Factors influencing nursing students’ continued success in a baccalaureate nursing program in Indonesia

Theoretical Framework

A holistic admission model, based on the Association of American Medical Colleges (2013), has been adapted for nursing by Scott and Zerwic (2015). Based on the model, an applicant to a nursing program should be reviewed based on experience factors (i.e., life, community, leadership, healthcare, and cultural/diversity experiences), attribute factors (i.e., demographic information, abilities, development, goals, and interests), and academic metrics (i.e., education experience, pre-admission test scores, grades, and standardized tests). The purpose of using a holistic admission model is to increase the diversity of the students being admitted.

A variation of the Scott and Zerwic (2015) model to review student success, as shown in Figure 1, was used in this study. As student success can include early academic success (Al-Alawi et al., 2020), in this study, it was defined as being able to continue in the nursing program and was measured as 1) being an active student at the end of the first year and 2) being an active student at the end of the second year.

The variables for this study were chosen based on previous research with nursing students. In previous research, the researchers found that the variables described only 28% - 37% of the variance in the GPA at the end of the first semester (Sommers & Park, 2017; Sommers & Wirawan, 2019). However, the variables that were significantly (p < .05) associated with the GPA at the end of the first semester were different in each of the research studies (see Table 1). The only variables that were consistently significant each year with the GPA at the end of the first semester were the scores on the admission exams in math, English, and Indonesian.

A description of each of the variables for this study is in Table 2. Several different variables were chosen because a combination of variables is a better predictor of student success (Al-Alawi et al., 2020).

Study Design and Sample

This study used an exploratory study research method. The already developed study data set that contained the admission and academic records variables of nursing students that were admitted in August 2018 to a large, private, baccalaureate nursing program in Tangerang, Indonesia, was used. In August of 2018, there were 351 students enrolled; 341 students were still enrolled at the end of the first semester. For this study, two new variables regarding student success were added to the existing dataset, and all identifying data was removed. In addition, data from the ten students no longer enrolled at the end of the first semester was excluded.

In the development of that database, based on previous studies of nursing students (Sommers & Park, 2017; Sommers & Wirawan, 2019), nursing students in Cohort 2018 were asked about grit (12-item Grit Scale), calling (4-item Brief Calling Score), and study habits (researcher developed question). Grit was chosen because, in previous research, it has helped to predict success in military education and business (Duckworth, 2016). Among college students, those that view their future as a calling tend to be more confident and comfortable in their career choices (Dik et al., 2012). It was unknown if a sense of calling had any association with nursing students’ GPA in the first semester. Studying in groups has been recommended for learning and long-term memory (Brown et al., 2014). It was unknown if studying with others had any association with nursing students’ GPA in the first semester.

As the Grit Scale and Brief Calling Score had not been used in Indonesia previously, the items were translated into Indonesian and back translated to English by educators fluent in both Indonesian and English. The back translated English version was evaluated by a native English-speaking educator. For the original study with Cohort 2018, after ethical clearance was obtained for that study, a pilot study with 46 students to establish face validity and clarity. The students had no difficulties understanding the Indonesian version of the questions.

Data Analysis

Data analysis was done using SPSS 26, with a level of significance set at .05. To describe the characteristics of the data, descriptive statistical analysis was used. Means and standard deviations were reported for continuous variables, and frequency distributions and percentages were reported for categorical and dichotomous variables. Binary logistic regression analysis was completed to determine if there were any relationships between the experience, attribute, and academic variables and student success (as measured by the end of the first academic year and at the end of the second academic year) of nursing students admitted in August 2018. Before conducting the binary logistic regression analysis, the data was reviewed for uncommon values, outliers, and missing data. There was no missing data.

Ethical Consideration

Ethical approval for this study was from the Mochtar Riady Institute of Nanotechnology Ethics Committee, protocol number: 1912016-04. Upon receiving ethical approval, the existing database from the original study in 2018 (Ethical Approval number for the protocol of the original study was from Mochtar Riady Institute for Nanotechnology Ethics Committee: 04.1709180) was enhanced by adding two new variables: 1) active student at the end of the first academic year (July 2019) and 2) active student at the end of the second academic year (July 2020). The privacy of the students was preserved and protected as all data were de-identified in the database. Confidentiality of the de-identified database was maintained as it was only accessible on secure servers retrieved from password-protected computers.

Description of Sample

As shown in Table 3, the majority of the participants were female (85.3%), from Sumatra Island (29.6%), and attended a Health/Science focused High School (80.6%). The majority did not attend the pre-nursing course (70.7%). Only a few of the participants exceeded the interview standards (28.7%). Most participants wanted to help or serve others or their families (67.4%) and had a previous leadership role (75.4%). Most of the students did not have previous healthcare experience (45.2%) or community service (57.8%) and studied with others (67.4%).

Relationship of Variables with Student Success

Binary logistic regression was completed to examine the association, if any, of independent variables with being an active student at the end of the first academic year (July 2019). The independent variables were tested to verify that there was no violation of the assumption. The Hosmer-Lemeshow Test of the Goodness of fit was not significant (p <.05), indicating that the model is correctly specified.

Based on an examination of the beta coefficients, the variables of first semester GPA, APM logic matrix admission exam, reason for entering nursing, and brief calling presence scale were found to contribute to the model significantly (p <.05, see Table 4). Controlling for other variables, the variable of students’ first semester GPA, in the logistic regression was found to contribute to the model. The unstandardized Beta weight for the variable first semester GPA; B = 6.594, SE = 1.882, Wald = 12.280, p <.05. The Estimated odd ratio favored an increase of nearly 730% [Exp (B) = 730.668, 95% CI (18.281, 29203.496)] for those still enrolled after the first year (July 2019) for one unit increase of first semester GPA.

Controlling for other variables, the variable of APM logic matrix admission exam, in the logistic regression was found to contribute to the model. The unstandardized Beta weight for the variable of APM logic matrix admission exam, B = -.200, SE = .100, Wald = 3.981, p <.05. The Estimated odd ratio favored an increase of nearly 0.8% [Exp (B) = .819, 95% CI (.672, .996)] for those still enrolled after the first year (July 2019) for one unit increase of APM logic matrix admission exam score.

Controlling for other variables, the variable reason for entering nursing, in the logistic regression was found to contribute to the model. The unstandardized Beta weight for the variable reason for entering nursing; B = 1.231, SE = .610, Wald = 4.069, p <.05. The Estimated odd ratio favored an increase of nearly 34.2% [Exp (B) = 3.424, 95% CI (1.036, 11.320)] for those still enrolled after the first year (July 2019) for one unit increase of reason for entering nursing.

Finally, controlling for other variables, the variable brief calling presence scale, in the logistic regression was found to contribute to the model the unstandardized Beta weight for the variable brief calling presence scale; B = .782, SE = .397, Wald = 3.876, p <.05. The Estimated odd ratio favored an increase of nearly 21.8% [Exp (B) = 2.186, 95% CI (1.004, 4.762)] for those still enrolled after the first year (July 2019) for one unit increase of brief calling presence scale.

The first semester GPA is the strongest factor that determines if the student will still be enrolled at the end of the first academic year (July 2019), followed by the two attribute factors of reason for entering nursing and brief calling presence scale and the one academic metrics factor of APM logic matrix admission exam. None of the experience factors were found to be significant in this study. The variables accounted for 40.3% of the variance of still be enrolled at the end of the first academic year (-2 Log likelihood = 86.83, Nagelkerke R Square = .403).

Binary logistic regression was also completed to examine the association, if any, of independent variables with being an active student at the end of the second academic year (July 2020). The independent variables were tested to verify that there was no violation of the assumption. The Hosmer-Lemeshow Test of the Goodness of fit was not significant (p >.05), indicating that the model is correctly specified.

Based on the analysis of the beta coefficients, only the variable of Grit was found to contribute to the model significantly (p <.05, see Table 5). The unstandardized Beta weight for the variable Grit: B = -3.470, SE = 1.575, Wald = 4.856, p <.05. The Estimated odd ratio favored an increase of nearly 3.1% [Exp (B) = .031, 95% CI (.001, .681)] for those still enrolled after the second year (July 2020) for one unit increase of Grit. The strongest factor that determined if the student was still enrolled at the end of the second year (July 2020) was the attribute factor of the Grit score. The other variables from experience factors and academic metric factors were not significant. The variables accounted for 31.0% of the variance of still be enrolled at the end of the first academic year (-2 Log likelihood = 42.365, Nagelkerke R Square = .310).

Implications of the Study

Answering the future challenge of nursing education, particularly in an endemic situation, evaluating the attributes of a nursing student applicant, especially calling and grit, as part of the admission process will be beneficial to the admission process. Applicants that have a sense of calling and grit may successfully complete a nursing program, even if they have lower admission exam scores, as part of holistic admission processes. The use of holistic admission processes also has implications for developing a diverse nursing workforce. To support diverse students that are admitted using a holistic admission process, a variety of support interventions to promote the success for all students is needed (Jeffreys, 2015). When teaching diverse nursing students, educators should consider personalizing approaches, utilizing available resources, promoting cultural diversity, and using a variety of activity teaching strategies (Sommers & Bonnel, 2020).