).
In 2003, a nursing education program was developed in Arizona specifically for students fluent in Spanish and English. This program, given the pseudonym in this study of the Spanish-English Nurse Education Program (SENE), was created with the goal of meeting the critical need for bilingual nurses. SENE was part of a concerted effort by several states to recruit bilingual students into educational programs that would lead to their licensure as nurses (Bagnato, 2004; Bathurst, 2007; Thacker, 2005). The students in the program often originated from Mexico, Puerto Rico, Peru, and Colombia, as well as other Latin American and Caribbean locations.
To be eligible for the SENE program, students needed to be fluent in both spoken and written Spanish and English and had to be United States citizens or have federal documents allowing them to work in this country legally. Fluency was determined during extensive interviews. Candidates were required to answer questions in Spanish and in English, write essays in the two languages, and interact with other applicants in group activities. Although no minimal prerequisite courses or grade point average (GPA) were required for entry into the program, the academic, professional, and, when relevant, personal backgrounds of applicants were reviewed by the admissions committee.
Individuals in the SENE program could potentially work as certified nursing assistants (CNAs) after one year of the program, complete an associate degree in nursing (ADN) and begin working as a licensed practical nurse (LPN) after two years of the program, or work as a registered nurse (RN) within three years of the program.
Three years were needed to finish the entire program because students often started with few required general education classes already completed, and because certain extra classes were required by SENE to improve student success. For example, SENE required the successful completion of Introductory Biology for Allied Health before students could take Human Anatomy and Physiology. General education classes, including all science classes, were provided by the community college involved in creating the program; clinical nursing classes were taught at another community college in the district, with labs and equipment specialized for nursing education.
This study addressed the following general question about the SENE program: How did the students' academic backgrounds create success or lack of success as measured by persistence in the program (persistence/nonpersistence) and performance (pass/not pass) on the National Council of State Boards of Nursing licensure examinations (NCLEX-PN[R] and NCLEX-RN[R])? More specifically, the study explored these questions: Did a relationship exist between success in these two areas and the characteristics of: a) entering-college GPA, b) number of college credits in mathematics and science upon entrance into the program, c) number of college credits in science upon entrance into the program, d) total number of college credits earned before entrance, and e) total number of developmental class credits in mathematics, English, and reading taken to reach college proficiency before the SENE program. Two null hypotheses were tested for these questions: the first, that no relationship existed between characteristics of students entering the SENE program and success as measured by persistence, and the second, that no relationship existed between characteristics of students entering the program and success in the program as measured by passing the NCLEX exams.
These characteristics were chosen because lack of academic preparedness is often cited as an important challenge facing Hispanic students attempting to pursue a science-intensive degree program (Gasbarra & Johnson, 2008; Hagedorn & Cepeda, 2004; Segura, 2002), and all but a few SENE students were Hispanic. The lack of a strong science and math academic background has been cited as a reason that Hispanic students are less successful in nursing programs (Seago & Spetz, 2003; Vogt & Taningco, 2008). The academic backgrounds of this population of SENE students varied widely, making them particularly well suited for this study.
Method The documents and institutional data of the SENE program were grounded in the language of the program and included student application folders containing incoming college GPAs; former college mathematics, science, and developmental classes; grades for each class; and other information regarding academic background and qualifications. Institutional documents provided information on the date of entry in and the date of exit out of the program (persistence), level of nursing education reached before exiting or graduating (certified nursing assistant, practical nurse, or RN), and reasons students may have left the program. Institutional data were also available on performance on the NCLEX-PN and the NCLEX-RN exams, with start and end dates of nursing certification if the exam was passed.
The institutional data and documents used dated from the start of the program (January 2003) through December 2007. Purposive sampling was used to select all files of students in the first three SENE cohorts (Groups 1, 2, and 3), since these students were all scheduled to complete the program by December 2007. Data were available on persistence and NCLEX scores for 71 participants (Group 1, 27 students; Group 2, 23 students; Group 3, 21 students).
To address the research questions, aspects of the SENE program pertinent to the questions were explored. The program involved three major phases. First, students took prerequisite classes (e.g., human anatomy and physiology, microbiology, English, chemistry, psychology, and algebra) as well as nursing assistant classes. At the end of the first year, students could take a standardized exam to gain their certified nursing assistant (CNA) certificate. During the second year, students took licensed practical nursing (LPN) classes, as well as any remaining general education classes. By the end of the second year, they were allowed to take the NCLEX-PN exam if they wanted to gain state licensure and work as LPNs; students were considered to have persisted to a nursing career if they reached this second stage of their education.
The SENE program also provided a third stage of instruction, in which students took registered nursing (RN) classes and completed the RN portion of the program. Successful students graduated at the end of their third year and were eligible to take the NCLEX-RN exam to gain state licensure. These students persisted through the entire program.
Administrators and instructors indicated that virtually all students took the NCLEX-PN exam and the NCLEX-RN exam after the end of second and third years of the program, respectively. Therefore, passing rates for both exams were used as measures of success. As information was not available on whether students chose not to sit for these exams, it is understood that students who took these exams but did not pass them may be grouped with students who did not attempt to pass these exams.
Results Quantitative data for answering questions related to persistence in the SENE program, the first dependent variable, came from the application and information folders of the first three cohorts and institutional data kept on the students. Analyses were made for both levels of persistence, making it through the LPN portion of the program or making it through the RN portion of the program, both of which resulted in practicing bilingual nurses. Nonpersistence was defined as exiting the program before either nursing portion was completed, either with no certification or after completion of the CNA portion of the program. Passing the NCLEX exam, the second dependent variable, was measured in three ways: a) students could pass the NCLEX-PN exam after completion of the second year; b) they could pass the NCLEX-RN exam, usually in the months following completion of the final RN portion of the program; or c) they could pass both the NCLEX-PN and NCLEX-RN exams.
The characteristics of incoming students, the independent variables in this study, included five measures of student academic strength: college grade point average (pre-GPA); number of college credits previously gained in mathematics (pre-math); number of college credits previously gained in science (pre-science); total number of college credits gained before entrance into the program (pre-credits); and total number of developmental (remedial) classes taken before the start of the program (pre-developmental). For mathematics, science, and all college credits, only a grade of "C" or better was counted since most programs and the state universities did not accept a grade of "D" for credit. Credits needed in developmental (remedial) classes were primarily in mathematics, reading, or English. As pre-GPA was not available for three students from cohort 3, the sample size for measurements, including pre-GPA, was 68 rather than 71.
Relationships between each of the five characteristics and the dependent variables (persistence and NCLEX exam results) were measured with a logistic regression statistical approach, using SPSS 15.0. Logistic regression is a statistical method used for prediction of the probability of occurrence of a categorical event using several predictor variables that may be either numerical or categorical (Pallant, 2007). In this study, the dependent variables were categorical, and the five characteristics were numerical. Logistic regression helped establish the relationship between the characteristics and the two dependent variables; in this way, a determination was made as to whether academic characteristics of students entering the SENE program were predictive of success in the program.
SUCCESS AS MEASURED BY PERSISTENCE Of the 71 students reviewed in this study, 49 students persisted through the second year (finishing LPN classes), and 37 completed all three years of the program (finishing RN classes). Logistic regression was performed to assess the impact of student academic backgrounds on the likelihood that the students would persist in the program through the LPN level. The full model containing all predictors was statistically significant, [chi square] (5, n = 68) = 11.56, p < .05, indicating that the model was able to distinguish between students who persisted to the LPN level and those who did not. The model as a whole explained between 15.6 percent (Cox and Snell [R.sup.2]) and 22.0 percent (Nagelkerke [R.sup.2]) of the variance in persistence to LPN level and correctly classified 70.6 percent of the cases.
As shown in the Table at left, two characteristics made a unique statistically significant contribution to the model (pre-GPA, pre-science). The strongest predictor of persisting to the LPN level was pre-GPA, with an odds ratio of 2.53. This indicated that students with higher GPAs were more than 2.5 times more likely to make it through the LPN level than those with lower GPAs, controlling for the other four factors in the model.
Logistic regression was also performed to assess the impact of student academic backgrounds on the likelihood that students will persist in the program through the RN level. The full model, containing all five predictors, was not statistically significant, [chi square] (5, n = 68) = 8.762, p = .119, indicating that the model was not able to distinguish between students who persisted and those who did not persist to the RN level of the program. When a logistic regression was performed to assess the impact of only pre-GPA scores on persistence to the RN level, however, the model was statistically significant [chi square] (1, n = 68) = 6.482, p < .01, indicating that this pre-GPA model was able to distinguish between students who persisted and those who did not persist to the RN level of the SENE program.
The model explained between 10.2 percent (Cox and Snell [R.sup.2]) and 13.6 percent (Nagelkerke [R.sup.2]) of the variance in persistence to RN level and correctly classified 61.8 percent of the cases. Pre-GPA affected the students' ability to persist in the program through the RN level, at a significance level of p = 0.016. The odds ratio of 3.37 indicated that students with higher GPAs (M = 3.2) were more than three times more likely to make it through the RN level of the program than those with lower GPAs (M = 2.5).
As illustrated in the Table, the first null hypothesis was rejected because pre-GPA was significantly related to persistence in the program through the LPN level (p = 0.050) as well as through the RN level (p = 0.016). For the first three cohorts of the SENE program, those who entered with higher GPAs were significantly more likely to persist in the program to both the LPN and RN levels than students with lower GPAs. This hypothesis was also rejected because the number of college science credits completed before the SENE program was significantly related to persistence through the LPN level of the program (p = 0.042), although the significance of this relationship did not hold for those who made it through the RN portion of the program (p = 0.708).
The number of pre-math credits completed before the program was not statistically related to persistence through the LPN portion (p = 0.062) or the RN portion of the program (p = 0.512). The number of pre-credits (LPN, p = 0.124; RN, p = 0.384) and the number of pre-developmental credits (LPN, p = 0.163; RN, p = 0.295) were also not significantly related to persistence in the program.
SUCCESS AS MEASURED BY PASSING THE NCLEX EXAM Of the 71 students in the SENE program reviewed in this study, 43 passed the NCLEX-PN exam and 31 passed the NCLEX-RN exam. To test the second null hypothesis regarding the relationship between characteristics of students and success in the program as measured by passing the NCLEX exams, logistic regression was first performed to assess the impact of student academic backgrounds on the likelihood that students would pass the NCLEXPN exam. However, the full model (pre-GPA, pre-math, pre-science, pre-credits, and pre-developmental) was not found to be significant; no single independent pre-program academic variable was shown to be significantly related to passing the LPN level exam. Therefore, the second null hypothesis could not be rejected with regard to passing the NCLEX-PN exam.
Logistic regression was also performed to assess the impact of student academic backgrounds on the likelihood that students pass the NCLEX-RN exam. The full model, containing all five predictors, was not statistically significant, [chi square](5, n = 68) = 8.728, p = .120, indicating that the model was not able to distinguish between students who passed the NCLEX-RN exam and those who did not pass the NCLEX-RN exam.
When a logistic regression was performed to assess the impact of only pre-GPA scores on passing the NCLEX-RN exam, however, the model was statistically significant, [chi square](1, n = 68) = 6.207, p < .05, indicating that the model was able to distinguish between students who passed the NCLEX-RN exam and those who did not pass this exam. The model explained between 8.7 percent (Cox and Snell [R.sup.2]) and 11.7 percent (Nagelkerke [R.sup.2]) of the variance in passing the NCLEX-RN exam and correctly classified 62.0 percent of cases. Pre-GPA affected students' ability to pass the NCLEX-RN exam at a significance level of p = 0.025.
The odds ratio of 3.06 indicated that students with higher GPAs were over three times more likely to pass the NCLEX-RN exam than those with lower GPAs. Therefore, because there was a significant relationship between the characteristics of pre-GPA and passing the NCLEX-RN exam, the second null hypothesis could be rejected in reference to this exam. Analyzed individually, the other pre-program factors did not have a significant relationship to passing either the NCLEX-PN or the NCLEX-RN exams.
Discussion Of the various pre-program academic factors that were analyzed, only entering GPA acted as a predictor of whether regular SENE students made it through the entire three-year program and also passed the NCLEX-RN exam. For colleges focused on maintaining a high persistence rate in their nursing programs to the RN level, the common requirement that entering students have a GPA of at least 2.5 may make sense. Gess-Newsome and Haden (2006) reported that students who were successful in science gateway classes had a self-reported GPA of 2.5 or greater in their non-science classes. Seago and Spetz (2003) reported that success in RN nursing programs could be predicted by a combination of high GPA and ability to pass core biology classes without repeating.
Most student files for cohorts 1, 2, and 3 of the SENE program did not have information on high school math and science credits, which may have been confounding variables with regard to academic preparation in science and mathematics for college level work. Although more than 66 percent of students (47 of 71) had at least 14 college credits with a "C" or better before starting the program, some students were just starting their college careers and had very few credits, while others were returning for a second career after completing a bachelor's degree with well over 100 credits. This great variance in credits did not appear to have a significant effect on either persistence in the program or performance on the NCLEX exams.
Other relationships may also be of interest to those attempting to determine academic indicators of who might persist through a bilingual nursing program. For example, the number of credits taken in developmental classes before entering the SENE program was not related to success in the program. Students who had needed no developmental classes and students who had needed several classes advanced to the LPN and RN levels in significantly similar numbers. For example, in cohort 2, one student who had taken eight developmental classes finished to the LPN level and passed the NCLEX-PN exam, while a student who had taken five developmental classes was pinned with her RN and subsequently passed the NCLEX-RN exam.
These results support the idea that developmental classes may allow certain students to successfully improve their academic skills before pursuing a college education and their chosen career. Attewell, Lavin, Domina, and Levey (2006) considered the practice of developmental placement a way for students of color to successfully attend both community colleges and four-year universities. They argued that developmental classes have allowed many students to ultimately complete their college education and pursue the career of their choice, a finding in line with this study.
Conclusion This study showed that for students in this bilingual nursing education program, entering GPA was the predominant factor related to student persistence. Persistence through both the LPN and RN levels of the program was significantly related to entry GPA level, while persistence to the LPN level was also significantly related to success in science classes before program entry. Of the various characteristics analyzed using logistic regression, only entering GPA showed a significant relationship to student persistence through the RN level of the program (p = 0.016). Persistence through the LPN level was also significantly related to pre-GPA scores (p = 0.050) as well as to pre-science credits (p = 0.042). For colleges focused on maintaining a high persistence rate, the requirement that entering students have a minimal GPA has statistical support, although no minimal GPA level was established by this study. Reviewing transcripts to note a history of successfully passing science classes may also serve as a predictor of success, at least through the LPN level of the program. Based on the results of this study, the total number of college credits gained before entering the program, the number of pre-program mathematics credits, and the number of pre-program developmental credits may not be predictive of success in terms of persistence.
The ability to pass the NCLEX-RN exam was also significantly related to pre-GPA (p = 0.025), although passage of the NCLEXPN exam was not. For the regular SENE students of this study, a higher GPA at the start of the program was related to both persistence through the LPN and RN portions of the program and the subsequent passing of the NCLEX-RN exam. GPA, therefore, seemed to be an important factor for potential SENE students who hoped to work as licensed registered nurses. The other four independent academic variables (pre-program math, pre-program science, and total pre-program college and developmental credits), were not significantly related to passing the NCLEX-PN and NCLEX-RN exams.
In the first three years of its existence, the SENE program was successful at graduating LPN- and RN-level students certified to help meet the community's need for bilingual nurses. This was accomplished without using a minimal GPA or minimal prerequisite class requirements for entrance into the program. However, since students with higher GPAs were more likely to succeed through the RN portion of the program and pass the NCLEX-RN exam, giving greater consideration to this component of an applicant's resume may improve the overall efficiency of the program, especially when openings and resources for educating future bilingual nurses are limited.
Anderson, E., & Kim, D. (2006). Increasing the success of minority students in science and technology. Washington, DC:American Council on Education. Retrieved from www.hmc.edu/files/ institutionalresearch/CPR%20%20Diversity/ACE_minority_STEM_2006.pdf
Attewell, R, Lavin, D., Domina,T., & Levey, T. (2006). New evidence on college remediation. Journal of Higher Education, 77(5), 886-924.
Bagnato, K. (2004). Racing for a cure. Community College Week, 17(10), 6-9.
Bathurst, P. (2007, April 1). Arizona faces lack of nurses. Arizona Republic, Section ECI, p. 1.
Curtis, S. M., & West, B. (1992, October). Expanding resources for working adults: A college/community/hospital collaboration. Paper presented at the University of South Carolina Conference,The Minority Student Today: Recruitment, Retention, and Success, San Antonio,TX. (ERIC Document Reproduction Service No. ED351049).
Drane, D., Smith, H., Light, G., Pinto, L., & Swarat, S. (2005).The gateway science workshop program: Enhancing student performance and retention in the sciences through peer-facilitated discussion. Journal of Science Education and Technology, 14(3), 337-352.
Fry, R. (2002). Latinos in higher education: Many enroll, too few graduate (Pew Hispanic Center Report).Washington, DC: Pew Research Center.
Gasbarra, R, & Johnson, J. (2008). Out before the game begins: Hispanic leaders talk about what's needed to bring more Hispanic youngsters into science, technology and math professions [Public Agenda Report]. Retrieved from http://inpathways.net/outbefore.PDF
Gess-Newsome, J., & Haden, C. (2006, January). Action research, student learning, and change: Reform in STEM college teaching. Paper presented at annual meeting of the Association of Science Teacher Educators, Portland, OR. Retrieved from www.wmich.edu/science/facilitatingchange/Products/ Gess-NewsomePoster.pdf
Glazer, G., & Alexandre, C. (2009). Legislative: The nursing shortage:A public health issue for all. Online Journal of Issues in Nursing, 14(1). Retrieved from www.nursingworld.org/MainMenuCategories/ ANAMarketplace/ANAPeriodicals/OJIN/Columns/Legislative/ NursingShortage.aspx
Hagedorn, L. S., & Cepeda, R. (2004). Serving Los Angeles: Urban community colleges and educational success among Latino students. Community College Journal of Research & Practice, 28(3), 199-211.
Kelley, J. (2007,June 10). Nursing programs struggle to expand. Arizona Republic, pp. B1-B7.
Lopez, M. H. (2009). Latinos and education: Explaining the attainment gap (Pew Hispanic Center Report).Washington, DC: Pew Research Center. Retrieved from http://pewhispanic.org/files/reports/115.pdf
Moore, J. L. (2005,July 30). Bilingual staff a big bonus for hospitals and Hispanic patients. The Morning News [Fayetteville,AR]. Retrieved from http://network.nshp.org/profiles/blogs/bilingual-staff-big-bonus-for
Pallant, J. (2007). SPSS survival manual (3rd ed.). Berkshire, England: Open University Press.
Passel, J. S., & Cohn, D.V. (2008). US population projections: 2005-2050 (Pew Hispanic Center Report).Washington, DC: Pew Research Center. Retrieved from http://pewhispanic.org/files/reports/85.pdf
Rivera-Goba M., & Wallen, G. R. (2008). Increasing the pipeline of Hispanic nurses:Are we making a difference? Hispanic Health Care International, 6(4), 170-171.
Seago, J.A., & Spetz, J. (2003). Admission policies and attrition rates in California community college nursing programs, No. 02-03. Berkeley, CA: California Policy Research Center and California Post-Secondary Education Committee.
Segura, D.A. (2002, April). Latinos and education in Ventura County. Paper presented at the conference, Immigration and the Changing Face of Rural California, Oxnard, CA. (ERIC Document Reproduction Service No. ED479362)
Smith, T. (2006, June 12). Spanish-speaking nurses becoming a medical necessity. Pdchmond-Times Dispatch. Retrieved from www.wtop.com/?nid=&sid=818454
Thacker, K. (2005).Academic-community partner ships: Opening the doors to a nursing career. Journal of Transcultural Nursing, 16(1), 57-63.
U.S. Census Bureau (2007). Minority population tops 100 million. Retrieved from www.census.gov/newsroom/releases/archives/population/cb07-70.html
U.S. Census Bureau (2008). Hispanic in the United States. Retrieved from http://www.census.gov/population/ www/socdemo/hispanic/files/Internet_Hispanic_in_US_2006.pdf
Velez-McEvoy, M. (2010). Faculty role in retaining Hispanic nursing students. Creative Nursing, 16(2), 80-83.
Vogt, R., & Taningco, M.T.V. (2008). Latina and Latino nurses:Why are there so few? Los Angeles: University of Southern California
Tomas Rivera Policy Institute. Retrieved from www.trpi.org/PDFs/Latino%20Nurses_Final.pdf
About the Authors Paul C. Bosch, EdD, is professor of biology, South Mountain Community College, Phoenix, Arizona. Sally A. Doshier, EdD, RN, CNE, is associate professor and assistant dean, Northern Arizona University School of Nursing, Flagstaff. Julie Gess-Newsome, PhD, is dean of the Graduate School of Education at
Willamette University, Salem, Oregon. For more information, contact Dr. Bosch at firstname.lastname@example.org.
Table. Logistic Regression Predicting the Likelihood of Persisting in the SENE Program Through the LPN Level. Independent Odds Variables B S.E. Wald df p Ratio Pre-program GPA 0.926 0.473 3.83 1 .050 * 2.53 Pre-program science credits 0.110 0.054 4.13 1 .042 * 1.12 Pre-program mathematics credits 0.148 0.079 3.48 1 0.062 1.16 Pre-program total credits 0.017 0.011 2.37 1 0.124 1.02 Pre-program developmental credits -0.061 0.043 1.95 1 0.163 0.94 Constant 1.95 1.45 1.81 1 0.209 0.15 * Shows significance