A study of factors affecting student outcomes in Vietnamese pedagogical tertiary institutions

ECTING QUALITY OF STUDENT OUTCOMES I.1. Literature reviews on factors affecting quality of student outcomes The literature review shows that many domestic and international studies have examined the content as well as the factors affecting the quality in the higher education sector. Factors measuring the quality in higher education services can be divided into 5 groups: (i) Training program; (ii) Teaching staff; (iii) Facilities; (iv) Learning support; and (v) Additional services. I.2. Literature review on criteria of assessing quality of student outcome Baumert and Kunter (2013) proposed a COACTIV model that evaluates professional competencies and professional competence models with determinants of specialized knowledge in a specific teaching context. Accordingly, employability is considered as a result of the interaction of factors: (1) Professional knowledge (competency in the narrow sense: knowledge and skills), (2) Professional values and beliefs and goals, (3) Motivational orientations, (4) Professional self-regulation skills. I.3. Literature review and the gaps Firstly, there is little research on the factors affecting student outcomes from a business administration perspective. Secondly, little research has measured the impact of the service quality in the higher education sector on student outcomes in pedagogical tertiary institutions in Vietnam. Thirdly, no quantitative research has considered the impact of factors on the quality of student outcomes in Vietnamese pedagogical tertiary institutions. Fourthly, this study is conducted to assess the factors affecting the quality of student outcomes in pedagogical tertiary institutions in the context that Vietnam is implementing a comprehensive renovation of higher education. 5 Chapter II: THEORETICAL BASIS AND RESEARCH MODEL TO MEASURE THE EFFECTS OF FACTORS ON THE QUALITY OF STUDENT OUTCOMES IN PEDAGOGICAL TERTIARY INSTITUTIONS II.1. Higher education and Pedagogical Tertiary Institutions II.1.1. Concepts and objectives of higher education There have been a number of concepts regarding higher education. In Vietnam, although there are no official definitions of higher education, it can be understood that higher education is a form of educational organization for the post-secondary education level with training levels: including college level, bachelor degree, Master's degree and Doctoral degree. Higher education is the period of education that usually takes place at universities, scientific research institutes within different professional fields, and colleges (Law on Education, 2013). II.1.2. Tertiary institutions and their roles in completing higher education objectives II.1.2.1. Concepts and categories of institutions Tertiary institutions are educational institutions belonging to the national education system, which perform the function of training the higher education levels, acting scientific and technological activities, and serving the community. Public institutions are owned, invested, and developed by the government. II.1.2.2. The roles of institutions in completing higher education objectives Tertiary institutions play an important role in implementing higher education objectives II.1.3. Pedagogical tertiary institutions and their activity II.1.3.1. Pedagogical tertiary institutions Pedagogical tertiary institutions are places to train individuals who participate in the career of training people, contributing to training human resources for society. II.1.3.2. The activity of pedagogical tertiary institutions The pedagogical tertiary institutions have the roles of: (i) teacher training; pedagogical training; (iii) connecting with high schools/ pre-schools; and (iv) doing scientific research. II.2. Quality and Quality of Higher education II.2.1. Concepts of quality and service quality Quality is a very broad and complex term, reflecting a combination of economic, technical and social issues. Parasuraman et al. (1988) states that the perceived service quality could be measured through the differences between a service performance and a client’s expectations. II.2.2. Quality of Higher education The International Higher Education Quality Assurance Organisation Network has defined quality of higher education, which is: (i) complying with the prescribed standards and (ii) achieving the set goals. The quality of higher education always changes to meet the needs of society. 6 II.2.3. Theory of service quality and training quality evaluation II.2.3.1. Service quality model by Gronroos (1984) II.2.3.2. Service quality model by Parasuraman et al. (1988) II.2.3.3. Service quality model by Cronin and Taylor (1992) II.2.3.4. 5-aspect service quality model of SEAMEO (1999) II.2.3.5. Model of evaluating training outcomes by Kirkpatrick (1975) II.2.3.6. Model of measuring service quality of higher education created by Abdullah (2006) II.2.3.7. COACTIV model of teacher professional competence II.3. Quality of student outcomes in pedagogical tertiary institutions II.3.1. Concept of Quality of student outcomes in pedagogical tertiary institutions The quality of student outcomes is the overall knowledge, skills and attitudes developed in higher education training, in accordance with the requirements of training human resources for local and national socio-economic development and in each period of time, ensuring and meeting the expectations of related stakeholders and the society regarding higher education human resources’ mentality, intellect and physical strength. II.3.2. Factors comprising pedagogical student outcomes The study used COACTIV model developed by Baumert and Kunter (2013) to evaluate the teacher professional competence. The model consists of four groups of factors with 14 variables. Those factors are (1) Professional knowledge (competency in the narrow sense: knowledge and skills), (2) Professional Values and beliefs and goals, (3) Motivational orientations, and (4) Professional self-regulation skills. II.4. Models and hypotheses II.4.1. Referenced models II.4.1.1. Le Ngoc Thang (2017) research model Le Ngoc Thang (2017) tested the HEdPERF scale created by Abullah (2006b) in the Vietnamese tertiary education context. II.4.1.2. Varana and co. Research model (2015) Varana et al. (2015) applied and proposed a revised HEdPERF model consisting of 05 groups of factors: (1) Academic aspects; (2) Facilities; (3) Training program; (4) Staff; and (5) Supporting services. II.4.1.3. Nguyen Minh Nha and Nguyen Thi Thanh Thuy (2018) model Nguyen Minh Nha and Nguyen Thi Thanh Thuy (2018) have developed and tested the influence of a number of factors on the service quality of Accounting major in Tien Giang University. II.4.1.4. COACTIV model of teacher professional competence Baumert and Kunter (2013) developed a model to evaluate the teacher professional competence which are demonstrated via four groups of factors with 14 variables. 7 II.4.2. Suggested research model The research model is the inheritance and expansion of scales in HEdPERF model created by Abdullah (2005) and the COACTIV model developed by Baumert and Kunter (2013). Figure 2.10 Model of factors affecting the quality of student outcomes II.4.3. Hypotheses Table 2.1 Summarises the study hypotheses Hypotheses Hypothesis expression H1 The training program has a positive impact on professional competence and pedagogy H2 The training program has a positive impact on professional values, beliefs and goals H3 Teaching staff has a positive impact on professional competence and pedagogy H4 Teaching staff has a positive impact on professional values, beliefs and goals H5 Facilities have a positive impact on professional competence and pedagogy H6 Facilities have a positive impact on professional values, beliefs and goals H7 Learning support service has a positive impact on professional competence and pedagogy H8 Learning support service has a positive impact on professional values, beliefs and goals H9 Additional services have a positive impact on professional competence and pedagogy Teaching staff Additional services Facilities Training program Learning support Professional competence and pedagogy Values, beliefs and goals 8 H10 Additional services have a positive impact on professional competence and pedagogy Chapter III: METHODOLOGY III.1. Research design The research process was conducted in two stages: preliminary and official stages. Preliminary stage is conducted with qualitative (preliminary) and quantitative (preliminary) methods. Based on the preliminary quantitative analysis, the revised preliminary questionnaire were used for official surveys. Data collected after the official survey were aggregated, coded and then analysed quantitatively. The steps in a linear structure model include: (i) analysis of the reliability of the scale; (ii) analysis of discovery factors; (iii) affirmation analysis; (iv) analysis and testing of the regression function. Next, based on the results obtained, the study proposed a number of recommendations to improve the quality of student outcomes in Vietnamese pedagogical tertiary institutions. III.2. Qualitative methodology III.2.1. Objectives The qualitative method used in this study is to adjust and supplement the impact factors and constituents of student outcome quality which were developed based on the literature review to fit Vietnamese context. III.2.2. Content of qualitative research Results obtained from qualitative data helped complete the scale and research model. III.2.3. Content of qualitative research Based on the feedback from the researched lecturers and literature review, the study has added the variable "professional values, beliefs and goals" in the official research model. III.3. Variables and scales Based on the inheritance of the model of HEdPERF's higher education service quality created by Abdullah (2006b) and COACTIV model to evaluate teacher professional competence by Baumert and Kunter (2013), the study has developed a scale of factors affecting and measuring the quality of pedagogical student outcomes. III.4. Preliminary quantitative research III.4.1. Questionnaire design The questionnaires consist of two parts. Part 1 consists of questions regarding participants’ general profiles. Part 2 includes scales of factors affecting quality of student outcome in Vietnamese pedagogical tertiary institutions. III.4.2. Preliminary quantitative research results The sample selected in the preliminary quantitative research with the convenient sampling with the participants are lecturers teaching at Hanoi University of Education and Hanoi Pedagogical University No.2. Sixty surveys were conducted but only 50 were collected and valid. For preliminary quantitative research, the reliability analysis of scales using Cronbach's alpha coefficient was conducted. The results show that the variables CSVC4 (Facilities), NH1 (Learning support), NH19-NH11 are excluded. 9 III.5. Official qualitative research III.5.1. Selecting and sampling In this study, the sample was selected based on the clustering sampling method and convenient sampling at the last sampling unit (cluster). The participants are lecturers and alumni of pedagogical tertiary institutions who are currently working at educational institutions. III.5.2. Collecting data The data collection is conducted via survey questionnaires. The questionnaires are sent directly to participants or via one lecturer who is working at the researched university. Totally, 321 surveys have been collected to analyse in the official stage. III.5.3. Data analysis The thesis used the factor analysis method, with the following steps: (i) Verifying the reliability of the scale; (ii) Analysing discovery factors; (iii) Analysing Affirmative factors; and (iv) Analysing Regression. 10 Chapter IV: FINDINGS IV.1. The reality of training and quality in pedagogical tertiary institutions IV.1.1. Training size Statistical results of the Ministry of Education and Training show an increase in the size of universities and colleges (public and non-public) in the period of 2010-2017. The student to faculty ratio is relatively high. Besides, there is an imbalance between the proportion of students studying in all majors. The number of students in Group III and IV accounts for 62.7% compared to the rest. In addition, there is a dispersion in State management of pedagogical tertiary institutions, which makes it difficult to manage and improve the quality of training. IV.1.2. Training programs The pedagogical tertiary institutions’ training programs set clear and specific goals, ensuring flexibility and meeting the diverse learning needs of learners. The training programs are designed to provide learners with knowledge, skills and essential qualities for teaching jobs. IV.1.3. Quality of lecturers and graduates The structure, ratio, number, and quality of lecturers in public tertiary institutions have not yet met the current and upcoming social expectations. IV.1.4. Status of facilities The researched pedagogical tertiary institutions are generally rationally planned, with sufficient quantities, necessary equipment and resources for teaching and learning. IV.2. The results of analysing the factors affecting the quality of student outcomes in pedagogical tertiary institutions IV.2.1. Statistical descriptions of surveyed samples The surveyed lecturer participants are mainly from Hanoi National University of Education and Hanoi Pedagogical University No.2; and Ho Chi Minh City University of Education at 17.4% and 10.6%, respectively. 170 lecturers participated in the survey, at 53%, and 151 alumni accounting at 47%. A high percentage (82.2%) of the participants had more than 5 years working in educational sites. Nearly 80% of the participants have postgraduate qualifications. IV.2.2. The results of the scales’ reliability testing The value(s) of the remaining variables in the scales is highly reliable with the correlation coefficient of the total variable is larger than 0.3 and Cronbach's Alpha coefficient is equal or larger than 0.8 after the removal of inappropriate variables including GV6 (Teaching staff), GV4, CSVC9 (Facilities), CSVC5, DV9 (Services), DV2, NL_NV8 (Competence - Pedagogy), NL_NV7. IV.2.3. Results from the discovery factors analysis IV.2.3.1. Analysis of discovery factors of dependent variables on the quality of student outcomes The EFA analysis indicates a KMO coefficient of 0.887 which is less than 1.0, demonstrating the suitability of the EFA model; the Bartlett test value is significant for Sig. = 0.000, indicating that the variables are correlated with respect to the total number of variables. The results of the EFA analysis showed that two explaining factors indicated 61.921%, which is larger than 50% 11 of the variation of the data set. Therefore, the dependent variables in the research model achieve the convergent and discriminant values. IV.2.3.2. Analysis of discovery factors of independent variables The EFA analysis for the independent variable resulted in a KMO coefficient of 0.995 <1.0, the Bartlett test value is significant for Sig. = 0.000 indicating that the variables are correlated with respect to the total number of variables. The results of the EFA analysis showed that five factors explained 55.955% > 50% of the variation of the data set. IV.2.4. Testing the means The study has conducted a mean testing with two groups of lecturer participants and alumni participants variables (GV_CSV) using the two independent sample testing method (t-test). The analytical results show that the p-value of the Levene test for the competence-pedagogy (NL_NV) and values, beliefs and goals (NT_YT) variables are both larger than 0.05, which makes a conclusion that the 2-sample variance for these variables are not different. The p-value at the t-test for professional competence – pedagogy (NL_NV) and values, beliefs and goals variables (NT_YT) is 0.158 and 0.896, respectively (which are larger than 0.05) demonstrating that groups of lecturers and alumni do not affect the quality of pedagogical student outcomes from the perspective of competence-pedagogy (NL_NV) and values, beliefs and goals (NT_YT) IV.2.5. Results of correlation and regression analysis IV.2.5.1. The correlation analysis of dependent and independent variables The correlation analysis results demonstrate that there is a linear correlation between the independent and dependent variables because the p-value is less than 5%. In addition, the Pearson coefficient between these variables is positive, indicating a positive relationship. This means that the value of the independent variables increases, so does the value of the dependent variables. IV.2.5.2. The impact of factors to student professional competence and pedagogy Table 4.20 shows that there are five factors affecting the dependent variable professional competence - pedagogy including training program (CTDT) (at 1% significance level), teaching staff (GV) (at 1% significance level), facilities (CSVC) (at 5% significance level), learning support (NH) (at 1% significance level) and services (DV) (at 1% significance level). Table 4.20 Estimated results of the regression coefficients with the dependent variable NL_NV Analysing variables Unstandardised coefficients Standardised coefficients t Sig. Multi-collinearity statistics Analysing variables B Standard errors Beta Zero- order B Standard errors Beta VIF (Constant) -.842 .213 -3.962 .000 CTDT .156 .044 .131 3.552.000 .486 .196 .115 .767 1.305 GV .279 .049 .243 5.760.000 .652 .309 .186 .588 1.700 12 CSVC .107 .045 .093 2.395.017 .486 .134 .077 .697 1.436 NH .331 .045 .321 7.300.000 .691 .380 .236 .540 1.850 DV .336 .046 .286 7.326.000 .633 .382 .237 .685 1.460 The coefficient of determination R2 adjusted at 0.647 demonstrating that training program (CTDT), teaching staff (GV), facilities (CSVC), learning support (NH), and services (DV) variables at 66.5% of the variation of the dependent variable. The VIF value is less than 10 and Durbin-Watson (1 <1.657 <3) values show that the model does not have multi-collinearity and there is no superlative autocorrelation between the adjacent errors. The regression model reflects the impact of the factors on professional competence and pedagogy (NL_NV) as follows: NL_NV = -0.834 + 0.131 * CTDT + 0.243 * GV + 0.093 * CSVC + 0.321 * NH + 0.286 * DV. According to this equation, the influence of learning support on professional competence and pedagogy is the strongest (+0.321), followed by services (+0.286), teaching staff (+0.243), training program (0.131), and facilities (+0.093). IV.2.5.3. The impacts of factors to students’ professional values, beliefs and goals Table 4.22 shows that there are three factors affecting dependent variable professional values, beliefs and goals (NT_YT) including teaching staff (GV) and services (DV) (at 5% significance level) and facilities (CSVC) (at 1% significance level) Table 4.1 Estimated results of the regression coefficients with the dependent variable NT_YT Analysing variables Unstandardised coefficients Standardised coefficients t Sig. Multi-collinearity statistics Analysing variables B Standard errors Beta Zero- order B Standard errors Beta VIF (Constant) .685 .302 2.268 .024 CTDT .081 .062 .071 1.310 .191 .263 .074 .062 .767 1.305 GV .172 .069 .154 2.500 .013 .396 .139 .118 .588 1.700 CSVC .413 .063 .369 6.532 .000 .484 .345 .308 .697 1.436 NH -.022 .064 -.022 -.346 .730 .352 -.019 -.016 .540 1.850 DV .164 .065 .143 2.518 .012 .351 .140 .119 .685 1.460 The coefficient of determination R2 adjusted is equal 0.49 demonstrating that variables teaching staff, services and facilities at 49% of the variation of the dependent variable values, beliefs, and goals. The VIF value is less than 10 and the Durbin-Watson (1 <1,895 <3) values show that the model does not have multi-collinearity and there is no superlative autocorrelation between adjacent errors. The regression model reflects the impact of factors on values, beliefs and goals as follows: NT_YT = 0.685 + 0.154 * GV + 0.369 * CSVC + 0.143 * DV. According to this equation, the effect of facilities on values, beliefs and goals is the strongest (+0.369), followed by teaching staff (+0.154), and services (+0.143). 13 IV.2.6. Linear structure model results IV.2.6.1. The results of the Affirmative factors analysis The results of CFA analysis in Figure 4.3 show the following conditions: Chi-square / df = 1,349 ≤ 3 with p ≤ 0.05; GFI = 0.862> 0.8, TLI, CFI ≥ 0.9 and RMSEA = 0.033 ≤ 0.05 are all qualified. As such, the model is considered to be good. IV.2.6.2. Linear structure model results Figure 4.4 shows that the coefficients in the model satisfy the conditions. Specifically, CMIN / df = 1,371 ≤ 2, GFI = 0.859> 0.8, TLI, CFI ≥ 0.9 and RMSEA = 0.034 ≤0.05. As such, the model is considered to be good. Table 4.24 illustrates that there are four factors affecting the dependent variable professional competence and pedagogy (NL_NV), including facilities (CSVC) (at 5% significance level), teaching staff (GV) (at 1% significance level), learning support (NH) (at 1% significance level) and services (DV) (at 1% significance level). Table 4.25 demonstrates that the impacts of learning support on professional competence and pedagogy is the strongest (0.366), followed by teaching staff (0.309), services (0.280) and training program (0.1104). In the linear structure model, the impact of facilities on professional competence and pedagogy is not statistically significant. While facilities have a statistically significant impact on professional competence and pedagogy in the regression analysis model, the impact level is not significant. Therefore, the analysis results from the two models are basically similar. Figure 4.4. The results analysing the impact of factors on the quality of student outcomes The analysis results indicate that there are two factors affecting the dependent variable values, beliefs and goals (NT_YT), including facilities (CSVC) (at 1% significance level) and services 14 (DV) (at 10% significance level). The degree of impact of facilities (CSVC) and services (DV) on values, beliefs and goals (NT_YT) are 0.534 and 0.149, respectively (Table 4.26). The results also show that the impact of the teaching staff (GV) on values, beliefs and goals (NT_YT) is not statistically significant. Meanwhile, in the multivariate regression analysis model, teaching staff (GV) is one of the factors affecting values, beliefs and goals (NT_YT). Table 4.2 The estimated results of the relationship between the factors Correlation Estimation Standard deviation C.R. P NT_YT <--- NH -.160 .101 -1.583 .113 NT_YT <--- DV .180 .095 1.898 .058 NL_NV <--- NH .365 .069 5.270 *** NT_YT <--- CTDT .085 .085 1.008 .314 NT_YT <--- CSVC .663 .115 5.777 *** NT_YT <--- GV .115 .093 1.242 .214 NL_NV <--- CSVC .036 .067 .547 .585 NL_NV <--- CTDT .128 .056 2.283 .022 NL_NV <--- DV .339 .066 5.111 *** NL_NV <--- GV .305 .063 4.816 *** 15 Chapter V: DISCUSIONS AND RECOMMENDATIONS TO IMPROVING THE QUALITY OF STUDENT OUTCOMES IN VIETNAMESE PEDAGOGICAL TERTIARY INSTITUTIONS V.1. Discussions V.1.1. Discussions of the current situation of training activities in Vietnamese tertiary institutions The study has investigated the current situation of Vietnamese tertiary institutions in general and the pedagogical tertiary institutions, in particular. The statistical results show that the structure, ratio, number, and quality of lecturers in public tertiary institutions have not yet met the society expectations. There is a dispersion in managements within the pedagogical tertiary institutions, which makes it difficult to manage and improve the quality of training. Most of the pedagogical tertiary institutions only focus on short-term training and retraining tasks rather than research. V.1.2. Discussions of the findings on measuring factors affecting the quality of student outcomes in Vietnamese pedagogical tertiary institutions Firstly, the training program (CTDT) has a positive impact on professional competence and pedagogy (NL_NV) in both models (multivariate regression analysis and linear struct