www.ijird.

com

May, 2014

Vol 3 Issue 5

ISSN 2278 0211 (Online)

Teaching Workload Analysis for Performance Contracting

and Service Delivery in the Academic Setting of Kenya
Philip Wambua Peter
Department of Business Administration
Human Resource Management, Kenyatta University, Nairobi, Kenya
Dr. Gorretti Ofafa
Department of Business Administration
Human Resource Management, Kenyatta University, Nairobi, Kenya
Dr. Samuel Otor
Department of Business Administration
Human Resource Management, Kenyatta University, Nairobi, Kenya
Cush L. Ngonzo
Department of Business Administration
Human Resource Management, Kenyatta University, Nairobi, Kenya
Abstract:
The study sought to analyze why there is a disconnect between the implementation of performance contracting as a management
tool in Public Universities. Empirical evidence was collected from three public universities, with consideration being put to
members of academic staff in three schools from each of the selected universities. The total staff who responded to the
questionnaire were 142. The descriptive findings showed that a greater percentage of the academic members of staff were aware
of performance contracting in the public universities and that they understood performance contracting in different versions and
terminologies. The tests of between-subjects effects established at least at 90% confidence intervals that a strong relationship
existed between the level of service delivery and the workload of the teaching members of staff of the universities selected in this
study.
1. Introduction
Traditionally, universities have defined the role of academic staff to three domains of teaching, research, and service with primary
emphasis placed upon the teaching and research aspects and secondary upon service or administration (Houston et al., 2006).
University academic staff does the complex work in an increasingly demanding environment. Universities are the only organizations
focused on dual core functions of knowledge creation and knowledge transmission through the processes of research and teaching
(Romainville,1996).The work life of university academic staff is predominantly framed and shaped by commitments and performance
in these functions.
Jenkins (2004) noted existing evidence that commitments to teaching and research can be synergistic and complementary or
antagonistic and competing. He argued that the relationship between research, teaching, broader work expectations, and rewards need
to be defined and managed at the institutional, departmental, and individual levels to avoid potentially undesirable effects and
counterproductive behaviors (Ibid). Harman(2001,2002,2003) investigated changes in academic staff roles in Australian universities
across 20-year period based on survey data gathered in 1977 and again in 1997.Leslie (2002) found that salary and job satisfaction
were uncorrelated and that faculty (who spent the majority of their time teaching) reported a preference for being rewarded for
teaching effectiveness.
Kerr (1975) noted that society hopes that (university) teachers will not neglect their teaching responsibilities, but rewards them
almost entirely for research and publications, consequently it is rational for university teachers to concentrate on research, even to the
detriment of teaching and at the expense of their students. McInnes (1999,2000) found that level of commitment remains high with
academic staff attributing this to intrinsic motivators rather than extrinsic factors such as salary and working conditions. Challenge,
variety, and autonomy are key elements of the academic staff to engage in core activities such as critical thinking, reflection, and

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 680

www.ijird.com

May, 2014

Vol 3 Issue 5

collegial interactions in the context of disciplinary interests and expertise (Winter et al.,,2000).Flexibility and autonomy are key
factors in becoming and remaining an academic (Bellamy et al., 2003).
In New Zealand, tertiary reforms have sought to refine the role of higher education and define university linkages to enhance national
economic development and to make universities more accountable to government, students as consumers, and the public generally
(Patterson,1996),while subjecting them to more centralized control mechanism. Where pursuit of the knowledge society has resulted
in increased pressures and performance expectations, workloads of academic staff have been affected directly. Coaldrake and Stedman
(1999) noted that as academic work expanded to meet growing expectations, universities and individual academics have responded
through accumulation and accretion rather than adaptation. McInnes (2000) highlighted the need to investigate workloads issues
such as increased stress on staff, development of creative solutions to ameliorate problems, and sustaining the primary sources of
work satisfaction that best promote quality. Coaldrake and Stedman (1999) noted that until recently the effect of change in
academic work has been a blind spot in policy terms for many universities and it remains so for most.
In the 1990s,research reports commissioned by university staff unions raised concerns regarding workloads and levels of stress
(Sullivan,1997).Chalmers (1998) found that staff were reporting increased stress associated with the academic work and more-workrelated illness or injuries in comparison to previous years. Consequently, workload systems management has increasingly been a
factor in recent contract negotiations and collective employment agreements.
The following sections present the methodology, key results of the study and their discussion, and a summary of the main findings and
conclusion.
2. Methodology
2.1. Research Design and Sampling
2.1.1. Research Design
The study adopted a descriptive design. Descriptive design helps to answer questions concerning the current status of the subjects
under study (Mugenda and Mugenda, 2003).Descriptive studies are aimed at finding out what is. Descriptive research can include
multiple variables for study(Borg and Gall,1996).Description emerges following creative exploration and serves to organize the
findings in order to fit them with explanations and then test or validate these explanations (Krathwoh,1998). Kothari(2009) points out
that descriptive research studies are concerned with specific predictions, narration of facts and characteristics concerning individuals,
groups or situations.
2.1.2. The Target Population
The target population comprised the seven (7) public universities. The category of staff focused on were the academic teaching staff
members ranging from tutorial fellows to professors. The total number of lecturers in the seven public universities was five thousand
six hundred and thirty(5630) (Table 1). The researcher targeted the teaching staff members since the main function of public
universities is dissemination of knowledge, research and community work which is formulated, implemented and monitored by them.
UNIVERSITY

Year of inception

Staff Members No.

Nairobi
1964
1429
Kenyatta
1985
879
Moi
1984
1286
Egerton
1987
543
Maseno
1991
320
Jkuat
1994
633
Masindemuliro
2007
540
Total
5630
Table 1: Universities Teaching Staff members schedule
Source: Human Resource Information Management Systems (2012)
2.1.3. Sampling Strategy
The researcher used purposeful random sampling since the members involved were the academic teaching staff at the main campuses
not satellite campuses, and the intention of the researcher was to find out whether performance contracting has had any positive effect
in service delivery in the state-run universities. Table 2 presents the results of such a sampling strategy.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 681

www.ijird.com

May, 2014

Vol 3 Issue 5

University

School

Total Pop.

Target (15%)

Kenyatta
University

Humanities and Social

Sciences
Business
Education
Sub-Total

205
74
151
430

30.75
11.10
22.65
64.50

66

Arts and Social Sciences

Business and Economics
Education
Sub-Total

121
72
114
307

18.15
10.80
17.10
46.05

55

Humanities and Social

Sciences
Business and Economics
Education
Sub-Total

49
30
32
111

7.35
4.50
4.80
16.65

21

848

127.2

142

Moi University

Maseno
University

Total

Actual number
of Respondents

Table 2: Teaching academic staff schedule of selected universities

Source: Field Data (Researcher 2013)
Table 2 shows that the researcher purposefully picked three universities with a total population of two thousand four hundred and
eighty five (2485) teaching academic members of staff. Kenyatta University the following was considered, schools (15) which make
up 62 academic departments. In Moi University the following was considered, schools (14) consisting of 74 academic departments
while in Maseno the following was considered ,schools (12) and one (1) faculty which make up 54 departments. The researcher chose
the biggest schools and faculties in terms of student population as evidenced by the proposed Joint Admissions Board (JAB)
admission of undergraduate students in the academic year 2012/2013.These schools were school of business and economics, school of
education and school of arts and social sciences for Moi university, while in Kenyatta university it was school of business, school of
education and school of humanities and social sciences and in Maseno it was school of business and economics, school of education
and school of humanities and social sciences .In effect,a total of 142 respondents were realized from a total of 180 questionnaires
issued to a total population of 848 academic staffers. This represented a proportion of 16.74%. The respondents to the questionnaire
were picked using simple random sampling.
2.2. Data Collection
2.2.1. Procedures for Data Collection
A self-administered questionnaire was dropped to each respondent and picked later. The questionnaire consisted of both open ended
and closed ended questions. Respondents to self-administered questionnaires are relatively unlikely to answer questions to please
(Mark et al., 2003).The questionnaire was divided into three parts covering demographic variables, practices and expectations from
performance contracting. And where additional information was required by the researcher, semi-structured interviews were
conducted.
The researcher used both qualitative and quantitative data. Qualitative data was applicable since meanings were based on expressions
through words and analysis was conducted through the use of conceptualization. Quantitative data was applicable since meanings
were derived from numbers and analysis was conducted through the use of diagrams and statistics (Mark et al., 2003).This information
was coded and analyzed with the help of statistical package for social sciences (SPSS) software package.
2.2.2. Validity of Research Instrument
The researcher employed construct validity as advocated by Cronbach(1955), in which it related the measuring instrument to the
general theoretical framework so as to determine whether the instrument was tied to the concepts and the theoretical assumptions.
Content validity was done to ascertain clarity and simplicity. The researcher used clear wording of the questions by using terms that
are likely to be familiar to, and understood by the respondents. The researcher engaged experts and his supervisors to ascertain
whether the content of the research instrument was up to standard, after which he administered it to the respondents.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 682

www.ijird.com

May, 2014

Vol 3 Issue 5

2.2.3. Reliability of Research Instrument

According to Easterby-Smith et al.(2002), reliability can be assessed by posing the following questions: will the measures yield the
same results on other occasions? Will similar observations be reached by other observers? And whether there is transparency in how
sense was made from raw data? Tabachnick and Fidel (2001) states that reliability relates to the constancy with which a measuring
instrument yields certain result, where the results of constructs measured demonstrate a high percentage of similar outcomes and is
without bias. This analysis was conducted for all statements structured on a likert point scale using Cronbach alpha score test.
Cronbachs alpha coefficient value for determining the internal consistency of the research instrument was defined by equation [3.1]:
K
2
K i 1 Yi

1
K 1
2X

[3.1]
2

Where K is the number of components (K-items or testlets), X the variance of the observed total scores and Y , the variance of
the component I for the current sample of persons. This is normally described as follows: Excellent, 0.9;Good,
0.9>0.8;Acceptable, 0.8> 0.7; Questionable, 0.7> 0.6; Poor, 0.6> 0.5; Unacceptable, < 0.5.
2.3. Data Analysis
2.3.1. Data Pre-processing and Processing
Under this section the following are examined: techniques of data analysis, issues of regression strength, problems of multicollinearity
of predictors and homogeneity of variances between predictant and predictors, and the calibration of the regression model. Regression
analysis tools were used so as to predict the results of the study. Results were interpreted and presented using descriptive and
inferential statistical methods. The following sub-sections explain the type of model specification, calibration and estimation
procedures involved in the study.
2.3.2. Model Specification
As an outcome of performance contracting, the quality of service delivery at university level is clearly assessed in terms of tangibles
(y1), reliability of services (y2), responsiveness of employees (y3), assurance given by lecturers (y4) as well as their empathy (y5).
These variables are impacted by several factors ranging from teaching workload (Objective 1) to working environment (Objective 4)
via the administrative work systems (Objective 2) and employees participation in community service (Objective 3). All these variables
and their relationships can be represented by equation [3.2]:
Yi = i+ 1jx1j+2jx2j+3jx3j+ 4jx4j + i
[3.2]
Where,
Yi are factors related to the quality of service delivery
X1j = Teaching workload factors
X2j =Administrative work systems factors
X3j = community service participation factors
X4j = the working environment factors
i = the intercept
i = the regression coefficients of the independent variables.
i = the error term
The following sub-sections deals with some econometric problems encountered in the study that could lead to biased predictions.
These encompass the testing of the multicollinearity among predictors and their homogeneity of variances with the dependent
variables.
2.3.3. Testing Multicollinearity
The multicollinearity test enabled to minimize the number of parameters involved in the model in order to generate reliable
predictions on the level of service delivery at university level within acceptable confidence limits. This model diagnostic was carried
out to rule out the assumption of high correlation between explanatory variables of different types of factors (predictants) related to the
quality of service delivered at university. Such an econometric problem was likely to lead to biased predictions of the performance of
public universities in Kenya. Consequently, one of the variables found to be highly correlated to another was removed from the model.
It shall be noted Cohen and Holiday (1998) declare that a multiple correlation coefficient is considered high when its value is equal to
or above 0.7. That was the cutting point for this study.
2.3.4. Homogeneity of Variances
Though having conducted the test of multicollinearity, an independent test for equal variances between the predictant and its
predictors would rule out any assumption of heterogeneity, which may possibly increase the presence of heteroskedasticity of errors in
the error term. Such rejection of heterogeneity would justify the use of normal distributions, namely the F and t tests. This hypothetical
homogeneity of variances was derived from an independent test of homogeneity of variances known as the Levenes test. The inferred
hypothesis test later on led to clustering the observed variables using heuristic method.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 683

www.ijird.com

May, 2014

Vol 3 Issue 5

2.3.5. Model Estimation

This stage of the study dealt with the identification of a specific model that could measure the level of service delivery at university
level. Hence, five categories of factors related to the quality of service delivery at university were selected, and their corresponding
causes (or variables) identified. The selected model can be represented by a system of equations defined by the algebraic description
below [3.3]:
Y= A+BX
[3.3]
Where,
Y = vector of predictants related to the level of service delivery
X = Matrix of predictors of the level of service delivery
B = Matrix parameters of the relevant predictors
A = vector of the model intercepts
The full regression model of the level of service delivery may be written as follows [3.4]

:
Table 3 displays the actual variables used in the modeling to predict the performance of university staffers vis--vis their contractual
academic duties.
Multiple logistic regressions (probit, logit or tobit) of these predictants by their relevant predictors were conducted to estimate the
regression parameters for each category of dependent variables (predictants). The latter were embedded in the Multivariate
Generalized Linear Model (GLM) procedure, which was run to that effect using the SPSS software package.
2.3.6. Model Evaluation and Validation
A diagnostic check-up was conducted to assess goodness of fit of the model and to rule out the presence of bias in the prediction. The
study used Pearsons Rho test to establish the correlation of various variables. The coefficient of determination (R2), the Beta weight
and the F and t statistics were also employed to get an appropriate set of parameters that determine the strength of ties between
subjects within the variables input in order to measure in the regression strength.
3. Results and Discussion
3.1. Key Descriptive of the Sample
The researcher in this section explored different parameters ranging from innovativeness, recommended work load, whether the
recommended workforce was more than recommended, the challenges faced in teaching more than recommended workload, how the
respondents overcome the challenges, the aspects of teaching workload which are hindrance to service delivery, the most attractive
aspect of the respondents work, the areas requiring innovativeness in the universities, whether the respondent has ever come up with
an innovative idea and how the customer responded to the implemented idea, the extent to which the suggested measures are effective
in custom satisfaction and whether the concept of performance contracting has improved effectiveness and efficiency in the
respondents work.These findings are presented in the following pages figures and tables.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 684

www.ijird.com

May, 2014

Vol 3 Issue 5

Table 3: Specific variables retained for analyzing performance contracting at university level
Source: Researcher, 2013

Figure 1: Management of public universities support of innovative teaching ideas

Source: Field Data (Researcher 2013)
As reflected in Figure 1 above 67.3% of the respondents expressed the feeling that there was support of innovative teaching ideas by
the management of public universities, while 32.7% felt there was no support.
Workloa
Frequency
Percent
Valid
Cumulative
d
Percent
Percent
1
1
.7
.7
10.6
2
2
1.4
1.4
12.0
3
16
11.3
11.3
23.2
4
32
22.5
22.5
45.8
5
8
5.6
5.6
51.4
6
69
48.6
48.6
100.0
Missing
14
9.9
9.9
9.9
Total
142
100.0
100.0
Table 4: Recommended teaching workload in the university per academic year
Source: Field Data (Researcher 2013)
The information in the above table4.24 reflects the responses in respect of the recommended teaching workload for each respondent in
that year. The majority (48.6%) had 6 units, while 0.7% showed those who had only 1 unit, the disparity in this could be explained by
the fact that some respondents could have been holding administrative positions either during the whole academic year or part thereof.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 685

www.ijird.com

May, 2014

Vol 3 Issue 5

Figure: 2: The teaching workload more than the recommended

Source: Field Data (Researcher 2013)
The researcher in Figure4.32 wished to ascertain whether the respondents had taught more units than recommended by the university
in that particular academic year, majority (60.0%) indicated that they had taught more, while 40.0% did not teach more than
recommended. This expresses the concern then how one is able to handle the extra units within the specified time and what the
university should then do. This also bring the question of whether this could be a contributing factor to the slow pace of PhD
completion by some staff member in certain schools.
Status
Yes
No
Total
Missing
Total

Frequency
Percent
Valid Percent
Cumulative Percent
66
46.5
73.3
73.3
24
16.9
26.7
100.0
90
63.4
100.0
52
36.6
142
100.0
Table 5: Challenges of teaching more than the recommended workload
Source: Field Data (Researcher 2013)

Table 5 draws its observations from figure 4.32,and it points out that there are challenges associated with teaching more than
recommended workforce with 73.3% stating that there are challenges and only,26.7% indicating that they did not find any challenges
It will be of great interest and research to explore this area and how universities are managing such extremes
The researcher wished to find out the challenges associated with the recommended workload, how the challenges were overcome, the
areas of teaching workload which are a hindrance to service delivery and which of the hindrances were more important, the aspects of
the teaching workload which were considered as more attractive and the areas requiring innovativeness in the public universities. The
researcher also sought to know whether the respondents had come up with any innovative idea during their period of service delivery
and implemented the same and how the customers responded to the implemented idea. Figure 3 illustrates the findings in respect of
the challenges associated with teaching recommended workload and,36.6% cited large classes,16.2% long and late teaching
hours,4.2% loss of time due to clashing time table and travels,2.1% limited time and short semesters, while 40.8% did not indicate.

Figure: 3: Challenges of teaching recommended workload

Source: Field Data (Researcher 2013)
Table 6 illustrates how the above cited challenges were overcome, 20.4% stated that they divided classes and group discussions,
20.4% stated flexible schedule and perseverance, 12.7% stated self-organization, 6.3% co-teaching and co-marking,while40.1% did
not indicate.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 686

www.ijird.com

May, 2014
Frequency

Percent

Vol 3 Issue 5
Valid
Percent
40.1
20.4

Cumulative
Percent
40.1
60.6

Missing
57
40.1
Divided classes and
29
20.4
group discussions
Flexible schedule and
29
20.4
20.4
81.0
perseverance
Self-organization
18
12.7
12.7
93.7
Co-teaching and co9
6.3
6.3
100.0
marking
Total
142
100.0
100.0
Table 6: How challenges of teaching more than recommended workload were overcome
Source: Field Data (Researcher 2013)
Figure 4 illustrates the aspects of teaching workload which are considered a hindrance to service delivery and the findings were 43.0%
cited inadequate teacher-student interaction,11.3% large material contents, 17.6% compressed teaching content, 27.5% cited
inappropriate marking of CATs and exams, while 0.7% did not indicate. Table 7 illustrates how the respondents ranked the responses
in respect of aspect of teaching workload considered a hindrance to service delivery, and43.0% ranked inadequate teacher-student
interaction as most important,11.3% ranked large material contents as most important, 17.6% ranked compressed teaching content as
most important, 27.5% ranked inappropriate marking of CATs and examinations as most important ,while 0.7% did not rank. In figure
5the researcher presents the most attractive aspect of work and26.1% stated that it was student-teacher interaction and prestige, 26.8%
learning by doing and knowledge acquisition, 8.5% ICT and automation of services, 28.9% academic freedom for research and parttiming, 6.3% socio-economic incentives,while3.5% did not indicate.

Figure: 4: Aspect of teaching workload considered a hindrance to service delivery

Source: Field Data (Researcher 2013)
Frequency
Percent
Valid
Cumulative
Status
Percent
Percent
Missing
1
.7
.7
.7
Inadequate teacher-student
61
43.0
43.0
43.7
interaction
Large material contents
16
11.3
11.3
54.9
Compressed teaching
25
17.6
17.6
72.5
content
Inappropriate marking of
39
27.5
27.5
100.0
CATs and exams
Total
142
100.0
100.0
Table 7: Aspects of teaching workload considered a hindrance to service delivery
Source: Field Data (Researcher 2013)
In respect of an area requiring innovativeness for service delivery table 8explains that 35.2% supported ICT and CCTV for exam
surveillance, 51.4% suggested innovative methods for teaching and marking exams, 3.5% suggested staff specialization, 8.5%
suggested research support, while 1.4% did not offer any suggestion.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 687

www.ijird.com

May, 2014

Vol 3 Issue 5

Figure 5: The most attractive aspect of work

Source: Field Data (Researcher 2013)
Frequency
Percent
Valid
Status
Percent
Missing
2
1.4
1.4
ICT and CCTV for exam
50
35.2
35.2
surveillance
Innovative methods for teaching
73
51.4
51.4
and making exams
Staff specialization
5
3.5
3.5
Research support
12
8.5
8.5
Total
142
100.0
100.0
Table 8: Type of innovativeness required at the university
Source: Field Data (Researcher 2013)

Cumulative
Percent
1.4
36.6
88.0
91.5
100.0

In respect of coming up with an innovative idea, Table 9explains that 54.2% had come, 39.4% had not while6.3% did not indicate. In
reference to response from customers about the implemented idea, the researcher in figure 6 established that 5.0% were not satisfied,
17.5% weresatisfied,28.8% fairly satisfied, 27.5% highly satisfied, and 21.3% excellently satisfied.
Status

Frequency

Percent

Valid
Cumulative
Percent
Percent
Missing
9
6.3
6.3
6.3
Yes
77
54.2
54.2
60.6
No
56
39.4
39.4
100.0
Total
142
100.0
100.0
Table 9: Innovativeness followed by implementation at university level
Source: Field Data (Researcher 2013)

Figure 6: Degree of customers response to implementation of innovated idea

Source: Field Data (Researcher 2013)
3.2. Test of Multicollinearity between Predictors
Suspicion of multicollinearity is strong in a regression when the Pearson correlation is equal to or above 0.7. Based on the correlation
matrix presented in Table 10, this statement was not confirmed. Pearson correlation test depicted a significant but moderate
relationship between Increase in demand of courses offered (PerCTS 13), Timely graduation (PerCTS 15) and Increase in number of
graduates (Per CTS 19) at 95% confidence interval. The relationship between all the other remaining variables was mostly significant

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 688

www.ijird.com

May, 2014

Vol 3 Issue 5

at 99% confidence interval but still weak or moderate. None of these correlations was beyond 0.69, thus providing an indication of
independence between different predictors retained in the study. Hence, the significance of their relationships could only be explained
by chance but not by collinearity. Consequently, the prediction of the quality of service delivery by workload effects could not
expectedly be tainted by a problem of multicollinearity.
3.3. Test of Homogeneity of Variances
The Levenes test confirmed the assumption of equality of error variances between the quality of service delivery measured by the
level of tangibles (EmploWS 11) and its corresponding predictors (Table 11). This test confirmed the existence of a strong relationship
between them based on their homogeneity. However, this test nullified the hypothesis of homogeneity between all other dependent
variables and their corresponding predictors. Therefore, the study confirmed the assumption that tangibles contribute to the quality of
service delivery in public universities by easing employees teaching workload.

Table 10: Correlation between independent variables for the teaching workloada
Source: Field Data (Researcher 2013)
Notes:
***. Correlation is significant at the 0.01 level (2-tailed).
**. Correlation is significant at the 0.05 level (2-tailed).
*. Correlation is significant at the 0.1 level (2-tailed).
a
Method: Pearson Correlation
Variable
Tangibles contributes to quality of service
offered at the university
Reliability contributes to quality of service
offered at the university
Responsiveness contributes to quality of
service offered at the university
Assurance contributes to quality of service
offered at the university
Empathy contributes to quality of service
offered at the university

df1

df2

Sig.

Conclusion

0.987

120

13

0.556

Homogeneity

2.726

120

13

0.022

Heterogeneity

11.012

120

13

0.000

Heterogeneity

6.856

120

13

0.000

Heterogeneity

3.485

120

13

0.007

Heterogeneity

Table 11: Results of Levene's Test of Equality of Error Variances a,b

Source: Field Data (Researcher 2013)
Notes:
a.
Tests the null hypothesis that the error variance of the dependent variable is equal across groups
b.
Weighted Least Squares Regression - Weighted by Age
c.
Design: Intercept + PerTCS13 + PerTCS14 + PerTCS15 + PertTCS16 + PerTCS17 + PerTCS18 + PerTCS19 + PerTCS20 +
PerTCS21

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 689

www.ijird.com

May, 2014

Vol 3 Issue 5

3.4. Model Estimation and Regression Strength Testing

3.4.1. Model Estimation
Table 12 presents the correlation coefficients computed between the dependent variables for the quality of service delivery at
university and the teaching workload factors (independent variables). Pearson Rho test established that the quality of service delivery
as measured by the level of tangibles (EmploWS 11) was statistically related at 99% confidence interval to Reduction in customer
complaints (PerCTS 17), Increase in customer compliments (PerCTS 18) and Increase in number of graduates(Per CTS 19).Tangibles
were also linked to Increased students enrollment(PerCTS 14) and Timely release of results(PerCTS 21) at 95% confidence interval;
and Timely graduation (PerCTS 15)at 90% confidence interval. Similarly, Reliability (EmploWS 12) was strongly correlated to
Reduction in customer complaints(Per CTS 17) at 99% confidence interval, to Increased students enrollment(PerCTS 14), Increase in
customer compliments (Per CTS 18),Increase in number of graduates(Per CTS 19) and Timely release of results(PerCTS 21) at 95%
confidence interval; and to Timely graduation (PerCTS 15)at 90% confidence interval.
The correlation test also upheld the relationship between Responsiveness (EmploWS 13) and the Increase in number of graduates(Per
CTS 19) at 99% confidence interval, EmploWS 13 and Timely graduation (PerCTS 15), Effective and efficient teaching
methodologies (PerCTS 16),Reduction in customer complaints(Per CTS 17) and Increase in customer compliments (Per CTS 18) at
95% confidence interval; and (EmploWS 13) and Timely graduation ceremonies (PerCTS 20) and Timely release of results(PerCTS
21)at 90% confidence interval. Assurance (EmploWS 14) was said to be significantly related to efficient teaching methodologies
(PerCTS 16),Increase in customer compliments (Per CTS 18) and Increase in number of graduates(Per CTS 19) at 90% confidence
interval, while Empathy (Emplo WS 15) was strongly correlated to Timely graduation(Per CTS 15) and Increase in number of
graduates(Per CTS 19) at 90% confidence interval. Hence, a strong relationship could be expected between the quality of service
delivery and the teaching workload effects.
The tests of between-subjects effects established at 90% confidence intervals that there existed a strong relationship between the level
of service delivery and the workload of the teaching members of staff of the universities selected in this study. The F test therefore
failed to reject the hypothesis stating: there is a positive relationship between employees teaching workload and the level of service
delivery in public universities in Kenya.

Table 12: Correlation between the quality of service delivery and the teaching workload factors
Source: Field Data (Researcher 2013)
Notes:
***. Correlation is significant at the 0.01 level (2-tailed).
**. Correlation is significant at the 0.05 level (2-tailed).
*. Correlation is significant at the 0.1 level (2-tailed).
3.4.2. Regression Strength
Table 13 shows that tangibles, including physical facilities, equipment and staff appearance, significantly contributed to the quality
of service delivery at the university (R2 = 0.623; Adjusted R2= 0.465; F= 3.955; Sig.= 0.000). This was explained by increased
demand for courses offered at selected universities (F=2.406; Sig.= 0.043), timely graduations ceremonies (F=2.295; Sig.= 0.065),
increased customer compliments (F=4.472; Sig.=0.002) and increased number of graduates (F=4.111; Sig.=0.004). Secondly,
reliability, that is the ability to perform service dependably and accurately, significantly contributed to level of service delivery at
these universities of Kenya (R2=0.657; Adjusted R2 = 0.514; F= 4.59; Sig.= 0.000) by means of timely graduation ceremonies
(F=2.96; Sig.= 0.024), reduction in customer complaints (F=2.134; Sig.= 0.083) and increase in customer compliments (F=9.143;
Sig.=0.000). Also is notable the contribution of responsiveness or willingness to help and respond to customer needs (R2= 0.563;
Adjusted R2 = 0.380; F= 3.079; Sig.= 0.000) by means of effective and efficient teaching methods (F=2.038; Sig.= 0.096), increase in
customer compliments (F=8.766; Sig.= 0. 000) and timely release of results (F=2.092; Sig.=0.088).
Regarding assurance, in other words the ability to inspire confidence (R2 = 0.534; Adjusted R2 = 0.339; F=2.741; Sig.= 0.000), its
contribution to level of service delivery at the university was largely explained by novel teaching methods (F= 4.33; Sig.= 0.003),
reduction in customer complaints (F=2.546; Sig.=0.045) and increase in customer compliments (F= 6.154; Sig.=0.000). Finally,
empathy or the extent to which individualized service is given (R2=0.486; Adjusted R2=0.272; F=2.268,Sig.=0.001) determined the
level of service delivery at the university via reduction in customer complaints (F=3.422; Sig.=0.012), increase in customer

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 690

www.ijird.com

May, 2014

Vol 3 Issue 5

compliments (F=2.36; Sig.= 0.059) and timely release of results (F=2.879; Sig.= 0.027). It shall however be noted that the increase in
customer compliments was the key explanatory variable predicting the level of service offered at the university. It was significantly
linked to tangibles, reliability, responsiveness, assurance and empathy. The remaining factors explained one or a few of the above
outcomes.

Table 13: Tests of between-subjects effects of teaching workload on the quality of service delivery
Source: Field Data (Researcher 2013)
a. R Squared = .623 (Adjusted R Squared = .465)
c.
b. R
R Squared
Squared =
= .563
.657 (Adjusted
(Adjusted R
R Squared
Squared =
= .380)
.514)
c. R Squared = .563 (Adjusted R Squared = .380)

*Significance at 10%; Confidence intervals of 90%

**Significance at 5%; Confidence intervals of 95%
***Significance at 1%; Confidence intervals of 99%

d. R Squared = .534 (Adjusted R Squared = .339)

e. R Squared = .486 (Adjusted R Squared = .272)
Table 14 presents results of the most significant parameters among different categories within a variable. The ttest mostly pointed out
to increase in demand of courses being offered as moderately important, important and very important, timely graduation ceremonies
as being important, increase in customer compliments as being both moderately important and very important, while timely release of
results was very important. Reduction in customer complaints was also explained as being moderately important while increase in the
number of graduates was not important. These categories were able to explain almost all the aspects of the level of service offered at
the university.
3.5. Discussion on Workload Analysis
In the literature review it is pointed out that universities have defined the role of academic staff to three domains of teaching, research,
and service and that university academic staff do complex work in an increasingly demanding environment. It is farther pointed out
that universities are the only organizations focused on dual core functions of knowledge creation and knowledge transmission, and
that academic staff members who spent the majority of their time teaching reported a preference for being rewarded for teaching
effectiveness. It is also noted that challenge, variety, and autonomy are key elements of the academic staff to engage in core activities
such as critical thinking, reflection, and collegial interactions in the context of disciplinary interests and expertise. Reforms have
sought to make universities more accountable to government, students as consumers, and the public generally. It is emphasized that
workload systems management has increasingly been a factor in recent contract negotiations and collective employment agreements
and therefore the need to investigate workloads issues such as increased stress on staff, development of creative solutions to
ameliorate problems, and sustaining the primary sources of work satisfaction that best promote quality.
It is the wish of each and every student to be admitted in a competitive course in a public university, more so the number of students
joining these universities has in the recent past increased due to the introduction of various degree programmes and the expansion of
teaching programmes through self-sponsorship, open and distance learning and evening classes in most of the major urban centres.
This number has also grown due to the opening up of satellite campuses and the recent uplifting of middle level colleges to fully
fledged universities.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 691

www.ijird.com

May, 2014

Vol 3 Issue 5

Table 14: The most significant teaching workload parameters explaining the level of service delivery
Source: Field Data (Researcher 2013)

Table 15: The most significant teaching workload parameters (Contd)

Source: Field Data (Researcher 2013)
Notes:
*Significance at 10%; Confidence intervals of 90%
**Significance at 5%; Confidence intervals of 95%
***Significance at 1%; Confidence intervals of 99%
The increase of the student has had to push the universities to innovate more effective means of satisfying them. This has led to
increased efficiency and effectiveness in various parameters in respect of student enrolment, timely graduation ceremonies so as to
join the labor market, teaching methodologies, timely examinations and timely release of examination results. Once all this is
addressed adequately then the universities would have to experience reduction in customer complaints, increase in customer
compliments and ultimately increase in the number of students enrolled.

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 692

www.ijird.com

May, 2014

Vol 3 Issue 5

4. Summary and Conclusion

That the level of service delivery in respect of the teaching workload and in consideration of the various variables used to measure
the level of service delivery was majorly determined by increased demand for courses offered at selected universities, timely
graduation ceremonies, increased customer compliments, increased number of graduates, reduction in customer complaints,
effective and efficient teaching methods, timely release of results and novel teaching methods. In view of these observations the
universities should try as much as possible to link the performance contracting practices to these variables, sustain them and equip
its staff with the necessary tools and skills besides giving feedback to its customers and also linking the courses offered to the
industry. It is worth noting that increase in student enrollment and timely examinations in regard to teaching workload was not
considered as affecting the level of service and this can be explained by the presence of many modules and programmes which have
been introduced to take care of the various needs of the customers, and that universities also continuously assess the students
through various types of examinations which include class presentations, discussion groups, take away assignments and continuous
assessment tests besides the final semester examinations and supplementary examinations.
5. References
1. Bellamy,S.,Morley,C.,and Watty,K. (2003).Why business academics remain inAustralianuniversities despite deteriorating
working conditions and reduced job satisfaction: An intellectual puzzle. Journal of Higher Education Policy
Management,25,13-28.
2. Borg,W.R and Gall,J.P(1996).Educational Research: An Introduction(6th edn),NewYork,Longman.
3. Chalmers,A. (1998).Workloads and stress in New Zealand universities. Report for the New Zealand Council for Educational
Research and The Association of University Staff of New Zealand. Wellington: Author.
4. Coaldrake,P.,and Stedman,L.(1999).Academic work in twenty first century: Changing roles and policies. Occassional paper
series, Department of Education, Training and Youth Affairs, Australia.
5. Harman, G.(2001).Academics and institutional differentiation in Australian higher education. Higher Education
Policy,14,325-342.
6. Harman, G.(2002).Academic leaders or corporate managers: Deans and heads in Australian higher education,1977 to
1997.Higher Education Management and Policy,14,53-70.
7. Harman, G.(2003).Australian academics and prospective academics: Adjustment to a more commercial environment. Higher
Education Management and Policy,15,105-122.
8. Houston, D., Luana H., Meyer and Shelley Paewai (2006).Academic Staff Workloads andJob Satisfaction: Expectations and
Values in academe.Journal of Higher EducationPolicy and Management Vol 28,No 1,March 2006,pp 17-30.
9. Jenkins,A.(2004).A Guide to Research Evidence on Teaching-Research Relations,The Higher Education Academy,383
10. Kerr,S.(1975).On the folly of rewarding A, while hoping for B.Academy of Management journal,18,769-782.
11. Kothari, C.R. (2009) Research methodology, methods and techniques, New age international publisher, New Delhi.
12. Krathwoh, D.R(1998).Methods of Educational and Social Science Research: An Intergrated Approach(2nd ed.).New York,
Longman.
13. Leslie,D.W. (2002).Resolving the Dispute:Teaching is Academes Core value. The Journal of Higher Education,73,49-73
14. McInnes,C.(1999).The work roles of academics in Australian universities. Canberra: Detya.
15. McInnes,C.(2000).Changing academic work roles:the everyday realities changing quality teaching.Quality in Higher
Education,6,143-152
16. Mugenda O. & Mugenda A. (2003).Research methods. Quantitative and Qualitative methods.Rev ed.Nairobi
17. Patterson, G.(1996).New Zealand universities in an era of challenge and change, occasional paper No.4.Palmerston
North,New Zealand: Department ofManagement Systems.
18. Romainville,M. (1996).Teaching and research at university:A difficult pairing.Higher Education Management,8,135-144.
19. (Sullivan,1997) Sullivan,K. (1997).What should count as Work in the Ivory Tower?:Determining academic workloads in
tertiary institutions, a university case sudy.NZARE monograph No.6.Wellington:New Zealand Association for Research in
Education.
20. Winter,R.,Taylor,T.and Sarros,J.(2000).Trouble at mill: Quality of academic work lifeIssues within a comprehensive
Australian university. Studies in Higher Education,25,279-294

INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH & DEVELOPMENT

Page 693