Studies in Higher Education

ISSN: 0307-5079 (Print) 1470-174X (Online) Journal homepage: https://www.tandfonline.com/loi/cshe20

How training and development practices

contribute to research productivity: a moderated
mediation model

Sanat Kozhakhmet, Kairat Moldashev, Aisulu Yenikeyeva & Assylbek

Nurgadeshov

To cite this article: Sanat Kozhakhmet, Kairat Moldashev, Aisulu Yenikeyeva & Assylbek
Nurgadeshov (2020): How training and development practices contribute to research productivity: a
moderated mediation model, Studies in Higher Education, DOI: 10.1080/03075079.2020.1754782

To link to this article: https://doi.org/10.1080/03075079.2020.1754782

Published online: 29 Apr 2020.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=cshe20
STUDIES IN HIGHER EDUCATION
https://doi.org/10.1080/03075079.2020.1754782

How training and development practices contribute to research

productivity: a moderated mediation model
Sanat Kozhakhmeta, Kairat Moldashevb, Aisulu Yenikeyevac and Assylbek Nurgadeshovd*
a
Department of Doctorate Studies, Business School, Narxoz University, Almaty, Kazakhstan; bBusiness School, Narxoz
University, Almaty, Kazakhstan; cBusiness School, Suleyman Demirel University, Kaskelen, Kazakhstan; dBusiness
School, HR Research Center, Narxoz University, Almaty, Kazakhstan

ABSTRACT KEYWORDS
The use of corporate management practices in higher education has led to Training and development;
spread of result-oriented practices to improve research productivity, research productivity;
including publication rewards and requirements. In this paper we stress knowledge sharing climate;
research self-efficacy;
the importance of input and process based approaches by examining
strategic human resource
the mediating role of research self-efficacy on the relationship between management
training & development practices and research productivity, and the
moderating role of knowledge sharing climate. This empirical work
provides a more complete understanding of how training &
development practices may build research self-efficacy which in turn
promotes faculty members’ research productivity. Finally, we contribute
to the emerging literature on the moderating effects of knowledge
sharing climate.

Introduction
The global higher education sector has changed significantly in the past two decades. Some of these
changes were induced by national governments to improve efficiency and effectiveness in various
fields, including policies to enhance research productivity (Billot 2010; Brew, Boud, and Malfroy
2017). According to Lucas and Murry (2011) research productivity became a requirement for research
institutions, as well as for faculty members in all types of institutions. This attracted scholarly attention
to understanding individual-level research productivity and factors that lead to its increase (Hem-
mings and Kay 2016; Nygaard 2017). In this study we focus our attention on HR practises, particularly
research training and development (T&D) practices. Governments and higher education institutions
(HEIs) in many developing countries devote significant resources to research T&D, including research
methodology workshops and ‘how to publish’ seminars. Previous studies have identified different
organizational factors that may predict faculty members’ research performance (Hedjazi and Behra-
van 2011; Wester et al. 2019). Bland (1992) revealed twelve factors from Productive Research Environ-
ments (PRE), based on meta-analysisincluding research emphasis, group climate, governance
mechanisms, availability of resources, and HR practices. A related concept, Research Training Environ-
ments (RTE) was studied as a predictor of research productivity among doctoral students (Brown et al.
1996; Mallinckrodt and Gelso 2002). On the post-doctoral level, there are a limited number of empiri-
cal studies which show the effectiveness of intervention in the form of research training programs in
improving the research performance (Anandarajah et al. 2016; Konstantakos et al. 2010; Kurahara
et al. 2012; Rothberg et al. 2014).

CONTACT Sanat Kozhakhmet [email protected], [email protected] Narxoz University, Zhandosov

street, 55, postal code 050035, Almaty, Kazakhstan
*Present address: Assistant Professor, Edinburgh Business School, Heriot–Watt University, Edinburgh, UK
© 2020 Society for Research into Higher Education
2 S. KOZHAKHMET ET AL.

In addition to the gap in the literature related to post-doctoral level research training, the recent
meta-analysis of the relevant studies indicates that the direct relationship between research environ-
ment and research productivity might be mediated or moderated by different organizational and
individual-level factors (Ajjawi, Crampton, and Rees 2018). This finding and our survey of the literature
suggest that there is a lack of empirical research regarding the potential role mediators and modera-
tors play between T&D and research productivity. Particularly, there is a need to examine individual-
level variables (e.g. research self-efficacy) as a mediator between T&D practices and research pro-
ductivity (Holttum and Goble 2006) and the boundary conditions of these relationships (Overall,
Deane, and Peterson 2011). The possibility of establishing boundary conditions can be explained
by the research evidence that the relationship between research self-efficacy (RSE) and its antece-
dents is not linear (Ajjawi, Crampton, and Rees 2018).
To address this research gap, we investigate whether T&D practices can affect research pro-
ductivity of faculty members, using the moderated mediation model. More specifically, we investi-
gate the potential mediating role of RSE and of a knowledge sharing (KS) climate, relying on social
exchange theory (SET) and social capital theory (SCT). The aim of the study is to analyze how T&D
practices and KS Climate interact together to predict both RSE and individuals’ research productivity.
This study attempts to add to the literature in several ways. First, the RSE construct is evaluated as a
mediator in the link between T&D practices and research productivity. Although the RSE construct is
becoming popular in educational and behavioral research, there is lack of studies that view it as a
mediator. The second contribution of our work is that it adds to the understanding of SET and
SCT interaction. To the best of our knowledge there is a lack of research that includes T&D practices,
KS Climate, and RSE within the models of reciprocal organizational relationships. Third, this paper will
widen our knowledge of social exchange relationships between individuals and organizations by
treating KS Climate as a moderating factor. Our findings provide credible support for the notion
that social capital theory might complement social exchange theory in explaining faculty
members’ research productivity. This suggests that several theories should be taken into account
as a framework in predicting faculty members’ research productivity.

Literature review and hypotheses development

Previous research works have revealed that human resource management (HRM) practices play a
unique role in facilitating research productivity of faculty members (Beerkens 2013). Studies have
shown that work autonomy (Edgar and Geare 2013), organizational counseling, a reward system
(Hedjazi and Behravan 2011), and other organizational practices may effectively promote research
performance. The major responsibility of HRD professionals is creating opportunities for learning
and development to help employees realize their potential (Peterson and Provo 1998; Swanson
and Holton 2008).
In our research, we follow this line of research by introducing training and development practices
as a mechanism for promoting research productivity. T&D refers to a bundle of activities or practices,
designed by an organization, to enhance employees’ skills, competencies, and knowledge (Kraiger
2003). We narrowed down this broad definition to research-related T&D practices that are intended
to equip employees with skills and competencies that enable them to become more productive in
their academic work. These practices include research workshops, master-classes, seminars, and
short courses for early career and senior researchers. As discussed before, SET is appropriate to
explain the impact of T&D practices on faculty members’ research productivity. Based on this theor-
etical framework, when employees feel that an organization cares about them by investing in T&D,
they are more likely to respond to such investment with desirable work outcomes (Blau 1968). In
other words, employees tend to reciprocate the favorable treatment they receive (Cropanzano and
Mitchell 2005). Other studies have also found that different kinds of T&D activities, formal and infor-
mal, play a crucial role in enhancing research productivity of academic staff (Phillips and Russell 1994;
Wood et al. 2018). Hence, the following hypothesis is posed:
 STUDIES IN HIGHER EDUCATION 3

Hypothesis 1: T&D practices have a positive impact on research productivity

T&D practices and research self-efficacy

In efforts to improve their research performance, higher education institutions invest in enhancing
the research skills and RSE of their faculty members. T&D practises increase research self-efficacy
in several ways. First, these practices make faculty members feel more confident in their abilities
to do research, leading to greater levels of RSE. Second, T&D practices can create collaboration
between faculty members, so that they can accumulate successful experience, which may further
enhance their self-efficacy in research. Third, T&D practices help individuals to gain a sense of
control over their career paths, due to organizational help and support for relearning old skills and
learning new ones (Lee and Bruvold 2003). This leads to the improvement of psychological states
such as self-efficacy and self-esteem. Fourth, T&D practices facilitate an organizational climate in
which feelings of trust, cooperation, and optimism occur (Kuvaas and Dysvik 2009). All of these
may further strengthen faculty members’ self-efficacy in doing their academic work. Overall, the lit-
erature suggests that the training environment had a positive effect on RSE (Phillips and Russell
1994). Previous studies have also reported that PhD students who had better training opportunities
were more likely to have higher RSE and interest in research (Lambie and Vaccaro 2011). Based on
these arguments, the following hypothesis was formulated:
Hypothesis 2: T&D practices have an impact on research self-efficacy

Research self-efficacy and research productivity

Past research on encouraging research productivity has emphasized organizational and individual
level factors (Teodorescu 2000). Recent empirical works on individual-level factors have revealed
that psychological factors are significant in explaining this variable (Hemmings and Kay 2016). There-
fore, it is vital to consider RSE as a potential predictor of research productivity.
In our study, it is assumed that faculty members with higher levels of RSE are more likely to be
productive in their research, for the following reasons. First, ones’ self-efficacy is most affected by
past experience in performing similar activities (Pajares 2002). The successful accomplishment of a
particular job enhances self-efficacy (Bandura 1982). When people repeat their success in completing
a certain task, they have higher confidence in their ability to accomplish this task again. For instance,
faculty members who have successfully published their work may have high research self-efficacy,
compared to those without such a track record. Second, a person highly confident of her/his abilities
is more likely to select challenging activities and complete them successfully (Bandura 1977). Hence,
research achievements of faculty members depend largely on each member’s psychological state.
Overall, past studies have viewed self-efficacy through the prism of social cognition theory
(Bandura 1982), and found it to be a significant predictor of individual outcomes (Bandura, 2001).
In a university context, RSE plays a major role in faculty members’ engagement in research activity
(Hemmings et al. 2012). In light of the above evidence, the following hypothesis is posed:
Hypothesis 3: Research self-efficacy positively affects research productivity

The mediating effect of research self-efficacy

One of the key contributions of our paper is the examination of the indirect effect of T&D practices on
research productivity via RSE. Building on our earlier arguments, we propose that RSE mediates the
association between T&D practices and research productivity.
Support for our hypothesis can be found in the job characteristics model of work motivation input-
process-outcomes (Hackman and Morris 1975). This framework provides insight into how T&D prac-
tices may affect employee-related outcomes via the mediation of psychological constructs (e.g. RSE).
4 S. KOZHAKHMET ET AL.

In the first stage (input), an organization implements T&D practices that are expected to improve
psychological states (e.g. RSE) of faculty members (process) in the second stage. Outcomes are
achieved in the third stage, in terms of enhanced research productivity. Hence, we expect that
T&D practices (input) positively affect RSE (process), which in turn facilitates research productivity
(outcome).
Additional theoretical support for our assumption is derived from SET, which views the employ-
ment relationship as an exchange process between employer and employee (Blau 1968). Employees
reciprocate organizational investment (e.g. T&D) with positive outcomes (e.g. RSE and research pro-
ductivity). In addition to the reciprocity, the sense of belonging can also explain the indirect relation
between T&D practices and research productivity. Investment in staff T&D might be perceived by
employees as a message that their employer cares about their development and professional
growth. It may have a positive impact on individuals’ sense of being an important part of the organ-
ization, which enhances their research self-efficacy. Enhanced RSE leads to active engagement in
research and increases productivity. Finally, a sense of belonging also can be used to explain the pro-
posed indirect relationship between T&D practices and research productivity. Namely, T&D practices
might be perceived as a message to academic staff that their HEIs value them and that the organiz-
ation cares about their professional development. Hence, it has a positive effect on individuals’ sense
of belonging to the organization, which in turn may increase faculty members’ research confidence
and RSE. Further, these positive capacities may help individuals become more productive in their aca-
demic activities. It can be concluded that T&D practices generate positive conditions required for RSE
and research productivity to flourish. These arguments enable us to formulate the following
hypothesis:
Hypothesis 4: Research self-efficacy mediates the link between T&D practices and research productivity

The moderating effect of a knowledge-sharing climate

The term KS refers to donating or collecting knowledge to assist others and to cooperate with others
to generate new ideas (Wang and Noe 2010). According to Connelly and Kevin Kelloway (2003), KS
Climate is related to the organizational environment in which employees see KS as a natural activity
in their daily work. KS climate is supported by personal belief and organizational structures that are
conducive to the free flow of information and knowledge sharing among employees (Bock et al.
2005). The effectiveness of research-related T&D practices can significantly increase if the skills and
knowledge obtained by trainees is shared with other colleagues within their organization.
In this paper, we are arguing that KS Climate plays a moderating role by strengthening the effect
of T&D practices on RSE, for the following reasons. First, when organizations provide employees with
T&D opportunities, alongside creating a work climate where knowledge and experience can be easily
shared, faculty members are more likely to have higher levels of self-efficacy in doing research.
Employees often acquire knowledge though KS activities and those with a greater level of knowledge
are more confident in their abilities to conduct research (Hu and Zhao 2016; Richter et al. 2012). By
contrast, an organization is unlikely to gain from T&D programs when people hoard their knowledge
and are reluctant to share it (Connelly et al. 2012).
The next reason is based on the fact that knowledge is a type of solid reasoning for data and infor-
mation which may improve employees’ learning and decision-making capabilities (Beckman 1999).
Hence, in an organizational climate where knowledge can flow freely, T&D practices have a much
stronger effect on faculty members’ RSE.
Third, in a KS Climate where employees can fearlessly share their knowledge and learn from
others, such positive group dynamics as closeness, cohesiveness, and brotherhood are more
likely to exist (Weinstein and Ryan 2010). In other words, this climate provides a framework for
the development of close social relationships among co-workers. Hence some organizational
members will see KS as opportunity to satisfy their needs for esteem, which may further
 STUDIES IN HIGHER EDUCATION 5

improve their self-efficacy. Therefore, KS Climate has the potential to strengthen the influence of
T&D practices on RSE.
Fourth, when knowledge is shared between co-workers, it becomes a public good and generates
new knowledge (van den Hooff and de Ridder 2004). This new knowledge may be beneficial by
improving the research skills and competencies of faculty members, and thereby further enhancing
their self-efficacy beliefs. Finally, we assume that KS Climate can strengthen T&D practices, because
jointly they send a positive message to employees that the employer cares about them and the work-
place environment, which may lead to improved RSE.
In sum, we hypothesize that the impact of T&D practices on RSE is conditional upon contextual
factors such as KS climate. Namely, we expect the moderating effect to occur between T&D practices
and KS Climate on RSE. Combing these arguments, we propose the following:
Hypothesis 5: The KS Climate moderates the relationship between T&D practices and research self-efficacy

Moderated mediation
In addition to the assumption that T&D practices are likely to promote research productivity through
RSE, it is expected that KS Climate may play a significant role in the relationship between T&D and
research productivity. On the basis of the arguments mentioned above, we suggest the moderated
mediation model which integrates the mediating role of RSE and the moderating role of KS Climate in
the link between T&D practices and research productivity (see Figure 1). In particular, the strength of
the mediated effect increased along with the level of KS climate. This led us to the following
hypotheses:
Hypothesis 6: KS Climate moderates the indirect relationship between T&D practices and research self-efficacy.

Methodology
Data and sample
A total of 600 questionnaires were randomly distributed among science, technology, engineering,
and mathematics (STEM) faculty members across eight leading public and private universities in
Astana and Almaty. These two cities are the main research and educational hubs of Kazakhstan. A
total of 165 questionnaires were returned to us, out of these, 140 responses were used for data analy-
sis; the remaining responses were excluded due to incomplete data or invalid responses, thus the
actual response rate was 23%. Among the faculty members, 43% were male and 57% female. Age
of respondents: 3% were 20–24 years old, 28% were 25–29 years old, 25% were 30–39 years old,

Figure 1. Research framework.

6 S. KOZHAKHMET ET AL.

32% were 40–49 years old, and 12% were over 50. In terms of educational level, almost all of the
respondents had a PhD degree (95%). The majority of the respondents (78%) held non-managerial
positions. The working experience of respondents varied from less than one year to more than
fifteen years.

Measures
Measures for each construct were derived from previously validated instruments. These items were
rated on a Likert scale ranging from strongly disagree to strongly agree.

T&D practices
The five-item scale of T&D practices was adopted from the research of Meyer, Allen, and Smith (1993),
which refers to the degree to which an organization enables employees to further develop their skills
and competencies. The questions in the original survey were slightly modified to refer to research
training. An example survey item is the following: ‘There are lots of (research) training opportunities
provided for me in this university/organization.’ The Cronbach’s α for these items was 0.89.

Research self-efficacy
In this paper, the RSE of faculty members was measured using a nine-item scale developed by Holden
and his colleagues (1999). It measures their confidence in their ability to complete specific research
activities. Each item begins with the phrase ‘How confident are you that you can … ?’ The Cronbach’s
α was 0.87. The sample item included: ‘effectively present your study and its implications?’

Knowledge-sharing climate
This construct was measured by five items, which were all derived from Connelly and Kevin Kelloway
(2003). The sample item included: ‘People with expert knowledge are willing to help others in this
organization.’ Cronbach’s α coefficient for this scale in the present research was 0.73.

Research productivity
Operationalizing research productivity is a subject of debate among scholars, as the scientific output
has both a tangible (e.g. publications, conferences, patents) and an intangible (e.g. consulting activity,
tacit knowledge) nature (Abramo and D’Angelo 2014). Due to possible time lag problems, we avoided
using the number of publications in WoS or Scopus databases, a common practice of ranking
agencies, to assess research productivity in this cross-sectional study. The indicators used to
measure other variables presented in this paper assess the current state of the respondents and
their environments. For consistency among the measures, we used the nine-item scale from Kahn
and Scott (1997) to measure research productivity. This scale covers a broad range of research activi-
ties (publications, conference participation, involvement in data collection and analysis) instead of
focusing only on the number of publications or citations. Example items are ‘How many unpublished
empirical manuscripts have you authored?’ and ‘How many local, regional, or national research con-
ventions have you attended?’ The value of Cronbach’s α for the scale was 0.76.

Data analysis
To analyze our research model, Partial Least Squares (PLS) path modeling was employed. The argu-
ments for using this approach are: (1) this study focuses on maximizing the variance explained in the
dependent variable rather than theory confirmation; (2) the proposed research model is complex,
and considers mediation and moderation (Hair et al. 2016). The SmartPLS 3.0 software was utilized
to perform our analysis. Additionally, the PROCESS macro was used to test the conditional indirect
effect. This macro used bootstrapping technique for testing mediated moderation hypotheses,
 STUDIES IN HIGHER EDUCATION 7

using a resampling procedure of 5000 bootstrap samples (Preacher, Rucker, and Hayes 2007). The
point estimate was considered significant when the confidence interval did not contain zero.

Results
Measurement model
Reliability and validity
Since all data were obtained from a single source, the potential impact of common method variance
(CMV) was addressed using the following steps. Two different post-hoc tests were utilized to assess
the impact of CMV. To minimize the risk of common method variance, Harman’s ex post one-factor
test was performed (Podsakoff and Organ 1986), by including all of the items used in the research into
un-rotated factor analysis. The outcomes revealed that no single factor appears to dominate, and the
biggest factor explained 24% of the variance, thus further confirming that CMV was not pervasive in
this research. Moreover, to reduce test anxiety, participants were assured that there were no right or
wrong answers. Moreover, the dependent variable was measured on a different scale to deal with
similar measurement issues. In sum, measurement bias was minimized by using these approaches.
To assess the measurement model, the reliability, convergent, and discriminate validities were
examined. As a measurement of reliability, Cronbach’s coefficient alpha was calculated for each con-
struct. The value of all constructs varied from 0.73 to 0.89, which is higher than the minimum
threshold of 0.7 (Nunnally and Bernstein 1967). Furthermore, composite reliability (CR) was used to
analyze the reliability of variables. All values were between 0.79 and 0.90, exceeding the threshold
of 0.7 (see Table 1). Similarly, convergent validity was assessed using the method outlined by
Fornell and Larcker (1981). Namely, average variance extracted (AVE) should exceed 0.50 to establish
convergent validity. All values of AVE (see Table 2) are above the advised threshold. In accordance
with Fornell and Larcker (1981), to ensure discriminant validity the square root of AVE should be
more than the correlation between study variables. Our outcomes indicate that discriminant validity
is achieved (see Table 2).

Results
Structural model
After validating the measurement model, SmartPLS 3.0 was used to test our hypotheses. A bootstrap-
ping technique (with 5000 resamples) was applied to examine the significance of the path coeffi-
cients (Chin 1998). In accordance with Preacher, Rucker, and Hayes (2007), bootstrap tests are non-
parametric simulations and are an efficient and reliable approach to evaluate indirect effects for
mediation and moderation models. This method is a better option than the Sobel test, because in
an indirect effect test the standard error estimate often does not follow a normal distribution and
yields a biased p-value (MacKinnon et al. 2002). Figure 2 shows the path coefficients, and significance
level of the endogenous variable. The outcomes point out that T&D practices have a significant

Table 1. Latent variable correlation matrix, internal consistency, and average variance extracted.
AVE
Standard KS Research Research T&D Composite Cronbach’s (square
Mean Deviation Climate self-efficacy productivity practices reliability coefficient AVE root)
KS Climate – 0.84 0.73 0.64 0.8
Research self- 0.27** – 0.80 0.87 0.65 0.8
efficacy
Research 0.17** 0.27** – 0.79 0.76 0.57 0.75
Productivity
T&D Practices 0.11* 0.21** 0.32** – 0.90 0.89 0.86 0.92
Note: t-values > 1.65* (p < 0.1); t-values > 1.96** (p < 0.05); t-values > 2.57*** (p < 0.001).
8 S. KOZHAKHMET ET AL.

Table 2. Testing mediation.

Standardized coefficients (t-values)
Paths Total effects Direct effects Indirect effects
Training -> Research productivity 0.25 (2.63***) 0.27 (2.80***) 0.10 (2.04**)
Training ->Research self-efficacy 0.14 (2.77***)
Research self-efficacy -> Research productivity 0.17 (4.34***)
Note: t-values > 1.65* (p < 0.1); t-values > 1.96** (p < 0.05); t-values > 2.57*** (p < 0.001).

Figure 2. Research model.

Note: t-values > 1.65* (p < 0.1); t-values > 1.96** (p;< 0.05); t-values > 2.57*** (p < 0.001).

impact on research productivity (Hypothesis 1: path coefficient = 2.94, t = 0.24). Also, the findings
revealed that T&D practices have a direct and positive effect on RSE (Hypothesis 2: path coefficient
= 2.53, t = 0.14), and RSE (Hypothesis 3: path coefficient = 0.15, t = 3.72) has positive impact on
research productivity.

Mediating role of RSE

As is shown in Table 2, the total effect is significant (β = 0.25; t = 2.63). The findings suggest the exist-
ence of both a direct and an indirect effect (mediation) (β = 0.10; t = 2.04), since the direct effect of
T&D practices on research productivity after introducing RSE is significant (β = 0.27; t = 2.80). In
sum, RSE is found to have a partial mediation effect on the association between T&D practices
and research productivity.

Moderating role of KS climate

In our analysis, the moderating variable (KS climate) was entered into the model, and the moderating
impact was added to the mediating variable (RSE). Then, bootstrapping was employed to examine
the statistical significance of the moderating effect. The outcomes of the moderation analysis are pre-
sented in Table 3. In line with our expectations, KS Climate increases the impact of training on RSE.
Therefore, Hypothesis 5 is fully supported.

Table 3. KS Climate as a moderator.

Hypothesis Relationship Interaction term (β) Interaction term (T- statistics)
H5 Trainings -> research self-efficacy 0.42 3.89***
Note: t-values > 1.65* (p < 0.1); t-values > 1.96** (p < 0.05); t-values > 2.57*** (p < 0.001).
 STUDIES IN HIGHER EDUCATION 9

Table 4. Conditional indirect effects results.

Outcome Indirect effect Index SE (Boot) 95% CI Lower limit 95% CI Upper limit
Research productivity 0.40 0.84 0.17 0.14 0.90
Note: Bootstrapping based on n = 5000 subsamples.

Moderated mediation
In order to examine moderated mediation, we have followed the suggestion of Hayes (2013). Con-
sistent with our expectations, KS Climate significantly moderates the link between T&D practices
and RSE (see Table 3). Next, Model 7 of the PROCESS macro was used to assess the conditional indir-
ect effect of T&D practices on research productivity when RSE is the mediator and KS Climate is the
moderator. PROCESS uses a bias-corrected 95% bootstrap confidence interval (CI) for the indirect
effect based on 5000 bootstrapped samples. If the confidence interval does not overlap zero, it
means that the indirect effect is significant. Consistent with our assumptions, the results support
Hypothesis 6, due to a significant indirect effect of 0.15 (Table 4).

Discussion and conclusion

This study followed the suggestion in the recent meta-analysis of literature in education research of
the necessity for further research on the interplay between mechanisms and context in fostering
research productivity (Ajjawi 2018). We added to the literature on the Research Training Environ-
ment (Brown et al. 1996; Mallinckrodt and Gelso 2002) by bringing in insights from HRM literature
with a focus on research-related T&D practices provided for academic staff at the post-doctoral
level. Based on the social exchange and social capital theories, a moderated mediation model
was specified to better understand the mediating and moderating mechanisms through which
T&D practices affect faculty members’ research productivity. Firstly, it was revealed that
research-related T&D practices are positively associated with research productivity. These
findings support the literature which states that T&D practices create the necessary conditions
for enhancement of individuals’ research productivity (Anandarajah et al. 2016; Konstantakos
et al., 2010; Kurahara et al., 2012; Rothberg et al. 2014). Secondly, it was found that RSE plays a
crucial role in bridging organizational interventions and research productivity. In other words,
T&D practices may enhance faculty members’ research productivity through promoting RSE,
which adds to the literature by indicating that RSE mediates this relationship (Lambie and
Vaccaro 2011; Lee and Bruvold 2003). Hence, it can be argued that to facilitate research pro-
ductivity, HEIs first need to utilize T&D practices to build individuals’ RSE. At the same time, to
get a more complete picture of factors that contribute to faculty members’ research productivity,
KS Climate was considered as a moderating construct. In line with our expectations, it was revealed
that KS Climate moderates the path between T&D practices and RSE. Further, a moderated
mediation effect was confirmed using a bootstrapping technique.
There are some plausible explanations for these outcomes. While T&D practices are more formal
ways of transferring knowledge, informal and spontaneous ways of sharing knowledge are also
important (Hu and Zhao 2016). Therefore, an organizational climate in which individuals feel free
to share their knowledge and ideas with others multiplies the effect of T&D practices on RSE and
research productivity. It is particularly important for the post-Soviet Kazakhstani context, where
such counterproductive phenomenon as ‘knowledge sharing hostility’ is apparent (Kozhakhmet
and Nazri 2017).

Theoretical implications
This manuscript makes several contributions to the educational literature on HRM. First, this paper
reinforces theoretical and empirical foundations for promoting faculty members’ research
10 S. KOZHAKHMET ET AL.

productivity from a human resource development perspective. Our results are consistent with
past research which reveals that certain HRM practices are effective tools for improving employee
outcomes (Beerkens 2013). More specifically, we found that T&D practices provide academic staff
with opportunities to improve their current skills or acquire new ones, making it possible to
enhance their research performance (Phillips and Russell 1994; Wood et al. 2018). The second
contribution is related to confirming the mediating role of RSE between T&D practices and
research productivity. It adds to the literature about the mechanism behind the impact of HRM
practices on various individual outcomes (Boxall, Hutchison, and Wassenaar 2015), and also
broadens the empirical basis of research on RSE. Finally, we add to the educational literature
on the Research Environment (Bland 1992; Ajjawi, Crampton, and Rees 2018) and Research Train-
ing Environment (Brown et al. 1996; Mallinckrodt and Gelso 2002) by integrating concepts from
HRM and organizational theory. The study broadens the literature on productive research
environments by exploration of the role of the KS Climate in influencing individuals’ RSE, and
by examining the moderating effect of KS Climate on the links between T&D practices–RSE–
research productivity. As hypothesized, we confirm moderation and moderated mediation
effects of the KS climate.

Practical implications
The corporate approach to university management has emphasized top-down approaches such as
minimum publication requirements and the monetary incentives of publishing, often at the
expense of investing in KS Climate and T&D of academic staff. This research has several practical
implications that argue for alternative HRM strategies in place of currently dominant ‘publish or
perish’ managerial practices. Our results suggest that T&D practices play a significant role in
improving RSE and the research productivity of academic staff. Particularly in non-anglophone
and developing countries such as Kazakhstan, scholars may need additional training and re-learn-
ing of research skills in order to fully join the global scientific community. Alongside investing in
T&D, HRM practitioners and executives in higher education should consider that the returns on
these investments depend on the psychological state (e.g. RSE) of their faculty members. The
RSE levels can be monitored and conditions for improvement should be created. KS Climate is
found to be an important condition that moderates the link between T&D practices and RSE.
Thus, HRM practices that enhance the KS Climate should be emphasized to improve research
productivity.

Limitations and future research

Several limitations and directions for future studies should be considered. The scope of this study is
limited to the Kazakhstani context. It would be interesting to replicate our findings in other settings.
Next, all surveys were based on single respondent and self-reported data, which may cause
common method bias. Despite the fact that we took required steps to address this issue, further
studies should collect data from multiple sources. In this paper, we have used cross-sectional
data which provide no evidence of causality or directionality. Hence, future research with longitudi-
nal data is required to address this limitation. Finally, the general scope of this article was to
examine the effect of T&D practices and RSE on research productivity. As an extension, our
study analyzes the effect of T&D practices, RSE, and KS Climate on research productivity only. In
the future, scholars may also examine other outcomes of research productivity to further clarify
the proposed framework.

Disclosure statement
No potential conflict of interest was reported by the author(s).
 STUDIES IN HIGHER EDUCATION 11

Funding
This work was supported by This study is part of the project “Integration of local researchers into the global scientific
community: The challenges and possible solutions in the case of Kazakhstan” (IRN: AP05133888) funded by the Govern-
ment of Kazakhstan under budget program 217 “d: [Grant Number IRN: AP05133888].

References
Abramo, G., and C. A. D’Angelo. 2014. “How Do You Define and Measure Research Productivity?” Scientometrics 101 (2):
1129–1144. doi:10.1007/s11192-014-1269-8.
Ajjawi, Rola, Paul ES Crampton, and Charlotte E Rees. 2018. “What Really Matters for Successful Research Environments? A
Realist Synthesis.” Medical Education 52 (9): 936–50. doi:10.1111/medu.13643.
Ajjawi, R., P. E. Crampton, and R. E. Rees. 2018. “What Really Matters for Successful Research Environments? A Realist
Synthesis.” Medical Education 52 (9): 936–950. https://doi.org/10.1111/medu.13643.
Anandarajah, G., C. Furey, R. Chandran, A. Goldberg, F. El Rayess, D. Ashley, and R. E. Goldman. 2016. “Effects of Adding a
New PCMH Block Rotation and Resident Team to Existing Longitudinal Training Within a Certified PCMH: Primary Care
Residents’ Attitudes, Knowledge, and Experience.” Advances in Medical Education and Practice 7: 457. 10.2147.
Bandura, Albert. 1977. “Self-Efficacy: Toward a Unifying Theory of Behavioral Change.” Psychological Review 84 (2): 191–
215. doi:10.1037//0033-295x.84.2.191.
Bandura, Albert. 1982. “Self-efficacy Mechanism in Human Agency.” American Psychologist 37 (2): 122–47. doi: 10.1037/
0003-066X.37.2.122.
Bandura, A. 2001. “Social Cognitive Theory: An Agentic Perspective.” Annual Review of Psychology 52 (1): 1–26. https://doi.
org/10.1146/annurev.psych.52.1.1.
Beckman, Thomas J. 1999. “The Current State of Knowledge Management.” Knowledge Management Handbook 1 (5): 1.
Beerkens, Maarja. 2013. “Facts and Fads in Academic Research Management: The Effect of Management Practices on
Research Productivity in Australia.” Research Policy 42 (9): 1679–93. doi:10.1016/j.respol.2013.07.014.
Billot, Jennie. 2010. “The Changing Research Context: Implications for Leadership.” Journal of Higher Education Policy and
Management 33 (1): 37–46. doi:10.1080/1360080x.2011.537010.
Bland, C. J. 1992. “Characteristics of a Productive Research Environment: Literature Review.” Academic Medicine: Journal of
the Association of American Medical Colleges 67 (6): 385–397. https://doi.org/10.1097/00001888-199206000-00010.
Blau, Peter M. 1968. “Social Exchange.” In International Encyclopedia of the Social Sciences, Vol. 7, 452–457. doi:10.1002/ncr.
4100570915.
Bock, G. W., R. W. Zmud, Y. G. Kim, and J. N. Lee. 2005. “Behavioral Intention Formation in Knowledge Sharing: Examining
the Roles of Extrinsic Motivators, Social-psychological Forces, and Organizational Climate.” MIS Quarterly 87–111.
doi:10.2307/25148669.
Boxall, Peter, Ann Hutchison, and Brigitta Wassenaar. 2015. “How Do High-Involvement Work Processes Influence
Employee Outcomes? An Examination of the Mediating Roles of Skill Utilisation and Intrinsic Motivation.” The
International Journal of Human Resource Management 26 (13): 1737–52. doi:10.1080/09585192.2014.962070.
Brew, Angela, David Boud, and Janne Malfroy. 2017. “The Role of Research Education Coordinators in Building Research
Cultures in Doctoral Education.” Higher Education Research & Development 36 (2): 255–68. doi:10.1080/07294360.2016.
1177812.
Brown, S. D., R. W. Lent, N. E. Ryan, and E. B. McPartland. 1996. “Self-efficacy as an Intervening Mechanism Between
Research Training Environments and Scholarly Productivity: A Theoretical and Methodological Extension.” The
Counseling Psychologist 24 (3): 535–544. https://doi.org/10.1177.
Chin, W. W. 1998. “Commentary: Issues and Opinion on Structural Equation Modeling.” MIS quarterly. https://www.jstor.
org/stable/249674.
Connelly, Catherine E., and E. Kevin Kelloway. 2003. “Predictors of Employees’ Perceptions of Knowledge Sharing
Cultures.” Leadership & Organization Development Journal 24 (5): 294–301. doi:10.1108/01437730310485815.
Connelly, Catherine E., David Zweig, Jane Webster, and John P. Trougakos. 2012. “Knowledge Hiding in Organizations.”
Journal of Organizational Behavior 33 (1): 64–88. doi:10.1002/job.737.
Cropanzano, Russell, and Marie S. Mitchell. 2005. “Social Exchange Theory: An Interdisciplinary Review.” Journal of
Management 31 (6): 874–900. doi:10.1177/0149206305279602.
Edgar, Fiona, and Alan Geare. 2013. “Factors Influencing University Research Performance.” Studies in Higher Education 38
(5): 774–792. doi:10.1080/03075079.2011.601811.
Fornell, Claes, and David F. Larcker. 1981. “Evaluating Structural Equation Models with Unobservable Variables and
Measurement Error.” Journal of Marketing Research 18 (1): 39–50. doi:10.1177/002224378101800104.
Hackman, Richard J., and Charles G. Morris. 1975. “Group Tasks, Group Interaction Process, and Group Performance
Effectiveness: A Review and Proposed Integration.” Advances in Experimental Social Psychology 8: 45–99. doi:10.
1016/s0065-2601(08)60248-8.
12 S. KOZHAKHMET ET AL.

Hair, Joseph F., G. Tomas, M. Hult, Christian M. Ringle, and Marko Sarstedt. 2016. A Primer on Partial Least Squares
Structural Equation Modeling (PLS-SEM). Sage Publications.
Hayes, Andrew F. 2013. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based
Approach. New York: The Guilford Press.
Hedjazi, Yousef, and Jaleh Behravan. 2011. “Study of Factors Influencing Research Productivity of Agriculture Faculty
Members in Iran.” Higher Education 62 (5): 635–47. doi:10.1007/s10734-011-9410-6.
Hemmings, Brian, and Russell Kay. 2016. “The Relationship Between Research Self-Efficacy, Research Disposition and
Publication Output.” Educational Psychology 36 (2): 347–61. doi:10.1080/01443410.2015.1025704.
Hemmings, Brian Colin, Russell Kay, John Sharp, and Claire Taylor. 2012. “A Transnational Comparison of Lecturer Self-
Efficacy.” Journal of Further and Higher Education 36 (3): 291–307. doi:10.1080/0309877x.2011.614932.
Holden, Gary, Kathleen Barker, Thomas Meenaghan, and Gary Rosenberg. 1999. “Research Self-Efficacy: A New Possibility
for Educational Outcomes Assessment.” Journal of Social Work Education 35 (3): 463–476. doi:10.1080/10437797.1999.
10778982.
Holttum, Sue, and Laura Goble. 2006. “Factors Influencing Levels of Research Activity in Clinical Psychologists: A New
Model.” Clinical Psychology & Psychotherapy 13 (5): 339–51. doi:10.1002/cpp.501.
Hu, B., and Y. Zhao. 2016. “Creative Self-efficacy Mediates the Relationship Between Knowledge Sharing and Employee
Innovation.” Social Behavior and Personality: An International Journal 44 (5): 815–826. https://doi.org/10.2224/sbp.2016.
44.5.815.
Kahn, Jeffrey H., and Norman A. Scott. 1997. “Predictors of Research Productivity and Science-Related Career Goals
among Counseling Psychology Doctoral Students.” The Counseling Psychologist 25 (1): 38–67. doi:10.1177/
0011000097251005.
Konstantakos, E. K., R. T. Laughlin, R. J. Markert, and L. A. Crosby. 2010. “Assuring the Research Competence of Orthopedic
Graduates.” Journal of Surgical Education 67 (3): 129–134. https://doi.org/10.1016/j.jsurg.2010.04.002.
Kozhakhmet, Sanat, and Mohammad Nazri. 2017. “Governing Knowledge Sharing Behaviour in Post-Soviet Kazakhstan.”
Journal of Workplace Learning 29 (3): 150–164. doi:10.1108/JWL-06-2016-0053.
Kraiger, Kurt. 2003. “Perspectives on Training and Development.” In Handbook of Psychology. doi:10.1002/0471264385.
wei1208.
Kurahara, D. K., K. Kogachi, M. Yamane, C. L. Ly, J. H. Foster, T. Masaki-Tesoro, D. Murai, and R. Rudoy. 2012. “A Pediatric
Residency Research Requirement to Improve Collaborative Resident and Faculty Publication Productivity.” Hawai’i
Journal of Medicine & Public Health 71 (8): 224.
Kuvaas, Bård, and Anders Dysvik. 2009. “Perceived Investment in Employee Development, Intrinsic Motivation and Work
Performance.” Human Resource Management Journal 19 (3): 217–36. doi:10.1111/j.1748-8583.2009.00103.x.
Lambie, Glenn W., and Nicole Vaccaro. 2011. “Doctoral Counselor Education Students Levels of Research Self-Efficacy,
Perceptions of the Research Training Environment, and Interest in Research.” Counselor Education and Supervision
50 (4): 243–58. doi:10.1002/j.1556-6978.2011.tb00122.x.
Lee, Chay Hoon, and Norman T. Bruvold. 2003. “Creating Value for Employees: Investment in Employee Development.”
The International Journal of Human Resource Management 14 (6): 981–1000. doi:10.1080/0958519032000106173.
Lucas, Christopher J., and John W. Murry. 2011. New Faculty: A Practical Guide for Academic Beginners. Basingstoke:
Palgrave Macmillan.
MacKinnon, D. P., C. M. Lockwood, J. M. Hoffman, S. G. West, and V. Sheets. 2002. “A Comparison of Methods to Test
Mediation and Other Intervening Variable Effects.” Psychological Methods 7 (1): 83–104. doi:10.1037/1082-989X.7.1.83.
Mallinckrodt, B., and C. J. Gelso. 2002. “Impact of Research Training Environment and Holland Personality Type: A 15-year
Follow-up of Research Productivity.” Journal of Counseling Psychology 49 (1): 60. https://psycnet.apa.org/doi/10.1037/
0022-0167.49.1.60.
Meyer, John P., Natalie J. Allen, and Catherine A. Smith. 1993. “Commitment to Organizations and Occupations: Extension
and Test of a Three-Component Conceptualization.” Journal of Applied Psychology 78 (4): 538–51. doi:10.1037//0021-
9010.78.4.538.
Nunnally, J. C., and I. H. Bernstein. 1967. Psychometric Theory. US.
Nygaard, Lynn P. 2017. “Publishing and Perishing: An Academic Literacies Framework for Investigating Research
Productivity.” Studies in Higher Education 42 (3): 519–32. doi:10.1080/03075079.2015.1058351.
Overall, Nickola C., Kelsey L. Deane, and Elizabeth R. Peterson. 2011. “Promoting Doctoral Students Research Self-Efficacy:
Combining Academic Guidance with Autonomy Support.” Higher Education Research & Development 30 (6): 791–805.
doi:10.1080/07294360.2010.535508.
Pajares, Frank. 2002. Overview of Social Cognitive Theory and of Self-Efficacy.
Peterson, Shari L., and Joanne Provo. 1998. “Profile of the Adult Education and Human Resource Development
Professoriate: Characteristics and Professional Fulfillment.” Adult Education Quarterly 48 (4): 199–215. doi:10.1177/
074171369804800401.
Phillips, Julia C., and Richard K. Russell. 1994. “Research Self-Efficacy, the Research Training Environment, and Research
Productivity among Graduate Students in Counseling Psychology.” The Counseling Psychologist 22 (4): 628–41. doi:10.
1177/0011000094224008.
 STUDIES IN HIGHER EDUCATION 13

Podsakoff, Philip M., and Dennis W. Organ. 1986. “Self-Reports in Organizational Research: Problems and Prospects.”
Journal of Management 12 (4): 531–44. doi:10.1177/014920638601200408.
Preacher, Kristopher J., Derek D. Rucker, and Andrew F. Hayes. 2007. “Addressing Moderated Mediation Hypotheses:
Theory, Methods, and Prescriptions.” Multivariate Behavioral Research 42 (1): 185–227. doi:10.1080/
00273170701341316.
Richter, Andreas W., Giles Hirst, Daan Van Knippenberg, and Markus Baer. 2012. “Creative Self-Efficacy and Individual
Creativity in Team Contexts: Cross-Level Interactions with Team Informational Resources.” Journal of Applied
Psychology 97 (6): 1282–90. doi:10.1037/a0029359.
Rothberg, M. B., R. Kleppel, J. L. Friderici, and K. Hinchey. 2014. “Implementing a Resident Research Program to Overcome
Barriers to Resident Research.” Academic Medicine 89 (8): 1133–1139. doi:10.1097/ACM.0000000000000281.
Swanson, Richard A., and Ed Holton. 2008. Foundations of Human Resource Development. San Francisco, CA: Berrett-
Kohler.
Teodorescu, Daniel. 2000. “Correlates of Faculty Publication Productivity: A Cross-National Analysis.” Higher Education 39
(2): 201–222. doi:10.1023/a:1003901018634.
van den Hooff, Bart, and Jan A. de Ridder. 2004. “Knowledge Sharing in Context: The Influence of Organizational
Commitment, Communication Climate and CMC Use on Knowledge Sharing.” Journal of Knowledge Management 8
(6): 117–30. doi:10.1108/13673270410567675.
Wang, Sheng, and Raymond A. Noe. 2010. “Knowledge Sharing: A Review and Directions for Future Research.” Human
Resource Management Review 20 (2): 115–31. doi:10.1016/j.hrmr.2009.10.001.
Weinstein, Netta, and Richard M. Ryan. 2010. “When Helping Helps: Autonomous Motivation for Prosocial Behavior and Its
Influence on Well-Being for the Helper and Recipient.” Journal of Personality and Social Psychology 98 (2): 222–44.
doi:10.1037/a0016984.
Wester, K. L., L. D. Borders, L. M. Gonzalez, and P. Waalkes. 2019. “Factors Contributing to Scholarly Productivity of
Assistant Professors in Counseling.” Counselor Education and Supervision 58 (3): 225–237. doi:10.1002/ceas.12152.
Wood, F. 2018. “Factors Influencing Research Performance of University Academic Staff.” Higher Education 19 (1): 81–100.
https://doi.org/10.1007/BF00142025.