Keywords

Remuneration, Job-Hopping Intention, A Teaching Engagement, University.

Introduction

In recent times, the term “World-class University” has become a catchphrase in emerging nations. This may be due to the rapid growth in the establishment of new universities and the desire to be ranked among the top-notch universities in the world. As the competition to be listed among top leading universities increases, their unique competitive advantage and sustainability reside in their academics’ engagement (Rathakrishnan et al., 2016). The attraction, maintenance and retention of academics’ in higher Institutions are key strategic issues in the higher educational system (Hundera, 2014; Salako, 2014).

Even though management of universities are making conscious efforts to ensure academics’ engagement, recent studies have shown that there is a steady increase in the rate at which academics move from one university to another (Akpa et al., 2016; Agbionu et al., 2018; Demetria, 2018). Akpa et al. (2016) argued that many universities around the world had lost significant numbers of their key faculty, not only in terms of their physical removal but in their work engagement. The departure of academics especially, desirable ones do not only have a damaging effect on the institution, both in terms of replacement costs and in terms of job disruption. But also affects the engagement of those who chose to stay (Rathakrishnan et al., 2016).

Corroborating this view, Akpa et al. (2016) argued that intention to quit has a significant negative effect on the engagement of academic staff, in terms of teaching engagement and research output. This challenges the tertiary institution management on the most appropriate means of engaging academic staff. Against this backdrop, studies have identified different factors contributing to academics’ intentions to leave from one university to another (Kyaligonza & Kamagara 2017; Onah, & Anikwe, 2016; Falola et al., 2018b; Koech, & Cheboi, 2018; Maneno, 2018; Demetria, 2018), but there is no consensus on basic factors influencing academics’ engagement especially in higher learning institutions.

In response to the challenges facing higher education in Africa, Kyaligonza & Kamagara, (2017) revealed poor management supports such as poor facilities, limited access to publishing facilities, poorly paid staff and lack of research grants as responsible for a high level of brain drain among academics of universities. Also, Imhonopi & Urim (2013) observed inadequate research support as the key factor that has stunted the growth of research in the Kenyan university system.

Others focused mainly on the perceived influence of career advancement on employee intention to stay or leave. Mustapha & Zakaria (2013) argued that employees who perceived a lack of fairness in promotion criteria might not likely be engaged, even though physically available. Falola et al., (2020a) studies the influence of management supports on the faculty engagement in Nigerian universities. In the same manner, Selesho & Naile (2014) identified poor working conditions and heavy workload, which is making it difficult for academics to meet their promotion requirements. Although these studies are indicative of the factors that determine academics intention to either stay or leave their current universities, there is a shortage of study that focused on remuneration as a determinant of job-hopping among academics of universities, especially in the Nigerian context. It is against this background that this study examined how remuneration can enhance job-hopping of academics and the implication on their teaching engagement.

In order to achieve the set objective of the study, the first part of the study focused on the background to the study while the second part focused on the review of literature, as well as a theoretical framework, the third part of the study focused on the methodology adopted, while data analysis, discussion of findings, conclusions, recommendations and area for further studies were captured in the last part of the study.

Research Hypothesis

H_o There is no significant relationship between remuneration and academics’ teaching engagement

Literature Review

Determinants of Job-hopping

Alzayed & Murshid (2017) examined the determinants of employees’ intention to quit and group them into three main categories. These include behavioural determinants, such as psychological contracts, work satisfaction and insecurity. The second type is economic determinants, which also include: pay and career advancement. The third category is demographic determinants, such as age, gender and educational qualification. Similarly, Al-Marri et al. (2018) established factors such as innovative remuneration systems, access to employee benefits, a secure work environment, career advancement opportunity, job security and recognition among others as determinants of employees’ intention to quit or stay.

Similarly, studies have also reflected on factors, such as employee's own attitude, organizational structure, well-matched job demands, remuneration system, job satisfaction, experience and knowledge acquired from the organization, social support and organizational policies as determinants of employee mobility (Lake et al., 2018).

Remuneration and Job-hopping

Remuneration is an attribute that cannot be ignored when referring to employees’ intention to leave or stay in an organization. It involves all financial rewards such as salary received at the end of performing a job, fringe benefits, and bonuses on workload and promotion benefits (Nyamubarwa, 2013; Ahmad & Ibrahim, 2015). Prior scholars have also investigated the influence of remuneration on employees’ intention to quit and agreed on the fact that remuneration serves as a motivating factor for employees’ intention to move from one organization to another (Al Mamun & Hasan, 2017).

In the academic environment, Rathakrishnan et al. (2016) revealed poor remuneration as a determinant of academics' mobility. Higher pay is often cited as the main reason why workers frequently change jobs. Employees find their way to higher levels of their career ladder, as each change of job carries with it pay increase and higher position (Al Mamun & Hasan, 2017). Singh and Loncar (2010) pointed out those payment disparities between employees with the same skills, qualifications and experiences, increase turnover intention and real turnover.

A study conducted by Chepchumba (2017) shows a negative association between remunerations and employees’ intention to quit. Internal rewards equity has also been described by Giunchi et al., (2016) as a driving force in employee turnover intentions. As puts it “When two or more employees do similar work and have similar responsibilities, pay differential can lead to the departure of lower-paying employees”. This is reinforced by the Expectancy Theory, as discussed in the theoretical context of the study.

Giunchi et al. (2016) reveals that employee engagement is characterized by how the employees think or view the remuneration they receive. With the rising numbers of higher education institutions, dissatisfied academics are willing to move from one university to another that offers higher remunerations. In support of the literature, Giunchi et al. (2016) revealed a negative relationship between salary and employees’ intention to quit. When there is lack of equity in terms of remuneration, academics’ intention to quit tends to increase, thereby affecting their level of commitment and level of engagement (Rathakrishnan et al., 2016). Thus, the need to further examine how remuneration determines job-hopping among academic staff, as well as perceived influence on teaching engagement.

Academics’ Teaching Engagement

Academics engagement in any University is characterized by three major functions: research, teaching and community engagement (Okpe et al., 2013; Sulaiman, 2018; Mutabuzi, 2019). Higher education institutions have a longstanding tradition of teaching, research and service. The recognition and advancement of academic staff in any university, rest largely on the quantity and quality of these key indicators and they also determine the performance and ranking of a University (Mushemeza, 2016)

Part of the key performance indicators of academics is teaching engagement. The findings of several studies have revealed this as a performance indicator of academics (Kasule et al., 2016; Ozurumba & Amasuomo, 2016; Atwebembeire & Malunda, 2019). Teaching engagement is measured in terms of length of teaching, delivery quality, coverage of course outline, number of courses taught as well as overall students’ performance (Collinson, 2000; Kasule et al., 2016; Mushemeza, 2016). The exact impact of teaching engagement on student learning and achievement has been debated over the past several decades by many educators, researchers and policymakers (Kasule et al., 2016).

Similarly, a variety of studies have shown a positive relationship between teaching engagement and students’ academic achievement (Ozurumba & Amasuomo, 2015; Walters & Openjuru, 2016). The teaching function includes the planning and delivery of lectures, the supervision of the final year student project and the grading of scripts. Other roles include the creation and promotion of new teaching techniques, student consultation and the production of teaching materials for students (Agbionu et al., 2018). Despite the importance of teaching engagement of academics to all stakeholders of the universities. Walters & Openjuru (2016) indicated that most academics fail to deliver on teaching engagement. Often time, they interact with students only half of the expected contact hours or are not regular in class. Hence this study examined the perceived effect of job-hopping intention on academics’ teaching engagement.

Theoretical Review

Equity theory

Adams (1965) propounded Equity Theory and argued that employee usually intends to maintain equity between inputs (effort) in the job (in the form of time, education, experience and commitment) and the outcomes in turn from the job (promotion, recognition, and pay rise) compared to perceived inputs and outcomes from other employees in the same organization or in other organizations. The theory further suggests that, if employees perceive to be under-rewarded, he or she will be demotivated and develop the intention to leave the organization. Also, if dissatisfied employees remain in the organization, they may perform by withholding their efforts in order to lower the quality or standard of that organization (Memon et al., 2017)

Olusola & Nathaniel (2019) pointed out that the inability to find equity leads to a variety of actions, one of which could be to leave the organisation. They also pointed out that the keystrength of this theory is that it acknowledges that individual inputs such as education, experience, effort should be recognized in such a way that equity is experienced. Also, when employees failed to achieve equity, they tend to target their hostility towards the organization, thereby producing negative behaviour in the hope of restoring justice. In relation to this study, Equity Theory guides in understanding what may influence academic staff to leave their jobs.

This theory, therefore, provides an understanding of the factors that make the academic staff leave their university by keeping in comparison to what academic staff in other institutions is benefiting to realize the ratio of input-output equilibrium. However, Kim (2017) suggested that the main weakness of this theory is the subjectivity of the comparison process. The study found that there is a propensity in human nature to distort their inputs, particularly in terms of effort. It is therefore difficult to compare their level of engagement to work.

This theory has been largely applied to motivation in organizations outside the academic environment. On the basis of this gap, the current study seeks to enquire how the Equity theory can be used to promote academic staff engagement or further strengthen existing staff engagement strategies in Nigerian public and private Universities.

Methodology

The study examined the relationship and resultant effect of remuneration on academics’ teaching engagement of selected universities in Southwest Nigeria. In assessing the subject matter, a survey research design was adopted. Remuneration was measured using the consistent payment of salary, equitable salary, payment of allowance and fringe benefits, priority on staff welfare and dissatisfaction with salary structure. This study focused on six (6) universities, two each, federal, state and private universities, using purposive sampling technique, influenced by year of establishment and top-ranked in their category, as appeared in webometrics ranking of Nigerian universities in 2019.

A sample size of six hundred and twenty (620) was derived using a published table by Gill, Johnson and Clark in 2010. The sample was further spread across selected universities using proportionate allocation formula by Bowler in 1996, as cited in Agbionu et al. (2018). Five hundred and forty-five (545) copies of the questionnaire reflecting (87.9%) response rate were returned and used for this analysis. The data collected was analyzed using Structural Equation Modeling (SEM)-Partial Least Square (PLS) to determine the impact of one variable on the other. Reliability and fitness were also carried out, while convergent and biased analyzes were used to determine construct validity.

The multistage sampling techniques which include purposeful (because it focused only on academics), stratified (because it cut across all academic strata) and simple random (because every academic staff was given equal opportunity of been selected) were used. The 5-point Likert scale was used to obtain responses from participants. This helps to assess the degree to which the respondents or participants agree with the specific item in the instrument.

Descriptive Statistics

Descriptive statistics was adopted to describe and interpret empirical data in a statistical form. Tables and Figures were used to present the data in a logical manner for concise understanding and interpretation. Five Likert scales were used for the descriptive statistics ranging from Strongly Agreed (5), Agreed (4), Undecided (3), Disagreed (2) and Strongly Disagreed (1). The most important thing required is to determine the level of satisfaction, as recommended by Mohammed (2016). The five scores of the answer were divided into five different levels of the Likert scale as calculated and presented in Table 1.

Table 1: Mean Score Criteria For Five (5) Point Likert Scale
Scale	Mean Score Categorization	Level of Satisfaction
1	1.00 – 1.80	Strongly Disagreed
2	1.81– 2.60	Disagreed
3	2.61 – 3.40	Undecided
4	3.41 – 4.20	Agree
5	4.21 – 5.00	Strongly Agreed

As depicted in Table 1, if the mean score is between 1.00-1.80, it suggested that the respondents strongly disagreed with the item on the research instrument, if the mean score is between 1.81- 2.60, it implies that respondents disagreed with the items on the questionnaire and if the mean is between 2.61- 3.40 it suggests that the respondents were indifferent, the mean score that ranges from 3.41-4.20 suggests that the respondents agreed with the statement while the means score that ranges from 4.21-5.00 indicates that the respondents strongly agreed with the items.

Table 2 Presents results of frequency distribution based on demographic characteristics of respondents. Regarding respondent’s gender, the total number of respondents was five hundred and forty-five (545). From this number, one hundred and forty-two (142: 26.1%) respondents were female, while four hundred and three (403: 73.9%) were male. The implication of this is that there is more male academic staff than the female staff and this suggests that the male are more likely to be involved in community engagement. Regarding respondent’s age, the findings as presented in Table 2 reveals that from 545 respondents that participated in the survey, 132(24.2%) were 20 years – 38 Years, 249(45.7%) were within the age bracket of 39-54 Years, while 55(164%) were 55 years and above. High responses of 164% were received from age 55 and above which indicates that majority of the respondents fall within the higher cadres. The years of teaching experience was also sought by the researcher. The findings revealed that majority of respondents have spent seven (7) years and above, which suggests that most of them have acquired much experience in teaching, research and community engagement.

Table 2: Demographic Characteristics Of Respondents
Characteristics		Frequency	Percentage
Gender	Male Female	403 142	73.9 26.1
Marital Status	Single Married Other	63 480 2	11.6 88.1 0.4
Age	20 – 38 years 39 – 54 years 55 – 73 years	132 249 164	24.2 45.7 30.1
Present Academics’ status/Cadre	Professor Associate Professor Senior Lecturer Lecturer 1 Lecturer II Assistant Lecturer Graduate Assistant	83 46 110 132 125 47 2	15.2 8.4 20.0 24.2 22.9 8.6 0.4
Year of Teaching Experience	0-3 years 4-6 years 7-9 years 10-12 years 13-15 years 16 years and above	4 30 124 124 90 173	0.7 5.5 22.8 22.8 16.5 31.7

In order to determine how selected institutions perceived remuneration package (Table 3), the combination of strongly agreed and agreed as expressed by the respondents shows that 148 (27.2%) of the respondents were of the opinion that the institution has good remuneration, 36 (6.6%) of the respondents were indifferent about the statement while 361 (66.3%) of the respondents had an opposing view. In addition, it was in the interest of the researcher to find out if the faculty in the selected institutions are well paid, the combination of strongly agreed and agreed as expressed by the respondents shows that 107 (19.4%) of the respondents were of the opinion and believe that they are well paid as the faculty of various institutions, 74 (13.6%) of the respondents were indifferent about the statement while 364 (66.8%) of the respondents had a contrary view.

Table 3: Descriptive Statistics For Remuneration Package
	Strongly Agree	Agree	Un- decided	Disagree	Strongly Disagree	Total
Perceived Remuneration Package
I am satisfied with the university remuneration package.	25 (4.6%)	123 (22.6%)	36 (6.6%)	171 (31.4%)	190 (34.9%)	545 (100%)
I feel this university pays well	13 (2.4%)	94 (17.2%)	74 (13.6%)	190 (34.9%)	174 (31.9%)	545 (100%)
This university offers a good benefits package compared to other University	15 (2.8%)	52 (9.5%)	137 (25.1%)	194 (35.6%)	147 (31.9%)	545 (100%)
Dissatisfaction with salary is one of the key factors undermining my engagement	62 (11.4%)	121 (22.2%)	57 (10.5%)	179 (32.8%)	126 (23.1%)	545 (100%)
The management of this university regards the welfare of staff as the first priority.	12 (2.2%)	61 (11.2%)	118 (32.3%)	195 (35.8%)	159 (29.2%)	545 (100%)

Also, to find out if the academics of these institutions have good reward benefits, the combination of strongly agreed and agreed as expressed by the respondents shows that 67 (12.3%) of the respondents were of the opinion that the university has good reward benefit for their faculties, 137 (25.1%) of the respondents were indifferent about the statement while 341(67.5%) of the respondents had an opposing view about the statement, in order to find out if the respondent's institutions have dissatisfaction with salaries. Having combining strongly agreed with agreed as expressed by the respondents, it was observed that 183 (33.6%) of the respondents were of the opinion that they have salaries dissatisfaction, 57 (10.5%) of the respondents were indifferent about the statement while 305 (62%) of the respondents had an opposing view about the statement.

However, in order to determine if there is a priority on staff welfare for faculties, the combination of strongly agree and agree show that 73 (13.4%) are of the opinion or believe staff welfare is the institution priority, 118 (32.3%) neither agree nor disagree with the statement while 354 (65%) of the respondent have contrary belief in the statement.

Table 4 shows the mean and standard deviation of each item perceived remuneration and academics teaching engagement on the research instrument across the six (6) selected University in Nigeria. The mean represents average that measures central tendency while standard deviation measures the extent of variation compared to mean. The decision rule for mean on a Likert scale of five (5) indicates that when the mean value is between 1.00-1.80, it is said to strongly disagree; the mean value between 1.81-2.60 is regarded as disagree, 2.61-3.40 it is undecided; 3.41-4.20 it is agreed; while the mean value between 4.21-5.00 is regarded as strongly agree. The standard deviation roles state that if the ratio of the standard deviation to mean is greater than 1, it indicates high variation compared to mean, but if it is less than 1, it suggests a low variation compared to mean.

Table 4: Mean Scores For Perceived Remuneration Packages And Academic Staff Teaching Engagement
Items	Uni#1	Uni#2	Uni#3	Uni#4	Uni#5	Uni#6	Mean Average
	UI	OAU	OOU	LASU	CU	BU
Perceived Remuneration Package
Equitable Salary	2.9	3.01	3.01	2.57	2.67	2.58	2.79
Consistent payment of salary	2.8	2.89	3.03	2.58	2.72	2.33	2.73
Adequate payment of allowances/ bonuses	2.88	2.93	3.24	2.58	2.48	2.13	2.71
Dissatisfaction with the salary structure	3.17	3.57	2.77	2.69	2.54	2.67	2.90
Priority on staff welfare	2.91	3.03	2.86	2.46	2.11	2.48	2.64
Average Mean for Remuneration Package	2.93	3.09	2.98	2.58	2.50	2.44	2.75

The average means score of the perceived remuneration package in Table 4 is in consonance with the frequency and percentage section. Using the criteria for understanding the mean scores of satisfaction level (see Table 1). It can be depicted that all university ranging from Uni#1 to Uni#6 were undecided with the statement (with an average mean score of 2.62) that the institution has good remuneration. Uni#3 (OOU), Uni#1 (UI) had the highest mean scores (3.01 and 2.90 respectively) the mean tendency show that the two universities neither agree nor disagree that the institution has good remuneration, Uni#2 (OAU), Uni#4, Uni#5 (CU), and Uni#6 (LASU) with a mean tendency (2.01, 2.57, 2.67 and 2.58) show that the respondents of these universities disagree except for the staff of CU what the institutions have good remuneration.

Basically, Uni#1 to Uni#6 disagrees with this statement with an average mean of 2.73, which also fall under undecided from the scale of preference. Uni#3 and Uni#2 had the highest mean scores (3.03 and 2.89 respectively) neither agree nor disagree that the institution is well paid. Uni#1 (UI) and Uni#5 (OOU) were also undecided, while Uni#4 (LASU) and Uni#6 (BU) with a mean (2.58 and 2.33 respectively) show that they disagreed with the statement that the university academic staff are well paid.

Nevertheless, Uni#1 to Uni#6 disagree with the opinion that the academics of these institutions have good reward benefits and the average mean is 2.71, which also fall under disagree from the scale of preference. Uni#3 (OOU) had the highest mean score of 3.24, and this implies that the faculty of the selected institution have good reward benefits. Uni#1 (UI), Uni#2 (OAU), Uni#4 (LASU), Uni#5 (CU) and Uni#6 (BU) had the least mean scores (2.88, 2.93, 2.58, 2.48 and 2.13 respectively) which fall under the disagree and undecided of the table of preference. The table shows that academic staff of Uni#4 (LASU), Uni#5 (CU) and Uni#6 (BU) disagreed that the faculty of the institution have good reward benefits.

In addition, Uni#1 to Uni#6 disagree with the opinion that the faculty of these institutions have dissatisfaction with salaries and the average mean is 2.57, which also fall under disagree from the scale of preference above. Uni#3 (OOU), Uni#4 (LASU) and Uni#6 (BU) had the highest mean scores (2.77, 2.69 and 2.67 respectively) are indifferent about the statement. Uni#1 (UI), Uni#2 (OAU), and Uni#5 (CU) had the least mean scores (2.17, 257, and 2.54 respectively) which fall under disagree of the table of preference. The Table 4 shows that majority of the academic staff are not satisfied with the salary structure and administration.

Therefore, all selected university disagree with the statement that the university gives priority to staff settlement with the (average mean = 2.48). Uni#1 (UI) had the highest mean scores (2.91) neither agree nor disagree with the statement that the institution gives priority on staff welfare, while Uni#2 (OAU), Uni#4 (LASU), Uni#5 (CU) and Uni#6 (BU) had the lowest mean scores (2.03, 2.46, 2.11 and 2.48 respectively) disagree with the statement.

Measurement Model for the Research Hypothesis

Both structural and measurement models were considered for data analysis. For the measurement model, all items are reflective, and the minimum acceptable value for a factor loading is 0.60 (Fornell & Larcker, 1981). Interestingly, all the constructs have values higher than 0.60. Few items that have a factor loading less than 0.5 were removed, and the results are presented in Figures 1, 2 and 3, respectively. The structural model is the inner model in structural equation modelling. It measures path coefficients (R2) values and significant values. Boots strapping method finds the significance (Vinzi et al., 2010; Sanchez, 2013). The default bootstrapping in PLS is 5000 subsamples to gain significant results (Wetzels et al., 2009). This study calculated 5000 subsamples in bootstrapping gains more precise results and path coefficient values to show the relationship between perceived remuneration and academic staff engagement in terms of teaching engagement. Results showed that selected institutions sampled had almost the same opinion. The hypothesis formulated thus:

H₀ Perceived remuneration does not affect teaching engagement of academic staff in the selected Institutions

The hypothesis has one exogenous variable (Perceived remuneration) and one endogenous variable (teaching engagement of academic staff). The coefficient of determination / r-square, the path coefficient (β value) and the T-statistic value, the size of the impact, the predictive significance of the model, and the Goodness-of-Fit (GOF) index were the core criteria for the evaluation of the structural model as presented in Figures 1, 2 and 3 respectively. All the research variables have been measured using a structured questionnaire with a five Likert scale. The perceived remuneration, which is the latent variable, was measured with five (5) items while teaching engagement of academic staff was measured with five (5) items as shown in Tables 4, 5 and 6 respectively.

Table 5: Factor Loading For Perceived Remuneration And Teaching Engagement
	Factor Loading	Error Variance	Composite Reliability	AVE	Cronbach's Alpha	No. of Indicators
Indicators	> 0.7	< 0.5	≥ 0.8	≥ 0.5	≥ 0.7
Perceived remuneration (PPO)			0.816	0.6528	0.7823	5
PR1	0.639	0.361
PR2	0.715	0.285
PR3	0.753	0.247
PR4	0.762	0.238
PR5	0.718	0.282
Teaching engagement of Academic (TE)			0.881	0.6283	0.8445	5
TE1	0.722	0.278
TE2	0.817	0.183
TE3	0.750	0.250
TE4	0.726	0.274
TE5	0.709	0.291

 

Table 6: Institutions Based Path Coefficients For Perceived Remuneration And Teaching Engagement Of Academic Staff
Variables and Cross Loading	Path Coefficient (O)	Indirect Effect (IE)	Std. Dev. (STDEV)	T Statistics (O/STDEV	P Values
Selected Federal Universitiesà Perceived remuneration	0.347		0.10	2.391	0.02
Selected Federal Universities à Teaching engagement		0.249	0.12	2.115	0.04
Selected State Universities à Perceived remuneration	0.292		0.11	2.112	0.04
Selected State Universities à Teaching engagement		0.220	0.09	2.124	0.03
Selected Private Universities à Perceived remuneration	0.268		0.11	2.451	0.02
Selected Private Universities à Teaching engagement		0.201	0.09	2.153	0.03

The items adapted for measuring perceived remuneration include equitable salary, payment of salary, payment of allowances/bonuses, dissatisfaction with salary structure, and priority on academic staff welfare. For this reason, data in this research were analyzed at the institution-level, model level and combined, using Partial Least Square-Structural Equation Modelling (PLS-SEM) technique for data analysis. PLS-SEM is often used on small sample size because this method does not consider distribution assumptions (Astrachan et al., 2014; Hair et al., 2011).

Figure 1 depicts the structural equation modelling of hypothesis 1 with standardized estimates that indicates the influence of perceived remuneration (PR0) on teaching engagement of academic staff (TE). It must be reported that factor loading depicted in Table 5 for all the items of perceived remuneration (PR) were above the minimum threshold of 0.60 and as well statistically significant at 0.05 level of significance as suggested by Fornell & Larcker (1981) & Newkirk & Lederer (2006).

Fornell & Larcker (1981) recommended the threshold for all the scales and measurement items. First, the factor loading must be above the minimum threshold value of 0.70. Second, the construct composite reliability must be equal to or greater than 0.80. Third, the construct average variance extracted estimate (AVE) must be above the minimum threshold of 0.50. Finally, the Cronbach Alpha must be equal or above 0.70 for the instruments to be reliable.

From Table 5, it can be deduced that all the constructs of perceived remuneration and teaching engagement of academic staff have values higher than 0.80 and 0.70, which means that they have composite and Cronbach Alpha reliability respectively. The factor loadings for the specific measures of construct ranged between 0.639 and 0.817. The instrument is adjudged reliable and valid since the entire requirement for the degree of fitness was satisfactorily met. None of the items had a factor loading less than 0.7, and the results of the inner structural model are presented in Figures 1, 2 and 3, respectively.

Evaluation of the Inner Structural Model

The structural model can be measured using the significant values of the path coefficient (R2). PLS-SEM was used to evaluate the path because PLS does not require any assumptions about the normal distribution. The use of bootstrapping has become critical for the determination of the level of significance (Chin, 2010; Sanchez, 2013). To achieve significant and accurate results, the default bootstrapping was conducted with 5000 subsamples (Wetzels et al., 2009). The path coefficient values of University of Ibadan (UI), Obafemi Awolowo University (OAU), Olabisi Onabanjo University (OOU), Lagos State University (LASU), Covenant University (CU) and Babcock University (BU) were presented with similar response rate. Results of structural models and path analysis for perceived remuneration (PR) on teaching engagement of academic staff (TE) have been presented in Table 5 and illustrated in Figures 1, 2 and 3 respectively.

Estimation of Path Coefficients (β) and T-statistics

The path coefficients in the PLS and the standard β coefficients in the regression analysis were related. The importance of the hypothesis was tested by the β value. The β denotes the predicted variation in the dependent construct for unit variation in the independent construct(s) (Astrachan et al., 2014). The β values of each path in the hypotheses model were determined, the higher the β value, the greater the significant impact on the endogenous latent construct.

However, the β value had to be checked for its significance level by means of the T-statistics test as shown in Figure 2.

This hypothesis predicted that perceived remuneration, which comprised perceived remuneration equitable salary, consistent payment of salary, payment of allowances/bonuses, dissatisfaction with salary structure, and priority on academic staff welfare significantly and positively influence teaching engagement of academic staff in the selected institutions. The path co-efficient (Table 6) affirmed that equitable salary #q1 indirectly and significantly influence teaching engagement of academic staff (β = 0.240, f2 = 0.178, p >0.05).

The indirect influence of consistent payment of salary was significant on teaching engagement of academic staff in the selected institutions #q2 (β=0.320, f2=0.224, p<0.05). Payment of allowances/bonuses #q3 also recorded a positive and significant impact on teaching engagement of academic staff in the selected institutions (β=0.320, f2=0.222, p <0.05). Dissatisfaction with salary structure #q4 has an inverse significant influence on teaching engagement (β=-0.112, f2=-0.086, p>0.05) while priority on academic staff welfare #q5 also have insignificant influence on teaching engagement (β = 0.189, f2 =0.150, p >0.05). Overall, the relationship between perceived remuneration and teaching engagement of academic staff in the selected institutions is confirmed to be directly significant with a beta value of 0.696, which also indicates a strong degree of association.

The path coefficient and bootstrapping of all constructs indicate significant relationships in the analysis at .05. The model found insignificant path co-efficient between equitable salary and teaching engagement of academic staff (β=0.178, Tval=2.251, p=0.01), consistent payment of salary and teaching engagement of academic staff (β=0.224, Tval=2.637, p=0.00); payment of allowances/bonuses and teaching engagement of academic staff (β=.222, Tval=3.781, p=0.01); the inverse relationship between academic staff that are dissatisfied with salary structure and their teaching engagement was also observed (β=-0.086, Tval=1.975, p=0.04); and finally, priority on academic staff welfare has a significant influence on the teaching engagement (β=0.150, Tval = 2.246, p=0.02). Hence, all path coefficients were of practical importance since the significance level is below .05. The result suggested that consistent payment of salary and payment of allowances/bonuses have the highest beta values among the constructs that best predict teaching engagement of academic staff; while there was an inverse relationship between academic staff that are dissatisfied with salary structure and their teaching engagement. The inverse relationship implies that an increase in dissatisfaction with salary structure will lead to a decrease in teaching engagement by 8.6%.

Specifically, the path analysis and bootstrapping were also developed to ascertain and assess how perceived remuneration influences teaching engagement of academic staff of selected institutions in Nigeria. This shows high predictive and explanatory power of the structural models and path analysis for perceived remuneration and teaching engagement of academic staff based on institutions.

Figure 4 indicates that the measurement model proved to be reliable as the magnitude of the factor loadings (above 0.6) were statistically significant and substantial. Thus, the model was free from bias, and the data supported the model adequacy in terms of convergence validity. In order to ascertain the predictive relevance, bootstrapping analysis was conducted. The β values of each path in the hypothesized model were determined for each institution, and the greater the β value, the greater the impact on the endogenous latent construct. However, the β value was checked for its significance level by means of the T-statistics test. The path co-efficient is presented in Table 6.

This hypothesis predicted that perceived remuneration which includes equitable salary, consistent payment of salary, payment of allowances/bonuses, dissatisfaction with salary structure, and priority on academic staff welfare significantly and positively influence teaching engagement of academic staff of selected institutions. The path co-efficient affirmed that perceived remuneration of selected federal universities indirectly and significantly influenced teaching engagement of academic staff (β=0.249, T=2.115, p< 0.04). The indirect influence of perceived remuneration was observed on teaching engagement of academic staff for selected state universities (β=0.220, T=2.124, p<0.03). Lastly, selected private universities evidently showed a significant indirect effect of perceived remuneration on teaching engagement of academic staff (β=0.201, T=2.153, p<0.00). Overall, the relationship between perceived remuneration for all the selected institutions and teaching engagement of academic staff is confirmed to be directly significant. This validates the similar finding of Falola et al. (2020b). They noted that the level of engagement of employees is a function of the management supports in terms of welfare, competitive pay and other motivational benefits that enhances engagement.

The indirect effect (MS) of the path co-efficient as depicted in Table 7 is often used to indicate the effect size of each item of the exogenous (independent) variables on the endogenous (dependent) in a PLS-SEM model. The effect size (also known as ƒ2) indicates the substantive impact of all the items. F2 measure the strength of each predictor variable in explaining endogenous variables. Cohen (1988) recommended the guiding principle for determining the effect size. The values of 0.02, 0.15, and 0.35 respectively, signify small, moderate, and large effects of an independent construct on dependent construct; while the ƒ2 values of less than 0.02 shows that there is no effect. Basically, the effect size of selected federal universities (ƒ2 = .249) is moderate; while selected state universities (ƒ2 = .220) is moderate; and selected private universities (ƒ2 = .201) is also moderate.

Table 7: Path Coefficients For Perceived Remuneration And Teaching Engagement
Variables and Cross Loading	Path Co-efficient (O)	Indirect Effect (IE)	Std. Dev. (STDEV)		T Statistics (O/STDEV	P Values
Equitable salary #q1 à Remuneration	0.240		0.09		2.556	0.01
Equitable salary #q1 à Academic Teaching engagement		0.178	0.07		2.251	0.02
Consistent payment of salary #q2 à Perceived remuneration	0.320		0.08		2.998	0.00
Consistent payment of salary #Q2 à Teaching engagement		0.224	0.08		2.637	0.00
Payment of allowances/bonuses #q3 à Remuneration	0.320		0.09		4.717	0.00
Payment of allowances/bonuses #q3 à Teaching engagement		0.222	0.06		3.781	0.01
Dissatisfaction with salary structure #q4 à Remuneration	-0.112		0.05		2.071	0.04
Dissatisfaction with salary structure #q4 à Academic Teaching engagement		-0.086	0.04		1.975	0.04
Priority on academic staff welfare #q5 à Perceived remuneration	0.189		0.10		2.346	0.02
Priority on academic staff welfare #q5 à Academic Teaching engagement		0.150	0.06		2.246	0.02
Perceived remuneration à Academic Teaching engagement	0.696		0.09		7.250	0.00
	R Square (R2)			R Square (R2) Adjusted
Perceived remuneration à Teaching engagement of Academic staff	0.484			0.457

Table 7 showed that the perceived remuneration of selected federal universities had the topmost path coefficient of β = 0.249 compared to the β values of other selected state and private universities in the model, which showed that it had a greater value of variance and high effect with regard to teaching engagement of academic staff. Whereas, the perceived remuneration of selected private universities had the least effect on teaching engagement of academic staff with β = 0.201. By implication, this means that management of selected private universities needs to develop appropriate strategies to ensure academics’ engagement. Hence, Ha indicates that the significant influence of perceived remuneration on the teaching engagement of academic staff of selected institutions in Nigeria was strongly supported.

The value of R2 in Table 7 and Figure 4 explains the variance between endogenous variables (Henseler et al., 2009; Hulland, 1999). According to Henseler et al. (2009) and Hair et al. (2011), an R2 value of 0.75 is considered substantial, an R2 value of 0.50 is regarded as moderate, and an R2 value of 0.26 is considered as weak. In this study, the inner path model of 0.64 was observed for perceived remuneration. This indicates that the five-factor loading items (perceived remuneration includes equitable salary, consistent payment of salary, payment of allowances/bonuses, dissatisfaction with salary structure, and priority on academic staff welfare) of the endogenous latent construct substantially explain 75% of the variance in perceived remuneration of the selected institutions. This simply means that 75% of the change in the perceived remuneration was due to five latent constructs in the model suggesting good explanatory power for the model.

In the same vein, the analysis showed that the indicators of exogenous (perceived remuneration) variable substantially explain 48.4% of the variability of teaching engagement of academic staff of selected universities. This simply means that 48.4% of the change in the teaching engagement of academic staff was due to five latent constructs in the model, suggesting good explanatory power for the model. In sum, the analysis provided evidence that the hypothesis is supported. Hence, the R2 value in this study was moderate. The established regression equation and model that shows the effect of perceived remuneration and teaching engagement of academic staff of selected institutions is expressed as:

Where:

Y = Teaching engagement of academic staff
ES = Equitable salary
CPS = Consistent payment of salary
PAB = Payment of allowances/bonuses
DSS = Dissatisfaction with the salary structure
PASW = Priority on academic staff welfare

Above all, the results established that perceived remuneration is a significant predictor of teaching engagement of academic staff of selected institutions.

Discussion

The relationship between perceived remuneration and teaching engagement of academic staff in the selected institutions was confirmed to be directly significant with a beta value of 0.696, which indicates a strong degree of association. The analysis showed that the indicators of exogenous (perceived remuneration) variable substantially explain 48.4% of the variability of teaching engagement of academic staff of selected universities. This simply means that 48.4% of the change in the teaching engagement of academic staff was due to five latent constructs in the model, suggesting good explanatory power for the model.

The result suggested that consistent payment of salary and payment of allowances/bonuses have the highest beta values among the constructs that best predict teaching engagement of academic staff; while there was an inverse relationship between academic staff that are dissatisfied with salary structure and their teaching engagement. The inverse relationship implies that increase in dissatisfaction with salary structure will lead to a decrease in teaching engagement by 8.6%. Hence, the null hypothesis, which indicates that perceived remuneration does not have a significant effect on teaching engagement of academics, was rejected.

The findings indicate that public and private universities’ ability to engage lecturers in terms of teaching engagement is determined by the competitiveness and attractiveness of remunerations and allowances that the universities offer. To a very large extent, selected federal universities had the topmost path coefficient of β =0.347. To encourage academics’ engagement, especially those in private universities, management should provide a competitive remuneration system, to include payment of allowances and fringe benefits to the level of satisfaction, so that when they compare their rewards to their colleagues in the similar institutions under the same job, they will be encouraged to stay with their institutions.

Consistent with the findings, several studies in related contexts have supported that pay and fringe benefits have impacts on the decision of employees to either stay or leave. Accordingly, past studies have supported the idea that lower salary and insufficient financial benefit often leads to employees leaving the organization and the vice-versa (Falola et al., 2018a; Al Mamun & Hasan, 2017). Moreover, a similar analysis revealed that fringe benefits are another factor influencing an employee's decision to continue working, and it has a critical role for employees at managerial level (Al Mamun & Hasan, 2017). Similarly, the relationship between compensation and employee intention to depart was established based on a sample of 60 employees from seven Safaricom dealers in Kenya (Chepchumba, 2017). The studies are as well in tandem with Salau et al. (2020) who emphasized that motivational factors such as salary, and fringe benefits influenced staff retention. Also, complimented on the findings that competitive compensation was the basic factor to retention while its ineffectiveness renders employees to quit the organization.

Conclusions

The relationship between perceived remuneration and teaching engagement of academic staff in the selected institutions was confirmed to be directly significant with a beta value of 0.696, which indicates a strong degree of association. The analysis showed that the indicators of exogenous (perceived remuneration) variables substantially explain 48.4% of the variability of teaching engagement of academic staff of selected universities.

The result suggested that consistent payment of salary and allowances/bonuses have the highest beta values among the constructs that best predict teaching engagement of academic staff. There was also an inverse relationship between academic staff that are dissatisfied with salary structure and their teaching engagement. The inverse relationship implies that an increase in dissatisfaction with salary structure will lead to a decrease in teaching engagement by 8.6%. To a very large extent, selected federal universities had the topmost path coefficient of β = 0.347. The study, therefore, concludes that a functional relationship exists between remuneration, job-hopping and teaching engagement of academic staff of selected universities. Hence, the following recommendations.

Recommendations

(i) The pay scale and allowances need to be improved for academics, especially for those in private universities. There should be a sense of uniformity in remunerations irrespective of the university. This is to ensure that there is both internal and external equity to retain highly experienced and qualified academics.

(ii) Management of university should endeavor to promote joint appointments for faculty members through university-industry linkage, that will enable them to work both in the university and industry in the absence of the university's ability to meet their basic financial needs.

(iii) Payment for workload and overtime, as well as other fringe benefits, should be encouraged, especially in private universities, to enhance academics' engagement.

Areas for further studies

Further studies may focus on other perceived determinants of job-hopping, such as demographic factors, management supports and career prospects and how they may affect academics’ teaching engagement.

Acknowledgements

Covenant University Center for Research, Innovation and Discovery is deeply appreciated for the financial support.

References

1. Adams, J.S. (1965). Inequity in social exchange. In Advances in experimental social psychology. Academic Press.
2. Agbionu, U.C., Anyalor, M., & Nwali, A.C. (2018). Employee engagement and performance of lecturers in Nigerian tertiary institutions. Journal of Education and Entrepreneurship, 5(2), 69-87.
3. Ahmad, H., & Ibrahim, B. (2015). Leadership and the characteristic of different generational cohort towards job satisfaction. Procedia-Social and Behavioral Sciences, 204, 14-18.
4. Akpa, V.O., Soetan, T.A., Nwankwere, I.A., & Magaji, N. (2016). An investigation of influence of job satisfaction on employees intention to leave: An empirical study of selected private universities in South West Nigeria. International Journal of Advanced Engineering and Management Research, 1(5), 592-613.
5. Al Mamun, C.A., & Hasan, M.N. (2017). Factors affecting employee turnover and sound retention strategies in business organization: A conceptual view. Problems and Perspectives in Management, 15(1), 63-71.
6. Al-Marri, M.T.S., Majid, A.H.A., & Abdullah, H.H.B. (2018). Examining the effects of work values on employee commitment in UAE via literature survey. International Journal of Advances in Agriculture Sciences, 60-69.
7. Alzayed, M., & Murshid, M.A. (2017). Factors influencing employees’ intention to leave current employment in the Ministry of Information in Kuwait. Australian Journal of Business Science Design & Literature.
8. Astrachan, C.B., Patel, V.K., & Wanzenried, G. (2014). A comparative study of CB-SEM and PLS-SEM for theory development in family firm research. Journal of Family Business Strategy, 5(1), 116-128.
9. Atwebembeire, J., & Malunda, P.N. (2019). Staff development and quality education in uganda: analysis of quality teaching and research in private chartered universities. The Ugandan, 77.
10. Chepchumba, T.R., & Kimutai, B.D. (2017). The relationship between employee compensation and employee turnover in small businesses among Safaricom dealers in Eldoret municipality, Kenya. International Journal of Economics, Commerce and Management, 5(2), 490-501.
11. Collinson, V. (2000). Staff development by any other name: Changing words or changing practices?. In The Educational Forum. Taylor & Francis Group.
12. Demetria, G. (2018). Academic staff retention in private universities in southern high zone Tanzania, a strategy of attaining sustainable development. International Journal of Humanities and Social Science 8(5) 148-158.
13. Falola, H.O., Adeniji, A.A., Adeyeye, J.O., Igbinnoba, E.E., & Atolagbe, T.O. (2020a). Measuring institutional support strategies and faculty job effectiveness. Heliyon, 6(3), e03461.
14. Falola, H.O., Ogueyungbo, O.O., & Ojebola, O.O. (2020b). Workplace management initiatives and talent engagement in the Nigerian pharmaceutical industry. F1000Research, 9(699), 699.
15. Falola, H.O., Olokundun, M.A., Salau, O.P., Oludayo, O.A., & Ibidunni, A.S. (2018a). Data article on the effect of work engagement strategies on faculty staff behavioural outcomes in private universities. Data in Brief, 18, 1383-1387.
16. Falola, H.O., Oludayo, O.A., Akinnusi, D.M., Osibanjo, A.O., & Salau, O.P. (2018b). Faculty commitment, effectiveness of job responsibilities and the moderating role of institutional support: A survey data set. Data in Brief, 19, 1120-1123..
17. Fornell, C., & Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of marketing research, 18(1), 39-50.
18. Giunchi, M., Emanuel, F., Chambel, M.J., & Ghislieri, C. (2016). Job insecurity, workload and job exhaustion in temporary agency workers (TAWs). Career Development International.
19. Hair, J.F., Ringle, C.M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing theory and Practice, 19(2), 139-152.
20. Henseler, J., Ringle, C.M., & Sinkovics, R.R. (2009). The use of partial least squares path modeling in international marketing. In New challenges to international marketing. Emerald Group Publishing Limited.
21. Hundera, M.B. (2014). Factors affecting academic staff turnover intentions and the moderating effect of gender. International Journal of Research in Business Management, 2(9), 2321-2886.
22. Imhonopi, D., & Urim, U.M. (2013). Factors affecting scholarly research output in Nigeria: perception of academics in South-Western Universities. Unilag Sociological Review (USR), 10, 24-58.
23. Kasule, G.W., Wesselink, R., & Mulder, M. (2016). Professional development status of teaching staff in a Ugandan public university. Journal of higher education policy and management, 38(4), 434-447.
24. Kim, W. (2017). Examining mediation effects of work engagement among job resources, job performance, and turnover intention. Performance Improvement Quarterly, 29(4), 407-425.
25. Kyaligonza, R., & Kamagara, E. (2017). Staff turnover in public universities in Uganda. Makerere Journal of Higher Education, 9(2), 59-72.
26. Lake, C.J., Highhouse, S., & Shrift, A.G. (2018). Validation of the job-hopping motives scale. Journal of Career Assessment, 26(3), 531-548.
27. Memon, M.A., Salleh, R., & Baharom, M.N.R. (2017). The mediating role of work engagement between pay satisfaction and turnover intention. International Journal of Economics, Management and Accounting, 25(1), 43-69.
28. Mushemeza, E.D. (2016). Opportunities and Challenges of Academic Staff in Higher Education in Africa. International Journal of Higher Education, 5(3), 236-246.
29. Mustapha, N., & Zakaria, Z.C. (2013). The effect of promotion opportunity in influencing job satisfaction among academics in higher public institutions in Malaysia. International Journal of Academic Research in Business and Social Sciences, 3(3), 20.
30. Mutabuzi, H. (2019). Factors affecting academic staff retention at the college of business education in Tanzania. Unpublished doctoral dissertation, Mzumbe University.
31. Nyamubarwa, W. (2013). I am considering leaving soon turnover intentions of academic librarians in Zimbabwe. Journal of Business Administration and Education, 4(1).
32. Okpe, I.J., Simisaye, A.O., & Otuza, C.E. (2013). Research Output and Pattern of Publication among Faculty in Nigerian Private Universities: Babcock University Experience. In Information and Knowledge Management.
33. Olusola, A.I., & Nathaniel, N.D. (2019). Dimensions of pay satisfaction as predictors of work engagement among military and civilian personnel. Journal of Reviews on Global Economics, 8, 1077-1085.
34. Ozurumba, C.N., & Amasuomo, J.O. (2015). Academic staff development and output in state Universities in south-south Nigeria. Makerere Journal of Higher Education, 7(2), 49-59.
35. Rathakrishnan, T., Imm, N.S., & Kok, T.K. (2016). Turnover intentions of lecturers in private universities in Malaysia.
36. Salau, O., Worlu, R., Osibanjo, A., Adeniji, A., Falola, H., Olokundun, M., Ibidunni, S., Atolagbe, T., Dirisu, J., & Ogueyungbo, O. (2020). The impact of workplace environments on retention outcomes of public Universities in Southern Nigeria. Sage Open, 10(2).
37. Selesho, J.M., & Naile, I. (2014). Academic staff retention as a human resource factor: University perspective. International Business & Economics Research Journal, 13(2), 295-304.
38. Sulaiman, A.A., Salihu, A.N., & Peter, A.A. (2018). Work-life balance and academic staff performance in nigerian universities. Ilorin Journal of Human Resource Management, 2(1), 102-113.
39. Vinzi, V.E., Chin, W.W., Henseler, J., & Wang, H. (2010). Handbook of partial least squares (Vol. 201, No. 0). Berlin: Springer.
40. Walters, S., & Openjuru, G. (2014). Knowledge, engagement and higher education in Africa. Global Universities Network for Innovation (GUNI).
41. Wetzels, M., Odekerken-Schröder, G., & Van Oppen, C. (2009). Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. MIS Quarterly, 177-195.