Unit I Ess

Interdisciplinary Description of Complex Systems 20(4), 349-374, 2022

*Corresponding author, : [email protected]; 098-562179; *Pavlinska 2, 42000 Varaždin, Croatia

PATH-GOAL THEORY – LEADERSHIP STYLES AND THEIR CHANGES DURING

THE COVID-19 PANDEMIC

Robert Fabac1, *, Karolina Kokot1 and Irena Bubalo2

1University of Zagreb, Faculty of Organization and Informatics 1Varaždin, Croatia

2University North 2Koprivnica, Croatia

DOI: 10.7906/indecs.20.4.4 Regular article

Received: 22 February 2022. Accepted: 24 May 2022.

ABSTRACT

The most significant leadership theory, within the frame of contingency and situational approaches, is

the path-goal theory. This theory is an extension of Vroom’s expectancy theory and defines four

styles of leadership: directive, supportive, participative, and achievement-oriented. The theory was

first defined in the ’70s, and since then, numerous studies have been conducted on leaders and

subordinates in different industries and countries. Research during the COVID-19 pandemic is

underrepresented, especially in Croatia, and the authors identified a research gap. Primary quantitative

research was conducted, and data were collected through the questionnaire developed by Northouse

(A Path-Goal Theory Investigation of Superior Subordinate Relationships). The research sample

encompassed 77 leaders of organizations in the private and public sectors in the Republic of Croatia,

and the researchers analysed self-assessment results about leadership styles before and during the

COVID-19 pandemic. Data were analysed to obtain insight into the frequency of leadership styles,

differences in leadership styles before and during the COVID-19 pandemic, and the impact of gender,

level of education, and organization size on leadership style. The results indicate that self-assessment

defines minor changes to leadership style preference before and during the pandemic. The results

indicate a difference in leadership styles according to gender; women more frequently choose the

supportive style of leadership. Furthermore, it emerges that the level of education and the organization

size have an impact on other leadership styles in the frame of the path-goal theory.

KEY WORDS

leadership, directive style, supportive style, participative style, achievement-oriented style,

COVID-19

CLASSIFICATION

APA: 3640, 3660

JEL: D83, M12

R. Fabac, K. Kokot and I. Bubalo

350

INTRODUCTION

According to Northouse’s [1] definition, leadership is a process in which an individual has an

impact on a group with the purpose of achieving a common aim. The path-goal theory is one

of the most significant approaches to leadership in the frame of contingency theories.

Contingency and situational theories are valuable for recommending what leaders must or

must not do in defined situations. Northouse [1] claims that this approach is useful in

numerous situations and highlights the need for adaptability of the leader. The path-goal

theory is focused on the behaviour style of leaders and consideration of the situation and

defines the regulations and recommendations that enable achieving group aims [2].

The path-goal theory was established by Evans [3], analysing a leader’s behaviour choices in

relation to their behaviour style, employees’ needs and conditions in the work environment.

According to this perspective, the leader is not an individual who uses their power but has a

role in support, and this approach is frequently defined as servant leadership. The most

significant advancement in path-goal theory was provided by House [4], who examined

situation elements within which leaders search for solutions to motivate subordinates. The

leadership styles that leaders can implement are directive, participative, supportive, and

achievement oriented. At the same time, the work environment has components such as the

characteristics of a workgroup, task structure, and formal authority system. Subordinates have

characteristics, abilities, and needs. The path-goal theory is based on the motivation theory

developed by Vroom [5] and named expectancy theory. Since Vroom theory enables the

identification of important things that can be done for the motivation of employees, it has

important implications from managers’ and leaders’ standpoints [6]. According to this theory,

subordinates will be motivated in the following situations: If they think that they are able to

manage the task, if they believe that their efforts will result in certain success, and if they

believe that compensation is in line with their effort [5].

Four types of leader behaviour according to the path-goal theory can be described as follows:

1) Supportive leadership includes a concern for the welfare of subordinates and the

organizational climate. This type of behaviour tries to achieve a friendly environment [2].

Leaders attempt to meet the requirements, needs, and preferences of subordinates through

the supportive style [7]. Subordinates are rewarded in the frame of this leadership style,

and their motivation is encouraged. This leadership style accomplishes great results when

subordinates are unsatisfied or frustrated.

2) Participative leadership enables subordinates to influence the decisions of leaders. Leaders

encourage subordinates on critical considerations and suggestions, and leaders consult

with them before decision-making [7].

3) Directive leadership includes specific guidelines for subordinates and explanations about

the expectations for employees. Directive leadership attempts to reduce role ambiguity

among subordinates and to clarify their perceptions, which is positive in situations when

tasks are unclear and problems are unstructured. When the requirements of subordinates

are clear, this type of leader’s guidance can have the opposite effect. 4) Achievement-

oriented leadership is defining challenging aims, searching for higher quality results,

followed by the trust that subordinates will achieve the aims [2].

Adequacy of implementation of leadership styles is in a relationship with the situation.

Leaders can change their style in line to best serve subordinates, and the same leader can

simultaneously implement different styles; however, great leaders need to know when to

apply which leadership style [8-10].

Path-goal theory – leadership styles and their changes during covid-19 pandemic

351

Supportive leadership is the most appropriate for subordinates who have the need for

belonging, while directive leadership is a good choice for a group of subordinates who tend to

need structure. For subordinates who have a need for control, participative leadership is

appropriate because it provides them with satisfaction in decision-making. Achievement-

oriented and directive leadership are appropriate when subordinates are focused on the

possibility of solving the task. Those kinds of subordinates require directive leadership when

they think that they have low capabilities, and with higher insight into their capabilities, these

subordinates have a lesser need for the directive style. The aim of this leadership style is to

boost subordinates’ self-confidence. Numerous authors refer to the abovementioned styles of

leadership, but Euwema et al. [11] point out directive and supportive behaviour because in

previous research, these concepts have been successfully applied for years.

Related to subordinates’ and leaders’ behaviour, subordinates are allocated to a given type of task,

which means that the type of task defines the leadership style. Northouse [12] states that directive

leadership is motivating when tasks are complex and rules are unclear. Supportive leadership is

motivating when tasks are repetitive. Unclear and unstructured tasks have a need for participative

leadership, and challenges and complex tasks require achievement-oriented leadership.

According to the path-goal theory, the right and successful work of subordinates is possible

when leaders clearly define tasks to them [12 p.500] and when leaders have the possibility to

provide the required education for this work. In addition, leaders must help subordinates in

work performance and provide a reward for successfully performing work.

Figure 1. Factors that have an impact on path-goal [12].

The main features of the path-goal theory are clearly defined aims, providing support in

performing tasks and rewards according to successful achievement of the aims. The approach

to subordinates is relatively personalized because leaders have to define the approach that

will be used for each individual and rewards for subordinates. Leaders must provide

subordinates with conviction about their ability to meet expectations. Figure 1 shows the

factors that impact subordinate motivation to achieve the goal.

LITERATURE REVIEW

In the field of path-goal theory, numerous articles have significantly contributed to this

theory. Ever since the establishment of this theory in the ‘70s, a significant number of studies

indicated empirical results in line with theory assumptions, although some results indicated

opposite results, which can be explained in applied research methodology [7, 14].

R. Fabac, K. Kokot and I. Bubalo

352

Farhan [15] analysed the application of path-goal leadership theory in a learning organization.

With the increasing importance of information and knowledge in the process of creating value,

numerous leaders revert to creating learning organizations, in which the majority of employees

are involved in the identification and solving of unstructured problems [16]. Farhan [15] states

that leadership style is the key support tool for creating learning organizations, especially the

participative style, which can be described as the architect of organizational learning. Sarti [17],

in their research conducted on Italian human service organizations, validated a hypothesis about

the positive relationship between work engagement (dedication and vigour) and

implementation of the participative style. Additionally, in this article, Sarti [17] analysed the

relationship between work engagement and directive leadership style, and the relationship

involved vigour and absorption but not dedication. Another research article [18] defines the

directive style as promoting task-oriented behaviour and a tendency to control discussions and

dominate interactions. Bell et al. [19] researched the impact of the directive and participative

style on organizational culture in university administrative departments in South Africa. The

results indicated a positive and significant impact of both styles on organizational culture,

which was measured by four cultural facets of involvement, consistency, adaptability, and

mission. Furthermore, the relationship between directive and participative leadership styles and

the performance of Israeli school staff teams were analysed in an article [20]. Somech [20]

states that the directive leadership style has value, especially for leaders (teachers) who operate

in an ambiguous work environment. Although participative leadership style is preferred in

modern educational policies, that style is not always related to improvement. Depoo &

Shanmuganathan [21] researched a multinational organization in the US and identified the

relationship between brand management and leadership style within the frame of path-goal

theory, and the most representative approaches are a coaching style (which contains the

supporting style) and the directive style.

Supportive leadership style is a characteristic behaviour within the frame of path-goal theory

and is indicated in most research articles. In the process of organizational changes and

learning, it is important that leaders implement a supportive leadership style to manage

subordinate anxiety [22]. Furthermore, House [7] indicated that causes of stress that can have

an influence on employees dictate the implementation of a supportive leadership style. Khalid

et al. [23], who analysed leadership in educational institutions in Pakistan, indicated that

“when employees serving in any organization receive support from their leaders, the level of

stress decreases and job performance is improved”.

La Rocco and Jones [24] researched the US Navy and indicated that supportive leadership

contributes to both job satisfaction and the level of satisfaction with the organization, which was

described in the book by Bass and Bass [25]. The relationship between the supportive leadership

style and interpersonal helping behaviours was analysed by Rafferty and Griffin [26], under the

guiding question “if it is supportive leadership style truly transformational?” Namely, supportive

leadership is focused on the satisfaction of subordinates’ wishes, but this can result in the absence

of organizational performance [7, 26]. Path-goal theory was analysed in the telecom industry of

Pakistan [27]. An analysis of the relationship between all four leadership styles and job

satisfaction in telecom companies in Chad was conducted by Dokony et al. [28].

Research in the field of leadership for several decades has focused on transformational leadership,

so one-half of the articles are about transformational leadership [29, 30]. Consequently, the study of

the path-goal model is not the focus of scientists; this is also the case in Croatia. Udovičić et al. [31]

analysed leadership style based on the leadership model developed and established by R. Likert in

Croatian entrepreneurship. Furthermore, Miloloža [32] researched the presence of autocratic,

democratic, and laissez-faire leadership styles and the relationship between leadership style and the

size, growth phase, and international orientation of the organization.

Path-goal theory – leadership styles and their changes during covid-19 pandemic

353

The impact of gender was researched by Van Emmerik et al. [33], who defined the

hypothesis “female managers score higher on consideration than male managers”, which

includes elements of classical leadership styles: consideration and initiating structure.

Consideration refers to friendly and interpersonally supportive supervisory behaviour [34],

while initiating structure refers to task-oriented and directive behaviour [34]. Van Emmerik et

al. [33] conducted research on a database with 64 000 subordinates from GLOBE societal

clusters. GLOBE is a project that addresses the research of practice and values in sectors of

industry, organizations, and society [35]. GLOBE indicated ten cultural clusters that involve

61 nations [36]. GLOBE researchers measure culture and try to define its relationship to the

effectiveness of leadership. Research results indicated a high mean of measures of

consideration (supportive style) among subordinates who have female managers in Nordic

countries in the frame of Globe Clusters. In line with expectations, research indicated female

leaders achieved better results on measures of consideration style than male leaders. One

insight that is outstanding is that female leaders also have a higher level of measures of

initiating structure (directive behaviour) than male managers. A key finding of the analysis is

“that female and male managers do not differ much in leadership behaviours”, which is

related to existing stereotypes regarding gender [33]. Research results conducted by Kent et

al. [37] indicated that men and women lead using the same behaviours.

Polston-Murdoch [10] researched the relationship between leadership style and commitment

of subordinates according to leaders, relative to leaders’ gender. The results indicated no

significant evidence that this relationship is dependent on leaders’ gender. Previous research

indicated that “women may be viewed as more supportive and affective with characteristics

involving the management of emotions” [38] as stated by Polston-Murdoch [10].

Thacker [39] states that expectations for female leaders are characterized by participative

leadership style and behaviours focused on relationships. Lord et al. [40], according to an

analysis of the literature, state that females are proven to be more successful leaders in the

task of developing supportive relationships with subordinates. Becker et al. [41] state “that

the female leaders work in nontraditional roles”. In recent times, the number of articles that

are focused on female leadership has increased, and especially prominent examples include

the book by Sally Helgesen [42] and the publication of Chamorro-Premuzik [43] on the topic

of women’s leadership competency gap. Chamorro-Premuzik [43] argued why male leaders

are more often chosen in practice and states that “leaderless groups have a natural tendency to

elect self-centred, overconfident and narcissistic individuals as leaders and that these

personality characteristics are not equally common in men and women”.

THE COVID-19 PANDEMIC AND LEADERSHIP

In December 2019, in the city of Wuhan, China, the appearance of a disease caused by a

coronavirus was reported, which quickly spread to the whole country and later to the world. In

February 2020, the meeting of the EU Health Council and Ministers of Health defined

conclusions about COVID-19. At the end of February, the impact of the disease COVID-19 on

the industry of the European Union was considered by the European Commission [44]. Later,

measures defined in all countries had an impact on people’s behaviour, especially in

organizations, by coordinating their movement, attendance in the workspace, and contacts, as

leaders in organizations and their subordinates had to adapt to new factors in the environment.

One of the rare articles that analyses the path-goal theory in the COVID-19 pandemic

situation is Mercedes & Burrell [45]. Their questionnaire (adapted by [46]) addresses

assessments of the statement “I create a clear path and vision concerning problems and

solutions,” which is related to ideas of the path-goal theory.

R. Fabac, K. Kokot and I. Bubalo

354

The majority of articles related to the COVID-19 environment are not structured within the

framework of leadership theory. Leader behaviours in COVID-19 pandemic time, especially

in the health system in England, were analysed in a article [47], where academics concluded

that “leaders who have excelled are those who have shown compassion, empathy,

connectedness and even vulnerability”. Furthermore, leadership in school in the COVID-19

pandemic was researched by Harris and Jones [48], where the authors state that “leading in

disruptive times means being able to navigate a different course, to create new pathways

through the disruption”. Research conducted by Lee [49] in the United Kingdom analysed the

need to create an environment where leaders support subordinates. Antonakis [50] claims that

they trust that scientists will recognize the need to study leadership because it is an effective

tool that has to be used in the fight against the viral pandemic.

Research that appreciates and refers to the leadership theory frame includes a article by Claus [51],

in which she defines a new approach to leadership research, New Leadership Practices and Skills Due

to COVID-19, with three dimensions: leading organizations in turbulent times, leading a distributed

workforce (individuals and teams), and being a resilient leader. All of these dimensions [51] elaborate

expectations from leadership. Watkins & Clevenger [52] analysed politicians in the US through the

prism of two dimensions, context and control, and stated that the framework has the aim of

demonstrating how leadership style affects crisis behaviour and communication. Behaviour in crisis

leadership was described by Boin et al. [53]. Eichenauer et al. [54] researched leadership behaviour

during COVID-19 through task assessment, activities (planning, clarifying, supporting, and

empowering), and differences between female and male leaders. Furthermore, academics

researched stereotypes that females lead to more communal behaviour and males lead to more

agentic behaviour [55] and indicated a small difference in behaviour.

Chen and Sriphon [56] researched leadership in organizations in Asia according to the

relationship between the following constructs: (1) trust, (2) communal relationships, and (3)

social exchange relationships and characteristic factors of COVID-19. The most important

results are that remote work interrupted trust and relationships between leaders and

employees during pandemics.

RESEARCH GAPS AND RESEARCH ISSUES

Leadership research in the COVID-19 pandemic is present and growing, although the number

of articles focusing on path-goal theory is not particularly exceptional; therefore, it is worth

covering this area, especially in regard to the situation in Croatia. Furthermore, research on

leadership styles of Croatian managers is relatively underrepresented. Among the potentially

interesting issues here is the segment related to women in the role of leaders. Guided by these

observed gaps, we formulated the following research questions:

RQ1. Under pandemic circumstances, did leaders in Croatian organizations change their

leadership style?

RQ2. Do leaders in Croatian organizations have a difference in leadership styles based on

gender?

RQ3. Is the dominant leadership style subject to the influence of other factors, such as the

leader’s age and level of education and the organization’s size?

METHOD AND RESEARCH

We conducted a survey of leadership practice in the Republic of Croatia through the PGT

model at the end of 2021, with the aim of addressing research issues. It should be noted that in

the middle of the last year, there were approximately 170 000 active business entities in the

country. The corresponding questionnaire was sent by e-mail to approximately 1950 randomly

Path-goal theory – leadership styles and their changes during covid-19 pandemic

355

selected addresses of experts and managers, and we had no knowledge of the positions of

individuals in their organizations. Therefore, it was suggested that the answer to the

questionnaire be provided only by managers or leaders in their companies. The questionnaire,

is presented in the Appendix section, is taken from the author Northouse [1, pp.145-146] and

includes 20 self-assessment questions regarding the frequency of application of characteristic

leadership styles, according to R. House theory [4].

We received 77 valid responses, of which seven respondents failed to give an overview of

leadership practices before the COVID-19 pandemic. Since we did not have information

about the positions of individuals in organizations when sending the questionnaire, the

questionnaire also received a large number of experts who are not in leadership positions.

They were not expected to send a response. Therefore, the real response rate of leaders is

significantly higher than the 77/1950 ratio. The basic demographics of the sample relevant to

the research questions is presented in Table 1.

Table 1. Leadership questionnaire – sample structure. Characteristics Number Percentage, %

Gender

Male 44 57,1

Female 33 42,9 Leader’s education

Up to Bachelor’s degrees 21 27,3

Bachelor’s, master’s, or doctoral degrees 56 72,7

Organizational size

Micro and small enterprises 40 51,9

Medium and large organisations 37 48,1

Sector

Public or state 11 14,3

Enterprises 66 85,7

RESEARCH INSTRUMENT

For the purposes of this research, a questionnaire of the PGT model by Northouse [1, pp.145-146]

was used, which is based on the self-assessment of the managers and is presented in the Appendix.

The author adapted the questionnaire according to the works of Indvik [56, 57]. The questionnaire

contains twenty statements relevant to leadership styles to which leaders responded by selecting

numbers in the range 1-7, where one indicates disagreement with the statement and seven, full

agreement. In concluding the frequency of application of a particular characteristic style, according

to the PGT, the following rules are followed [1]: first reverse the scores for Items 7, 11, 16, and 18;

then calculate for the directive style assessment: sum the scores on Items 1, 5, 9, 14, and 18; for

supportive style assessment: sum the scores on Items 2, 8, 11, 15, and 20; for participative style

assessment: sum the scores on Items 3, 4, 7, 12, and 17; for achievement-oriented style assessment:

sum the scores on Items 6, 10, 13, 16, and 19. According to suggestions [1], the assessment of

achievement (high, common, low) is presented in Table 2.

Table 2. Classifying achievements by styles.

Considered high Common score Considered low

Directive style above 28 23 below 18

Supportive style above 33 28 below 23

Participative style above 26 21 below 16

Achievement-oriented style above 24 19 below 14

R. Fabac, K. Kokot and I. Bubalo

356

METHODS

The collected data were analysed using statistical methods, IBM SPSS FP v28 and JASP v0.16.

Principal component analysis was applied to determine the components that describe and factorize

the respondents’ answers. Then, the t test was applied to the variants paired samples test and

independent samples test. The paired samples test was used to determine whether there was a

change in the frequency between the prepandemic and COVID-19 pandemic cases. In investigating

the potential significant differences between groups (Table 1) regarding the frequency of choosing a

particular leadership style, the independent samples t test was applied. Since the assumption of

normal distribution was not met when testing some groups in the sample, the Mann–Whitney U test

was also used to determine significant differences between groups in arrangement with independent

groups. As an alternative to the t test, in the absence of conditions for distribution normality, the

nonparametric Wilcoxon signed ranks test method was applied. This test was used in the analysis of

data for related dependent groups, i.e., for the purpose of determining differences in the frequency

of leadership styles in two time periods among the same respondents.

RELIABILITY AND VALIDITY ANALYSIS

Respondents’ answers to 2 questionnaires (prepandemic leadership status and leadership

styles during the COVID-19 pandemic) were tested using the Alpha-Cronbach method. These

include 70 answers to the first questionnaire of 20 questions and 77 answers to the second

questionnaire. It turns out that this factor indicating consistency for the overall set of

responses results in a Cronbach’s alpha = 0,92 (Table 3), which can be considered an

excellent indicator of reliability, according to, for example, Cortina [58].

Table 3. Reliability Statistics.

Reliability Statistics

Cronbach’s Alpha Cronbach’s Alpha Based on Standardized Items N of Items

,924 925 40

We also checked the internal consistency among the respondents’ answers which focused on

leadership during the COVID-19 pandemic (77 of them), and the result as a factor indicating

consistency was Cronbach’s alpha = 0,84. This can be considered a very good fulfilment of

the conditions of reliability.

To determine the adequacy of the data, we also applied a validity analysis through the

Kaiser–Meyer–Olkin (KMO) test. The idea behind this check is to gain insight into the

adequacy of the sample data before performing the factor analysis procedure. The preferred

value of the KMO test is 0.5 or slightly more [59], and Bartlett’s aspherical value should have

significance. We see that these conditions are met since the KMO factor is 0,711 (Table 4).

This score suggests that there are sufficient items for each factor.

Table 4. Sampling adequacy – KMO test.

KMO and Bartlett’s Test

Kaiser–Meyer–Olkin Measure of Sampling Adequacy 0,711

Bartlett’s Test of Sphericity

Approx. Chi-Square 660,070

df 190

Sig. < 0,001

For the results of the leadership questionnaire during the COVID-19 pandemic, we performed

an analysis using the SPSS tool, applying principal component analysis. In doing so, we

preferred to identify four factors, expecting the agreement of the individual components

Path-goal theory – leadership styles and their changes during covid-19 pandemic

357

(Table 5) with the characteristic items of individual leadership styles. In addition, the rotation

method varimax with Kaiser normalization (rotation converged in 6 iterations) was applied.

Table 5. Principal Component Analysis.

Rotated Component Matrixa

Component

1 2 3 4

Q1 0,049 0,384 0,656* 0,049

Q2 0,522* -0,456 0,254 0,068

Q3 0,802* -0,089 0,065 -0,035

Q4 0,708* 0,110 0,235 0,011

Q5 0,423 0,072 0,600* 0,007

Q6 0,038 0,722* 0,385 -0,009

Q7 0,493* 0,106 -0,339 0,352

Q8 0,447 0,160 0,192 0,099

Q9 0,152 0,141 0,797* 0,116

Q10 0,152 0,677* 0,228 -0,282

Q11 0,030 -0,111 0,146 0,814*

Q12 0,751* 0,329 -0,068 -0,102

Q13 0,573* 0,226 0,334 -0,083

Q14 0,358 0,616* 0,242 0,047

Q15 0,607* 0,093 0,341 0,067

Q16 -0,002 0,033 0,048 0,746*

Q17 0,701* 0,342 -0,010 -0,183

Q18 0,021 0,019 -0,027 0,689*

Q19 0,159 0,840* -0,004 0,157

Q20 0,709* -0,025 0,019 0,173

The obtained results can be interpreted in such a way that if we set the threshold of the

importance of impact/weight to approximately 0.5 (allowing 0.49; marked with an asterisk *),

then we determine the state of components in terms of leadership styles, taking into account

the description of the research instrument by style:

• component 1 – participative style (questions 3, 4, 7, 12, and 17) and supportive style

(questions 2, 8, 11, 15, and 20) explain 8 of the 9 main contributions of component 1,

which enter with weights ≥ 0, 5; (listed questions marked bold),

• component 2 – achievement-oriented style (questions 6, 10, 13, 16, and 19) explains 3 of

the 4 main contributions of component 2, which enter with weights ≥ 0.5; (listed questions

marked bold),

R. Fabac, K. Kokot and I. Bubalo

358

• component 3 – directive style (questions 1, 5, 9, 14, and 18) explains 3 of the 3 main

contributions of component 3, which enter with weights ≥ 0.5; (listed questions marked

bold),

• component 4 – inverted answers (questions 7, 11, 16, and 18) explain 3 of the 3 main

contributions of component 4, which enter with weights ≥ 0.5; (listed questions marked

bold).

Thus, the questionnaire distinguishes directive style and achievement-oriented style, while

supporting and participative leadership styles place in the same factor (in component 1). What

obviously had an impact on the respondents was the appearance of inverted answers where high

agreement with the statement was converted into a small contribution to the corresponding

leadership style and vice versa. Namely, if a certain model of selection that includes a frequent

preference for higher grades (e.g., 6 and 7) suddenly asked for a grade that should be low (for

example, 1 or 2), it is possible that this created some confusion among respondents.

Overall, we can assume that the self-assessment questionnaire on the frequency of leadership

styles [1] contains relatively separate components that describe these styles.

RESULTS

THE DOMINANT STYLE IN THE REPERTOIRE OF LEADER BEHAVIOUR

Considering all the specific leadership styles that an individual leader can present in his or her

repertoire of behaviours over time, we have formulated Table 6, which considers three levels of

frequency of application of individual styles. It should be noted that the highest level regarding

direct leadership is achieved in as many as 66.2 % of cases, while for this style, there are no cases

of low performance. Results are similar to the style of achievement orientation, where as many as

74 % of respondents have a high level of achievement. The participative style has a highest level

in 41.5 % of cases. It should be emphasised again how the path–goal questionnaire provides

information about which style of leadership is applied how often [1, pp. 147].

Table 6. Frequency of leadership styles by level.

Directive style

Supportive style

Participative style

Achievement oriented style

Average score 29,662 29,338 25,610 26,714 High level

(No.) 51 9 32 57

Low level (No.) 0 2 0 0 High l. (%) 66,23 11,69 41,56 74,03

The supportive style stands out because only 11,7 % of respondents represent this style with

the highest level of frequency, and in addition, there are established results of very low

representation (in 2 cases), which was not noted in other styles. Therefore, this style has a

low frequency given the basic, pre-survey expectations.

THE IMPACT OF THE COVID-19 PANDEMIC ON LEADERSHIP STYLE

Respondents answered questions about their approach to leadership during the pandemic

(also when the questionnaire was implemented) and before the pandemic period. To

determine possible changes in preferences regarding leadership style, in this part of the

results, we analyse for each of the 20 questions — is there a statistically significant difference

between the answers of respondents? The t test method was therefore applied for the same

sample but with records for two time periods, meaning “the paired samples t test” (SPSS),

which compares the mean values of two measurements taken from the same respondent.

Path-goal theory – leadership styles and their changes during covid-19 pandemic

359

Since we are not only interested in strict statistical significance, we performed this test first

without checking the sample normality prerequisites (Table 7).

Table 7. Paired Samples Test (before the pandemic and during the pandemic). Paired Samples Test

Pair

Paired Differences

t df

Significance

Mea

n

Std.

Dev.

Std.

Error

Mean

95 %

Confidence One-

Sided

p

Two-

Sided

p Lower Upper

1 Q1 BCov - Q1 ,000 ,482 ,058 -,115 ,115 ,000 69 ,500 1,000

2 Q2 BCov - Q2 ,129 ,588 ,070 -,012 ,269 1,830 69 ,036 ,072

3 Q3 BCov - Q3 -,057 ,447 ,053 -,164 ,049 -1,070 69 ,144 ,288

4 Q4 BCov - Q4 -,014 ,399 ,048 -,109 ,081 -,300 69 ,383 ,765

5 Q5 BCov - Q5 -,071 ,393 ,047 -,165 ,022 -1,522 69 ,066 ,133

6 Q6 BCov - Q6 ,000 ,417 ,050 -,099 ,099 ,000 69 ,500 1,000

7 Q7 BCov - Q7 ,057 ,508 ,061 -,064 ,178 ,942 69 ,175 ,349

8 Q8 BCov - Q8 -,043 ,464 ,055 -,154 ,068 -,772 69 ,221 ,443

9 Q9 BCov - Q9 -,057 ,376 ,045 -,147 ,033 -1,270 69 ,104 ,208

10 Q10 BCov - Q10 ,086 ,474 ,057 -,027 ,199 1,514 69 ,067 ,135

11 Q11 BCov - Q11 ,086 ,474 ,057 -,027 ,199 1,514 69 ,067 ,135

12 Q12 BCov - Q12 -,071 ,428 ,051 -,173 ,031 -1,396 69 ,084 ,167

13 Q13 BCov - Q13 -,029 ,380 ,045 -,119 ,062 -,630 69 ,265 ,531

14 Q14 BCov - Q14 -,057 ,336 ,040 -,137 ,023 -1,425 69 ,079 ,159

15 Q15 BCov - Q15 ,029 ,339 ,041 -,052 ,109 ,705 69 ,242 ,483

16 Q16 BCov - Q16 ,014 ,399 ,048 -,081 ,109 ,300 69 ,383 ,765

17 Q17 BCov - Q17 ,014 ,434 ,052 -,089 ,118 ,276 69 ,392 ,784

18 Q18 BCov - Q18 -,071 ,393 ,047 -,165 ,022 -1,522 69 ,066 ,133

19 Q19 BCov - Q19 ,000 ,241 ,029 -,057 ,057 ,000 69 ,500 1,000

20 Q20 BCov - Q20 -,043 ,316 ,038 -,118 ,032 -1,136 69 ,130 ,260

In Table 7 and further in the article, BCov stands for “Before COVID-19“. Regarding stricter

conditions, if we take into account the threshold p < 0.1 with two 2-sided p, we obtain the

result that a significant difference exists only in question number 2. This second question

concerns maintaining friendly working relations with subordinates (Appendix). With the

criteria mitigated, considering one-sided p with significance (p < 0.1), i.e., looking at the

indication in which direction the changes are going, then the following results are obtained

for queries: Q5 (I inform subordinates about what needs to be done and how it needs to be

done), Q9 (I ask subordinates to follow standard rules and regulations.), Q10 (I set goals for

subordinates’ performance that are quite challenging), Q11 (I say things that hurt

subordinates’ personal feelings), Q12 (I ask for suggestions from subordinates concerning

how to carry out assignments), Q14 (I explain the level of performance that is expected of

subordinates), and, at the end, Q18 (I give vague explanations of what is expected of

subordinates on the job).

However, testing the normality for differences between pairs of scores on the same questions

where significant differences are inferred, we obtain a result that indicates that there is not a

normal distribution (Table 8).

R. Fabac, K. Kokot and I. Bubalo

360

Table 8. Normality-Paired Samples.

Tests of Normality

Kolmogorov–Smirnov* Shapiro–Wilk

Statistic df Sig. Statistic df Sig. Statistic

Q2 0,472 70 < 0,001 0,503 70 <0,001

Q5 0,486 70 < 0,001 0,445 70 <0,001

Q10 0,500 70 < 0,001 0,392 70 <0,001

Q11 0,515 70 < 0,001 0,295 70 <0,001

Q12 0,466 70 < 0,001 0,514 70 <0,001

Q14 0,482 70 < 0,001 0,448 70 <0,001

Q18 0,529 70 < 0,001 0,181 70 <0,001

*Lilliefors Significance Correction

Therefore, we cannot even accept the results from Table 7 for the t test, and consequently, we

consider the solution provided by the nonparametric test below in question Q2. or the case of

related samples it is possible to apply the Wilcoxon signed rank test (Table 9), which does

not require the condition of a normal distribution of differences. The obtained result shows a

statistically significant difference at the level of p < 0.1.

Table 9. Related samples – Q2 (PGT questionnaire) Wilcoxon Signed Rank Test.

Test Statisticsa

2. I maintain a friendly working relationship with subordinates. – 2. I

maintain a friendly working relationship with subordinates. (BCov)

Z ̶1,812b

Asymp. Sig. (2-tailed) 0,070

aWilcoxon Signed Rank Test bBased on positive ranks

The next thing that was done as part of the examination of the impact of the COVID-19 pandemic

on leadership style was to summarize the answers to specific groups of questions to determine the

results for specific leadership styles. Therefore, a calculation was made according to the instructions

from the questionnaire [1] for each individual leadership style, and a characteristic sum is relevant,

both at the time of the COVID-19 pandemic and before the pandemic period. Table 10 contains the

results for the average values of the four PGT leadership styles.

Table 10. Paired Samples – Typical PGT leadership styles before and during a pandemic.

Paired Samples Statistics

Mean N Std. Deviation Std. Error

Mean

Pair 1 Directive style 29,44 70 3,713 0,444

Directive style BCov 29,19 70 3,589 0,429

Pair 2 Supportive style 29,23 70 3,477 0,416

Supportive style BCov 29,39 70 3,427 0,410

Path-goal theory – leadership styles and their changes during covid-19 pandemic

361

Table 10. Paired Samples – Typical PGT leadership styles before and during a pandemic.

(continuation from p. 360)

Pair 3 Participative style 25,33 70 4,416 0,528

Participative style BCov 25,26 70 4,393 0,525

Pair 4 Achievement-oriented style 26,43 70 4,060 0,485

Achievement-oriented style BCov 26,69 70 3,794 0,453

Below, we analyse the overall results according to the preferred leadership styles because we

are interested in the difference in the situation before and after the pandemic. Using two tests

of normality, Kolmogorov–Smirnov and Shapiro–Wilk, we find differences in related pairs

belonging to certain styles (before the pandemic and at the time of the pandemic) that do not

meet the condition of normal distribution.

Therefore, we approached the application of nonparametric methods, specifically the

Wilcoxon Signed Rank Test, for two samples of connected subjects. The final result

indicates a statistically significant difference in the directive style, where the average

frequency of this style before the pandemic was higher than during the COVID-19

pandemic. We interpret the results in Table 11 through the rule that a positive z score

indicates a negative difference and vice versa.

Table 11. PGT leadership styles: Related Samples – Wilcoxon Signed Rank Test.

Test Statisticsa

Directive style

BCov – Directive style

Supportive style BCov – Supportive

style

Participative style BCov –

Participative style

Achievement-oriented style BCov –

Achievement-oriented style

Z –2,517b 1,093c –0,353b –1,659c

Asymp. Sig. (2-tailed)

0,012 0,274 0,724 0,097

aWilcoxon Signed Ranks Test bBased on positive ranks cBased on negative ranks

Using the nonparametric Wilcoxon Signed Rank Test, which is equivalent to the t test for

dependent samples, we confirm a statistically significant difference in the frequency change of

the directive style (with p < 0.05) and an indication of the change in the achievement-oriented

style frequency (with p < 0.1) within the answers to both questionnaires (70 of them).

THE DIFFERENCE IN LEADERSHIP STYLES WITH RESPECT TO GENDER

To research whether gender influences the outcomes regarding the choice of leadership

style, i.e., the frequency of certain styles according to PGT, a t test was applied to the

answers related to the period of work and leadership in the COVID-19 pandemic. Data on

the groups “male” and “female” are contained in Table 1, while in Table 12 there is a label

1 for the group of male respondents (44 of them) and a label 2 for the group of female

respondents (33 of them).

R. Fabac, K. Kokot and I. Bubalo

362

Table 12. The influence of gender on leadership styles – group statistics.

Group Statistics

Gender N Mean Std. Deviation Std. Error Mean

Directive style 1 44 29,20 3,508 ,529

2 33 30,27 3,891 ,677

Supportive style 1 44 28,70 3,638 ,548

2 33 30,18 3,066 ,534

Participative style 1 44 25,66 4,446 ,670

2 33 25,55 4,459 ,776

Achievement-oriented

style

1 44 26,36 3,792 ,572

2 33 27,18 4,482 ,780

The results of the t test method for independent samples are shown in Table 13. Given the

prevalence of the supportive style, there is a visible difference between the two groups with a

significance of p <0.1 regardless of the assumption: “Equal variances (not) assumed“. The

group of female respondents more often applies a supportive leadership style, at the level of

significance p <0.1.

Table 13. The influence of gender on leadership style: t test.

F Sig. t df

Significan

ce Mean

Differe

nce

Std.

Error

Differe

nce

95 %

Confidence

One-

Side

d p

Two-

Sided

p

Lower Upper

Directive

style

EV

assumed ,461 ,499 -1,262 75 ,105 ,211 -1,068 ,847 -2,755 ,618

EV not

assumed -1,243 64,943 ,109 ,218 -1,068 ,859 -2,785 ,648

Supportive

style

EV

assumed ,699 ,406 -1,883 75 ,032 ,064 -1,477 ,784 -3,040 ,085

EV not

assumed -1,930 73,920 ,029 ,057 -1,477 ,765 -3,002 ,048

Participativ

e style

EV

assumed ,043 ,836 ,111 75 ,456 ,912 ,114 1,025 -1,928 2,156

EV not

assumed ,111 68,974 ,456 ,912 ,114 1,025 -1,932 2,159

Achieveme

nt-oriented

style

EV

assumed ,925 ,339 -,866 75 ,195 ,389 -,818 ,944 -2,699 1,063

EV not

assumed -,846 62,226 ,200 ,401 -,818 ,967 -2,752 1,115

In Table 13, “EV“ means equal variances. For the application of the t test, we wanted the

tested and validated samples to meet the necessary prerequisites, such as independence,

normality, and approximately equal variance [61]. The normality of the distribution was

checked by the Shapiro–Wilk test (Table 14). Since the null hypothesis for the Shapiro–Wilk

test is that the variable is normally distributed in some populations and that the null

hypothesis is rejected if p <0.05, we note here that the normality condition for the directive

Path-goal theory – leadership styles and their changes during covid-19 pandemic

363

style distribution is not met. However, for the supportive style, the condition of the normal

distribution of the variable is satisfied.

Table 14. The influence of gender on leadership style – test of normality (Shapiro–Wilk).

Test of Normality (Shapiro–Wilk)

GG G W p

Directive style 1 0.949 0.051 2 0.905 0.007

Supportive style 1 0.971 0.337 2 0.957 0.219

Participative style 1 0.971 0.328 2 0.960 0.266

Achievement-oriented style 1 0.982 0.716 2 0.955 0.189

*Significant results suggest a deviation from normality

Thus, looking at genders, we obtained a statistically significant difference p <0.1 in favour of

females for the frequency of applying a supportive leadership style. No statistically

significant difference was found between groups of respondents formed by gender for other

leadership styles.

INFLUENCE OF OTHER FACTORS ON PARTICULAR LEADERSHIP STYLES –

THE SIZE OF THE ORGANIZATION AND THE DEGREE OF LEADER

EDUCATION

Investigating the impact of environmental factors related to the size of the organization, we

identified two independent groups of respondents, 40 from the category “Micro and small

enterprises“ (label 1 in Table 15) and 37 from the category “Medium and large organizations“

(label 2 in Table 15).

Table 15. The influence of organizational size on leadership style – group statistics.

Group Statistics

Size N Mean Std. Deviation Std. Error Mean

Directive style 1 40 28,90 3,740 ,591

2 37 30,49 3,501 ,576

Supportive style 1 40 29,75 3,402 ,538

2 37 28,89 3,518 ,578

Participative style 1 40 26,20 4,416 ,698

2 37 24,97 4,400 ,723

Achievement-oriented style 1 40 26,38 3,940 ,623

2 37 27,08 4,278 ,703

R. Fabac, K. Kokot and I. Bubalo

364

Using the t test method for independent samples formed on the basis of size, the results

obtained on the frequency of styles are shown in Table 16.

Table 16. The influence of organizational size on leadership style: t test.

F Sig. t df

Significanc e Mean

Differe nce

Std. Error

Differe nce

95 % Confidence

One- Side d p

Two- Side d p

Lower Uppe

r

Directive style

EV assumed

1,004 ,320 -1,917 75 ,029 ,059 -1,586 ,827 -3,235 ,062

EV not assumed

-1,922 74,988 ,029 ,058 -1,586 ,825 -3,230 ,057

Supportiv e style

EV assumed

,111 ,739 1,088 75 ,140 ,280 ,858 ,789 -,713 2,430

EV not assumed

1,086 74,065 ,140 ,281 ,858 ,790 -,716 2,432

Participati ve style

EV assumed

,168 ,683 1,220 75 ,113 ,226 1,227 1,005 -,776 3,230

EV not assumed

1,221 74,575 ,113 ,226 1,227 1,005 -,776 3,230

Achievem ent-

oriented style

EV assumed

,090 ,765 -,754 75 ,227 ,453 -,706 ,936 -2,572 1,159

EV not assumed

-,752 73,110 ,227 ,455 -,706 ,939 -2,578 1,166

There was a statistically significant difference between the groups in terms of the frequency

of application of the leadership style, with p <0.1. For other styles, there was no difference

with the feature of statistical significance.

Furthermore, we apply the test of normality (Shapiro–Wilk) to test the assumption of

distribution normality, and here, we obtain a specific deviation result for Group 2 (medium and

large enterprises) in the distribution for the directive leadership style (Table 17). Since the

normality test for directive style suggests a deviation, we must seek confirmation of the results

by another method. We therefore apply the nonparametric Mann–Whitney U test, which is

commonly used when there is doubt about the normality of the sample distribution. This test is

suitable for ordinal variables, such as scoring outcomes using the Likert 7-point scale, and can

be applied to independent samples when there is no normal distribution of variables.

Table 17. The influence of size on leadership style – test of normality (Shapiro – Wilk).

Test of Normality (Shapiro–Wilk)

G W p

Directive style 1 0.967 0.282

2 0.877 < .001*

Supportive style 1 0.966 0.276

2 0.957 0.164

Participative style 1 0.970 0.370

2 0.978 0.645

Achievement-oriented style 1 0.976 0.536

2 0.976 0.578

*Significant results suggest a deviation from normality

Path-goal theory – leadership styles and their changes during covid-19 pandemic

365

The results obtained using both student’s t test and the Mann–Whitney U test are shown in

Table 18. The JASP statistical tool was used to apply and compare both methods. The Mann–

Whitney test also suggests a statistically significant difference in groups formed by

organization size and for the frequency of directive leadership style, even with a stricter

parameter for statistical significance (p <0.05).

Table 18. The influence of size on leadership style – Student test and Mann–Whitney test.

Independent Samples T Test Test Statistic df p

Directive style Student -1.917 75 0.059

Mann–Whitney 537.500 0.038

Supportive style Student 1.088 75 0.280

Mann–Whitney 846.500 0.278

Participative style Student 1.220 75 0.226

Mann–Whitney 848.500 0.269

Achievement-oriented style Student -0.754 75 0.453

Mann–Whitney 654.000 0.382

The application of directive leadership style in larger organizations is more frequent in the

sample, with statistical significance of p <0.05. Frequency in other leadership styles does not

“feel” the effect of the contingency size factor (Table 18). Since one of the conditions for the

application of the Mann–Whitney U test states that the shapes of distributions by groups must

meet the requirement of similarity, based on the analysis of Figure 2, we can say that a

certain degree of similarity exists.

Figure 2. Sample similarity (size) – Mann–Whitney test.

Determining the impact of education level on the frequency of leadership styles according to

PGT was the next task. As noted in Table 1, we identified two independent groups of

respondents, 21 from the category “Up to Bachelor’s degree“ and 56 from the category

“Bachelor’s, master’s, and doctoral degrees“. In Table 19, mark 1 is for the group “Up to

Bachelor’s degree“, while mark 2 is for the group “Bachelor’s, master’s, and doctoral degrees“.

R. Fabac, K. Kokot and I. Bubalo

366

Table 19. The influence of education on leadership. Group Statistics

Educational

level N Mean Std. Deviation

Std. Error

Mean

Directive style 1 21 30,57 4,106 ,896

2 56 29,32 3,501 ,468

Supportive style 1 21 29,38 3,918 ,855

2 56 29,32 3,314 ,443

Participative style 1 21 26,24 5,674 1,238

2 56 25,38 3,887 ,519

Achievement-oriented

style

1 21 28,29 4,256 ,929

2 56 26,13 3,908 ,522

Table 20 contains the results of the t test, assuming a normal distribution. A statistically

significant result was recorded in the PGT leadership style “achievement-oriented style“. The

group of respondents who have lower formal education estimate a higher frequency of

applying achievement-oriented leadership style compared to the more educated group. A

statistically significant difference was defined as p≤ 0.05.

Table 20. The influence of education on leadership style – t test.

F Sig. t df

Significance

Mean Difference

Std. Error Difference

95 % Confidence

One- Sided

p

Two- Sided

p Lower Upper

Directive style

EV assumed

,102 ,750 1,330 75 ,094 ,187 1,250 ,940 -,622 3,122

EV not assumed

1,237 31,542 ,113 ,225 1,250 1,011 -,810 3,310

Supportive style

EV assumed

1,223 ,272 ,067 75 ,473 ,947 ,060 ,892 -1,717 1,836

EV not assumed

,062 31,356 ,476 ,951 ,060 ,963 -1,903 2,022

Participative style

EV assumed

6,027 ,016 ,761 75 ,225 ,449 ,863 1,135 -1,397 3,124

EV not assumed

,643 27,353 ,263 ,526 ,863 1,343 -1,890 3,616

Achievement- oriented style

EV assumed

,245 ,622 2,109 75 ,019 ,038 2,161 1,025 ,120 4,202

EV not assumed

2,028 33,433 ,025 ,051 2,161 1,066 -,006 4,327

Path-goal theory – leadership styles and their changes during covid-19 pandemic

367

Furthermore, we apply the test of normality (Shapiro–Wilk) to test the assumption of

distribution normality, and here, we obtain the result of the deviation of both groups in the

directive style and Group 2 in the supportive style. The results are shown in Table 21. The

variables in the achievement-oriented style assessment satisfy the normality test.

Table 21. The influence of education on leadership style – test of normality (Shapiro–Wilk).

Test of Normality (Shapiro–Wilk)

G W p

Directive style 1 0.861 0.007

2 0.955 0.035

Supportive style 1 0.937 0.191

2 0.956 0.038

Participative style 1 0.956 0.439

2 0.977 0.345

Achievement-oriented style 1 0.968 0.697

2 0.969 0.159

*Significant results suggest a deviation from normality

Since in assessing the impact of a leader’s level of education on styles, we have a group with

N = 21 members, this is an additional argument that the conclusions are checked using the

Mann–Whitney test, as a nonparametric test for which there are no strict prerequisites.

Table 22. The influence of leadership style – Mann–Whitney test.

Hypothesis Test Summary

Null Hypothesis Test Sig.a Decision

1

The distribution of Directive

style is the same across

categories of education level

Independent-Samples

Mann–Whitney U Test ,089

Retain the null

hypothesis.

2

The distribution of Supportive

style is the same across

categories of education level

Independent-Samples

Mann–Whitney U Test ,918

Retain the null

hypothesis.

3

The distribution of Participative

style is the same across

categories of education level

Independent-Samples

Mann–Whitney U Test ,315

Retain the null

hypothesis.

4

The distribution of Achievement-

oriented style is the same across

categories of education level

Independent-Samples

Mann–Whitney U Test ,041

Reject the null

hypothesis.

aThe significance level is ,050, asymptotic significance is displayed

Table 22 contains the results of the independent samples test, and the conclusion is

unchanged, so with a significance level of p <0.05, the hypothesis of equal distribution for

achievement-oriented style in groups of more educated and less educated leaders is rejected.

Finally, regarding the similarity of the distribution samples, we analysed Figure 3, and we

can conclude that a degree of similarity between the samples exists. The results for Group 1

on the y-axis appear at slightly higher levels and end at higher levels, and the highest

frequency is also at higher levels compared to Group 2.

R. Fabac, K. Kokot and I. Bubalo

368

Figure 3. Sample similarity (education) – Mann–Whitney test.

DISCUSSION

The first goal of the research in this article was to determine whether and to what extent the

impacts of the COVID-19 pandemic are reflected in leadership styles, defined through the

PGT model. Using a questionnaire presented by Northouse [1], we examined the observations

of leaders in Croatian organizations regarding the frequency of applying certain leadership

styles. Looking at the individual questions contained in the instrument, only question Q2

(Appendix) was marked by a statistically significant difference (p <0.1), so the statement “I

maintain a friendly working relationship with subordinates“ in the period of a pandemic is

less accurate, and the frequency of such behaviour of leaders is less present. Such a

conclusion is in agreement with the discovery made by Chen, et al. [61] that during the

pandemic, telework undermined trust and relationships between managers and employees.

In addition, indications of changes of no statistical significance relate to differences in

behaviour with respect to statements Q5 (higher frequency), Q9 (higher frequency), Q10 (lower

frequency), Q11 (lower frequency), Q12 (higher frequency), Q14 (higher frequency) and Q18

(higher frequency), and details of these queries can be seen in the Appendix section.

Using the nonparametric Wilcoxon signed rank test, we confirmed a statistically significant

difference in the frequency of use of the directive style (with p < 0.05) and an indication of the

change in the frequency of achievement-oriented style (with p < 0.1) for the behaviour of

respondents who answered both questionnaires. The frequency of the directive style increased

somewhat during the COVID-19 pandemic, which means that leaders, on average, are more

likely to give specific guidance and various explanations to followers. Of course, during the

COVID-19 pandemic, basic instructions were relatively often given to citizens and employees

regarding the new rules, so the growth of the presence of the directive style was expected.

There are certain similarities between leadership in the general, undifferentiated work

environment noted by the author [20] and the circumstances of the COVID-19 pandemic,

which favour a task-oriented, directive leadership style. At the same time, with the onset of

the COVID-19 pandemic period, the frequency of achieving the achievement-oriented style

decreased. Since this style reflects a leader’s behaviour where he poses challenges seeking

greater achievement with confidence in employees [20], it is not surprising that this approach

has decreased. The conditions of the COVID-19 pandemic imposed healthcare priorities,

models of working from home, isolation of individuals, and the like, which reduced attempts

to raise the ladder of achievement.

Path-goal theory – leadership styles and their changes during covid-19 pandemic

369

In the general assessment of the achieved levels of PGT leadership styles at the present time

of pandemic rule, the supportive style stands out. Although it is a style characterized by

concern for employee welfare and the creation of a desirable organizational climate, with this

model, only 11.7 % of respondents reported applying this style the most often. Furthermore,

in accordance with the interpretation model for the questionnaire [1, pp. 146] in the

supportive leadership style part, we also found a very low representation of this behaviour.

Surprisingly, low levels were not observed in the remaining three styles of the PGT model.

Good effects of the application of supportive style are recorded in the studies of Schein [22],

Khalid, et al. [23], and Bass and Bass [25].

By testing groups of leaders formed by gender, the frequency of supportive style leadership

results in a statistically significant difference (p < 0.1), in favour of female leaders. This

result is in line with the findings of a study of subordinates from GLOBE societal clusters

conducted by Van Emmerik, et al. [33], where a higher level of friendly supportive behaviour

was found in female managers and is also consistent with the findings of Lord, et al. [40].

The influence of the organization’s size in the conducted research proved to be significant in

terms of the frequency of the directive style of the behaviour of leaders. The application of

this style in leaders from larger organizations in the sample is more frequent with statistical

significance (p < 0.05). It is possible that this circumstance has caused the initiation of

structure, as noted [34], or perhaps the need to maintain structure. It is known that structure is

more important in larger organizations.

A statistically significant result (with p <0.05) was obtained for the analysis of the

achievement-oriented style of behaviour of groups of more educated and less educated leaders

using the Mann–Whitney test. Less educated leaders are more likely to apply an achievement-

oriented style that Northouse [12] describes as desirable and challenging with complex tasks.

With regards to limitations of research instrument, we should stress that the applied Path –

Goal Leadership Questionnaire [1] is based on self-assessment of leaders, thus providing

“self-reported data“, potentially yielding undesirable biases such as selective memory and

others. As far as we know based on the analysis of other publications, no more advanced

version of this tool for verifying leader responses by subordinate employees or even

colleagues has been applied. A comprehensive consideration of the application of 360-degree

assessments can be found in study [63], while for example for management skills assessment

research [62] is relevant.

CONCLUSION

Consistent with self-judgement, leaders in Croatian organizations have changed their

leadership style in the circumstances of the COVID-19 pandemic, albeit not dramatically. To

a lesser extent, it strengthened the directive leadership style. The application of the directive

style by leaders from larger organizations is more frequent. A leadership style that is in some

ways deficient in representation is supportive leadership. For this style, female leaders in

Croatian organizations declare a higher frequency of application than their male counterparts.

Leaders of Croatian companies with lower education prefer an achievement-oriented style.

The frequency of using a participative leadership style has changed the least during pandemic

conditions. In addition, for this leadership style, no differences were found between groups

formed according to demographic criteria.

Future research on leadership in Croatian companies will be directed towards recent

theoretical concepts such as team leadership and multilevel leadership.

R. Fabac, K. Kokot and I. Bubalo

370

APPENDIX: QUESTIONNAIRE (PATH–GOAL LEADERSHIP QUESTIONNAIRE [1])

Indicate how often each statement is true of your own behaviour.

1 = Never 2 = Hardly ever 3 = Seldom 4 = Occasionally 5 = Often 6 = Usually 7 = Always

1. I let subordinates know what is expected of them.

2. I maintain a friendly working relationship with subordinates.

3. I consult with subordinates when facing a problem.

4. I listen receptively to subordinates’ ideas and suggestions.

5. I inform subordinates about what needs to be done and how it needs to be done.

6. I let subordinates know that I expect them to perform at their highest level.

7. I act without consulting my subordinates.

8. I do little things to make it pleasant to be a member of the group.

9. I ask subordinates to follow standard rules and regulations.

10. I set goals for subordinates’ performance that are quite challenging.

11. I say things that hurt subordinates’ personal feelings.

12. I ask for suggestions from subordinates concerning how to carry out assignments.

13. I encourage continual improvement in subordinates’ performance.

14. I explain the level of performance that is expected of subordinates.

15. I help subordinates overcome problems that stop them from carrying out their tasks.

16. I show that I have doubts about subordinates’ ability to meet most objectives.

17. I ask subordinates for suggestions on what assignments should be made.

18. I give vague explanations of what is expected of subordinates on the job.

19. I consistently set challenging goals for subordinates to attain.

20. I behave in a manner that is thoughtful of subordinates’ personal needs.

REFERENCES

[1] Northouse, P.G.: Leadership-Theory and Practice. 4th ed. In Croatian. MATE d.o.o., Zagreb, p.127, 2010,

[2] House, R.J. and Mitchell, T.R.: Path-goal theory of leadership. Journal of Contemporary Business 5, 81-94, 1974,

[3] Evans, M.G.: The effects of supervisory behavior on the path-goal relationship. Organizational Behavior & Human Performance 5(3), 277-298, 1970,

http://dx.doi.org/10.1016/0030-5073(70)90021-8,

[4] House, R.J.: A path-goal theory of leader effectiveness. Administrative Science Quaterly 16(3), 321-338, 1971,

http://dx.doi.org/10.2307/2391905,

[5] Vroom, V.H.: Work and motivation. Jossey-Bass, San Francisco, 1964,

[6] Lunenburg, F.C.: Goal-Setting Theory of Motivation. International Journal of Management, Business, and Administration 15(1), 43-48, 2011,

[7] House, R.J.: Path goal theory of leadership: lesson legacy and a reformulated theory. Leadership Quarterly 7(3), 324-352, 1996,

http://dx.doi.org/10.1016/S1048-9843(96)90024-7,

[8] Mohammad Mosadegh Rad, A. and Hossein Yarmohammadian, M.: A Study of

Relationship between Managers’ Leadership Style and Employees’ Job Satisfaction. Leadership in Health Services 19(2), 11-28, 2006,

http://dx.doi.org/10.1108/13660750610665008,

[9] Alanazi, T.R.; Khalaf, B.H. and Rasli, A.M.: Overview of Path-Goal Leadership Theory. Jurnal Teknologi (Social Sciences) 64(2), 49-53, 2013,

http://dx.doi.org/10.11113/jt.v64.2235,

Path-goal theory – leadership styles and their changes during covid-19 pandemic

371

[10] Polston-Murdoch, L.: An Investigation of Path-Goal Theory, Relationship of Leadership

Style, Supervisor-Related Commitment, and Gender. Emerging Leadership Journeys 6(1), 13-44, 2013,

http://dx.doi.org/10.1097/01.NUMA.0000437595.66556.08,

[11] Euwema, M.C.; Wendt, H. and Van Emmerik, H.: Leadership Styles and Group

Organizational Citizenship Behavior across Cultures. Journal of Organizational Behavior 28, 1035-1057, 2007,

http://dx.doi.org/10.1002/job.496,

[12] Northouse, P.G.: Leadership-Theory and Practice. 8th ed. SAGE Publications Ltd., Thousand Oaks, 2019,

[13] Sikavica, P.; Bahtijarević-Šiber, F. and Pološki Vokić, N. Fundamentals of management.

In Croatian. Školska knjiga, Zagreb, 2008,

[14] Yukl, G.: Leadership in Organizations. 3rd ed. Prentice Hall, Upper Saddle River, 1994,

[15] Farhan, B.Y.: Application of path-goal leadership theory and learning theory in a

learning organization. Journal of Applied Business Research 34(1), 13-22, 2018,

http://dx.doi.org/10.19030/jabr.v34i1.10088,

[16] Daft, R.L.: Leadership: Theory and Practice. The Dryden Press, New York, 1999,

[17] Sarti, D.: Leadership styles to engage employees: Evidence from human service

organizations in Italy. Journal of Workplace Learning 26(3), 202-216, 2014,

http://dx.doi.org/10.1108/JWL-09-2013-0066,

[18] Cruz, M.A.; Henningson, D.D. and Smith, B.A.: The impact of directive leadership on

group information sampling, decisions, perceptions of the leader. Communication Research 26, 349-370, 1999,

http://dx.doi.org/10.1016/S0160-7383(98)00103-0,

[19] Bell, C.; Chan, M., and Nel, P.: The Impact of Participative and Directive Leadership on

Organisational Culture: An Organisational Development Perspective. Mediterranean Journal of Social Sciences 5(23), 1970-1985, 2014,

[20] Somech, A.: Directive Versus Participative Leadership: Two Complementary

Approaches to Managing School Effectiveness. Educational Administration Quarterly 41(5), 777-800, 2005,

http://dx.doi.org/10.1177/0013161X05279448,

[21] Depoo, T., and Shanmuganathan, G.D.: An Empirical Investigation on Leadership Styles

Applied for Global Brand Management of Multinational Organizations in the United States. Open Journal of Business and Management 1, 11-17, 2013,

[22] Schein, E.H.: Organizational Culture and Leadership: A Dynamic View. 3rd ed. Jossey-Bass, San Francisco, 2018,

[23] Khalid, A., et al.: Role of Supportive Leadership as a Moderator between Job Stress and

Job performance. Information Management and Business Review 4(9), 487-495, 2012,

http://dx.doi.org/10.22610/imbr.v4i9.1004,

[24] La Rocco, J.M., and Jones, A.P.: Co-worker and leader support as moderators of

stressstrain relationships in work situations. Journal of Applied Psychology 63(5), 629-634, 1978,

http://dx.doi.org/10.1037/0021-9010.63.5.629,

[25] Bass, B.M. and Bass, R.: The Bass handbook of leadership: theory, research, and

managerial applications. 4th ed. Free Press, New York, 2008,

R. Fabac, K. Kokot and I. Bubalo

372

[26] Rafferty, A.E. and Griffin, M.A.: Dimensions of transformational leadership:

Conceptual and empirical extensions. Leadership Quarterly 15, 329-354, 2004,

http://dx.doi.org/10.1016/j.leaqua.2004.02.009,

[27] Malik, S.H.; Aziz, S. and Hassan, H.: Leadership Behavior and Acceptance of Leaders

by Subordinates: Application of Path Goal Theory in Telecom Sector. International Journal of Trade, Economics and Finance 5(2), 170-175, 2014,

http://dx.doi.org/10.7763/IJTEF.2014.V5.364,

[28] Dokony, H.; Singh, J.S. and Arumugam, T.: The Influence of Leadership Behaviors

Based on the Path-Goal Theory towards Employees’ Satisfaction in a Developing

Nation. A Study in the Telecommunication Sector in N’djamena, Chad. International Journal of Psychosocial Rehabilitation 24(2), 1324-1336, 2020,

[29] Antonakis, J. and House, R.J.: Instrumental leadership: Measurement and extension of

transformational–transactional leadership theory. The Leadership Quarterly 25(4), 746-771, 2014,

http://dx.doi.org/10.1016/j.leaqua.2014.04.005,

[30] Hunt, J.G.: Task leadership. In: Goethels, G.R.; Sorensen, G.J. and Burns, J.M, eds.: Encyclopedia of leadership IV. Sage,

Thousand Oaks, pp.1524-1529, 2004,

[31] Udovičić, A.; Požega, Ž. and Crnković, B.: Analysis of leadership styles in Croatia. Ekonomski Vjesnik 27(1), 71-80, 2014,

[32] Miloloža, I.: Analysis of the Leadership Style in Relation to the Characteristics of

Croatian Enterprises. Interdisciplinary Description of Complex Systems 16(2), 249-264, 2018,

http://dx.doi.org/10.7906/indecs.16.2.5,

[33] Van Emmerik, H.; Euwema, M.C. and Wendt, H.: Leadership behaviors around the

world: The relative importance of gender versus cultural background. International Journal of Cross-Cultural Management 8(3), 297-315, 2008,

http://dx.doi.org/10.1177/1470595808096671,

[34] Yammarino, F.J.; Dionne, S.D.; Uk Chun, J. and Dansereau, F.: Leadership and Levels

of Analysis: A State-of-the-Science Review. Leadership Quarterly 16(6), 879-919, 2005, http://dx.doi.org/10.1016/j.leaqua.2005.09.002,

[35] House, R.J., et al.: Culture, Leadership, and Organizations: The GLOBE Study of 62 Societies. Sage, Thousand Oaks, 2004,

[36] Gupta, V.; Hanges, P.J. and Dorman, P.: Cultural Clusters: Methodology and Findings. Journal of World Business 37(1), 11-15, 2002,

http://dx.doi.org/10.1016/S1090-9516(01)00070-0,

[37] Kent, T.W.; Blair, C.A.; Rudd, H.F. and Schuele, U.: Gender Differences and Transformational

Leadership Behavior: Do Both German Men and Women Lead in the Same Way? International Journal of Leadership Studies 6(1), 52-67, 2010,

[38] Lopez-Zafra, E.; García-Retamero, R. and Martos, M.P.B: The Relationship Between

Transformational Leadership and Emotional Intelligence from a Gendered Approach. The Psychological Record 62, 97-114, 2012, http://dx.doi.org/10.1007/BF03395790,

[39] Thacker, R.A.: Gender, influence tactics, and job characteristics preferences: New

insights into salary determination. Sex Roles 32, 617-638, 1995,

http://dx.doi.org/10.1007/BF01544215,

[40] Lord, R.G., et al.: Leadership in applied psychology: three waves of theory and research. Journal of Applied Psychology 102(3), 434-451, 2017,

http://dx.doi.org/10.1037/apl0000089,

Path-goal theory – leadership styles and their changes during covid-19 pandemic

373

[41] Becker, J.; Ayman, R. and Korabik, K.: Discrepancies in self/subordinates perceptions of

leadership behavior: Leader’s gender, organizational context, and leader’s self-monitoring. Group & Organization Management 27(2), 226-244, 2002,

http://dx.doi.org/10.1177/10501102027002004,

[42] Helgesen, S.: The Female Advantage: Women’s Ways of Leadership. Doubleday Currency, New York, 1990,

[43] Chamorro-Premuzic, T.: Why Do So Many Incompetent Men Become Leaders? (And

How to Fix It). Harvard Business Review Press, 2013,

[44] European Council: Chronology – activities of the European Council on COVID-19. https://www.consilium.europa.eu/hr/policies/coronavirus/timeline,

[45] Mercedes, D. and Burrell, D.N. Managerial adaptability and business strategic change

in age of COVID-19. PSU Research Review, Emerald Publishing Limited, 1-15, 2021,

http://dx.doi.org/10.1108/PRR-12-2020-0046,

[46] Burrell, D.N.: Understanding the talent management intricacies of remote cybersecurity

teams in covid-19 induced telework organizational ecosystems. Land Forces Academy Review 25(3), 232-244, 2020,

http://dx.doi.org/10.2478/raft-2020-0028,

[47] Gerada, C.: Reflections on leadership during the Covid pandemic. Postgraduate Medicine 133(7), 717-720, 2021,

http://dx.doi.org/10.1080/00325481.2021.1903218,

[48] Harris, A. and Jones, M.: COVID 19 – school leadership in disruptive times. School Leadership & Management 40(4), 243-247, 2020,

http://dx.doi.org/10.1080/13632434.2020.1811479,

[49] Lee, M.M.: Covid-19: agnotology, inequality, and leadership. Human Resource Development International 23(4), 333-346, 2020,

http://dx.doi.org/10.1080/13678868.2020.1779544,

[50] Antonakis, J. and House, R.J.: Instrumental leadership: Measurement and extension of

transformational–transactional leadership theory. The Leadership Quarterly 25(4), 746-771, 2014,

http://dx.doi.org/10.1016/j.leaqua.2014.04.005,

[51] Claus, L.: Do We Need a New Leadership Paradigm Due to Covid-19? International Journal of Business and Management Research 9(2), 162-167, 2021,

http://dx.doi.org/10.37391/ijbmr.090206,

[52] Watkins, D.V. and Clevenger, A.D.: US Political Leadership and Crisis Communication

During COVID-19. Cogent Social Sciences 7, No. 1901365, 1-33, 2021,

[53] Boin, A.; Hart, P.; Stern, E. and Sundelius, B.: The Politics of Crisis Management:

Public Leadership Under Pressure. Cambridge University Press, New York, 2005,

http://dx.doi.org/10.1017/CBO9780511490880,

[54] Eichenauer, C.; Ryan, A. and Alanis, J.M.: Leadership During Crisis: An Examination of

Supervisory Leadership Behavior and Gender During COVID-19. Journal of Leadership & Organizational Studies 29(2) 1-18, 2022,

http://dx.doi.org/10.1177/15480518211010761,

[55] Zehnter, M.K.; Olsen, J. and Kirchler, E.: Obituaries of female and male leaders from

1974 to 2016 suggest change in descriptive but stability of prescriptive gender stereotypes. Frontiers in Psychology 9(1), 1-5, 2018,

http://dx.doi.org/10.3389/fpsyg.2018.02286,

[56] Indvik, J.: A path-goal theory investigation of superior-subordinate relationship. PhD Thesis. University of Winsconsin, Madison, 1985,

R. Fabac, K. Kokot and I. Bubalo

374

[57] Indvik, J.: A more complete testing of path-goal theory. Academy of Management Annual Meeting, Anaheim, 1988,

[58] Cortina, J.M.: What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology 78(1), 98-104, 1993,

http://dx.doi.org/10.1037/0021-9010.78.1.98,

[59] Field, A. Discovering statistics using IBM SPSS Statistics. SAGE Publications Ltd., London, 2013,

[60] Kim, T.K.: T test as a parametric statistic. Korean Journal of Anesthesiology 68(6), 540-546, 2015,

http://dx.doi.org/10.4097/kjae.2015.68.6.540,

[61] Chen, J.K.C. and Sriphon, T.: Perspective on COVID-19 Pandemic Factors Impacting

Organizational Leadership. Sustainability 13(6), No. 3230, 2021,

http://dx.doi.org/10.3390/su13063230,

[62] Mihalcea, A. and Mihalcea, D.: Management Skills Assessment Using 360° Feedback –

Msf 360. Procedia – Social and Behavioral Sciences 187, 318-323, 2015,

http://dx.doi.org/10.1016/j.sbspro.2015.03.059,

[63] Hardison, C.M., et.al.: 360-Degree Assessments: Are They the Right Tool for the U.S. Military?

RAND Corporation, 2015.

Copyright of Interdisciplinary Description of Complex Systems is the property of Croatian Interdisciplinary Society and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.