FORECASTING NURSING STUDENT SUCCESS AND FAILURE ON THE NCLEX-RN USING PREDICTOR TESTS by Lawrence A. Santiago Copyright 2013 A Dissertation Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Education University of Phoenix ABSTRACT A severe and worsening nursing shortage exists in the United States. Increasing numbers of new graduate nurses are necessary to meet this demand. To address the concerns of increased nursing demand, leaders of nursing schools must ensure larger numbers of nursing students graduate. Prior to practicing as registered nurses in the United States, graduates of nursing schools must pass the National Council Licensure Examination for Registered Nurses (NCLEX-RN). Various companies, such as the Assessment Technologies Institute (ATI) and Kaplan have created NCLEX-RN predictor tests that report candidates’ chances of passing the NCLEX-RN. ATI created a test called the RN Comprehensive Predictor and Kaplan created an NCLEX-RN predictor examination called the Kaplan Readiness Test. Students with less than optimal scores on the predictor can remediate to improve their knowledge of nursing, critical thinking, and test-taking skills. The intent for the ATI RN Comprehensive Predictor and the Kaplan Readiness Test is to predict the probability of success on the NCLEX-RN. The focus of the quantitative study was the ability of the aforementioned examinations to predict both success and failure on the NCLEX-RN. The overall TEAS score did not have a significant relationship with NCLEX-RN results. However, the ATI TEAS Math score was significantly higher (p = .005) for students who passed the NCLEX-RN. Data analysis showed no significant relationship between the Kaplan NCLEX-RN Readiness Test and NCLEX-RN results (p > .05). A significant relationship between the ATI RN Comprehensive Predictor and the NCLEX-RN scores existed in both the total sample (p = .001) and the BSN subsample (p = .001). In the MSN student subsample, all 37 students passed the NCLEX on the first attempt. iii DEDICATION This work is dedicated to my parents, Mildred and Larry. Thank you for helping me become who I am today. ACKNOWLEDGMENTS I would like to take this opportunity to thank all of the individuals who have supported me in this doctoral program. First of all, I would like to thank God for all of the blessings He has given me in this life. Through the past four years, my family’s love, support, and encouragement has sustained me through this journey. My wife Myla has been by my side through each course, each residency, and each stage of the dissertation process. Asawa, mahal na mahal kita! A special thanks to my son Joshua. You are the best son a father could have. I would also like to thank my dissertation committee. Dr. Phan, thank you for your continued inspiration throughout the dissertation process. Dr. Tidwell, thank you for sharing your expertise and experience with me. Finally, Dr. McAtee, thanks for being so reliable over the years. I would also like to thank the nursing department, including Dr. Susan Drummond, Dr. Geneva Oaks, Jeanette Russell, and Lynne Bosch. Thanks for helping with my study and answering my many emails. Finally, I would like to thank my colleagues for their steadfast support: Millicent De Jesus, Caroline Alfonso, Jeffrey Woods, and Karen Silva. Thanks for always being there for me! I cannot forget the support of my director, Peachy Hain. You have truly been my cheerleader through this process. ii TABLE OF CONTENTS Contents Page List of Tables ......................................................................................................... vi Chapter 1: Introduction ............................................................................................1 Problem Statement .......................................................................................2 Purpose of the Study ....................................................................................6 Significance of the Problem .........................................................................7 Significance of the Study .............................................................................7 Nature of the Study ......................................................................................9 Research Questions ....................................................................................11 Hypotheses .................................................................................................12 Theoretical Framework ..............................................................................13 Definition of Terms....................................................................................14 Assumptions...............................................................................................15 Scope, Limitations, and Delimitations .......................................................16 Summary ....................................................................................................17 Chapter 2: Review of the Literature.......................................................................19 Title Searches, Articles, Research Documents, and Journals ....................20 Bloom’s Taxonomy and Revision .............................................................20 Nursing Shortage .......................................................................................22 Societal Need .............................................................................................25 History of Nursing Licensure Examinations..............................................26 NCLEX-RN ...............................................................................................26 iii Quality of Nursing Programs .....................................................................27 Predictive Validity .....................................................................................28 High-Stakes Testing ...................................................................................29 Test Anxiety ...............................................................................................31 Computerized Testing ................................................................................31 Costs of NCLEX-RN Failure .....................................................................32 Historical NCLEX Predictor Research ......................................................32 Current NCLEX Predictor Research ..........................................................33 Beyond the NCLEX-RN ............................................................................40 Attitudinal Change .....................................................................................41 Gaps in the Literature.................................................................................41 Summary ....................................................................................................42 Chapter 3: Research Methods ................................................................................44 Research Method and Design Appropriateness .........................................44 Population ..................................................................................................45 Sampling ....................................................................................................45 Geographic Location ..................................................................................46 Data Collection ..........................................................................................46 Instrumentation ..........................................................................................47 Validity and Reliability ..............................................................................48 Data Analysis .............................................................................................49 Research variables .........................................................................49 Logistic regression analysis ...........................................................49 iv Summary ....................................................................................................49 Chapter 4: Results ..................................................................................................51 Research Questions ....................................................................................51 Data Analysis .............................................................................................52 Testing of the Hypotheses ..........................................................................54 Findings......................................................................................................57 Testing of the Hypotheses ..........................................................................58 Chapter Summary ......................................................................................60 Chapter 5: Conclusions and Recommendations ....................................................61 Purpose of the Study ..................................................................................61 Research Questions ....................................................................................62 Summary of Key Findings .........................................................................63 Implications................................................................................................64 Limitations .................................................................................................65 Suggestions for Future Research ...............................................................66 Significance................................................................................................67 Practitioner Recommendations ..................................................................68 Conclusion .................................................................................................69 References ..............................................................................................................71 Appendix A: Data Collection Tool ........................................................................83 Appendix B: Permission to Use Premises Form ....................................................84 Appendix C: Letter of Collaboration Among Institutions .....................................85 v LIST OF TABLES Table 1: Frequency Counts for Selected Demographic Variables (N = 251) ........53 Table 2: Descriptive Statistics for Selected Variables (N = 251) ..........................53 Table 3: Distribution of Predicted and Actual NCLEX Scores and Pass Rates for 2007 Predictor (n = 162) .................................................................................................54 Table 4: Distribution of Predicted and Actual NCLEX Scores and Pass Rates for 2010 Predictor (n = 57) ...................................................................................................55 Table 5: t Test Comparisons Based on NCLEX Outcome for Selected Variables 56 Table 6: Relationship Between Outcome of ATI Screening Test and NCLEX Outcome ................................................................................................................................58 vi Chapter 1 Introduction A severe and worsening nursing shortage exists in the United States (American Association of Colleges of Nursing [AACN], 2011) and more new-graduate nurses are necessary to meet this demand. To address the concerns of increased nursing demand, larger numbers of nursing students must graduate from nursing schools (Roa, Shipman, Hooten, & Carter, 2011). Prior to practicing as registered nurses in the United States, graduates of nursing schools must pass the National Council Licensure Examination for Registered Nurses (NCLEX-RN), which is the national registered nurse licensing examination (National Council of State Boards of Nursing [NCSBN], 2011). Various companies such as the Assessment Technologies Institute (ATI) and Kaplan have created NCLEX-RN predictor tests that report a candidate’s chances of passing the NCLEX-RN (ATI, 2012; Kaplan, n.d.). Leaders at ATI created a test called the RN Comprehensive Predictor (ATI, 2012) and leaders at Kaplan created an NCLEXRN predictor examination called the Kaplan Readiness Test (Kaplan, n.d.). Students with less than optimal scores on the predictor can remediate to improve their knowledge of nursing, critical thinking, and test-taking skills. The intent for the ATI RN Comprehensive Predictor and the Kaplan Readiness Test is to predict the probability of success on the NCLEX-RN. The focus of the current quantitative study is the ability of the aforementioned examinations to predict both success and failure on the NCLEX-RN. Leaders at ATI also created the Test of Essential Academic Skills (TEAS). Educators at various schools of nursing use this test as part of the admissions requirements. Students at these schools, including the school in the current study, must 1 achieve a certain score on the test, determined by each school, to be eligible for admission to their school of nursing (ATI, 2011). The purpose of the test is to ensure admitted students are more likely to be successful in nursing school and to pass the NCLEX-RN than those students who scored below the indicated passing mark on the TEAS. The focus of the current quantitative study was also on the ability of the TEAS to predict success and failure on the NCLEX-RN. Chapter 1 contains the problem and purpose of the quantitative, retrospective study. The study focus was on the ability of the TEAS, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test to forecast the outcome of NCLEX-RN results for the total population of registered nurse candidates from a school of nursing in California. Due to reasons unknown to the researcher, only 219 of the 251 graduates took the ATI RN Comprehensive Predictor, and only 100 of the 251 graduates took the Kaplan NCLEX-RN Readiness Test. Literature on NCLEX-RN predictor accuracy of nursing board examination failure is lacking. Therefore, determining the accuracy of these standardized tests to predict success and failure might assist in establishing the degree of usefulness for prospective registered nurses. Research such as this quantitative study could lead to earlier NCLEX-RN success and alleviation of the nursing shortage. When nursing students pass the NCLEX-RN earlier, they will be able to obtain employment and lessen the nursing shortage. Problem Statement A gap exists in the literature regarding the ability of the ATI RN Comprehensive Predictor or the Kaplan NCLEX-RN Readiness Test to forecast failure on the NCLEXRN accurately. Twenty percent of nursing schools in the United States will not allow 2 students to graduate and receive their nursing degree unless they pass the predictor test (National League for Nursing [NLN], 2012). A review of the EBSCOhost and ProQuest databases revealed no studies on the predictive accuracy of the Kaplan NCLEX-RN Readiness Test. EBSCOhost and ProQuest are use-for-fee research database providers that allow students and researchers to view articles from thousands of peer-reviewed journals (EBSCO Industries, 2012; ProQuest, 2012). The quantitative, retrospective study involved examining the ATI TEAS scores, the ATI RN Comprehensive Predictor scores, and the Kaplan Readiness Test scores for the total population of graduates from a nursing school in California. The purpose of the study was to determine the extent to which these standardized tests can accurately predict both passing and failure on the NCLEX-RN. Background of the Problem The nationwide nursing shortage continues to worsen. As baby boomers age, the need for nurses to care for this population will increase (AACN, 2011). The U.S. Census Bureau (2006) defined baby boomers as Americans born between 1946 and 1964. Baby boomers include Americans born in foreign countries who immigrated to the United States (U.S. Census Bureau, 2006). As of 2006, the U.S. Census Bureau reported that 77,980,296 baby boomers resided in the United States. The nursing shortage is even worse in California because the state has mandatory nurse-to-patient ratios (Aiken et al., 2010). These ratios include five patients per nurse in a medical/surgical acute setting and two patients per nurse in an intensive/critical care setting (California Department of Public Health, 2010). Nurses in California care for one less patient on average than other states (California Department of Public Health, 2010). Because of this mandatory ratio, it 3 is more difficult to staff hospitals in California with sufficient numbers of nurses than in other states (California Department of Public Health, 2010). California is the only state in the United States with mandatory nurse ratios (American Nurses Association, 2011). Insufficient staffing decreases patient safety, increases work-related stress and injuries for nurses, and causes nurses to leave the profession (AACN, 2011). Large numbers of registered nurses will retire over the next 25 years, contributing to a projected shortage of 265,000 nurses by 2025 (Buerhaus, Staiger, & Auerbach, 2009). The demand for registered nurses will continue to increase, which will create the largest shortage of nurses since the 1960s (Buerhaus et al., 2009). The average life expectancy in the United States is 77.8 years, and the United States will experience a substantial increase in its elderly population over the next 30 years, from now until 2033 (Orman & Guarneri, 2009). By 2030, all baby boomers will be over 65 years old (Buerhaus et al., 2009). Increasing the number of prospective registered nurses who can successfully pass the NCLEX-RN will assist in alleviating the nursing shortage. The NCLEX-RN passing standard has increased in recent years, which could potentially place more graduate students at risk for failure (Carrick, 2011). Although nursing school leaders in California should always ensure their graduating students are ready to pass the NCLEX-RN and join the nursing workforce, the worsening shortage further indicates the importance of first-time NCLEX-RN success (AACN, 2011). The purpose of an NCLEX-RN predictor test is to identify readiness for students to sit for the exam and areas of weakness to assist with remediation. A significant number of nursing schools in the United States use NCLEX-RN predictor tests to determine whether students can receive approval to take the NCLEX4 RN (Spurlock & Hunt, 2008). The primarily reason for designing exit examinations is to predict readiness for the NCLEX-RN. Many schools use the exit exam as a high-stakes test. Nichols and Berliner (2008) defined high-stakes testing as “the practice of attaching important consequences to high stakes test scores” (p. 41). High-stakes nursing tests require students to pass before they can obtain approval to take the NCLEX-RN (Harding, 2010). If nursing instructors are failing students based on the results of a highstakes test, then the test must accurately predict failure on the NCLEX-RN (Giddens, 2009; Spurlock & Hanks, 2004). A limited number of NCLEX-RN predictor studies have shown prediction of failure rates to be less than 50% accurate (Benefiel, 2011; Spurlock & Hanks, 2004). Potentially, some students held back would pass the NCLEXRN with the first attempt. Students capable of passing the NCLEX-RN should be able to sit for the test, pass the national board examination, and begin their careers as registered nurses (Spurlock & Hanks, 2004). Nursing leaders should identify variables that affect NCLEX-RN pass rates (Ukpabi, 2008). An NCLEX-RN predictor test can help students to determine the knowledge and concepts that were deficient in their ability to pass the NCLEX-RN (Ukpabi, 2008). For the ATI RN Comprehensive Predictor, nursing instructors have the option to provide students with a breakdown of the specific topics for which the students require remediation (ATI, 2011). Remediation should be the primary purpose of a predictor examination, not being a singular measure of progression in a nursing program (Ukpabi, 2008). 5 Purpose of the Study The purpose of the quantitative study was to determine if the TEAS, the ATI RN Comprehensive Predictor, and the Kaplan NCLEX-RN Readiness Test could accurately predict both success and failure on the NCLEX-RN at one nursing program in Southern California. The quantitative study took place at one nursing program in the Southern California region. This program offers both a baccalaureate degree program (bachelor of science in nursing, BSN) and an accelerated entry level master’s degree program (master of science in nursing, MSN). If these tests can accurately predict student outcomes on the NCLEX-RN, they will be valuable tools for selecting students for remediation (Uyehara, Magnussen, Itano, & Zhang, 2007). Faculty can identify students at risk for failing the NCLEX-RN and develop an individualized tutoring and mentoring program for them (Uyehara et al., 2007). The interventions are one method of increasing the nursing school’s first-time NCLEX-RN pass rate (Uyehara et al., 2007). The quantitative study involved a statistical analysis that measured the ability of the students’ ATI TEAS results, their ATI RN Comprehensive Predictor results, and their Kaplan Readiness Test results to predict both success and failure on the NCLEX-RN. The school required students to take all three examinations. Due to reasons unknown to the researcher, only 219 of the 251 graduates took the ATI RN Comprehensive Predictor. In addition, only 100 of the 251 graduates took the Kaplan NCLEX-RN Readiness Test. Because the purpose of the study was to determine the extent to which the predictor tests can individually forecast success on the NCLEX-RN, a quantitative, retrospective study design was appropriate. Because the study involved comparing the NCLEX-RN result to the NCLEX-RN predictor result, a retrospective design was necessary. 6 Significance of the Problem Although employment of registered nurses is increasing rapidly, the growth might not be enough to meet the increasing demand (Bureau of Labor Statistics, 2012). In 2011, leaders at the AACN reported 135,000 registered nurse vacancies. Of all registered nurse positions in the United States, 8.1% were vacant. This is an alarming statistic because the nursing shortage persists in a time of economic recession (AACN, 2011). As nurses retire, the number of newer nurses entering the profession is not sufficient to meet the demand (Buerhaus et al., 2009). Nursing schools must ensure they graduate every student who is able to safety practice as a registered nurse. The number of elderly will dramatically increase because the baby boomer generation is aging (Buerhaus et al., 2009). More nurses will be necessary to care for the older population (Bureau of Labor Statistics, 2009). If nursing school prorgression policies prevent a large number of nursing students capable of passing the NCLEX-RN from graduating, this group of prospective nurses is a delayed resource because these individuals will not be able to assist in alleviating the nursing shortage in a timely manner. More research on NCLEX-RN predictor failure accuracy is necessary because of the insufficient number of studies on this topic. In addition, a review of the literature revealed published studies on the Kaplan NCLEX-RN Readiness Test, which opened an opportunity for further research to validate this test. Significance of the Study The quantitative study involved comparing student results of the TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test to their result on the NCLEX-RN. The findings might assist nursing leaders with developing policies for 7 remediation. Although other researchers have measured the accuracy of exit exams to predict passing the NCLEX-RN, few researchers have sought to measure the accuracy of these exams to predict failure (Abbott, 2008; Adamson & Britt, 2009; Landry, Davis, Alameida, Prive, & Renwanz-Boyle, 2010). Prediction to pass relates to the result of the predictor test. If the predictor test result is 60%, the prediction is that the student will have a 60% chance of passing the NCLEX-RN and a 40% chance of failing the NCLEXRN (ATI, 2011; Kaplan, n.d.). A lack of literature exists on the topic of NCLEX-RN predictor failure. Also, a review of literature did not reveal any studies measuring the predictive accuracy of the Kaplan NCLEX-RN Readiness Test. The predictor tests provide scores after students complete the examinations. A table is provided for faculty that attaches the scores to percentages that indicate the chances of passing the NCLEX-RN if the students were to take it that day. If a highstakes test cannot accurately predict failure on the NCLEX-RN, passing such a test should not be a requirement for graduation from the nursing program (Harding, Rateau, & Heise, 2011). The primary purpose of the proposed study is to examine the ability of the ATI RN Precdictor Test and the Kaplan NCLEX-RN Readiness Test to determine failure on the NCLEX, as well as passing. Although it is important for educators to determine if students are ready for the NCLEX-RN examination, it is critical to choose a valid research design to confirm these predictions. This research design includes the ability of the research instruments to predict NCLEX-RN success as well as failure. Through careful scrutiny of research design, researchers can implement an accurate study (Burns & Grove, 2009). Using the appropriate research design will further validate the findings of the study and increase the 8 ability of the results to apply to other academic settings. The next section will provide a brief overview of the chosen research design for the study. Nature of the Study The quantitative, retrospective study involved examining the predictive accuracy of the ATI TEAS test scores, the ATI RN Comprehensive Predictor scores, and the Kaplan Readiness Test scores to forecast both passing and failing the NCLEX-RN at one school of nursing in Southern California. Retrospective studies are nonexperimental, and no manipulation of the subjects occurs (Burns & Grove, 2009). In a retrospective study, a researcher attempts to link an event to another event that occurred prior to the first one (Burns & Grove, 2009). The current study involved linking NCLEX-RN results to NCLEX-RN predictor tests taken before the students’ graduation. A multiple regression analysis served to determine if a relationship exists between the test scores and the NCLEX-RN outcomes. The software used to perform the analysis was Statistical Package for the Social Sciences (SPSS) 21. A retrospective, quantitative design was the ideal method to compare standardized test data to the NCLEX-RN. Retrospective studies involve measuring current circumstances that might have a relationship with past circumstances (Polit & Beck, 2008). The current study involved analyzing NCLEX-RN scores with ATI TEAS scores, ATI RN Comprehensive Predictor scores, and Kaplan Readiness Test scores to determine the extent of their relationship. Qualitative research designs were not appropriate for the study. A Southern California baccalaureate degree nursing program was the site for the retrospective study. Data collection consisted of examining archived data. The data 9 included ATI TEAS scores, ATI RN Comprehensive Predictor scores, and Kaplan NCLEX-RN Readiness Test scores. Prospective nursing students take the TEAS before admission to nursing school and take the exit tests during the last semester of the nursing program. The researcher collected the data on the campus. He received archived student test results and he recorded the results onto the Data Collection Tool (see Appendix A). Archived data analysis is cost-effective, accessible, and available (Creswell, 2005). Data collection proceeded in an organized, systematic manner that maximized the generalization of the findings (LoBiondo-Wood & Haber, 2008). The data gathered were from one nursing program in Southern California after receiving approval from the school’s institutional review board (IRB) as well as the IRB from University of Phoenix. All applicants to the nursing school must take the TEAS to receive consideration for admission to the nursing school. All students take both the ATI RN Comprehensive Predictor and the Kaplan Readiness Test prior to graduation. Due to reasons unknown to the researcher, only 219 of the 251 graduates took the ATI RN Comprehensive Predictor. In addition, only 100 of the 251 graduates took the Kaplan NCLEX-RN Readiness Test. The study involved measuring only the first attempt of these predictor examinations. Students who fail one or both of these examinations must enter a remediation course. After the course, the student will retake the test. Students must pass this examination to graduate. The study involved comparing results data from five student cohorts (2008-2011) to the students’ pass–fail data on the NCLEX-RN. Student consent was not a requirement because the study involved analyzing existing data sets. No personal contact with any nursing students or graduates took place. Data collection occurred during a review of retrospective files at the school of nursing. The 10 use of codes rather than student names served to protect students’ identity. The researcher consistently matched all scores with the correct code number. Research Questions The research project included two questions. The questions were appropriate to the study and highlighted the dependent and independent variables. The study involved determining the extent to which three standardized nursing exams could predict both success and failure on the NCLEX-RN at one nursing school in Southern California. 1. How are ATI RN Comprehensive Predictor and Kaplan Readiness Test examination scores (independent variable) related to results on the NCLEX-RN (dependent variable)? Recent studies have shown the ability of NCLEX-RN predictor examinations to forecast success on the NCLEX-RN accurately (Harding et al., 2011; McGahee, 2010; Ukpabi, 2008). The predictor tests examined included the Health Education Systems, Inc. (HESI) Exit Exam, the ATI RN Comprehensive Predictor test, and the Educational Resources Inc. (ERI) test. The ERI test was an NCLEX-RN predictor test that no longer exists (Bondmass, Moonie, & Kowalski, 2008). Lacking in the literature were studies on the ability of NCLEX-RN predictor exams to forecast failure on the NCLEX-RN accurately. The current study involved analyzing both success and failure predictions. A multiple regression analysis assisted with the study (LoBiondo-Wood & Haber, 2008). In some schools of nursing, including the school in the current study, students must pass an NCLEX-RN predictor examination prior to graduation and prior to receiving permission to sit for the nursing board examination. If the predictor test cannot accurately forecast failure on the NCLEX-RN, the progression policy might not be fair to the students. The 11 focus of the second research question was NCLEX-RN failure and NCLEX-RN predictor accuracy. 2. How are nursing school admissions test scores (independent variable) related to results on the NCLEX-RN (dependent variable)? Researchers recently examined the relationship between high scores on standardized nursing school admissions examinations and positive results on the NCLEX-RN, but a gap exists in the literature regarding the relationship between low scores on the admissions exams and negative results on the NCLEX-RN. The two research questions served as guides during data collection. The focus of the study was the results of the ATI TEAS test, the ATI RN Comprehensive Predictor test, and the Kaplan Readiness Test compared with the NCLEX-RN results. A statistical analysis assisted in determining the answers to the study questions. Hypotheses Parallel to the research questions were the following null and alternative hypotheses: H10: The ATI RN Comprehensive Predictor test score is not significantly related to results on the NCLEX-RN. H1a: The ATI RN Comprehensive Predictor test score is significantly related to results on the NCLEX-RN. H20: The Kaplan Readiness Test score is not significantly related to results on the NCLEX-RN. H2a: The Kaplan Readiness Test score is significantly related to results on the NCLEX-RN. 12 H30: The ATI TEAS nursing school admissions test score is not significantly related to results on the NCLEX-RN. H3a: The ATI TEAS nursing school admissions test score is significantly related to results on the NCLEX-RN. The study determined the extent to which the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test could foretell both success and failure on the NCLEX-RN. The preceding hypotheses were appropriate for the quantitative, retrospective nature of the study. Theoretical Framework For both nursing schools and the NCSBN, one of the primary educational goals is for students to develop didactic and clinical competency. Bloom’s taxonomy has historically served as a guide for faculty to determine what is the appropriate level at which a nursing student must function. Bloom, Engelhart, Furst, Hill, and Krathwohl (1956) first developed the taxonomy to classify learning outcomes. Nursing faculty typically focus on the cognitive domain of Bloom’s taxonomy because this framework is the basis of the formulation for the NCLEX-RN test plan (NCSBN, 2011). NCLEX-RN predictor test plans must be congruent with Bloom’s taxonomy or they risk a decrease in compatibility with the NCLEX-RN. Although the original 1956 publication of Bloom’s taxonomy was not research based, the 1999 revision was evidence based. Research conducted after the 2001 publication further validated the concepts. Nasstrom (2009) conducted a study in which the findings indicated Bloom’s revised taxonomy is an acceptable tool for understanding mathematical concepts. Bumen (2007) conducted a study that involved comparing the 13 initial 1956 taxonomy to the revision. Bumen found the revised taxonomy was superior to the initial taxonomy when applied to lesson planning skills. Bloom’s taxonomy is an excellent guide for teaching students about the nursing process (Duan, 2006). This is particularly true for the taxonomy’s cognitive domain. Nursing faculty often use these concepts to develop nursing program and course objectives. The levels of the taxonomy such as knowledge and comprehension are appropriate for beginning nursing students, and instructors gradually integrate the higher concepts such as application and analysis into instruction (Duan, 2006). For test questions, instructors should be aware to which cognitive level the question relates. Test developers must weave the elements of the taxonomy into NCLEX-RN predictor test plans, as the pattern of the actual NCLEX-RN aligns with Bloom’s taxonomy. Definition of Terms Some of the terminology in the study was specific to undergraduate nursing education. Such terms are not common knowledge to society. This section includes a discussion of the terms. Bachelor of science in nursing (BSN) program: A BSN program is a 4-year higher education program that prepares students to become licensed registered nurses (Chitty & Black, 2011). Graduates are eligible to sit for the NCLEX-RN. The sample in the current quantitative study included graduates from a BSN program. ATI RN Comprehensive Predictor: Students typically take this test toward the end of the nursing program (ATI, 2011). This examination is computerized and has 180 items. Test developers at ATI developed this examination. The quantitative study included the ATI RN Comprehensive Predictor as a predictor variable. 14 National Council Licensure Examination for Registered Nurses (NCLEXRN): The NCLEX-RN is a computer adaptive test (NCSBN, 2011). The computer-based examination assesses for minimum competency levels for candidates seeking to become registered nurses in the United States. Test developers at the NCSBN strive to create a test that will ensure safe nursing practice. California Board of Registered Nursing (BRN): The California Board of Registered Nursing (BRN, 2011) is the government organization that approves and monitors nursing programs in the state of California. Test of Essential Academic Skills (TEAS): The TEAS is a standardized examination typically given either prior to admission to nursing school or at the beginning of a nursing program (ATI, 2011). The examination has 170 questions, including 55 English and language use items, 30 science items, 45 mathematics items, and 40 reading items. Assumptions Assumptions are present in all research studies (Neuman, 2003). Identifying the assumptions in a study will assist the researcher to better comprehend the various concepts in the study (Neuman, 2003). Several assumptions existed in the proposed quantitative study. The first assumption was that the data collection would be accurate. The rationale for this assumption included the researcher following protocol for retrospective data collection. The second assumption was the execution of the data analysis would be correct. The rationale for this assumption included the researcher performing an accurate data analysis. The third assumption was that the administration of the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness 15 Test would be appropriate to the nursing students. The fourth assumption was that the nursing students in the study would follow the same or a similar curriculum during nursing school. Assumptions about the research design also existed. One assumption was no unknown or unacknowledged variables influenced the study. An example of this was the assumption that all students from the nursing program studied had fully met the admission criteria required by the institution. Another assumption was the admission criteria from the nursing program were similar to other programs in the state of California. Scope, Limitations, and Delimitations The scope of the study included graduate nurses from a prelicensure nursing program in Southern California. The population included program graduates from the past 10 cohorts. These include seven cohorts from the traditional baccalaureate program and three cohorts from the entry level master’s program. The data collected were the archived results of the students’ TEAS test, ATI RN Comprehensive Predictor, Kaplan Readiness Test, and NCLEX-RN. Data collected were from the graduating cohorts between 2006 and 2012. Although NCLEX-RN scores exist, the study involved collecting only pass–fail data. The dependent variable was the outcome of the NCLEXRN. The sample was a limitation because the nursing students all came from the same school. The program was relatively new as it began in 2006; therefore, the data gathered were from 10 graduating cohorts of students. Also, the sample was a convenience sample. A convenience sample is a group of individuals chosen at the discretion of the 16 researcher (Burns & Grove, 2009) and does not involve an attempt to capture an accurate representation of the population. Selecting the sample did not involve randomization, which is the assignment of study participants by chance, rather than at the discretion of the researcher (Burns & Grove, 2009). The scores of students with previous health care careers remained part of the sample. The data collected were for the total population of graduated cohorts. Delimitations are the defined boundaries of a study (Burns & Grove, 2009). The data collection tool used in the study was one such delimitation. The geographical location of the study was also a delimitation (Burns & Grove, 2009). The study took place at one baccalaureate nursing school in Southern California. Summary The nursing shortage in the United States is severe and worsening. By 2020, the national shortage could increase to more than 1 million full-time equivalent registered nurses (Health Resources and Services Administration, 2010). NCLEX-RN predictor examinations forecast both success and failure on the board examination. Some nursing schools mandate that students pass predictor examinations to progress or graduate. If the predictor test is not accurate, then it should be a determination of competence for nursing students. The purpose of the quantitative retrospective study was to compare the results of NCLEX-RN predictor tests to actual NCLEX-RN results on 10 cohorts of undergraduate nursing students at a Southern California university. The dependent variable was the NCLEX-RN results. The independent variables were the predictor test results. 17 Chapter 2 will include a literature review on NCLEX predictor research. Also included will be discussion on the study’s chosen theoretical framework, Bloom’s taxonomy. The chapter also contains a discussion on the history of nursing shortages in the United States, as well as the current nursing shortage. 18 Chapter 2 Review of the Literature The purpose of the retrospective, quantitative study was to determine the extent to which the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test could accurately forecast both success and failure on the NCLEX-RN. The proposed study included one baccalaureate degree and entry-level master’s degree nursing program in Southern California. A program requirement is that students achieve a set passing score on the TEAS prior to admission to the program. Students must also pass both the ATI RN Comprehensive Predictor and the Kaplan Readiness Test to graduate and take the NCLEX-RN. A high first-time student pass rate on the NCLEX-RN is crucial to validate a successful nursing program (BRN, 2011). The BRN (2011) mandates that schools maintain at least a 75% first-time pass rate. Approximately one third of nursing schools in the United States use NCLEX-RN predictor tests to determine whether students can graduate from their programs (NLN, 2012). Exit examinations, including the ATI RN Comprehensive Predictor and the Kaplan Readiness Test, primarily predict readiness for the NCLEX-RN and do not predict failure on the test (Spurlock & Hunt, 2008). Many schools use the exit exam as a high-stakes test. High-stakes tests such as NCLEX-RN predictor examinations require students to pass before they can receive approval to take the NCLEX-RN (Spurlock & Hunt, 2008). The focus of this chapter is a review of peer-reviewed publications in which researchers have addressed the nursing shortage and NCLEX predictor examinations. The literature review consists of six sections. The first section includes a review of 19 nursing shortages throughout the history of the United States and the societal need for nurses. The second section includes a discussion on the history of nursing licensure examinations in the United States. The third section contains a description of the concept of predictive validity and recent studies that include a measure of the predictive validity of an examination. The fourth section will contain a review of historic and recent research from peer-reviewed publications pertaining to NCLEX predictor tests. The fifth section will include a discussion of gaps in the literature. The sixth section will consist of a discussion on leadership. Title Searches, Articles, Research Documents, and Journals An electronic literature search using EBSCOhost provided articles published after 2007. The search included the following keywords: NCLEX, HESI, ATI, Kaplan, nursing, student, shortage, and Bloom’s taxonomy. The search yielded 70 referenced articles and documents. Of the total references in the study, the dates for 85% were after 2007. Bloom’s Taxonomy and Revision One of the chief directives of the NCSBN is for nursing students to develop didactic and clinical competency (NCSBN, 2011). Since the 1950s, Bloom’s taxonomy has functioned as a tool for nursing faculty to establish the appropriate cognitive level to which a student must perform. Bloom, Englehart, Furst, Hill, and Krathwohl (1956) initially developed the taxonomy to categorize learning outcomes. Nursing faculty typically focus on the cognitive domain of Bloom’s taxonomy because the NCLEX-RN test plan is formulated on this framework (NCSBN, 2011). To maximize the predictive 20 ability of an NCLEX-RN predictor test, Bloom’s taxonomy is necessary to assist in formulating its test plan. In 1999, Anderson and Krathwohl (2001) revised the taxonomy to include the following levels: remember, understand, apply, analyze, evaluate, and create. In the cognitive domain, three of the six levels changed. The levels are no longer hierarchical in nature (Ari, 2011). Nursing faculty expect that students will gradually master each level, beginning with remembering (Duan, 2006). Remembering involves the ability to recall knowledge from memory, such as with anatomical sites and names of medications. The next level is understanding, wherein the learner can interpret meaning from information. If a patient has chest pain, the student with understanding can state that the patient possibly had a heart attack. Applying is the next level. A nursing student who masters this level can successfully write a care plan. If a patient complains of constipation, the student can assess for signs and symptoms of constipation, diagnose the patient, set a goal to restore normal bowel function, and implement interventions to relieve constipation. Analyzing is the next level. A student is able to separate a concept into individual parts and learn how the parts relate to each other. A student is also able to determine the overall purpose of these parts. Evaluation is the next level, wherein a student is able to make decisions and determine if the intervention is effective or ineffective. The final level is creating. A student is able to put the pieces together and create a logical structure, discuss a patient’s plan with care with the patient’s family, perform discharge planning, and produce discharge teaching instructions for the patient. The NCSBN (2011) reported, “NCLEX-RN and NCLEX-PN examinations may include items written at various cognitive levels. The majority of items are written at the 21 application or higher levels of cognitive ability” (p. 11). This section includes a discussion on both the 1956 work by Bloom and the 1999 update. Because Bloom’s taxonomy is the basis of the NCLEX-RN test plan, nurse educators should also consider lining up Bloom’s instructional theory with their own curriculum (Duan, 2006). NCLEX-RN predictor test developers must also base their test plan and test items on Bloom’s taxonomy. Writing instructional outcomes using the keywords from Bloom’s taxonomy, such as analyze and evaluate, will help students understand the level of competency at which they must function to meet the objectives of the particular course (Boland, 2009). This practice will create a congruency between the nursing school’s curriculum and the NCLEX-RN test plan. Using Bloom’s taxonomy will also increase similarity between NCLEX-RN predictor tests and the actual NCLEX-RN. In addition to the six cognitive levels of Bloom’s taxonomy, the 1999 update includes another four kinds of knowledge. The additions were factual, conceptual, procedural, and metacognitive elements (Anderson & Krathwohl, 2001). The elements allow instructors to engage students in activities that, while maintaining focus on the cognitive domain, include affective and psychomotor aspects. Using this array of elements in instruction will prepare students for the NCLEX-RN and for becoming safe, effective, and caring nurses. In the clinical area, instructors must determine at what cognitive level the student is functioning. Senior nursing students should demonstrate evidence of thinking at the analysis, application, and evaluation levels (Duan, 2006). Nursing Shortage Several nursing shortages have occurred since the post-World War II era (1945present). In 1947, the Hill-Burton Act became law (Buerhaus, 1987). The law allowed a 22 large increase in the number of hospitals in the United States. The Hill-Burton Act greatly increased the demand for nurses (Buerhaus, 1987) and significantly influenced the modern health care system, especially regarding in-hospital patient care. The U.S. government has taken various steps to alleviate the nursing shortages throughout history. In 1964, President Johnson signed the Nurse Training Act into law (Yett, 1975). The law enabled nurses to receive hospital-managed loans by which nursing students would receive funding for school. The nurses would pay back the loans interest-free over 10 years after completing nursing school, as long as they worked fulltime as a registered nurse. The intent for the law was to entice potential nursing students to enter school who otherwise would not be able to afford nursing school. Buerhaus (1991) analyzed hospital wage increases during times of nursing shortage compared to times of adequate staffing. The intent for the increases was to entice licensed nurses who were not working. Hospitals offered wage increases during times of shortage. Hospitals that could not afford such wage increases hired lower cost employees, such as licensed practical nurses and nurses’ aides. The current nursing shortage began in 2000 (Chitty & Black, 2011). Several factors make the current shortage unique compared to previous crises. There are 77 million baby boomers compared to 44 million people in the next generation following, known commonly as Generation X (Americans born between 1965 and 1985; AACN, 2011). Increasing health care resources will be necessary for baby boomers as they age. Typically, as a person ages, they will require more health care needs, including hospitalizations (Buerhaus et al., 2009). In 2006, there were 37 million people over 65 in 23 the United States (Federal Interagency Forum on Aging Related Statistics, 2009). This number will more than double by the year 2030. Another unique issue of the current nursing shortage is that the number of nursing schools in the United States is inadequate to supply the health care system with the number of nurses needed (Buerhaus et al., 2009). For the 2008-2009 school year, 36,511 qualified students applied for admission to nursing programs in California, and the programs did not accept 61.7% of the applicants (BRN, 2011). The statistics clearly indicated the number of applicants far exceeded the seats available. The situation is disturbing, considering the high demand for nurses in health care (Buerhaus et al., 2009). Increasing numbers of nurses are leaving the profession compared to the past. The stress and intensity of the work causes many nurses to leave their jobs (Chitty & Black, 2011). Nurses also experience a number of physical ailments rendering them unable to work, including back, knee, and wrist injuries (Chitty & Black, 2011). Technological advances have also caused some nurses to leave their jobs because they do not feel comfortable adjusting to changes such as computerized documentation (Chitty & Black, 2011). Technologies include computerized documentation, electronic medication storage, and barcode medication administration. Rapid advances are challenging for some older nurses (Chitty & Black, 2011). Several solutions to alleviate the current nursing shortage exist. One such solution was the implementation of nurse–patient ratios to reduce work-related stress and improve patient care (Tevington, 2011). The nurse–patient ratio had the opposite effect to some degree, as a decreased nurse–patient ratio requires more nurses for each hospital unit (Tevington, 2011). Another solution for nursing shortages has been for hospitals to 24 recruit nurses from foreign countries, although the cost of recruiting foreign nurses is high and the result is not always positive (Ross, Polsky, & Sochalski, 2005). Societal Need The largest contributing factor to the nursing shortage over the next 20 years will be the baby boomer population in the United States (AACN, 2011). As this population ages, it will become a large obligation for the American health care system that will require a greater number of registered nurses in the health care workforce. Schools of nursing must ensure they will graduate the largest number of qualified students possible (Giddens, 2009). In addition to aging patients, registered nurses are also aging. In 2007, the average age of registered nurses in the United States was 43.7 years old (AACN, 2011). By the year 2025, many nurses from the baby boomer generation will retire. These nurses need replacing by greater numbers of nurses from succeeding generations. By 2025, the United States might experience a nursing shortage of 260,000 nurses, which is double the number of nurses needed in 2007 (AACN, 2011). Adequate nursing staffing is essential to patient safety. Kane, Shamliyan, Mueller, Duval, and Wilt (2007) performed a meta-analysis of 28 studies that focused on the relationship between registered nurse staffing and patient deaths and complications. Kane et al. found when registered nurses are present on patient units, length of stay on intensive care units and postsurgical units was significantly less. To address the concerns of increased nursing demand and an aging workforce, nursing schools must have larger numbers of nursing students graduate (Roa et al., 2011). Nursing programs should not have attrition policies that fail potentially capable students. 25 Through increased support, counseling, guidance, and remediation, nursing faculty can guide greater numbers of nursing students to graduate nursing school and pass the NCLEX-RN the first time (Uyehara et al., 2007). History of Nursing Licensure Examinations Before 1944, every state had its own nursing licensure examination (MatassarinJacobs, 1989). Beginning in 1944, nursing graduates sat for the State Board Test Pool Examination developed by the NLN. During this time, each state maintained its own pass rate, but this become problematic when nurses moved from one state to another (Birnbach, 1982). By 1978, the NCSBN began to oversee the State Board Test Pool Examination, redesigned the test to include criterion referencing, and set a standardized passing score at 1600. The NCSBN renamed the examination to the NCLEX-RN. In 1988, the examination changed to a pass–fail format (Matassarin-Jacobs, 1989). NCLEX-RN To work as a registered nurse in the United States, each nursing school graduate must pass the NCLEX-RN (NCSBN, 2011). The intent for the NCLEX-RN was to ensure nurses in the United States can practice safe patient care (NCSBN, 2011). The examination initially began as a paper and pencil examination. Since 1994, the NCLEXRN has been a computerized adaptive test, which is a test that adapts to the ability level of the examinee (NCSBN, 2011). Every 3 years, the NCSBN reexamines the rigor of the NCLEX-RN. Since 1994, the NCSBN has modified the NCLEX-RN four times, based on a practice analysis survey. Aucoin and Treas (2005) studied the practice analysis survey, as the results of the survey lead to modifications on the NCLEX-RN Test Plan. The survey has a return 26 rate of approximately 30%. Aucoin and Treas found the practice analysis is a useful tool for nurse educators, as it can make clear the necessary changes to nursing education. The authors recommend that nursing educators adjust curricula to include more information on topics such as advanced cardiac life support and electrocardiograms (Aucoin & Treas, 2005). The NCLEX-RN passing standard has increased in recent years. The NCSBN assesses passing candidates’ ability to perform safely in health care settings and uses surveys to educators and employers to determine the passing standard (NCSBN, 2011), which could potentially place more graduate students at risk for failure. Carrick (2011) used two theories to identify interventions for the NCLEX-RN candidates: systems theory and the student’s approach to learning theory. Although nursing instructors might feel the need to increase the difficulty of questions on their class examinations, it is important to first assess the students’ learning needs. Although assessment testing is common, research to improve NCLEX-RN prediction is inadequate. Instructors must acknowledge that personal or situational issues might influence students’ test-taking abilities (Spurlock & Hunt, 2008). In addition to assessment testing, instructors must use a variety of methods to assist students at risk for failure. Quality of Nursing Programs State boards of nursing often measure the quality of a nursing program by its NCLEX-RN pass rate. The National League for Nursing Accrediting Commission and the Commission on Collegiate Nursing Education cite the NCLEX-RN as primary evidence for achieving program outcomes (Morrison, 2005). Some have questioned the wisdom of only counting the NCLEX-RN first-time pass rate in the integrity of the 27 nursing program. This practice has led to the implementation of strict progression policies in some nursing programs (Giddens, 2009), including the requirement of passing standardized examination that predict success on the NCLEX-RN prior to progressing to the next semester of classes. Giddens (2009) noted this practice is potentially unfair and could be an unethical policy. She recommended boards of nursing and accrediting bodies consider the possibility of including the second NCLEX-RN attempt when judging the integrity of a nursing program. Morrison (2005) noted the developers of the NCLEX-RN did not design the test as an evaluation tool for the quality of a nursing program or its curriculum. Predictive Validity Predictive validity is the value of a test that determines a person’s achievement on a future measure (Polit & Beck, 2008). The study involved measuring the predictive validity of the ATI TEAS, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test. The proposed study will focus on the aforementioned examinations’ ability to predict success and failure on the NCLEX-RN. Ali, Zaman, and Alamgir (2011) measured the predictive validity of admissions tests for dental schools in Pakistan and found significance between the admissions test and the academic achievement of the dental students. Students with higher scores on the admissions test had better grades and a lower attrition rate than students with lower scores. Ali et al. advocated for the inclusion of additional variables for admissions consideration, including live interviews. Additional admissions criteria would increase the likelihood of admitting the best students for each program (Ali et al., 2011). 28 Pishghadam and Khosropanah (2011) measured the predictive validity of the English Language Teacher Competency Test. The test assessed examinees’ ability to read, write, and speak the English language. English teachers in Iran took the test. The test scores were compared to the final examination scores of each teacher’s students. The statistical analysis revealed a significant relationship between the English Language Teacher Competency Test and the students’ final examination scores. High-Stakes Testing Tests have been a vital part of human society for thousands of years (Madaus & Russell, 2009). A high-stakes test is an examination in which the outcome holds great significance for the examinee. In the United States, 12% of nursing schools, including the school in the current study, require students to pass an NCLEX-RN predictor test before allowing them to take the NCLEX-RN (NLN, 2012; Spurlock & Hunt, 2008). Twenty percent of schools have policies that will not allow students to graduate and receive their nursing degree unless they pass the predictor test (NLN, 2012). About one third of nursing schools in the United States require registered nursing students to achieve a minimum score on a predictor test to progress in the nursing program (NLN, 2012). The NCLEX-RN predictor examinations are not the only high-stakes tests. Exams such as the SAT and the Graduate Record Examination are also high-stakes tests (Duncan & Stevens, 2011). The SAT is a part of the admission requirements for most undergraduate universities in the United States, and the Graduate Record Examination is an admission requirement for most U.S. graduate schools (Duncan & Stevens, 2011). Duncan and Stevens (2011) surveyed public school teachers in Texas about their attitudes toward high-stakes testing. Of the respondents, 92% believed that high-stakes tests 29 forced them to prioritize tested subjects over nontested material. Seventy-five percent of the respondents believed the high-stakes tests did not measure instructional effectiveness accurately (Duncan & Stevens, 2011). Duncan and Stevens (2011) believed standardized testing should continue but teachers should not have to limit the curriculum to material that appears on the standardized examination. In the United States, tests often become a remedy to solve educational as well as societal dilemmas (Madaus & Russell, 2009). High-stakes testing promotes American values, including competitiveness, accountability, and standardization. However, highstakes tests often compel teachers to convert lectures into test preparation sessions (Madaus & Russell, 2009). Creators of predictor tests do not take random circumstances into account when determining test design and scoring. Such circumstances include student illness, or movement and noises occurring within the testing environment. In February 2012, the NLN Board of Governors published The Fair Testing Imperative in Nursing Education on their website. Leaders at the NLN became aware of the pressure schools of nursing feel to generate high first-time NCLEX-RN pass rates (NLN, 2012). An association exists between a high first-time pass rate and the need to ensure safe nursing practice for the public. The NLN’s primary concern is the use of standardized predictor tests to prevent nursing students from taking the NCLEX-RN. Because of this concern, leaders at the NLN developed national fair testing guidelines to ensure educators and administrators follow evidence-based practices for progression and graduation. The first guideline addresses the ethics of high stakes testing. The NLN leaders implore nurse educators to use predictor tests that are well-researched and validated 30 across each nursing course. They must also ensure the tests lack age, gender, ethnic, religious, sexual, or regional bias (NLN, 2012). Faculty members must recognize they have the primary responsibility to ensure their students can practice nursing competently, not pass the standardized test (NLN, 2012). Also, the NLN leaders recommend multiple means for assessing competence and knowledge. Nursing faculty must use test results to evaluate the curriculum continually and as a guide to improve instruction, not just as a tool to measure student achievement. Faculty members must review all information regarding the comprehensive test prior to implementing policies on high stakes testing. Students must be fully aware of the purpose of the test and the consequences of not passing a high-stakes test (NLN, 2012). Test Anxiety Test anxiety is a prevalent issue among college students (Neuderth, Jabs, & Schmidtke, 2009). Low academic performance and psychological problems often accompany the phenomenon. High-stakes tests often generate feelings of severe anxiety for examinees. It is not the test that creates this anxiety but the stakes linked with the test scores (Madaus & Russell, 2009). Computerized Testing Computerized testing has increased exponentially since the 1990s (Madaus & Russell, 2009). By the 21st century, standardized tests had combined with advanced technology (Madaus & Russell, 2009). Many standardized tests are computerized, including the predictor tests measured in the current study and the NCLEX-RN (ATI, 2011; Kaplan, 2011; NCSBN, 2011). The next section will include a review of the existing literature on NCLEX predictors. 31 Costs of NCLEX-RN Failure A recent update of the NCLEX-RN occurred in April 2010. The result stemmed from a national practice analysis survey on the current practices of new graduate nurses (NCSBN, 2010). The survey included evaluations of 155 nursing bedside interventions. Also, health care administrators expressed the need for greater competence from entrylevel nurses. Leaders at NCSBN updated the NCLEX-RN again in April 2013. Roa et al. (2011) estimated the cost of NCLEX-RN failure for nursing school graduates could be as much as $11,426. The cost includes the loss of salary for a minimum period of the months in California and remediation costs (BRN, 2012). About 15,000 new graduate nurses in the United States fail the NCLEX-RN each year (NCSBN, 2009) on the first attempt. Historical NCLEX Predictor Research Deardorff, Denner, and Miller (1976) looked at 5 years of associate degree in nursing (ADN) graduates at one school of nursing who graduated between 1969 and 1974. Deardorff et al. compared scores on the State Board Examination (SBE) with scores on the NLN Achievement Tests using correlation analysis and multiple regression. The tests consisted of examinations on each nursing subject. The Medical, Pediatric, and Postpartum tests accurately predicted success on the SBE. Washburn (1980) performed a study on one nursing diploma program. The sample consisted of 166 graduates. She also found a significant correlation between the SBE and specific NLN Achievement Tests: Pediatric, Psychiatric Nursing, and Nursing Care of Patients I, II, and III. Bell and Martindill (1976) also tested the predictive accuracy of the NLN Achievement Tests to predict success on the SBE by analyzing the 32 test results of 101 students from one BSN program in Houston, Texas. Bell and Martindill found the Pediatric and Obstetric test scores were the best predictors of the students’ outcome on the SBE. Fowles (1992) compared the Mosby Assess Test results to NCLEX-RN results for 192 graduates of one BSN program. Using multiple regression analysis, Fowles (1992) determined that the Mosby Assess Test accurately predicted NCLEX-RN success. She recommended using the data to help at-risk students enter intervention programs. Current NCLEX Predictor Research Simon, McGinniss, and Krauss (2013) examined predictor variables for NCLEXRN results, NLN-readiness examination results, and nursing program success. The researchers analyzed student records at one urban nursing school in the northeastern United States. Initially, grade point average (GPA), biology grades, and chemistry grades seemed to be predictive of success in the nursing program, but the researchers found the content from the core nursing curriculum overrode these findings. The first nursing course, called Nursing 1, was highly predictive of success on the NLN-readiness examination. Simon et al. (2013) also found transfer students had higher success rates on the NCLEX-RN than the traditional students. The transfer students tended to be older, second-career adults. Grossbach and Kuncel (2011) conducted a meta-analysis on 31 independent samples with 7,159 study participants. The participants consisted of baccalaureate nursing students. The focus of the meta-analysis was the ability of 13 different predictors to forecast student performance on the NCLEX-RN. Effective predictors included SAT/ACT results and baccalaureate nursing course GPA. The second year nursing GPA 33 predictor success was comparable to standardized NCLEX-RN predictor tests such as the NLN and Mosby tests. Bondmass, Moonie, and Kowalski (2008) compared nursing students’ scores on the Nursing Entrance Test (NET) and ERI standardized examinations and their results on the NCLEX-RN. The NET is a standardized nursing school admissions test, and the ERI is a standardized NCLEX-RN predictor test. Because of a low first-time pass rate, the leaders at one study school instituted the NET and ERI tests in the hopes of increasing the pass rates and having more graduate students achieve their goal of becoming registered nurses. The results revealed an increase in NCLEX-RN pass rates after introducing the standardized testing program. Bondmass et al. found the NET and ERI scores significantly correlated to NCLEX-RN success, whereas the students’ GPAs did not. Bondmass et al. did not examine the ability of the NET and ERI to predict failure on the NCLEX-RN, but focused on the probability statistics for success on the NCLEX-RN. Leaders at many nursing schools in the United States use NCLEX-RN predictor tests to determine whether students can graduate from their programs. Spurlock and Hunt (2008) performed a study in which they demonstrated that one school’s HESI Exit Exam results did not accurately predict success on the NCLEX-RN. If the test cannot reliably predict both success and failure on the NCLEX-RN, it should not be a means of determining success in nursing school (Spurlock & Hunt, 2008). Nibert, Young, and Adamson (2006) performed multiple studies measuring the ability of the HESI Exit Exam to accurately predict success on the NCLEX-RN. The HESI Exit Exam is a standardized NCLEX-RN predictor test produced by Elsevier. Nibert et al. disclosed the predictive accuracy of the HESI Exit Exam to determine 34 passing the NCLEX, but not the accuracy of the exam to determine failure. Spurlock and Hanks (2004) noted the positive predictive value of the HESI Exit Exam is only 19%, which means approximately 81% of the students predicted to fail actually passed the boards. McGahee (2010) sought to determine the factors that contributed to success on the NCLEX-RN. The studied variables were SAT or ACT scores, prerequisite nursing course grades, nursing course grades, and standardized assessment tests. McGahee examined two cohorts of students from one nursing school. The science course grades and the RN Assessment test score positively related to success on the NCLEX-RN. Morris and Hancock (2008) performed a study to determine if the HESI Exit Exam helped improve outcomes at a new nursing program. The study had two research questions: (a) How did the new curriculum impact learning as measured by the HESI Exit Exam and (b) Is there a relationship between the HESI Exit Exam score and the first-time pass rate on the NCLEX-RN? Morris and Hancock examined data from two cohorts of students. The data included the results of the HESI Exit Exam and students’ NCLEX-RN results. A relationship did exist between HESI Exit Exam results and NCLEX-RN results. All students who scored 900 or higher on the HESI Exit Exam also passed the NCLEX-RN. Ukpabi (2008) performed a study to identify the variables attributed to success on the NCLEX-RN at a baccalaureate nursing program. The researcher looked at one cohort and examined 18 variables, including ATI tests and several tests created by the NLN. The other variable was the students’ GPA. The ATI and NLN tests were the strongest 35 predictors of success on the NCLEX-RN. The tests can assist leaders in schools of nursing to predict student success and improve quality in their respective programs. Alameda et al. (2011) measured the relationship between the NCLEX-RN and the ATI RN Comprehensive Predictor. The researchers examined two different versions of the examination: Version 3.0 and Form A and B. Form A and B are two versions that intend to predict results from the 2007 NCLEX-RN test plan, and the basis of Version 3.0 is the 2007 test plan. Alameda et al. analyzed the test results from 589 students, including 367 students for Version 3.0 and 222 students for Form A and B. On Version 3.0, the test result accounted for 3% of the course grade. On Form A and B, the test result increased to 10% of the course grade. Alameda et al. discovered the Form A and B group averaged a higher score on their exam than the Version 3.0 group. The researchers attributed this to the higher stakes of the Form A and B test. Harding et al. (2011) measured the ability of the HESI midcurricular examination to predict nursing school success, success on the HESI Exit Exam, and success on the NCLEX-RN. Harding et al. used a retrospective descriptive correlational design to examine the data and found a positive correlation between the midcurricular exam and the exit exam. The midcurricular exam had a predictive value of 95% for passing the NCLEX-RN. These findings might help guide the remediation policy for nursing schools concerning the HESI midcurricular exam. Tipton et al. (2008) studied variables that possibly predicted success in the nursing program and on the NCLEX-RN. The study involved examining entrance test scores, nursing course grades, and GPA for five cohorts of students from one nursing school. The students who had higher nursing course grades were more successful on the 36 NCLEX-RN than those students with lower nursing course grades. The entrance test scores and GPA had no association with student success. Abbott (2008) examined predictors of NCLEX-RN success for nursing students in an accelerated bachelor degree in science program. Abbott looked at preprogram degree, course grades, and the results of the HESI Exit Exam and compared these variables to the students’ NCLEX-RN results. The students who had preprogram science degrees performed better in program courses and on the NCLEX-RN than the other students. The students who performed well on the HESI were more likely to pass the NCLEX-RN. Adamson and Britt (2009) and their associates published several studies regarding the accuracy of the HESI Exit Exam. In previous studies, the researchers focused on the accuracy of the first HESI Exit Exam to predict NCLEX-RN success, but did not examine the ability of subsequent exit exam attempts to predict NCLEX-RN success. Adamson and Britt examined the accuracy of the first, second, and third attempts. The first attempt was 96.44% accurate and the second attempt was 92.94% accurate. However, the third attempt was only 82.5% accurate. Adamson and Britt concluded that students who required three attempts to pass the HESI had greater risk for failing the NCLEX-RN than those who passed following the first or second attempt. Murray, Merriman, and Adamson (2008) measured the ability of the HESI Admission Assessment Test (A2) to predict nursing school success. Murray et al. examined A2 results and course grades for 286 students from two different nursing programs and found that the A2 reliably predicted student success in both the associate’s degree in nursing and the bachelor of science in nursing programs studied. The A2 was a 37 more reliable predictor than the students’ preprogram GPA. The A2 can be a preprogram determinant of admission for prospective nursing students. Lavandera et al. (2011) sought to determine the effectiveness of the HESI tests to predict first-attempt nursing licensure. The researchers compared HESI test results to NCLEX-RN results at one private university in the southeastern United States. They excluded those students who took the NCLEX-RN more than 140 days after graduation, as the lag time can create an increased chance for failure. Using logistic regression to determine predictive accuracy, the researchers found that the HESI Exit Exam accurately predicted success on the NCLEX-RN. Nurse educators can use the results of the HESI Exit Exam to determine necessity for academic remediation prior to sitting for the NCLEX-RN. Lavandera et al. acknowledged that the HESI Exit Exam is a modest predictor of failure on the NCLEX-RN at best. They found that students who received a single D or F on one nursing, science, or mathematics course were more likely to fail the NCLEX-RN than those students who failed the HESI Exit Exam. Zweighaft (2011) evaluated the effectiveness of the HESI specialty exams to predict success on the HESI Exit Exam and the NCLEX-RN at 63 nursing programs across the United States, consisting of six diploma programs, 31 ADN programs, and 26 BSN programs. Of the programs, 43 used both specialty exams and the exit exam, while 20 used only the exit exam. Zweighaft found that the schools that used specialty exams scored significantly higher scores on the exit exam than the schools that did not use specialty exams. Also, some of the HESI specialty exams strongly associated with NCLEX-RN success, including the Medical/Surgical exam, the Fundamentals exam, the Maternity exam, and the Critical Care exam (Zweighaft, 2011). Zweighaft recommended 38 using the results of the specialty exams to determine which students required remediation to ensure the best possible readiness for the HESI Exit Exam and the NCLEX-RN. Kleber (2010) compared exit exam results to National Council Licensure Examination for Practical/Vocational Nurses (NCLEX-PN) results for 411 students from 14 practical nursing programs throughout the state of Kentucky. Using a logistic regression model, Kleber found the first-attempt exit exam result was significant in predicting success on the NCLEX-PN. Subsequent attempts on the exit exam were not significant predictors. Kleber suggested using the exit exam results as a tool to assist atrisk students. These students should receive resources and interventions to increase chances of success on the licensure examination. Benefiel (2011) examined five nursing program and 11 preadmission variables at two Central California community colleges to determine which variables significantly predicted results on the NCLEX-RN. These variables included TEAS, ATI Comprehensive Predictor, and NCLEX-RN results. Benefiel (2011) analyzed the data using logistic regression for four different models. The first model included the traditional and contract students, as well as the overall TEAS score. The model had 92.3% accuracy. The second model was the same as the first, except using the TEAS subject test scores instead of the overall score. This model had 91.7% accuracy. The focus of the third model was the LVN to RN students minus the TEAS variables. This model had 100% accuracy. The fourth model included all students minus the TEAS variables. This model had 90.2% accuracy. The variables did not significantly predict NCLEX-RN failure, with Model 4 having the highest predictive accuracy at 47.1% (Benefiel, 2011). 39 The Fair Testing Imperative in Nursing Education In February 2012, the NLN Board of Governors published The Fair Testing Imperative in Nursing Education on its website. The NLN became aware of the pressure schools of nursing feel to generate high first-time NCLEX-RN pass rates (NLN, 2012). A high first-time pass rate is associated with the need to ensure safe nursing practice for the public. The NLN’s primary concern is the use of standardized predictor tests to prevent nursing students from taking the NCLEX-RN. Because of this concern, the NLN developed national fair testing guidelines to ensure educators and administrators follow evidence-based practices for progression and graduation. The guidelines address the ethics of high stakes testing. The NLN implores nurse educators to use predictor tests that are well-researched and validated across each nursing course. They must also ensure the tests are free of age, gender, ethnic, religious, sexual, or regional bias (NLN, 2012). Beyond the NCLEX-RN Valiga and Ironside (2012) were critical of the current rigor in educational nursing research. A gap exists in the literature that measures outcomes between prelicensure nursing programs and postlicensure practice settings. Measurement of competence should go beyond the NCLEX-RN and NCLEX-RN predictor tests. Valiga and Ironside called for a national agenda focusing on nursing education research. More comprehensive research on postlicensure competence will better assist nursing faculty to ensure prelicensure education will lead to postlicensure safe and effective patient care. 40 Attitudinal Change Carr (2011) acknowledged her students did not believe their university’s first-time NCLEX-RN pass rate was a priority for them. Students tended to be satisfied with graduating nursing school or postponed sitting for the NCLEX-RN even though the chances of passing decrease the longer a student waits (Eich & O’Neill, 2007). However, in 2002, the Long Island University School of Nursing’s first-time NCLEX-RN pass rate averaged only 73.2%. The school had to raise the pass rate or face loss of accreditation from the New York State Education Department. The education department required annual reports and a strategic plan to demonstrate how the school would improve the pass rate. School leaders implemented a standardized exit examination and subject tests, but the exams did not improve the pass rate. When the school began to require students to pass the exit exam before graduation, and the pass rate increased to 93%. However, because the New York State Education Department felt a single criterion for graduation was not acceptable. The policy changed so that the exit exam was only part of the graduation requirements. Carr (2011) believed educating students about the importance of taking the NCLEX-RN early made a difference in their scores. Despite the school’s stoppage of high-stakes testing, the school’s pass rate remained high in 2011. Gaps in the Literature An examination of the extent to which the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test can forecast both success and failure on the NCLEX-RN was the focus of the conducted study. A gap in the literature existed in the exploration of the ability of NCLEX predictor tests to forecast failure on the NCLEX-RN. Many researchers have examined the ability of NCLEX predictor tests 41 to forecast success on the NCLEX-RN, but few have focused on their ability to predict failure on the NCLEX-RN. Few researchers have examined the predictive validity of the Kaplan Readiness Test. Many schools use NCLEX-RN predictor tests to determine if students should pass nursing school, so a failure on the predictor test should have a strong relationship with a failure on the NCLEX-RN (Spurlock & Hunt, 2008). The study involved determining the ability of the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test to predict success and failure on the NCLEX-RN at one location. Summary Several nursing shortages have occurred in the modern American hospital era, but the United States is approaching a shortage unlike any previous one. The baby boomer generation is aging, and every baby boomer will be 65 or older by 2030. Nursing schools should aspire to achieve a high first-time NCLEX-RN pass rate. Tests that accurately predict both success and failure on the NCLEX-RN would help nursing students and faculty achieve high first-time NCLEX-RN pass rates. Leaders in the U.S. government, hospitals, and schools of nursing are continuously involved in a search for solutions to the national nursing shortage. Increasing nursing school graduation rates is one such solution (Uyehara et al., 2007). Decreased attrition will also assist in alleviating the nursing shortage. Chapter 3 will contain details of the design of the study, the appropriateness of the study design, and how the data collection occurred for the retrospective, quantitative study. Chapter 3 will include a description of the methodology selected to collect quantitative data from archived data at one school of nursing in Southern California. The 42 study involved comparing the results of the students’ ATI TEAS, ATI RN Comprehensive Predictor, and Kaplan Readiness Test scores to the first-time results of the NCLEX-RN. 43 Chapter 3 Research Methods The purpose of the retrospective quantitative study was to determine the degree of accuracy that the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test can predict both success and failure on the NCLEX-RN at one nursing program in Southern California. The study included three associated hypotheses. The study is unique because the focus was on measuring the ability of two NCLEX-RN predictor tests and one admission test to forecast both success and failure on the NCLEXRN. Although numerous studies exist that measure the accuracy of NCLEX predictor tests, none include a focus on the ability of the tests to predict failure on the NCLEX-RN. Given the fact that the leaders of some nursing programs use these predictor tests as a graduation requirement, the gap in the literature should be addressed. This chapter will include a discussion on the methodology selected to collect quantitative data from archived data of the last 10 graduating cohorts of a nursing program. The chapter also includes a discussion on the appropriateness of a retrospective, quantitative design for the study, followed by a description of the design and data collection. This chapter contains six sections: (a) research method, (b) research questions, (c) study population, (d) confidentiality, (e) geographic description, and (f) methodology of the data collection. Research Method and Design Appropriateness The quantitative design for the study was appropriate because the research involved analyzing variables to gauge the strength of a relationship between them. The retrospective approach was also appropriate for the study. Retrospective studies involve 44 examining a phenomenon that exists in the present that has a relationship with phenomena from the past (Polit & Beck, 2008). Researchers examine an outcome from the present and determine if a relationship exists with the past phenomenon. The study involved examining the results of the NCLEX-RN and comparing them to the results of the ATI TEAS test, the ATI RN Comprehensive Predictor, and the Kaplan Readiness Test. Students took all these tests prior to the NCLEX-RN. The data collection stage of the study included the cognitive portion of Bloom’s taxonomy. The designers of the NCLEX-RN formulated the test from the cognitive process categories of Bloom’s taxonomy. The NCLEX-RN test plan is the basis of both the ATI RN Comprehensive Predictor and the Kaplan Readiness Test. Population The target population for the retrospective, quantitative study included graduates of a baccalaureate degree nursing program in Southern California. The college has a health sciences division that includes a nursing program. The program has full approval by the California BRN and accreditation by the Commission on Collegiate Nursing Education. Graduates of the program are eligible to sit for the NCLEX-RN after successfully completing all classes and requirements. The requirements include passing the ATI RN Comprehensive Predictor and the Kaplan Readiness Test. The demographics of the nursing program are similar to those of other schools of nursing in the state of California. Sampling The only source of data was archived records of the last 10 graduating cohorts of the nursing program. Approval from the University of Phoenix’s IRB was necessary to 45 proceed with the study. A representative of the participating school signed a permission to use premises form (see Appendix B) and a letter of collaboration to provide evidence of cooperation with the study (see Appendix C). The University of Phoenix IRB provided approval for this research prior to the beginning of data collection. The researcher held and will continue to hold all collected information in the strictest confidentiality. No names appeared on data collected, computer files, or presentations. Three years after the completion of the study, the researcher will shred all student data and erase all electronic files. Geographic Location The geographic location for the study was on the campus of a private university in Southern California. This university has a baccalaureate nursing program that has been in existence since 2006. During this time, students had to achieve a set score on the TEAS examination to be admitted to the nursing program and had to pass both the ATI RN Comprehensive Predictor and the Kaplan Readiness Test to graduate from the program and take the NCLEX-RN. Data Collection Researchers accumulate research data for quantitative studies through an organized procedure that includes details about the information they will collect and how to collect it. Researchers usually cannot collect data from the entire population under study, and therefore will study a representative sample of the group (Polit & Beck, 2008). Collection of data consisted of an organized review of students’ archived records from the past seven cohorts of baccalaureate graduates and the past three cohorts of entry-level MSN graduates from the nursing program. The predictor variables included results from 46 the ATI TEAS test, results of the ATI RN Comprehensive Predictor, and results from the Kaplan Readiness Test (see Appendix C). The study involved comparing the predictor results to the actual NCLEX-RN result, which is a pass or fail result. Students take the ATI TEAS test prior to admission, and the results are a part of the admission qualification process. The results include reading, math, and English scores. Students take the ATI RN Comprehensive Predictor and the Kaplan Readiness Test prior to graduation and must achieve a minimum passing score on each test prior to sitting for the NCLEX-RN. The University of Phoenix IRB provided permission to perform the study. In addition, the dean of nursing signed a permission to use premises letter and a personal letter of permission. The nursing program data technician created a code for each student name to maintain compliance with student data privacy regulations. The nursing program data technician was not part of the study and is an employee of the nursing school. No student names or identifying data were a part of the data collected. All nursing student data were deidentified, provided to the researcher, and analyzed. All data will remain confidential, and all records will be secure for 3 years after completion of the study. After the 3 years, the researcher will destroy all records. Instrumentation The instrument used for data collection was simple, organized, and systematic. The instrument was a table created specifically for collecting data for the study. The only data entered into the table were nominal, ordinal, and interval data. For each item in the study, a clear definition exists (see Appendix A). 47 Validity and Reliability Internal validity ensures the independent variable caused the outcome, rather than other factors (Polit & Beck, 2008). Researchers must develop approaches that will prevent other aspects outside a study causing the outcome. The study included measures to make certain the interval validity is intact. Selecting the entire population of graduates in the nursing program eliminates selection bias, which is one of the greatest threats to internal validity (Polit & Beck, 2008). Mortality is another threat to internal validity, but was not an issue in the study because of the retrospective nature of the study. All nursing graduates who took the NCLEX-RN took the other predictor tests previously. The study did not provide a direct benefit to any company, institution, or individual during the study period. External validity involves the ability to apply the observed relationships in a study to various settings and conditions (Polit & Beck, 2008). Evidence gained from a controlled study with external validity could apply to real-world settings. The researcher addressed this concept in the study. The university that was the setting for the study had a curriculum that was fully compliant with the guidelines set forth by the California BRN. The design of the study did not require any changes to the nursing program environment. The curriculum at the school is similar to other curricula throughout the state of California, so the results of the study will be applicable to other schools in California with a similar nursing program environment (BRN, 2011). Because the study included only archival data, the results were generalizable to similar nursing schools in California, and because the study sample 48 did not involve human subjects, subject bias was not a consideration. The reactions of subjects were not a threat to external validity in the study. Data Analysis The study involved a statistical analysis to determine the significance of research findings. The analysis in the study served to determine if research findings are significant by confirming either the hypothesis or the null hypothesis. The research design serves to determine the type of analysis performed (Burns & Grove, 2009). The study included SPSS 21 for all statistical analyses. Research variables. The research variables included the ATI TEAS reading score, the ATI TEAS math score, the ATI TEAS English score, the ATI Comprehensive Predictor score, and the Kaplan Readiness Test score. The study included point biserial correlation to determine the level of significance between the aforementioned variables with the pass and fail rates on the first-time results of the NCLEX-RN test. Logistic regression analysis. Logistic regression analysis helped to determine the accuracy of predictor variables related to the actual results of the first-time NCLEXRN test. “Logistic regression analyzes the relationship between multiple independent variables and a dependent variable and yields a predictive equation” (Polit & Beck, 2008, p. 629). A logistic regression converts the probability of an event occurring into odds between 0 and 1. The higher the number is, the greater is the probability of an event occurring (Polit & Beck, 2008). Summary The quantitative study involved examining nursing students’ ATI TEAS test scores, ATI RN Comprehensive Predictor scores, and Kaplan Readiness Test scores and 49 the ability to predict students’ success and failure on the NCLEX-RN. Chapter 3 contained a description of the study method, design, and data collection process. The retrospective design was the quantitative method chosen. The study involved performing a statistical analysis with the data collected from the school of nursing. Chapter 4 will include the methodology implemented in this study. The chapter also includes a discussion on the design, study, variables, data collection, statistical analyses, and limitations. 50 Chapter 4 Results A gap existed in the literature regarding the ability of the ATI RN Comprehensive Predictor or the Kaplan NCLEX-RN Readiness Test to accurately forecast failure on the NCLEX-RN. The purpose of the quantitative study was to add this information into the literature. The independent variables included the baccalaureate and ATI TEAS scores, the ATI RN Comprehensive Predictor scores, and the Kaplan NCLEX-RN Readiness Test scores of the entry-level master’s students. The dependent variable was the BSN and MSN students’ first-time results on the NCLEX-RN. The examination included archived records from 251 BSN and MSN student records. Chapter 4 will include a description of the screening procedures and data collection of the study, a synopsis of the statistical analyses used to address the research questions, and the results of the analysis in the quantitative study. The study involved analyzing all the independent and dependent variables. Analysis results will then be aligned with the hypotheses. Research Questions The research project included two questions. The questions were appropriate to the study and highlighted the dependent and independent variables. The study involved determining the extent to which two standardized nursing exams can predict both success and failure on the NCLEX-RN at one nursing school in Southern California. 1. Are NCLEX-RN predictor examination scores (independent variable) significantly related to results on the NCLEX-RN (dependent variable)? 51 2. Are nursing school admissions test scores (independent variable) significantly related to results on the NCLEX-RN (dependent variable)? The associate dean of the university under study granted permission to use the premises and the name of the university (see Appendix A) as well as permission to use the data (see Appendix B). Data analysis included SPSS 21. The level of statistical significance was set at p = .05 to determine if a significant relationship existed between the independent variables and first-time success or failure on the students’ first attempt on the NCLEX-RN. Data Analysis The purpose of the quantitative study was to determine if the TEAS, the ATI RN Comprehensive Predictor, and the Kaplan NCLEX-RN Predictive Exit Test could accurately predict both success and failure on the NCLEX-RN at one nursing program in Southern California. The data gathered for the study came from 251 students. Table 1 shows the frequency counts for selected variables. The sample included more female students (90.0%) than male students (10.0%). All but three (98.8%) had U.S. citizenship. The most common racial and ethnic categories were Caucasian (62.2%), Asian/Pacific Islander (15.5%), and Hispanic (14.7%). For the entire sample, 90.8% of the students passed the NCLEX on their first attempt (see Table 1). Table 2 shows the descriptive statistics for selected variables. The statistics included the student’s age (M = 27.73, SD = 5.26), the TEAS total score (M = 82.40, SD = 4.59), and the ATI score (M = 71.59, SD = 6.76). The table also includes the seven TEAS subscale scores (see Table 2). 52 Table 1 Frequency Counts for Selected Demographic Variables (N = 251) Variables and category Gender Female Male Citizenship Other United States Race/ethnicity African American Asian/Pacific Islander Caucasian Hispanic Other NCLEX outcome Failed Passed n % 226 25 90.0 10.0 3 248 1.2 98.8 11 39 156 37 8 4.4 15.5 62.2 14.7 3.2 23 228 9.2 90.8 Table 2 Descriptive Statistics for Selected Variables (N = 251) Variable M SD Low High Age 27.73 5.26 22.00 61.00 TEAS total score 82.40 4.59 69.30 93.50 Reading 91.25 5.93 54.80 100.00 Inferences 87.82 12.69 33.30 100.00 Math 78.10 9.91 44.40 100.00 Diagrams 77.93 15.83 33.30 100.00 Science 75.84 8.20 53.30 93.30 Science reasoning 68.50 20.26 20.00 100.00 English 83.43 5.71 63.30 100.00 a ATI score 71.59 6.76 55.30 89.30 a For this variable, n = 219 because school leaders did not administer it some semesters. Table 3 shows the distribution of predicted and actual NCLEX scores and pass rates for the students who took the 2007 ATI RN Comprehensive Predictor. For all but one of the predicted score ranges, the students in the sample either had similar or better pass rates than expected based on the predictions provided by ATI (2007; see Table 3). 53 Table 3 Distribution of Predicted and Actual NCLEX Scores and Pass Rates for 2007 Predictor (n = 162) RN Comprehensive Predictor score range (%)a ≥82.0 79.3-81.3 76.0-78.7 74.0-75.3 72.0-73.3 70.7-71.3 68.7-70.0 67.3-68.0 65.3-66.7 62.0-64.7 56.7-61.3 0.0-56.0 Predicted probability of passing the NCLEX (%) 99 98 96-97 94-95 91-93 89-90 85-88 82-84 76-80 63-73 38-60 1-36 Actual nb 10 19 17 8 27 14 12 7 10 11 10 1 Actual pass rate (%)c 100.0 100.0 100.0 100.0 100.0 100.0 81.8 90.9 78.9 78.9 82.4 100.0 Table 4 displays the distribution of predicted and actual NCLEX scores and pass rates for students who took the 2010 ATI Comprehensive Predictor. For all but one of the predicted score ranges, the students in the present sample had either similar or better pass rates than expected based on the predictions provided by ATI (2010; see Table 4). Testing of the Hypotheses Null Hypothesis 1 was as follows: The ATI RN Comprehensive Predictor test score is not significantly related to results on the NCLEX-RN. Table 4 shows the chisquare tests used to test the relationship between the two variables. This table includes the chi-square tests for the total sample (n = 219), the traditional subsample (n = 182), and the accelerated subsample (n = 37). A significant relationship existed in both the total sample (p = .001) and the traditional subsample (p = .001). In the accelerated 54 student subsample, all 37 students passed the NCLEX on the first attempt. This combination of findings provided support to reject Null Hypothesis 1 (see Table 4). Table 4 Distribution of Predicted and Actual NCLEX Scores and Pass Rates for 2010 Predictor (n = 57) RN Comprehensive Predictor Predicted probability of Actual pass score range (%)a passing the NCLEX (%) Actual nb rate (%)c ≥80.7 99 2 100.0 78.0-80 98 7 100.0 74.7-77.3 96-97 4 100.0 72.0-74.0 94-95 9 89.0 70.0-71.3 91-93 6 100.0 68.7-69.3 89-90 7 100.0 66.0-68.0 84-88 6 100.0 64.7-65.3 81-82 3 100.0 62.7-64.0 75-79 5 100.0 59.3-62.0 63-73 4 75.0 53.3-58.7 37-60 4 75.0 0.0-52.7 1-34 0 100.0 a NCLEX predictions based on data supplied by ATI RN Comprehensive Predictor 2010 Proctored Assessment. bNumber of students in current study who passed the NCLEX but had a predicted score in that specific range. cNCLEX pass rate for students with a predicted score in that specific range. Null Hypothesis 2 was as follows: The Kaplan NCLEX-RN Readiness Test score is not significantly related to results on the NCLEX-RN. To examine this hypothesis, an analysis of variance (ANOVA) test was suitable for the 100 students who took the Kaplan NCLEX-RN Readiness Test. The ANOVA yielded no significant differences between Readiness Test results and NCLEX-RN results (p < .05). These findings provided support to fail to reject Null Hypothesis 2. Null Hypothesis 3 was as follows: The ATI TEAS nursing school admissions test score is not significantly related to results on the NCLEX-RN. Table 5 displays the t 55 tests for independent means comparing whether students passed the NCLEX on their first attempt with the TEAS total score along with the seven TEAS subscale scores. Only the TEAS Math score was significantly higher (p = .005) for students who passed the NCLEX. This combination of findings provided limited support to reject Null Hypothesis 3. Table 5 t Test Comparisons Based on NCLEX Outcome for Selected Variables Variable and outcome n M SD rpb t p Age .07 1.12 .27 Fail 23 26.57 3.13 Pass 228 27.85 5.42 TEAS Score .10 1.60 .11 3.35 Fail 23 80.95 Pass 228 82.55 4.67 Reading .02 0.29 .78 Fail 23 90.91 4.71 Pass 228 91.28 6.05 Inferences .03 0.41 .68 Fail 23 88.85 12.37 Pass 228 87.72 12.74 Math .18 2.86 .005 Fail 23 72.55 8.01 Pass 228 78.66 9.93 rpb = Point biserial correlations used as a measure of the strength of the relationship. Also in Table 5 is the t test for independent means comparison of the students’ ATI score with whether they passed the NCLEX on their first attempt. Students who passed had significantly higher ATI scores (p = .001). The table also includes the pointbiserial correlation (rpb) used as a measure of the strength of that relationship. The correlation (rpb = .27, rpb2 = .073) found that the ATI score accounted for 7.3% of the variance in whether the student passed the NCLEX on the first attempt. 56 In conclusion, the study included data from 251 students to attempt to determine if the TEAS, the ATI RN Comprehensive Predictor, and the Kaplan NCLEX-RN Predictive Exit Test could accurately predict both success and failure on the NCLEX-RN. The overall NCLEX pass rate for this sample was 90.8% (see Table 1). Hypothesis testing resulted in supporting Hypothesis 1 (ATI test with the NCLEX test result; see Tables 4, 5, and 6). It was not possible to test Hypothesis 2 (Kaplan score with NCLEX test result) due to limited available data. Hypothesis 3 (TEAS scores with NCLEX test result) received limited support, with only the Math subscale score being significantly higher for successful NCLEX students (see Table 6). The final chapter will contain a comparison of these findings to the literature, as well as conclusions, implications, and a series of recommendations. Findings The sample in the quantitative study consisted of 251 BSN and MSN students: 45 traditional BSN students who graduated in 2009, six entry-level MSN students who graduated in 2009, 48 traditional BSN students who graduated in 2010, 19 entry-level MSN students who graduated in 2010, 52 traditional BSN students who graduated in 2011, 13 entry-level MSN students who graduated in 2011, and 38 traditional BSN students who graduated in 2012. Of the students, 89% passed the NCLEX-RN on their first attempt. The study involved calculations for the minimum, maximum, mean, and standard deviations for the ATI TEAS scores, the ATI RN Comprehensive Predictor scores, and the Kaplan NCLEX-RN Readiness Test scores. The ATI TEAS scores ranged from a low of 25 to a high of 100, with a mean ATI TEAS score of 50. The ATI RN 57 Comprehensive Predictor scores ranged from a low of 25 to a high of 100, with a mean ATI RN Comprehensive Predictor score of 50. The Kaplan NCLEX-RN Readiness Test scores ranged from a low of 25 to a high of 100, with a mean Kaplan NCLEX-RN Readiness Test score of 50. Testing of the Hypotheses Null Hypothesis 1 was as follows: The ATI RN Comprehensive Predictor test score is significantly related to results on the NCLEX-RN. Table 6 displays the chisquare tests used to test the relationship between the two variables. This table includes the chi-square tests for the total sample (n = 219), the traditional subsample (n = 182), and the accelerated subsample (n = 37). A significant relationship was found in both the total sample (p = .001) and the traditional subsample (p = .001). In the accelerated student subsample, all 37 students passed the NCLEX on the first attempt. This combination of findings provided support to reject Null Hypothesis 1 (see Table 6). Table 6 Relationship Between Outcome of ATI Screening Test and NCLEX Outcome Failed n % a Passed n % Sample and ATI outcome Total sample (n = 219) b Fail 16 16.0 84 84.0 Pass 3 2.5 116 97.5 Traditional subsample only (n = 182) c Fail 16 18.0 73 82.0 Pass 3 3.2 90 96.8 Accelerated subsample only (n = 37) d Fail 0 0.0 11 100.0 Pass 0 0.0 26 100.0 a Passing the ATI screening test meant achieving at least the 90th percentile. bTotal sample: χ2 (1, n = 219) = 12.46, p = .001. Cramer’s V = .24. cTraditional subsample: χ2 (1, n = 182) = 10.59, p = .001. Cramer’s V = .21. dAccelerated subsample: no chi-square or Cramer’s V test because all passed the NCLEX. 58 Null Hypothesis 2 was as follows: The Kaplan Predictive Exit Test score is significantly related to results on the NCLEX-RN. Given that less than half the students took the Kaplan test, the researcher did not test this hypothesis. Null Hypothesis 3 was as follows: The ATI TEAS nursing school admissions test score is not significantly related to results on the NCLEX-RN. To examine this hypothesis, Table 5 displayed the t tests for independent means comparing whether students passed the NCLEX on their first attempt with the TEAS total score along with the seven TEAS subscale scores. Only the TEAS Math score was significantly higher (p = .005) for students who passed the NCLEX. This combination of findings provided limited support to reject Null Hypothesis 3. Table 5 also showed the t test for independent means between the students’ ATI score and whether they passed the NCLEX on their first attempt. Students who passed had significantly higher ATI scores (p = .001). The table included the point-biserial correlation (rpb) used as a measure of the strength of that relationship. The correlation (rpb = .27, rpb2 = .073) found that the ATI score accounted for 7.3% of the variance in whether students passed the NCLEX on their first attempt. This study included data from 251 students to determine if the TEAS, the ATI RN Comprehensive Predictor, and the Kaplan NCLEX-RN Predictive Exit Test could accurately predict both success and failure on the NCLEX-RN. The overall NCLEX pass rate for this sample was 90.8% (see Table 1). Hypothesis testing revealed support for Hypothesis 1 (ATI test with the NCLEX test result; see Tables 4 and 5). Testing of Hypothesis 2 (Kaplan score with NCLEX test result) did not occur due to limited available data. Hypothesis 3 (TEAS scores with NCLEX test result) found limited 59 support, with only the Math subscale score being significantly higher for successful NCLEX students (see Table 5). Chapter Summary Chapter 4 contained the findings for the statistical analyses. The statistical analyses determined if the predictor variables had significant relationships between passing and failing the NCLEX-RN examination on the students’ first attempt. The predictor variables included ATI TEAS scores, ATI RN Comprehensive Predictor scores, and Kaplan NCLEX-RN Readiness Test scores. Chapter 5 contains a discussion of the research findings, interpretations, and conclusions. The chapter also includes recommendations for future research studies. 60 Chapter 5 Conclusions and Recommendations Chapter 5 contains a discussion of the research findings, interpretations, and conclusions. Also reviewed will be findings from other studies that are in the line with the results from the current study. The chapter also includes recommendations for future research. The severe and worsening nursing shortage in the United States has increased the need to examine progression policies in American schools of nursing. The specific problem discussed in this study was the use of NCLEX-RN predictor examinations as high-stakes tests to ensure satisfactory first-time NCLEX-RN pass rates at schools of nursing in the United States. If predictor examinations do not correctly predict failure on the NCLEX-RN, then leaders of nursing schools should not hold nursing students back based on the results of an inaccurate predictor test. To help alleviate the severe and worsening nursing shortage, nursing schools in the United States must graduate all students who can demonstrate safe and effective clinical practice in the classroom and clinical settings. Purpose of the Study There is a gap in the literature regarding the ability of the ATI RN Comprehensive Predictor or the Kaplan NCLEX-RN Readiness Test to forecast failure on the NCLEXRN accurately. The purpose of the study was to determine if the ATI TEAS, the ATI RN Comprehensive Predictor, and the Kaplan NCLEX-RN Readiness Test could accurately predict both success and failure on the NCLEX-RN at one baccalaureate degree nursing program in Southern California. The independent variables were BSN students’ ATI 61 TEAS scores, ATI RN Comprehensive Predictor scores, and Kaplan NCLEX-RN Readiness Test scores. The dependent variable was the BSN students’ first-time results on the NCLEX-RN. The archived records examined were from 251 BSN and entry-level MSN students. Research Questions 1. Are NCLEX-RN predictor examination scores (independent variable) significantly related to results on the NCLEX-RN (dependent variable)? 2. Are nursing school admissions test scores (independent variable) significantly related to results on the NCLEX-RN (dependent variable)? Hypotheses The study also included six hypotheses. H10: The ATI RN Comprehensive Predictor test score is not significantly related to results on the NCLEX-RN. H1a: The ATI RN Comprehensive Predictor test score is significantly related to results on the NCLEX-RN. The study involved testing Hypothesis 1 using ANOVA to determine if a significant relationship existed between the ATI RN Comprehensive Predictor test and NCLEX-RN results. H20: The Kaplan Readiness Test score is not significantly related to results on the NCLEX-RN. H2a: The Kaplan Readiness Test score is significantly related to results on the NCLEX-RN. 62 The study involved testing Hypothesis 2 using ANOVA to determine if a significant relationship existed between the Kaplan Readiness Test and NCLEX-RN results. H30: The ATI TEAS nursing school admissions test score is significantly related to results on the NCLEX-RN. H3a: The ATI TEAS nursing school admissions test score is not significantly related to results on the NCLEX-RN. The study involved testing Hypothesis 3 using ANOVA to determine if a significant relationship existed between the ATI TEAS nursing school admissions test and NCLEXRN results. Summary of Key Findings Null Hypothesis 1 was as follows: The ATI RN Comprehensive Predictor test score is not significantly related to results on the NCLEX-RN. Table 6 showed the chisquare tests used to test the relationship between the two variables. This table includes the chi-square tests for the total sample of students who took the ATI RN Comprehensive Predictor (n = 219), the traditional subsample (n = 182), and accelerated subsample (n = 37). A significant relationship existed in both the total sample (p = .001) and the BSN subsample (p = .001). In the MSN student subsample, all 37 students passed the NCLEX on the first attempt. This combination of findings provided support to reject Null Hypothesis 1. The quantitative method was appropriate for identifying the ability of the independent variables to predict success and failure on the NCLEX-RN. The findings offered substantiation to reject Null Hypothesis 1 for the ATI RN Comprehensive 63 Predictor. The higher the students scored on the ATI RN Comprehensive Predictor, the more likely they were to pass NCLEX-RN on the first try. Null Hypothesis 2 was as follows: The Kaplan Predictive Exit Test score is not significantly related to results on the NCLEX-RN. Given that less than half the students took the Kaplan test, the stud did not involve testing this hypothesis. Null Hypothesis 3 was as follows: The ATI TEAS nursing school admissions test score is not significantly related to results on the NCLEX-RN. Table 5 displayed the t tests for independent means comparing whether the students passed the NCLEX on their first attempt with the TEAS total score along with the seven TEAS subscale scores. Data analysis revealed the TEAS Math score was significantly higher (p = .005) for students who passed the NCLEX. This combination of findings provided limited support to reject Null Hypothesis 3. The higher the student scored on the TEAS mathematics test, the more likely they were to pass NCLEX-RN on the first try. Because ATI does not directly measure the relationship between TEAS scores and NCLEX-RN results, nursing faculty and administrators should find these data to be helpful to their programs. These results are similar to other studies, such as one conducted by Ukpabi (2008). Ukpabi (2008) performed a study to identify the variables attributed to success on the NCLEX-RN at a baccalaureate nursing program. The ATI and NLN tests were the strongest predictors of success on the NCLEX-RN. Implications This study resulted in numerous contributions to the literature. Few published researchers have studied the predictive ability of the Kaplan Readiness Test. This study 64 involved examining the predictive ability of both the ATI tests and the Kaplan predictor at one nursing school. Data from the study can assist in shaping admission and progression policies for nursing programs that use these predictor tests. The 219 students who took the ATI RN Comprehensive Predictor, both 2007 and 2010 versions, had similar or better pass rates than the rates predicted by ATI for all but one of the predicted score ranges (ATI, 2007, 2010). One hundred students failed the ATI RN Comprehensive Predictor on the first attempt. Of these 100 students, 84 passed NCLEX-RN on the first attempt, which translated to 84% passing. The ATI scale predicted these students would have only a 71% chance of passing NCLEX-RN. These statistics call into question the predictive ability of the ATI RN Comprehensive Predictor to forecast failure on the NCLEX-RN. Data analysis showed no significant relationship between the Kaplan NCLEX-RN Readiness Test and NCLEX-RN results (p > .05). However, because only 100 students took the Kaplan Readiness Test, further study will be necessary to validate findings. Nursing administration and faculty should use the Kaplan Readiness Test with caution until researchers can independently validate its validity. The overall TEAS score did not have a significant relationship with NCLEX-RN results. However, the ATI TEAS Math score was significantly higher (p = .005) for students who passed the NCLEX-RN. Nursing administration and faculty should take a close look at this score for future applicants. Limitations The study included several limitations. The study had a nonrandomized sample. The data collected were from only one nursing program in California. The students all 65 had gone to school under the same support system. Another limitation was the different numbers of students who passed the first attempt of the NCLEX-RN versus those who failed the first time. In this study, 200 students were in the NCLEX-RN pass group, whereas only 19 students were in the fail group. The fail group size might not have been large enough to accurately measure the prediction of failure on the NCLEX-RN. These factors can limit the generalizability of the findings of this study to other nursing programs across the United States. The study did not involve accounting for individual differences among the students. The differences included age, ethnicity, gender, and socioeconomic status. Suggestions for Future Research The results of the current study are in agreement with the literature that the NCLEX predictor examinations often predict NCLEX-RN success accurately. The study has also found that the accuracy of failure prediction is significantly less accurate. However, researchers should replicate this type of study at many nursing schools across the United States. Future researchers should study differences in predictive accuracy between students of different ethnicities. Researchers should also study student test results from ADN, BSN, and direct-entry MSN programs, as well as the differences in test results for students of various age groups. Another gap in the literature is the NCLEX predictors’ ability to predict a successful career as a registered nurse. A researcher should conduct a longitudinal study on NCLEX predictor results and a nurse’s patient care outcomes. The study would require collaboration between the nursing school and the facilities where the graduates work. 66 Significance The significance of the quantitative study was the contribution of evidence-based practice for restructuring progression policies in nursing schools across the United States. In February 2012, the NLN Board of Governors published The Fair Testing Imperative in Nursing Education on its website. The NLN became aware of the pressure schools of nursing feel to generate high first-time NCLEX-RN pass rates (NLN, 2012). An association exists between high first-time pass rates and the need to ensure safe nursing practice for the public. The NLN’s primary concern is using standardized predictor tests to prevent nursing students from taking the NCLEX-RN. Because of this concern, the NLN developed national fair testing guidelines to ensure educators and administrators follow evidence-based practices for progression and graduation. The guidelines address the ethics of high-stakes testing. Leaders at the NLN implore nurse educators to use predictor tests that are well-researched and validated across each nursing course. They must also ensure the tests are free of age, gender, ethnic, religious, sexual, or regional bias (NLN, 2012). First-time NCLEX-RN pass rate is the measure used to determine the quality of nursing schools in the United States. However, the school’s attrition rate is not often a measure of quality. The first-time NCLEX-RN pass rate for California schools of nursing is readily available on the California BRN website. The only other statistic provided on the site is the number of students who sat for the NCLEX-RN in that particular academic year. The website does not include the attrition rate. Government authorities and regulatory agencies should report nursing school attrition rates to 67 encourage nursing schools to reduce their attrition rates. Remeasuring the NCLEX-RN passing standard every 3 years may not be enough to ensure the safety of the public. “Just as grades are not the only result of the college experience, the environment is not the only cause of learning. Measuring only the components of students and structure overlooks the complete equation: input plus structure equals outcomes” (Cornell, 1985, p. 357). Structure (NCLEX-RN results) should not be the sole determinant of the quality of a nursing program. Researchers have not extensively studied the consequences of low NCLEX-RN pass rates for nursing programs (Spurlock, 2013). Although measuring the input level of a nursing student is more complicated than measuring the structure, its value toward the safety of the public cannot be undermined. Practitioner Recommendations Nursing faculty should create course examinations in a similar style to the NCLEX-RN. Consistently performing well on course examinations but struggling on NCLEX predictor tests could demonstrate that the course examinations are dissimilar in style or content to the predictor test. Faculty should focus on graduating high-quality nurses who can alleviate the nursing shortage. The focus should not be on passing any one test that could lead to increased attrition rates. Nursing school administrators strive to obtain a high first-time NCLEX-RN pass rate. However, they typically do not seek to maintain lower attrition rates. The first-time NCLEX-RN pass rate for California nursing schools is easily available on the California BRN website. Including the attrition rate on the website would present a clearer picture of the school’s quality. In this scenario, nursing school administrators would probably pay greater attention to their attrition rate. Recommendations include a decreased 68 reliance on NCLEX-RN predictors to determine readiness for the NCLEX-RN. Another recommendation is the inclusion of the percentage of examinees who passed the NCLEXRN on the second attempt. Another contributing factor to reliance on NCLEX-RN predictors is the inability for nursing instructors to write items that are comparable to NCLEX-RN questions. Even experienced faculty members have difficulty developing reliable and valid test questions to adequately evaluate student learning (Halstead, 2013). Most nursing faculty would benefit from attending item-writing workshops, such as those offered by NCSBN (NCSBN, 2013). Just as hospitals use continuous quality improvement for patient outcomes, faculty must continuously improve testing and remediation procedures. If more nursing faculty would learn to write test items that are similar to NCLEX-RN items, there would be fewer reasons to rely on NCLEX-RN predictors and administer them as high-stakes tests. Many nursing programs do not rely on high-stakes tests to achieve excellent NCLEX-RN pass rates (Spurlock, 2013). More research is necessary to determine the reasons for their success. Conclusion The study involved investigating the effectiveness of the ATI TEAS test, the ATI RN Comprehensive Predictor examination, and the Kaplan Readiness Test to predict NCLEX-RN outcomes. The findings of this study might help to reexamine progression policies at undergraduate schools of nursing. The results of the study demonstrated that the NCLEX predictors studied were not accurate in predicting failure on the NCLEX-RN. Researchers should study students from various schools to obtain a larger sample size and 69 should examine the differences between students of various ages, ethnicities, genders, and socioeconomic statuses. 70 References Abbott, A. A. (2008). Predictors of success on National Council Licensure Examination for registered nurses for accelerated baccalaureate nursing graduates. Nurse Educator, 33, 5-6. doi:10.1097/01.NCN.0000336453.62659.06 Adamson, C., & Britt, R. (2009). Repeat testing with the HESI exit exam-sixth validity study. CIN: Computers, Informatics, Nursing, 27(6), 393-397. doi:10.1097/NCN.0b013e3181c1367d Aiken, L. H., Sloane, D. M., Cimiotti, J. P., Clarke, S. P., Flynn, L., Seago, J. A., . . . Smith, H. L. (2010). Implications of the California nurse staffing mandate for other states. HSR: Health Sciences Research, 45, 904-921. doi:10.1111/j.14756773.2010.01114.x Alameda, M. D., Prive, A., Davis, H. C., Landry, L., Renwanz-Boyle, A., & Dunham, M. (2011). Predicting NCLEX-RN success in a diverse student population. Journal of Nursing Education, 50(5), 261-267. doi: 10.3928/01484834-20110228-01 Ali, A., Zaman, A., & Alamgir. (2011). Correlation between admission criteria and academic performance of students in dental colleges of Khyber Pukhtunkhawa. FWU Journal of Social Sciences, 5, 1-16. Retrieved from http://www.fwu.edu.pk/downloads/ American Association of Colleges of Nursing. (2011). Nursing shortage fact sheet. Retrieved from http://www.aacn.nche.edu/Media/FactSheets/NursingShortage.htm 71 American Nurses Association. (2011). Nurse staffing plans and ratios. Retrieved from http://www.nursingworld.org/MainMenuCategories/PolicyAdvocacy/State/Legislative-Agenda-Reports/State-StaffingPlansRatios Anderson, L. W., & Krathwohl, D. R. (Eds.). (2001). A taxonomy for learning, teaching, and assessing: A revision of Bloom’s taxonomy of educational objectives. New York, NY: Longman. Ari, A. (2011). Finding acceptance of Bloom’s revised cognitive taxonomy on the international stage and in Turkey. Educational Sciences: Theory & Practice, 11, 767-772. Retrieved from http://www.edam.com.tr/kuyeb/en/default.asp Assessment Technologies Institute. (2011). Comprehensive assessment and review program. Retrieved from https://atitesting.com/Solutions/DuringNursingSchool /ComprehensiveAssessmentAndReviewProgram.aspx Aucoin, J. W., & Treas, L. (2005). Assumptions and realities of the NCLEX-RN. Nursing Education Perspectives, 26(5), 268-271. Retrieved July 24, 2011, from http://www.nln.org/nlnjournal Bell, J. A., & Martindill, C. F. (1976). A cross-validation study for predictors of scores on state board examinations. Nursing Research, 26(7), 278-281. Retrieved August 2, 2011 from http://journals.lww.com/nursingresearchonline/pages/default.aspx Benefiel, D. (2011). Predictors of success and failure for ADN students on the NCLEXRN (Doctoral dissertation). Available from ProQuest Dissertations and Theses database. (UMI No. 3456526) 72 Birnbach, N. (1982). The genesis of the nurse registration movement in the United States, 1893-1903. Ed.D. diss., Teachers College, Columbia University. Bloom, B. S., Engelhart, M. D., Furst, E. J., Hill, W. H., & Krathwohl, D. R. (1956). Taxonomy of educational objectives: The cognitive domain. New York, NY: Longman. Boland, D. (2009). Developing curriculum: Frameworks, outcomes, and competencies. In D. M. Billings & J. A. Halstead (Eds.), Teaching in nursing: A guide for faculty (pp. 137-153). St. Louis, MO: Saunders. Bondmass, M., Moonie, S., & Kowalski S. (2008). Comparing NET and ERI standardized exam scores between baccalaureate graduates who pass or fail the NCLEX-RN. International Journal of Nursing Education Scholarship, 5, Article 16. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/18384274 Buerhaus, P. (1987). Not just another nursing shortage. Nursing Economic$, 5(6), 267279. Retrieved from http://www.nursingeconomics.net Buerhaus, P. (1991). Economic determinants of the annual number of hours worked by registered nurses. Medical Care, 29, 1181-1195. Retrieved from http://journals.lww.com/lww-medicalcare/pages/default.aspx Buerhaus, P., Staiger, D. O., & Auerbach, D. I. (2009). The future of the nursing workforce in the United States: Data, trends, and implications. Boston, MA: Jones & Bartlett. Bumen, N. T. (2007). Effects of the original versus revised Bloom’s taxonomy on lesson planning skills: A Turkish study among pre-service teachers. Review of Education, 53, 439-455. doi:10.1007/s11159-007-9052-1 73 Bureau of Labor Statistics. (2012). Occupational outlook handbook (OOH), 2012-13 edition. Retrieved from http://www.bls.gov/ooh/healthcare/registered-nurses.htm Burns, N., & Grove, S. K. (2009). The practice of nursing research: Appraisal, synthesis, and generation of evidence (6th ed.). Philadelphia, PA: W. B. Saunders. California Board of Registered Nursing. (2011). 2009-2010 annual school report. Retrieved from http://www.rn.ca.gov/pdfs/schools/schoolrpt09-10.pdf California Department of Public Health. (2010). Nurse to patient ratio. Retrieved from http:// www.cdph.ca.gov/services/DPOPP/regs/Pages/N2PRegulations.aspx Carr, S. M. (2011). NCLEX-RN pass rate peril: One school’s journey through curriculum revision, standardized testing, and attitudinal change. Nursing Education Perspectives, 32(6), 384-388. Retrieved February 12, 2012 from http://www.nln.org/nlnjournal Carrick, J. A. (2011). Student achievement and NCLEX-RN success: Problems that persist. Nursing Education Perspectives, 32(2), 78-83. doi:10.5480/1536-502632.2.78 Chitty, K. K., & Black, B. P. (2011). Professional nursing: Challenges and concepts (6th ed.). Maryland Heights, MO: Saunders Elsevier. Cornell, G. (1985). The value-added approach to the measurement of educational quality…measuring student gains. Journal of Professional Nursing, 1(6), 356363. Creswell, J. W. (2005). Educational research: Planning, conducting, and evaluating qualitative and quantitative research (2nd ed.). Upper Saddle River, NJ: Pearson, Merrill, Prentice-Hall. 74 Deardorff, M., Denner, P., & Miller, C. (1976). Selected National League of Nursing Achievement Test scores as predictors of state board examination scores. Nursing Research, 25, 35-38. Retrieved August 2, 2011 from http://journals.lww.com/nursingresearchonline/pages/default.aspx Duan, Y. (2006). Selecting and applying taxonomies for learning outcomes: A nursing example. International Journal of Nursing Education Scholarship, 3, Article 10. doi:10.2202/1548-923X.1154 Duncan, B. A., & Stevens, A. (2011). High-stakes standardized testing: Help or hindrance to public education. National Social Science Journal, 36(2), 35-43. Retrieved from http://www.nssa.us/journals/pdf/NSS_Journal_36_2.pdf EBSCO Industries. (2012). About us. Retrieved from http://www.EBSCO.com Eich, M., & O’Neill, T. (2007). NCLEX delay pass rate study. Retrieved from http://www.ncsbn.org Federal Interagency Forum on Aging Related Statistics. (2009). Data sources on older Americans 2009. Hyattsville, MD: Author. Fowles, E. R. (1992). Predictors of success on NCLEX-RN and within the nursing curriculum: Implications for early intervention. Journal of Nursing Education, 31(2), 53-57. Retrieved from http://www.healio.com/journals/jne Giddens, J. (2009). Changing paradigms and challenging assumptions: Redefining quality and NCLEX-RN pass rates. Journal of Nursing Education, 48(3), 123-124. doi:10.3928/01484834-20090301-04 75 Grossbach, A., & Kuncel, N. R. (2011). The predictive validity of nursing admission measures for performance on the National Council Licensure Exam: A metaanalysis. Journal of Professional Nursing, 27, 124-128. doi: 10.1016/j.profnurs.2010.09.010 Halstead, J. A. (2013). The NLN’s fair testing imperative and implications for faculty development. Nursing Education Perspectives, 34(2), 72. Retrieved August 7, 2013 from http://www.nln.org/nlnjournal Harding, M. (2010). Predictability associated with exit examinations: A literature review. Journal of Nursing Education, 49(9), 493-497. doi:10.3928/01484834-2010073001 Harding, M., Rateau, M., & Heise, J. L. (2011). Efficacy of a midcurricular examination for predicting nursing student academic success. CIN: Computers, Informatics, Nursing, 29(10), 593-598. doi:10.1097/NCN.0b013e3182066458 Health Resources and Health Administration. (2010). HRSA study finds nursing workforce is growing and more diverse. Retrieved from http://bhpr.hrsa.gov/healthworkforce/rnsurveys/rnsurveyinitial2008.pdf Kane, R. L., Shamliyan, T. A., Mueller, C., Duval, S., & Wilt, T. J. (2007). The association of registered nurse staffing levels and patient outcomes: Systematic review and meta-analysis. Medical Care, 45, 1195-1204. Retrieved July 24, 2011, from http://journals.lww.com/lww-medicalcare Kaplan. (n.d.). Kaplan-LWW NCLEX-RN Integrated Testing. Retrieved from http://www.kaptest.com/NCLEX/Deans-and-Directors/RN-Programs/rntesting.html 76 Kleber, M. A. (2010). Determination of variables which predict success on the National Council Licensure Examination (NCLEX-PN) (Doctoral dissertation). Available from ProQuest Dissertations and Theses database. (UMI No. 3451104) Landry, L. G., Davis, H., Alameida, M. D., Prive, A., & Renwanz-Boyle, A. (2010). Predictors of NCLEX-RN success across 3 prelicensure program types. Nurse Educator, 35(6), 259-263. doi:10.1097/NNE.0b013e3181f7f1c9 Lavandera, R., Whallen, D. M., Perkel, L. K., Hackett, V., Molnar, D., Steffey, C., . . . Harris, J. (2011). Value-added of HESI exam as a predictor of timely first-time RN licensure. International Journal of Nursing Education Scholarship, 8, 1-12. doi:10.2202/1548-923X.2152 LoBiondo-Wood, G., & Haber, J. (2008). Nursing research: Methods and critical appraisal for evidence-based practice. St. Louis, MO: Mosby. Madaus, G., & Russell, M. (2009). The paradoxes of high stakes testing: How they affect students, their parents, teachers, principals, schools, and society. Charlotte, NC: Information Age. Matassarin-Jacobs, E. (1989). The nursing licensure process and the NCLEX-RN. Nurse Educator, 14(6), 32-35. Retrieved from http://journals.lww.com/nurseeducatoronline/pages/default.aspx McGahee, T. W. (2010). NCLEX-RN success: Are there predictors? Southern Online Journal of Nursing Research, 10(4), Article 13. Retrieved from http://snrs.org/publications/SOJNR_articles2/Vol10Num04Art13.html Morris, T., & Hancock, D. (2008). Program exit examinations in nursing education: Using a value added assessment as a measure of the impact of a new curriculum. 77 Educational Research Quarterly, 32(2), 19-29. Retrieved from http://www.erquality.org Morrison, S. (2005). Improving NCLEX-RN pass rates through internal and external curriculum evaluation. In M. H. Oermann & K. Heinrich (Eds.), Annual review of nursing education (Chap. 5, pp. 77-94). New York, NY: Springer. Murray, K., Merriman, C., & Adamson, C. (2008). Use of the HESI admission assessment exam to predict student success. CIN: Computers Informatics Nursing, 26(3), 167-172. Retrieved from http://www.nursingcenter.com/library/JournalArticle.asp?Article_ID=818352 Nasstrom, G. (2009). Interpretation of standards with Bloom’s revised taxonomy: A comparison of teachers and assessment experts. International Journal of Research and Method in Education, 32, 39-51. doi:10.1080/17437270902749262 National Council of State Boards of Nursing. (2011). 2011 NCLEX Examination Candidate Bulletin. Retrieved from https://www.ncsbn.org/11_NCLEX_Candidate_Bulletin.pdf.pdf National Council of State Boards of Nursing. (2012). 2013 NCLEX Examination Test Plan. Retrieved from http://www/ncsbn.org National Council of State Boards of Nursing. (2013). Test development webinar. Retrieved from https://www.ncsbn.org/4282.htm National League for Nursing Board of Governors. (2012). The fair testing imperative in nursing education. Retrieved from http://www.nln.org/aboutnln/livingdocuments/pdf/nlnvision_4.pdf 78 Neuderth, S., Jabs, B., & Schmidtke, A. (2009). Strategies for reducing test anxiety and optimizing exam preparation in German university students: A preventionoriented pilot project of the University of Wurzburg. Journal of Neural Transmission, 116, 785-790. doi: 10.1007/s00702-008-0123-7 Neuman, W. L. (2003). Social research methods: Qualitative and quantitative approaches (5th ed.). Boston, MA: Allyn and Bacon. Nibert, A. T., Young, A., & Adamson, C. (2006). Predicting NCLEX success with the HESI Exit Exam: Fourth annual validity study. Computers, Informatics, Nursing: CIN, 20(6), 261-267. doi: 10.1097/01.NCN.0000336439.16918.8b Nichols, S. L., & Berliner, D. C. (2008). Why has high-stakes testing so easily slipped into contemporary American life? Education Digest, 74(4), 41-47. Retrieved from http://www.eddigest.com/ Orman, J. M., & Guarneri, C. E. (2009). United States population projections: 2000 to 2050. Retrieved from http://www.census.gov/population/www/projections/analytical-document09.pdf Pishghadam, R., & Khosropanah, F. (2011). Predictive validity of the English Language Teacher Competency Test. International Journal of Education, 3, 1-19. Retrieved from http://profdoc.um.ac/ir/articles/a/1022785.pdf Polit, D. F., & Beck, C. T. (2008). Nursing research: Generating and assessing evidence for nursing practice (8th ed.). Philadelphia, PA: Lippincott Williams & Wilkins. ProQuest. (2012). Library types. Retrieved from http://www.ProQuest.com Roa, M., Shipman, D., Hooten, J., & Carter, M. (2011). The costs of NCLEX-RN failure. Nurse Education Today, 31, 373-377. doi: 10.1016/j.nedt.2010.07.009 79 Ross, S. J., Polsky, D., & Sochalski, J. (2005). Nursing shortages and international nurse migration. International Nursing Review, 52(4), 253-262. Retrieved July 24, 2011 from http://onlinelibrary.wiley.com/journal/10.1111/%28ISSN%2914667657 Simon, E. B., McGinniss, S. P., & Krauss, B.J. (2013). Predictor variables for NCLEXRN readiness exam performance. Nursing Education Perspectives, 34, 18-24. Retrieved August 7, 2013 from http://www.nln.org/nlnjournal Spurlock, D. R., & Hanks, C. (2004). Establishing progression policies with the HESI Exit Examination: A review of the evidence. Journal of Nursing Education, 43(12), 539-545. Retrieved July 24, 2011 from http://www.healio.com/journals/jne Spurlock, D. R., & Hunt, L. A. (2008). A study of the usefulness of the HESI Exit Exam in predicting NCLEX-RN failure. Journal of Nursing Education, 47(4), 157-166. Retrieved July 24, 2011 from http://www.healio.com/journals/jne Spurlock, D. (2013). The promise and peril of high stakes tests in nursing education. Journal of Nursing Regulation, 4, 4-8. Retrieved June 8, 2013 from http://jnr.metapress.com/content/u3uk17m21h014161/ Tevington, P. (2011). Mandatory nurse-patient ratios. MEDSURG Nursing, 20(5), 265268. Retrieved February 27, 2012 from http://www.medsurgnursing.net Tipton, P., Pulliam, M., Beckworth, C., Illich, P., Griffin, R., & Tibbitt, A. (2008). Predictors of associate degree nursing students’ success. Southern Online Journal 80 of Nursing Research, 8, Article 2. Retrieved from http://snrs.org/publications/SOJNR_articles2/Vol08Num01Art02.html Ukpabi, C. V. (2008). Predictors of successful nursing education outcomes: A study of the North Carolina Central University’s nursing program. Educational Research Quarterly, 32(2), 30-40. Retrieved July 24, 2011 from http://eric.ed.gov/?id=EJ847443 U.S. Census Bureau. (2006). Selected characteristics of baby boomers 42 to 60 years old in 2006. Retrieved from http://www.census.gov/population/www/socdemo/age /2006%20Baby%20Boomers.pdf Uyehara, J., Magnussen, L., Itano, J., & Zhang, S. (2007). Facilitating program and NCLEX-RN success in a generic BSN program. Nursing Forum, 42, 31-38. Retrieved July 24, 2011 from http://onlinelibrary.wiley.com/journal/10.1111/%28ISSN%291744-6198 Valiga, T. M., & Ironside, P. M. (2012). Crafting a national agenda for nursing education research. Journal of Nursing Education, 51, 3, 6. doi:10.3928/0148483420111213-01 Washburn, G. (1980). Relationship of achievement tests scores and state board performance in a diploma nursing program. Doctoral dissertation, Indiana University at South Bend. Yett, D. E. (1975). An economic analysis of the nurse shortage. Lanham, MD: Lexington Books. 81 Zweighaft, E. L. (2011). The impact of the Elsevier/HESI specialty exams on student performance on the National Council Licensing Exam for registered nurses (Doctoral dissertation). Available from ProQuest database. (UMI No. 3459906) 82 Appendix A Data Collection Tool 83 Appendix B Permission to Use Premises Form 84 Appendix C Premises, Recruitment, and Name Use Permission 85 Author Biography Lawrence Santiago has been a registered nurse for 16 years. He has been a nursing instructor for nine years, including four years of teaching undergraduate nursing students. He has also taught BSN and MSN students for University of Phoenix for five years. Lawrence obtained his Bachelor of Science in Nursing degree from Azusa Pacific University. He received a Master of Science in Nursing degree from Cal State University Dominguez Hills. He has completed an Ed.D. program at University of Phoenix. He is currently an Education Program Coordinator at Cedars-Sinai Medical Center in Los Angeles. He has been married for ten years and they have an eight year old son. 86