Impact of Supportive Supervisor on Doctoral Students’ Research Productivity: The Mediating Roles of Academic Engagement and Academic Psychological Capital

Abstract

This research investigates the impact of supportive supervisors on doctoral students’ research productivity, with parallel mediation effects of academic engagement and academic psychological capital as two mediators. Data has been collected through an online survey from international doctoral students (N = 415) studying in six research-oriented universities in mainland China. Confirmatory factor analysis and structural modeling were used in the analysis and mediation analysis conducted by adopting the 04 Model in PROCESS. The results indicated that a supportive supervisor is positively related to research productivity. Student psychological factors—academic engagement and academic psychological capital—partially mediate the relationship between supportive supervisors and research productivity. The findings suggest that the supervisor’s supportive behavior is essential for encouraging students’ academic engagement and academic psychological capital. Furthermore, students are more productive, and their engagement and psychological resources are increased under supportive supervision, which ultimately significantly increases their research productivity.

Full text

Original Research
SAGE Open
July-September 2023: 1–15
ÓThe Author(s) 2023
DOI: 10.1177/21582440231185554
journals.sagepub.com/home/sgo
Impact of Supportive Supervisor
on Doctoral Students’ Research
Productivity: The Mediating Roles
of Academic Engagement and
Academic Psychological Capital
Waqas Khuram
1
, Yanqing Wang
1
, Mudassar Ali
2
, Aisha Khalid
2
,
and Heesup Han
3
Abstract
This research investigates the impact of supportive supervisors on doctoral students’ research productivity, with parallel
mediation effects of academic engagement and academic psychological capital as two mediators. Data has been collected
through an online survey from international doctoral students (N=415) studying in six research-oriented universities in main-
land China. Confirmatory factor analysis and structural modeling were used in the analysis and mediation analysis conducted
by adopting the 04 Model in PROCESS. The results indicated that a supportive supervisor is positively related to research
productivity. Student psychological factors—academic engagement and academic psychological capital—partially mediate the
relationship between supportive supervisors and research productivity. The findings suggest that the supervisor’s supportive
behavior is essential for encouraging students’ academic engagement and academic psychological capital. Furthermore, stu-
dents are more productive, and their engagement and psychological resources are increased under supportive supervision,
which ultimately significantly increases their research productivity.
Keywords
supportive supervisor, research productivity, academic engagement, academic psychological capital, international doctoral
student
Introduction
Research productivity (RP) through doctoral students’
performance and contribution has recently emphasized
research in higher education settings (Brew et al., 2016 ).
The scientific literature has identified the numerous fac-
tors influencing research performance. However, a doc-
toral supervisor’s role is essential among various factors
(i.e., cultural, personal, contextual) that contribute to the
student’s research performance (Bui, 2014; Devine &
Hunter, 2017; Peng, 2015). The significance of doctoral
supervision is considerably determined by a vast body of
concern among scholars (Bui, 2014; Khuram, Wang,
Khan, & Khalid, 2021; Khuram et al., 2017). Doctoral
supervision plays a crucial role in information sharing,
motivating, and helping students to become independent
researchers (Halse & Malfroy, 2010), with comprehensive
and detailed experience in theory and knowledge to carry
out research and achieve the set targets (Peng, 2015).
Therefore, Gong et al. (2009) claimed that the supervisor
role is getting attention among scholars, especially super-
visory styles. Existing literature described that doctoral
supervisor plays a crucial role in academic success during
doctoral candidature (Gruzdev et al., 2020; Mason, 2018;
1
Harbin Institute of Technology, Harbin, China
2
Harbin Normal University, Harbin, China
3
Sejong University, Gwangjin-gu, Republic of Korea
*Mudassar Ali is also affiliated to Capital University of Science &
Technology, Islamabad, Pakistan
Corresponding Author:
Heesup Han, College of Hospitality and Tourism Management, Sejong
University, 825-B Gwanggetto-Kwan, Gwangjin-gu 05006, Republic of
Korea.
Email: heesup@sejong.ac.kr
Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License
(https://creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of
the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages
(https://us.sagepub.com/en-us/nam/open-access-at-sage).

Yang et al., 2020). Students have recently begun to con-
sider various aspects of supervision styles, such as pas-
toral and contractual (Gatfield & Alpert, 2002),
supportive and directive (Gu et al., 2015), to have a role
in student RP in higher education settings. Among such
styles, supportive supervisors (SS) may be an essential
predictor of research productivity. Although supportive-
ness is a critical characteristic of a doctoral supervisor
(Lindqvist, 2018), the empirical relationship between SS
and RP has yet to be explored. SS has been defined as
friendly behavior with subordinates and exhibiting con-
cerns for mental/physical preferences and satisfaction,
enabling them toward high productivity and performers
(Gu et al., 2015).
Prior studies have found that a supportive supervisor
significantly influences an individual’s commitment that
enables them to perform high and achieve goals
(Khuram et al., 2021b; Yang et al., 2020), ultimately
leading to productivity in doctoral research. Therefore,
this study aims to examine the relationship between SS
and research productivity. Assessment of the association
between SS and RP shows how the supportive supervisor
supports and motivates doctoral students toward suc-
cessful research objectives with high productivity. We
propose academic engagement (AE) and psychological
capital (PsyCap) as two parallel mediating mechanisms
between SS and research productivity. AE is a psycholo-
gical state of mind, that is rooted and characterized by
vigor, dedication, and absorption that reflects an individ-
ual’s connection with their tasks (Bakker et al., 2008).
Previous studies have explicitly found an empirical rela-
tionship between supervisor-supportive behavior and
students’ AE (Ahmed et al., 2017; Kahu et al., 2015).
The supervisor’s supporting and guiding behavior is
related to doctoral students’ performance (Mainhard
et al., 2009), as it intrinsically motivates students to think
innovatively in research-related work (Yidong & Xinxin,
2013). Such supervision gave them the confidence that
their knowledge and insight could significantly contrib-
ute; such appreciation and recognition motivate
researchers to continue to work hard, especially in a
research-oriented task (Gonza
´lez-Ocampo & Castello
´,
2019). These research studies indicate that a supportive
supervisor is likely to enhance students’ AE by support-
ing students’ efforts, and strengths, providing feedback,
and acknowledging their contribution and knowledge to
achieve academic objectives. Therefore, we propose that
AE has a mediating role between a supportive supervisor
and a student’s research productivity. Another possible
intervening mechanism is the student’s academic
PsyCap. PsyCap is the developmental psychological state
of a person consisting of hope, efficacy, resilience, and
optimism (HERO) (F. Luthans & Youssef-Morgan,
2017; K. W. Luthans et al., 2019). Students used their
psychological abilities to assess, develop, and manage to
improve academic performance during academic tenure
(Martı
´nez et al., 2019).
Supervisors supporting and encouraging behavior can
enhance supervisees’ confidence and recognize their
motivational efforts toward research activities, which
give them positive feelings that their work and learnings
impact their productivity (Khuram et al., 2021a; Platow,
2012). PsyCap can lead the students to overcome tense
situations and resolve research-oriented problems, which
becomes an essential component for their academic suc-
cess. It suggests that doctoral students’ academic PsyCap
can mediate the relationship between the supportive
supervisor and research productivity. The present study
presents a theoretical model based on the previous dis-
cussion. The supervisor is, directly and indirectly, related
to the RP via AE and academic PsyCap. This research
aims to introduce the framework and contribute to the
existing research literature. This research would expand
the understating of previous research on the supportive
behavior of supervisors in supervising doctoral students
(Fan et al., 2019); It will increase our awareness of the
human aspect that is central to research productivity.
This research would also expand the literature on the
student’s psychological factors in research-oriented doc-
toral education (Ahmed et al., 2017).
Moreover, this research enhances the understanding
of Hobfoll’s ‘‘conservation of resource theory’’ (COR)
by applying the theoretical assumptions in this study.
Hobfoll (2001) claimed that the supervisor considered a
resource supporting, motivating, and encouraging stu-
dents to handle their resources. Therefore, we propose
that the COR theory model enhances our understanding
of how SS is related to resource management strategies
representing RP improvement. In particular, this
research aims to contribute to the existing literature by
introducing a COR theoretical model to show how a
supportive supervisor is, directly and indirectly, related
to RP through AE and academic psychological capital.
Theoretical Background
This study draws from COR theory (Hobfoll, 1989,
2001) as its theoretical lens to explain the SS-RP rela-
tionship and mediating role of AE and psychological
capital between SS and RP. COR is a comprehensive
motivational and stress theory based on concepts and
principles of resources (e.g., loss, gain, and investment)
associated with related outcomes.
It describes the effects of stressful/ motivational (sup-
portive) surroundings/ circumstances on individuals’
actions/outcomes. The COR model’s fundamental con-
cept is for individuals to develop, secure, preserve, and
maintain their resources (Ali et al., 2020). In this regard,
2SAGE Open

Hobfoll (2001), therefore, suggests that resources could
be the objects, positions, characteristics, qualities or
potentials, and energy valued by the individuals that can
be increased in a conducive environment; however, loss
of such resources is most detrimental to a person (Ali,
Li, Khan, et al., 2021).
From the standpoint of COR theory, RP is likely to
increase when the gain in resources follows the resources
investments. In other words, students might feel moti-
vated and perform well when their supervisor supports
them in handling academic activities. Supportiveness is
an essential interpersonal feature of the supervisor, and
it is intended to improve doctoral learning skills, under-
standing, and trust, which contribute to building a con-
ducive learning and collaboration environment, leading
to RP (i.e., resources gain). In fact, the sustained sense of
supportiveness increases the motivational and psycholo-
gical resources of the supervisee, which enhances their
productivity. Precisely, SS increases students’ RP by
investing their resources (e.g., expertise sharing, guide-
lines, time) and providing support to increase student
engagement and develop their psychological resources to
acquire more resources and improve their research
productivity.
Literature Review and Hypothesis Development
Supportive Supervisor. A supportive supervisor refers to
relations-oriented behavior and expresses concerns for
subordinates’ psychological preferences and well-being
(Devine & Hunter, 2017; Fan et al., 2019). Accordingly,
relation-oriented behavior mainly aims to enhance the
relationship through helping, cooperation, and identifi-
cation (Gruzdev et al., 2020). However, such behavior
considers the interpersonal quality of supervisors, which
recognize by supervisees during the interaction. The
behavioral attributes of a supportive supervisor, such as
an emphasis on helping and satisfaction of subordinates,
needs, and preferences, caring, welfare, and creating a
friendly psychological environment (R. J. House, 1996;
Platow, 2012), appreciating the efforts, and contribution
(Amabile et al., 2004), recognize their skills and strengths
(Dangel & Tanguay, 2014). Moreover, supervisors are
open to sharing new ideas, providing constructive feed-
back, and encouraging and facilitating subordinates in
the skill development process by providing support
(Khuram & Wang, 2018). Extant literature indicates that
a supportive supervisor (leadership) seems to be more
influential than a directive leadership approach task-
oriented; a supportive supervisor is unique from a direc-
tive supervisor in several aspects. For instance, J. S.
House (1983) and Gu et al. (2015) comprehensively
described how the supportive style of supervision is
unique regarding (behavior, support, and outcomes) as
compared to the directive style of supervision, which
may have adverse effects on students due to its control-
ling, dominating, and directing supervising style.
Therefore, Devine and Hunter (2017) mentioned that
supervision in the academic setting is similar to a non-
academic context, whose behaviors significantly influ-
ence subordinates’ performance. In the context of higher
education, supportive supervision usually entails per-
sonal, academic, and autonomy support (Gu et al., 2015)
for improving the student’s productivity. Accordingly,
doctoral supervisors must be proactive and supportive,
clearly understanding research plans to achieve higher
productivity and performance (Fan et al., 2019).
Doctoral supervisors play different roles (e.g., teacher,
guide, coach, mentor, and even critic), so the supervisee’s
performance can be improved (Gruzdev et al., 2020;
Khuram & Wang, 2018). Thus, research showed that
supervisors’ support and appreciation could also increase
students’ self-confidence, motivating them to improve
their performance (Gu et al., 2015). In a similar vein, a
review study (Johansson & Yerrabati, 2017) stated that
doctoral students showed satisfaction with the supervi-
sors’ professional and obliging behavior in assisting them
and cooperating with research scholars to seek and
understand the knowledge and think innovatively.
Moreover, they support students in achieving their
objectives (Gu et al., 2015) and develop their skills and
make them self-sufficient, creative, and innovative con-
tributors (Fan et al., 2019).
In contrast, the directive supervisor in the academic
context emphasizes that the supervisee must follow their
instructions and commands. Specifically, directive super-
visors are inclined to control, dominate, and direct stu-
dents and compel them to think and behave in specific
ways rather than their supervisee’s initiatives and crea-
tive opinions during doctoral research (Gu et al., 2015).
Therefore, such a supervision style may decrease the
supervisee’s knowledge exploration eagerness, and inno-
vative productivity (Fan et al., 2019). Precisely, SS helps
improve graduate students’ creativity, and it will increase
their exploratory and seeking spirit while continuing to
work effectively. These characteristics of a supportive
supervisor are assumed to impact doctoral students’
research performance and productivity substantially.
Supportive Supervisor and Research Productivity. Existing
research indicates that a supervisor supports significant
impacts on doctoral students’ academic outcomes, such
as high-grade achievements and academic performances
(i.e., creativity and innovations) (Ahmed et al., 2017; Lee
et al., 2020). In higher education settings, a supervisor’s
supportive behavior develops the supervisee’s knowl-
edge, understanding, research capabilities, encourage-
ment, and engagement in conducting scientific research,
enabling students to become professionals and
Khuram et al. 3

accomplish innovative achievements during an academic
tenure (Peng, 2015). High-level supervisors’ support,
research competence, communication, and academic gui-
dance cultivate an atmosphere where students and super-
visors can both be interconnected and continuously exert
efforts to accomplish their set research targets (Peng,
2015). Supervisor support enables their research associ-
ates by appreciating their ideas and inputs, increasing
their confidence and interaction to frequently discuss
their problems with supervisors and actively engage in
learning-oriented activities to meet specific objectives
(Gruzdev et al., 2020). Past studies on doctoral educa-
tion have broadly highlighted the significance of supervi-
sion in guiding students’ learning, promoting researcher
development, and increasing productivity during the
doctoral journey (Gonza
´lez-Ocampo & Castello
´, 2019).
However, scholars defined RP as several scholarly publi-
cations’ quality, impact factor, and collecting or gather-
ing data of research articles (Abramo & D’Angelo, 2014;
Kahn & Scott, 1997; Khuram et al., 2017).
Therefore, since the peer-reviewed journal’s research
publication becomes a graduation criterion (Mason,
2018), both supervisor and student are compelled to
publish research work during doctoral education. In
such situations, supervisors collaborate and share skills
to fulfill graduation requirements and develop the doc-
toral student’s ability to perform scientific research
competently. Furthermore, the supervisor’s role in aca-
demic (higher education) settings is viewed as a
resource that provides more support for the positive
growth of subordinate skills (Ahmed et al., 2017;
Devine & Hunter, 2017). This phenomenon is known
as the conservation of resource (COR) theory (Hobfoll,
1989), and recently this has become important and
widely applied in organizational and human psychol-
ogy studies (Ali et al., 2020). The COR model’s funda-
mental concept is for individuals to develop, secure,
preserve, and maintain their resources (Ali et al., 2020).
However, suggest that resources could be the objects,
positions, characteristics, qualities or potentials, and
energy valued by the individuals that can be increased
in a conducive environment; however, loss of such
resources is most detrimental to a person (Ali, Li,
Khan, et al., 2021). Accordingly, we assume that sup-
portiveness is an essential interpersonal characteristic
of a supervisor and devoted to enhancing doctoral stu-
dents’ learning skills, understandings, and confidence,
which helps to format a conducive environment of
learning and cooperation, resulting in RP (i.e.,
resources gain).
The above discussion indicates that the doctoral
supervisor’s supportive behavior may increase a doctoral
student’s research performance, increasing research pro-
ductivity. Therefore, it is proposed that:
H1: Supportive supervisor is positively related to
research productivity.
The Mediating Role of Academic Engagement. Existing
educational research studies have shown that supervisor
support is positively related to AE (Ahmed et al., 2017).
Student engagement in higher education settings is
broadly recognized as a significant and essential compo-
nent of academic success (Hughes & Coplan, 2010;
Kahu et al., 2015). However, Schaufeli et al. (2002) and
Bakker et al. (2008) defined ‘‘engagement as a positive
and psychologically fulfilling state of mind that is char-
acterized by vigor, dedication, and absorption.’’ Vigor
refers to high levels of strength and mental endurance in
spending effort and resilience in facing difficulties
(Schaufeli et al., 2002). Dedication shows the involve-
ment, eagerness, encouragement, pride, challenge, and
sense of a person’s importance in their work (Schaufeli
et al., 2002). Absorption refers to being entirely focused
and delightedly immersed in work as time passes fast
and challenging for an individual to disengage from the
task (Schaufeli et al., 2002). Therefore, engaged persons
are high in energy, enthusiastic, strongly recognized, and
connected with their tasks (Bakker et al., 2008). In addi-
tion to this, (Swanberg et al., 2011) stated that supervisor
support is one of the critical characteristics of cultivating
the environment for student engagement.
The supervisor’s guidance and support enrich student
engagement (Ahmed et al., 2017). SS gives autonomy in
conducting scholarly research, which creates a sense of
self-confidence and efficacy in students to alleviate per-
formance levels and encourage them to freely interact
with followers (Overall et al., 2011). According to
(Gonza
´lez-Ocampo & Castello
´, 2019), SS views students’
failure as an opportunity to improve and achieve. Thus,
doctoral students tend to engage in risk-taking activities
(Xu & Grant, 2020). The doctoral supervisor supports
such engagement by giving the appropriate award or
refusing to penalize if targeted results are not achieved
(Lee et al., 2020). Doctoral students feel confident and
frequently share matters and ideas with supervisors and
get constructive responses (Lindqvist, 2018). The super-
visor enhances such confidence by creating a conducive
climate where the supervisee feels free to share innova-
tive concepts that motivate individuals to develop inno-
vative skills to solve problems creatively (Gu et al.,
2015). Furthermore, the supervisor supports doctoral
students through guiding and suggestion, enabling them
to follow the experimental approach to doing new things
(Devine & Hunter, 2017; Mainhard et al., 2009), which
promotes their AE (Ahmed et al., 2017)
Prior studies indicate that engagement is positively
related to student achievements, performance, and satis-
faction (Hughes & Coplan, 2010; Qureshi et al., 2021).
4SAGE Open

Similarly, in a non-academic study, Swanberg et al.
(2011) stated that engaged individuals are highly moti-
vated, energized, and thoroughly involved in their work,
positively relating to productivity. Accordingly, a doc-
toral student’s engagement in scholarly activities is relied
upon to enhance research performance and productivity.
In a similar vein, O’Keeffe (2020) stated that in doctoral
education, individuals actively contribute and fill the
knowledge gaps through scholarly research and publish-
ing their academic publications.
Fan et al. (2019) asserted that student engagement in
innovative research activities must be appreciated. It is a
valuable method to encourage them to explore new
knowledge and pursue innovation in a competitive
research environment. Students’ engagement in various
scholarly activities can provide innovative insights and
creative expertise that promote learning and academic
work (Nguyen et al., 2018). By engaging themselves in
scholarly activities, students valued their efforts and rea-
lized that such engagement behavior leads to academic
achievements (Hughes & Coplan, 2010).
The previous discussion indicated that a supportive
supervisor encourages student engagement (Bakker
et al., 2008), and significantly influences performance.
Specifically, SS increases students’ RP by facilitating
their engagement in research activities. Furthermore, a
COR helps understand the mediating role of students’
AE between the supportive supervisor and the research
productivity. A supportive supervisor is an essential
resource at the workplace that encourages subordinates
to devote discretionary behavior (i.e., AE), which even-
tually supports them in accomplishing desired research
productivity. Consequently, the rationality of the philo-
sophy of conversation of resources reveals that people
with vast resources can invest resources to gain adequate
resources in return (Ali et al., 2020; Hobfoll, 1989).
Therefore, according to the above discussion, we pro-
pose that
H2a. A supportive supervisor is positively related to
academic engagement.
H2b. Academic engagement is positively related to
research productivity.
H2c. Academic engagement mediates the relationship
between supportive supervisors and research
productivity.
The Mediating Role of Academic Psychological
Capital. Psychological capital describes in an academic
context as the study of human resources and abilities to
use psychological resources by students to assess posi-
tively, develop, and improve performance during aca-
demic tenure (K. W. Luthans et al., 2019). A
psychological developmental state is characterized by
hope, efficacy, resilience, and optimism (F. Luthans
et al., 2007). Hope is a ‘‘positive motivational state based
upon an interactive sense of the successful (a) agency
(‘‘willpower’’) and (b) the path (‘‘way power’’)’’ (Snyder
et al., 2002). Self-efficacy is derived from social cognitive
theory (Bandura, 1997) and is defined as individuals with
higher (confidence) efficacy tend to carry out activities
and stay calm while confronting challenges or problems
in performing any task. Optimism refers to the positive
attribution of an individual that builds and uses the
exploratory style in response to specific situations and
events (Seligman, 2006). Resilience ‘‘refers to positive
adaptation in the light of severe difficulties or harm’’
resilience ability helps individuals to learn to tackle and
overcome setbacks, and failures and keep focusing on
performing at a high level (K. W. Luthans et al., 2019).
Supportive behavior of the supervisor can enhance the
student’s psychological capital (Ahmed et al., 2017); simi-
larly, in the non-academic study, (Schaufeli et al., 2002)
found that coaching has a significant impact on workers’
psychological capital. For example, the supervisor recog-
nizes the supervisee’s motivational efforts toward
research activities (Johansen et al., 2019). Providing
timely guidelines and support to subordinates with kind-
ness, reverence, and appreciation (Fan et al., 2019) gives
them positive feelings. It increases their hope and confi-
dence that their work and learning will impact their pro-
ductivity (Platow, 2012). The supervisor acknowledges
students’ expertise and knowledge to boost their self-
belief and efficacy and encourage them to achieve the
desired objectives (Johansen et al., 2019; Johansson &
Yerrabati, 2017). Therefore, highly confident and effica-
cious exert more effort to achieve high research perfor-
mance and compete in every situation (B. C. Luthans
et al., 2012). Moreover, the supervisor’s supportive beha-
vior provides an atmosphere where students can gain
academic autonomy and get valuable input from supervi-
sors, which will promote students’ productivity (Gu
et al., 2015). Scholars have also found that psychological
resources positively influence academic performance
(Martı
´nez et al., 2019). Previous research studies linked
each PsyCap component with positive academic out-
comes and students using such resources during their
academic tasks (K. W. Luthans et al., 2019).
Moreover, Youssef-Morgan and Youssef-Morgan
and Luthans (2013) found that students’ academic
PsyCap resources lead them to success in different ways;
for example, cognitively evaluating the situation and
retaining positivity and self-motivation, self-belief, and
trust to assess their academic aim and putting more
effort to achieve it. A higher level of PsyCap resources
may build various strategies to overcome the challenges
and learn from mistakes (Youssef-Morgan & Luthans,
2013). Furthermore, PsyCap students accept and
Khuram et al. 5

overcome cognitive problems and become resilient and
satisfied with their performance (B. C. Luthans et al.,
2012). Similarly, doctoral students with a high PsyCap
capability will contribute significantly to their research
accomplishments by performing well and increasing their
research productivity.
The above discussion suggests that academic psycho-
logical capital plays a mediating role between SS and
RP. Moreover, our arguments for the mediating effects
of academic psychological capital can be understood
through the COR theory (Ali, Li, Durrani, et al., 2021;
Hobfoll, 2001). We believe that SS invests its resources in
developing student resources (e.g., academic PsyCap) to
acquire more resources and improve its research produc-
tivity. Academic psychological capital seems to mediate
the effects of SS on the supervisee’s research productiv-
ity. Based on the above discussion, we propose the fol-
lowing hypothesis:
H3a. A supportive supervisor is positively related to
academic psychological capital.
H3b. Academic psychological capital is positively
related to research productivity.
H3c. Academic psychological capital mediates the
relationship between supportive supervisors and
research productivity.
Thus, we propose the empirical model, as shown in
Figure 1.
Method
Research Settings and Participants
According to the Chinese Ministry of Education (2018)
report, approximately 492,000 international students are
studying all over China in different degree programs
offered by Chinese higher education institutions.
Therefore, this study has only focused on international
doctoral students and targeted populations, as they are
pursuing education in research-oriented degree pro-
grams. With a few exceptions, most international doc-
toral students conduct scientific research under a
Chinese professor (supervisor), contributing significantly
to enhancing new knowledge during their doctoral candi-
dature (Fan et al., 2019), a suitable research population
for this study.
Sample and Procedure
An online survey (WeChat) has been conducted from the
international doctoral students enrolled in a league of
nine Chinese research-oriented universities (C9 League).
The C9 League includes Tsinghua University (TU),
Peking University (PU), Shanghai Jiao Tong University
(SJTU), Fudan University (FU), Zhejiang University
(ZU), Nanjing University (NU), Harbin Institute of
Technology (HIT), University of Science and Technology
(USTC), Xi’an Jiao Tong University (XJTU). We consid-
ered drawing a sample from C9 League universities
based on their ranks, academic research achievements,
and relatively high concentration and enrollment of
international students. The questionnaire accompanied a
cover letter to assure the anonymity of the respondents.
The respondents were assured that their information
would be kept confidential and not used for any other
purpose other than this research. The respondents were
also told that there were no right and wrong answers,
and they were free to choose an option to rate a ques-
tion. It aimed to reduce the evaluation apprehension and
social desirability bias of the respondents.
Currently enrolled international doctoral students
from only six universities (XJTU, HIT, FU, NU, HIT,
ZU) participated in an online survey conducted through
social networking application named (WeChat). A total
of 470 answers were recorded without data missing; how-
ever, 55 of the total responses (470) were deleted because
the respondents were careless or inattentive, as is appar-
ent by having the same score on almost all questions in
the survey questionnaire. Finally, 415 (88%) out of 470
responses were considered in this study. Detailed demo-
graphics of respondents are shown in Table 1.
Measure
Supportive Supervisor (SS). Supportive supervisors were
measured with the 4-item scale adopted from (Parker
et al., 2006). A sample item is included, ‘‘My supervisor
encourages us to expect a lot from ourselves,’’ and ‘‘My
supervisor encourages us to be aware of our level of per-
formance.’’ All items were assessed on a seven-point
Likert scale, ranging from 0 (strongly disagree) to 6
Figure 1. The empirical model of this study.
6SAGE Open

(strongly agree). The Cronbach’s afor this scale was
0.83.
Academic Engagement (AE). With its three dimensions
(vigor, dedication, and absorption), academic engage-
ment was measured using the 9-items Utrecht Work
Engagement Scale (Schaufeli et al., 2006). A sample item
included, ‘‘When I am doing my work as a student, I feel
bursting with energy.,’’ ‘‘My studies inspire me.’’ and ‘‘I
am immersed in my studies.’’ All items were rated on a
seven-point Likert scale that ranged from 0 = ‘‘strongly
disagree’’ to 6 = ‘‘strongly agree.’’ The Cronbach’s afor
this scale was 0.93.
Academic Psychological Capital (APC). In an academic
context, psychological capital was measured with the 12-
item scale adapted version of the Psychological Capital
Questionnaire (Avey et al., 2011). A sample item is
included ‘‘I feel confident contributing to discussions
about strategies in my studies.’’ ‘‘I can think of many
ways to reach my current goals regarding my studies.’’ ‘‘I
am optimistic about what will happen to me in the future
as it pertains to my studies.’’ All items were rated on a
seven-point Likert scale ranging from 0 (strongly dis-
agree) to 6 (strongly agree). The Cronbach’s afor this
scale was 0.94.
Research Productivity (RP). RP and its operationalization
are scholars’ primary concerns, as it has both (tangible
and intangible) scientific outcomes (Abramo &
D’Angelo, 2014). Instead of pointing to the number of
publications using WoS and Scopus, we prefer to use a
9-item scale to assess the RP adapted by (Kozhakhmet
et al., 2022), developed by (Kahn & Scott, 1997). The
scale covers a wide range of research activities (i.e., pub-
lications, conferences, and data gathering). A sample
item is included ‘‘A total number of published articles as
authored or co-authored in refereed journals.’’ ‘‘How
many presentations you have made in research confer-
ences or conventions locally, regionally, or internation-
ally.’’ The Cronbach’s afor this scale was .93.
Control Variable. Age, gender, major, country, and year
in current degree programs have been demonstrated to
impact research productivity (Khuram et al., 2022; Horta
et al., 2018); therefore, these control variables are consid-
ered and included in our study.
Data Analysis
SPSS 23 and AMOS 23 have been used to analyze the
data in this research study. The analysis was performed
in two steps: initially, confirmatory factor analysis
(CFA) is performed to measure the scale items underly-
ing hypothesized latent variables (Kline, 2015), and then
structural equation modeling (SEM) tests are performed
to test the hypothesized relationships.
Measurement Model Assessment
Based on Hair et al. (2006), a model measurement was
used to confirm the reliability and validity of constructs.
All the 34 indicators of this study were determined to be
intact from exclusion, as the factor’s loadings were found
greater than recommended 0.60 value. Furthermore, the
KMO Kaiser-Meyer-Olkin (KMO) measure of sampling
adequacy is 0.944, p\.05; Chi-square = 9795.415, p
\.05, while conducting a factor analysis and eigenvalues
are reported higher than 1. Table 2 exhibits the factor
loadings of constructs with mean, standard deviation,
KMO, and eigenvalues.
Confirmatory Factor Analysis
The CFA results demonstrate that the hypothesized
four-factor model was appropriate for the data. The
results of the CFA indicated an excellent model fitness
with the results of (X
2
= 933.78, df =521, X
2
/df = 1.82,
p\.001, CFI = 0.96, TLI = 0.95, SRMR = 0.03,
RMSEA = 0.04). The standardized factor loadings
results were found higher than 0.7. After that, the com-
posite reliability, convergent validity, and discriminant
validity were evaluated in four latent variables using
Fornell and Larcker (1981). As a result, CR values were
found higher than 0.7 for all constructs, indicating a high
Table 1. Demographic Profile of Respondents.
Measure Items Frequency Percentage
Gender Male 247 59.5
Female 168 40.5
Age 20–25 129 31.1
26–30 157 37.8
31–35 111 26.7
36–40 13 3.1
41 years and above 5 1.2
University Fudan University 83 20
Harbin Institute of
Technology
92 22.2
Nanjing University 62 14.9
Shanghai Jiao Tong
University
59 14.2
Xian Jiao Tong
University
61 14.7
Zhejiang University 58 14
Year Freshman 62 14.9
Second Year 113 27.2
Third Year 184 44.3
Fourth Year 49 10.6
Fifth Year 7 1.5
Khuram et al. 7

level of internal consistency (Bagozzi, 1983; Fornell &
Larcker, 1981). The convergent validity was confirmed
by the average variance extracted (AVE). AVE values
for all constructs are higher than 0.5, indicating no con-
vergent validity issues for these constructs. The AVE val-
ues of all models have a 0.5 significance (Sarstedt et al.,
2016). Average variance extracted (AVE) values confirm
the convergent validity. In order to achieve convergent
validity among the constructs, the AVE values should be
greater than 0.5 (Sarstedt et al., 2016). For all constructs,
AVE values were higher than 0.5 and reported no con-
vergent validity issues between constructs of the study.
The method by Fornell and Larcker (1981) was used for
verifying the discriminant validity. According to stan-
dard standards, the AVE square root value must be
higher than the correlation value of all the constructs, as
demonstrated in Table 3.
Model Fit
After confirming the scale’s reliability and validity, we
move next to check the model’s fitness. To assess and
analyze the adequacy of the structural model, well-
known fit statistics four models containing X
2
, the root
mean square error of approximation (RMSEA), the
Tucker–Lewis index (TLI), and the comparative fit index
(CFI), have been applied as a model fit indicator (Kline,
2015). The model fitness is considered acceptable when
the test results indicate that the value of X
2
is significant,
CFI and TLI values are higher than 0.9, and RMSEA
Table 2. Item Loading and Mean.
LMSDKMO Eigenvalues
Supportive Supervisor 3.94 1.5 8.04 2.69
SS_1 0.759
SS_2 0.785
SS_3 0.774
SS_4 0.712
Academic Engagement 4.47 1.21 0.95 6.061
AE_1 0.802
AE_2 0.716
AE_3 0.744
AE_4 0.815
AE_5 0.811
AE_6 0.833
AE_7 0.814
AE_8 0.816
AE_9 0.802
Academic Psychological Capital 4.68 0.97 0.948 5.686
ACP_1 0.760
ACP_2 0.778
ACP_3 0.824
ACP_4 0.756
ACP_5 0.773
ACP_6 0.815
ACP_7 0.845
ACP_8 0.814
ACP_9 0.764
ACP_10 0.864
ACP_11 0.847
ACP_12 0.726
Research Productivity 4.45 1.17 9.50 5.58
RP1 0.786
RP2 0.723
RP3 0.732
RP4 0.811
RP5 0.802
RP6 0.796
RP7 0.790
RP8 0.812
RP9 0.773
8SAGE Open

should be less or equal to 0.08 (Kline, 2015). We per-
formed a series of model comparison tests to confirm the
hypothesized model based on the indications and metho-
dology used (Ali, Li, Khan, et al., 2021; Ali, Li, Durrani,
et al., 2021). The (M0) model values showed a good fit-
ness result (X
2
= 881.116, df = 459, RMSEA = 0.047,
CFI = 0.953, and TLI = 0.949). We eliminated the direct
path between SS and RP in the alternative model (M1);
however, this model results also displayed good fit val-
ues, which indicate its model fitness (X
2
= 881.116, df =
459, RMSEA = 0.047, CFI = 0.953, and TLI = 0.949).
Compared to the hypothesized model, X
2
increased by
0.006, CFI reduced by 0.003, and TLI reduced by 0.001,
whereas the degree of freedom added 1. We have
detached the mediating role and have removed the path-
ways from APC and AE to RP in the second alternative
model (M2). The model was also a good fit
(X
2
= 885.122, df = 461, RMSEA = 0.048, CFI = 0.951,
and TLI = 0.948). Compared to the M0 model, the X
2
values increased by 13.029, the DF improved by 2, and
CFI and TLI reduced by 0.002 and 0.001; this indicates
a decreasing trend for CFI and TLI. We have also
sequentially regressed all the variables in the third alter-
native model (M3). The results suggest that the values of
this model (M3) were not better fit as a model (M0) M0
(X
2
= 873.881, df = 461, RMSEA = 0.047, CFI = 0.954,
and TLI = 0.950). Accordingly, the (M0) was observed
the better among alternative (M1, M2, M3) models (see
Table 4).
Structural Model Testing
The reliability and validity of the measurement model
have been established; a structural model testing, the sub-
sequent step was to test the structural model (Hair et al.,
2013) to determine the model’s productiveness and its
association with proposed structures. Structural model
testing in this analysis was conducted on two steps mod-
els. Only control variables are entered in Model 1 in the
first step; in the second step, main and control variables
are entered in Model 2. The main effects analysis results
showed that the supportive supervisor was significantly
related to RP (b= .13, p\.001), supporting H1 of the
study. However, test results of supportive supervisors in
H2a, and H3a were shown as a significant predictor of
AE (b= .22, p\.001) and academic psychological capi-
tal (b= .16, p\.001). The test results of H2b AE
(b= .12, p\.001) and H3b academic PsyCap (b= .12,
p\.001) also displayed a significant association with
research productivity. It is vital to note that the results
also indicate that controls were insignificantly linked to
research productivity; these findings have not influenced
the primary variable’s relationships (see Table 5).
Further, the relationship between H2c and H3c analy-
sis is conducted through the bootstrapping method
(Preacher & Hayes, 2008); the bootstrapping test helps
identify the mediating effects of sample distribution
skewed from 0 (Shrout & Bolger, 2002). Besides, Model
4 in PROCESS has been adopted to test the mediation
effects (Preacher & Hayes, 2008). It is estimated that
Table 4. Model Comparison.
Model X
2
df X
2
/df CFI TLI RMSEA
SS-RP, SS-APC-RP, SS-AE-RP (M0) 881.116 459 1.92 0.953 0.949 0.047
SS-APC-RP, SS-AE-RP (M1) 885.122 460 1.95 0.950 0.948 0.047
SS-RP, SS-APC, SS-AE (M2) 894.145 461 1.94 0.951 0.948 0.048
SS-APC-AE-RP (M3) 873.881 461 1.89 0.954 0.950 0.047
Note. SS = Supportive Supervisor; APC = Academic Psychological Capital; RP = Research Productivity; AE = Academic Engagement.
Table 3. Descriptive Statistics and Correlation.
CR AVE aAE APC RP SS
AE 0.92 0.57 .93 0.76
APC 0.94 0.63 .94 0.345 0.79
RP 0.93 0.62 .93 0.175 0.23 0.79
SS 0.80 0.56 .83 0.260 0.29 0.193 0.75
Note. The bold diagonal letters indicate the AVE’s Square root. AVE = average variance extracted; a= Cronbach’s alpha; CR = composite reliability;
SS = Supportive Supervisor ; AE = Academic Engagement; APC = Academic Psychological Capital; RP = Research Productivity. Variance extracted is on the
diagonal: Correlations are off-diagonal.
Khuram et al. 9

95% of the confidence intervals (CI) are bootstrapped
with a sample of 5,000 data. According to the conven-
tional standard of significance, upper and lower bound
findings remove 0 for AE and academic PsyCap. The
results of the bootstrap, as demonstrated in Table 6,
shows a positive mediation effect of AE between the sup-
portive supervisor and the RP (b= .03, SE = 0.01, p
\.05, CI 95% [0.007, 0.053]). Additionally, academic
PsyCap has significantly and positively mediated the
relationship between SS and RP (b= .02, SE = 0.01, p
\.05, 95% CI [0.007, 0.087]). These results show the sig-
nificant direct relationship between supportive supervisor
and research productivity, which indicates that media-
tion effects of AE and academic PsyCap partially affect
the supportive supervisor and RP (see Table 6).
Discussion
The current study’s findings are consistent with previous
studies, as discussed in the literature review section. The
current research has examined the relationship between
supportive supervisors and RP with the mediating role
of academic engagement and academic psychological
capital. The data was collected from international stu-
dents enrolled in doctoral degree programs in China.
The results have shown that a supportive supervisor was
directly and indirectly (via academic engagement and
academic psychological capital) related to research pro-
ductivity. The present study addresses the significant
gaps found in the literature by conducting an empirical
study to determine the relationship between the suppor-
tive supervisor and research productivity. However, the
relationship between other elements (e.g., creativity and
innovation persuasion) and supportive supervision has
already been explored in the academic context. The cur-
rent study results suggest that supportiveness should be
a vital characteristic of the doctoral supervisor to carry
out the research projects successfully, thereby supporting
the previous study (Ali et al., 2020; Lindqvist, 2018).
Moreover, given the supervisor’s role in international
doctoral students’ research productivity, this study advo-
cated the definite link between supervisor-student rela-
tionships in research-oriented practices (Franke &
Arvidsson, 2011). Furthermore, the study follows a
Table 5. Structural Equation Model Path Analysis Results.
Mode-1 Model-2
Structural path bSE t-value bSE t-value
Age !RP .02 0.07 0.29 .01 0.08 0.11
Country !RP .07 0.04 2.08 .07 0.04 2.08
Gender !RP –.21 0.13 –1.58 –.17 0.13 21.35
Major !RP .02 0.06 3.14 .01 0.06 2.75
Year !RP .04 0.07 0.64 .03 0.07 0.46
University !RP –.014 0.03 –0.36 –.06 0.04 20.15
SS!RP .13*** 0.05 2.36
SS !AE .22*** 0.04 4.88
SS !ASC .16*** 0.03 5.46
AE !RP .12*** 0.05 2.11
ASC !RP .20** 0.09 2.17
R
2
.34
a
˜R
2
0.4
Note. Model 1: Controls were regressed on the dependent Variable (Note: no other variable was added to analysis). Model 2: Complete model was run,
and the dependent variable was controlled. SE = Standard error; SS = Supportive Supervisor; APC = Academic Psychological Capital; RP = Research
Productivity; AE = Academic Engagement
*p\.05. **p\.01. ***p\.001.
Table 6. Mediation Estimation Effects.
Percentile 95% CI
Standardized estimate (b) Standard error (SE) Lower bound, Upper bound Significance level (p)
SS !AE !RP .03 0.01 0.007, 0.053 \.05
SS !ASC !RP .02 0.01 0.007, 0.087 \.05
Note. SS = Supportive Supervisor; APC = Academic Psychological Capital; RP = Research Productivity; AE = Academic Engagement.
10 SAGE Open

previous research call to investigate the effect of suppor-
tive supervisors on international doctoral students’ inno-
vation and research performance (Fan et al., 2019).
The current research also shows that AE partially
mediates the relationship between supportive supervisors
and research productivity. Apart from the direct effect, a
supportive supervisor increases performance and produc-
tivity by engaging students in research activities. These
findings confirm the main argument of the previous
assertion that a supportive supervisor is unlikely to affect
doctoral students’ research performance unless they are
deeply engaged in research-oriented activities (Ahmed
et al., 2017). This finding also shows that when students
experience supervisors’ supportive behavior during their
interaction, they become intrinsically motivated and
enjoy engaging themselves in research activities under
supportive supervisors (Yidong & Xinxin, 2013). Thus,
Students feel devoted and motivated to the high perfor-
mance of research projects (Mainhard et al., 2009).
Overall, the results obtained in this study indicate that
doctoral students’ AE is a significant factor in RP that
can be enhanced and achieved by having a highly suppor-
tive doctoral supervisor. Similarly, we also found similar
mediation effects of academic PsyCap between suppor-
tive supervisors and research productivity. Although the
mediating effect was partial, indicating that a supportive
supervisor could directly and indirectly increase and
enhance RP through academic PsyCap.
Accordingly, the present study also contributes to pre-
vious research indicating a more substantial impact of
supervisors’ supportive behavior on students’ psychologi-
cal capital (Ahmed et al., 2017). Considering the suppor-
tive supervisor’s role in the supervisee’s AE (Kahu et al.,
2015) and academic PsyCap (Ahmed et al., 2017), this
study takes further steps by integrating the two processes
from a single perspective. It suggests that a supportive
supervisor may promote AE and academic PsyCap to
motivate supervisees toward productivity in research.
Practical Implication
The study has several practical implications. This study’s
findings underline the importance of the supervisor’s
supportiveness for RP in higher education settings.
Supervisors’ supportive behavior is essential in supervi-
sion that could be improved and learned (Gu et al.,
2015). Therefore, this suggests that higher education
institutions, mainly research-oriented ones, should
encourage supervisors’ supportive behavior from their
surroundings, colleagues, and supervisees by taking spe-
cific measures. For example, such research-oriented
HEIs should train both (senior and junior) supervisors to
promote supportive behavior in supervising international
students by designing and organizing different training
sessions (Dangel & Tanguay, 2014). Supportiveness is an
interpersonal relation-oriented attribute of a person.
Therefore, academics can improve social interaction
among international students and supervisors to
exchange and share knowledge (i.e., scholarly expertise,
knowledge) formally and informally. The results also
reveal that a supportive supervisor improves students’
AE and ultimately leads to research productivity. The
supportive supervisor gives academic freedom to their
subordinates, strengthening their self-confidence and
self-efficacy to manage their knowledge and engage in
tasks that are important for achieving the research objec-
tives. Academic engagement, therefore, plays an essential
part in defining the individual’s connectedness with their
takes, reflecting a student’s mental endurance, enthusi-
asm, and strength to overcome difficulties (Bakker et al.,
2008). As such, doctoral students are encouraged to be
more confident performers. This further implies that
higher productivity is achievable when the elements of
academic engagement are used appropriately. The litera-
ture has suggested a supportive environment cultivated
by a supervisor to provide social and psychological
resources that influence an individual’s psychological
state of engagement, leading to performance and effi-
ciency (Swanberg et al., 2011).
As mentioned earlier, supportive supervisors are
friendly and take care of supervisees’ preferences and
satisfaction (Ahmed et al., 2017; Fan et al., 2019),
enabling them to enhance their skills and performance.
A supervisor’s supporting trait encourages supervisees to
develop and utilize their psychological resources during
learning and conducting research activities. Such suppor-
tive behavior enhances the trust and strengthens the rela-
tionship between the supervisee and supervisor, essential
for developing PsyCap resources (Ahmed et al., 2017).
However, psychological resources are used to develop
and manage the performance and assess their resilience
to control and overcome difficulties. Supervisors and
educators should develop a supportive and appreciative
learning environment where students feel confident
about their innovative ideas or present and share freely.
The literature shows that the supervisor and academic
support encourage researchers to do maximum work
and maintain high-quality values, ultimately resulting in
high productivity (Vuong et al., 2019).
Limitations and Future Research
As with every research, this study had some limitations,
which we discussed alongside potential future research
directions. First, the tested model in this study was theo-
retically based; however, its scope is narrow. The only
outcome for a student’s doctoral journey (i.e., RP) was
examined. Although the presence of RP as an outcome
Khuram et al. 11

indicator is consistent with the rationales, the RP reflects
the core outcome of students during their doctoral candi-
dature (Abramo & D’Angelo, 2014) under a supportive
supervision style. However, future scholars are urged to
investigate another supervising style and its influence on
students’ productivity/outcomes. Such future studies will
enable them to investigate distinct relationship trends of
supervision style and performance components and the
mediating role of their psychological factors (i.e., aca-
demic engagement and academic psychological capital)
(Lindqvist, 2018). Similarly, the mediating role of stu-
dents’ academic commitment and adjustment with buf-
fering effects of stressors/strains may also need to be
examined; such factors are asserted to influence learning
significantly. It may be a potential topic for future study
since studies investigate students’ perceptions of supervi-
sion quality and support. Second, this research is a
cross-sectional, designed study that is broadly applied in
higher education research. However, given the psycholo-
gical factors (i.e., academic engagement and PsyCap)
may change with time and environment, a longitudinal
research design may uncover some interesting findings.
Third, although CFA has identified four different
constructs, the potential common method variance can-
not be ignored due to the study’s design (cross-sectional).
Nevertheless, we attempted to address this issue using
Harman’s one-factor test (Podsakoff et al., 2003). The
test results indicated the 22.9% variance reported in
single-factor tests below 50% of the threshold (Ali et al.,
2020). They suggested that the common method bias is
not a severe problem. However, future researchers can
use longitudinal design and multi-sources for data collec-
tion to address the potential biases connected with the
data of cross-sectional studies. Fourth, we only choose
international doctoral students studying in Chinese uni-
versities as a respondent to measure the supervisor’s sup-
portive behavior during doctoral candidature. Doctoral
supervisors and supervisees may have different expecta-
tions and perceptions regarding doctoral supervision. In
this sense, both supervisors and students can be included
from other Chinese HEIs in future studies and further
improve our understanding.
Moreover, researchers in future studies can replicate
the current study from any other (i.e., cultural and orga-
nizational) perspectives. In particular, as supporting
supervision is a relation-oriented supervision style, more
research studies could be done to assess the impact of
RP compared with culture to culture from high to low
and relation-oriented supervision to the task. It will be
interesting to see if supportive supervision contributes to
opposite outcomes; Fan et al. (2019) stated that more
support could affect students’ independence and perfor-
mance. Accordingly, the literature indicates that most
empirical studies on supportive behavior demonstrate
positive effects (Gu et al., 2015). However, it is uncertain
whether supportive behavior can cause adverse out-
comes, such as slower or underperformance of the
followers.
Finally, it is worth noting that the present study’s
sample size was not predetermined. As sample size may
influence the heftiness of findings and results; therefore,
future scholars are urged to use the priori power analysis
method for determining an adequate sample.
Conclusion
This study established that a supportive supervisor is sig-
nificantly influencing the RP of doctoral students.
Moreover, the current study’s findings are consistent
with previous studies and revealed that SS could develop
supervisees’ capabilities by increasing their psychological
factors (i.e., AE and PsyCap) (Ahmed et al., 2017).
Therefore, individuals become more resilient and
engaged in handling academic challenges by taking tasks
seriously and performing scholarly activities. The find-
ings encourage HEI’s authorities to design training ses-
sions for supervisors to develop their supervising skills
that improve the quality of supervision, doctoral educa-
tion, and students’ development (Halse & Malfroy,
2010). This research showed that a productive doctoral
supervisor is highly supportive through such motivation
pushing the students to a high productivity level. It indi-
cates that supporting supervisors’ quality makes them
useful as their followers get inspiration, recognition, and
appreciation. The research study founds two primary
characteristics of students: academic engagement and
psychological capital with a substantial impact on
research productivity; these attributes thrive under the
guidance of a supportive supervisor.
In sum, all the results reported in this study are aligned
with COR principles (Hobfoll, 1989, 2001) that expand
our understanding of the relationship between supportive
supervisor and student research productivity through par-
allel mediation effects of AE and PsyCap. This research is
noteworthy because it widens the supervision-style litera-
ture in higher education settings and tests the mediating
effects of AE and PsyCap on international doctoral stu-
dents’ research productivity in China.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
12 SAGE Open

Ethical Procedure
The research meets all applicable standards concerning the
ethics of experimentation and research integrity, and the
following is being certified/declared true.
As an expert scientist and co-authors of the concerned field,
the paper has been submitted with full responsibility, fol-
lowing the due ethical procedure. There is no duplicate
publication, fraud, plagiarism, or concerns about animal or
human experimentation.
Data Availability Statement
The datasets collected and analyzed during the current study
are available from the corresponding author upon reasonable
request.
ORCID iDs
Waqas Khuram https://orcid.org/0000-0002-8940-6802
Yanqing Wang https://orcid.org/0000-0002-5961-0482
Mudassar Ali https://orcid.org/0000-0002-8869-2421
Heesup Han https://orcid.org/0000-0001-6356-3001
References
Abramo, G., & D’Angelo, C. A. (2014). How do you define
and measure research productivity? Scientometrics,101(2),
1129–1144. https://doi.org/10.1007/s11192-014-1269-8
Ahmed,U.,Umrani,W.A.,Pahi,M.H.,&Shah,S.M.M.
(2017). Engaging Ph.D. Students: Investigating the role of
supervisor support and psychological capital in a mediated
model. Iranian Journal of Management Studies,10(2), 283–306.
https://doi.org/10.22059/ijms.2017.220219.672364
Ali, M., Li, Z., Durrani, D. K., Shah, A. M., & Khuram, W.
(2021). Goal clarity as a link between humble leadership and
project success: The interactive effects of organizational cul-
ture. Baltic Journal of Management 16, 407–423. https://doi.
org/10.1108/bjm-09-2020-0341
Ali, M., Li, Z., Khan, S., Shah, S. J., & Ullah, R. (2021). Link-
ing humble leadership and project success: The moderating
role of top management support with mediation of team-
building. International Journal of Managing Projects in Busi-
ness,14, 545–562. https://doi.org/10.1108/ijmpb-01-2020-
0032
Ali, M., Zhang, L., Shah, S. J., Khan, S., & Shah, A. M. (2020).
Impact of humble leadership on project success: The mediat-
ing role of psychological empowerment and innovative work
behavior. Leadership & Organization Development Journal,
41(3), 349–367. https://doi.org/10.1108/lodj-05-2019-0230
Amabile, T. M., Schatzel, E. A., Moneta, G. B., & Kramer, S. J.
(2004). Leader behaviors and the work environment for crea-
tivity: Perceived leader support. The Leadership Quarterly,
15(1), 5–32. https://doi.org/10.1016/j.leaqua.2003.12.003
Avey, J. B., Avolio, B. J., & Luthans, F. (2011). Experimentally
analyzing the impact of leader positivity on follower positiv-
ity and performance. The Leadership Quarterly,22(2),
282–294. https://doi.org/10.1016/j.leaqua.2011.02.004
Bagozzi, R. P. (1983). Issues in the Application of Covariance
Structure Analysis: A Further Comment. Journal of Con-
sumer Research,9(4), 449–450. https://doi.org/10.1086/
208939
Bakker, A. B., Schaufeli, W. B., Leiter, M. P., & Taris, T. W.
(2008). Work engagement: An emerging concept in occupa-
tional health psychology. Work and Stress,22(3), 187–200.
https://doi.org/10.1080/02678370802393649
Bandura, A. (1997). Self-efficacy: The exercise of control. Jour-
nal of Cognitive Psychotherapy, 13(2).
Brew, A., Boud, D., Namgung, S. U., Lucas, L., & Crawford,
K. (2016). Research productivity and academics’ concep-
tions of research. Higher Education,71(5), 681–697.
Bui, H. T. M. (2014). Student–supervisor expectations in the
doctoral supervision process for business and Management
Students. Business and Management Education in HE,1(1),
12–27. https://doi.org/10.11120/bmhe.2014.00006
Dangel, J. R., & Tanguay, C. (2014). Don’t leave us out there
alone’’: A Framework for supporting supervisors. Action in
Teacher Education,36(1), 3–19. https://doi.org/10.1080/
01626620.2013.864574
Devine, K., & Hunter, K. H. (2017). PhD student emotional
exhaustion: The role of supportive supervision and self-
presentation behaviours. Innovations in Education and
Teaching International,54(4), 335–344. https://doi.org/10.
1080/14703297.2016.1174143
Fan, L., Mahmood, M., & Uddin, M. A. (2019). Supportive
Chinese supervisor, innovative international students: A
social exchange theory perspective. Asia Pacific Education
Review,20(1), 101–115. https://doi.org/10.1007/s12564-018-
9572-3
Fornell, C., & Larcker, D. F. (1981). Evaluating structural
equation models with unobservable variables and measure-
ment error. Journal of Marketing Research,18(1), 39–50.
Franke, A., & Arvidsson, B. (2011). Research supervisors’ dif-
ferent ways of experiencing supervision of doctoral students.
Studies in Higher Education,36(1), 7–19. https://doi.org/10.
1080/03075070903402151
Gatfield, T., & Alpert, F. (2002). The supervisory management
styles model.
Gong, Y., Huang, J. C., & Farh, J. L. (2009). Employee Learn-
ing Orientation, transformational leadership, and Employee
Creativity: The mediating role of employee creative self-effi-
cacy. Academy of Management Journal,52(4), 765–778.
https://doi.org/10.5465/amj.2009.43670890
Gonza
´lez-Ocampo, G., & Castello
´, M. (2019). How do doctoral
students experience supervision? Studies in Continuing Edu-
cation,41(3), 293–307. https://doi.org/10.1080/0158037x.
2018.1520208
Gruzdev, I., Terentev, E., & Dzhafarova, Z. (2020). Superhero
or hands-off supervisor? An empirical categorization of PhD
supervision styles and student satisfaction in Russian univer-
sities. Higher Education,79(5), 773–789. https://doi.org/10.
1007/s10734-019-00437-w
Gu, J., He, C., & Liu, H. (2015). Supervisory styles and graduate
student creativity: The mediating roles of creative self-efficacy
and intrinsic motivation. Studies in Higher Education,42(4),
721–742 https://doi.org/10.1080/03075079.2015.1072149
Khuram et al. 13

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., &
Tatham, R. (2006). Multivariate data analysis. Pearson Pre-
ntice Hall.
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2013). Partial least
squares structural equation modeling: Rigorous applica-
tions, better results and higher acceptance. Long Range Plan-
ning,46(1-2), 1–12.
Halse, C., & Malfroy, J. (2010). Retheorizing doctoral supervi-
sion as professional work. Studies in Higher Education,
35(1), 79–92. https://doi.org/10.1080/03075070902906798
Hobfoll, S. E. (1989). Conservation of resources: A new attempt
at conceptualizing stress. American Psychologist,44(3),
513–524.
Hobfoll, S. E. (2001). The influence of culture, community, and
the Nested-Self in the Stress Process: Advancing Conserva-
tion of Resources Theory. Applied Psychology,50, 337–421.
Horta, H., Cattaneo, M., & Meoli, M. (2018). PhD funding as
a determinant of PhD and career research performance.
Studies in Higher Education,43(3), 542–570. https://doi.org/
10.1080/03075079.2016.1185406
House, J. S. (1983). Work stress and social support. Addison-
Wesley Series on Occupational Stress.
House, R. J. (1996). Path-goal theory of leadership: Lessons,
legacy, and a reformulated theory. The Leadership Quar-
terly,7(3), 323–352.
Hughes, K., & Coplan, R. J. (2010). Exploring processes link-
ing shyness and academic achievement in childhood. School
Psychology Quarterly,25(4), 213–222.
Johansen, B. T., Olsen, R. M., Øverby, N. C., Garred, R., &
Enoksen, E. (2019). Team supervision of Doctoral Students:
A Qualitative Inquiry. International Journal of Doctoral
Studies,14, 069–084. https://doi.org/10.28945/4177
Johansson, C., & Yerrabati, S. (2017). A review of the litera-
ture on professional doctorate supervisory styles. Manage-
ment in Education,31(4), 166–171. https://doi.org/10.1177/
0892020617734821
Kahn, J. H., & Scott, N. A. (1997). Predictors of research pro-
ductivity and science-related career goals among counseling
psychology doctoral students. The Counseling Psychologist,
25(1), 38–67.
Kahu, E., Stephens, C., Leach, L., & Zepke, N. (2015). Linking
academic emotions and student engagement: Mature-aged
distance students’ transition to university. Journal of Further
and Higher Education,39(4), 481–497.
Khuram, W., Wang, Y., & Khalid, A. (2021a). Moderating
impact of positive emotion between academic stress, and aca-
demic performance of students: A conceptual framework.Pro-
ceedings of INTCESS.
Khuram, W., Wang, Y., & Khalid, A. (2021b). Moderating
impact of supervisor support on knowledge seeking intention
of international students: A conceptual framework [Paper pre-
sentation]. 8th international conference on education and
education of social sciences, Turkey.
Khuram, W., Bhutto, A., & Jabeen, A. (2017). Analyzing the
impact of higher education commission policies on motiva-
tion of the faculty member of Jamshoro Education City.
International Journal of Academic Research in Business and
Social Sciences,7(6), 232–257.
Khuram, W., & Wang, Y. (2018). Sharing knowledge through
sharing advisor in age of sharing economy: A conceptual
model [Conference session]. International Conference on
Economics, Business, Management and Corporate Social
Responsibility (EBMCSR 2018).
Khuram, W., Wang, Y., Anjum, M. A., & Khalid, A. (2022).
Factors affecting academic performance of international
students in China: A Theory of reasoned action approach.
Croatian Journal of Education: Hrvatski c
ˇasopis za odgoj i
obrazovanje,24(3), 831–859. https://doi.org/10.15516/cje.
v24i3.4317
Khuram, W., Wang, Y., Khan, S., & Khalid, A. (2021). Aca-
demic attitude and subjective norms effects on international
doctoral students’ academic performance self-perceptions: A
moderated-mediation analysis of the influences of knowl-
edge- seeking intentions and supervisor support. Journal of
Psychology in Africa,31(2), 145–152. https://doi.org/10.
1080/14330237.2021.1903188
Kline, R. B. (2015). Principles and practice of structural equation
modeling. Guilford publications.
Kozhakhmet, S., Moldashev, K., Yenikeyeva, A., & Nurgabde-
shov, A. (2022). How training and development practices
contribute to research productivity: A moderated mediation
model. Studies in Higher Education,47, 437–449. https://doi.
org/10.1080/03075079.2020.1754782
Lee, A., Legood, A., Hughes, D., Tian, A. W., Newman, A., &
Knight, C. (2020). Leadership, creativity and innovation: A
meta-analytic review. European Journal of Work and Organi-
zational Psychology,29(1), 1–35. https://doi.org/10.1080/
1359432x.2019.1661837
Lindqvist, M. H. (2018). Reconstructing the doctoral publish-
ing process. Exploring the liminal space. Higher Education
Research & Development,37(7), 1395–1408. https://doi.org/
10.1080/07294360.2018.1483323
Luthans, B. C., Luthans, K. W., & Jensen, S. M. (2012). The
impact of business school students’ psychological capital on
academic performance. Journal of Education for Business,
87(5), 253–259.
Luthans, F., Youssef, C. M., & Avolio, B. J. (2007). Psycholo-
gical capital: developing the human competitive edge. Oxford
University Press.
Luthans, F., & Youssef-Morgan, C. M. (2017). Psychological
capital: An evidence-based positive approach. Annual
Review of Organizational Psychology and Organizational
Behavior,4(1), 339–366. https://doi.org/10.1146/annurev-
orgpsych-032516-113324
Luthans, K. W., Luthans, B. C., & Chaffin, T. D. (2019). Refin-
ing grit in academic performance: The mediational role of
psychological capital. Journal of Management Education,
43(1), 35–61.
Mainhard, T., van der Rijst, R., van Tartwijk, J., & Wubbels,
T. (2009). A model for the supervisor–doctoral student rela-
tionship. Higher Education,58(3), 359–373. https://doi.org/
10.1007/s10734-009-9199-8
Martı
´nez, I. M., Youssef-Morgan, C. M., Chambel, M. J., &
Marques-Pinto, A. (2019). Antecedents of academic perfor-
mance of university students: Academic engagement and
14 SAGE Open

psychological capital resources. Educational Psychologist,
39(8), 1047–1067. https://doi.org/10.1080/01443410.2019.
1623382
Mason, S. (2018). Publications in the doctoral thesis: Chal-
lenges for doctoral candidates, supervisors, examiners and
administrators. Higher Education Research & Development,
37(6), 1231–1244. https://doi.org/10.1080/07294360.2018.
1462307
Nguyen, T. D., Cannata, M., & Miller, J. (2018). Understand-
ing student behavioral engagement: Importance of student
interaction with peers and teachers. Educational Research
eJournal,111(2), 163–174.
Overall, N. C., Deane, K. L., & Peterson, E. R. (2011). Promot-
ing doctoral students’ research self-efficacy: Combining aca-
demic guidance with autonomy support. Higher Education
Research & Development,30(6), 791–805. https://doi.org/10.
1080/07294360.2010.535508
O’Keeffe, P. (2020). PhD by publication: Innovative approach
to social science research, or operationalisation of the doc-
toral student .or both? Higher Education Research &
Development,39(2), 288–301. https://doi.org/10.1080/
07294360.2019.1666258
Parker, S. K., Williams, H. M., & Turner, N. (2006). Modeling
the antecedents of proactive behavior at work. Journal of
Applied Psychology,91(3), 636–652.
Peng, H. (2015). Assessing the quality of research supervision in
mainland Chinese higher education. The Quality of Higher
Education,21(1), 89–100. https://doi.org/10.1080/13538322.
2015.1049441
Platow, M. J. (2012). PhD experience and subsequent out-
comes: A look at self-perceptions of acquired graduate attri-
butes and supervisor support. Studies in Higher Education,
37(1), 103–118. https://doi.org/10.1080/03075079.2010.
501104
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff,
N. P. (2003). Common method biases in behavioral research:
a critical review of the literature and recommended remedies.
Journal of Applied Psychology,88(5), 879–903. https://doi.
org/10.1037/0021-9010.88.5.879
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resam-
pling strategies for assessing and comparing indirect effects
in multiple mediator models. Behavior Research Methods,
40(3), 879–891. https://doi.org/10.3758/brm.40.3.879
Qureshi, M. A., Khaskheli, A., Qureshi, J. A., Raza, S. A., &
Yousufi, S. Q. (2021). Factors affecting students’ learning
performance through collaborative learning and engage-
ment. Interactive Learning Environments,31(4), 2371–2391.
https://doi.org/10.1080/10494820.2021.1884886
Sarstedt, M., Hair, J. F., Ringle, C. M., Thiele, K. O., & Guder-
gan, S. P. (2016). Estimation issues with PLS and CBSEM:
Where the bias lies! Journal of Business Research,69(10),
3998–4010.
Schaufeli, W. B., Bakker, A. B., & Salanova, M. (2006). The
measurement of work engagement with a short question-
naire: A cross-national study. Educational and Psychological
Measurement,66(4), 701–716.
Schaufeli, W. B., Martı
´nez, I. M., Pinto, A. M., Salanova, M.,
& Bakker, A. B. (2002). Burnout and engagement in univer-
sity students: A cross-national study. Journal of Cross-Cul-
tural Psychology,33(5), 464–481.
Seligman, M. E. (2006). Learned optimism: How to change your
mind and your life. Vintage.
Shrout, P. E., & Bolger, N. (2002). Mediation in experimental
and nonexperimental studies: New procedures and recom-
mendations. Psychological Methods,7(4), 422–445.
Snyder, C. R., Shorey, H. S., Cheavens, J., Pulvers, K. M.,
Adams, V. H., & Wiklund, C. (2002). Hope and academic
success in college. Journal of Education & Psychology,94(4),
820–826. https://doi.org/10.1037/0022-0663.94.4.820
Swanberg, J. E., McKechnie, S. P., Ojha, M. U., & James, J. B.
(2011). Schedule control, supervisor support and work
engagement: A winning combination for workers in hourly
jobs? Journal of Vocational Behavior,79(3), 613–624. https://
doi.org/10.1016/j.jvb.2011.04.012
Vuong, Q. H., Napier, N. K., Ho, T. M., Nguyen, V. H.,
Vuong, T. T., Pham, H. H., & Nguyen, H. K. T. (2019).
Effects of work environment and collaboration on research
productivity in Vietnamese social sciences: Evidence from
2008 to 2017 Scopus data. Studies in Higher Education,
44(12), 2132–2147.
Xu, L., & Grant, B. (2020). Doctoral publishing and academic
identity work: two cases. Higher Education Research &
Development,39(7), 1502–1515. https://doi.org/10.1080/
07294360.2020.1728522
Yang, W., Hao, Q., & Song, H. (2020). Linking supervisor sup-
port to innovation implementation behavior via commit-
ment. Journal of Managerial Psychology,35(3), 129–141.
https://doi.org/10.1108/jmp-04-2018-0171
Yidong, T., & Xinxin, L. (2013). How Ethical Leadership Influ-
ence Employees’ Innovative Work Behavior: A perspective
of intrinsic motivation. Journal of Business Ethics,116(2),
441–455. https://doi.org/10.1007/s10551-012-1455-7
Youssef-Morgan, C. M., & Luthans, F. (2013). Psychological
Capital Theory: Toward a positive holistic model. Advances
in Positive Organizational Psychology.1, 145–166. https://
doi.org/10.1108/s2046-410x(2013)0000001009
Khuram et al. 15