Data collection tools,
samples and sampling technique
The data for this study was collected through a
questionnaire survey, which from the university teachers of two public
universities in the Mymensingh district of Bangladesh. A well-structured
questionnaire with five point Likert-type scale, where 1= strongly disagree and
5= strongly agree was used for the questions. The questionnaire was distributed
to the respondents through hand to hand and through mail survey. Based on the
previous literature, the researcher described and accumulated 37 different
issues (factors), or elements that indicate the innovation performance of an
individual (see appendix). The questionnaire consists of those elements with
the above stated five point scaling.
Data analysis tools
The study analyzed 185 data collected through the questionnaire
survey. To analyze the collected data from the respondents, SPSS version 25
software and several sets of statistical analyses were used. Descriptive
statistics and the exploratory factors analysis were used to analyze the
collected data from the respondents.
Reliability of data
The reliability of the data was assessed by measuring
the Cronbach’s alpha. The alpha value of the 30 items questionnaire was .942
(Table 1). The detailed reliability statistics are shown in the table 1 in the
Appendix I. It shows the individual item reliability value and the scale value
if any of the items is deleted.
Analysis
of Findings
Demographic profile of
the respondents Kaiser-Meyer-olkin (KMO) and bartlett’s test
The KMO measures the sampling adequacy which should be
greater than 0.5 for a satisfactory factor analysis is to proceed. If any pair
of variables has a value less than this, consider dropping one of them from the
analysis. The off-diagonal elements should all be very small (close to zero) in
a good model. Looking at the table (Table 2) below, the KMO measure is 0.705.
The value 0.5 for KMO test is minimum and barely accepted, values between
0.7-0.8 are acceptable, and values above 0.9 are superb. Bartlett's test is
another indication of the strength of the relationship among variables. This
tests the null hypothesis that the correlation matrix is an identity matrix. An
identity matrix is matrix in which all of the diagonal elements are 1 and all
off diagonal elements are 0. From the same table, we can see that the
Bartlett's test of sphericity is significant That is, its associated
probability is less than 0.05. In fact, it is actually 0.000, i.e. the
significance level is small enough to reject the null hypothesis. This means
that correlation matrix is not an identity matrix.
Descriptive statistics
and communalities
Following table 4 shows the descriptive statistics and
communalities (variances) of all of the items uses in this study for factor
analysis. The table shows that the item knowledge of teaching and learning
methods has the highest mean value (4.46) where the item sufficient resources
has the lowest mean value (2.51). Standard deviation measures the variability
of data. Following (Table 3) shows that the item rewords and recognition has
the highest variability of responses (1.108) on the other hand the item
knowledge of teaching and learning methods had the lowest variability of
responses (.571). The communalities are commonly used in factor analysis to
show how much of the variance in the variables has been accounted for by the
extracted factors (Alam and Bhuiyan, 2014). For instance in the above table 4,
over 88% of the variance in innovative work behavior, over 87% of the variance
in collaboration is accounted for while 46.5% of the variance in Utilization of
knowledge and skills is accounted for.
Number of factors to be
extracted
Total variance explained and the scree plot are
commonly used to identify the number of factors extractable from the analysis.
The factor (component) which has the eigen value (the scree plot in the
Appendix II shows all the components with their eigen values in a single graph)
more than 1 is normally considered to be
extracted as factor (Alam and Bhuiyan, 2014; Talukder et al, 2014). In the
following (Table 4), it is seen that only 8 of the factors have the eigenvalues
over 1 and all other remaining are not significant (>1). So 8 factors can be
extracted in this study.
Factor (Component)
Matrix
The table (Table 5) below shows the loadings of the 30
variables on the 8 factors extracted. The higher the absolute value of the
loading, the more the item contributes to the factor. The gap on the table
represent loadings that are less than 0.4, this makes reading the table easier.
The researchers suppressed all loadings less than 0.4. The following (Table 6)
shows that nine items are loaded in the factor 1, six items are loaded in
factor 2 and four items are loaded in the factor 3, 4, 5, and 6. The table also
shows that some of the items are loaded in several factors such as the item
application of management/leadership styles is loaded in both the factors 1 and
6; item quality of educational system is loaded in both the factors 5 and 8.
These types of items are extracted based on their highest loaded value in a
particular factor. For example, the item application of management/leadership
styles has the higher loading value in the factor 1, so it is extracted as an
item of the factor 1.
Naming of the factors
Based on the factor matrix in the above table 6 the
researchers named the factors considering the loaded items in each of the
factors. The following table 6 shows the loaded items in each factor and their
names.