Impact of leadership style in leadership style. Research at universities in Vietnam

Leadership style of the head of the organization has a great influence on the success or failure of an organization, today in modern management there are many different leadership styles such as democratic leadership., authoritarian leadership style, liberal leadership style, transaction leadership style, transformation, destination leadership style, ethical leadership style

Topic : Impact of leadership style in organizational innovation Research at universities inVietnam

Chapter 1: Introduction (1-2 pages)

1.Reason for choosing topic

Leadership style of the head of the organization has a great influence on the success orfailure of an organization, today in modern management there are many differentleadership styles such as democratic leadership., authoritarian leadership style, liberalleadership style, transaction leadership style, transformation, destination leadership style,ethical leadership style, According to the author Karin & e (2010) said that the rapidlychanging practice requires more and more new models of leadership styles, which aremore flexible and suitable, bringing high efficiency to the organization's operations.Therefore, when applying any model of leadership style, the organization also needs tocarefully consider specific factors such as geography, culture, religion, customs toadjust for fit and achieve the organization's goals

In fact, the above leadership style models have been applied very effectively inbusinesses around the world Mary (2004) said that these leadership style models showdiversity, many orientations, multiple subjects, multiple objects and even othercompetitive connotations In that same spirit, inheriting the study of Downton (1973) etal., 1985 Bass, North use (2007) extended free, transitional and transactional models byacknowledging them as an interdependent process is customary and has given birth to acontinuous leadership style model In this style, Bass points out that leaders inspire theiremployees to do better through: Increasing employee awareness of the organization'sgoals; improve the interests of each individual on the basis of the common interests of theorganization; address needs at a higher level This leadership style model demonstratesthe connection between transformational leadership styles, transactional leadership stylesand freelance leadership styles that have a direct impact on the performance of theorganization To further confirm the above conclusion, according to the most recent

research by Maria L Perkins (2014), once again, the influence of leadership style on theorganization's performance is applied, when applying the style model leadershipcontinuity, one model converges many advantages over others In addition, empiricalstudies on the influence of leadership style on the performance of Glynis M Breakwell et

al (2010), or on the satisfaction and working motivation of lecturers Anam Siddique &colleagues (2010) in university Therefore, it is theoretically necessary to have empiricalresearch on an organization's leadership style in universities to clarify the impact of anorganization's leadership style on innovation in that organization

In fact, up to now, in Vietnam, there are no empirical studies on the influence ofleadership style on the performance in universities For the reasons mentioned above,the "The influence of leadership style on innovation results in organization - research atVietnamese universities"

4 Research Methodology

To achieve the above objectives, the author uses a combination of research methodsqualitative and quantitative research

Qualitative method

Qualitative research process is as follows:

Based on the theory of leadership style and the results of related research review, theauthor proposes research model and hypotheses On the basis of the questionnaire that theauthors previously researched, the graduate student inherited to build the questionnaire tosuit the context of Vietnamese education The research student conducted in-depthinterviews to collect ideas, supplement and complete the official survey (Subject and list

of universities conducting in-depth interviews are shown in appendix 2 and appendix 4)After interview:

- Data will be translated and recorded to word file;

- Identify the main topics covered by keywords;

- Filter separation of new ideas;

- Encrypt data for grouping;

- Identify groups of similar and distinct data;

Find relationships between information to build hypothetical models Complete officialsurvey

Quantitative methods

- The collected information will be quantified, show common points and differences,synthesize into control variables, independent variables, dependent variables, and encodenecessary information in the questionnaire asked multi-factor leaders (Avolio & Bass,2004) to input data and analyze data using SPSS software Using model descriptionmethod, correlation regression to evaluate the impact and relationship between thevariables

- Carry out the statistics to describe the collected data, then proceed with the followingsteps: Verifying the value of the variable by EFA factor analysis method; Evaluate thereliability of the scale with Cronbach alpha reliability coefficient; Multivariate regressionanalysis

The specific contents are explained as follows:

Sample descriptive statistics: Describe the characteristics of the research sampleaccording to predefined distinguishing signs

Scale test: Because the research variables are built from many different observedvariables To check the reliability of variables, use Cronbach's Alpha coefficients, with aminimum test standard of 0.7 To check the suitability of a question, it is necessary toconsider the total variable correlation coefficient with the test standard of at least 0.3

Explore factor analysis (EFA): This step will reduce the data from many observedvariables into fewer factors while still reflecting the significance of the research data.Some standards when the EFA is the minimum KMO coefficient of 0.5, the Bartlett testhas p-value less than 0.05, the minimum explanation variance is 50%.

Factor analysis (CFA): is an analytical method based on the perceived understanding oflatent factors in the model through theory or experiment (through EFA) The estimationmethod used is the estimate by the maximum rational function The model is considered

to be suitable with the research data when the indexes: Chi - square adjusted for degrees

of freedom are less than 2, in some new cases it may be less than 3 (Camines & Mever,

1981 cited by Nguyen Khanh Duy, 2009), RMSEA is good at 0.05, in Vietnam authorsNguyen Dinh Tho and Nguyen Thi Mai Trang (2008) suggest RMSEA is below 0.08

5 Contribution of the topic

As stated, research on leadership styles in practical application is always lacking andneeds to be supplemented The relationship and influence level of the components in theleadership model are carefully researched, helping the organization administrator to havemore references to improve leadership skills Contribution

The most important part of this study is the construction of a scale that tests theleadership style components in the continuum of leadership model, which can help thenext researches to use this scale your research

In theory: by the results of survey data from 127 universities in Vietnam, the thesis hasmade new contributions to the theory of leadership style in Vietnamese universities,specifically:

(a) affirms factors such as transformative leadership style and trading leadership style,liberal leadership style that has a positive relationship with university performance andthe degree of influence of each factor;

(b) affirming the relationship between the characteristics of universities such as: type,size and culture influencing leadership styles and performance in universities;

(c) the thesis has confirmed the positive relationship between professional results andfinancial results in universities in Vietnam

+ In practice: from the main results of the thesis, the author has (a) assessed the current

state of leadership in Vietnamese universities and thereby (b) made recommendations toThe Ministry of Education and Training and the universities refer to improve theefficiency of school change in the context of university autonomy under the influence ofleadership style

6 The layout of the thesis

The layout of the thesis is divided into the main sections as follows:

-Heading

- Chapter 1: Research overview of leadership style, performance results

movement in university

- Chapter 2: Literature Review (9-10 pages)

- Chapter 3: Research results (Impact of leadership style in organizational innovation.Research at universities in Vietnam)

- Chapter 4: Conclusions and recommendations

- References

- Appendixes

Chapter 2: Literature Review (9-10 pages)

2.1 Leadership style in university

2.1.1 Leadership style in the organization

Leadership styles are the methods or ways a leader often uses to influence the subjectsbeing led (Bryman, A (2007) Personally, leadership style is the way The leadershipstyle of an individual is the form of the person's behavior that demonstrates attempts toinfluence the performance of others.As a whole, leadership style is the system.characteristic signs of management activities of leaders, are determined by theirpersonality traits Currently in the world, there are many authors studying leadershipstyles in different organizations Specifically:

Approach according to the essential characteristics and qualities of leaders:

In the overall perspective, DSL is a system of characteristic signs that governmanagement activities of leaders, defined by their personality characteristics Currently,there are many authors researching on DSL in organizations and approaching manydifferent directions, specifically:

According to the author House and Podsakoff (1994), "Research on leadership styles insociety", the author argues that the personality model of typical leaders should have: 1 /Vision; 2 / Passion and sacrifice; 3 / Trust, determination and persistence; 4 / Build goodimage; 5 / Exemplary; 6 / Social communication; 7 / Able to actuate when needed; 8 /The ability to arrange well; 9 / Ability to inspire According to the author, Vision is theability to foresee future problems so that leaders can plan These are the most importantcharacteristics of a good leader because it clearly distinguishes it from other mediocreleaders, those characteristics are: Passion and self-sacrifice for work and for others in theteam function This characteristic represents the love and sacrifice of the leader; Trust,determination and persistence are factors that bring success to a leader The perseverance

to pursue the ultimate goal, the employee's absolute confidence in the realization of workand life; Build a good image for employees to show exemplary status, build trust andreputation with their employees; A capable leader needs to be exemplary in order for hissubordinates to respect; Social communication is a characteristic a talented leader shouldhave; Ability to mobilize and gather people when needed; Good sorting ability;Inspirational ability, These are important skills that bring success to a leader In this

study, the author has pointed out the necessary characteristics of a leadership style on thebasis of theory, the study has not considered the specific impact of each factor on theactivities in the organization This is one of the empirical studies, research has shown theessential characteristics of leadership in organizations On that basis, it is necessary tofocus on more prominent characteristics to help leaders have good leadership styles andachieve high efficiency in higher education.

Author McShane and Von Glinow (2013), the authors point out that, to become a leader,

it is necessary to converge the following 7 factors: Sensitivity; Integrity; Energy;Confident; Motivated to be a leader; Intelligence; Expertise According to the author,these factors are defined specifically as follows: "sensitive" - the leader always feelsquickly about the employee's attitude, emotions and desires; "Integrity" - is one of thefactors that are expected by employees This integrity makes employees feel confident; animportant factor is the basis for employees to decide whether to trust in leadership or not;

"Will" - help leaders overcome internal and external difficulties, dare to face challenges.Demonstrating the difference between leadership and ordinary employees and this is thefactor that helps employees admire the leadership; "Confidence" is a very necessaryfactor in the working process in general and used in special cases in particular such asexplaining the problem in public; “Leadership Motivation” - Leaders can show whetherthey are ambitious or not, but in reality they always need leadership motivation to be atrue leader; "Intelligence" - Just average or above Good leaders do not require too much

of this factor However, it is essential for leaders to be able to analyze problems andopportunities; "Expert knowledge" - this is only a decent factor, mainly to assist thedecision-making process The capacity of each person is limited, if the leaders are tooprofessional, they hardly have enough time to be leaders All the above authors mainlyresearch according to qualitative methods and stop at only the basic characteristicsrequired of a leader and leadership style to achieve the goals set out of the organization

2.1.2 Leadership style in university

The role of leadership style is confirmed by many authors to be very necessary, especiallycreating a competitive advantage for the development of the organization (Brandy Smith,2016) In the university, the role of leadership becomes even more important because foruniversities with a mission to create, transfer knowledge and develop human resourcesfor the society, not just limited to development knowledge to develop your ownorganization The world higher education environment is changing in a more competitivedirection, prompting universities to soon develop into active learning organizations toimprove their teaching and research results (Dill, 1999) The organization's leadershipstyle has a direct influence on the behavior and attitude of university staff and faculty and

is one of the key factors in the school's outstanding performance ( Nafei et al., 2012)

In the study of Bass & Avolio (1994), the author also points out the specific influence oftransition, transactional and liberal leadership styles in universities Research shows thatleaders of this style are willing to take risks to make a fundamental change in theorganization This approach emphasizes a principled understanding: change is animportant part of an organization and therefore plays an integral role in leadershipsuccess The author conducts research at university With a sample of 143 staff, lecturersand accepted research hypothesis confirming the positive effect of transition leadershipstyles compared to other styles on work commitment of workers at school (0.678 and issignificant with p <0.05)

Studies of leadership styles in universities are of interest to many authors In Australia, inrecent years, RamSDEN (1998), according to the author's research, in the near future it iscertain that universities face change Basic changes in the environment - specific highereducation such as: increasing the amount of knowledge, increasing scientific products,emphasis on working skills for students after graduation, pressure on the public scientificresearch, changing the way it operates and operates At the same time, research andteaching standards will be stricter, leading to more ongoing university competition Atthat time, the leadership style in the university also needs to change to create trust for

employees, maintain working morale, improve work efficiency, and help employeesadopt important change weight easily.

According to author Radam (1998), in the theory of leadership styles, the leadership styleunder the continuous model (transitional, transitional and free leadership styles) isappropriate to the educational context education because with changes in highereducation innovation and this style has the potential to have a high impact on theperformance of the university These leadership styles demonstrate encouragement, andinspiration that increases motivation in teaching and research Specifically, thetransformational leadership style has been studied by many authors to have a positiveimpact on the changing organizational conditions of the school

Researching each style individually, the author Northouse (2001) has identified theleadership role of transformation as the ability to lead the employees to change andimprove in a more positive direction Although “leadership transformation style” is asimple concept, it helps to assess personal motivation, meet the needs of the organizationmembers and is a style that is highly appropriate to the context change in organization.Inheriting Burns (1978) research describes transformational leadership as a way to meetworkers' needs and assist them in achieving greater work efficiency; This leadership stylealso promotes organizational engagement through a work environment where respect isvalued and participation is encouraged

Following the circuit of research, in 2007, author Northouse pointed out the suitability ofthe model of transformative, transactional and liberal leadership styles suitable in theeducational context, especially in the changing innovation environment Researchemphasizes innovation in universities, which means that successful educational leadersneed to have a deeper understanding of the working environment in universities, aboutthe products of higher education specificity and unlike other products

2.2 Innovation

2.2.1 General theories of innovation

2.2.2 Is Innovation always good?

2.3 Innovation and leadership

2.3.1 Visionary Leadership to drive innovation

2.3.2 Learning organization to achieve innovation

2.3.3 Resources spent in innovation

2.3.4 Incentives to boost innovation

The study of the influence of leadership styles on university organizational innovation is

a new study in Vietnam and the research objectives set out in the overall research design,Research method selected by the author is a combination of qualitative research andquantitative research Qualitative research is used to help the author discover leadershipstyles in Vietnamese universities and is also an adjunct to the official quantitativeresearch of the thesis Quantitative research is done through the method of surveying theopinions of leaders, staff, department managers, lecturers of universities according to theresearch content The dissertation is done on the basis of experimental research methods,the author starts the research by defining research objectives, then conducting anoverview of the research on theory and reality of the university in Vietnam to identifyresearch hypotheses Next, the author conducts investigative research at a university toclearly define leadership styles in universities in Vietnam and complete the variables

measuring research concepts Finally, quantitative research helps collect and process data

to test research hypotheses and report research results

The collected data is the basis for evaluating the quality of the scale, testing the modeland research hypotheses Data analysis techniques such as exploratory factor analysis(EFA), affirmative factor analysis (CFA) and linear structural model (SEM) have beenused by the author to evaluate the quality of the scale and testing research hypotheseswith the support of SPSS and AMOS software were used by the author to evaluate thequality of the scale and test research hypotheses with the support of SPSS 22.0 software

3.2 Study sample design

3.2.1 Sample population

As of August 2019, the country had 235 universities and institutes: including 170 publicschools, accounting for 72.3%; 60 non-public schools, accounting for 25.5%; 5 schoolswith 100% foreign capital, accounting for 2%; Scientific research institute (according tothe summary conference of the Ministry of Education and Training) If divided by regionsand regions, the universities are divided into 06 regions and the number of schoolscorresponding to the following table:

Table 1 Statistics of the number of universities by region

universitiesNotes

Notes

total 235

Source: Ministry of Education and Training, (Aug 2019)

In addition to the above ownership classification, universities in Vietnam can beclassified according to their scale including: “(1) University (National University,Regional University), (2) Universities and Institutes by majors and (3) LocalUniversities ”(National Assembly, 2012)

3.2.2 Sampling method

With the overall sample of 235 universities and the purpose of the study is also expected

to reflect clearly on the leadership style of the organization in universities in Vietnam Inorder to study in-depth about each group of schools according to the type of organization,the author sent questionnaires to all managers, lecturers, departmental staff of universities

in the overall research sample as above To ensure the reliability of the research data that

is representative of the population (95% confidence) and the quality of the study shown

by the tolerance level of +/- 5% as well as the numerical requirement The number ofresearch samples for exploratory factor analysis (EFA) techniques as published by Wolf

et al (2013), the author expects to receive feedback from 127 universities (accounting forover 60% of the in the overall study sample) This number of samples is also consistentwith the formula for calculating the sample size when the overall research is identified as

a finite number as presented in the textbook "Survey method: principles and practice" ofthe authors of the National Economics School (Nguyen Thi Tuyet Mai et al., 2015).The sample size as above is also consistent with the method of proving the researchhypothesis using multivariate regression according to Hair et al (2006) For multivariateregression method, the minimum sample size is calculated by the formula: 50 + 8 * m (m

is the number of independent variables) In this study, the number of independentvariables is 7, so the minimum sample size is: 50

+ 8 * 7 = 106 observations Therefore, the author obtained the feedback results from 127schools in the total 235 universities of the country This result is consistent with theformula for calculating sample size above, with 95% confidence and 5% sampling error,

as well as greater than the number of samples required for the multivariate regressionmethod (Hair et al , 2006)

In the study of the influence of leadership style on the organization's performance, theauthor uses survey data using quantitative research methods based on collectingevaluation opinions of team members That function, previous studies in enterprises inother countries, other scholars get data from the top leader himself (Alan Bryman (2007)and some other studies surveyed through the management staff (Alberta Thrash (2012).Optimal case, data collected in this study will come from the Leadership representatives

of the surveyed universities organizations in these universities do not have to get votesfrom the management board but can get votes from managers and senior lecturersbecause in this type of organization there are specific knowledge Like universities, theabove individuals are also suitable for the survey to ensure the research quality.Researching organizational content like this study is also very difficult to collect datafrom university leaders and it will be more objective when collecting data from manypeople in the university Therefore, about the research framework in each university, theauthor collected comments from the managers at all levels and lecturers and researcherswho have worked for> 5 years at the university, each university includes Thequestionnaires belong to both groups of respondents above The above framework is alsoconsistent with the results of previous empirical research in businesses and universities(Jeremy Dutschke, Ph.D Dallas Baptist University (2005) on organizational leadershipstyles

In this thesis, the author conducts research on the influence of leadership style onorganizational innovation in universities in Vietnam and the analytical unit here at theorganizational level (university) learn) To collect data for the organization's leadershipstyle analysis at the university level, after collecting responses from individuals

(managers, lecturers) of each In schools, the author checked the collected data of eachschool, removed the inconsistent samples and aggregated the data by each university byaveraging the values of the observed variables to get the data is aggregated by field) Thismethod of data synthesis was carried out by Maria L Perkins (2014) in her doctoralthesis on organizational leadership styles in US businesses at Capella University in 2014,

as well as according to research by Alberta Thrash (2012) in universities in severalcountries around the world In the above studies, to study the organizational unit, theauthors synthesized the opinions of middle managers in the company by averaging thevalues obtained from the variables observed through people research has proved thatalthough this synthetic method has limited error, it has been accepted to match the actualconditions of collecting research data on the team function

2.3 Scale development

2.3.1 The process of developing the scale

On the basis of research reviews and theoretical bases of previously published studiesrelated to leadership styles in universities, the author has proposed a research model withthe central variable being style Leadership, the author uses the continuous leadershipstyle model with 3 styles: transformational leadership style, transactional leadership style,liberal leadership style

The results of the research review show that the observed variables measuring leadershipstyle and performance have been researched and published by the authors but mainlyapplied to businesses and a few for universities learn The above studies were all doneabroad, where the research context has many differences with Vietnam

Qualitative research to understand the nature of leadership style in universities inVietnam as well as assess the appropriateness of the relevant scales of previous studies.Specifically, the author has conducted qualitative research by in-depth interviews with theadministrators, faculty, and managers of 20 universities to understand the characteristics

of leadership styles in universities In which, the author studies both fully autonomous

and partially autonomous public and non-public schools to better understand themanagement mechanism, leadership style characteristics in each relevant school to theprocess of higher education reform in Vietnam

The case study was conducted by the author at a public university in Vietnam asdescribed above This result combined with qualitative research phase helps the author toadjust the scales to ensure it is suitable with the research context in universities inVietnam Specifically, the list of observed variables used in this study is presented inappendix 7 of the thesis

All variables in the above research model used the statements and Likert 5 scale todescribe frequency (1 = Very rarely; 5 = Very often) to measure based on perceptions ofmanagers, lecturers at universities These questions are based on scales, according to Bass

& Avolio's original MLQ questionnaire (2004), and have been verified in previous relatedstudies, translated from English into Vietnamese and back-translated to ensure thelanguage conversion is correct, does not change the meaning of the original scale Somescales based on the results of interviews and group discussions with experts, researchersand management experience in universities in Vietnam have been adjusted and perfected

to suit the actual situation

2.3.2 Leadership style scale

Transformational leadership

Transformational leadership is a style of leadership that does work by evoking andtransforming individual emotions, values, ethics, standards, and long-term goals To dothis, leadership is through fostering credibility and extending the foresight throughout theleadership process (Northouse, 2007; Bass & Avolio, 2004)

According to Bass & Avolio (2004), the transformative leadership style consists of fourmain elements:

Influenced by ideal (idealized influence); (2) an insperational motivation; (3) intellectualstimulation; (4) individualized consideration.

Idealized influence

LT1- Leadership makes employees feel confident working by their side

LT2- By fairness in solving work Leadership creation

gain trust with employees

LT3- Employees feel proud of working with leaders

LT4- Leaders have a sense of vision when managing employees, listening, sharing andrecording comments and exchanges from employees

Inspirational motivation

ĐV1- The leader always gives instructions to employees in a simple way

ĐV2- Leadership describes jobs to employees in an attractive way Unit3- Leaders helpemployees find meaning in the work Unit4- Leaders care, share, and help employees atwork

ĐV5- Leadership builds solidarity to create a comfortable working spirit for employeesĐV6- Leadership builds working spirit based on the principle of respecting staff

Individualized consideration

QT1- Leadership helps employees develop themselves (including: advancing at work,learning to improve qualifications, creating prestige in the profession, building self-branding)

QT2- Leaders express views to employees about their way of working

QT3- Leadership explained clearly to employees why they did not achieve the desiredresults

QT4- Leadership always ensures employees have a good income to stabilize their lifeQT5- Leadership creates a friendly working environment for employees to work mostcomfortably

QT6- Leadership sharing personal issues of employees

Transaction leadership style

Trading leadership style: is a style of leadership that focuses on supervision, organizationand performance; Trading leadership is a style of leadership in which leaders promotefollower compliance through both reward and punishment Unlike transition leaders, whouse a transactional approach who do not want to change the future, they keep everythingthe same (Northouse, 2007; Bass & Avolio, 2004)

According to Bass & Avolio (2004), the trading leadership style consists of two mainelements:

Trading leadership style is measured by these factors: Commendation

Unexpectedly, managed by exception

KT1- Leaders tell employees what to do if they want to be rewarded for their work

KT2- Leadership rewards employees when they finish a job well KT3- Leadershipalways makes employees happy when they finish a job well Management by exception

QLNN1- With the viewpoint of "thinking is doing", the leaders are ready to change

QLNN2- Leadership clearly tells employees what standards they must know to do a goodjob

QLNN3- Leaders explain clearly to employees when they are not finished, do not finishwork before giving disciplinary action

Liberal leadership is also known as representative leadership, which is a type ofleadership in which leadership is handed over and allows team members to makedecisions

According to Bass & Avolio (2004), liberal leadership styles include:

TDO1- Leaders are pleased to keep employees working in the way that employeesnormally do

TDO2- Leadership respects the employee's creativity in work

TDO3- Leadership creates conditions for employees to practice their creativity at workTDO4- Leadership does not ask for more from employees if it is not really essential

2.3.3 Scale of university performance

According to the previous research results of Chen (2009) and Wang (2010), the authorclassified the school's performance by two factors: professional results and financialresults In particular, the professional results are measured by the results of teachingactivities and the results of research activities of the school The scale is built to measurethe change in performance in the past 3 years (2015-2017) according to the assessment ofleaders, staff, lecturers, researchers of the house school

Professional results: Professional results are one of the most important criteria whenmeasuring university performance to reflect the reputation and quality of the school It isthese component results that help classify the school according to the respectivedevelopment orientations: research orientation, application orientation or practice

orientation The Ministry of Education and Training has given a set of universityevaluation standards and criteria that include many indicators related to the professionalquality of the school to serve as the basis for grading and grading respectively Based onthe research results of Chen (2009), the Ministry of Education and Training (2007), andthese scales are also tested at Vietnamese universities in the thesis of author Tran QuangHuy (2017).

Professional results are divided into: teaching results and scientific research results:KQGD 1 - The university's organic faculty to student ratio has increased over the past 3years;

KQGD 2 - The school facilities are good for the teaching of the teachers

KQGD 3 - Students are satisfied with the quality of their teaching

KQGD 4 - Employers appreciate students' competencies after graduation

KQGD 5 - After graduation, students can find jobs suitable for their majors

KQGD 6 - Students are interested in the modules of the training program

KQNCKH1- The number of articles published in domestic journals of the school tended

to increase in the last 3 years

KQNCKH2- The number of articles published in international journals of the schooltended to increase in the last 3 years

KQNCKH3- The number of textbooks, monographs, and lectures of the school tends toincrease in the last 3 years

KQNCKH4- The number of scientific research topics (school, ministry, State level) of theschool tends to increase in the last 3 years

The financial results show the ability of the school to balance finance, ensuring thequality of teaching and research activities and ensuring the lives of staff and lecturers

Three specific observed variables used based on the study of Chen (2009) and theMinistry of Education and Training (2017) include:

2.4 Design of survey questionnaires and data collection

2.4.1 Design of survey questionnaires

On the basis of a research model with variables and scales, the original MultifactorLeadership Questionnaire (MLQ) questionnaire was developed over 20 years ago (Bass,1985) Various versions of MLQ have been used in the United States and more than 30countries, and the MLQ translation has been completed in multiple languages (Avolio &Bass, 2004) The author has conducted to build survey questionnaires through manydifferent drafting, editing and finishing steps Specifically, the author has madediscussions in the expert group and educational administrators in universities (list ofschools discussing expert groups according to annex 4), during qualitative research andModifying the first survey questionnaire in a way to refine the translation of thequestions, adjust them to suit the Vietnamese context, but do not change the content ofthe question Next, the author used this edited questionnaire for case studies at 01 publicuniversity in Vietnam and the second revision in the direction of adding some newquestions and completing questions in the first questionnaire Finally, before Officiallydeploying the author to test on a small sample and the third adjustment, clarifying thequestions to help the respondents easily and accurately The questionnaire contentincludes 3 main parts (details are shown in attached Appendix 3):

Introduction: This content introduces the purpose and meaning of the study and providesguidance on how to answer the questionnaire

Key information about the statements: This section will include statements in recognition

of the respondent's judgment A total of 45 questions in 2 research contents, in whichLeadership style in university consists of 31 questions (reflecting 3 leadership styles inthe continuum model: leadership transformation style, leadership style trading leadership

and liberal leadership style) and 14-question performance (reflecting 2 factors ofprofessional results and financial results).

Statistical information: to collect other information related to the respondents and theuniversities they are working in to statistic, describe samples and serve as a basis for in-depth research by each group of universities

2.4.2 Collect survey data from universities

Data collection in this study was done by survey method via online questionnaire (onlinequestionnaire on google docs at: https://goo.gl/forms/ZtnfDdxLSyqbqr8U2) and paperquestionnaire On the basis of university information from statistical information aboutuniversities and colleges on the Internet, the author has compiled a list of all universities

in Vietnam classified by region , size and type of school are consistent with the overallresearch sample as above On the basis of research at the University of Labor and SocialAffairs, is a research student university that is working to send dispatch withquestionnaires to universities by post Through information about implementation clues,and the author goes directly to the schools for comments, combined with sending links toonline survey questionnaires via email to serve as a basis for data collection through bothchannels: online and paper questionnaires for data collection

2.5 Data analysis

After receiving the questionnaires, the author cleaned the data and encoded the necessaryinformation in the questionnaire and entered data Data analysis techniques such asdescriptive statistics, assessing reliability, accuracy of the scale, checking the suitability

of research models and testing research hypotheses are done on the software SPSS 22

2.5.1 Descriptive statistical analysis

Descriptive statistics are the first step in analyzing quantitative data to determine thedistribution of the data set under study Two statistical methods of inclination (Skewness)

and sharpness (Kurtoris) are performed to check whether the data follows the normaldistribution law.

Tilt is a measure of the degree of asymmetry of the probability distribution of a randomvariable

Kurtosis measures the degree of flatness of the distribution relative to the normaldistribution (zero sharpness) The distribution is acute when the kurtosis is positive andflat when the kurtosis is negative

For the normal distribution, the value of the inclination and the sharpness is zero Based

on the ratio of the inclination value to the taper value and its standard error, we can judgewhether the distribution is normal or not (when this parameter is less than -2 and greaterthan +2, the distribution is abnormal)

The study conducted statistical descriptive analysis of observed variables in the modelthrough criteria including: maximum value, minimum value, average value and standarddeviation

The thesis analyzes the SEM model with AMOS software with the tests on the suitability

of the model and the Bootstrap test for the analytical model to evaluate the errors inresearch data

2.5.2 Test scale

To ensure the quality of the scale ready for testing research hypotheses in the nextsection, the author uses a combination of two methods of exploratory factor analysis(EFA)

Exploratory Factor Analysis (EFA)

The reliability of the scale is assessed through Cronbach's Alpha coefficients According

to Hoang Trong and Chu Nguyen Mong Ngoc (2008), scales with Cronbach's Alphacoefficients from 0.6 to 1 are acceptable, scales with Cronbach's Alpha coefficients from0.7 to 0.8 are usable and the scales with Cronbach's Alpha coefficient from 0.8 - close to

1 are good scales In this study, to ensure good reliability, scales with Cronbach's Alphacoefficients of 0.6 or more were retained for further analysis.

However, Cronbach's Alpha's coefficients only show if the scales are linked, but doesn'thelp in deciding which observational variable to keep or remove Therefore, the studyuses an additional total correlation (item - total correlation) coefficient to have more basisfor making decisions whether to keep or remove the observed variable

The total variable correlation coefficient is the correlation coefficient of a variable withthe mean of other variables in the same scale, so the higher the coefficient, the higher thecorrelation of this variable with other variables in the scale high According to Nunn allyand Burnstein (1994), variables with total variable correlation coefficients <0.3 areconsidered garbage variables and remove this variable from the scale

Therefore, according to the above mentioned variable selection criteria, to ensure thereliability of the scale, the study will select the scales that ensure Cronbach's Alphacoefficient ≥ 0.6 and have total variable correlation coefficients > 0.3

- Verify the value of the scale by exploratory factor analysis

- EFA: The method of factor analysis explores that EFA belongs to interdependentmultivariate analysis group, that is, there are no dependent variables and independentvariables, but it is based on the correlation between the variables EFA is used to reduce aset of k observations into a set F (F <k) of more significant factors The basis of thisreduction is based on linear relationships of factors with observed variables (NguyenDinh Tho, 2011)

Factor analysis will be used to verify convergence of component variables To confirmthat the analytical data is appropriate and highly reliable, the following criteria must bemet:

KMO coefficient (Kaiser - Meyer - Olkin) must reach a value of 0.5 or more(0.5≤KMO≤1), showing that factor analysis is appropriate Bartlett test is statistically

significant (Sig value <0.05), which shows that the observed variables are correlated witheach other in the whole Factor loading factor is an indicator to ensure practicalsignificance of EFA ≥ 0.5.

Based on the above criteria, research and select the appropriate variables for theconditions: KMO test coefficient: 0.5≤ KMO ≤ 1

Bartlett's Test (Bartlett's Test is statistically significant (sig ≤ 0.05), proving that theobserved variables are correlated with each other in the population

Extracting variance> 50%: For conventional research models on a small sample (<100), it

is necessary to apply this standard at the level of 0.5, but for the investigation results to

be practical and research still use this tight test

Next, the correlation between the independent variables is also checked to ensure thatthere is no strong correlation when the regression coefficient is not too large (> 0.8) andfinally the sub-regressions as well as the coefficients VIF has been tested and confirmedthat there is no multicollinearity phenomenon in the regression Details of the multi-collinearity test results are presented in Appendix 8 of the thesis

2.5.3 Explore factor analysis

This section will test the value of the variable to evaluate the convergence of scales ofeach factor by the method of exploratory factor analysis (EFA) In this process, theobserved variables that do not meet the requirements of the load factor (> 0.5) will berejected After screening the observed variables by EFA as above, we will evaluate thereliability of the scale based on the required Cronbach's alpha coefficient> 0.6 according

to Hair et al (2006) as well as the scale will reach required when the total varianceextracted> 50% Through the use of SPSS software version 22, the results of EFAanalysis for each factor in the research model are as follows:

The results of exploratory factor analysis (EFA) for all the above research concepts showthat many observed variables have been used and tested for quality in previous studies in

national contexts Other or other fields or disciplines were not relevant to the highereducation context in Vietnam.

Model suitability level and model appropriateness indicators

The suitability of the model is determined by the correspondence between the observedcovariance matrix and an estimated proposed model of covariance matrix

The SEM model uses metrics to describe how research hypotheses interpret input data.The input data forms the matrix of covariance between the measurement variables

Table 2 Conformity testing standards of the model

Chi-square / df (cmin / df) <3 is good, <5 is sometimes acceptable

Source: Hair et al (2010)

If all the main correlations that are inherent in the data set can be explained (related to thevariables in the model), there will be a good fit If there is a discrepancy between thesuggested correlations and the observed correlations, then there will be a poor fit model.The proposed model is not "consistent" with the observed model, the estimated model

- Indicators for measuring the suitability of the model

The goodness of fit metrics is listed below, along with threshold acceptable values.Appropriateness of the inverse model with sample size and number of observed variables

in the model The rating thresholds are shown in the table below (Hair et al., 2010)

The built models may or may not ensure the above conditions, in case the modelconformity indicators are not good, the study uses the MI index to improve the suitability

of the model

-How to improve the suitability of the model

Modification indices (MI): provides corrective measures for discrepancies between theproposed model and the estimated model In factor analysis, CFA cannot add regressionlines to correct the model, because all regression lines between latent and observedvariables are available Therefore, in the CFA, revised indices for covariance areconsidered The principle is: do not use covariance’s arrows between error terms andobserved or latent variables; cannot be used between error terms that do not belong to thesame factor Therefore, the most suitable way to use the MI index is to connect the two-way covariance arrows between the remainders in the same factor

The smaller the Chi-square, the better The MI column suggests which two-input errorpairs should be hooked to in order to improve the Chi-square Then GFI, TLI, CFI willalso be improved The study will select cases where there is a large MI to prioritize hookfirst Then, run the model again, and see which two errors should continue to improve….This section will test the value of the variable to evaluate the convergence of scales ofeach factor by the method of exploratory factor analysis (EFA) In this process, theobserved variables that do not meet the requirements of the load factor (> 0.5) will berejected After screening the observed variables by EFA as above, we will evaluate thereliability of the scale based on the required Cronbach's alpha coefficient> 0.6 according

to Hair et al (2006) as well as the scale will reach required when the total varianceextracted> 50% Through the use of SPSS software version 18, the results of EFAanalysis for each factor in the research model are as follows:

Table 3 The results of factor analysis and the reliability of the scale with the performancevariables of the university

9international Magazin

0.477Scientific research 3 - The number oftextbooks and specialized books

0.575KQGD4- Students find jobs suitable for majorsKQGD5- Lecturers always exceed teachinghours

0.510

0.670during the school year

0.638Financial

results

KQGD6- Students are interested in themodules in the training program

0.543

0.797

TC1- Sources of income from student tuitionfees are still available

0.535

0.633ensuring stability and growth in the past years 0.40

8KMO = 0.781> 0.5 and Bartlett's Test is statistically significant (p <0.005)

Source: Survey results of the author

According to the results in Table 3.4 and Table 3.5 on the results of the discovery factoranalysis of leadership styles and performance in universities above, we can see that theEFA results of the load coefficients of each observed variable in each Correspondingfactors are> 0.5 and Cronbach's alpha coefficients are greater than 0.6) In addition, thetotal variable correlation coefficients are all greater than 0.3, meeting the requirements ofthe scale quality

The results of exploratory factor analysis (EFA) for all the above research concepts showthat in fact many observed variables have been used and tested in previous studies innational contexts, fields, other industries There are also some studies in universities, butnot suitable for the current context of higher education innovation in Vietnam

CHAPTER IV: RESEARCH RESULTS

4.1 Research background

The author presents an overview of the current Vietnamese higher education context,thereby helping to study the nature of leadership styles in universities and testing therelationship between the factors in this research

Vietnam's higher education context

During the process of human development, especially in the recent period when theknowledge economy is the choice of most countries, education plays an important role,the driving force behind the economy economic development Many studies have long

shown that human capital plays an important role in economic growth and education isthe basic way to accumulate human capital The role of Vietnam's higher education alsoplays a great role in contributing to the country's development Therefore, Vietnam'shigher education is also very interested by the government, in 2012, the NationalAssembly of the Socialist Republic of Vietnam passed the Law on Higher Education toregulate the organization, duties and rights terms of higher education institutions,training activities, science - technology and other activities as well as regulations onlecturers, learners and the state management of higher education Accordingly, thegeneral goal of higher education is to research science and technology to create newknowledge, new products to serve the requirements of socio-economic development,ensuring national defense and security In addition, higher education helps train learnerswith political qualities, ethics and knowledge, professional practice skills as well asprofessional responsibility and a sense of serving the people (National Assembly, 2012) .Thus, with the above general objective, it can be seen that higher education institutions ingeneral, or universities in particular will have two main activities: (1) Scientific research

to create new knowledge and (2) Teaching to transfer knowledge to learners In addition

to these two main activities, the university may have other activities such as providingservices for learners and society

Mechanism of university management

For Vietnam's higher education, the classification of universities depends on the form ofownership (public and non-public), so the management mechanism of the universities hascorresponding differences Thus, in terms of property ownership, our country divides theuniversities into: (i) Group of public schools, state-owned, invested by the State, andbuilding facilities; (ii) The group of universities with foreign investment includes: (a)Higher education institutions with 100% foreign capital; (b) A joint venture between aforeign investor and a domestic investor; (iii) A group of private schools, owned by socialorganizations, socio-professional organizations, private economic organizations orindividuals, organized by social organizations, socio-professional organizations and

organizations private economy or individuals investing, building facilities, with privateownership According to the current management mechanism of non-business units,depending on the degree of financial autonomy, the state also decentralizes and givesautonomy in: finance, personnel organization and respective expertise.

About the mechanism of action

For public universities in Vietnam, the Ministry of Education and Training manages allprofessional activities related to: enrollment, training, testing and education qualityassurance Line ministries and provinces and cities manage the schools in terms oforganization, personnel and other aspects of work Besides scientific research activitiesrelated to the specialized management of the Ministry of Science and Technology;activities related to finance of the Ministry of Finance In terms of organization, atpresent, public universities are operating under the management regulations andorganization of public non-business units directly under the ministries and localities and

in compliance with legal documents relate to

Reform Vietnamese higher education towards autonomy

In 2003, the University Charter, issued in the Prime Minister's Decision No 153/2003 /QD-TTg affirmed that “universities have autonomy and self-responsibility in accordancewith the law on school development planning, plan, organization of training activities,science, technology, finance, international relations, organization and personnel ” In July

2005, in Article 14, the Education Law mentioned the implementation of assignment anddecentralization of education management, enhancing the autonomy and self-responsibility of educational institutions Resolution No 14/2005 / NQ-CP of theGovernment dated November 2, 2005 on the fundamental and comprehensive innovation

of higher education in Vietnam in the 2006-2020 period also affirms the importance ofimprovement policies to develop higher education towards ensuring the autonomy andsocial responsibility of higher education institutions, ensuring the State's managementand the role of social supervision and evaluation for higher education Accordingly, the

reform of the management mechanism needs to shift the public higher educationinstitutions to operate under the mechanism of autonomy, full legal status, the right todecide and take responsibility for training, research, organization,

With the content of the resolution providing autonomy for public higher educationinstitutions when committed to ensuring recurrent costs and investment costs as above up

to now, the TTCP has approved for 23 institutions higher education institutions (22institutions under ministries, central agencies; 01 local institution) and 3 vocationaleducation institutions (vocational education) (2 institutions under ministries, centralagencies; 01 local institution ) pilot reform of operating mechanism under Resolution 77(see list in Appendix 5 attached)

4.2 Descriptive statistics of research samples

After sending emails, and questionnaires directly to the schools The author obtained thenumber of responses from about 127 schools (accounting for 54.04%) compared to thetotal of 235 universities nationwide (list in Appendix 6 of the thesis)

Table 4 Results of survey sample collection

Object Forms sent Universities in

Vietnam

Number ofresponses

Total 235 127 54,04%

Source: author's survey results

Table 5 Statistical data by survey object

Workplace

Other

From 6 - 10 years

Source: Survey results of the author

Table 6 Statistics data by university classification

Type of organization

In addition to the public

Less than 100 instructors 969 49.9

Source: Survey results of the author

According to the statistical results, the collected data showed that they basically met therequirements of the original study design The collected research sample is equallydistributed among the following subjects: administrators, departmental managers,teachers, and

On the basis of data collection of individuals in this research, the analysis of data will beaccording to the analytical unit being the organization and is done through synthesizingthat individual opinion after testing and evaluating the quality number of personal survey

to ensure that the results are not misleading when analyzing

4.3 Results of testing the distribution format of the scales

Through the data collected as above, it shows that: the minimum and maximum values ofthe scales are in the range 1 to 5, this implies that there is no limit on the scale variationused These values fluctuate around an equilibrium value of 3.9; standard deviation valueless than 1.0 Finally, the two absolute values of inclination (Skewness) and Kurtosis arewithin the corresponding allowable limits within +/- 2 These results confirm that thescale has a normal distribution, ensuring that Inspection and analysis requirements in thefollowing sections

Table 7: Statistical description of the study sample

nhỏnhất

Giá trịlớnnhất

Giá trịtrungbình

Độ lệchchuẩn