NURSING ADMISSION PRACTICES TO DISCERN “FIT”: A CASE STUDY EXEMPLAR

NURSING ADMISSION PRACTICES TO DISCERN “FIT”: A CASE STUDY EXEMPLAR by JAIME M. SINUTKO

A dissertation submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY IN EDUCATION: EDUCATIONAL LEADERSHIP

2014

Oakland University Rochester, Michigan

APPROVED BY: ___________________________________________ Dawn M. Pickard, Ph.D., Chair Date ___________________________________________ Kerri Schuiling, Ph.D. Date ___________________________________________ Kristina Shiree Aaron, Ph.D. Date ___________________________________________ C. Suzanne Klein, Ph.D. Date

© by Jaime M. Sinutko, 2014 All rights reserved

ii

This dissertation is dedicated to my family. My husband and children patiently allowed me space to read, think, and write. My parents consistently asked about the dissertation process with actual curiosity and even attempted to read drafts.

Thank you all for your love and understanding.

iii

ACKNOWLEDGMENTS

I owe my gratitude to Dr. Sandra Packard who initially sparked my interest in educational leadership. Her powerful stories will always stay with me. I also gratefully acknowledge Dr. Julia Smith and Dr. Jana Nidiffer’s positive influences. They faithfully supported my educational goal in a no-nonsense manner. Thankfully, my dissertation committee championed my research efforts. Dr. Dawn Pickard, Dr. Kerri Schuiling, Dr. Suzanne Klein, and Dr. Kristina Aaron nudged me along to improve my work. Behind the scenes, many others tirelessly listened to my research findings to help me work through the process. I am grateful to Laura Bochenek Klein, Dr. Gordon MacKinnon, Dr. Brian Stogner, Dr. Debi Hoggatt, Dr. Chris Pacini, and Dr. Patrick Rombalski. I feel utterly blessed to be writing an acknowledgment section and humbled that you all walked this path with me. I will be sure to take your wisdom with me as I continue to work with future college students.

Jaime M. Sinutko

iv

ABSTRACT

NURSING ADMISSION PRACTICES TO DISCERN “FIT”: A CASE STUDY EXEMPLAR by Jaime M. Sinutko Adviser: Dawn M. Pickard, Ph.D. Admission to a baccalaureate nursing school in the United States is currently a challenging proposition for a variety of reasons. This research explored a holistic nursing school admission process at a small, private, baccalaureate college using a retrospective, mixed-method, approach. The holistic method included multiple admission criteria, both qualitative and quantitative student data, in an attempt to view the whole applicant. Sixtyeight nursing students admitted through the 2011-2013 admission cycles comprised the study population. Holland’s Person-Job Fit Theory (1992) framework was applied to the student’s holistic data, which included college grade point averages, essays, and reviewer notes from group interviews. The findings revealed that, for the study population, interviews were the best predictor of semester-to-semester nursing student retention. A statistical relationship was not found when the college admission grade point averages were compared to the nursing student retention. Likewise, the admission essays, rescored seeking fit according to Holland, were not predictive of actual retention in the program.

v

TABLE OF CONTENTS

ACKNOWLEDGMENTS

iv

ABSTRACT

v

LIST OF TABLES

ix

LIST OF FIGURES

x

CHAPTER ONE INTRODUCTION

1

Personal Anecdote

1

Background Research

2

Definitions

5

Research Questions

7

Significance

8

Conclusion

8

CHAPTER TWO LITERATURE REVIEW AND THEORETICAL FRAMEWORK

10

Introduction to the History of Nursing Program Admissions

10

Admission Practices Unique to Nursing Programs

13

Traditional Nursing School Admission

15

Alternative Nursing School Admission Practice

22

Theoretical Framework

28

Holland’s Person-Job Fit Theory Conclusion

28 33

vi

TABLE OF CONTENTS – Continued

CHAPTER THREE METHODOLOGY

35

Research Questions

36

Participants

37

Student Demographics

38

Case Study Method

39

Case Study Qualitative Data Collection

41

Case Study Quantitative Data Collection

48

Case Study Data Strategy

49

Projected Findings

50

Conclusion

50

CHAPTER FOUR DATA ANALYSIS

52

Data Analysis Procedure

52

Quantitative Data Analysis

53

Qualitative Data Analysis

56

Aggregated Analysis

57

Individual Analysis

58

Inter-Rater Reliability

62

Summary

64

vii

TABLE OF CONTENTS – Continued

CHAPTER FIVE FINDINGS AND CONCLUSIONS

66

Overview

66

Researched Questions

66

Summary of Findings

67

Limitations

68

Future Research

69

Conclusion

71

APPENDICES A. IRB Approval

73

B. Re-Scored Interview and Essay Data Sheet

75

C. Inter-Rater Reliability Scoring Sheet

78

REFERENCES

79

viii

LIST OF TABLES

Table 2.3

Holland’s Person-Job Fit Theory

30

Table 3.2

Nursing School Admission Variables

42

Table 3.3

Group Interview Questions

46

Table 3.4

Improved Admission Variables

48

Table 4.1

Sample Essay Excerpts

59

Table 4.2

Sample Interview Note Excerpts

61

Table 4.3

Inter-Rater Reliability Session Scores

63

ix

LIST OF FIGURES

Figure 2.1

U. S. Historical Nursing Admission

11

Figure 2.2

Common U. S. Admission Practices Based on Credentials

17

Figure 3.1

Baccalaureate Nursing Student Diversity

40

x

CHAPTER ONE INTRODUCTION

Personal Anecdote I was admitted to a baccalaureate nursing school during an era when applicants with minimal entry-level qualifications were accepted. All applicants with a pre-nursing college grade point average (GPA) above 2.7 were admitted. Despite my lack of a higher early college GPA, I was a successful nursing student and became a professional registered nurse (RN) with bachelor’s degree in the science of nursing (BSN). I enjoyed working collaboratively, thinking critically, and caring for my patients. My nursing career began with direct patient care in a hospital; however, it soon shifted to the nursing education sector within my first five years as an RN. Within the past ten years I have had the opportunity to educate future nurses and observe the changes that are occurring in the admission process. The major change in nursing school admission, I experienced, was increased competition; it seems that it is becoming harder and taking longer to gain entry. In 2010, I became the inaugural director for a new baccalaureate nursing school. A small team (n=3) of nursing professors and I were tasked to create a nursing school admission policy within four months. I began with a literature review, which identified some contradictory results, which led me to the conclusion that the nursing profession has not been able to agree upon the practices leading to the admission of nursing students 1

who stay in college. However, using what we found, my colleagues and I created an admission policy that sought more than just the admission of high academic achievers into our program. We concluded that RNs should be more than academically intelligent; we wanted our graduates to be caring and strong communicators. We created, what I called, a holistic admission process that used admission essays, group interviews, coupled with traditional GPA criteria. After three years of using the holistic admission process (2011-2013), the time came to evaluate and assess if the process could be improved given issues such as students struggling to stay in school, the fact that the holistic admission process was timeconsuming for everyone involved, and it was not grounded in a theory. Theoretical frameworks can help explain complex problems and offer an alternative way of viewing a situation. Viewing the admission process differently might be able to improve the retention rate. It seems appropriate to apply a theoretical framework when seeking the outcome of increased student retention. This research investigated the holistic admission process utilized for three years at one small, private, baccalaureate nursing school.

Background Research In a health care profession such as nursing, “the single greatest hurdle in terms of likelihood of success is, overwhelmingly, admission” (Reiter, Salvatore, Rosenfeld, Trinh, & Eva, 2006, p. 40). Undergraduate college students, who wish to pursue nursing, face a challenge when their chosen program has limited access. “Preliminary AACN data show that 52,212 qualified applicants were turned away from 566 entry-level baccalaureate nursing programs in 2012” (American Association of Colleges of Nursing, 2

2013, p.2). Many students have determined their own personal preference for the nursing profession, however; they might not be able to gain admission into an accredited nursing school. To complicate the limited access, research shows that U. S. nursing schools lose approximately 24% of nursing students after being admitted and beginning coursework, but before graduation (Kovner & Djukic, 2009). In 2012, 174,644 students were enrolled in bacclaureatte nursing education in the U. S. (AACN, 2013). These statistics imply that 41,915 non-admitted, but qualified applicants could have taken the place of those students who did not remain in their nursing programs. Therefore, research within this purview might assist U. S. undergraduate nursing schools in making better admission decisions. One admission change may effect the retention probibility of students who enter baccalaureate programs and thus more nursing degrees could be earned. Kovner and Djukic (2009) analyzed student attrition from U. S. undergraduate nursing programs and concluded that admitting qualified applicants is among the most challenging endeavors since precise data are difficult to obtain. Admitting students who will graduate and succeed is the ultimate goal, and achieving that goal is both a social and intellectual process (Cortes, 2013). Since the research is conflicted, with respect to admission practices, leading to retention, it seemed appropriate to consider alternative approaches to the admission process such as application of Holland’s Person-Job Fit Theory (1992). Holland identified that satisfaction and productivity can be improved when an employee is matched with a congruent work environment. Utilizing this theoretical framework requires a holistic admission process seeking applicant fit characteristics and not merely admitting students with one quantitative indicator, such as 3

GPA. This retrospective research study applied Holland’s theory to the admission process that already occurred at a small, private baccaluarette nursing school and attempted to discover if the application process can be improved to better predict student retention so nursing students have the opportunity to take the national board exam (NCLEX) to become an RN. National Council Licensure Exam. The United States employs a computerized national proficiency exit exam for licensed registered nurses (RN) called the NCLEX (National Council Licensure Exam). Graduates from state approved nursing schools, whether a diploma nursing program, associate degree program (ADN), or baccalaureate degree program (BSN), are eligible to take the exam. The NCLEX is administered on a computer in an adaptive testing format that mimics a realistic professional environment, complete with audio and visual questions (Catalano, 2012; Zara, 1999). The technology began in the 1970s, when the U. S. army began administering computerized exams, and was previously shaped by educational and psychological measurement theories from the late 1950s (Zara, 1999). The NCLEX has stakeholder buy-in and serves the purpose of protecting the public from potentially unsafe RNs (Catalano, 2012; Zara, 1999). Although a rigorous RN licensure process is in place, the profession does not have a standardized process to admit nursing students.

4

Definitions For the purpose of this research, the following definitions apply. Nursing school was defined as a professional baccalaureate science degree in the United States. The researched school was a small private college. An applicant was defined as a student who was applying to a nursing school. High-stakes testing was defined as an assessment of student performance that affected an admission decision (Spurlock, 2006). Attributes were defined as applicants’ character traits, i.e. social. Caring attributes may also be assessed, and were defined as commitment to care for self, allowing for the capacity to care for others (Watson, 2007). Holistic was defined as a view of a whole person or the entire picture of character traits that form behaviors and influence actions. A holistic admission process includes more than one piece of admission criteria and usually involves interaction with the applicant, possibly in the form of an interview. Fit was defined as a proper match when a student’s personality matches the environment or academic program (Holland, 1992). O*NET stands for Occupation Information Network and is an online database. Retention is the ability or capacity of an individual student to progress, stay in school, and return for a subsequent semester. Success, for the purpose of this study, was defined as retention at a small private nursing school. Year to year, improving student retention increases graduation rates and could place additional nurses in the workforce. 5

O*NET (2103) provides the following interest codes and workplace value definitions that apply to registered nurses. Social: Social occupations frequently involve working with, communicating with, and teaching people. These occupations often involve helping or providing service to others. Investigative: Investigative occupations frequently involve working with ideas, and require an extensive amount of thinking. These occupations can involve searching for facts and figuring out problems mentally. Conventional: Conventional occupations frequently involve following set procedures and routines. These occupations can include working with data and details more than with ideas. Usually there is a clear line of authority to follow. Relationships: Occupations that satisfy this work value allow employees to provide service to others and work with co-workers in a friendly non-competitive environment. Support: Occupations that satisfy this work value offer supportive management that stands behind employees. Achievement: Occupations that satisfy this work value are results oriented and allow employees to use their strongest abilities, giving them a feeling of accomplishment.

6

Research Questions This dissertation sought to answer the following research questions: 1. To what extent does an applicant's college GPA, in their pre-nursing courses for their major, predict semester-to-semester retention at a small, private baccalaureate nursing school? 2. If student fit, as articulated by Holland’s Person-Job Fit Theory (1992), based on registered nurse interest codes and common workplace values in O*NET and, is assessed during the baccalaureate nursing school admission process, will retention improve? 2a. To what extent does an applicant's admission essays align with registered nurse’s work values from Holland (1992) as applied by the O*NET, if nursing school fit is established, is semester-to-semester retention greater, based upon these fit characteristics? 2b. During an admission interview, if an applicant describes his or her interests to be congruent with registered nurse’s interest codes from Holland (1992), as applied by the O*NET, will nursing school fit be established and lead to improved semester-to-semester retention? In summary, this retrospective case study required quantitative and qualitative student data anaysis and applied Holland’s fit theory in the attempt to better predict nursing student semester-to-semester retention.

7

Significance This study was conducted in a retrospective manner by analyzing past applicant’s admission packets from 2011-2013. The application of a new theoretical framework, from Holland (1992), to an existing set of holistic admission student data was utilized so that no students were placed in jeopardy due to the new admission criteria. The American Educational Research Association (AERA) (1999) classified admission decisions as highly impactful for the student and the institution. This dissertation research has the potential to possibly increase student retention in similar nursing schools. If Holland’s Person-Job Fit Theory (1992) has a place in admission decisions, and is a predictive measure of student retention, it may increase graduation rates of nurses entering the profession or workforce. This study has possible positive impacts but there was no direct benefit to the students who were studied in this research because they have already completed the nursing school admission process. Increasing future retention rates is important to many institutions and affected students, due to factors such as the financial burden and lost time when students change majors or are not successful in the field.

Conclusion It is widely recognized that admission to U. S. colleges and universities is traditionally based on tests and past academic records (AERA, 1999). These admission practices may not be optimal for nursing retention. Based on the theoretical and literature research, a retrospective, mixed-method case study methodology was employed. One small, private baccalaureate nursing school’s admission practices were examined in this 8

study in order to better understand the application of the principles of Holland’s PersonJob Fit Theory (1992) as it applied to nursing school, with a goal of semester-to-semester student retention.

9

CHAPTER TWO LITERATURE REVIEW AND THEORETICAL FRAMEWORK

Introduction to the History of Nursing Program Admissions Nursing school admission practices vary significantly and tend to be influenced by current trends of supply and demand (Catalano, 2012; Porter & Bean, 2004). There are currently (2014) three different routes in the U. S. to become a registered nurse (RN) with the length of time in the program lasting up to five years. The final degree earned can be a diploma, associate’s degree, or a bachelor’s degree in the science of nursing. Upon degree completion, the graduate takes the national board exam (NCLEX) to demonstrate competency leading to the title RN. An RN title does not indicate an educational degree. Educational path and board exams are inconsequential if an applicant is presented with a challenging admission process and cannot gain access. Two trends influencing present (2014) admission practices are the U. S. nursing shortage of the 1960s and the increased demand for registered nurses in the 1990s (Catalano, 2012). The increased demand for nurses in the 1990s was attributed to changes in health care, increased population growth, and the advancing average age of U. S. citizens (Caron, 2004; Catalano, 2012). Concomitantly, the level of individual interest in the nursing profession has fluctuated. These conditions have resulted in competitive practices of nursing admissions (Caron, 2004). The current (2014) demand for nurses remains high yet space seems to be limited (Figure 2.1). Space in nursing school can be determined by the 10

1990s Increased demand for RNs despite low interest in nursing school. 1960s Shortage of trained nurses. BSN higher education is widely available now.

Admission was open to those interested as admission was not stringent.

Rolling admission practice where all qualified are admitted based on basic admission criteria. Some were admitted into future cohorts.

2010s Increased demand for nursing school causing a shortage of available nursing school seats.

Competitive or selective admission policies. More and more qualified applicants cannot gain access to nursing school.

Figure 2.1. U. S. Historical Nursing Admission (Adapted from Caron, 2004; Catalano, 2012)

11

availability of hospital-based learning opportunities and qualified Master’s-prepared professors who have a nursing license and experience in the necessary specialty. The local hospitals offer the required hands-on clinical learning opportunities for nursing students and may not have hospital units available or possibly are not willing to accept nursing student groups. In a local nursing program, random sampling of ten Midwest nursing school’s admission practices, seven represented the baccalaureate programs, described earlier, and three are associate nursing degree programs. It was found that none of the nursing schools within a 50 mile radius of the program studied for this research currently use an admission interview. Three, all of which were baccalaureate, include an admission essay along with other criteria such as college grade point average (GPA). The majority (n=7) currently (2013) admit nursing students based on GPA or a combination of GPA and an entrance exam only. The ten nursing program’s websites describe a competitive admission era where more nursing students are seeking entrance than availability allows. In the program admission process, “rolling” admission practice means that all applicants who met the set admission criteria were admitted to nursing school. During approximately twenty years (1980-2000), this admission practice was used since classes were not being filled. During the “rolling” admission era, students could be admitted to a nursing cohort that was more than a year in the future. An applicant in 1999 may have met the basic admission criteria in 1999 and then subsequently admitted into a cohort to begin nursing course work in 2001 attributed to limited space. Space in nursing school is dependent upon hospital-based clinical availability and available professors. Nonetheless, 12

the “rolling” admission process was not a competitive or highly selective since all qualified applicants ultimately were allowed to enter the program.

Admission Practices Unique to Nursing Programs Admission practices can vary between healthcare disciplines. This research focused on nursing programs as a unique discipline. A search for relevant research was conducted initially via the CINAHL (Cumulative Index to Nursing and Allied Health Literature) database and was accessed through the University’s library website. The search terms used were “nursing” + “school” AND “admission”, which produced twenty articles published between 1980 and the present (2014). Approximately 75% of these articles focused on non-English speaking students or students with disabilities, or targeted the vocational level of nursing, such as licensed practical nurses (LPN). A subsequent search of PsycInfo and the ERIC databases used the same search terms of “nursing school” AND “admission” and revealed additional articles published between 1988 and the present, two of which pertained to the educational arena. Final searches of “nursing” + “education” AND “admission” were conducted on the One Search database, and resulted in the 28 studies which were reviewed in this analysis. The question exists: do admission policies and practices actually play a role in determining student success in the form of retention? The ability to predict student retention rates through admission policies is uncertain, given that multiple variables contribute to a student’s success in nursing and college (Williams, 2010). Some variables, such as financial instability or personal illness, emerge after admission decisions have been made. International research has identified other issues in admission practice 13

decisions regarding barriers to student success such as poor social support and inadequate knowledge regarding their chosen profession (Al-Kandari & Ajao, 1998). It appears that collecting information regarding the characteristics of a successful nursing student might have the potential to influence admission outcomes. Researchers attempt to better understand the characteristics of successful nursing students. Beeson and Kissling’s (2001) study of 505 nursing students’ records and scores showed that students’ first attempt at passing the national licensure exam for Registered Nurses in the U. S. (National Council Licensure Examination or NCLEX) is a reliable academic predictor of nursing school success. According to Beeson and Kissling (2001), academic predictors are the norm because “few researchers have looked at nonacademic predictors of success” (p. 122). Their study revealed that “students who performed well reported fewer family responsibilities” and that the students most likely to pass the NCLEX on the first attempt were older men who had already obtained a bachelor’s degree in another field (Beeson & Kissling, 2001, p. 122). In contrast, Timer and Clauson (2010) established that older nursing students had lower grade point averages (GPA). Timer and Clauson also provided the link that “passing grades” in nursing school resulted in the increased likelihood of passing the final nursing licensure exam (NCLEX). Beeson and Kissling concluded their study with a call for additional research to identify other variables, which may predict nursing success, since student selection has the potential to affect retention.

14

Traditional Nursing School Admission There are three routes in the U. S. to become a registered nurse. Nursing school can vary from one to five years in length. The resultant degree can be a diploma, associate’s degree, or a bachelor’s degree in the science of nursing. Upon degree completion, the graduate takes the national board exam to become an RN. An RN title does not indicate an education degree, as there is much variation. There are no educational programs that graduate RNs due to the fact that it is a license obtained after a board exam. Across programs, the most commonly used admission criteria are high school or early college GPA, possibly combined with high-stakes testing. Grade point average. GPA is the most commonly used admission practice for undergraduate nursing schools (Carpio & Brown, 1993; Codier & Odell, 2013; JonesSchenk & Harper, 2013). Beale and McCutcheon (1980) found that diploma programs weigh quantitative data, such as GPA and high-stakes testing, more heavily for admission than do the associate degree in nursing (ADN) or bachelor of science in nursing (BSN) programs. Regardless of the pre-licensure academic preparation, including diploma through baccalaureate prepared, all state approved registered nursing school’s graduates sit for the same national competency exam (NCLEX) to become a registered nurse. There are very few diploma schools in existence in the U. S. The primary reason for this is that employers are demanding higher education for their registered nurses and a diploma program is traditionally hospital based and not degree granting. In fact, hospital-based employers are showing signs of preferential hiring for bachelor’s degree (BSN) graduating RNs (AACN, 2013). 15

In summary, admission practices in the U. S. can differ, based on graduating credentials, as depicted in Figure 2.2, where areas of similarity overlap, such as highstakes testing and college GPA. The researched small, private BSN program admitted students with all three common practices; college GPA, essays, and interviews. In contrast to the United States, Australian, and Czechoslovakian higher education institutions tend to rely on a high-school senior’s grades to blindly rank graduates for all college admission (Jurajda & Münich, 2010; Pascoe, 1999). Admission teams deemed these blind rankings as objective data, equivalent to the academic rigor needed to succeed in nursing school (Knauss & Willson, 2013). Beyond objective data some programs explore students’ personal or professional experiences or background characteristics during the admission period. Timer and Clauson (2010) found that student background information was relevant to academic success. Their findings were similar to Newton and Moore’s (2006) study of admission essays, which showed non-GPA admission criteria were valuable. Timer and Clauson (2010) also reported that many Canadian schools use GPA as the only admission criterion, due to the assumption that it is deemed quick and fair. Timer and Clauson (2010) studied an accelerated baccalaureate nursing school and hypothesized that “certain clinical aptitudes (i.e. motivation, empathy, and communication skills) are as necessary in becoming a professional nurse as is academic performance” (p. 601). Timer and Clauson (2010) assessed clinical aptitude through the student’s application package (résumé, personal statement, letters of recommendation, and an interview) and found a statistically

16

Diploma Nurse

High School Grades

Associate Degree Nurse (ADN)

High Stakes Testing

ACT Score

College GPA

Interview and Essay

Bachelor Degree Nurse (BSN) Figure 2.2 Common U. S. Admission Practices Based on Credentials.

17

significant correlation with older students from minority ethnic backgrounds who were more well-rounded, but who had lower nursing school GPAs. The researchers rejected their hypothesis regarding clinical aptitudes and concluded that the non-academic benchmarks in their admission process were unreliable and that they did not relate to nursing school success; they concluded that the traditional measure of GPA alone was sufficient. In an effort to find predictors of success, other than testing, Wong and Wong (1999) compared various science courses and how the student’s GPA may be related to nursing programs for successful nursing students. Their study of 258 Canadian nursing students failed to demonstrate a significant relationship between high school and college science grades relating to a passing score on their final nursing licensure exam, which was the researchers’ proxy for academic success in nursing school. However, it is important to note that high school and college courses lack standardization and, therefore, may teach different topics. Similarly, in the U. S., Wolkowsitz and Kelley (2010) reviewed nursing student test scores (in science, math, reading, and English) to predict nursing school success. In a review of 4,105 nursing students’ scores from varying educational programs, they concluded that it was unclear whether reading or English scores were useful admission criteria. High-stakes testing. High-stakes testing is defined as an assessment of student performance that affects an admission decision (Spurlock, 2006). In a health care profession such as nursing, “the single greatest hurdle in terms of likelihood of success is, overwhelmingly, admission” (Reiter et al., 2006, p. 40). If high-stakes tests are used in 18

making admission decisions, there should be evidence that the resultant scores could be a reliable predictor of student success. Pascoe (1999) explained that high-stakes college admission practices are more commonly used in the U. S. than internationally. The American Educational Research Association (AERA) (1999) classified admission decisions as highly impactful for the student and the institution. Therefore, it was recommended that admission never be determined by one criterion, such as high-stake tests, since the validity of a decision increases when additional information is considered. Discipline-specific tests are used to determine admission in many nursing programs (AERA, 1999; Reiter et al., 2006). They are most commonly administered through nursing-specific multiple-choice question entrance exams (Tarrant, Knierim, Hayes, & Ware, 2006). An example of a nursing-specific test is the Health Education Systems, Inc. (HESI). HESI is a computer-based exam that tests math, vocabulary, reading comprehension, and grammar usage. Knauss and Willson (2013) found a correlation between such admission tests and the associate-degree (ADN) students’ grades in their first two semesters of nursing school. Lengacher and Keller (1990) also found that ADN nursing students’ final GPAs were similar to their pre-admission ACT (American College Testing) composite scores of math, reading, science, and English. ACT and SAT (Scholastic Assessment Test) test scores indicate mastery of memorized or low-level analytical skills, but not all the skills needed for college or professional success (Sternberg, 2010). A question that must be considered is whether high-stakes testing should be required for undergraduate baccalaureate (BSN) admission. 19

Tarrant et al. (2006) studied the efficacy of the multiple-choice questions used in nursing assessments and concluded that almost half of the questions contained itemwriting flaws. According to researchers, the questions had not been generated by nursing professionals and were written at too low a cognitive level. This raises the potential question of whether multiple-choice tests should be administered for nursing admission decisions. Negative consequences of testing. The validity of high-stakes testing is questionable when nursing educators consider the potential negative consequences and not the potential benefits (Spurlock, 2006). For example, U. S. BSN program admission policies are biased in favor of English-speaking students without performance-inhibiting disabilities because they are often not available in other languages or accommodating forms (Spurlock, 2006). High-stakes tests can also negatively affect English-as-secondlanguage students. ADN and vocational nursing schools that do not use such tests are more effective at recruiting linguistically diverse students than BSN programs that do not require high-stakes admission testing (Brown & Marshall, 2008). According to Cortes, “test-oriented admission practices…only increase the obstacles to access without improving an institution’s capacity to predict student success” (2013, p. 61). If a minority nursing school applicant is seeking a BSN education, he/she may face a potentially biased high-stakes test to gain access. “Attempts to recruit, retain, and graduate a diverse cadre of professional nurses remain a goal for nurse educators and nursing programs” (Melillo, Dowling, Abdallah, Findeisen, & Knight, 2013, p. 100). The research seems clear: if

20

program diversity is valued, high-stakes tests should not be critical admission criteria (Adnett, 2011; Spurlock, 2006). Beyond the diversity concerns, a potential for inaccurate tests occurs when written by new nursing faculty. New nursing faculty can be recruited based solely on their training in a health care specialty and not their abilities as educators. For example, new nursing faculty may lack sound test-writing knowledge (Tarrant et al., 2006). Reiter et al. (2006) discussed an added concern for potential security breaches, such as cheating, that exists when a test is utilized during a highly competitive admission process. Ultimately, high-stakes tests are challenging to construct, they need to be nondiscriminatory, and they must be safeguarded. When properly utilized, high-stakes tests must be a valid and reliable measure of what is desired (Huch, Leonard, & Gutsch, 1992). Wolkowsitz and Kelley (2010) studied scores from the Test of Essential Academic Skills (TEAS), a high-stakes test used for nursing school admission, which includes reading, math, science, and English. It was found that science scores were the strongest predictor of academic success in nursing school, at 3% of the variance per the R2 value, regardless of the program type or length. English, math, and reading scores were, in that order, less useful for predicting nursing student success. Unfortunately, employees of ATI, the test manufacturer, conducted the study and the data have not been confirmed by independent sources. High-stakes admission tests and early college GPA calculation is a common practice emerging from the literature (AERA, 1999; Reiter et al., 2006). Since descriptions of a successful nursing student differ, essays and personal interviews may 21

serve as a unique and effective admission practice it seems. Ultimately, high-stakes testing and college grades are not strong enough predictors of nursing student success to be relied upon exclusively.

Alternative Nursing School Admission Practice Nursing school program differences begin during the admission process because admittance policies and procedures vary among institutions (Gilmore, 2008). Admission to nursing school has historically gone unnoticed during times when supply and demand were congruent or when there was a smaller supply of nursing applicants compared to available space. During historical times of declining nursing applicants, such as the 1990s, for example, some schools revised their admission policies to become “rolling”, meaning all applicants who met the admission criteria were admitted (Newton, Smith, & Moore, 2007). Carpio and Brown (1993) explained that selective nursing admission processes occur when there are larger numbers of applicants or a lack of educational space. Carpio and Brown recognized that “as nursing education has evolved, academic requirements for admission have become progressively more stringent” (p.42). Currently popular (2014), a ranked or competitive admission strategy admits students based on college grades alone, meaning that a school of nursing chooses the applicants with the highest grades to fill the class (Codier & Odell, 2013; Jones-Schenk & Harper, 2013; Newton et al., 2007). It may be that understanding the history of admission practices can influence future admission practices (Figure 2.1). The question for nursing administration becomes how do nursing school admission committees measure the quality of a student’s ability to demonstrate or 22

develop the professional skills and qualities necessary to become a successful nurse through grade assessment alone. Nursing professionals believe that GPA alone cannot account for a successful nurse (Codier & Odell, 2013; Jones-Schenk & Harper, 2013; Leners, Beardslee, & Peters, 1996; McNelis et al., 2010). “The sole reliance on objective tests and past performance records cannot improve the student selection and performance in nursing programmes” (Ehrenfeld & Tabak, 2000, p.105). Yet prediction of nursing school success, as mentioned previously, is generally based on scholastic markers, such as GPA, or unreliable, standardized nursing tests (Leners, Beardslee, & Peters, 1996; Newton, Smith, Moore, & Magnan, 2007). For example, the majority (n=7) of a random sample (n=10) of Midwest schools were currently (2014) admitting nursing students with scholastic markers only. Attribute seeking. It is a common belief that student qualities or attributes cannot be completely taught during nursing school. Attributes were defined as applicants’ character traits, such as being organized or caring. Exploration of an applicant’s attributes may help admission committees better understand a person, and thus more reliably predict his or her future nursing school success (Huch, Leaonar, & Gutsch, 1992). Though not studied in the U. S., Ehrenfeld and Tabak (2000) concluded that admission interviews increased retention in nursing school by screening out applicants who displayed personal barriers to success. It was concluded that interviews were a viable method to identify applicant attributes. While nursing skill competence is mandated in nursing, the attribute of caring is also desirable. Caring attributes were defined as commitment to care for self, which 23

allows for the capacity to care for others (Watson, 2007). A nursing applicant should have foundational skills and personality traits that support nursing school completion and, in turn, a greater likelihood of success in the nursing profession (Huch, Leaonar, & Gutsch, 1992). Huch, et al. (1992) using a pre-admission personality screening, identified that successful nurses are independent, realistic, responsible, mature, and calm. Jones (1975) identified successful ADN students as having more than intellectual aptitude; the graduates were also mature, nurturing, and emotionally aware. Personal attributes can likely predict nursing student success, especially during times when the applicant pools are similar in terms of high academic marks (Carpio & Brown, 1993). However, it is still difficult for nursing admission committees to determine which applicant criteria are most important. Predicting retention. Nursing school success begins with a strong admission policy directed at retention. Al-Kandari and Ajao (1998) studied 54 Kuwait nursing students from a diploma, ADN, and a BSN school, to recommend admitting nursing students who demonstrated strong social support and prior awareness of the demands of the program and profession. Williams (2010) found similar results amongst ten BSN nursing students from the Midwestern United States with in-depth interviews. Melillo et al. (2013) found increased diversity with pre-admission essays and interviews, when seeking best practices for diverse BSN nursing students in the U. S. Carpio and Brown (1993) also encouraged admission interviews, in a semi-structured team approach, to discern the applicant’s awareness of the nursing program, future career aspirations, motivation, and critical thinking abilities. 24

Accelerated second-degree in nursing. Accelerated, second-degree BSN programs may attract a different student demographic than a traditional BSN program. Does it warrant a unique admission process that includes an interview or essay to discern readiness for the program’s inevitable demands? The development of an accelerated method to achieve a second bachelor’s degree in nursing was initially created to address the nursing shortage (Seldomridge & DiBartolo, 2005). The accelerated second-degree program can usually be completed in less than eighteen months and predominantly attracts students with a baccalaureate background in biology or psychology (Seldomridge & DiBartolo, 2005). Many accelerated second-degree programs have higher GPA standards and require additional pre-requisite coursework for admission consideration. Seldomridge and DiBartolo (2005) reported that graduates from such programs tended to be high academic achievers who reported high levels of stress that hindered retention. The reported retention challenges could lead accelerated admission committee members to seek students with strong support systems and financial stability, along with a solid academic background. Admission essays and interviews. The use of admission essays or student narratives seems to be a prominent theme in nursing admission research. It is a more prevalent admission practice for a BSN program than for an ADN or diploma program (Beale & McCutcheon, 1980). In some circumstances, qualitative or non-cognitive data have been collected along with quantitative indicators, such as GPA. This mixed methods approach was called holistic because it was more comprehensive, assigning value, and considering the whole person instead of just one part. A holistic admission includes 25

character traits that form behaviors and influence nursing practices (McNelis et al., 2010). The approach involves multiple sources and in-depth interpretation of these sources and is deemed ideal for undergraduate eligibility assessment (AERA, 1999). The holistic admission approach considers both qualitative and quantitative student data, and seems well suited to the study of nursing school admission practices. Sadler (2003) conducted a mixed methodology study (n=236 baccalaureate nursing students in Michigan) that recommended a review of admission essays. Newton and Moore (2006) and Katz, Chow, Motzer, and Woods (2009) also recommended the use of admission essays. Sadler (2003) asked applicants “to describe their understanding of professional nursing and to explain the ways in which their thinking has developed and/or changed during the past year” (p. 623). This essay question was relevant since the researchers used the essays for graduate school admission where the applicants were already nurses, similar to Newton and Moore (2006). Sadler’s research looked for a relationship among and between the essays’ score, the entering GPA, and whether the student had completed nursing school. Study results indicated that several noncompleters wrote about having no personal experience with nursing: “most students in the group of non-completers wrote about nursing as external to themselves, as something to ‘do’ rather than to ‘be’” (Sadler, 2003, p. 625). Sadler’s work affirmed the notion that nursing is more than a vocation; to many, it is a calling or way of life. GPA differences between those who successfully completed nursing school and those who dropped out were not statistically significant, however. While the study ultimately concluded that neither GPA nor admission essay scores could fully predict attrition, Sadler (2003) 26

concluded that new research methods are needed to establish a connection between students’ emotional intelligence and their subsequent school success. Newton and Moore (2006) performed a retrospective qualitative study (n=157) and reported the results in a mixed-method fashion on graduate level nursing students. The background information on admission practices suggested, “no consistent method or criteria was encountered and no empirical data were found to support one method or criteria over another” (Newton & Moore, p. 206). The authors found more students focused on short-term goals and did not follow the essay instructions. This finding led the faculty to revise the directions for subsequent applicants and concluded there was a significant relationship between the quality of the essay writing and the student’s final nursing course grade. Newton and Moore (2006) demonstrated a relationship between nursing school success and non-GPA admission criteria and encouraged faculty to “consider the written goal statements of graduate applicants to be at least as important as the more objective measures (e.g. GPA, GRE scores) for them to admit the best applicants and to facilitate program outcomes success” (p. 209). Katz et al. (2009) drew similar conclusions when they researched graduate nursing students’ written documents for admission. McNelis, et al. (2010) described the holistic admission process to admit BSN students at Indiana University. The admission interviews were deemed fair, successful, and a good opportunity to identify applicant attributes such as motivation, awareness, problem solving, and career goals (McNelis et al., 2010). Williams (2010) and Leners et al. (1996) also identified that a clearly written career path was instrumental for successful nursing 27

students and could be completed during the admission process. These research findings offer an opportunity to improve the admission process in future studies since the field is currently undecided about best practices. The nursing admission practice literature review revealed 28 relevant studies producing conflicting data and varied recommendations. Undergraduate admission use of both essays and interviews was identified as a gap in the literature. These findings influenced my research of a U. S. baccalaureate nursing school that utilized essays and interviews for admission. Theoretical Framework Holland’s Person-Job Fit Theory Fit theory can be found across theorists and throughout centuries as it spans vocational psychology and organizational theory. Holland’s (1992) theory is based on the work of Murray (1959) and the other researchers at the Harvard Psychological Clinic who explored vocational and personality attributes. Holland’s background in vocational counseling influenced his early work. An early tool that Holland developed was the Vocational Preference Inventory, which supported his position to categorize people by their personalities (1992). Holland began publishing in the 1950s and focused on the kind of work performed and not the possible levels or prestige of work. Holland advanced other researcher’s work of society behavior when he articulated the concept of assessing environments by characterizing the people in the environment, where the majority of the environmental characteristics are the result of people’s influence (1992). Holland continued to research personality types and was able to 28

characterize and calculate distributions that related to environmental preferences. Holland began his research with high-achieving youth. He then expanded to college students and surveyed working adult men and women. According to Holland, a person who fits in one’s environment, school, or profession can be described as productive, satisfied, and successful. This is important for the profession of nursing since job productivity and satisfaction can affect the work which in turns touches lives. His person-job fit theory can help researchers understand why some students thrive in a professional program, such as nursing, while others fail to thrive. This organizational theory considers both an individual’s personality and the environment, to determine whether there was a proper fit. Alignment with one’s personality type into the congruent work environment, as seen in Table 2.3, leads to positive outcomes (Holland, 1992). Fit theory characterizes people based on their alignment with six different personality types and environment categories of the same name: Realistic, Investigative, Artistic, Social, Enterprising, and Conventional (Holland, 1992). Holland concluded that people choose environments that allow them to express themselves and utilize their abilities. The pairing of personality and environment “leads to outcomes that we can predict and understand from our knowledge of the personality types and the environmental models” (Holland, 1992, p. 2). Some of these outcomes include career choice, achievements, continuity, and stability since “behavior is determined by interaction between personality and environment” (Holland, 1992, p. 3). The aforementioned outcomes are highly relevant to the work of an admission committee as students choose nursing and then the committee chooses them. 29

Application of Holland’s fit theory in O*NET. Holland’s work with identifying personality and environment types had broad application across professions (1992). For example, the National Center for O*NET Development (2013) is a web-based database for occupational information, such as worker attributes and job characteristics. The U. S. Department of Labor was involved in the development of O*NET. It is an online repository for a national data set of randomized survey data. The database assists job seekers by detailing the educational and skill requirements for employment in any given field. The database also assists researchers via the worker characteristic section, which can be filtered by specific job titles, such as registered nurse. This area of O*NET describes consistent worker abilities, occupational interests, work values, and work styles. The occupational interests section discerns a worker’s preference for work environment and is drawn directly from Holland’s (1992) personality types. Table 2.3 Holland’s (1992) Person-Job Fit Theory. Personality Types

Work Environments

Realistic

Realistic

Investigative

Investigative

Artistic

Artistic

Social

Social

Enterprising

Enterprising

Conventional

Conventional

30

Application of Holland’s fit theory in the nursing profession. Nurses share some common values and interests that shape the professional arena. The National Center for O*NET Development (2013) reports that registered nurses indicate primary occupational interests in the codes of social, investigative, and conventional fit characteristics. These data indicated that nurses prefer to be in groups and enjoy communicating with others. The dataset indicated that a nurse is also a critical thinker, responsible for solving problems. Finally, the data indicated that the profession of nursing remains conventional possibly due to of the hierarchical nature of hospitals that have set procedures and routines. Social, investigative, and conventional are the three interest codes reported in O*NET for a registered nurse and were researched through the interview notes in Chapter Four. O*NET further reports that registered nurses have a common set of work values such as relationships, support, and achievement. The three aforementioned work values were the values listed in the dataset and were researched via student admission essays in Chapter Four. Nurses value collegiality, service, supportive employers/companies, and view results, such as a patient’s return to health, with a sense of accomplishment. Using Holland’s (1992) fit theory, the interest codes (n=3) and work values (n=3), within the nursing profession, can be assessed during undergraduate nursing school admission using applicant essays and interviews. Holland’s application in the nursing admission arena has yet to be published thus opening the door for work in this area. A theoretically-driven admission process in a career field has the potential to seek

31

applicants who express the work value of social interaction, investigative thinking, fulfilling conventional roles, and who are achievement driven. Admission decisions, which are based on judgments of an individual’s written values or stated interests, are not without concerns or critics. Murray (1959) raised the concern that a candidate’s motives may be purposefully hidden. An applicant, who wishes to be perceived positively during the admission process, may present differently than his/her usual demeanor. The applicant may even choose to mislead the decisionmakers. Personalities can frequently vary and can also be quite situational in nature. Holland (1992) disclosed that person-job fit theory is a fulfillment model where the assumption is made that all people seek fulfillment. The theory assumes that people gain fulfillment through goal achievement. This is not a universally held sentiment perhaps since some people may not be goal-oriented. Therefore, caution should be taken when making broad inferences about personalities since it has not been fully researched. Holland’s Person-Job Fit Theory (1992) application is uncertain and warranting investigation. The chosen theoretical application was chosen due to the fact that nursing school was defined as a professional baccalaureate science degree within this research. However, nursing school retains a prominent vocational component since students participate in clinical immersion experiences during each semester of nursing school. During these clinical experiences, students attend a registered nurse’s job site and fulfill components of the role. Thus, it can be argued that the academic coursework in nursing school requires a vocational and professional fit as students take numerous classes throughout the curriculum, including clinical practice in a hospital setting. 32

According to Holland (1996), as stated previously, people thrive in an environment when there is a good fit between their own personality and the characteristics of the environment. Lack of academic effort may occur when a student is in a program that does not fit their personality. A registered nurse’s environment has been identified by O*NET, based on three out of the six possible common work values and three out of the six possible employee interest codes. A student applying for nursing school could potentially complete a personality assessment as part of an admission essay or interview to determine a degree of congruence. If nursing admission committees utilize nursing student personality factors in combination with academic benchmarks, such as college grades, could a successfully retained student gain admission? Personality and environmental congruence, as assessed during the admission process, can identify students who are likely to put forth the necessary effort to succeed.

Conclusion Variables such as nursing school curriculum and time-to-degree completion contribute to the current (2014) variability with respect to admission practices. Baccalaureate nursing (BSN) education can vary in length from one year in an accelerated format to five years in a more traditional format. BSN curriculum differs from associate degree nursing (ADN) coursework, since it usually contains additional liberal arts studies along with nursing theory, research, and leadership components. Personality traits may also differ in a nursing student who chooses an ADN program versus a BSN program. However, all graduates must prove minimal competency on the same licensure exam (NCLEX) to become a registered nurses (Catalano, 2012). The 33

variation in nursing education also occurs during the admission process as schools adopt unique practices to predict achievement. Nursing student achievement predictability, coupled with academic benchmark scores, is recommended for admitting the most suitable students (Shulruf, Wang, Zhao, & Baker, 2011). A common nursing admission practice of utilizing only quantitative data was a theme in the literature. Yet, a consensus has not been reached about the topic of optimal nursing admission criteria. The opportunity exists for future exploration at the baccalaureate nursing level. This study of U. S. baccalaureate nursing applicants applied Holland’s Person-Job Fit Theory (1992) during a mixed-method admission process to answer the overarching question: do admission practices play a role in determining student success in the form of semester-to-semester retention. Ehrenfeld and Tabak stated “the aim of nursing schools is to educate people for a profession in which personality and professional conduct play a central role, thus mandating more stringent admission criteria” (2000, p.105).

34

CHAPTER THREE METHODOLOGY

Nursing is described as both an art and a science. Both of these components can be a part of the nursing school admission process. A nurse must have competent skills, as well as the ability to think critically, care authentically, and communicate adequately. While early college grade point average is the most commonly used admission practice for baccalaureate nursing schools (Carpio & Brown, 1993; Codier & Odell, 2013; JonesSchenk & Harper, 2013), U. S. nursing schools lose about 24% of admitted nursing students after beginning coursework but before matriculation to graduation (Kovner & Djukic, 2009). It is also important to recognize the institutional and student repercussions, such as the financial burden and lost time for those who change majors (Borges, Gibson, & Karnani, 2005). In a complex profession such as nursing, adequate student fit is worth researching. Assessing qualitative and quantitative admission factors, to discern fit, is considered a mixed- methods approach and can be called holistic because it is more comprehensive (McNelis et al., 2010). The approach involves multiple sources and indepth interpretation of these sources and is deemed ideal for undergraduate eligibility assessment (AERA, 1999). The holistic admission approach seems well suited to the study of nursing school admission practices.

35

The college research location granted permission to access the data through an exempt institutional review board decision on March 11, 2014. The University’s institutional review board, likewise, determined that this dissertation research was exempt on April 16, 2014. See Appendix A for this document.

Research Questions This chapter presents the methodology, using multiple data collection strategies, to address the following research questions: 1. Quantitative: To what extent does an applicant's college GPA, in their prenursing courses for their major, predict semester-to-semester retention in a small, private baccalaureate nursing school? 2. Qualitative: If student fit, as articulated by Holland’s Person-Job Fit Theory (1992), based on registered nurse interest codes and common workplace values in O*NET and, is assessed during the baccalaureate nursing school admission process, will retention improve? 2a. To what extent does an applicant's admission essays align with registered nurse’s work values from Holland (1992) as applied by the O*NET, if nursing school fit is established, is semester-to-semester retention greater, based upon these fit characteristics? 2b. During an admission interview, if an applicant describes his or her interests to be congruent with registered nurse’s interest codes from Holland (1992) as applied by the O*NET, will nursing school fit be established and lead to improved semester-to-semester retention? 36

Qualitative and quantitative data (n=68) in the form of interviews, essays, and cognitive markers such as grade point average (GPA) were researched focusing on one small private baccalaureate nursing school’s admission process. The qualitative theme analysis occurred through the lens of Holland’s Person-Job Fit Theory (1992) and the quantitative GPA data underwent a logistic regression. These data findings contributed to the growing body of knowledge and attention in this field as described in the 28 literature review items.

Participants A sample (n = 60 retained and n=8 non-retained) of nursing students attending a small liberal arts, faith-based, Midwest, private college classified as baccalaureatediverse composed the study population. The researched college had been in existence for over fifty years, but primarily served the purpose of religious education or theology. In the most recent decade (2014), the college expanded baccalaureate degree offerings in the fields of psychology, business, and education. The sample was comprehensive because all nursing students at the time of data collection, early 2014, were included. The data admission packets include demographic indicators, a 300 word essay, interview notes, and GPA calculations. The dataset represented students’ admission packets to nursing school for the 2011 (17 students), 2012 (20 students), and 2013 (23 students) school years as well as the eight non-retained student admission packets from the same years. During May 2014, the first 17 nursing students graduated with a BSN and subsequently scheduled their national board

37

examinations. This study did not include pass rates on the national registered nurse board examination.

Student Demographics Eight percent of the 60 retained nursing students were male and 92% were female, with an average age of 25 at admission. The students self-identified their race as follows: 76% Caucasian, 11% African American, 10% Asian, and 3% Hispanic. Approximately 50% entered as transfer students. The other half began pre-nursing science coursework at the same college as nursing school. The American Association of Colleges of Nursing (2012) reported that 11.4% of nursing students were male. The research report continued to identify that U. S. baccalaureate nursing students were 72% Caucasian, 10% African American, 9% Asian, and 7% Hispanic. Overall, the national average was slightly more diverse than the retained research population. Eight nursing students had not been retained, due to course failure or personal non-academic reasons. The non-retained students (11.8% of the total) were 25% male. This is larger than the 8% male students in the retained population. Also, 75% were Caucasian and 25% Asian, which is larger than the 10% Asian students in the retained group, as depicted in Figure 3.1. It was difficult with the small population (n=8) to identify diversity distinctions between the retained students, non-retained students, and the national averages. The national data set from the AACN (2012) does not offer insight into student’s religious affiliation. The sample pool attended a faith-based institution and, therefore, 38

may be diverse in regards to other metrics. Based on anecdotal experiences, the sample pool was quite similar to other institutions and with that of the nursing profession. The similarity may be partially attributed to the fact that half the researched nursing students had transferred in from another institution. The majority of the institutions from which the research’s nursing students transferred were secular.

Case Study Method This research required a mixed-method instrumental case study approach. Effective social science disciplines combine qualitative and quantitative methods and produce strong case studies (Flyvberg, 2006). The case study method was well suited to this research due to its holistic nature (Baxter & Jack, 2008; Flyvberg, 2006). Flyvberg (2006) explained that human knowledge is context-dependent and that case studies are optimal for that reason. In health science research, case studies can be done for purposes of theory generation and program evaluation (Baxter & Jack, 2008). An instrumental case study is best accomplished when bounded by time. This research examines the first three cohorts of nursing students admitted in 2011-2013. It is considered instrumental since understanding about how to best admit nursing students by studying a particular case was sought (Stake, 1995). The practical steps to accomplish a strong case study, as identified by Stake (1995), includes access to confidential data and researcher time commitment.

39

Non-Retained Nursing Students (n=8) Caucasian Asian

Retained Nursing Students (n=60) Caucasion African American Asian Hispanic

AACN (2012) Nursing Students (n=161,900) Caucasian

African American

Asian

Hispanic

Figure 3.1 Baccalaureate Nursing Student Diversity.

40

The process of admission to nursing school at this research site represented an exemplar case, which was why a case study method was appropriate. The goal of a case study is to better understand a complex phenomenon which can provide insight or direction for future work. This method complements the research questions, increases validity, and produces a professional exemplar. In addition, the nursing students’ emergent viewpoints were investigated, using interviews and essays, along with grade point averages used for the admission process. The qualitative and quantitative pieces of the case study follow.

Case Study Qualitative Data Collections The first data collection point was an in-depth theme investigation of the students’ admission essays and interview notes. These secure data were collected over the past three years and are accessible. All admission data were examined manually to distinguish salient themes. The qualitative portion of the case study allowed exploration prior to data analysis which further allowed the case to unfold naturally. The qualitative case study portion also offered the researcher the opportunity to evaluate a complex phenomenon within the natural settings (Baxter & Jack, 2008). Pre-admission prediction of clinical capabilities is quite unclear as it has been studied less often (Salvatori, 2001). Also found in the literature review section was the need for clarity and consensus. The essay and interview data were converted into a weighted numerical scoring system totaling ten, where all components are worth two points each. Table 3.2 shows the variables in the current and past admission process and the percentage weight in

41

admission decisions. This admission process was employed for three years (2011-2013) to admit nursing students. At this research site, applicants with the highest scores were admitted until all available spots are filled. Therefore, students with a score of ten and then nine and so on were admitted until capacity is reached. Capacity is determined by the State’s Board of Nursing in this case. The school of nursing coordinator tabulated the scores and kept a data sheet. Each admission year produces a unique cohort. College grade point averages account for 40% of the admission decision since the first two criteria are related to GPA. This was the largest factor affecting the admission decision. The essay and interview each account for 20% of the admission decision. See Table 3.2 for a visualization of these variables. Table 3.2 Nursing School Admission Variables. _____________________________________________________________________ Components for Nursing School Admission Weighted Percentage _____________________________________________________________________ Cumulative college GPA to date

20%

Re-calculated college GPA for major specific courses such as Biology

20%

Previous college credits at the same institution (considered a residency legacy)

20%

Essay assessing caring, commitment, and community involvement

20%

Interview assessing professionalism, 20% communication, and dedication _____________________________________________________________________ 42

The residency legacy was enacted prior to the start of the school of nursing and could not be altered. Residency was calculated based on a students’ prior attendance at the same institution. For example, those applicants that took their pre-admission science courses at the researched institution were granted the two admission points from that category. This criterion is not under review in this study. The College had expressed desire to remove this component for future admission decisions. Based on Holland’s Theory (1966) and its application in O*NET (2013), the admission essay and interview notes were researched and analyzed according to registered nurse’s common workplace values and interest codes. The O*NET is considered public domain and has many purposes, such as to inform research. However, it was ultimately designed with the intent of helping people understand different occupational options. The National Center for ONET Development created the online O*NET database. It is located in the public domain and expresses a freedom of use statement on the website. O*NET (2103) provides the following interest and work value definitions that apply to registered nurses. Social – Social occupations frequently involve working with, communicating with, and teaching people. These occupations often involve helping or providing service to others. Investigative - Investigative occupations frequently involve working with ideas, and require an extensive amount of thinking. These occupations can involve searching for facts and figuring out problems mentally.

43

Conventional - Conventional occupations frequently involve following set procedures and routines. These occupations can include working with data and details more than with ideas. Usually there is a clear line of authority to follow. Relationships - Occupations that satisfy this work value allow employees to provide service to others and work with co-workers in a friendly non-competitive environment. Support - Occupations that satisfy this work value offer supportive management that stands behind employees. Achievement - Occupations that satisfy this work value are results oriented and allow employees to use their strongest abilities, giving them a feeling of accomplishment. Admission process. The original admission variables were determined by a small group (n=3) of nursing professors and were based upon past academic and professional experiences. The three nurses who devised the original admission rubric had previous experience in the hospital and academia. At the research site, community involvement was an institutional tenet and a commitment to caring was highly valuable to a hospital partner. Therefore, examples of caring and community service were sought through the original interview and essay. Table 3.4 highlights the area to be researched, prior to adoption or use in an actual or subsequent nursing school admission process. The essays and interviews scoring rubrics were altered by examining theoretically-based interest codes and workplace values of registered nurses found in O*NET, such as an interest in being social and valuing relationships. 44

For the purpose of this study, the first three cohorts were admitted to nursing school under the same criteria. Prior to making such an admission scoring practice change, which could affect future applicant admission trajectory, a theory-based research initiative was warranted. Adding theory to the work of admissions offers the team a new way of viewing the process. As part of the admission process, each essay has been securely saved, along with the notes that were taken during the interviews by an average of six interdisciplinary admission committee professionals. The team who scored the essays also participated in the group interview session. The team was comprised of supporters of the school of nursing such as a local nurse educator from the partnering hospital. Essay. The essay questions remained unchanged during the three year research period. The applicants were given approximately one month to complete a three hundred word essay for admission consideration. The three essay questions posed were: What caring abilities do you have and how do you show them, describe your spiritual values and how are they evident in your life, and how are you involved in community activities and/or volunteer work? Initially, the review team found that many students did not follow the instructions. Approximately 25% wrote well beyond the 300 word maximum or wrote on a completely different topic. Although the questions posed are quite different than the research fit characteristics, many applicants wrote about relationships, support, or achievement.

45

Interviews. Interviews are ideal for assessing non-cognitive traits and fit with a program’s mission (Salvatori, 2001). The interviews were conducted in a group setting, where the admission committee members varied somewhat throughout the years, but always included nursing faculty, local community members, and staff or faculty from other disciplines at the College. The questions asked are displayed in Table 3.3. The interviews were considered group interviews due to not only was a panel of interviewers present, but also there were multiple candidates present. For example, a group interview session lasted approximately one hour and contained six interviewers and six student candidates. This methodology was chosen not only to be time conscious, but it also offered the interviewers insight into the candidate’s social interactions with one another. For example, a group-interview candidate may look at her cell phone or file her nails while other candidates are speaking and that behavior would be noted negatively in the notes. Table 3.3 Group Interview Questions. Why did you choose nursing?

What are your strengths in regards to interpersonal relationships?

What is your vision for your future nursing

What are your past educational

career?

experiences?

What tools do you use when you encounter

What has been your most recent or

unexpected stressors?

memorable experience helping someone?

46

The group interviewers (n=6) all asked one question during the session. Each candidate had the opportunity to start the round of answers, so the same student would not always have the advantage or disadvantage of answering last. The interviewers drafted their own questions, but shared them in advance with the group for consistency. The interview questions asked can be found in Table 3.3. Data regarding the students not admitted during the past three years were not investigated since nursing school retention data are not discernable. Admitted nursing students were utilized, including those who were not fully retained, to calculate an actual retention percentage. Currently (2014), 88% (60/68) of admitted nursing students have been retained. How would retention have been affected by altering or improving the admission process? The researchable changes occur in the last two admission-variable components in Table 3.4, the essay and interview. The new words, relationship, support, achievement, social, investigative, and conventional, are shown in bold text. The essays and interview notes were scored to reflect the potential or researched admission change. The first three components— cumulative GPA, re-calculated GPA, and previous college credits earned at the institution or residency—were not altered nor were any weighted percentages changed.

47

Case Study Quantitative Data Collection The following quantitative analysis investigates the admission GPA as the predictor and students’ retention in nursing school as the outcome. Pre-admission academic scores predict didactic academic performance for all professional disciplines (Salvatori, 2001). The predictor was deemed continuous with a categorical outcome. This analysis used a logistic regression with the log-odds of the dichotomous outcome to preserve the assumption of a linear relationship in regression. Thus, it more accurately estimates the incremental relationship between a continuous predictor and the likelihood of shifting value in the dichotomous outcome from not transferring to transferring. Like a linear regression analysis, the logistic regression provides two critical points of information: (1) the amount of variance in the probability of the outcome explained by the predictor Table 3.4 Improved Admission Variables. Components for Nursing School Admission

Weighted Percentages

Cumulative college GPA to date

20%

Re-calculated college GPA for major specific courses such as biology

20%

Previous college credits at the same institution (residency legacy)

20%

Essay assessing relationships, support, and achievement values

20%

Interview assessing social, investigative, and conventional interests

20%

48

(Nagelkerke R2); and (2) the amount of estimated linear relationship between the predictor and log of the odds of the outcome (B). To make sense of this coefficient, the exponential conversion shows the percent difference from a 1:1 odds relationship. Logistic regression analysis also supplies each estimate with an associated significance, tested with a Wald test statistic, to determine the extent to which the observed estimate could have occurred by chance. For this analysis, a standard alpha level of .05 to argue for significance was used.

Case Study Data Strategy Final data interpretation occurred as an instrumental case study by piecing together the qualitative and quantitative data from the original and modified admission criteria variables. The quantitative data were not changed between comparison groups due to the GPA piece remained static in this research. Retrospective data collection can reduce researcher bias and potential risk for the population under study. Finally, all previous interview and essay review team members (n=6) were invited for an inter-rater reliability session. During this voluntary session, which lasted approximately one hour, the previous admission team re-scored ten randomly selected and de-identified essays and interview notes. This served as a validity tool to check the degree of agreement between the team and myself, as I was be the only researcher who re-scored all 68 admission packets.

49

Projected Findings Understanding about the relationship between nursing student retention and applicant attributes upon admission, whether it is cognitive or non-cognitive in nature was researched. This study may have implications for nursing schools, students, and higher education administrators. A more reliable admission process has the potential to increase retention and perhaps, become a model for admission processes in other nursing schools. Considerable resources are exhausted as students and nursing school admission committees work to discern fit and attempt to justify a holistic admission process. A holistic assessment incorporates the personality attributes that have been influential within one’s home, social, and work environment (Borges et al., 2005). This study had the potential to understand the applicant outside of an academic environment and possibly project into the profession.

Conclusion Nursing is considered an art and a science, yet this balance has not been fully articulated during the admission process. Researching the balance between cognitive and non-cognitive indicators has the potential to result in a published study for future exploration and possible adoption. Based on the prior theoretical and literature searches this research was built on the foundation of Holland’s Fit Theory (1992) and employed a mixed-method case study methodology. A retrospective approach was applied, revisiting the past three cohorts of applicant admission essays and interview notes, followed by a logistic regression of the pre-admission grade point averages. The research answered the following questions: 50

1. To what extent does an applicant's college GPA, in their pre-nursing courses for their major, predict semester-to-semester retention in a small, private baccalaureate nursing school? 2. If student fit, as articulated by Holland’s Person-Job Fit Theory (1992), based on registered nurse interest codes and common workplace values in O*NET, is assessed during the baccalaureate nursing school admission process, will retention improve? Over the past three years, 68 nursing students were admitted to the research population and 60 remain; eight were not retained. The complete admission packets for these 68 students were researched to garner insight, which has the potential to possibly improve student retention or success in nursing schools if revised admission processes are subsequently adopted.

51

CHAPTER FOUR DATA ANALYSIS

This research was comprised of admission data from nursing students attending a small liberal arts, faith-based, Midwest, private college, classified as baccalaureate diverse. The sample included archival data of students enrolled during 2011 through 2013. At the time of data collection, all admitted nursing students were included (n=68, 8 of whom did not complete the program). The normal admission process, for this small pivate college, requires applicants to submit their college grade point average (GPA), write an essay, and participate in a group interview. All identifying information was removed in January, 2014, from the admission packets, which included GPA, essay, and interview notes. This information was immediately coded and archived for the purpose of this research study. The packets were also numbered and coded as either retained with a 0, or not retained with a 1.

Data Analysis Procedure The deidentifed archival data were re-read and scored 0-6 according to Holland's Person-Job Fit Theory (1966), as seen in Table 3.3. Each of the fit characteristics were assigned one point. The average score was 4.18 where “1” was the lowest score and “6” was the highest. For example, a student would score a point by writing about their desire for achievement or for expressing a desire to fulfill a role on a team. 52

According to Holland, a person who fits in one’s environment, school, or profession can be described as productive, satisfied, and successful. His person-job fit theory can help researchers understand why some students thrive in a professional program, such as nursing, while others fail to thrive. This theory considers both an individual’s personality and the environment, to determine whether there is a proper fit. The National Center for O*NET Development (2013) is a web-based research database for occupational information that applied Holland’s theory. A registered nurse’s environment has been identified by O*NET, based on the common work values and employee interest codes: valuing relationships, support, and achievement, and a desire to be social, investigative, and conventional. The School of Nursing being researched had only been in existence for three years at the time of this study. Thus, a total of three years’ worth of essays and interview notes (2011-2013) from all possible applicants were researched by re-scoring based on the new theoretical criteria from Holland (1966) and O*NET (2013). Appendix B reflects the re-scored sheet.

Quantitative Data Analysis A quantitative measure, such as grade point average (GPA) is the most commonly used admission practice for undergraduate nursing schools (Carpio & Brown, 1993; Codier & Odell, 2013; Jones-Schenk & Harper, 2013). The first research question posed: To what extent does an applicant's "major specific" college GPA predict semester-tosemester retention in a baccalaureate nursing school? The small private college’s major specific prerequisite courses were Chemistry, Biology, Anatomy/Physiology, 53

Psychology, and Microbiology. The researched college utilized a cohort model where a small group of nursing students took classes together for three years. The averaged pre-nursing school college grade point average (GPA) was the continuous predictor. The minimum GPA found was 2.72 and the maximum was 4.0. This resulted in a mean GPA of 3.381 with a 0.272 standard deviation (n=68). The mean GPA for the retained students was 3.389 (n=60) and the mean GPA for non-retained students was 3.323 (n=8). For this analysis, a logistic regression tested the impact of GPA on whether or not the student was retained in nursing school. This analysis used the log-odds of the dichotomous outcome to preserve the presumption of linear relationship in regression. Thus, it more accurately estimated the incremental relationship between a continuous predictor and the likelihood of shifting value in the dichotomous outcome, from not transferring to transferring. Similar to a linear regression analysis, the logistic regression provided two critical points of information: (1) the amount of variance in the probability of the outcome explained by the predictor (Nagelkerke R2); and (2) the amount of estimated linear relationship between the predictor and log of the odds of the outcome (B). Logistic regression analysis also supplied each estimate with an associated significance, tested with a Wald test statistic, to determine the extent to which the observed estimate could have occurred by chance. The GPA predictor was not statistically significant (0.513), indicating that GPA was not correlated to nursing school retention. Therefore, the admission college GPA scores, within this study, do not have a significant impact on the outcome of nursing 54

student retention and are not recommended as the sole admission criteria when retention is sought. The variable of GPA had an un-standardized exponential coefficient of 2.556, which are the impacted odds. After standardization, the impact observed was that for every one-point increase in GPA there was a 156% increase in the likelihood that the student was retained. A statistical power analysis was conducted to determine how likely the logistic regression had detected an effect. A power analysis can assist a researcher by providing a level of confidence or information about the necessary sample size that would often effect the given outcome. The outcome, in this study, was retention in nursing school. An effect size above 0.5 is considered large. A score below 0.3 is considered a small or trivial effect. It was determined that there was a strong finding (0.882 probability) that the 60 retained students’ GPA scores were linked to retention. However, the eight non-retained students’ GPA scores were found to be a weak finding (0.118 probability), as linked to retention. This is most likely attributed to the small sample size. Only eight students were coded as not retained. The majority (n=60) were coded as retained. The non-retained student sample did not show a statistically significant effect with respect to semester-to-semester retention. It is possible that a larger sample size would be statistically significant however, in this quantitative analysis, no statistically significant relationship was discovered. For the quantitative analysis, two outside experts checked the data and validated the findings.

55

Qualitative Data Analysis A nursing applicant should have the personality traits that support nursing school completion and, in turn, a greater likelihood of success in the nursing profession (Huch et al., 1992). Student fit was based upon Holland’s Person-Job Fit Theory (1966). If student fit, according to Holland, is based on registered nurse interests and common workplace values as researched by O*NET (2013), is assessed during the baccalaureate nursing school admission process, will retention improve? The registered nursing field demonstrates the following six fit characteristics: valuing relationships, support, and achievement, and a desire to be social, investigative, and conventional (O*NET). The first three fit characteristics (relationships, support, and achievement) were in the workplace values category and better suited for assessment with the essay. The final three fit characteristics (social, investigative, and conventional) were considered interest codes and were better suited to be assessed through the interview notes because of the applicant’s ability to verbally or nonverbally express these characteristics. For example, the interviewers also had the opportunity to witness and document social interactions. The fit characteristics, according to O*NET, were defined as: Relationships - Occupations that satisfy this work value allow employees to provide service to others and work with co-workers in a friendly non-competitive environment. Support - Occupations that satisfy this work value offer supportive management that stands behind employees. Achievement - Occupations that satisfy this work value are results-oriented and allow employees to use their strongest abilities, giving them a feeling of accomplishment. 56

Social – Social occupations frequently involve working with, communicating with, and teaching people. These occupations often involve helping or providing service to others. Investigative - Investigative occupations frequently involve working with ideas, and require an extensive amount of thinking. These occupations can involve searching for facts and figuring out problems mentally. Conventional - Conventional occupations frequently involve following set procedures and routines. These occupations can include working with data and details, more than with ideas. Usually there is a clear line of authority to follow.

Aggregated Analysis The aggregate scored essays and interview notes from three years or three cohorts of nursing students identified five students who scored low with only “0”, “1”, or “2” fit characteristics. This indicated that five students displayed less than half of the six possible sought-after fit characteristics during the admission process. Of these five lowest scoring nursing students, as identified in this new admission process, four were not retained in nursing school. Half of the total not retained nursing students (4 of 8) could have been identified during admission as having a poor fit for the profession and subsequently not been admitted. Students with a stronger fit alignment, gaining 3-6 fit characteristic alignment, with the surveyed registered nurses from O*NET (2013) could have possibly been admitted. One student would have been erroneously identified as a poor fit with the new criteria, since that student was successfully retained in the program. Also, four of the 57

non-retained students were not identified as poor-fit during this new process, indicating that the new admission process would not clearly identify all non-successful students. Other factors or variables could have influenced the attrition rate. Some of these factors may not be identifiable during the admission process. For example, one student suffered a devastating car accident in the beginning of nursing school and could not be retained. A post-hoc opportunity for deeper analysis was considered. Detailing how many applicants scored points or fit alignment within each of the six characteristics could have been possibly interesting. The data could have been visually represented as a scatter plot. It is possible that certain fit characteristics were more prevalent than others. Unfortunately, in accordance with the IRB regulations, the data is secured at the research location’s campus and not currently (2014) available for re-analysis.

Individual Analysis The subsequent data analysis separated the essays and interview notes to interpret the results individually. This portion of data analysis seeks to answer the last two of the four research questions: 2a. To what extent does an applicant's admission essays align with registered nurse’s work values from Holland (1992) as applied by the O*NET, if nursing school fit is established, is semester-to-semester retention greater, based upon these fit characteristics? 2b. During an admission interview, if an applicant describes his or her interests to be congruent with registered nurse’s interest codes from Holland (1992) as applied by

58

the O*NET, will nursing school fit be established and lead to improved semester-tosemester retention? Essays. Applicants were identified who did not express any of the fit characteristics: relationships, support, or achievement, and those who displayed only one of the possible three desired values in their written essays. Sample excerpts from the applicants are highlighted in Table 4.1. These applicants with “0” or “1” fit characteristics in their admission essay were identified as not having fit with the Registered Nurse’s work environment. The third research question examines the relationship between fit and retention that can be researched with the essay portion of the admission process. The new fit characteristics being assessed were written descriptions that exhibited valuing relationships, support, and achievement. The fit characteristic of relationships described Table 4.1 Excerpts from the Essays. Admission Essay Re-scored as a poor fit for the Registered Nurse’s work environment. This applicant lost one point in the achievement category.

“My spirituality plays an integral role in choosing a human service profession. I have a life’s worth to give to this profession and I am hoping to have the opportunity to do so.”

Re-scored as meeting one of the three fit theory characteristics. This applicant scored one point in the achievement category.

“Becoming a pediatric registered nurse is the highest priority in my life. I have taken many classes in preparation for applying to nursing school.”

59

an applicant who was satisfied serving others and working with a team. Support-seeking employees value a responsive management style that stands behind them. Finally, the work value called achievement was defined as satisfaction with results, which leads to a feeling of accomplishment. The re-scored essays, seeking fit, only identified two of the eight non-retained students due to the score of a “0” or “1”. The re-scored essays, seeking fit, did not reveal a relationship with retention since the majority of the non-retained students (n=8) were not identified as having fit according to Holland (1992) during the re-scored essay process. Interviews. Applicants who did not express the social, investigative, or conventional fit characteristics and those who displayed only one of the possible three desired values during their interview were identified and coded. Sample excerpts from the interview notes are highlighted in Table 4.2. These applicants scored a “0” or “1” fit characteristic in their admission interview notes and were subsequently identified as not having fit with the RN’s work environment. This was determined due to the fact that an applicant with “2” or “3” fit characteristics displayed more than half (n=3) of the possible registered nurse work values and would constitute fit. The final research question focused on the admission interview. If an applicant described his or her interests to be congruent with nursing fit theory characteristics, accoring to Holland as applied by the O*NET, during the admission interview, could nursing school fit be established and lead to improved semester-to-semester retention?

60

These interests included a verbal desire to be social, investigative, and conventional. A social employee would seek an environment that involves helping people, communicating frequently, and, possibly teaching. An investigative occupation requires a large amount of thinking, searching for data, and solving problems. Finally, a conventional occupation follows set procedures and has a detailed routine with, possibly, a clear line of authority. The interview notes identified seven of the eight total non-retained students did not possess two or three of these characteristics. The majority (n=7) of the non-retained nursing students (n=8) had 0 or just 1 fit characteristic identified through the interview notes. It is possible to infer that the interview is capable of distinguishing the applicant’s semester-to-semester retention potential. This finding is much stronger than that of the rescored essays. Table 4.2 Excerpts from the Interview Notes. Re-score as a poor fit for the Registered Nurse’s work environment. This applicant lost one point in the investigative category.

Re-scored as meeting one of the three fit theory characteristic. This applicant scored one point in the investigative category.

61

Interview Notes When asked about critical thinking as a future nurse, the applicant stated, “I plan to do whatever the doctor tells me to do.”

The interviewer documented that the applicant described how she effectively problem solves to manage her life and schoolwork demands.

Inter-Rater Reliability Past members of the inter-professional admission team (n=6) were invited to spend one hour to rescore 10 sets of essays and interview notes. This team worked on a non-incentivized voluntary basis since they were all employed in a full-time capacity. Three of the past six team members agreed to participate in the inter-rater reliability session. A smart phone application was used, which generates random numbers, to assist in choosing ten random admission packets. The essays and interview notes were already de-identified for the team and were, therefore, confidential. An inter-rater reliability scoring sheet, in Appendix C, included the new fit characteristic definitions as described earlier in this chapter. The inter-rater reliability (n=3) team was instructed about the case study methodology and Holland’s Person-Job Fit Theory (1966). A reliability analysis can help determine consistency or degree of agreement between the different raters. The re-scored essays and interview notes were scored 0-3 each for a total of 0-6 for each rater, including myself, for a total of four raters. The essays and interview notes were scored separately allowing addition of the two scores to create an applicant’s aggregate score. The inter-rater reliability score was 0.77. When considering the number of scoring options and number of observations, this Cronbach’s Alpha score indicated adequate reliability. Therefore, the scores appeared similar to the way others would have scored the data three out of four times, thus strengthening the results and confirming the analysis. Results from the inter-rater reliability session are shown in Table 4.3.

62

Table 4.3 Inter-Rater Reliability Session Scores. Researcher's Scores

Rater #1's Scores

Rater # 2's Scores

Rater #3's Scores

5

0

3

4

3

0

2

0

5

0

5

1

5

0

5

3

1

1

2

0

4

1

3

4

4

0

3

3

2

1

0

1

5

3

6

5

3

0

2

3

63

Summary All four research questions were investigated using the most appropriate method, either quantitative for the first question, or a qualitative analysis for the last three research questions. The first question required a quantitative analysis and the other research questions were answered with a qualitative theme analysis. A statistical relationship was not found when comparing college admission GPA and nursing student retention. Considering the research location was a small, private, and faith-based institution, early college course grades were not correlated with nursing student success and therefore should not be the only admission criteria or a heavily weighted indicator of retention. Similar to GPA, the re-scored essays, demonstrating fit characteristics, were not predictive of actual nursing student success. The applicants seeking admission to the research location had the potential to write inauthentic essays. It is also possible that the essay instructions were not clear enough. In this research situation, the admission essays should not justify admission when semester-to-semester retention was the goal. Unlike GPA and the essay, the interview notes proved to be the most reliable in predicting nursing student semester-to-semester retention. Similarly, Ehrenfeld and Tabak (2000) concluded that admission interviews increased retention in nursing school by screening out applicants who displayed personal barriers to success. Based on the case study of a small private baccalaureate college, nursing student admission interviews were the most valuable and when applied, according to Holland’s Person-Job Fit Theory (1992), could possibly lead to improved semester-to-semester retention. 64

CHAPTER FIVE FINDINGS AND CONCLUSIONS

Overview Qualitative and quantitative data including interviews, essays, and grade point averages were researched using a retrospective mixed-method case study. The case study highlighted the admission process at one small, private U. S. baccalaureate nursing school to address the goal of semester-to-semester retention. The analysis procedure included a logistic regression and power analysis, coupled with a re-scored theme analysis of essay and interview notes. The research was grounded in Holland’s Person-Job Fit Theory (1992) and used aggregate criterion data from the National Center for O*NET Development (2013). The O*NET surveyed registered nurses to identify that their primary interest codes were in the areas of social, investigative, and conventional traits. These demonstrated a common set of workplace values of relationships, support, and achievement (O*NET).

Researched Questions 1. To what extent does an applicant's college GPA, in their pre-nursing courses for their major, predict semester-to-semester retention in a small, private baccalaureate nursing school?

65

2. If student fit, as articulated by Holland’s Person-Job Fit Theory (1992), based on registered nurse interest codes and common workplace values in O*NET, is assessed during the baccalaureate nursing school admission process, will retention improve? 2a. To what extent does an applicant's admission essays align with registered nurse’s work values from Holland (1992) as applied by the O*NET, if nursing school fit is established, is semester-to-semester retention greater, based upon these fit characteristics? 2b. During an admission interview, if an applicant describes his or her interests to be congruent with registered nurse’s interest codes from Holland (1992) as applied by the O*NET, will nursing school fit be established and lead to improved semester-to-semester retention?

Summary of Findings It was found that college grade point averages (GPA), in this admission practice study, did not have a statistically significant impact on the outcome of nursing student retention. Overall, GPA was found to not be correlated to nursing school retention. However, it is believed that a larger sample of non-retained nursing students, in a different setting, may have altered the findings. Academic benchmarks, such as GPA according to the literature has an impact on retention prediction, but was not a strong enough predictor to stand alone, based on the research findings, utilizing the sample pool from a small, private, and faith-based institution. 66

The admission essays and interview notes were re-scored using criteria based upon Holland’s Fit Theory (1992) applied in the Registered Nurses’ work environment survey from O*NET (2013). The re-scored essays lacked the ability to determine student retention since only two out of eight were identified as lacking fit. However, the interviews had the most potential and accuracy in predicting retention after the application of fit theory. The inter-rater reliability session lasted one hour and was able to add a level of reliability to the qualitative theme analysis procedure, as it confirmed the qualitative analysis. In summary when seeking retention at a similar nursing school consideration should be given to the importance of the interview and Holland’s Fit Theory as admission criterion.

Limitations There were three limitations of the study identified. The study’s location was a limitation, as it was my former place of employment. Likewise, the sample was extracted from only one small baccalaureate nursing school in the Midwest, which hampered generalizability. The research location was a small, private, faith-based institution. The subject pool may vary significantly, if the research had occurred at a large or public institution. Future research at a larger and/or public university could expand the methodology potential and body of knowledge. Also, the interviews were not recorded or scripted. This would have allowed for more detailed or accurately transcribed notes. It also might have been able to add insight for future use at a different institution if the process was more formalized and replicable.

67

Lastly, it is important to note the difficulty of comparing college grade point averages from different institutions and over various time periods. Ideally, the preadmission grade variable would remain constant to decrease variability. However, this would require that all applicants have a similar academic background and educational experiences from one institution.

Future Research There are many prospective leads for future research that have emerged. The options are limitless, but it’s prudent to build upon the current work to alter variables one at a time. These options include applying a new theoretical framework, a larger sample size, admission criteria correlation to clinical skills as a success indicator, or other success indicators. During the comprehensive phase of this dissertation process, I explored additional theoretical frameworks. I plan to revisit the theories of motivation to enlighten my future nursing admission research initiatives. Could a scripted interview, designed to seek a motivated applicant, produce a successful registered nurse? It is also possible that the essay has merit after the questions are re-written. The essay questions could be re-written with a new theoretical focus since the researched essays were not grounded in a theory nor were they seeking fit characteristics. Fit theory, as it is applied in the O*NET may prove useful in other bodies of future research as well. During my theoretical discovery in this paper, I found the O*NET to be full of rich data pertaining to numerous occupational categories. I began to ponder 68

the use of fit theory from the O*NET to improve faculty retention. Nursing Instructor is an occupational category in the online database. The interest codes are the same as for registered nurses, but the work values are vastly different. I found this query interesting and piqued my interest for future research. Does Holland’s Person-Job Fit Theory (1992) matter in regards to faculty retention? The sample (n=68) was similar to the national results in regards to diversity. Therefore, I do not believe that the students at a baccalaureate faith-based school differed from those attending a secular institution. This may also be due to the fact that half the students were transfer students from predominantly secular institutions. At the place of research, students were accepted from all backgrounds, as religious affiliation was not a requirement. I have been in discussion with a nursing admission testing company. The company is interested in publishing research about their unique admission test, which has been designed to predict student retention. I think an important next step is to bring my research to a larger population and the assistance of a testing company may prove more feasible. There is also an opportunity to change the indicator of student success from semester-to-semester retention to research first-time NCLEX pass rates for future research. However, I am more interested in a longitudinal approach to research the graduate’s future in the nursing profession. What early assessment variables will predict long-term professional dedication? There is a growing concern about the numbers of nursing graduates who leave the field. 69

A true theme analysis of the essays and interview notes is another future research direction. Instead of seeking fit characteristics, I could re-visit the qualitative data and conduct a theme analysis. This would give me the opportunity to find similarities amongst the retained and non–retained applicants. During the data collection process, I discovered that my current employer, which is no longer the research location, has made nursing student retention their highest priority for the upcoming fiscal year. My current employer has some other existing data, indicating similar findings regarding the lack of correlation between admission GPA and retention, but on a much larger research scale. I am very interested in joining their research team, continuing to pursue this topic, and seeking dissemination. Ultimately, I hope to positively impact nursing student retention beyond graduation and into the profession.

Conclusion Nursing student retention is a research area of great interest and value to the profession, public, students, members of academia, and accreditation teams. Student applicants, as well as the profession, are multi-faceted and complex. Seeking optimal fit can be challenging, thus the lack of consensus in the literature review. One such challenge is the time dedication required for a holistic admission process. However, it is important to note that proactively working on the admission criteria phase of an admission process has the potential to decrease future work. For example, a nursing student with adequate fit in the profession might require less support and be more engaged during the program. Ultimately, the application of an operational theory, such as 70

Holland’s Person-Job Fit Theory (1992), and an admission process that includes an interview, appears to have the most prediction power for semester-to-semester nursing student retention in a small, private college.

71

APPENDIX A IRB APPROVAL

72

73

APPENDIX B RE-SCORED INTERVIEW AND ESSAY DATA SHEET

74

Nursing Essay assessing Student relationships, support, ID # and achievement values 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529

Interview assessing social, investigative, and conventional interests 3 2 2 2 2 2 2 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 3 2 1 2 3 2 2 3 2 2 2 2 2

Total 2 2 2 2 1 2 2 2 2 2 2 3 0 2 2 2 2 2 3 2 2 2 2 3 0 2 3 2 2 2 2 2 0 2 2

75

5 4 4 4 3 4 4 5 5 5 4 5 2 4 4 4 4 4 5 4 4 4 5 5 1 4 6 4 4 5 4 4 2 4 4

530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562

3 2 1 2 2 3 2 2 2 3 3 3 2 3 2 2 2 2 3 2 3 3 3 2 2 2 2 2 2 2 2 2 2

2 0 1 2 2 2 2 2 2 2 3 2 1 2 2 2 2 2 2 2 2 3 2 3 2 2 2 3 2 2 2 2 2

76

5 2 2 4 4 5 4 4 4 5 6 5 3 5 4 4 4 4 5 4 5 6 5 5 4 4 4 5 4 4 4 4 4

APPENDIX C INTER-RATER RELIABILITY SCORING SHEET

77

Interview total out of 3

Interview: conventional

Interview: investigative

Interview: social

Essay total out of 3

Essay: achievement

Essay: support

Essay: relationships

Nursing Student ID #

Inter-Rater Reliability Scoring Sheet

495 499 506 517 519 522 526 532 541 542

Relationships - Occupations that satisfy this work value allow employees to provide service to others and work with co-workers in a friendly non-competitive environment. Support - Occupations that satisfy this work value offer supportive management that stands behind employees. Achievement - Occupations that satisfy this work value are results-oriented and allow employees to use their strongest abilities, giving them a feeling of accomplishment. Social – Social occupations frequently involve working with, communicating with, and teaching people. These occupations often involve helping or providing service to others. Investigative - Investigative occupations frequently involve working with ideas, and require an extensive amount of thinking. These occupations can involve searching for facts and figuring out problems mentally. Conventional - Conventional occupations frequently involve following set procedures and routines. These occupations can include working with data and details more than with ideas. Usually there is a clear line of authority to follow. 78

REFERENCES Adnett, N., et al. (2011). Achieving “transparency, consistency, and fairness” in English higher education admissions: Progress since Schwartz? Higher Education Quarterly, 65, 12-33. Al-Kandari, F. H. & Ajao, E. (1997). Recruitment and retention of nursing students in Kuwait. International Journal of Nursing Studies, 35, 245-251. American Association of Colleges of Nursing. (2013). AACN releases preliminary from 2012 annual survey: Press release. Retrieved from: http://www.aacn.nche.edu/news/articles/2012/enrolldata American Association of Colleges of Nursing. (2012). New AACN data show an enrollment surge in baccalaureate and graduate programs amid calls for more highly educated nurses. Retrieved from: http://www.aacn.nche.edu/news/articles/2012/enrollment-data American Educational Research Association (AERA), American Psychological Association, & National Council on Measurement in Education. (1999). Standards for educational and psychological testing. Washington, DC: American Education Research Association. Baxter, P. & Jack, S. (2008). Qualitative case study methodology: Study design and implementation for novice researchers. The Qualitative Report, 13(4), 544-559. Beale, A. V. & McCutcheon, A. F. (1980). On becoming a nurse. The Journal of Nursing Education, 19(4), 28-32. Beeson, S. A. & Kissling, G. (2001). Predicting success for baccalaureate graduates on the NCLEX-RN. Journal of Professional Nursing, 17(3), 121-127. doi: 10.1053/jpnu.2001.23382 Borges, N. J., Gibson, D. D., & Karnani, R. M. (2005). Job satisfaction of physicians with congruent versus incongruent specialty choices. Evaluations & The Health Professions, 28(4), 400-413. doi: 10.1177/0163278705281071 Brown, J. & Marshall, B. L. (2008). A historically black university’s baccalaureate enrollment and success tactics for registered nurses. Journal of Professional Nursing, 24(1), 21-29. doi: 10.1016/j.profnurs.2007.06.006

79

Butler, C. S. (2013). Personality as a key to nursing retention: The temperament and character patterns of registered and student nurses by specialty. (Unpublished doctoral dissertation). Oakland University, Rochester, MI. Caron, V. F. (2004). The nursing shortage in the United States: What can be done to solve the crisis? In Schmidt Labor Research Center Seminar Paper Series. University of Rhode Island. Carpio, B. & Brown, B. (1993). The admission process of a bachelor of science in nursing program: Initial reliability and validity of the personal interview. Canadian Journal of Nursing Research, 25(3), 41-52. Catalano, J. T. (2012). Nursing now! Today’s issues, tomorrow’s trends (6th ed.). Philadelphia, PA: F. A. Davis. Codier, E. & Odell, E. (2013). Measured emotional intelligence ability and grade point average in nursing students. Nurse Education Today, 34(4), 608-612. doi: 10.1016/j.nedt.2013.06.007 Cortes, C. M. (2013). Profile in action: Linking admission and retention. New Directions for Higher Education, 161, 59-69. doi: 10.1002/he.20046 Crede, M. & Phillips, L. A. (2011). A meta-analytic review of the motivated strategies for learning questionnaire. Learning and Individual Differences, 21, 337-346. doi: 10.1016/j.lindif.2011.03.002 Creswell, J. W. (2008). Educational research: Planning, conducting, and evaluating quantitative and qualitative research. Upper Saddle River, NJ: Pearson. Ehrenfeld, M. & Tabak, N. (2000). Value of admission interviews in selecting of undergraduate nursing students. Journal of Nursing Management, 8, 101-106. Flyvbjerg, B. (2006). Five misunderstanding about case-study research. Qualitative Inquiry, 12(2), 219-245. Gilmore, M. (2008). Predictors of success in associate degree nursing programs. Teaching and Learning in Nursing, 3, 121-124. doi: 10.1016/j.teln.2008.04.004 Holland, J. L. (1996). Exploring careers with a typology. American Psychologist, 51(4), 397-406. Holland, J. L. (1992). Making vocational choices (2nd ed.). Odessa, FL: Psychological Assessment Resources, Inc. 80

Holland, J. L. (1966). The psychology of vocational choice. Waltham, MA: Blaisdell Publishing Company. Huch, M. H., Leonard, R. L., & Gutsch, K. U. (1992). Nursing education: Developing specification equations for selection and retention. Journal of Professional Nursing, 8(3), 170-175. Jones, C. W. (1975). Why associate degree nursing students persist. Nursing Research, 24(1), 57-59. Jones-Schenk, J. & Harper, M. G. (2013). Emotional intelligence: An admission criterion alternative to cumulative grade point averages for prelicensure students. Nurse Education Today. doi: 10.1016/j.nedt.2013.03.018 Jurajda, S. & Münich, D. (2010). Admission to selective schools, alphabetically. Economics of Education Review, 29, 1100-1109. doi: 10.1016/j.econedurev.2010.06.005 Katz, J. R., Chow, C., Motzer, S. A., & Woods, S. L. (2009). The graduate record examination: Help or hindrance in nursing graduate school admissions? Journal of Professional Nursing, 25(6), 369-372. doi: 10.1016/j.profnurs.2009.04.002 Knauss, P. J. & Willson, P. (2013). Predicting early academic success: HESI admissions assessment exam. Journal of Professional Nursing, 29(2), 28-31. doi: 10.1016/j.profnurs.2012.07.001 Kovner, C. T. & Djukic, M. (2009). The nursing career process from application through the first 2 years of employment. Journal of Professional Nursing, 25(4), 197-203. doi: 10.1016/j.profnurs.2009.05.002 Larsson, R. (1993). Case study methodology: Quantitative analysis of patterns across case studies. Academy of Management, 36(6), 1515-1546. Leners, D., Beardslee, N. Q., & Peters, D. (1996). 21st century nursing and implications for nursing school admissions. Nursing Outlook, 44, 137-140. Lengacher, C. A. & Keller, R. (1990). Academic predictors of success on the NCLEXRN examination for associate degree nursing students. Journal of Nursing Education, 29(4), 163-169. McDaniel, M. J., Thrasher, A., & Hiatt, T. (2013). The use of noncognitive factors in physician assistant admission. The Journal of Physician Assistant Education, 24(1), 15-23. 81

McNelis, A. M., Wellman, D. S., Krothe, J. S., Hrisomalos, D. D., McElveen, J. L., & South, R. J. (2010). Revision and evaluation of the Indiana University school of nursing baccalaureate admission process. Journal of Professional Nursing, 26(3), 188-195. doi: 10/1016/j.profnurs.2010.01.003 Melillo, K. D., Dowling, J., Abdallah, L., Findeisen, M., & Knight, M. (2013). Bring diversity to nursing: Recruitment, retention, and graduation of nursing students. Journal of Cultural Diversity, 20(2), 100-104. Murray, H. A. (1959). Explorations in personality: A clinical and experimental study of fifty men of college age. New York: Oxford University Press. National Center for O*NET Development. 29-1141.00. O*NET OnLine. Retrieved June 26, 2013, from http://www.onetonline.org/link/summary/29-1141.00 Newton, S. E., Smith, L. H., Moore, G., & Magnan, M. (2007). Predicting early academic achievement in a baccalaureate nursing program. Journal of Professional Nursing, 23(3), 144-149. doi: 10.1016/j.profnurs.2006.07.001 Newton, S. E. & Moore, G. (2006). The significance of graduate admission written goal statements. Journal of Professional Nursing, 22(3), 205-209. doi: 10.1016/j.profnurs.2006.03.010 Pascoe, R. (1999). Admission to Australian universities. Journal of Higher Education Policy and Management, 21(1), 17-30. Porter, M. L. & Bean, J. P. (2004). Organizational lifecycle in a school of nursing. Western Journal of Nursing Research, 26(4), 444-460. doi: 10.1177/0193945904263282 Reiter, H. I., Salvatore, P., Rosenfeld, J., Trinh, K., & Eva, K. W. (2006). The effects of defined violations of test security on admissions outcomes using multiple miniinterviews. Medical Education, 40, 36-42. Sadler, J. (2003). Effectiveness of student admission essays in identifying attrition. Nurse Education Today, 23, 620-627. doi: 10.1016/S0260-6c17(03)00112-6 Salvatori, P. (2001). Reliability and validity of admission tools used to select students for the health professions. Advances in Health Sciences Education, 6, 159-175. Seldomridge, L. A. & DiBartolo, M. C. (2005). A profile of accelerated second bachelor’s degree nursing students. Nurse Educator, 30(2), 65-68. 82

Shulruf, B., Wang, Y. G., Zhao, Y. J., & Baker, H. (2011). Rethinking the admission criteria to nursing school. Nurse Education Today, 31, 727-731. doi: 10.1016/j.nedt.2010.11.024 Spurlock, D. (2006). Do no harm: Progression policies and high-stakes testing in nursing education. Journal of Nursing Education, 45(8), 297-302. Stake, R. E. (1995). The art of case study research. Thousand Oaks, CA: Sage Publications. Sternberg, R. J. (2010). College admissions for the 21st century. Cambridge, MA: Harvard University Press. Tarrant, M., Knierim, A., Hayes, S. K., & Ware, J. (2006). The frequency of item writing flaws in multiple-choice questions used in high stakes nursing assessments. Nursing Education in Practice, 6, 354-363. Timer, J. E. & Clauson, M. I. (2010). The use of selective admissions tools to predict students’ success in an advanced standing baccalaureate nursing program. Nurse Education Today, 31, 601-606. doi: 10.1016/j.nedt.2010.10.015 Watson, J. (2007). Watson’s theory of human caring and subjective living experiences: Caritive factors/caritas processes as a disciplinary guide to the professional nursing practice. Texto Contexto Enferm, Florianopolis, 16(1), 129-135. Williams, M. G. (2010). Attrition and retention in the nursing major: Understanding persistence in beginning nursing students. Nursing Education Perspectives, 31(6), 362-367. doi: 10.1043/1536-5026-31.6.362 Wolkowsitz, A. A. & Kelley, J. A. (2010). Academic predictors of success in a nursing program. Journal of Nursing Education, 49(9), 498-503. doi: 10.3928/0148483420100525-09 Wong, J. & Wong, S. (1999). Contribution of basic sciences to academic success in nursing education. International Journal of Nursing Studies, 36(4), 345-354. Zara, A. R. (1999). Using computerized adaptive testing to evaluate nurse competence for licensure: Some history and forward look. Advances in Health Sciences Education, 4, 39-48. doi: 10.1023/A:1009866321381

83