A structural equation modelling approach for massive blended synchronous teacher training.
In recent years, Information and Communication Technology (ICT) enabled education is seen as a way to improve the quality of education at all levels. Although ICT enabled education is considered as a necessary step to scale up the educational activities, its implementation is complex. One has to take into consideration many factors, such as availability of technology, time, training and support, coordination and management, individual attitude, belief and motivation, characteristics and ethos of the organization (Tearle, 2004). It also costs a lot of money to build the ICT infrastructure and to maintain them.
Most developing countries have limited infrastructure and fewer experts. Therefore, in future, many programmes are likely to be conducted in the distance mode--in a synchronous or asynchronous fashion. Since 2009, in order to train a large number of engineering college teachers in the country, the Indian Institute of Technology (IIT) Bombay, under the National Mission on Education through ICT (NMEICT, www.sakshat.ac.in), uses ICT to conduct teacher training workshops. These workshops are a first of its kind in the country, conducted in a blended synchronous distance mode. Thousands of teachers are engaged simultaneously through Internet based video conferencing software called AVIEW (www.aview.in) for a period of two-weeks. These workshops are scalable and more than 10,000 teachers can be trained simultaneously. Therefore, understanding the factors associated with such a system are very important for delivering future blended synchronous mode training workshops. The methodology used for conducting blended teacher training workshop is briefly explained in the following section. More details can be found in our earlier studies: Kannan and Narayanan (2011); Kannan and Narayanan (2012).
This paper helps find some critical success factors in a blended synchronous teacher training program. In our study, a research model for a synchronous distance mode teacher training program was developed based on the theoretical framework of Technology Acceptance Model (TAM) by Davis (1989) and the Self Determination Theory (Deci & Ryan, 1985). We redefined the constructs from TAM and added constructs of motivation--intrinsic and extrinsic motivation to explain the relationship between various factors. Although TAM has been tested extensively for technology acceptance by users, to the best of our knowledge, there are hardly any studies carried out so far in India on such a large scale blended synchronous training workshop.
This study analyses the data drawn from 523 participants of a blended synchronous distance mode workshop, conducted for engineering college teachers, in June 2012. Specifically, the subject matter covered the field of "Computational fluid dynamics" (CFD). In the workshop, there were about 1200 engineering college teachers from about 275 cities in India, representing 550 colleges. Therefore, the sample had a good representation from all parts of the country and from various engineering colleges.
In our study, the effectiveness of the distance mode blended synchronous teacher training workshop was measured with the help of a dependent variable - Participants' satisfaction (SAT). The independent variables were: Instructor effectiveness (IE), Technology effectiveness (TE), Intrinsic motivation (IM) and Extrinsic motivation (EM). The mediating variables were: Perceived usefulness (PU) and Behaviour intention to use (BIU). The current research focuses on two specific objectives: (1) To test the proposed research model for blended synchronous teacher training workshop using SEM technique, (2) To understand the factors that contribute to effective blended synchronous distance mode training.
Technological advances in the last two decades have led to increasing use of technology for teaching and learning. According to Bower et al. (2013), as the technology and bandwidth continue to improve, in future, we may have ubiquitous teaching-learning. Remote participants may be represented in any classroom via video interactions or any mixed mode with a similar experience of a real classroom. As we proceed further towards online education, understanding the factors that influence successful online education can provide useful insights to both teachers and learners.
Critical success factors in online learning
According to Dillon and Guawardena (1995), the three main variables that affect the effectiveness of e-learning environment are: technology, instructor characteristics and student characteristics. According to Willis (1994) and Collis (1991), the instructor plays a central role in the effectiveness and success of e-learning based courses. They believed that it was not the information technology but the instructional implementation of the ICT that determined the effectiveness of e-learning. Volery and Lord (2000) identified three critical success factors in e-learning: Technology (ease of access and navigation, interface design and level of interaction); Instructor (attitude towards students, instructor technical competence and classroom interaction) and previous use of technology from a students' perspective. According to Papp (2000), e-learning critical success factors included intellectual property, suitability of the course for e-learning environment, building the e-learning course, e-learning course content, e-learning course maintenance, e-learning platform, and measuring the success of an e-learning course. Papp (2000) suggested studying each one of these critical success factors in isolation and also as a composite to determine which factors influence and impact e-learning success.
TAM in online learning
Technology Acceptance Model abbreviated as TAM was proposed by Davis (1989); Davis et al. (1989). TAM was developed based on the theory of reasoned behaviour (TRB) by Fishbein and Ajzen (1975). TAM is widely used in empirical research to explain and predict user acceptance of information system. Literature shows that in education, TAM has been applied by researchers in various studies such as, e-learning and online course delivery. Chen and Tseng (2012) modified TAM for a web-based e-learning system and found that motivation to use and self-efficacy were positively associated with Behaviour intention to use (BIU). Liaw (2008) studied the effectiveness of a Blackboard e-learning system in a university and found that Perceived usefulness (PU) and Perceived satisfaction contributed to BIU e-learning system. Liu et al. (2010) evaluated the web based interactive language learning community among senior high school students in Taiwan and found that Perceived usefulness, Perceived ease of use, and Perceived interaction positively influenced intention to use an online learning community. Teo (2011) tested a model to explain school teachers' Behaviour intention to use technology. Most of the studies predicted the users' behaviour intention to use information technology.
Motivation in online environment
As motivation plays an important role in learning, especially in an online environment, Self Determination Theory (SDT) serves as an important framework for addressing motivation in an online environment. According to SDT (Deci & Ryan, 1985), self-determined form of motivation (intrinsic and integrated regulation) may lead to positive outcomes, while the non self-determined form of motivation (Amotivation, external regulation, introjection regulation and identified regulation) may lead to negative outcomes (Deci & Ryan, 1991). SDT has been widely applied across various fields, such as, physical education (Moreno et al., 2010), politics (Losier, et al., 2001), health care (Williams et al., 2006) and general education (Chen & Jang, 2010).
Synchronous and asynchronous interactions
According to Offir et al. (2008), students prefer synchronous to asynchronous mode of interaction, as the transactional distance in asynchronous mode causes poor quality interaction which leads to decreased learning. Educational researchers propose several benefits of using blended synchronous learning approaches. It enables equity of access for learners who are geographically isolated and better completion rate than asynchronous (Norburg, 2012). It also promotes discussion and community learning (Roseth et al., 2013; Lidstone & Shield, 2010).
The concept of satisfaction is derived from Marketing, where customer satisfaction refers to the state of mind when the expectations of the customer are met from a product or a service (Anisor et al., 2010). Satisfaction is a construct used by many researchers in the past to measure the effectiveness of online learning (Hermans et. al., 2009; Arbaugh, 2000; Kanthawongs, 2011; Marks et al., 2005; Eom et al., 2006). Arbaugh (2000) found that both perceived usefulness and ease of use were positively associated with students' satisfaction in online MBA courses. Kanthawongs (2011) found that learners' satisfaction with the instructor, perceived ease of use, commitment, and perceived flexibility were positively related to the learners' satisfaction in a web-based ERP course in Thailand. Drawing from the literature in our study, we have taken participants' satisfaction as a dependent variable.
Teacher training methodology
In order to train a large number of teachers spread across the length and breadth of India, IIT Bombay uses a hub and spoke model. IIT Bombay acts as the hub and a number of remote centres (RCs) located at various engineering colleges in the country act as spokes. RCs are equipped with the necessary infrastructure to accommodate 30 to 60 teachers each. The infrastructure consists of a minimum of 1MBps bandwidth to receive live transmission and two way interaction, open source software AVIEW, different types of cameras, an LCD projector, a classroom, and a laboratory. RCs are located at short distances from several engineering colleges. RCs form a convenient and effective ecosystem for interactions between the college teachers and IIT instructors and for hands on training. There are about 250 such RCs. The model is diagrammatically represented in Figure 1. It can accommodate 10,000 to 15,000 participants at a time.
The training workshop is conducted in two parts: coordinators' workshop and the main workshop. The coordinators' workshop is conducted about two months prior to the main workshop, where the workshop coordinators from every remote centre come to the hub (IIT Bombay) for a one week face-to-face workshop. During this workshop, the coordinators are given an orientation on various teaching-learning technologies, the laboratory facility required, and software and hardware required to conduct the main workshop. A common syllabus of the subject to be taught in the main workshop is also finalized. These coordinators help conduct the main workshop in each of the RCs.
The main workshop is conducted for a duration of two-weeks during the winter and summer vacation. Live transmission of lectures and two-way interaction between the hub and RCs happen through an Internet based video conferencing software AVIEW. The screenshot of lectures delivered through AVIEW is given in Figure 2. The AVIEW screen has three windows--A, B and C. Window-A shows the video of the course instructor. In window-B, the participants' video is displayed. Window-B has two other modes: the chat window and users' window, which can be selected at any time. Whenever participants have a query, they can electronically raise their hand through the users' window. This appears as a question mark against the RC in Window B. The instructor at the IIT Bombay hub can choose this RC and carry out live interactions, which are seen by all RCs. Window-C is used for displaying the presentation slides or as a whiteboard. Laboratory sessions and tutorials are conducted locally at each RC with the help of coordinators. All lectures and live interactions are recorded and the final content is released as open content to the teaching-learning community through T10KT website (www.it.iitb.ac.in/nmeict/home.html).
The blended mode workshop has:
* Synchronous interactions between the hub and RCs through AVIEW and chat
* Asynchronous interactions between instructors, participants and coordinators through discussion forum (Moodle) and email
* Face-to-face and peer group interactions between participants and coordinators in all RCs.
Thus, the model is unique, as the combination of all three interactions makes it very effective and engaging.
Research model and hypotheses
This section describes the research model that was tested using SEM technique. The research model is given in Figure 3. The main components in the research model are given below:
* Online environment factor consists of instructor and technology. Instructor is the subject experts at the hub and technology consists of both synchronous and asynchronous ICTs used for lecture delivery and interactions.
* Participants' motivation consists of both intrinsic motivation and extrinsic motivation to attend the workshop.
* Participants' perception consists of perceived usefulness of the workshop and behaviour intention to use/ adopt the teaching-learning material in the classroom after the workshop.
* Participants' satisfaction is a measure of effectiveness of blended synchronous training program.
Operation definitions of constructs
Operation definitions of constructs used in the study are given below:
Instructor effectiveness (IE)--Instructor effectiveness was measured in terms of organization of course material, time management, content knowledge, pedagogy and query handing from participants.
Technology effectiveness (TE)--Technology effectiveness was measured in terms of quality of audio and video transmission, clarity of live interaction, ease of use of synchronous and asynchronous mode of interaction during the teacher training workshop conducted in the distance mode.
Participants' motivation (IM and EM)--Motivation of participants to attend a distance mode workshop was measured by intrinsic motivation (to learn from subject experts, to learn new teaching methodology and to become accomplished teacher) and extrinsic motivation (certification, promotion, sponsored by government).
Perceived usefulness (PU)--Perceived usefulness refers to whether the participating teachers believed that attending the workshop was useful to enhance his/her subject knowledge and would benefit his/ her teaching performance in future.
Behaviour intention to use (BIU)--Behaviour intention to use refers to the participating teachers' intention to use the knowledge gained from the workshop and adopting the teaching-learning material from the workshop in their teaching.
Satisfaction (SAT)--Satisfaction refers to the confirmation of participants' expectations from the workshop.
From the research model, we formulated the following hypotheses and tested the same using SEM technique.
* H1: Instructor effectiveness will have a positive effect on Perceived usefulness of workshop.
* H2: Technology effectiveness will have a positive effect on Perceived usefulness of workshop.
* H3 and H4: Participants' motivation will have a positive effect on Perceived usefulness.
* H5: Perceived usefulness will have a positive effect on Behaviour intention to use.
* H6: Behaviour intention to use will have a positive effect on Satisfaction.
* H7: Instructor effectiveness will have a positive effect on Satisfaction.
* H8: Technology effectiveness will have a positive effect on Satisfaction.
* H9 and H10: Participants' motivation will have a positive effect on Satisfaction.
In the field of education, we often have many independent variables and dependent variables and need to examine the interplay between the variables. Therefore, the appropriate technique used in explaining such complex models is multivariate analysis using the structural equation modelling (SEM). All hypotheses in our study were tested using SEM.
Online survey methodology was adopted to collect data from the participating teachers. We adopted the online survey method, as the number of participants was large and geographically scattered. In order to measure the variables in the study, an instrument was developed. The items that were used to measure the constructs are given in Appendix A. Some items were adopted from prior research work (Young & Lewis, 2008; Davis, 1989) and were changed to suit the context of the study. The questionnaire was first tested in a pilot study and some items were modified and re-worded. All items were measured using a five-point Likert scale from "Strongly disagree = 1" to "Strongly agree = 5." Literature shows that Likert scale is a popular method used by researchers to measure people's attitudes, perceptions, preferences, opinions etc. Therefore, we chose Likert scale to measure the items used in the study.
In order to achieve adequate power in SEM, the recommended sample size is at least one hundred observations (Hair et al., 2006). We had a total sample size of 923. We randomly split the data into two parts. We performed reliability and exploratory factor analysis (EFA) with the sample of 400 and with the remaining data (523) we performed confirmatory factor analysis (CFA) and tested the structural model.
Data collected through the online questionnaire were analyzed using Statistical Package for Social Sciences (SPSS) version-18 and AMOS version-18.
Demography of the sample
The demography of the sample is tabulated in Table 1. The sample consisted of 85% male and 15% female. Majority of the respondents, about 84% were below 40 years and only about 16% were above 40 years. About 75% were postgraduates, 15% were undergraduates and 10% of the respondents were Doctorates. About 67% had a work experience of less than 10 years and about 33% had worked for more than 10 years. About 38% of the respondents belonged to colleges located in rural areas and about 62% were from colleges situated in urban areas.
The descriptive statistics of the constructs are shown in Table 2. All means were above the midpoint of 2.5. The standard deviations indicate a narrow spread around the mean. The normality assumption using skewness and kurtosis indices was checked for each of the variables using the criteria given by Kline (2005). According to Kline (2005), lack of normality occurs when absolute value of skewness index is > 3 and absolute value of kurtosis index is > 10. In our sample, the skewness and kurtosis of all the variables were within the expectable range. The skewness ranged between -2.15 and 0.64 and kurtosis ranged between -0.99 and 7.81. Therefore, the normality assumption was met for the purpose of structural equation modelling. Except for the extrinsic motivation (EM) variable, all other variables had significant correlations among the variables. The correlation matrix is given in Table 2.
Exploratory factor analysis
Exploratory Factor Analysis (EFA) on a sample of 400 observations was performed. Principal axis factoring (PAF) method with direct oblimin rotation was used. Four factors were extracted. The total variance explained by the factors was 56.7% and Kaiser-Meyer-Olkin (KMO) value was 0.870. The factor loadings of the factors were in the range 0.384 to 0.848. According to Hair et al. (2006) if the sample size is above 350, factor loadings of 0.3 and above are considered significant for interpreting the results. Therefore, we used the value of 0.3 as the threshold for factor loading. Items with low factor loading were removed from the data set.
Reliability and validity
In our study, we established the reliability of the instrument with the help of internal consistency. Cronbach's alpha (a) was calculated for all the constructs in the study. Cronbach's alpha values ranged from 0.62 to 0.82. Except for the Intrinsic motivation construct ([alpha] = 0.62), all other constructs had satisfactory internal consistency ([alpha] [greater than or equal to] 0.70). One of the reasons for low value of (a) for Intrinsic motivation construct could be high mean and low standard deviation.
The face validity or content validity of the instrument was done by three experts and through pilot testing. The construct validity was established by convergent and discriminant validity. According to Hair et al. (2006) convergent validity is established with composite reliability (CR) and average variance extracted (AVE). Convergent validity is judged to be adequate when composite reliability exceeds 0.70 and average variance extracted exceeds 0.50 (Hair et al., 2006). As shown in Table 3, CR and AVE for all the constructs were adequate. Thus, convergent validity for all the constructs was established.
Discriminant validity is established when the variance shared between a construct and any other construct in the model is less than the variance that a construct shares with its own indicators (Fornell & Larcker, 1981). To get satisfactory discriminant validity, the square root of AVE for each construct should be greater than the squared correlation between the constructs (Fornell & Larcker, 1981). The diagonals in Table 3 represent the square root of average variance extracted and other entries represent squared correlations. Table 3 shows acceptable discriminant validity between each pair of constructs. Thus, discriminant validity for all the constructs was established.
Confirmatory factor analysis (CFA) for the measurement model was performed using Maximum Likelihood Estimation (MLE) method using AMOS version 18. A total of 26 items of the measurement model were subjected to CFA. The initial estimation model was modified with the help of modification index with a few error terms in the model to covary. The standardized factor loadings ranged from 0.46 to 0.78 and were significant at p < 0.001 (tvalues ranged from 4.9 to 12.7). The Goodness of fit (GOF) results of the measurement model was found within the acceptable range with [chi square]/df = 2.5 ([chi square] = 725, df = 281, p = .000), CFI = 0.87, GFI = 0.91, AGFI = 0.88, TLI = 0.85, all values were [greater than or equal to] 0.85 (recommended value for GOF is [greater than or equal to] 0.9) and RMSEA = 0.05, recommended value is [less than or equal to] 0.10. As most of the observed GOF indices satisfied the recommended values of the indices, the results of CFA suggests that the measurement model had a good fit.
The structural model was tested using AMOS. The results showed that the model was recursive with 351 distinct sample moments and 70 distinct parameters to be estimated. The model was identified and it had 277 degrees of freedom. The fit indices indicated that the model had a good fit. The [chi square] statistics was 494 and it was significant (p = .000). The [chi square] value should be non-significant, but in our model due to the large sample size it was significant. Therefore, we checked the value of [chi square]/df. The [chi square]/df was 1.7 (recommended value is [less than or equal to] 3), implying that the observed covariance matrix and implied covariance matrix were statistically identical.
The GFI and AGFI were 0.93 and 0.91 respectively (the recommended value is [greater than or equal to] 0.90) (Hair et al., 2006). CFI and TLI were 0.93 and 0.92 respectively. The parsimonious fit index PNFI and PCFI were 0.74 and 0.79 respectively (recommended value is [greater than or equal to] 0.6). The RMSEA value was 0.03 (recommended value is less than 0.05 for good fit and [less than or equal to] 0.10 for moderate fit). The results of the model indicated that overall the model had a good fit as all the indices were within the recommended values.
Ten hypotheses were tested using path analysis. The regression weights for the parameters along with the critical ratios are given in Table 4. The research model with the standardized path coefficients and the explanatory power ([R.sup.2]) for each dependent variable is displayed in Figure 4. Except for four paths, all other paths were statistically significant. The results of the model supported H1, H5, and H8 with path coefficients of 0.57, 0.89 and 0.36 (p [less than or equal to] .001) respectively and H2, H6 and H10 with path coefficients of 0.19, 0.18 and 0.14 (p [less than or equal to] 0.05) respectively. However, the results indicated rejecting H3, H4, H7 and H9.
A coefficient linking one construct with another directly in the path diagram (Figure 4) represents the direct effect. An indirect effect is represented when the determinants affect the endogenous variable through one or two mediating variables. Table 5 shows standardized total effects, direct and indirect effects associated with each of the variables in the model. The total effect on a variable is the sum total of respective direct and indirect effects. In our model, the standardized total effects of predictor variables on the dependent variables ranged from 0.00 to 0.89.
Three endogenous variables were tested in the model. In terms of the explanatory power, the model adopted in the study explained 58% of variance in Perceived usefulness, 80% of variance in Behaviour intention to use and 34% of variance in Satisfaction.
Among the three endogenous variables, the highest amount of variance (80%) was explained by the determinants of Behaviour intention to use (BIU). The most dominant determinant was Perceived usefulness (PU) with a total effect of 0.891, followed by Instructor effectiveness (IE) with a total effect of 0.509 and Technology effectiveness (TE) with a total effect of 0.171. Both Intrinsic motivation (IM) and Extrinsic motivation (EM) had insignificant effect on BIU.
The model explained 58% of variance in Perceived Usefulness (PU). Among all the variables, Instructor effectiveness (IE) had the largest effect, with total effect of 0.572; Technology Effectiveness (TE) had a total effect of 0.192. However, Intrinsic Motivation (IM) with a total effect of 0.113 and Extrinsic Motivation (EM) with a total effect of 0.001 had statistically insignificant effect on Perceived Usefulness (PU). This implies that the participants would find the course useful if the instructor is very effective in making the subject matter interesting, easy to understand and interacts regularly with the distant participants.
The model explained about 34% of variance in Satisfaction (SAT). The most dominant determinant was Technology Effectiveness (TE) with a total effect of 0.392 followed by Behaviour intention to use (BIU) with a total effect of 0.185. Among other variables, Perceived Usefulness (PU) had a total effect of 0.165, Instructor Effectiveness (IE) had a total effect of 0.144, Extrinsic Motivation (EM) had a total effect of 0.143. However, Intrinsic Motivation (IM), with a total effect of 0.134, was insignificant as both direct and indirect paths were non-significant.
Discussion, limitations and future work
The results imply that Technology Effectiveness (TE) defined in terms of "Quality of audio and video transmission, clarity of live interaction and ease of use of synchronous and asynchronous mode of interaction" during the teacher training workshop emerged as the most important factor for Satisfaction. If one improves the technology component by one point, then it will result in 0.39 standard deviation increase in the satisfaction level of the participants. Instructor Effectiveness (IE) had an indirect effect (0.094) through the mediating variables PU and BIU on participants' satisfaction, which was more than the direct effect (0.05). Therefore, we can infer that Instructor Effectiveness (IE) was fully mediated through PU and BIU. Extrinsic Motivation (EM) had a total effect of 0.143 on Satisfaction, which was entirely a direct effect. This implies that some kind of incentive, such as certification, sponsorship is required for the participating teachers for greater satisfaction. However, Intrinsic Motivation (IM) had insignificant effect on Satisfaction, as there was not much variation in the participants' response that contributed to the variance.
The results of our study were consistent with the TAM model proposed by Davis & Venkatesh (1996), where PU is positively related BIU. In our study, PU and BIU had a statistically significant correlation (P = 0.89, p < .001). The findings of our study were consistent with studies done by Ramayah and Lee (2012) in Malaysia, where system quality positively affected user satisfaction in an e-learning environment. The findings were also consistent with the study done by Eom et al. (2006), where the course structure, self-motivation, instructor feedback, interaction, instructor facilitation had a positive effect on students' satisfaction in an online course.
The importance of technology component in a blended synchronous mode is stressed by other researchers in the literature. According to Bower et al. (2013), the main issues that instructors confronted when teaching blended synchronous lessons were communication issues and issues related to cognitive overload. According to White et al. (2010), the success of blended synchronous learning activities depends on the quality of audio, use of teaching associates and support staff. Our study confirms that in a blended synchronous mode teaching-learning, the quality of audio-video, clarity of live interaction and ease of use of technology were important factors.
Although care was taken to ensure that the methodology used in the study was sound, there were some limitations. First, as data were collected online, just after the workshop through a self-rating questionnaire, some participants could have given higher ratings to some of the variables, which could have affected the results. Second, Intrinsic Motivation (IM) construct had a low Cronbach's alpha (0.62); therefore, some more items need to be added or modified. Third, the six variables in the model explained only 34% of variance in Satisfaction leaving 66% unexplained. The findings of the study suggest that there could be other variables in the blended synchronous training that could influence the Satisfaction. Nevertheless, such an attempt to systematically evaluate the effectiveness of a blended synchronous teacher training program has not been done so far in the Indian context. To that extent, this paper makes an important contribution.
In order to further improve the design and implementation of blended synchronous training, we looked at the qualitative feedback from the participants. Some of the feedback was:
* "This experience of distance learning is good and innovative."
* "It is my first experience of distance education and it is good to see that understanding is as good as the lectures conducted in a lecture hall in the conventional manner."
* "The RC facility was excellent and the centre coordinator was very helpful. He was knowledgeable and we could clarify most of the doubts."
* "Laboratory session was very helpful for understanding the topic which was taught during the theory session."
* "During the online quiz there was bandwidth problem in our RC."
* "Sometimes there was problem with the audio and video."
* "Facilities at my RC were not good."
From the participants' feedback, we found that in some RCs the workshop coordinators were very helpful and knowledgeable. They helped the participants in clearing doubts during laboratory sessions and encouraged discussions among the participants. While in some other RCs, the coordinators did not take much initiative. Facilitating conditions at the RCs also varied from one centre to another. Therefore, we infer that workshop coordinators, facilitating conditions at RC and peer group interactions could also influence satisfaction in the blended training model. Therefore, in future, one may consider introducing the following additional factors in the SEM: (1) Coordinator effectiveness (Educational qualification, subject knowledge, motivation, query handling, workshop management and coordination) (2) Facilitating conditions at RC (Audio-video setup, laboratory facility, Internet bandwidth and support staff) (3) Peer group interactions (formal and informal discussions with peers in RCs).
In this paper, a research model for a blended synchronous mode teacher training program was developed based on the theoretical framework of TAM (Davis, 1989) and Self Determination Theory (Deci & Ryan, 1985). The research model was tested with empirical data from 523 teachers, who had participated in a blended synchronous workshop in June 2012. Ten hypotheses were tested with the help of path analysis. Six hypotheses were supported and four hypotheses were not supported. The model adopted in the study explained 58% of variance in Perceived usefulness (PU), 80% of variance in Behaviour intention to use (BIU), and 34% of variance in Satisfaction (SAT). The results imply that in a blended synchronous mode teacher training the technology factor is the most significant factor that contributes to the participants' satisfaction.
This study has many implications for educational institutions, administrators and policy makers who are involved in ICT enabled education. First, the study shows that Technology effectiveness is an important factor, the technology that is used for blended synchronous mode has to be good and easy to use for the participants. Second, as PU and BIU had a statistically significant correlation, it implies that participants of the workshop are likely to adopt the teaching-learning material in their teaching if they find the training workshop to be useful and relevant. Third, as Extrinsic motivation (EM) had a direct effect on Satisfaction, it implies that some kind of incentive, such as certification, sponsorship etc., would increase the satisfaction level of the participants. Fourth, apart from the six factors considered in the research model for blended synchronous mode training, coordinator effectiveness, facilitating conditions at RC, and peer group interactions could also influence participants' satisfaction.
We would like to thank Prof. D. B. Phatak, Principal Investigator of the project, and Prof. A. Sharma and Prof. B. Puranik, Instructors of the workshop. Prof. P. Purang gave valuable inputs while writing the paper. Financial support for conducting the workshop was made available by a grant from the NMEICT, Ministry of Human Resource Development, Government of India. AVIEW support was provided by Amrita Vishwa Vidyapeetham.
Anisor, N., Adela-Eliza, D., & Luciana, C. (2010). The method used for measuring the customers' satisfaction. In H. Fujita & J. Sasaki (Eds.), Proceeding of 9th WSEAS International Conference on Education and Educational Technology (EDU'10) (pp. 197-200). Iwate, Japan: WSEAS Press.
Arbaugh, J. B. (2000). Virtual classroom characteristics and student satisfaction with internet-based MBA courses. Journal of Management Education, 24(1), 32-54.
Bower, M., Kenney, J., Dalgarno, B., Lee, M. J. W., & Kennedy, G E. (2013). Blended synchronous learning: Patterns and principles for simultaneously engaging co-locator and distributed learners. In H. Carter, M. Gosper & J. Hedberg (Eds.), Electric Dreams, Proceedings of 30th ASCILITE conference (pp. 92-102). Sydney, Australia: Macquarie University. Retrieved from http://ascilite.org.au/conferences/sydney13/program/papers/Bower.pdf
Chen, H. R., & Tseng, H. F. (2012). Factors that influence acceptance of web-based e-learning systems for the in-service education of junior high school teachers in Taiwan. Evaluation and Program Planning, 35(3), 398-406.
Chen, K. C., & Jang, S. J. (2010). Motivation in online learning: Testing a model of Self-Determination Theory. Computers in Human Behavior, 26, 741-752.
Collis, B. (1991). Anticipating the impact of multimedia in education: Lessons from the literature. International Journal of Computer in Adult Education and Training, 2(2), 136-149.
Davis, F. D. (1989). Perceived usefulness, perceived ease of use and user acceptance of information technology. MIS Quarterly, 13(3), 319-339.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User acceptance of computer technology: A Comparison of two theoretical models. Management Science, 35(8), 982-1003.
Davis, F. D., & Venkatesh, V. (1996). A critical assessment of potential measurement biases in the technology acceptance model: Three experiments. International Journal of Human Computer study, 45, 19-45.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behaviour. New York, NY: Plenum.
Deci, E. L., & Ryan, R. M. (1991). A motivational approach to self: Integration in personality. In R. Dienstbier (Ed.), Nebraska symposium on motivation: Perspectives on motivation (Vol. 38, pp. 237-288). Lincoln, NE: University of Nebraska press.
Dillon, C. L., & Guawardena, C. N. (1995). A Framework for evaluation of telecommunications based distance education. In D. Sewart (Ed.), Proceedings of 17th Congress of the International Council for Distance Education (Vol. 2, pp. 348-351). Milton Keynes, UK: The Open University.
Eom, S. B., Wen, H. J., & Ashill, N. (2006). The Determinants of students' perceived learning outcomes and satisfaction in university online education: An Empirical investigation. Decision Sciences Journal of Innovative Education, 4(2), 215-235.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intension and behavior: An Introduction to theory and research. Reading, MA: Addison Wesley.
Fornell, C., & Larcker, D. F. (1981). Evaluation structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39- 50.
Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate Data Analysis (6th ed.). Upper Saddle River, NJ: Pearson Prentice Hall.
Hermans, C. M., Haytko, D. L., & Mott-Stenerson, B. (2009). Student satisfaction in web-enhanced learning environments. Journal of Instructional Pedagogies, 1(1), 1-19.
Kannan, K., & Narayanan, K. (2011). Model and learner feedback for a large scale technology enabled teacher training program. In Proceedings of IEEE International conference on Technology for Education, T4E 2011, Chennai, India (pp. 250-253). doi:10.1109/T4E2011.51
Kannan, K., & Narayanan, K. (2012). ICT-enabled scalable workshops for engineering college teachers in India. In R. Clothey, S. Austin-Li & J. C. Weidman (Eds.), Post-Secondary Education and Technology--A Global Perspective on Opportunities and Obstacles to Development (pp. 127-146). New York, NY: Palgrave Macmillan.
Kanthawongs, P. (2011). A Structural model of student satisfaction in web-based ERP--simulated learning environments. International Journal of Education and Information Technologies, 5(2), 166-173.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York, NY: Guilford.
Liaw, S. S. (2008). Investigating Students' perceived satisfaction, behaviour intention, and effectiveness of e-learning: A Case study of the Blackboard system. Computers and Education, 51(2), 864-873.
Lidstone, J., & Shield, P. (2010). Virtual reality or virtual real: Blended teaching and learning in a master's level research methods class. In Y. Inoue (Ed.), Cases on online and blended learning technologies in higher education: Concepts and practices (pp. 91-111). Hershey, PA: Information Science Reference, IGI Global.
Liu, I. F., Chen, M. C., Sun, Y S., Wible, D., & Kuo, C. H. (2010). Extending the TAM model to explore the factors that affect Intention to use an online learning community. Computers and Education, 54(2), 600-610.
Losier, G F., Perreault, S., Koestner, R., & Vallerand, R. J. (2001). Examining individual differences in the internalization of political values: Validation of the self-determination scale of political motivation. Journal of Research in Personality, 55(1), 41-61.
Marks, R. B., Stanley, D. S., & Arbaugh, J. B. (2005). A Structural equation model of predictors of effective online learning. Journal of Management Education, 29(4), 531-563.
Moreno, J. A., Gonzalez-Cutre, D., Sicilia, A., & Spray, C. M. (2010). Motivation in the exercise setting: Integrating constructs from approach--avoidance achievement goal framework and self-determination theory. Psychology of Sport and Exercise, 11(6), 542-550.
Norburg, A. (2012). Blended learning and new education logistics in northern Sweden. In D. Oblinger (Ed.), Game Changers: Education and information technologies (pp. 327-330). Boulder, CO: Educause.
Offir, B., Lev, Y, & Bezalel, R. (2008). Surface and deep learning processes in distance education: Synchronous versus asynchronous systems. Computers and Education, 51, 1172-1183.
Papp, R. (2000). Critical success factors for distance learning. In Proceedings of Americas Conference on Information Systems (aMcIS) (pp. 1858-1861). AIS Electiric Library (AISeL). Retrieved from http://aisel.aisnet.org/cgi/viewcontent.cgi?article=1570&context=amcis2000
Ramayah, T., & Lee, J. W. C. (2012). System characteristics, satisfaction and e-learning usage: A Structural Equation Model (SEM). The Turkish online Journal of Educational Technology, 11 (2), 196-206.
Roseth, C., Akcaoglu, M., & Zellner, A. (2013). Blending synchronous face-to-face and computer supported cooperative learning in a hybrid doctoral seminar. TechTrends, 57(3), 54-59.
Tearle, P. (2004). A Theoretical and instrumental framework for implementing change in ICT in education. Cambridge Journal of Education, 54(3), 331-351.
Teo, T. (2011). Factors influencing teachers' intention to use technology: Model development and test. Computers and Education, 57(4), 2432-2440.
Volery, T., & Lord, D. (2000). Critical success factors in online education. The International Journal of Education Management, 14(5), 216-223.
White, C. P., Ramirez, R., Smith, J. G, & Plonowski, L. (2010). Simultaneous delivery of face-to-face course to on-campus and remote off-campus students. TechTrends, 54(4), 34-40.
Williams, G C., McGregor, H. A., Sharp, D., Levesque, C. S., Kouides, R. W., Ryan, R. M., & Deci, E. L. (2006). Testing a self-determination theory intervention for motivating tobacco cessation: Supporting autonomy and competence in a clinical trail. Health Psychology, 25, 91-101.
Willis, B. (1994). Enhancing faculty effectiveness in distance education. In B. Willis (Ed.) Distance education: Strategies and tools (pp. 277-290). Englewood cliffs, NJ: Education Technology publication Inc.
Young, A., & Lewis, C. W. (2008). Teacher education programmes delivered at a distance: An Examination of distance student perceptions. Teaching and Teacher Education, 24, 601-609.
Appendix A Construct Type Item Statement Instructor Independent IE1 Instructor used the time effectiveness variable effectively to meet workshop (IE) objectives IE2 The course material was well organized IE3 Instructor gave effective examples, illustrations and guidelines IE4 Instructor was able to explain difficult concepts IE5 Instructor used innovative teaching methods IE6 Instructor was responsive to participants' queries Technology Independent TE1 The quality of audio effectiveness variable transmission was good (TE) TE2 The quality of video transmission was good TE3 The live interaction session with the instructor was clear and understandable TE4 It was easy to interact with the instructor through 'AVIEW' technology (synchronous) TE5 I found it easy to interact with fellow participants through "Moodle" (asynchronous) Intrinsic Independent IM1 To learn new teaching motivation variable methodology (IM) IM2 To become an accomplished teacher IM3 To learn from subject expert Extrinsic Independent EM1 For certification motivation variable (EM) EM2 It will help in promotion EM3 Management of your institute asked you to attend EM4 Because, the course is fully sponsored Perceived Dependent PU1 I found the lectures to be usefulness (Mediating) useful for my understanding (PU) of the subject PU2 I found the lab/tutorials to be useful for my understanding of the project PU3 The workshop would enable me to teach better Behaviour Dependent BIU1 I will adopt the material from intention to (Mediating) the lectures in my teaching use (BIU) BIU2 I will adopt the material from the lab/tutorial in my teaching Satisfaction Dependent SAT1 I enjoyed the distance mode (SAT) variable workshop SAT2 I would recommend such distance mode workshops to others SAT3 The workshop provided a good learning experience Construct Type Item Source Instructor Independent IE1 Young & effectiveness variable Lewis (2008) (IE) IE2 -do- IE3 -do- IE4 self IE5 self IE6 Young & Lewis (2008) Technology Independent TE1 self effectiveness variable (TE) TE2 self TE3 Davis (1989) TE4 self TE5 self Intrinsic Independent IM1 self motivation variable (IM) IM2 self IM3 self Extrinsic Independent EM1 self motivation variable (EM) EM2 self EM3 self EM4 self Perceived Dependent PU1 self usefulness (Mediating) (PU) PU2 self PU3 self Behaviour Dependent BIU1 self intention to (Mediating) use (BIU) BIU2 self Satisfaction Dependent SAT1 Young and (SAT) variable Lewis (2008) SAT2 -do- SAT3 self
Kalpana Kannan * and Krishnan Narayanan
Department of Humanities and Social Sciences, IIT Bombay, Indiafirstname.lastname@example.orgemail@example.com
* Corresponding author
Table 1. Demography of the sample (N = 523) Variable N Percentage Gender Male 447 85% Female 76 15% Age group 23 to 30 219 41.8% 31 to 40 222 42.4% 41 to50 67 12.8% above 50 15 3% Educational qualification Undergraduate 80 15% Postgraduate 390 75% Doctorate 53 10% Work experience less than 2 years 108 20.7% 2 to 5 years 123 23.5% 5 to 10 years 119 22.8% more than 10 years 173 33% Location Rural 201 38.4% Urban 322 61.6% Table 2. Descriptive statistics and correlation matrix Construct Item Mean SD [alpha] IE TE IE 6 4.15 0.53 0.82 1.00 TE 5 4.14 0.56 0.81 .454 *** 1.00 IM 3 4.41 0.56 0.62 .295 *** .248 *** EM 4 2.73 0.80 0.69 .038 -.085 PU 3 4.18 0.57 0.72 .568 *** .417 *** BIU 2 4.12 0.61 0.76 .445 *** .351 *** SAT 3 3.94 0.60 0.69 .337 *** .362 *** Construct IM EM PU BIU IE TE IM 1.00 EM -.061 1.00 PU .287 *** -.016 1.00 BIU .242 *** -.033 .614 *** 1.00 SAT .197 *** .105 ** .35 *** .288 *** Note. *** Correlation is significant at the .01 level (2-tailed), ** Correlation is significant at the .05 level (2-tailed) SD--Standard deviation, [alpha]--Cronbach's alpha, IE--Instructor effectiveness, TE--Technology effectiveness, PU--Perceived usefulness, IM--Intrinsic motivation, EM--Extrinsic motivation, BIU--Behaviour intention to use, SAT--Satisfaction. Table 3. Convergent and discriminant validity Construct CR AVE IE Instructor effectiveness (IE) 0.85 0.501 0.707 Technology effectiveness (TE) 0.83 0.500 0.206 Intrinsic motivation (IM) 0.74 0.500 0.087 Extrinsic motivation (EM) 0.79 0.500 0.001 Perceived usefulness (PU) 0.74 0.497 0.322 Behaviour intention to use (BIU) 0.83 0.495 0.198 Satisfaction (SAT) 0.74 0.498 0.113 Construct TE IM EM Instructor effectiveness (IE) Technology effectiveness (TE) 0.707 Intrinsic motivation (IM) 0.061 0.707 Extrinsic motivation (EM) 0.007 0.003 0.707 Perceived usefulness (PU) 0.173 0.087 0.000 Behaviour intention to use (BIU) 0.123 0.058 0.001 Satisfaction (SAT) 0.131 0.038 0.011 Construct PU BIU SAT Instructor effectiveness (IE) Technology effectiveness (TE) Intrinsic motivation (IM) Extrinsic motivation (EM) Perceived usefulness (PU) 0.704 Behaviour intention to use (BIU) 0.376 0.706 Satisfaction (SAT) 0.122 0.082 0.707 Note. Composite reliability (CR) = (Square of the summation of factor loadings)/(square of the summation of factor loadings) + (summation of error variances). Average variance extracted (AVE) = (Summation of the square of factor loadings)/(summation of the square of factor loadings) + (summation of variances). Diagonals represent the square root of average variance extracted and other entries represent squared correlations. Table 4. Results of hypotheses testing Hypotheses Path Path coefficient ([beta]) H1 IE [right arrow] PU 0.57 H2 TE [right arrow] PU 0.19 H3 IM [right arrow] PU 0.11 H4 EM [right arrow] PU .001 H5 PU [right arrow] BIU 0.89 H6 BIU [right arrow] SAT 0.18 H7 IE [right arrow] SAT 0.05 H8 TE [right arrow] SAT 0.36 H9 IM [right arrow] SAT 0.11 H10 EM [right arrow] SAT 0.14 Hypotheses Critical Results ratio H1 5.8 *** Supported H2 2.3 * Supported H3 1.7 (n.s.) Not supported H4 0.02 (n.s.) Not supported H5 9.5 *** Supported H6 2.1 *** Supported H7 0.49 (n.s.) Not supported H8 3.6 *** Supported H9 1.5 (n.s.) Not supported H10 2.3 * Supported Note. *** indicates p [less than or equal to] .001. ** indicates p [less than or equal to] .01. * indicates p [less than or equal to] .05. (n.s.) indicates non-significant. Table 5. Direct, indirect and total effects of the research model Endogenous Determinant Standardized estimates variable Direct Indirect Total Perceived IE 0.572 -- 0.572 usefulness TE 0.192 -- 0.192 ([R.sup.2] IM 0.113 -- 0.113 = 0.58) EM 0.001 -- 0.001 Behaviour IE -- 0.509 0.509 intention TE -- 0.171 0.171 to use IM -- 0.101 0.101 ([R.sup.2] = 0.80) EM -- 0.000 0.000 PU 0.891 -- 0.891 Satisfaction IE 0.050 0.094 0.144 ([R.sup.2] TE 0.360 0.032 0.392 = 0.34) IM 0.115 0.019 0.134 EM 0.143 0.000 0.143 PU -- 0.165 0.165 BIU 0.185 -- 0.185
|Printer friendly Cite/link Email Feedback|
|Author:||Kannan, Kalpana; Narayanan, Krishnan|
|Publication:||Educational Technology & Society|
|Date:||Jul 1, 2015|
|Previous Article:||Active Learning Spaces: New Directions for Teaching and Learning.|
|Next Article:||ET4ET: a large-scale faculty professional development program on effective integration of educational technology.|