Job satisfaction among university academics: Do academic rank and age make a difference?

Abstract The principal objectives of the present study were threefold—to examine the possible significant group differences in job satisfaction based on age and academic rank; to examine the extent to which university academics were satisfied with their job; and more importantly, to further validate the Minnesota Job Satisfaction Questionnaire (MSQ) among university academics in Tanzania. The study was quantitatively driven with a cross-sectional survey design in which the MSQ along with a demographic information questionnaire was administered to a convenient sample (N = 411) drawn from a total of four universities (i.e. two public and two private universities) in Tanzania. Both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were performed to examine the factor structure and reliabilities respectively. Additionally, the analysis of variance (ANOVA) was conducted to examine whether or not demographics had any difference in academics’ job satisfaction. Three key findings were established as a result of the current study. First, good psychometrics for the MSQ were obtained confirming a two-factor model with the best absolute fit indices. Second, academics were moderately satisfied with their job. Third, the two demographic variables (i.e. age and academic rank) examined, had a significant difference in job satisfaction. This study highlights some areas the university management and policy-makers of the higher education institutions can examine more closely and develop innovative approaches to enhance the job satisfaction of the university lecturers.

The landscape of the academic profession in many countries around the world has undergone a drastic deterioration since the 1980s (Escardíbul & Afcha, 2017;Kinman & Jones, 2008;Shin & Jung, 2014;Webber, 2018), especially after the introduction of managerial reforms with an exclusive focus on performance-based management, budget cuts, and efficiency-oriented management (see e.g., Fredman & Doughney, 2012, Kallio, Kallio, Tienari, & Hyvönen, 2016;Houston et al., 2006).As such, the academic profession in recent years has been pre-occupied with increased workloads as a result of massification policies emerging across many parts of the world (Ellis, 2013;Mgaiwa, 2021aMgaiwa, , 2023;;O'Meara et al. 2019).At the same time, some studies (see e.g., Fredman & Doughney, 2012;Houston et al., 2006;Mgaiwa, 2021a;Winefield et al., 2003), show that the work environment in which academics work has been getting less favourable because of the new public management reforms.In their work, Locke and Bennion (2013) call the deteriorating academic work environment as academic proletarianization.
Of particular interest, is the existence of some literature (e.g., Fredman & Doughney, 2012;Lyons & Ingersoll, 2010) showing that in terms of pay, compared to other professions, academics are paid relatively less, but surprisingly, they are satisfied with their job (Harman, 2001;Houston et al., 2006).Fredman and Doughney (2012), as well as Lyons and Ingersoll's (2010) findings, suggest that some employees may be more satisfied by factors other than pay.Some other existing studies show that job satisfaction can be explained by factors other than pay including employees' work environment (see e.g., Mgaiwa, 2021a;Røssberg et al., 2004) and leadership styles (see e.g., Mgaiwa, 2023).Overall, the work environment in which academics work differs significantly across higher education systems (HE), and so do job satisfaction (Shin & Jung, 2014).For example, in the United Kingdom and Australia, academics are less satisfied compared to their counterpart in Japan who are comparatively more satisfied (Bentley et al., 2012;Shin & Jung, 2014).Since academics' work environment differs across countries, universities, and institutional types, it is expected that job satisfaction would be different across countries, and universities as well as based on university type (i.e., public or privately owned).Although studies on academics job satisfaction are informative and illuminating, the existing studies in the context of Tanzania both academic and nonacademic (see e.g., Amani & Komba, 2016;Mgaiwa, 2021bMgaiwa, , 2023;;Mgaiwa & Hamis, 2022) have not yet examined the extent to which academics are satisfied with their job.There are some existing studies in both academic and non-academic professions (see e.g., Ashraf, 2020;Wu, & Zhou, 2020) on the role of demographics in employees' job satisfaction.As such, Escardíbul and Afcha (2017) examined among others, the role of gender on Spanish academic staff's job satisfaction.The study results did not indicate any significant differences in academics' job satisfaction by gender.Although studies on academics' job satisfaction concerning other factors including demographics are quite informative, however, no single study, particularly in Tanzania, has ever examined the role of demographics (i.e., age and academic rank) in academics' job satisfaction.Based on such research gap, the present study was pioneered to address three key research objectives: (1) To explore the possible significant group differences in job satisfaction based on age and academic rank, (2) to examine the extent to which Tanzania University academics are satisfied with their job and (3) To further validate the Minnesota Satisfaction Questionnaire (MSQ) in the Tanzanian university academics population.

Job satisfaction: a critical review
Higher education remains one of the critical pillars for the economic development of any country, especially at times when country economies are towards knowledge economies (Abrahams & FitzGerald, 2015;Wirba, 2021).As such, some scholars argue that most countries today invest a lot in research and innovation intending to reap a strong and sustainable knowledge economy (Mgaiwa, 2021b;Mihaela & Ţiţan, 2014).To achieve a country's knowledge economy, academics remain critical for propelling sustainable research and innovation.However, existing literature suggests that effective role performance and job commitment of faculty is subject to the extent to which they are satisfied with their job (Toker, 2011).Researchers have pointed out two key reasons why job satisfaction to employees is important for effective role performance.Firstly, job satisfaction is effective in fixing employees' absenteeism (Schaumberg & Flynn, 2017;Siu, 2002), addressing the issue of fluctuation in production, and also capable of solving the problem of organizational inefficiencies, such as counter-productive behaviour and sabotage (Dormann & Zapf, 2001;Mount et al., 2006) as well increasing employees' organisational commitment (see e.g., Mwesigwa et al., 2020;Nanjundeswaraswamy, 2023).Secondly, job satisfaction is relevant to employees' motivation to work (see e.g., Ayalew et al., 2019) and to those interested in evaluating such working conditions as task diversity, work requirements, and communications (Demerouti, 2006).Smith et al. (1969), as well as Locke (1976), suggested that pay, promotions, recognition, co-workers, supervision, the company and its management, the work itself, and working conditions are fundamental factors in determining employees' job satisfaction.
In the context of the present study, job satisfaction refers to the positive and favourable attitudes and feelings that academics have about their jobs (Armstrong, 2006).Researchers have suggested numerous models to elucidate why some workers are more satisfied than others.To explain why some employees are more satisfied than others, Herzberg (1987) in his Two Factor Theory contends that numerous workplace factors promote job satisfaction, while other factors cause dissatisfaction.In their Job Characteristics Model (JCM), Hackman and Oldham (1980) proposed five key job features that affect five work-related outcomes-task identity, skill variety, autonomy, task significance, and feedback.The Job Characteristic Model further postulates that these characteristics affect five other job-related outcomes-i.e., satisfaction, motivation, absenteeism, performance, and turnover.Similarly, Locke's (1976) Value Percept Theory postulates that job satisfaction is contingent on whether an employee feels that the work offers ideal pieces of stuff.
Job satisfaction is essential in virtually all organizations, including educational organisations.It influences employees ' organizational commitment (Ayalew et al. 2019;Breaugh et al., 2018), in the sense that satisfied workers are more apt to stay dedicated to the job and the institution.Empirical evidence has likewise revealed a strong positive relationship between a satisfied employee and that employee's organisational commitment and retention (O' Leary-Kelly & Griffin, 2013).For example, Riggio (2009) found that the nature and variability of work, quality of the social affairs at work, freedom involved in the job, returns and remuneration, promotion opportunities, and professional development in the organisation are factors that affect employees' job commitment.Available evidence (e.g., Platis et al., 2015) further indicates a positive relationship between employees' job satisfaction with job performance.This again suggests that, for employees to register better performance (as prescribed in their job descriptions), job satisfaction is essential.
Today, a key challenge for policymakers in the higher education sector in Africa and probably many other developing countries is a shortage of academics, especially in higher ranks.For a long time, research has pointed to academics' brain drain (see e.g., Adeyemi et al., 2018;Altbach, 2013;Fussy, 2017;Odhiambo, 2013) as one of the major challenges of universities in developing countries as they move to developed countries to search for green pastures and favourable work environment (Odhiambo, 2013).Consequently, academics in these countries are dominated by those with lower academic ranks and fewer in higher academic ranks such as professorial ranks (Fussy, 2017;Mgaiwa, 2021b;Peter, 2018).As such, the few in higher academic ranks are those with ordinary academic profiles and profiles because most of the best African academic talents emigrate to developed countries where they are better paid.Academics also move from academia to other non-academic jobs which seemingly are either more paying, or they are more fulfilling than academia.For example, in Tanzania, existing literature shows that at the University of Dodoma alone, a total of 102 academics left the university appointments either by resignation or termination within five years (Hokororo & Michael, 2019) as cited in Nyahongo (2015).Similarly, at the University of Dar es Salaam, the country's flagship university, 72 university lecturers left the university appointments within four years (Amani & Komba, 2016).Scholars have attributed the existing academic turnover in Tanzania and elsewhere in developing countries to a global decline in the attractiveness of academic jobs owing to an increase in work stress, reduced academic freedom and autonomy, increased workload as a result of HE massification policies currently in place as well as comparatively low pay (Hokororo & Michael, 2019;Nyahongo, 2015).
Although universities by their very nature are supposed to operate within institutional autonomy and academic freedom of their faculty members, the key professional challenge for university faculty in Tanzania is about the limitations to academic freedom and university autonomy.Since the inception of the first university institution in the early 1960s to date, universities have witnessed some elements of state control due to fear of losing political control if the academia will criticise the government in power.Because of this fear, currently, universities have witnessed controlled public academic debates and the introduction of censorship laws that appears to limit the use of statistics.At times, the use and publication of statistics are only possible upon government endorsement.There are several cases where both overseas and local lecturers were dismissed from university appointments for criticizing government policies (Ngirwa et al., 2014).This state of the art not only repressed the academic freedom of the research community and institutions but may also have stemmed in the production of a questionable type of research, which consequently may affect research independence and objectivity as well as opportunities for a research partnership with other local and foreign research organizations.Generally, constraints to academic freedom and autonomy may affect academics' job satisfaction, which may consequently restrict their efficiency and institutional effectiveness.

Age and job satisfaction
Extant research has established that employees' age influences their job satisfaction.For instance, in the United States, Lee and Wilbur (1985) examined the association of age with employees' job satisfaction in a sample of 1707 public workers.The study findings revealed that job satisfaction increased with age suggesting that older employees were more satisfied than their counterparts.Supporting earlier studies, Bishay (1996) established that employees' job satisfaction increased with age, while Van Maele and Van Houtte (2012) and Eliofotou-Menon and Athanasoula-Reppa (2011) pointed out that younger employees are associated with low experience in handling workrelated stress and workload, and therefore tend to be less satisfied than older employees.Another study among academics in North Cyprus by Saner and Eyüpoğlu (2012), examined the age-job satisfaction relationship in higher education.Employing a sample of 412 academics and Analysis of Variance (ANOVA) to understand the relationship between job satisfaction and age, results indicated that older employees had higher levels of job satisfaction than other age groups of academics suggesting a linear relationship with age.
Although previous studies have consistently indicated a small positive relationship between age and employees' job satisfaction, a study by O'Brien and Dowling (1981) among employees in metropolitan Adelaide on the influence of age on employees' job satisfaction showed that age alone did not account for employees' job satisfaction.In addition, in a study of secondary school teachers in Cyprus, Eliofotou-Menon and Athanasoula-Reppa (2011) revealed that young teachers are associated with low retention and high attrition, due to low job satisfaction, while older teachers tended to experience more job satisfaction and better correlation with both workmates and leaders, due to their age (Eliofotou- Menon & Athanasoula-Reppa, 2011).Furthermore, Saiti and Papadopoulos (2015) conducted a study on school teachers' job satisfaction as it relates to their features in the metropolitan city of Athens, Greece, and found age significantly correlated with such dimensions of teachers' job satisfaction as the type of job, administration, colleagues, and potential rewards.
Employing several control variables in their influential work, Clark, Oswald and Warr (1996) examined how employees' job satisfaction differed with age using a survey response from British employees.Findings have shown a strongly significant U-shape relationship suggesting that job satisfaction steadily decreases in the early career age and increase towards retirement age.Supporting the earlier work of Clark, Oswald and Warr (1996), and statistically controlling for supervisor and position tenure, gender, quadratic tenure terms, and affective disposition, Hochwarter et al. (2001) assessed the association between age and job satisfaction.Using a hierarchical polynomial regression to examine the age-satisfaction relationship.Their findings confirmed a U-shape relationship between age and job satisfaction suggesting that job satisfaction levels are relatively higher in the early career age, steadily decrease towards the mid-career age and increase towards retirement age.
Overall, the findings of research on the relationship between age and job satisfaction have provided mixed results, contradictory, and inconclusive.This call for further research to examine more on such relationship with careful choice of methodology while using a larger sample, particularly from Tanzania, a developing country where little research has been conducted among university academics and their job satisfaction.

Measuring job satisfaction
Several inventories have long been developed by scholars to measure employees' job satisfaction.For example, the Minnesota Satisfaction Questionnaire (MSQ-short), originally designed by Weiss et al. (1977), is a widely-used inventory for measuring workers' perceived job satisfaction.This is evident in being used by scholars in thousands of doctoral theses, master's dissertations, and research projects to produce successful results related to employees' job satisfaction (Ahmadi & Alireza, 2007;Mansoor et al., 2011;Nguni et al., 2006;Shyji & Santhiyavalli, 2014;Spitzbart, 2013;Unzicker, 2012).Originally, the MSQ short version was a product of the Work Adjustment Project carried out at the University of Minnesota.The philosophy behind this inventory assumes work suitability is contingent upon the resemblance between one's abilities and the existing supports in the work environment (Weiss et al., 1977).
However, a major difficulty in assessing employees' job satisfaction is that employees may be contented with certain dimensions of work, and discontented with others (Spagnoli et al., 2012).This has presented some challenges to many existing inventories in assessing job satisfaction, which has mainly taken two major approaches.The first is a macro perspective in which respondents are required to show their overall feelings or perceptions about their current job (Wanous et al., 1997).The second approach focuses on asking employees to reveal the extent to which they are contented or pleased with different dimensions of a job, with overall job satisfaction being determined by a summation of employees' satisfaction with several dimensions or facets.The MSQ generally measures three proportions of job satisfaction-intrinsic satisfaction, extrinsic satisfaction, and general satisfaction.One of the items representing extrinsic satisfaction is "The opportunities to do different things from time to time.
While there are other existing inventories for measuring job satisfaction, the MSQ has several advantages.For example, scholars have shown, through its psychometric properties, that the MSQ is steady over time.Moreover, earlier studies have indicated satisfactory Cronbach's coefficient alpha values, extending from .85 to .92 (Martins & Proença, 2012).However, this instrument is widely used in developed countries and in particular among hospital nurses, but it has yet to be used in East African Countries (i.e., emerging economies), particularly among academics.Therefore, this study aimed to address three research issues: Firstly, explore the psychometric properties of the MSQ among academics in Tanzania HE.Secondly, to examine the extent to which academics in Tanzania are satisfied with their academic job.Thirdly to examine the influence of demographic factors on academics' job satisfaction in Tanzania.

Research hypothesis
Three hypotheses were to be examined in light of the study's objectives: H1.Based on age and academic rank, academics' job satisfaction would differ in a statistically significant way.
H2: There would be a statistically significant higher level of job satisfaction among academics.

H3:
Job satisfaction as measured by MSQ, presents acceptable levels of reliability and validity among Tanzanian academics

Participants
The sample for the present study constituted of academics (N = 411) conveniently sampled from four universities in Tanzania-Ruaha Catholic University, the St. Augustine University of Tanzania, the University of Dodoma, and the University of Dar es Salaam.These four universities embodied both private and public universities in the country, respectively.The academics from these four universities were suitable for this study because they met the inclusion criterion of having served in their respective universities for at least one year.As such, they were in a good position to assess the extent to which they are satisfied with their academic job.Of these total participants, 75.4% (N = 310) were males while 24.6% (N = 101) were female.This sampling typically reflected the representation of sex in academia between males and females.According to available data, females constitute only 24.6% of the total academics while 75.3% of the total academics in the country are males (TCU, 2019).Of the 411 participants, 122(29.9%)were in mathematics and natural sciences while 286 (70.1%) were in humanities and social sciences.More importantly, the sample of this study constituted 2(0.5%) full professors, 4 (0.9%) associate professors, 14 (3.4%) senior lecturers, 97 lecturers (23.8%), 223 assistant lecturers (54.7%) and 68 (16.7%) tutorial assistants.Largely, this sample was a true reflection of academic profiles of academics in Tanzanian universities being more skewed towards the lower academic rank with a very little number of academics in higher academic ranks such as associate professors and professors.

Measures
To measure the academics' job satisfaction, a short form of the Minnesota Satisfaction Questionnaire (MSQ) was employed.MSQ is a widely used inventory for measuring employees' job satisfaction and has been used in a variety of research contexts and proved to be valid and reliable.Apart from collecting demographic information, all participants responded to the selfreport MSQ comprising 20 items by Weiss et al. (1977), to assess academics' job satisfaction.The participants used a seven-point Likert-type scale to rate themselves by indicating the degree to which the statement fits their level of satisfaction with their job (i.e., from 1 indicating "Absolutely disagree" to 7 indicating "Absolutely Agree").To show each participant's level of job satisfaction, scores are established by summing all scores ranging from 20 to 100.According to the scoring manual Weiss et al. (1977), the most meaningful scores to use in interpreting the MSQ are the percentile scores for each scale obtained from the most appropriate norm group for the individual group.More importantly, when percentile scores are used, the group scoring an average score of 50 or better, is regarded to be satisfied.As such, an average percentile score of 75 or better is considered to represent a higher degree of satisfaction.A percentile score of 25 or lower would represent a low level of satisfaction.Moreover, scores in the middle range of 26 to 74 would represent an average satisfaction.

Procedures
To ensure that this research complied with the ethical principles governing research, before research data collection, the author obtained ethics approval from the Human Research Ethics Committee of the University of Hong Kong (i.e., ethics approval number EA 1,805,028).The ethical approval among other things, allowed the researcher to gain access to the research sites (i.e., the four participating universities in Tanzania) in Tanzania.Before the actual fieldwork, the university management of each study university was contacted, offered research approval to the researcher for data collection, and introduced the researcher to the participants.Academics who met the criterion of having worked with the university for at least one year as academic a member of staff were contacted through their immediate supervisors, staff meetings, and individually in their offices.They were then informed of the drive of the present research.Potential participants who were willingly ready to take part in this study were then invited and offered an information sheet (instead of an informed consent form) to inform them about their rights and simplify them to choose whether or not to take part in this research.
Academics who wishfully accepted the invitation to take part in the study were given a paper questionnaire and a sheet to capture demographic information such as gender, age, work experience, marital status, the primary field of specialization, and university type.Additionally, the MSQ comprised preliminary information guiding participants on how to respond to all items.Participants self-assessed the extent of their job satisfaction with their current job.The researcher refrained from using any identifiable information in the questionnaires.As such, during the process of data analysis, the researcher used the data on a group basis in order further protect participants.Only academics with at least one year of work experience were allowed to take part in this research.On average, questionnaire completion took from 10 to 15 minutes.Finalised questionnaires were then collected from all four study universities until the desired number of surveys was attained.

Analyses
I analysed the data by using a statistical software package for social sciences (SPSS) version 25.0.Firstly, I computed the descriptive statistics to examine the total score in MSQ, range, mean, and standard deviations to assess the extent to which academics were satisfied based on the MSQ scoring manual.To establish the extent to which academics were satisfied, I employed the distributions of the cut-off scores as used in the MSQ scoring manual by Weiss et al. (1977).Based on the scoring manual, the scores in the lowest quartile below 60 were designated as "low satisfaction", while that of 60 were designated as "moderately satisfied".As such, a score ranging from 61 to 79 was designated as "moderate to not fully satisfied" and a score of 80 or higher was designed as highly satisfied.Secondly, I performed both EFA and CFA to examine the validity and reliability of MSQ among university academics.Thirdly, I conducted an Analysis of Variance (ANOVA) to examine for differences in the key research variables (i.e., job satisfaction) based on demographics (i.e., age, academic rank).

Validity and reliability of the MSQ
The first objective of this study was to validate the MSQ among Tanzanian academics by examining its psychometric properties.Therefore, both factor structure and reliability tests were examined given that this inventory was used for the first time among academics in Tanzania.Whilst exploratory factor analysis (EFA) was used to explore the factor structure of the MSQ, Cronbach's alpha coefficient values were used to estimate the reliability of the MSQ.

Scale factor structure
Based on the prior theoretical model and the scree plot (see Figure 1) as a standard criterion for the choice of components, two factors (i.e., intrinsic and extrinsic components) were forcibly extracted using principal component analysis as an extraction method with direct Oblimin as a rotation method.The two extracted factors accounted for 58.7% of the total variance in the Minnesota Satisfaction Questionnaire (MSQ).As such, the Kaiser-Meyer-Olkin measure confirmed the sampling adequacy for the analysis, KMO = 0.83.At the same time, Bartlett's test of sphericity χ2 (414) = 6514.68,p < 0.001, showed that associations between items were sufficiently large for PCA.Based on Kaiser's criterion of 1 and an analysis of the scree plot and eigenvalue (see Figure 1), two components were extracted.

Scale reliability
Based on the reliability test through Confirmatory Factor Analysis (CFA), overall, the inventory had satisfactory reliability, due to its calculated Cronbach's Alpha value of .92 with retention of 17 question items (N = 411).However, three items were excluded from the subsequent analysis owing to their loading value below .3(Worthington, & Whittaker, 2006, p. 823).According to George and Mallery (2003), all two extracted scales had acceptable internal scale reliability (i.e., intrinsic satisfaction with an alpha value of .92 while extrinsic satisfaction had an alpha value of .86).Table 1 indicates the two-factor structure and item loadings.The two-factor model is comparable to previous studies carried out in a diverse population sample (see e.g., Hirschfeld, 2000;Martins & Proença, 2012;Mgaiwa, 2021aMgaiwa, , 2023;;Negussie, & Demissie, 2013;Sharp, 2008;Saner & Eyüpoğlu, 2012Hirschfeld, 2000).Overall, these findings suggest that MSQ is a reliable inventory for assessing employees' job satisfaction across diverse population samples and cultures.

The level of academics' job satisfaction in higher education
The second aim of this study was to examine the extent to which university academics in Tanzania are satisfied with their job.It is critical at this point to delineate how the job satisfaction level is measured.According to the Minnesota Satisfaction Questionnaire scoring manual, the most meaningful scores to use in interpreting the MSQ are the percentile scores for each scale obtained from the most appropriate norm group for the individual group.More importantly, when percentile  scores are used, the group scoring an average score of 50 or better, is regarded to be satisfied.As such, an average percentile score of 75 or better is considered to represent a higher degree of satisfaction.A percentile score of 25 or lower would represent a low level of satisfaction.Moreover, scores in the middle range of 26 to 74 would represent an average satisfaction.Therefore, after converting the scores into percentile scores, the results indicated the average scores were 63 with an SD = 11.6 suggesting that academics in Tanzania's HE were moderately satisfied with their academic job according to the MSQ scoring manual.These findings are comparable to previous studies in both academic and non-academic professions (see e.g., Graeff et al., 2014;Masath, 2015;Oshagbemi, 2000bOshagbemi, , 2001)).The policy and managerial implications of these findings are discussed in Section 5 of this paper.

Differences in academics' job satisfaction based on age and academic rank
Demographic factors (i.e., age and academic rank) and job satisfaction were examined to assess the possible significant group differences in job satisfaction based on age and academic rank.To find out whether or not group differences exist, an analysis of variance (ANOVA) was performed (see Tables 2 and 3).Before conducting an ANOVA, the assumption of normality was evaluated using skewness and kurtosis and was found tenable for all groups (see Table 4).
Scholars (see e.g., Tabachnick & Fidell, 2007) argued that conventional inference tests are not always appropriate for large samples.They further argued that, in large samples, if conventional p values are used (i.e., .01 or .001), it is likely to be regarded as non-normality with only slight variations.Therefore, to avoid such weakness, in the present research with a large sample size of above 400 academics, absolute values of skewness and kurtosis were employed to establish the data normality.To determine considerable data normality for a sample size larger than 300, Mishra, et al., (2019) argued that an absolute skewness value ≤ 2 or an absolute kurtosis ≤4 May be used as a determinant of normality.Furthermore, Byrne (1998) argued that for a larger sample size, the skewness and kurtosis values less than 2.00 and 7.00 respectively would suggest that data are normally distributed.Therefore, using any of the standards stated above, suggests that all the data in this research were normally distributed.
The assumption of homogeneity of variances was also tested by using Levene's test and found tenable.To evaluate the pairwise difference among group means, post hoc multiple comparisons were conducted with the use of Tukey's Honestly Significant Differences (HSD) test since equal variances were tenable.Overall, tests revealed a significant pairwise difference between the mean scores of job satisfaction as a role of academic rank and age.Sections 4.3.1 and 4.3.2,present the differences in academics' job satisfaction based on academic rank and age respectively.

Differences in academics' job satisfaction based on academic rank
Results indicated that academic rank played a statistically significant role in academics' job satisfaction.Specifically, academics in tutorial assistant academic rank were more satisfied on the scale of intrinsic job satisfaction than those in lecturer academic rank (Mean T/Assistant = 50.68;SD = 9.84; Mean A/Lecturer = 46.37;SD = 9.23), F (5:403) = 2.56; p = 026.Results further indicated that academics in professorship rank were more satisfied with their job than those in lecturer academic rank (Mean Professorship = 69.50;SD = 8.86; Mean Lecturer = 61.82;SD = 11.70),F (5:406) = 2.76; p = 018 and assistant lecturer position (Mean T/Assistant = 62.25; SD = 11.96;F (5:406) = 2.76; p = 026.These findings are similar to Paul and Phua (2011) who revealed a group difference in job satisfaction between lecturers' job satisfaction and academic position (i.e., senior lecturers and lecturer academic ranks).These results suggest that university lecturers in lower and higher academic ranks are more satisfied than those in mid-academic ranks.One plausible reason for such difference in job satisfaction is that perhaps, academics in lower ranks have more expectations which are why they are more satisfied with their job than their counterparts in mid-academic rank.On the part of those in professorship rank maybe are more satisfied because they achieved their career plans and expectations than their colleagues in mid-academic ranks.Those in mid- academic rank are less satisfied with their jobs possibly because they are still struggling to achieve their career plans and are more likely to have more workload, more responsibilities including leadership roles and diminishing expectations as they cross the mid-career age towards retirement age with unmet expectations.

Differences in academics' job satisfaction based on age
Results in

Discussion
The present study used an Exploratory Factor Analysis (EFA) with a principal component as an extraction method and direct oblimin as a rotation method.With the help of Eigenvalues greater than 1 and the scree plot (see Figure 1), the analysis confirmed the two-factor solution of the MSQ.The two-factor structure of the Minnesota Satisfaction Questionnaire (MSQ) short form supported the results of some previous studies of two-factor model solutions (see e.g., Martins & Proença, 2012;Mgaiwa, 2021aMgaiwa, , 2023;;Saner & Eyüpoğlu, 2012).The two factors model of the MSQ were named Intrinsic Job Satisfaction and Extrinsic Job Satisfaction as in the original MSQ-short form (Weiss et al., 1967).While extant research (see e.g., Fields, 2002;Martins & Proença, 2012;Sousa et al., 2011) has testified to two-factor structures for the MSQ short form as in its original form, the findings of this study and the two-factor model reported in the current study offer further evidence on the two-factor model for MSQ-short form.The results of the validation of the MSQ-short and the underlying theory were indicated not only in a cross-cultural context of a developing country but also administered in a sample of university academics.Although the results of the factor analysis essentially confirmed the construct validity of the Minnesota Satisfaction Questionnaire for a sample of Tanzanian academics, the current study only extracted two components that accounted for 58.7% of the MSQ's total variance.However, the third factor (i.e., general satisfaction) in the three-factor model represents both internal and external job satisfaction.
Regarding the reliability of the Minnesota Satisfaction Questionnaire (MSQ), these findings suggest that the MSQ is reliable in measuring job satisfaction in the context of sampled academics in Tanzania and probably in other research samples with a similar context to that of Tanzania.As such, the internal reliability indices for MSQ were acceptable with Cronbach's alphas ranging from .92 to .86 for intrinsic and extrinsic satisfaction scales respectively.Such internal reliability indices for MSQ support the existing literature (see e.g., Dirani & Kuchinke, 2011;Lakatamitou, Lambrinou, Kyriakou, Paikousis, & Middleton, 2020;Mgaiwa, 2021bMgaiwa, , 2023)).While the MSQ was used for the first time among academics in Tanzania, its validity and reliability reported in the present study build up the confidence of this inventory in the sense that can be used in a diverse research context with a diverse group of participants including academics in developing countries such as Tanzania.
The extent to which academics in Tanzania are satisfied with their job, this article offers new insights into the job satisfaction of academics in a developing country.The moderate satisfaction of academics in Tanzania's higher education reported in this study although not low, yet shed light on what university management should do in terms of policy imperatives and practices to ensure the level of job satisfaction among academics is augmented to a higher level such that it is capable to address issues of brain drain (see e.g., Adeyemi et al., 2018;Altbach, 2013;Fussy, 2017;Odhiambo, 2013), academics turnover (Hokororo & Michael, 2019;Nyahongo, 2015) and inefficiencies in academia (Dormann & Zapf, 2001;Mount et al., 2006).The present findings on the moderate satisfaction of academics may be attributable to the long-standing debate on the ongoing academic brain drain mostly from developing countries to the developed world, especially the OECD countries where they are likely to work while contented with their job (Altbach, 2013;Cattaneo et al., 2019;Dodani & LaPorte, 2005).
Regarding the differences in academics' job satisfaction based on demographic characteristics (i.e., academic rank and age), the findings showed that academics in the senior academic rank of full professorship and those in the lowest academic rank (i.e., teaching assistants) were more satisfied than any other group.The plausible explanations for such findings could be that perhaps academics in full professorship positions are more satisfied because they have managed to climb to the topmost rank of their career position(i.e., full professorship academic rank) and so do the remuneration package or pay in such rank.Since academic rank is determined by one's publications and subsequent promotion that also determines salary level.Thus, a higher academic rank most likely means higher academic job satisfaction with promotions and salary levels.It is therefore possible to appreciate the role which higher academic rank plays in influencing the overall job satisfaction of university faculty.Of course, these results support the findings in the previous studies.For example, several scholars (see e.g., Eyupoglu & Saner, 2009;Oshagbemi, 1997) Oshagbemi (1997) and Eyupoglu and Saner (2009) revealed that academic rank had a significant difference in job satisfaction among university teachers.As such, these findings also lend support to Ssesanga and Garrett (2005) who also established that academic rank significantly predicts academic job satisfaction.Interestingly, the findings of the present study also lend support to earlier research by Oshagbemi (2000a) who revealed that female academics at higher academic ranks namely, senior lecturers, readers and professors were more satisfied with their job.For the case of teaching assistants with a higher level of job satisfaction, the possible plausible explanations could be that this group category is in the early stages of their career and therefore they have a lot of career expectations.Contrary to previous studies, the higher levels of job satisfaction among those in lower academic ranks (i.e., tutorial or teaching assistants), these findings offer a new insight that even those in lower levels of academic ranks are likely to be satisfied in the same way as those in higher academic ranks.
Similarly, in the same vein, academics with ages between 21 to 30 and 51 to 60 years as well as those with 60 and above years were reported to have higher levels of job satisfaction compared with other sampled age groups.The current study's findings are consistent with earlier studies (see, e.g., Eliofotou-Menon & Athanasoula-Reppa, 2011;Lee & Wilbur, 1985;Saner & Eyüpoğlu, 2012;Van Maele & Van Houtte, 2012), which found that employees in older age groups generally have greater job satisfaction levels than those in younger age groups.Similarly, other earlier studies (see e.g., Oshagbemi, 2003) reported a U-shape relationship between age and job satisfaction accounted for by changing individual expectations throughout their career.The results of the current study lend credence to earlier studies (e.g., Ssesanga & Garrett, 2005) that found that age strongly predicted academic job happiness.Academics with higher ages reported higher levels of job satisfaction than those in middle age.Arguably, one conceivable explanation could be that employees at a young age could feel their job more satisfying not only due to the novelty of their situations but also due to the raising unemployment rate among youth, therefore, those with jobs at a young age feel contented with their job compared to their peers.As such, perhaps older academics (i.e., aged 40 years or above) may become progressively frustrated, realizing that their aspirations and expectations are seemly more and more restricted as a result of the process of accommodation and resignation.Additionally, perhaps employees in their mid-career age may experience increased job stress and pressure from factors such as role overload, changing technologies, or an increasing emphasis on objective productivity measures (see e.g., Luthans & Thomas, 1989;Oshagbemi, 2003).On the contrary, employees as getting towards the midcareer age, acquire more information about the real nature of their jobs in comparison with their expectations and at this point, they are likely to find their jobs dissatisfying.The higher levels of job satisfaction reported at these ages between 51 to 60 years as well as those with 60+ years are likely to come from reduced ambitions and expectations after the realization that there are limited alternative existing jobs once an employee's career is founded.These findings lend support to the existing literature that employees are more satisfied as age increases especially when associated with job autonomy (see e.g., Cavanagh et al., 2020), and skill variety (see e.g., Besen et al., 2013).

Implications for university management
The present study has both practical and policy implications for university management.First, from a management point of view, these findings imply that university management needs innovative, distinctive, and accommodating strategies as well as desirable work environments that will make academics in all academic ranks and age groups feel contented with their jobs.
The findings of the present research could be suitable for university managers and leaders particularly those in charge of human resources management as they could use these findings to propose institutional policies to improve academics job satisfaction, especially in those group categories such as academics in mid-career academic ranks and those in mid-career age to ensure that all university academics enjoy their job.
From a practical point of view, as a cumulative body of work on employees' job satisfaction, the current research offers valuable advice to university policy-makers and administrators alike on the critical elements they need to address to augment faculty job satisfaction in their universities.The academic rank that can be attained by successful research production and how it consequently affects academics' job satisfaction are some of the study's findings that may be of most assistance to policy-makers and educators.In this study, I discovered that demographic parameters such as academic rank and age have the biggest effects on faculty members' job satisfaction.The present research particularly challenges the long-standing notion of linearity in employees' job satisfaction influenced by age by demonstrating and theorizing that employees' job satisfaction is influenced by age but with a non-linear relationship to an age characterized by a U-shaped and a specific increase and decrease pattern.

Research limitations
The present study had several limitations both methodological and technical.First, methodologically, the current study employed a cross-sectional research design through a single survey.Therefore, the shortcomings of the present research provide an opportunity for future research to endeavour to amass data longitudinally to assess the stability of academic job satisfaction over time.Second, the analysis of the findings was limited to data from the self-reported MSQ.Selfreported data are not always reliable in the same way as data collected across behavioural measures since the self-reported data may be overstated by socially favourite responses.Although the sample for this study was fairly reasonable, such a sample was not obtained through random sampling procedures, due to practical research constraints; therefore, future research should focus on randomized sampling procedures.

Figure 1 .
Figure 1.Scree plot for the extracted components.

Table 3 . Differences in academics' job satisfaction based on academic rank N Mean Std. Deviation Std. Error 95% Confidence Interval for Mean Minimum Maximum Lower Bound Upper Bound
Table 5 further indicated that age played a significant difference in academics' job satisfaction.Specifically, academics with age between 21 to 30 and 51 to 60 as well as 60 and above years scored higher in all two scales of job satisfaction than those aged 31 to 40 years [(Mean 21-30 years = 66.19;SD = 10.09),Mean 31-40 years = 61.44;SD = 12.07), F (4:407) = 2.56; p = .001.Similarly, those aged between 51 and 60 years as well as those with the age of 60 and above years scored higher on academics' job satisfaction than those aged between 31 to 40 (Mean 51-60 years = 65.92;SD = 8.66; Mean 61+ years = 67.26;SD = 6.19),F (4:407) = 2.18; p = 05].Table 5 present the differences in academics' job satisfaction based on age.