Factors associated with job satisfaction among graduate nursing faculties in Nepal

A cross-sectional study was conducted to explore the level of job satisfaction and factors associated with it among graduate nursing faculties in Nepal. At the time of the study initiation, 54 nursing colleges affiliated to seven universities offered baccalaureate or masters’ level nursing degrees. All the faculty members who held at least a master’s degree in nursing with current teaching responsibility at bachelors’ level or higher for more than one year at the same institution were considered the population for the study. A census sampling technique was applied for the study. The name of the college was identified through a web search. Among the 54 colleges, four colleges could not be reached due to scheduling conflict, seven did not consent to participate in the study, and four colleges did not have faculties with a master’s degree. The number of nursing faculties with at least a masters’ degree from the remaining 39 colleges was 357.

As the participants were from different institutions, ethical clearance was obtained from the institutional review committees of both Nepal Medical College (NMC), and Nepal Health Research Council (NHRC) before the initiation of the study. Upon approval from IRB, the campus chief from each nursing college was sent a formal letter to receive their institutional approval to conduct the study with their faculties. The respondents were asked to read the instructions and given an opportunity to decide whether would accept or decline the survey.

The instrument was divided into three sections – the first and second sections included sociodemographic and organization related information. These sections were constructed by the researchers through literature search [10] and discussion with the experts. Third section consisted of a 36-item ‘job satisfaction survey’ developed by Spector [11] and used with his permission. Each question included 6-point Likert scale responses (1: disagree very much, 2: disagree moderately, 3: disagree slightly, 4: agree slightly, 5: agree moderately, and 6: agree very much). These questions covered nine different domains and included four questions in each domain. The total score ranged from 108 to 216 and domain scores ranged from 4 to 24.

The questionnaire was posted online with the help of an expert in information technology. A link to the instrument was emailed to 30 randomly selected faculties as a pretest. They were followed up via email on a weekly basis for one month. A valid response was received from 56.6% of the faculties during pretesting. Data from these faculties were excluded from the main study.

The final survey was sent to 327 faculties via a link to the online questionnaire (after excluding 30 pre-test sample). At least one facilitator was identified from each college, who would coordinate the timely delivery of responses from nursing faculties. These facilitators explained the research to participants and collected the mailing addresses of the nursing faculties, which were then sent to researchers. The researchers prepared a list of the faculties and then sent the questionnaire by mail. The researchers remained in frequent contact with the facilitators throughout the data collection period. The facilitators were provided with a remuneration of Nepalese rupees 500 (equivalent to USD 5) for coordinating the research at local level.

Responses were received from 178 (54.4%) of the population. A retrospective cleaning was done to remove those with less than one year of service, those submitted a ‘declined’ response, or those currently not teaching at a graduate (bachelor and above) program. After this cleaning, 171 usable responses (52.3%) were included in the final analysis (Fig. 1).

Due to the design constrain during the delivery of online questions, a few demographics and institutional-related questions had some missing values. The missing rate was never higher than 7.5%. Simple imputation was done by replacing the missing values by the mean (continuous variables such as age) or largest class (such as involvement in departmental decision making). There were no missing values in the response for job satisfaction questions.

Out of 327 nursing faculties who were sent the questionnaire, 171 (52.3%) were included in the final analysis. The average age was 36.8 ± 7.0 years (range 26 to 57 years). Only 2.9% were male faculties. Most of the faculty had Master of Nursing (MN) degree (55.6%), followed by Master of Science in Nursing (MSN, 40.9%) and Doctorate in Nursing (PhD, 3.5%).

Slightly over one third (36.8%) of the graduate nursing faculties were satisfied with their job. Dissatisfaction about their current job was observed only in 14.6% faculties, and a majority (48.5%) had ambivalent feeling towards their job. Among the nine job satisfaction domains, highest job satisfaction was observed in coworkers (81.3%), followed by the nature of job (71.3%), communication (70.8%), and supervision (63.2%). Most dissatisfaction was towards lack of promotion (56.1%), contingency rewards (44.4%), operating condition (44.4%), pay (40.9%), and fringe benefits (35.1%). Only two domains, pay and supervision, have acceptable reliability score (Cronbach’s alpha > = 0.70). Hence, individual domain scores are not analyzed further, and the satisfaction categories are presented for informational purpose. (Table 1).

The primary objective of this study was to explore factors associated with overall job satisfaction among the graduate nursing faculty. Since majority of the respondents were ambivalent with few respondents expressively dissatisfied, we grouped these two categories as not satisfied as opposed to the respondents who were satisfied. This would allow us to run a multiple logistic regression model with a binary dependent variable (satisfied vs not satisfied).

The details of sociodemographic and organizational characteristics of the faculties are shown in Table 2.

Majority of the nursing faculties (60.8%) were from private organizations and 39.2% from the public institutions. Public institutions usually offer provident fund, grade, and promotional opportunities for their employees. In recent years, private institutions are also catching up in providing these benefits to both their tenured and non-tenured employees. For example, 76.6% faculties responded positively about grade opportunity and 63.7% reported having provident fund benefits even though the overall employment in public colleges was 39.2% and tenure (permanent) rate was only 58.5%.

A grade in Nepalese context is a periodic increment in the basic salary of employees after they have completed certain years of service. In other word, it can be defined as steps within the same grade level in the USA. On the other hand, a provident fund is a pension fund scheme for employees of both the public and private sectors in Nepal. This fund is managed by a government institution which invests the money to generate profit for the depositors. Under this scheme, eligible employees contribute 10% of their basic salary and employer matches the fund equally. This scheme operates similar to a 401(K) savings plan in the USA.

Healthcare benefits were reported by 74.3% of the respondents. Healthcare benefits in Nepalese context is an employer-provided welfare scheme that provides limited medical care within their network for an employee and their dependents. Immediate family members including spouse and children plus parents of both spouses are usually covered by this benefit.

A majority of respondents reported the availability and easy access to textbooks (74.9%), reference books (66.1%), nursing and medical journals (57.3%) and internet facilities (83.6%). (Table 3).

The descriptive results showed that some of the variables have a stronger association with job satisfaction than others. In order to evaluate the effect of each of these factors while simultaneously controlling for other factors, a multiple logistic regression model was used. All variables with an effect size greater than 0.15 (see the correlation coefficient or Cramer’s V statistics in Tables 2 and 3 above) were included in the initial model. Univariate logistic regression (not shown in these tables) also suggested the selection of these variables. All variables that were selected based on the high correlation with job satisfaction were further evaluated for possible collinearity and strong association among each other. Cramer’s V test statistics were used to check for each pair of categorical variables. A cutoff point of 0.50 was established for unusually high correlation at which point two variables were probably measuring the same concept.

Being tenured had a strong correlation with provident fund (0.70) and grade (0.54). In Nepalese context, whether it is a public or private institution, a permanent employment (tenured status) usually requires the employer to provide both provident fund and grade (0.59) but employers can extend these benefits to non-permanent employees at any time. Since majority of the faculty worked in private sector colleges, an employer matched provident fund may have greater value towards their saving for the future. Therefore, only provident fund benefit was chosen to enter the model. Similarly, strong association was detected between current position and level of primary teaching responsibility (0.57) and total compensation above median (0.52). However, effect size between master’s program and total compensation was much lower (0.37) and both primary responsible program and total compensation were entered in the model without current position. Adequate textbook and reference book were also strongly related (0.64) and only adequate reference book which had higher correlation with outcome was selected (Fig. 2).

Table 4 shows the results from multiple logistic regression model where the job satisfaction is the function of many variables. The area under curve of the receiver operating curve showed very strong discriminatory power (AUC = 0.855) of the model. This means that the logistic regression model can accurately classify the predicted outcome as satisfied or not-satisfied for every possible pair of observations for 85.5% of the pairs. The likelihood ratio test (p < 0.0001) and Hosmer and Lemeshow goodness of fit statistics (p = 0.609) also indicate that the model is well behaved. However, a majority of the individual variables included in the model do not have a statistically significant coefficient and have a very wide odds ratio interval.

Table 4 showed that faculties involvement in decision making (higher job satisfaction with higher involvement levels) and adequate reference books came out to be the only strongly significant variables (p < 0.05) in the model. Faculties who were involved in departmental decision-making processes ‘sometimes’ and ‘often or always’ were 3.6 and 4.83 times more likely to be satisfied respectively than those who were never or rarely involved. It is interesting to note that the lowest job satisfaction (Table 1) was in the domain “operating conditions”. It suggests that job satisfaction is high when faculties are included in the decision-making process and they feel part of the ‘system’. Similarly, their job satisfaction is higher (OR = 2.92, p = 0.050) when reference books were adequately available for themselves and the student.

Although the adjusted odds ratios were not significant at 5% level of significance and 95% confidence interval, there were positive indicators of job satisfaction with various factors identified during the descriptive analysis. The nursing faculties were more likely to be satisfied with availability of medical or nursing journals (OR = 2.32, p = 0.090) and internet access (OR = 3.92, p = 0.100). They were 64% more likely to be satisfied (OR = 1.64, p = 0.378) when they were involved in setting questions for the final exams, which is often regarded a greater respect to the faculty in Nepal. They were more likely to be satisfied with their job when they did not have evening clinical (OR = 3.30, p = 0.069), could start their AM clinical at 8 am or later instead of earlier in the morning (OR = 1.54, p = 0.352), had a weekly work load of less than 42 h (OR = 1.98, p = 0.198), had an annual teaching load less than 125 h (OR = 2.41, p = 0.082), and were provided with more than 5 days of professional development opportunities (OR = 2.34, p = 0.072).

One of the limitations of this study was that only highly motivated faculties were willing to participate in the study which may have introduced some outcome bias. Moreover, few of the participants did not have easy access to internet and did not respond to emails frequently, which limited the timely delivery of the response. Due to the lack of national registry system with the contact details of the nursing faculties working in different organization in the country, it was difficult to retrieve the contact details of the participants and some of the surveys were returned as undeliverable.