Midlevel administrators working at colleges and universities in the United States rated the degree to which they perceived their supervisor to exhibit syn-ergistic supervisory behaviors. They also responded to questions regarding their core self-evaluation and commitment to their supervisors. The results suggest a significant positive correlation exists between synergistic supervision and core self-evaluation and between synergistic supervision and supervisor-related commitment. The impact of demographic characteristics on these variables was also examined.
Experts [
Supervision has a significant impact on employee retention and satisfaction [
In higher education, student affairs’ greatest resource is human capital and a large portion of an institution’s budget is personnel [
Midlevel administrators make up the vast majority of administrative employees and according to researchers [
Researchers [
Four researchers [
The authors [
This study, taken from dissertation research [
Core self-evaluation theory originated with [
Core self-evaluation has been linked to a variety of important outcomes such as job satisfaction, engagement, popularity, and job performance within employees [
The CSES uses a 5-point Likert-type scale to indicate the level of agreement or disagreement with each of the statements. The 12 statements address the composite personality traits of self-esteem, self-efficacy, locus of control, and emotional stability and the sum score of the items represents the overall value the individual has of themselves.
While the research results vary, there appears to be a large association between commitment and performance when individual bases of commitment were distinguished [
In summary, there are a large number of midlevel administrators on college campuses around the country and a significant amount of time is spent supervising these individuals [
This study examined the relationship between synergistic supervision and job performance indicators (core self-evaluation and supervisor-related commitment) with midlevel administrators in student affairs. The following questions were examined:
Question 1: To what extent are the 22 behaviors associated with synergistic supervision perceived to be practiced by the supervisors of midlevel student affairs administrators? 1A) To what extent is the length of the supervisory relationship associated with the perceived levels of the 22 synergistic supervision behaviors? 1B) To what extent is the gender make-up of the supervisory dyad associated with the perceived levels of the 22 synergistic supervision behaviors?
Question 2: What is the relationship between the perceived level of synergistic supervision received by midlevel student affairs administrators and core self-evaluation? 2A) To what extent is the length of the supervisory relationship associated with the relationship between perceived level of synergistic supervision received and core self-evaluation? 2B) To what extent is the gender make-up of the supervisory dyad associated with the relationship between perceived level of synergistic supervision received and core self-evaluation?
Question 3: What is the relationship between the perceived level synergistic supervision received by midlevel student affairs administrators and supervisor-related commitment? 3A) To what extent is the length of the supervisory relationship associated with the relationship between perceived level of synergistic supervision received and supervisor-related commitment? 3B) To what extent is the gender make-up of the supervisory dyad associated with the relationship between perceived level of synergistic supervision received and supervisor-related commitment?
The population was midlevel student affairs administrators who were members of Student Affairs Administrators in Higher Education (NASPA) [
Appropriate permissions were obtained from the respective authors to use the three instruments: Synergistic Supervision Scale (SSS) [
Prior to initiating the research, the authors addressed the ethical issues of participant confidentiality, informed consent processes and secured IRB approval. The lead researcher then sent an e-mail to each of potential participants. The e-mail explained the study’s purpose, described participation incentives (30 randomly selected participants would each receive a $10 Amazon gift certificate), asked participants to access and complete the online survey via a link, and suggested the estimated time to complete the survey. Accessing the survey included accessing the informed consent page and confidentially statement. The online survey included demographics questions, the SSS, the CSES, and the SRCS. Demographic items included gender, geographic area of employment, years reporting to supervisor, supervisor’s position level, institution type, and functional area of employment. At the completion of the survey, participants provided their name and e-mail address if they wished to participate in the drawing. This information was not connected to the actual data submitted. A reminder e-mail was sent 8 days later to participants who had not completed the survey. The survey site was closed 16 days after the first email. Data was collected using the Campus Labs Baseline program.
Synergistic supervision was most often used as the independent variable, with scores on the SSS being used to operationalize the variable in most research questions. The dependent variable changed based on the question. Research Question 2 used the scores on the CSES as the dependent variable to examine the relationship between core self-evaluation and synergistic supervision. Research Question 3 used scores on the SRCS as the dependent variable to examine the relationship between supervisor-related commitment and synergistic supervision. All questions featured certain demographic characteristics as an independent variable and examined how the characteristics were associated with scores on the SSS, CSES, and SRCS.
The data was reviewed to ensure accuracy and missing data. Before data analysis began, applicable items were reverse coded as instructed by authors of the three scales. Descriptive statistics and frequency counts were calculated to verify that all means and standard deviations seemed realistic. Descriptive statistics included measures of central tendency, variability, and position and were run to test the skewness for each variable. Results indicated skewness for the independent variable, scores on the SSS. Therefore, the Spearman rho statistic was used.
For Question 1, the means for each of the 22 behaviors were examined and a mean score of 4 or above suggested that the behavior was perceived to be practiced at a meaningful rate. For Question 2, Pearson correlation coefficients of the means of the composite scores of the SSS and CSES were computed to determine if a relationship existed between the two variables. Since direction of correlations was unknown, a two-tailed test was completed with statistical significance indicated at p < 0.05. For Question 3, Pearson correlation coefficients of the means of the composite scores of the SSS and SRCS were computed. As the direction of correlations was unknown, a two-tailed test was used with statistical significance indicated at p < 0.05. To answer questions exploring the impact of the demographic characteristics on the relationship between synergistic supervision and performance indicators, the demographic variable was broken down into sub-categories. Correlation coefficients were used to determine if a significant relationship existed and statistical significance (p < 0.05) suggested a relationship existed for the sub-category of the identified variable. Alpha coefficients were calculated using the data collected during this study and needed to be positive and greater than .70 to ensure the reliability of the instrument [
Of the 2316 individuals contacted, 770 responded. Of those, 66 surveys were eliminated. Sixty-four surveys were incomplete and two participants were outliers related to gender. One person identified as transgender another identified as “other.” These two surveys were not included in the analysis because a subgroup containing only one person is too small to have statistical significance. The final count was 704 which means a response rate of 30.3%.
The following demographic information was obtained: gender of participant; gender of participant’s supervisor; institutional type of employment, institutional size; functional area of responsibility; and number of years reporting to supervisor. A majority of the participants indicated they were female. A majority of the respondents (N = 405) reported to work for a female supervisor. See
Gender/Perceived gender identity of supervisor | N | % |
---|---|---|
Female Participants | 474 | 67.3 |
Male Participants | 230 | 32.7 |
Participants reporting to female supervisor | 405 | 57.5 |
Participants reporting to male supervisor | 299 | 42.5 |
Participants indicated the type of institution where they were employed and a majority indicated a 4-year institution. See
Participants’ Type of Institution | N | % |
---|---|---|
4-year public institution | 404 | 57.4 |
4-year private institution | 255 | 36.2 |
2-year institution | 246 | 34.9 |
Institutional enrollment size was also a demographic item. The size of the institution and participant numbers are indicated in
Enrollment | N | % |
---|---|---|
1999 and under | 58 | 8.2 |
2000 - 4999 | 118 | 16.8 |
5000 - 9999 | 104 | 14.8 |
10,000 - 20,000 | 178 | 25.3 |
Over 20,000 | 246 | 34.9 |
Participants indicated their geographic area of employment The largest group of participants identified their place of employment in the Mid-Atlantic States. See
Geographic Region | N | % |
---|---|---|
Mid Atlantic (DE, MD, NJ, NY, PN, VA, WV) | 150 | 21.3 |
Southeast (AL, AR, FL, GA, KY, LA, MS, NC, SC, TN) | 133 | 18.9 |
Midwest (IL, IN, MI, OH, WI) | 109 | 15.5 |
Pacific West (AK, CA, HI, OR, WA) | 99 | 14.1 |
New England (CT, ME, MA, NH, RI, VT) | 65 | 9.2 |
Southwest (AZ, NM, OK, TX) | 58 | 8.2 |
Heartland (IA, KS, MN, MO, NB, ND, SD) | 49 | 7.0 |
Mountain West (CO, ID, MT, NV, UT, WY) | 41 | 5.8 |
Participants were asked about their area of responsibility. No single area accounted for more than 18% and 25% separate areas were identified, with 16 areas having representation from at least 10 participants. See
Functional Area | N | % |
---|---|---|
Residence Life | 126 | 17.9 |
Student Activities | 96 | 13.6 |
Other | 85 | 12.1 |
General Student Affairs | 45 | 6.4 |
Multicultural Student Services | 38 | 5.4 |
Judicial Affairs | 34 | 4.8 |
Academic Support Services | 32 | 4.5 |
Advising | 31 | 4.4 |
Health/Drug and Alcohol Education | 26 | 3.7 |
Leadership Development | 25 | 3.6 |
Assessment/Research | 22 | 3.1 |
---|---|---|
Greek Life | 21 | 3.0 |
Career Planning/Placement | 19 | 2.7 |
Orientation | 16 | 2.3 |
Disabled Student Services | 16 | 2.3 |
Admissions/Enrollment Management | 15 | 2.1 |
Student Center/Union | 14 | 2.0 |
Service Learning | 11 | 1.6 |
Counseling | 6 | 0.9 |
LGBT Student Services | 5 | 0.7 |
Adult Learner Services | 4 | 0.6 |
International Student Services | 4 | 0.6 |
Religious Programs | 4 | 0.6 |
Campus Recreation/Intramural Sports | 3 | 0.4 |
Commuter Services | 3 | 0.4 |
Financial Aid | 3 | 0.4 |
The last demographic item was the number of years participants have reported to their supervisor. A small majority have reported to their supervisor less than 1.5 years. See
Time | N | % |
---|---|---|
Less than 1.5 years | 258 | 36.6 |
1.5 - 3.4 years | 249 | 35.4 |
3.5 - 7.4 years | 144 | 20.4 |
7.5 or more years | 53 | 7.5 |
Question 1: To what extent are the 22 behaviors associated with synergistic supervision perceived to be practiced by the supervisors of midlevel student affairs administrators? Participants rated their supervisor regarding the perceived frequency of synergistic supervision behaviors using a 5-point Likert-type scale (see
Question | Mean | SD |
---|---|---|
My supervisor includes me in a significant way when making decisions that affect my area of responsibilities. | 4.16 | 1.015 |
My supervisor works with me to gather the information needed to make decisions rather than simply providing me the information he/she feels is important. | 3.96 | 1.049 |
My supervisor criticizes staff members in public. (R) | 4.29 | 0.945 |
---|---|---|
My supervisor makes certain that I am fully knowledgeable about the goals of the division and institution. | 3.83 | 1.079 |
My supervisor willingly listens to whatever is on my mind, whether it is personal or professional. | 4.23 | 1.013 |
My supervisor shows interest in promoting my professional or career advancement. | 3.87 | 1.192 |
My supervisor is personally offended if I question the wisdom of his/her decisions. (R) | 3.87 | 1.143 |
My supervisor shows that she/he cares about me as a person. | 4.08 | 1.078 |
My supervisor speaks up for my unit within the institution. | 4.09 | 1.031 |
My supervisor expects me to fit in with the accepted ways of doing things, in other words, “don’t rock the boat”. (R) | 3.33 | 1.146 |
My supervisor has favorites on the staff. (R) | 3.19 | 1.363 |
My supervisor breaks confidences. (R) | 4.15 | 0.975 |
My supervisor takes negative evaluations of programs or staff and uses them to make improvements. | 3.43 | 1.002 |
When faced with a conflict between external constituents (for example parent or donor) and staff members, my supervisor supports external constituents even if they are wrong. (R) | 3.84 | 0.977 |
My supervisor is open and honest with me about my strengths and weaknesses. | 3.73 | 1.058 |
If I’m not careful, my supervisor may allow things that aren’t my fault to be blamed on me. (R) | 4.20 | 1.022 |
My supervisor rewards teamwork. | 3.50 | 1.160 |
When the system gets in the way of accomplishing our goals, my supervisor helps me to devise ways to overcome barriers. | 3.46 | 1.143 |
My supervisor looks for me to make a mistake. (R) | 4.51 | 0.857 |
My supervisor and I develop yearly professional development plans that address my weaknesses or blind spots. | 3.13 | 1.341 |
When problem solving, my supervisor expects staff to present and advocate differing points of view. | 3.52 | 1.097 |
In conflicts with staff members, my supervisor takes students’ sides (even when they are wrong). (R) | 4.05 | 0.986 |
Response options: 1 = never or almost never; 2 = seldom; 3 = sometimes; 4 = often; 5 = always or almost always. (R) = Reversed items―these items were changed before computations.
The mean for each behavior was examined and a score of 4 or above suggested that the behavior was perceived to be practiced at a meaningful rate. All 22 behaviors received mean scores above 3.0 (sometimes), 17 out of 22 had mean scores above 3.5, and 9 out of 22 had mean scores above 4.0 (often). The overall mean score on the SSS was 84.39 (N = 704), which equates to an average behavior score of 3.84. The median was 87, and mode 97. This finding suggests most of the supervisors were perceived to practice synergistic supervision at considerable frequencies. The 22 behaviors can be divided into 14 positive behaviors and 8 negative behaviors. Positive behaviors practiced most frequently included: supervisors listening to employees (mean = 4.23) and supervisors including their employees in the decision making process (mean = 4.16). Positive behaviors practiced least frequently included: developing yearly professional development plans (mean = 3.13) and using negative evaluations of programs and staff to make improvements (3.43). Negative behaviors practiced least frequently included: criticizing employees in public (reversed mean = 4.29) and looking for employees to make mistakes (reversed mean = 4.51). Negative behaviors practiced most frequently included: supervisors having favorites on staff (reversed mean = 3.19) and supervisors expecting employees to fit in with accepted norms (3.33).
Question 1A: To what extent is the length of the supervisory relationship associated with the perceived levels of the 22 synergistic supervision behaviors? The analysis consisted of sorting the respondents into four separate groups based on the length of time the participants had reported to their direct supervisor (see demographic results). The respondents in Group 3 reported the highest mean score, 85.54. Group 1 reported the second highest mean score, 85.09 followed by Group 4 with a mean score of 83.79. The lowest mean score was Group 2, which reported a mean score of 83.06.
Question 1B: To what extent is the gender make-up of the supervisory dyad associated with the perceived levels of the 22 synergistic supervision behaviors? The respondents were sorted into four groups based on the gender make-up of the dyad and the mean scores were reviewed. Group 1-females to female (N = 273); Group 2-females to male (N = 201); Group 3-males to female (N = 132); and Group 4-males to males (N = 98). Mann-Whitney U tests comparing the scores on the SSS examined differences between genders and results indicated higher mean ranks for females but no statistical significance existed. Female supervisors received higher overall mean scores (84.75) on the SSS than male supervisors (83.91). Data also indicated that female administrators provided higher overall mean scores (85.04) for their supervisors than the male administrators provided (83.06) for their supervisors. Results also found that those relationships featuring a male reporting to another male had the lowest reported overall mean score on the SSS (82.15) and those relationships featuring a female reporting to another female had the highest overall mean score on the SSS (85.32).
Regarding the relationship between synergistic supervision and key performance indicators, participants responded to the CSES and the SRCS with the purpose to examine the relationship between synergistic supervision (independent variable) and core self-evaluation and supervisor-related commitment variables (both dependent variables).
Question 2: What is the relationship between the perceived level of synergistic supervision received by midlevel student affairs administrators and core self-evaluation? On the CSES, participants rated themselves regarding their composite personality traits of four core evaluations using a 5-point Likert-type scale (see
Question | Mean | SD |
---|---|---|
I am confident I get the success I deserve in life. | 3.80 | 0.859 |
Sometimes I feel depressed. (R) | 3.12 | 1.204 |
When I try, I generally succeed. | 4.33 | 0.558 |
Sometimes when I fail I feel worthless. (R) | 3.38 | 1.159 |
I complete tasks successfully. | 4.38 | 0.544 |
Sometime, I do not feel in control of my work. (R) | 2.77 | 1.153 |
Overall, I am satisfied with myself. | 4.12 | 0.746 |
I am filled with doubts about my competence. (R) | 3.85 | 1.010 |
I determine what will happen in my life. | 3.91 | 0.807 |
I do not feel in control of the success in my career. (R) | 3.80 | 0.996 |
I am capable of coping with most of my problems. | 4.26 | 0.656 |
There are times when things look pretty bleak and hopeless to me. (R) | 3.94 | 1.064 |
Response options: 1 = never or almost never; 2 = seldom; 3 = sometimes; 4 = often; 5 = always or almost always. (R) = Reversed items - these items were changed before computations.
Question 3: What is the relationship between the perceived level synergistic supervision received by midlevel student affairs administrators and supervisor-related commitment? Participants responded to nine statements on the SRCS regarding their direct supervisor using a 7-point Likert-type scale (see
Question | Mean | SD |
---|---|---|
When someone criticizes my supervisor, it feels like a personal insult. | 3.49 | 1.819 |
When I talk about my supervisor, I usually say “we” rather than “they”. | 4.33 | 1.786 |
My supervisor’s successes are my successes. | 4.51 | 1.707 |
When someone praises my supervisor, it feels like a personal compliment. | 3.73 | 1.758 |
I feel a sense of “ownership” for my supervisor. | 4.07 | 1.867 |
If the values of y supervisor were different, I would not be as attached to my supervisor. | 4.66 | 1.731 |
My attachment to my supervisor is primarily based on the similarity of my values and those represented by my supervisor. | 4.47 | 1.773 |
Since starting my job, my personal values and those of my supervisor have become more similar. | 3.84 | 1.668 |
The reason I prefer my supervisor to others is because of what he or she stands for, that is, his or her values. | 4.52 | 1.852 |
Response options: 1 = strongly disagree; 2 = disagree; 3 = slightly disagree; 4 = neutral; 5 = slightly agree; 6 = agree; 7 = strongly disagree.
As the data for the scores on the inventories were negatively skewed; Spearman Rho correlation coefficients were used to analyze the data. The analysis found a positive correlation (r = 0.314, p < 0.01) between the perceived level of synergistic supervision received and core self-evaluation. This correlation supports the idea that individuals who perceived their supervisor to practice synergistic supervision were also more likely to score higher on the core self-evaluation scale. The effect size is considered medium or typical [
Independent Variable | Dependent Variable | (r) | (p) | Effect Size |
---|---|---|---|---|
Synergistic Supervision | Core-Self Evaluation | 0.314 | <0.01 | Medium |
Synergistic Supervision | Supervisor Related Commitment | 0.632 | <0.01 | Large |
Scale | Mean | Median | Mode | High | Low |
---|---|---|---|---|---|
Synergistic Supervision Scale | 84.39 | 87 | 97 | 30 | 110 |
Core Self-Evaluation Scale | 45.65 | 45 | 45 | 19 | 60 |
Supervisor-Related Commitment Scale | 37.61 | 39 | 39 | 9 | 63 |
To address the sub questions under Question 2 and 3, the lead researcher used the groupings established for Question 1. Specifically for 2A and 2B, the Spearman Rho correlation coefficients of the mean scores on the SSS and CSES were used. A p value of <0.05 suggested a statistically significant relationship existed between the variables.
Question 2A) To what extent is the length of the supervisory relationship associated with the relationship between perceived level of synergistic supervision received and core self-evaluation? Positive correlations between the perceived level of synergistic supervision received and core self-evaluation were indicated in all groups (see
Length of Supervisory Relationship | (r) | (p) | Effect Size |
---|---|---|---|
Under 1.5 years | 0.261 | <0.01 | Medium |
1.5 - 3.4 years | 0.250 | <0.01 | Medium |
3.5 - 7.4 years | 0.439 | <0.01 | Large |
Over 7.5 years | 0.395 | <0.01 | Medium to Large |
Question 2B) To what extent is the gender make-up of the supervisory dyad associated with the relationship between perceived level of synergistic supervision received and core self-evaluation? The analysis indicated positive correlations between synergistic supervision and core self-evaluation in all groups (see
Gender Make-up of the Dyad | (r) | (p) | Effect Size |
---|---|---|---|
Female Employee with Female Supervisor | 0.288 | <0.01 | Medium |
Female Employee with Male Supervisor | 0.297 | <0.01 | Medium |
Male Employee with Female Supervisor | 0.356 | <0.01 | Medium |
Male Employee with Male Supervisor | 0.469 | <0.01 | Large |
As before, the lead researcher used the groupings established for Question 1 to examine the Questions 3A and 3B. The Spearman Rho correlation coefficients of the mean scores on the SSS and SRCS were used. A p value of <0.05 suggested a statistically significant relationship existed between the variables.
3A) To what extent is the length of the supervisory relationship associated with the relationship between perceived level of synergistic supervision received and supervisor-related commitment? The analysis indicated large positive correlations between the perceived level of synergistic supervision received and supervisor-related commitment in all groups (see
Length of Supervisory Relationship | (r) | (p) | Effect Size |
---|---|---|---|
Under 1.5 years | 0.607 | <0.01 | Large |
1.5 - 3.4 years | 0.594 | <0.01 | Large |
3.5 - 7.4 years | 0.723 | <0.01 | Very Large |
Over 7.5 years | 0.639 | <0.01 | Large |
Question 3B) To what extent is the gender make-up of the supervisory dyad associated with the relationship between perceived level of synergistic supervision received and supervisor-related commitment? The analysis indicated large positive correlations between the perceived level of synergistic supervision received and supervisor-related commitment (see
Gender Make-up of the Dyad | (r) | (p) | Effect Size |
---|---|---|---|
Female Employee with Female Supervisor | 0.620 | <0.01 | Large |
Female Employee with Male Supervisor | 0.622 | <0.01 | Large |
Male Employee with Female Supervisor | 0.650 | <0.01 | Large |
Male Employee with Male Supervisor | 0.627 | <0.01 | Large |
In summary, the study found that most of the synergistic supervisory behaviors were perceived to be practiced at high frequencies. The overall mean score on the synergistic supervision scale for all 704 participants was 84.39. The supervisory relationships between 3.5 - 7.4 years in length reported the highest overall mean scores (85.54) on the synergistic supervision scale. Additionally, the supervisory relationships consisting of female supervisees and female supervisors reported the highest overall mean scores (85.32) on the synergistic supervision scale.
In addition, the study found that there was a statistically significant relationship between synergistic supervision and core self-evaluation. The length of the supervisory relationship was found to influence the relationship between synergistic supervision and core self-evaluation such that a relationship was strongest after the employee and supervisor had been together for at least 3.5 years. The gender make-up of the supervisory dyad was found to influence the relationship between synergistic supervision and core self-evaluation such that a relationship was stronger for male employees than female employees.
The study also found a statistically significant relationship between synergistic supervision and supervisor-related commitment. The length of the supervisory relationship was found to influence the relationship between synergistic supervision and supervisor-related commitment such that a relationship was strongest after the employee and supervisor had been together for at least 3.5 years. The gender make-up of the supervisory dyad was found to influence the relationship between synergistic supervision and supervisor-related commitment such that a relationship was stronger for male employees than female employees.
There are some clear similarities between this study and the original study on the Synergistic Supervision Scale [
In addition, the findings of this study support the results of another study [
The findings of this study support earlier work [
Regarding synergistic supervision and indicators of performance, the results indicate a statistically significant relationship between the perceived level of synergistic supervision and core self-evaluation. The increase in core self-evaluation is important for supervision practices as it should lead to improved job satisfaction, employee engagement, and job performance. The results also support the notion that the longer the supervision relationship the stronger the relationship between the two variables to some extent as it revealed that the strongest correlation was for those supervisory dyads that had existed between 3.5 and 7.4 years. Those relationships over 7.5 years saw a small decline in the strength of the correlation. This decline could be due to the smaller N in this group.
Regarding gender, the results indicated that the largest relationship between synergistic supervision and core self-evaluation was for dyads consisting of two males and the smallest relationship was for those consisting of two females. The stronger relationships were found for those featuring at least one male employee, either as supervisor or supervisee. These findings support some prior research [
As previously addressed, the increase in supervisor-related commitment is beneficial for employees as well as institutions as it should lead to longevity with the organization, improved attendance, and enhanced job performance [
The Synergistic Supervision Scale measures perceptions not necessarily reality [
The impact of synergistic supervision on staff members and the organization as a whole warrants further investigation. A qualitative approach to understanding the impact of synergistic supervision and what staff members expect and desire from their supervisors would add to the literature. A supervisor may demonstrate the synergistic supervision behaviors, but these behaviors may not be appreciated or desired by employee. It could have been beneficial to collect that information to be able to connect the desires of employees with the positive outcomes related to performance. In addition, a study could examine annual staff evaluations or appraisals and levels of perceived synergistic supervision received. Researchers could examine the connection between supervision received and supervision given. It would be interesting to investigate whether there is any consistency between what midlevel administrators receive and the supervision they provide [
The authors declare no conflicts of interest regarding the publication of this paper.
Morgan, D. and Anderson, S.K. (2019) The Relationship between Perceived Level of Synergistic Supervision Received and Key Job Per- formance Indicators of Midlevel Student Affairs Administrators. Open Access Library Journal, 6: e5226. https://doi.org/10.4236/oalib.1105226