Only on Eduzaurus

Assessing Model Of Teaching Beliefs And Practices: Using Structural Equation Modelling

Download essay Need help with essay?
Need help with writing assignment?
writers online
to help you with essay
Download PDF

Teaching reforms need changes in lecturers’ beliefs about teaching and practices for continuing professional development. This study aims to assess the measurement model for teaching beliefs functions and practices among lecturers. The sample of this study was 103 lecturers from Community Colleges, Yemen. Structural Equation Modeling was used to assess and analyse the proposed model for lecturers’ beliefs on teaching functions and practices. Findings of the modified model showed the goodness fit indices of proposed measurement model was improved and showed good goodness of fit. Based on the findings, a tested model assessment can be used as a recommended model for Lecturers’ Beliefs on Teaching Functions and Lecturers’ Teaching Practices among community colleges’ lecturers.

Lecturers’ classroom practices and their professional growth influenced by their beliefs of teaching and learning (Prawat, 1992; Harste and Burke, 1977; Kuzborska, 2011; Mamsour, 2009; Pajares and Nespor, 1992; Latshaw, 1995; Albion, 2001). Lectures decide about classroom instruction theoretical beliefs about teaching and learning. Therefore, many researchers pointed the need of examining lecturers’ thinking and teaching practices (Pajares, 1992; Pajares and Pomeroy, 1993; Clark, 1988, Kennedy, 1997; Mansour, 2009; Standen, 2002; Mansour, 2010; Abelson, 1979). Several reasons explain the complexity of the relationship between lecturers’ beliefs and practices including their knowledge, goals, educational context, and pedagogy (Mansour, 2009; Gahin, 2001; Abell and Roth, 1992, Mofreh ,, 2013).

Essay due? We'll write it for you!

Any subject

Min. 3-hour delivery

Pay if satisfied

Get your price

Community Colleges (CC) in Yemen (CC) recognized that effective lecturers are an important factor to continue its mission and to build skilled graduated students for the labor market. There is a need to understand how lecturers think about their teaching functions and practices. Beliefs of lecturers may play an important role in explaining the individuals’ change of their academic performance of. These beliefs were used to assess new thoughts and concepts about teaching that lecturers confront in their teaching practice in classes (Kennedy, 1997). Therefore, those teachings that are shaped their beliefs which are recognized and characterized as “what is new?” (Kenndey, 1997; Bruner, 1996; Raths, 2001). These beliefs provides lecturers with possible examples of ways to practice those promoted thoughts, solving conflicts between different beliefs, organizational supports, constraints and similar practices.

Currently, carrying out the Community Colleges agenda for a wider participation lecturers’ teaching and learning approaches, needing them to adjust and explore strategies to help students. These developments require the lecturers to actively improve existing practices to improve their professionalism, to discover a firm understanding of the pedagogy of their subjects and of how students learn, but above all else to become reflective practitioners. Thus, this study aims to investigate how the bellies on teaching functions influence their teaching practices among the lecturers in CC.

Lecturer’s Beliefs and Teaching Practices

Many researchers claim that implementing any reform program heavily be influenced by lecturers (Haney, Czerniak, and Lumpe, 1996; Levitt, 2002; Pajares, 1992; Jarvis, 2006; Campbell et al.., 2004; Campbell, 2007). Lecturers play an important role in educational institutions and classrooms change (Prawat, 1992). However, lecturers are also seen as the core obstacles to change their traditional beliefs. According to Bandura (1986), the decisions of an individual through his / her life are strongly influenced by his / her beliefs. Pajares argued that beliefs are “best indicators of the decisions individuals make throughout their lives”. Lecturers’ beliefs play an important role in deciding about curriculum and lecturers instruction program (Nespor and Pajares, 1992). In short, educational researchers have argued the need for a more detailed and direct research of the relationship between lecturers’ beliefs and practices in education (Pajares, 1992 and Pomeroy, 1993). Therefore, the relationship between beliefs and teaching practices is well documented in the literature of science education.

Series of researches have studied the relation of lecturers’ beliefs and teaching practices. Pajares (1992) in his study supported the idea of the influences of lecturers’ beliefs on their performance in the classroom. Similarly, the value of a person who guides behaviors of life was developed by the person’s beliefs (Ajzen, 1985). The beliefs of lecturers have a strong influence on teaching practices (Ernest, 1989). The beliefs and theories of lecturers were described as “the wealth of knowledge that lecturers have that affects their planning, interactive thoughts, and ideas and decisions” (Clark and Peterson, 1986).

Structural Equation Modelling (SEM) was used based on the objective of this study to assess the measurement model for the Lecturer’s Beliefs on Teaching Function (LBTF) and Lecturers’ Teaching Practice (LTP) at community colleges. Several ordered steps were followed to test the model. These included developing the theoretical model, conducting the CFA, constructing a path diagram, assessing model identification, evaluating estimates and model fit, interpreting and analysing the model, and the final model (Stevens, 2002; Norirs, 2005; Kenny, 2006; Garson, 2009; Byrane, 2010; Kline, 2011; Brown, 2011; Zainuldeen, 2012). The relationships between indicators or observed variables and latent variables are indicated by arrows. The path model depicts directional relationships among variables. A straight-arrow is used to specify a recursive relationship.

To decide if the model will be accepted or rejected, at least 3 to 4 tests are recommended. Goodness of fit was evaluated using chi-square for the null hypothesis significance test (Haire et al., 1995; Holmes-Smith, 2006; Zainuldin, 2012). Chi-square (x2) is an absolute fit index. A nonsignificant chi-square showed the parameters that were estimated for the model fit the data.

For this study, the comparative fit index (CFI) and standardized root mean was used. The CFI had a cut-off value of equal to or greater than 0.90 for an acceptable fit and equal to or greater than 0.95 for a good fit (Hu and Bentler, 1999; Byrane, 2010; Zainuldin, 2012) less than .05 was used to show a good model. The root mean square error of approximation (RMSEA) is less than .05 for a good fit or less than .08 for an acceptable fit (Kline, 2011 and Norris, 2005).

For adequate theory testing, the model needed to be over identified. To achieve this, three or more indicators were used for each of the latent variables (Garson and Norris, 2005). SEM includes CFA, which was used to test the measurement model as previously showed. Parameter estimates were used to show how well the indicators corresponded to the latent variables (factors). Parameter estimates used for this include variance and covariance of the indicators and factor loadings and residuals. Indicators should have coefficients (factor loadings) of 0.6 or higher on the latent factors (Awang, 2012).

Measurement Model

Because of the need of explaining a fit model, analysing the initial model was made by calculating estimates of the model. The initial model, as explained in Figure 1, is based on the unidimensionality, validity, and reliability analysis. The unidimensionality was achieved when measuring items having acceptable factor loading equal or higher than the value of 0.5 for respective latent construct (Awang, 2012). As shown in Figure 1, construct items had good satisfactory factor loadings hence representing unidimensionality. The validity of the measurement model analysed the convergent validity, construct validity, and discriminative validity. According to Awang (2012), the convergent validity could be verified through AVE (Average Variance Extracted) and the AVE should be greater or equal to 0.5. The AVE were calculated for the measurement model by calculating the sum of variance of constructs and then dividing it by the number of constructs of the Lecturers’ Beliefs on Teaching Functions (LBTF) and Lecturers’ Teaching Practices (LTP). The AVE of LBTF constructs was 0.73, and the AVE of LTP was 0.65. The results of AVE indicated that all items in the measurement model were statistically significant. The discriminative validity of the measurement model was achieved when the measurement model was free from redundant items, or when the correlation between each pair of latent exogenous construct is less than 0.85 (Awang, 2012). Figure 1 showed the good discriminative validity of the initial measurement model.

Table 1 explained the goodness of fit indexes used to evaluate the initial measurement model. As shown in Figure 4.6, the Chi-square was significant, value of CFI was value was 0.93, value of TLI was 0.92, IFI Value was 0.93, NFI value was 0.89, SEMR value was 0.56, RMSEA value was 0.11 and Chi-square/df value was 2.23. The values of Chi-square, CFI, TLI, NFI and Chi-square/df showed acceptable goodness fit of measurement. However, the goodness fit indexes of NFI was 0.89, SRMR was 0.56 and the RMSE was 0.11 which showed low goodness fit. Therefore, the initial measurement model needed modification. There is a series of goodness of fit indices that reflect the fitness of the model. It was recommended to use at least three fit indexes by including at least one index from each category of model fit (Norris, 2005; Garson, 2009; Awang, 2012).

Name of category Name of index Level of acceptance Index Level results

  • Absolute fit Chisq P > 0.05 Significant
  • RMSE RMSE < 0.08 0.11
  • Incremental fit CFI CFI > 0.90 0.93
  • TLI TLI > 0.90 0.92
  • NFI NFI > 0.90 0.89
  • Parsimonious fit Chis/df Chis/df < 5.0 2.23

Modification indices were used to improve model fitness (Garson, 2009). However, modification needed to be consistent with the theory used to propose the model. Figure 2 and Table 2 depicted a new measurement model. In the modification model, the goodness fit indices of proposed measurement model were improved and showed good fit as showed in Figure 2 and Table 2.

  • Name of category Name of index Index in initial model Indexes in proposed model
  • Absolute fit Chisq Significant Significant
  • RMSE 0.11* 0.076*
  • Incremental fit CFI 0.93 0.97
  • TLI 0.92 0.96
  • NFI 0.89* 0.93*
  • Parsimonious fit Chisq/df Chisq/df 1 = 2.23 < 5.0 Chisq/df = 1.58 < 5.0

The measurement model, therefore, provides an integrated model of teaching functions and practices. Findings of the modified model showed the goodness fit indices of proposed measurement model was improved and showed good goodness of fit. The proposed hierarchical model is made up two levels with LBTF construct variables being the first level while the LTP constructs make up the second level. Therefore, this model provides a conceptual background for future analysis of beliefs on teaching functions and practices in community colleges.


This essay has been submitted by a student. This is not an example of the work written by our professional essay writers. You can order our professional work here.

We use cookies to offer you the best experience. By continuing to use this website, you consent to our Cookies policy.


Want to get a custom essay from scratch?

Do not miss your deadline waiting for inspiration!

Our writers will handle essay of any difficulty in no time.