bs_bs_banner

International Journal of Training and Development 23:2

ISSN 1360-3736
doi: 10.1111/ijtd.12150

Team-based learning and training

transfer: a case study of training for the
implementation of enterprise resources
planning software

Sharon L. Hart, Brigitte Steinheider and

Vivian E. Hoffmeister

While traditional models of training such as behavioral

modeling (BMT) have been found to enhance training transfer,
research suggests that more active learning strategies such as
error management (EMT) and team-based learning (TBL) may
be more effective. This paper analyzes BMT, EMT and TBL
strategies to train employees on new enterprise resources plan-
ning (ERP) software and discusses which training leads to suc-
cessful procedural and declarative knowledge transfer,
knowledge retention and application, and tangible business
outcomes. TBL was predicted to be the most effective training
type, as it models several components needed to use ERP soft-
ware in the actual job setting. Overall and procedural knowl-
edge as well as knowledge application scores improved most
for TBL participants, while declarative knowledge improved
the most in the EMT condition. During training, all conditions
showed significant improvement in knowledge application;
however, the TBL condition showed the highest knowledge
application gains. This paper discusses the elements of TBL
that support its use as an effective strategy to increase knowl-
edge transfer in an organizational context.

Introduction
Although organizations continue to spend increasingly large amounts of resources
on employee training, research suggests that employees use only a fraction of that
training in their everyday jobs (Fitzpatrick, 2001; Freifeld, 2014; Perino & Michael,
2013).This ‘transfer problem’ (Saks & Belcourt, 2006, p. 630) has implications for the

❒❒Sharon L. Hart, Phillips 66, Bartlesville, OK USA. Email: sharon.l.hart@p66.com.

Brigitte Steinheider and Vivian E. Hoffmeister, University of Oklahoma, Tulsa, OK USA.
Email:bsteinheider@ou.edu; vhoffmeister@ou.edu.

© 2019 Brian Towers (BRITOW) and John Wiley & Sons Ltd.

The power of the team   135

implementation of enterprise resources planning (ERP) software which requires exten-
sive training of employees working with these systems. Training, however, must also
remain cost-effective and relevant, while ensuring that the skills being trained will
transfer to the job contexts. Although traditional behavioral modeling techniques have
been found to enhance training transfer (Burke & Hutchins, 2007), research suggests
that more active learning strategies as well as teamwork in a realistic work setting
may be more effective for adaptive knowledge transfer (Emke, Butler, & Larsen, 2016;
Huang & Lin, 2017; Keith & Frese, 2008; Michaelsen & Sweet, 2008).
This paper compares the effects of two commonly used training approaches (behav-
ior modeling and error management) with team-based training used by an aerospace
company in the Midwestern United States who aimed to train employees on the ERP
software using SAP, a German-based software company which offers a behavioral
modeling approach to training (Allen, 2005). This study reports results of each train-
ing type (behavioral modeling, error management and team-based learning) and dis-
cusses how training design elements may influence successful knowledge transfer,
knowledge application and tangible business outcomes.

Training transfer
According to Baldwin and Ford’s (1988) transfer model, training transfer occurs when
training inputs, such as the training design and trainee characteristics influence out-
comes, such as learning and knowledge retention. Using this model, Grossman and
Salas (2011) reviewed the training literature and identified the trainee characteris-
tics, training design and work climate factors that most strongly enhance transfer
in organizations. They propose future research use this list as a ‘springboard’ (p.
117) for empirical investigations into training methods at the organizational level;
specifically, how organizations can incorporate multiple factors efficiently. Other
meta-analyses and reviews of the training literature confirmed these factors as criti-
cal for the successful transfer of training (Bhatti & Kaur, 2010; Blume, Ford, Baldwin,
& Huang, 2010; Grohmann et al., 2014; Leimbach, 2014).
Trainee characteristics, such as cognitive ability, motivation and self-efficacy have
been shown to have moderate to high associations with successful training transfer
(Blume et al., 2010; Grohmann et al., 2014; Grossman & Salas, 2011; Leimbach, 2014).
Blume et al.’s (2010) meta-analysis of the transfer literature identified cognitive ability
as the strongest predictor of training transfer, showing that those with higher cogni-
tive ability are significantly more likely to retain and generalize training concepts; to
mediate low cognitive abilities research suggests increasing trainee motivation and
self-efficacy (Bhatti & Kaur, 2010; Grohmann et al., 2014). Motivation and self-efficacy
can be enhanced through the structure of the training program itself, by promoting
strategies such as behavioral modeling, which involves a guided explanation of the
new skills by providing models to show the correct use and teaching participants how
to avoid errors (Grossman & Salas, 2011), error management, a more active technique,
which encourages participants to make mistakes and use them as learning opportu-
nities (Cullen, Muros, Rasch, & Sackett, 2013), and implementing a realistic training
environment (Bhatti & Kaur, 2010; Grossman & Salas, 2011; Leimbach, 2014).
Lastly, the work climate in which training takes place should have several layers of
support in place to influence successful training transfer (Grossman & Salas, 2011).
Peer and supervisor support for learning transfer have consistently been shown to be
one of the strongest predictors, as they increase trainee self-efficacy and motivation to
transfer (Blume et al., 2010; Grohmann et al., 2014; Grossman & Salas, 2011; Leimbach,
2014). Grossman and Salas (2011) specifically identify the use of support behaviors,
such as goal-setting techniques, peer-to-peer coaching and feedback as strong predic-
tors for training transfer. Additionally, Lim and Johnson (2002) found that providing
ample opportunities to perform the learned skills both in and outside of training was
rated the most important sign of support from employees.
These three factors (i.e. characteristics, design and climate) function collaboratively,
meaning that if one or two factors are present without the others, training transfer can

136    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
be limited. For example, if a training program is designed to reflect accurate job con-
texts and the training environment supports feedback and coaching, but the cognitive
ability of the trainees does not enable them to generalize the information, transfer will
be limited. Similarly, if trainees are cognitively able to generalize information and the
training is designed to teach them to do so, but the climate of the work environment
does not provide peer or supervisor support, then transfer will be limited. Thus, it is
imperative that the training approaches that are implementing adequately address
each factor of training in tandem to promote successful knowledge transfer.

Training for ERP implementation

The implementation of new ERP systems presents a substantial challenge for organi-
zations. The extensive training necessary to introduce the system signals a massive
change in the organizational climate and requires employees to develop increased
analytical and social skills to be able to engage with the new software (Marler & Liang,
2012). Additionally, ERP implementation requires employees to work cross-function-
ally, meaning employees must be aware of functions of the software that may not be
explicitly relevant for their job context (Marler, Liang, & Dulebohn, 2006). Through
the integration of training design and ERP implementation research, we propose
that trainings that promote peer-to-peer engagement, continued learning and pro-
vide opportunities to apply concepts to realistic situations should be most effective
for training transfer in ERP implementation (Marler & Liang, 2012; Marler et al., 2006;
Saks & Belcourt, 2006; Salas, Tannenbaum, Kraiger, & Smith-Jentsch, 2012).
Behavioral modeling training (BMT) is based on Bandura’s (1977) social learning
theory, which is efficient in quickly disseminating information and leading students
through the course material by modeling the specific behaviors necessary to reach
the intended goal (Grossman & Salas, 2011). On the other hand, error management
instructions are designed to reframe errors as a natural, instructive part of the learning
process, therefore increasing self-efficacy (Keith & Frese, 2005). When both elements
are utilized in a training environment that accurately reflects the work setting, trainees
can better connect the new skills with their actual job functions, thus increasing their
motivation to learn and transferring their skills (Bhatti & Kaur, 2010; Grohmann et al.,
2014; Grossman & Salas, 2011).
However, trainings differ in terms of their effectiveness for adaptive knowledge
transfer. For example, BMT has been used extensively in SAP implementation and
was found to be more effective for trainees with lower cognitive ability or with previ-
ous experience with the training subject (Allen, 2005; Cullen et al., 2013; Reisslein et al.,
2006). When the task or topic to be trained covers a relatively small amount of material
that is highly structured and less complex, behavioral modeling may be an economical
approach to teach the correct strategies but should be paired with opportunities for
the trainees to use the new skills in ways that relate to their job contexts (Grossman &
Salas, 2011; Keith & Frese, 2008).
BMT also seems to be most effective when performance tasks are similar, if not
identical, to the ones practiced in training, and when it utilizes positive and nega-
tive models (Grossman & Salas, 2011; Keith, Richter, & Naumann, 2010). Because BMT
focuses on task-based as opposed to cognitive-based learning, individuals often have
difficulty adapting their knowledge and skills when concepts or processes change in
their environment (Devine & Kozlowski, 1995; Sternberg & Frensch, 1992). This can be
problematic for ERP implementation, as employees need to learn how to adapt to use
the software in situations that may not have been specifically addressed in the train-
ing. Further, trainees with higher cognitive abilities or with more experience with the
subject may view the guided instruction as redundant, hindering their motivation to
learn as it does not provide as much room for active exploration as compared to other
training methods (Cullen et al., 2013).
Error management training (EMT) employs behavioral modeling principles and
encourages the learning experience that occurs when trying to resolve a mistake
(Grossman & Salas, 2011). EMT exercises are designed to be complex to expose

The power of the team  137

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
trainees to many error situations in order to train adaptive learning (Heimbeck, Frese,
Sonnentag, & Keith, 2003). As it is less structured than BMT, it places more emphasis
on exploration. This approach is most useful when the learning material cannot be
covered completely, resulting in the need for students to ‘learn to learn’ when con-
fronted with new tasks (Keith & Frese, 2008), which has great implications for ERP
training, as trainings cannot realistically be comprehensive, but must teach employees
how to figure out the system independently.
Research has found that EMT is more effective for higher ability individuals; how-
ever, because trainees work alone, it is important that training exercises are clear. This
empowers students to continue to solve their own mistakes instead of being over-
whelmed by them (Gully, Payne, Koles, & Whiteman, 2002; Keith & Frese, 2008).
Additionally, EMT can be implemented with intentional peer or supervisor support
strategies to further increase the successful transfer of training. Although EMT has
been found to significantly predict transfer of software skills, research has focused
mainly on training common administrative software such as Microsoft Office prod-
ucts, not necessarily ERP software (Caputi et al., 2011; Keith & Frese, 2008; Nordstrom,
Wendland, & Williams, 1998). Additionally, EMT research has utilized undergraduate
student samples, but when used in workplace contexts, research has shown mixed
results due to increased variety in cognitive ability, age and other organizational con-
straints (Carter & Beier, 2010; Keith & Frese, 2008).
While used mostly in higher education settings, particularly medical education
(Haidet, Kubitz, & McCormack, 2014), team-based learning (TBL) incorporates many
of the key training factors outlined by Grossman and Salas (2011). TBL shifts the
trainee workload for reading and learning the concepts to before class time to leave
more time for application and reinforcement in the training session. At the beginning
of the session, trainees are divided into teams with diverse levels or areas of expertise
in which they remain for the duration of the training, so that they can learn from each
other in addition to the material and the instructor (Michaelsen & Sweet, 2008). This
could be highly useful for ERP implementation, as it gives employees a chance to train
with their future cross-functional partners, an essential skill needed for successful ERP
implementation (Marler et al., 2006).
TBL administers two knowledge assessments: an individual readiness assessment
(IRAT), which is completed alone, and a group readiness assessment (GRAT), which
is completed as a team (Michaelsen & Sweet, 2008). The rest of the course is designed
to include decision-based application exercises, so that trainees can practice applying
the material to tackle complex, real-world problems (Kibble, Bellew, Asmar, & Barkley,
2016). Team activities promote active learning, involving discussions about different
approaches to problems and benefitting individuals who do not memorize well and
need to apply the concepts in order to learn. Trainees are able to practice their new
skills in a realistic training environment and can increase their self-efficacy through the
team performance aspect (Grossman & Salas, 2011).
TBL is also beneficial for lower performing trainees, as it provides social support
for knowledge retention that may not always be present in lecture-based approaches
(Koles et al., 2005; Michaelsen & Sweet, 2008). This team engagement models the
social nature of ERP software use, especially between employees across the organi-
zation (Marler & Liang, 2012). Further, the team aspect trains employees to develop
a transactive memory system, where employees work interdependently, yet collabo-
ratively, to apply their knowledge to problem-solving tasks (Sharma & Yetton, 2007).
Overall, research has shown that TBL is associated with increased learner engagement,
self-efficacy, motivation (Huang & Lin, 2017; Kelly et al., 2005; Loftin & West, 2017),
improved problem-solving skills (Hunt et al., 2003; Kelly et al., 2005), high knowledge
retention due to active learning (Emke et al., 2016) and successful transfer of training to
actual job contexts (Haidet et al., 2014). A recent analysis of 40 TBL studies concluded
that there is initial evidence for positive educational outcomes in terms of knowledge
acquisition, participation and engagement, and team performance, but that more rig-
orous testing is needed to confirm the effects and explore the underlying mechanisms
(Haidet et al., 2014).

138    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
 While TBL has been mostly used in higher education settings (Haidet et al., 2014),
this training models the kind of team communication and peer support needed to use
ERP software in an adaptive, interdepartmental context. The current study investigates
three training approaches (BMT, EMT and TBL) in terms of overall, declarative (defi-
nitional) and procedural (adaptive) knowledge transfer when used to train employees
on SAP software. BMT is the current training method for SAP software, yet the cur-
riculum only includes positive models of software use. While EMT incorporates both
positive and negative models, this training has been shown to have mixed results and
may be an overwhelming approach when used in the context of ERP training, where
trainees are working alone to learn a complex system (Carter & Beier, 2010; Keith &
Frese, 2008). However, BMT and EMT do not account for the interdependence needed
to use ERP systems in the actual job context. This study also investigates the tangible
business outcomes associated with each training approach, measuring outcomes for
a 6-month period before, during and after the training concludes. It is predicted that
TBL will be more effective compared to BMT and EMT training, as TBL incorporates
the most effective elements of both trainings and meets organizational needs during
ERP implementation.

Methods
This study was implemented at a medium-sized privately held aerospace repair
and manufacturing company based in the Midwestern United States. The company
decided to replace all the various business systems in use at all their facilities with the
ERP software system SAP, one of the leading ERP systems in the world and the most
common system in use by the company’s larger customers and vendors.
The traditional training method used for the SAP system was BMT-focused and
included some of the most common business scenarios and transactions that an end
user would encounter. Although it only supplied trainees with positive models, it had,
for the most part, successfully equipped employees with the procedural knowledge to
process transaction scenarios that were presented in the training materials. However,
the effectiveness of the BMT design with regards to self-sufficiency had questionable
results for the company, due to the continued dependency on the deployment team and
length of time business metrics for the deployed facilities have remained in decline.

Design
The study employed a quasi-experimental repeated measurement design with the
training condition (BMT, EMT and TBL) as the independent variable and declara-
tive and procedural knowledge, knowledge retention and application and Customer
Return Turnaround Time as the dependent variables with three measurement points
(pre-, within- and post-training). Peer/supervisor support, opportunity to use, expe-
rience with business intelligence and age group were assessed as control variables.

Training procedures
The training was administered by two instructors, both of whom were ERP imple-
mentation team members at the organization. Both instructors attended an off-site
foundation course recommended by the SAP Education training curriculum. The
organization’s implementation team then developed and delivered in-house courses
modeled after the SAP curriculum but customized to the needs of the company.
The training assignment presented to participants required them to create and
interpret Customer Return Turnaround Time (TAT) reports from the SAP Business
Warehouse. This TAT process is an important performance metric that measures the
time it takes to receive a returned part, repair and return it to the customer, which
depends on the communication between different departments. Each condition
was administered with the same course introduction, debrief sections and pre- and
post-training measures, and the time for each condition was consistent. Conditions

The power of the team  139

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
differed only in the actual delivery of the training instructions. After completion of
pre-training measures, participants received the same company-specific reference
document with general navigation and report-building information about the SAP
Business Warehouse.
Pre-training procedure
Pre-training procedures remained constant for each training condition. Before the
course began, participants were briefed on the course agenda and completed pre-train-
ing assessments. These assessments included a survey of experience and background,
a declarative and procedural knowledge test, the Value of Teams assessment, and an
assessment measuring peer/supervisor support and opportunity to use the training
concepts.
Within-training procedure
The training consisted of three exercises that asked participants to: (1) access a stan-
dard Business Intelligence (BI) report and information about a customer return,
(2) locate a standard BI report to identify the tasks and task status for a specific cus-
tomer return and (3) find the individual department and total turnaround time for
a specific customer return and identify delay points in the process. Each condition
received different instructions for how to learn the Customer Return TAT processes.

BMT condition. In the BMT condition, participants began the course by watching
their instructor explain the Customer Returns course material and demonstrate
each exercise. Participants completed an individual readiness assessment test (IRAT)
to measure knowledge retention after the instructor-led walkthrough and were
provided with written step-by-step directions (a navigation guide) and solutions to
complete the three course exercises. Participants were instructed to carefully follow
the instructions to complete the task and were informed that it would help them learn
the key functions in the shortest amount of time. They were told that following the
navigation guide would teach them the correct way to complete the exercises and to
notify the instructor if any errors occurred.

EMT condition. In the EMT condition, participants also began by watching the
instructor explain the Customer Returns course material and demonstrate each
exercise. Participants then completed the IRAT and were asked to complete the course
exercises without any written step-by-step directions to provide them with an ‘intensive
interaction with the program’. They were advised to consult their navigation guide if
they had questions but were also encouraged to make errors and to try to figure out
solutions on their own. Trainees were told to explore the software, even if they were
unsure what they were doing, and were reassured that any errors that occurred were
natural, fixable and would help them learn the software more effectively.

TBL condition. In the TBL condition, participants were first divided into teams with
representatives from multiple functional business areas. Participants individually
completed the pre-training assignment by reading and working through the Customer
Returns course material. The instructor followed with a quick overview and answered
questions as necessary. Participants then completed an IRAT to test individual
learning, followed by re-taking the same test as a team (group readiness assessment
test; GRAT). Participants were given the same error management instructions, except
they were instructed to solve problems and complete course exercises as a team.
Post-training procedure
Participants in all conditions participated in a question and answer period followed
by the administration of post-training measures. Post-training measures included a

140    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
post-training declarative and procedural knowledge assessment test and the retaking
of the Value of the Team assessment.

Participants
The entire population were 84 aerospace company employees from different func-
tional work areas (accounting, customer service, receiving and production). The
training was required for every employee involved with the use of SAP software.
Fifteen participants were unable to attend the training classes and data from two par-
ticipants were discarded because of incomplete surveys, resulting in a final sample
size of N = 67 with the following age distribution: age group 1 (20–29 years) = 6%,
age group 2 (30–39 years) = 25.4%, age group 3 (40–49 years) = 28.4% and age group 4
(50 + years) = 38.8%.
Computer experience among employees ranged from 5 to 40 years (M = 20.08 years,
SD = 6.99); participants reported using a computer daily (98.5%) and 79% had never
created a BI report with the SAP software. To ensure representation of each functional
business area in all the training groups, the participants were first stratified by business
function, and then randomly assigned to one of the three training conditions. Schedule
conflicts and reschedule requests resulted in the following distribution of participants
in each course: BMT (n = 18), EMT (n = 22) and TBL (n = 27), with Instructor #1 having
9, 13 and 12 students in the BMT, EMT and TBL conditions, respectively.

Measures
Dependent variables
Performance was assessed using company developed exercises and test questions. To
accommodate measurement of conceptual and adaptive learning in this study, declar-
ative and procedural test questions throughout the course were designed to be pro-
gressively complex at each test point. Questions were coded by the instructors based
on their complexity factor (1 = low, 2 = medium and 3 = high) prior to calculating
the overall test score. This design has been used in several studies to measure prob-
lem-solving skills beyond the simple recall of relevant information (Bell & Kozlowski,
2008; Heimbeck et al., 2003; Keith & Frese, 2008).
Individual performance was assessed two ways. First, to assess declarative and pro-
cedural transfer, a baseline pre-test was conducted at the beginning of the training and
compared to a more difficult post-test at the end of the training. The baseline pre-train-
ing test was comprised of four declarative questions (two low and two medium com-
plex; e.g. ‘True/False: BI Reports provide “real time” access to transactional data in
SAP’.) to measure their knowledge of the software, and two procedural questions (one
low and one medium complex; e.g. ‘Which of the following do you use to set a report
filter?’) that measured their existing ability to use the tool. The post-test was com-
pleted individually by the participants and consisted of three declarative (one low and
two medium complex) and three procedural questions (two medium and one highly
complex).
Second, to assess within-training knowledge retention, participants completed an
IRAT during training after the review of course materials and before working the
course exercises. Scores on the course exercises were used to measure application of
the knowledge taught to the actual scenarios. To assess adaptive transfer, the IRAT
was designed to be more complex than the pre-training test and consisted of seven
procedural questions of medium to high complexity (‘What could you do to expedite
the processing of [return QN 20030305]?’). In the TBL condition, a GRAT was also
administered. The GRAT was identical to the IRAT and was used to assess training
transfer changes due to the team-based design of the condition. The course exercises
were comparable to the IRAT and consisted of six procedural questions of medium to
high complexity.
The business performance goal of the training program was to teach employees
how to monitor tasks that have been assigned to them, so they would be able to more

The power of the team  141

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
effectively manage and reduce TAT. To measure the effect of the training program,
performance metrics were observed for 6 months, before, during, and after training.
Baseline business performance TAT metrics for the total population were extracted
prior to training (December and January) from the SAP Business Warehouse and
recorded. Metrics were then taken during training (February and March) and after
training (April and May).
Covariates
Age, peer/supervisor support, experience with Business Intelligence and attitudes
toward team work were assessed as covariates. Peer/supervisor support and opportu-
nity to use the applied concepts were measured using the work environment subscale
adapted from Holton, Chen, and Naquin’s (2003) Learning System Transfer Inventory
(LTSI). The subscales assessed peer support (‘My colleagues encourage me to use the
skills I have learned in training’, 4 items, α = 0.79), supervisory support (‘My supervi-
sor shows interest in what I learn in training’, 6 items, α = 0.92) and opportunity to use
(‘The resources I needed to use what I learned were available to me after training’, 4
items, α = 0.70). Participants responded to each item using a 5-point Likert scale rang-
ing from 1 (strongly disagree) to 5 (strongly agree).
Attitudes toward team work were assessed using the Value of Teams survey devel-
oped by the Baylor College of Medicine (Fund for the Improvement of Postsecondary
Education [FIPSE], 2003) which measures a student’s appreciation of learning within
a group with 12 items (‘I have a positive attitude about working with my peers’); one
distractor item (‘Memorization is an important part of learning’) was discarded. The
tool was developed and tested with participants of the FIPSE team-based learning
project and has been used in numerous TBL studies to measure attitudes toward team
work (e.g. Kelly et al., 2005; O’Malley et al., 2003). The data were assessed at the begin-
ning and the end of the training. Cronbach’s αs were 0.84 and 0.88 at time 1 and time 2,
respectively. Participants responded to each item using a 5-point Likert scale ranging
from 1 (strongly disagree) to 5 (strongly agree).
Control variables
As control variables, participants were asked to indicate their age group, prior experi-
ence with the training subject and tenure at the company.
Manipulation check Changes in attitudes toward teamwork were assessed before and
after the training by the Value of Teams survey (FIPSE, 2003). Additionally, differences
in IRAT and GRAT scores for the TBL groups were compared.

Results
The small size and high diversity of the population negatively impacted the power
required to statistically test differences between training conditions for this pilot
study. Therefore, analyses were limited to paired t-tests comparing pre- and post-test
data and calculations of effect sizes; however, effect sizes can be more informative
than significance testing, especially in business applications (Valentine & Cooper,
2003).
Analysis showed that the Value of Teams was significantly correlated with scores on
the IRAT (r = 0.25, p < 0.05; see Table 1). Peer support was significantly correlated with
supervisor support (r = 0.43, p < 0.001) and opportunity to use the training (r = 0.41,
p = 0.001), as well as tenure (r = −0.30, p < 0.015), indicating that those who have
worked at the company for a shorter amount of time perceived more peer support.
Additionally, age was significantly correlated with tenure (r = 0.44, p < 0.001), as well
as with computer experience (r = 0.29, p = 0.02).
Conditions did not differ in terms of peer/supervisor support (F(2,64) = 0.72,
NS), Value of Teams perception (F(2,64) = 0.54, NS), years of computer experience
(F(2,64) = 0.35, NS), experience with BI (F(2,64) = 0.51, NS), tenure (F(2,64) = 0.03, NS)
or age group (F(2,63) = 0.02, NS). The average pre-test scores on overall knowledge
(declarative and procedural questions) for the BMT (M = 51.85, SD = 27.48) condition

142    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
 Table 1: Descriptive statistics and correlations

Variable M SD 1 2 3 4 5 6 7 8 9 10 11 12

1 Overall pre-test 45.27 23.79 –

grade
2 IRAT 37.14 25.54 0.10 –
3 Course exercises 62.26 28.07 −0.05 0.33** –
4 Overall post-test 54.60 26.56 0.21 0.36** 0.31** –
grade
5 Value of teams 3.97 .45 0.05 0.25* 0.16 0.17 (0.84)
6 Peer support 3.89 .59 0.06 −0.15 0.08 0.13 0.24 (0.79)
7 Supervisor support 3.36 .80 0.10 −0.14 0.14 −0.11 0.07 0.43** (0.92)
8 Opportunity to use 3.59 .54 0.05 0.00 0.07 0.19 0.24 0.41** 0.40** (0.70)
9 Tenure 10.08 9.18 −0.16 −0.16 −0.08 −0.09 −0.14 −0.30* −0.17 −0.09 –
10 Age 3.02 .95 −0.03 −0.24 −0.18 −0.09 −0.10 −0.11 −0.09 −0.07 0.44** –
11 Experience 20.08 6.99 0.16 0.13 0.04 0.21 0.01 0.23 −0.11 0.07 −0.22 0.29* –
– computer
12 Experience – BI 2.58 1.01 0.16 0.10 −0.01 0.13 −0.13 0.06 −0.02 0.13 −0.02 0.23 0.19 –
report

Note: Internal consistency reliability coefficients (α) reported in parentheses when appropriate.
*p < 0.01; **p < 0.001.

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
The power of the team  143
were higher than EMT (M = 45.45, SD = 20.55) and TBL (M = 40.74, SD = 23.47); how-
ever, due to the small sample size this difference is not significant (F (2,64) = 1.19, NS).
Participants indicated an intermediate level of perceived BI experience (M = 2.58 out
of 5, SD = 1.02); pre-test scores, however, showed that participants had a slightly lower
than average knowledge of BI (M = 45.27 out of 100, SD = 23.79).
Paired sample t-tests showed that attitudes toward teamwork (Value of Teams)
improved the most for participants in the TBL condition. While the BMT (t(17) = 0.21,
p = 0.837, d = 0.05) and EMT conditions (t(20) = 0.802, p = 0.432, d = 0.18) showed no
improvement, a small effect was observed for the TBL group (t(24) = 1.53, p = 0.140,
d = 0.31). In the TBL condition, GRAT scores (M = 73.20, SD = 24.38) were signifi-
cantly higher than previous IRAT scores (M = 33.77, SD = 23.68; t(26) = 6.42, p < 0.001,
d = 1.24). Scores on the test improved by 117% when participants were allowed to take
it as a team, indicating a successful manipulation of teamwork in the TBL groups.
Table 2 reports mean ratings for pre- and post-training assessments, as well as per-
centage of improvement and effect sizes for all conditions. To examine effects of the
conditions on the dependent variables, we conducted paired sample t-tests.
Pre/post comparison of overall performance showed improvements of 2% in the BMT,
23% in the EMT and 35% in the TBL condition. Differences for the BMT (t(17) = 0.12,
NS, d = 0.02, 95% CI [−15.97, 17.84]) and EMT groups were not significant (t(21) = 1.43,
NS, d = 0.30, 95% CI [−4.67, 25.13]), whereas TBL groups (t(26) = 2.56, p = 0.017,
d = 0.49, 95% CI [2.81, 25.59]) revealed significant differences and a medium effect size
(Figure 1).
Declarative knowledge improved by 18% for BMT, 42% for EMT and 22% for TBL par-
ticipants (Figure 2). Pre/post differences were not significant for the BMT (t(17) = 1.68,
NS, d = .40, 95% CI [−2.66, 23.40]) and the TBL condition (t(26) = 1.65, NS, d = 0.31,
95% CI [−2.46, 22.46]) and effect sizes were small. Declarative knowledge scores in the
EMT condition improved significantly and the effect size was medium (t(21) = 3.14,
p = 0.005, d = 0.67, 95% CI [7.07, 34.74]).
Procedural knowledge improved by 3% for BMT, 23% for EMT and 82% for TBL par-
ticipants. Pre/post differences were not significant for BMT (t(17) = 0.09,NS, 95%
CI [−29.98, 32.62]) and EMT (t(21) = 0.68, NS, 95% CI [−17.30, 34.18]), and there was
no effect in the BMT (d = 0.02) and EMT condition (d = 0.15), whereas TBL scores
improved significantly (t(26) = 2.46, p = 0.021, 95% CI [3.97, 44.70]) with a medium
effect size (d = 0.47) (Figure 3).
Within training knowledge retention was tested by comparing IRAT scores to the pro-
cedural pre-test questions for each training condition. Performance did not improve
for the BMT condition but improved by 6% and 14% for EMT and TBL participants,
respectively. None of the conditions significantly improved (BMT: t(17) = 0.02, NS,
95% CI [−27.20, 26.76]; EMT: t(21) = 0.22,NS, 95% CI [−18.04, 22.31]; TBL: t(26) = 0.478,
NS, 95% CI [−13.68, 21.96]) and effect sizes were negligible (d = 0.00; d = 0.05; d = 0.09).
Within training knowledge application compared average exercise scores to the aver-
age IRAT scores (both administered during training) and resulted in performance
improvements between 28% (EMT) and 103% (TBL; BMT: 69%). Differences were
significant for BMT and TBL conditions (BMT: t(17) = 3.69, p = 0.002, 95% CI [12.04,
44.21]; TBL: t(26) = 6.12, p = 0.000, 95% CI [23.07, 46.43]) and marginally significant for
the EMT condition (t(21) = 1.82, p = 0.084, 95% CI [−1.58, 23.27]), with a small effect
size for the EMT condition (d = 0.39) and large effect sizes for the BMT and TBL condi-
tions (d = 0.87; d = 1.18, respectively).
There were no significant differences between both instructors for the dependent
variables, except for GRAT scores. Instructor #1’s TBL condition scored higher than
Instructor #2’s on the GRAT. However, these differences did not affect both post-tests
and worked examples.
TAT averages were recorded in the months before, during and after training. The
average TAT was 22 days for the months prior to the training intervention, and a
substantial reduction in TAT to an average 12 days (approximately 50% reduction)
occurred during the months when the training was conducted. The TAT during the
months immediately following training returned to a pre-training average of 23 days.

144    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
 Table 2: Improvement scores and effect sizes for each condition

Behavioral modeling training Error management training Team-based learning

Knowledge Pre Post % Effect Pre Post % Effect Pre Post % Effect
Type M SD M SD Improved d size M SD M SD Improved d size M SD M SD Improved d size
Overall 51.9 27.5 52.8 26.4 2% 0.02 NE 45.5 20.6 55.7 24.5 23 0.30 Sm 40.7 23.5 54.9 29.2 35 0.49 Med
Declarative 57.4 28.7 67.8 31.5 18% 0.40 Sm 50.0 21.8 70.9 27.4 42 0.67 Med 46.3 28.2 56.3 36.0 22 0.31 Sm
Procedural 40.7 40.5 42.1 34.8 3% 0.02 NE 36.4 35.5 44.8 41.7 23 0.15 NE 29.6 40.7 54.0 38.1 82 0.47 Med
Retention 40.7 40.5 40.5 25.8 0% 0.00 NE 36.4 35.5 38.5 28.1 6 0.05 NE 29.6 40.7 33.8 23.7 14 0.09 NE
Application 40.5 25.8 68.7 31.2 69% 0.87 Lg 38.5 28.1 49.4 30.9 28 0.39 Sm 33.8 23.7 68.5 19.6 103 1.18 Lg

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
The power of the team  145
 Figure 1: Pre- and Post-test scores.

Figure 2: Declarative and procedural pre- and post-test scores.

Figure 3: Procedural questions, IRAT, GRAT and exercise scores.

Note: Within-training knowledge retention was measured by comparing scores on
procedural questions to IRAT score. Within-training knowledge application was
measured by comparing IRAT and exercise scores.

146    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
 Figure 4: TAT averages.

To review effects between training methods, the average TAT for the total popula-
tion metric was filtered by the participants of each training method and compared to
employees who did not complete any training. Task assignments and distributions are
a function of the types of occurring customer returns and cannot be easily controlled
or planned. During the research period, the resulting task distribution was heavily
skewed toward the participants in the TBL condition with few tasks for BMT partici-
pants, reducing the reliability of the measure (Figure 4).
During training, the average TAT for the BMT condition was reduced by 58% (38
to 16 days); however, immediately post-training, BMT groups did not maintain this
reduction, with a TAT average of 28 days. The average TAT for the EMT group did
not change during the training period (11 days) and subsequently trended upwards
during April and May (28 to 32 days) toward December pre-training levels (49 days).
The average TAT for the TBL group was reduced by 45% (from 18 to 10 days) during
the training period; TBL was the only group that maintained improvement during the
months immediately following training (8 days and 10 days). Notably, a 62% reduction
in average TAT for the remainder of the population that did not participate in training
occurred during the training period, but subsequently trended upwards during April
and May (19 to 62 days) exceeding pre-training levels.
Unsolicited negative feedback was received by the first author (instructor # 1) via
email from two of the participants receiving the EMT condition. These participants did
not feel that the unguided individual-based format of the course was ‘fun’ or ‘produc-
tive’. Additionally, unsolicited positive feedback about the different course formats
was also received from the second instructor who noted a much higher level of par-
ticipant interaction and interest in the TBL format. Instructor# 2 preferred this format
over the other two (BMT, EMT) and indicated an interest in converting some existing
training courses to the TBL format.

Discussion
This study examined the effects of traditional BMT approaches versus error encourage-
ment approaches, such as EMT and TBL training. Our study investigated these effects
in terms of overall performance gains, declarative and procedural performance gains,
and within-training knowledge retention and application. Post-test average scores
improved the most for TBL participants followed by EMT participants and BMT par-
ticipants with an effect size most pronounced for TBL training (see Table 2). TBL scores
improved the most for procedural questions with a moderate effect size, indicating a

The power of the team  147

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
successful transfer of adaptive (procedural) knowledge in the TBL groups, whereas
no effect was found for the BMT condition. For declarative knowledge, EMT scores
improved the most with a large effect size. During training, all conditions showed sig-
nificant improvements on knowledge application; however, the TBL groups showed
the highest knowledge application gains with a very large effect size.
The BMT condition had the highest overall pre-test score, but scores did not improve
significantly in the post-test; whereas both other conditions improved significantly. A
ceiling effect was also not present, as the highest score was 72.3% out of 100% (GRAT,
TBL). For within training knowledge application, BMT post scores were higher than
EMT (see Table 1); however, BMT scores did not differ from TBL scores. It is surpris-
ing, though, that the BMT scores were not even higher, as they were guided through
worked exercises and were able to refer to their notes. This is inconsistent with pre-
vious research, as the guided structure of the BMT groups should have resulted in
higher knowledge application (Heimbeck et al., 2003; Keith & Frese, 2005). However,
the guided structure can be stifling for knowledge application, as individuals often
have difficulties adapting the knowledge to different or more complex problems in the
future (Devine & Kozlowski, 1995; Sternberg & Frensch, 1992).
Both the BMT and EMT groups also scored high on declarative knowledge. Because
the BMT condition was instructed to avoid errors and both conditions were able to use
their notes to inform their answers, it is probable that declarative questions (which con-
sisted of basic definitions) were the easiest to locate and answer correctly, compared
to the procedural questions that involved more problem-solving. This may explain
why TBL groups subsequently improved at a higher rate on procedural questions, as
trainees were in an environment conducive to problem-solving. Comparisons of pre-
post training scores revealed that subjects in the TBL group had the greatest overall
improvement, especially on procedural knowledge and knowledge application, while
the EMT condition improved the most for declarative knowledge.
TAT averages showed that BMT groups fluctuated greatly before, during and after
training, although TBL groups showed a 45% reduction in TAT during training and
continued to show consistent results after training. They were the only group to main-
tain reduced turnaround time post-training, indicating longer term effects, whereas
BMT groups improved knowledge during and after training but this knowledge trans-
fer did not extend to tangible business outcomes. Previous research suggests that the
EMT condition should have shown improvements post-training while remaining con-
sistent within-training (Dimitrova, van Dyck, van Hooft, & Groenewegen, 2015); our
results partially support this finding with groups not significantly improving from
pre- to within-training (9 to 11 days). However, the EMT condition’s average rose in
the two months post-training (11 days to 28–32 days), instead of improving.
Thus, the TBL groups showed the highest improvement scores on all knowledge
types, including overall, and procedural knowledge as well as knowledge application,
except declarative knowledge with the EMT condition having the greatest improve-
ment and knowledge retention (no condition improved in knowledge retention).
Additionally, the TBL groups showed consistent reduction in TAT average after train-
ing, while both other conditions did not.

Implications
Our results suggest that TBL training leads to successful knowledge transfer, as foun-
dational elements of the training strategy influence trainee self-efficacy, perceived
support and motivation. The teamwork structure may enhance students’ self-efficacy
and motivation throughout the training process and offers a realistic training envi-
ronment in that it models real-world business settings with employees functioning
in team or group settings where their actions are dependent on and influenced by
others. TBL also primes continued team learning, another important characteristic
of real-world business scenarios and a component that has been found to increase
successful learning transfer (Grossman & Salas, 2011). Employees not only learn how
to learn (i.e. adaptive transfer), but learn by being part of a team. Additionally, these

148    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
teams were diverse in business expertise, grouping together employees from various
departments which also aligns with the foundational purpose of an ERP system –
to integrate various business functions, helping them problem-solve together in one
cohesive system.
TBL training also offers more autonomy than BMT or even EMT methods, because
the responsibility for learning the declarative knowledge is shifted to the individual.
TBL enables trainees to decide how they study the material beforehand and gives them
control over the speed at which they learn, which may increase their feelings of auton-
omy that may have been weakened by the implementation of the ERP system (Marler
et al., 2006). Additionally, instead of learning and working through examples with an
instructor, employees are instructed to work through them with their team, therefore
not only modeling behaviors of the instructors, but also of their peers. These benefits
– peer support, realistic training environments and positive and negative behavioral
modeling – have been highlighted as essential factors in the success of training transfer
(Grossman & Salas, 2011; Saks & Belcourt, 2006; Salas et al., 2012).
The kind of learning that takes place in TBL is similar to the theory of situated learn-
ing, which posits that learning takes place as part of humans’ social engagement with
communities of practice (Lave, 1991; Wegner, 2000). Thus, the social interaction and
negotiation that occurs in TBL settings is necessary for learning and knowledge trans-
fer. In this model of learning, trainees do not only ‘learn by doing’, but rather learn by
interacting with each other, focused on authentic problems (Lave, 1991; Wegner, 2000).
Additionally, Wegner (2000) notes that organizational success is dependent upon if
organizations can establish themselves as social learning systems or, as Grossman and
Salas (2011) define it, if they can establish a positive transfer climate where situational
and social cues encourage the transfer of knowledge. Through the social support and
teamwork structure present in TBL, employees develop many of the dimensions of
communities of practice (e.g. alignment of goals, engagement in addressing knowl-
edge gaps; Wegner, 2000), which could have contributed to the increase of procedural
knowledge and tangible business outcomes seen in the TBL condition.
The work environment itself is also more conducive to TBL training approaches
than traditional behavioral modeling approaches by offering an inherent positive and
supportive transfer climate and built-in follow-up opportunities. A positive transfer
climate consists of support from a social network, including supervisors and peers,
opportunities to correct mistakes and opportunities to receive constructive feedback
(Grossman & Salas, 2011), which are innate characteristics of TBL training. Also,
follow-up opportunities occur as part of the TBL curriculum, and are further enhanced
as employees take learned skills and apply them to their day-to-day job functions.
Overall, TBL seems to integrate the most positive and effective aspects of each training
approach studied – it incorporates behavioral modeling and error management prin-
ciples, but in a manner that cues students for successful learning transfer.

Limitations
The relatively small population size resulted in insufficient statistical power to test
performance differences between conditions. The study utilized randomized assign-
ment as much as possible, but situational constraints, such as schedule changes and
participant availability, limited the ability to have a truly randomized sample. During
the 6-month research period, the task distribution between conditions was heavily
skewed toward the participants in the TBL condition with few tasks for BMT par-
ticipants, reducing the reliability of the measure. However, this study modeled how
implementation of various training methods would function in a real-world business
setting, where constraints cannot always be avoided.

Future research
To our knowledge, this is one of the first studies to test TBL training in a business
setting. So far, studies have been conducted primarily in academic environments

The power of the team  149

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
with younger and less diverse populations (e.g. Dunaway, 2005; Haidet & Fecile, 2006;
Huang & Lin, 2017; Hunt et al., 2003; Kibble et al., 2016; Koles et al., 2005). Future research
should continue to examine how companies can offer cost-effective and efficient
opportunities for employees to engage in continued learning to strengthen training
concepts and prevent knowledge decay. Research on measuring long-term knowledge
retention in a business setting is also limited; future research should investigate what
components of practical business training reduce decay and strengthen retention.

Conclusion
This paper analyzes three training approaches for the implementation of ERP soft-
ware and identifies alternative and possibly more effective training approaches for
procedural and declarative knowledge transfer, knowledge application and tangible
business outcomes. Additionally, this study investigated components present in TBL
that may contribute to its effectiveness as a training method. The collaborative struc-
ture of TBL fosters team skills and achievement which may provide a continuous
support system for the material and procedures learned during training. Therefore,
TBL could be an effective approach to combine the strengths from both approaches
by incorporating the error encouragement instructions present in EMT methods but
offering employees a chance to learn from each other while practicing real-world
application. Further, these strengths have been found to be the key components pres-
ent in successful learning transfer (Grossman & Salas, 2011). Our explorative study
found that TBL successfully initiates transfer of procedural knowledge and knowl-
edge application, leading to enhanced tangible business outcomes.

References
Allen, G. F., Sr. (2005). The impact of enterprise resource planning on business processes in
allied aerospace corporation. Available from ProQuest Dissertations & Theses Global.
(305395457).
Baldwin, T. T. and Ford, J. K. (1988), ‘Transfer of training: a review and directions for future
research’, Personnel Psychology, 41, 1, 63–105.
Bandura, A. (1977), Social Learning Theory (Englewood Cliffs, NJ: Prentice Hall).
Bell, B. and Kozlowski, S. W. J. (2008), ‘Active learning: effects of core training design elements
on self-regulatory processes, learning, and adaptability’, Journal of Applied Psychology, 93,
2, 296–316.
Bhatti, M. A. and Kaur, S. (2010), ‘The role of individual and training design factors on training
transfer’, Journal of European Industrial Training, 34, 7, 656–72.
Blume, B. D., Ford, J. K., Baldwin, T. T. and Huang, J. L. (2010), ‘Transfer of training: a meta-ana-
lytic review’, Journal of Management, 36, 4, 1065–105.
Burke, L. A. and Hutchins, H. M. (2007), ‘Training transfer: an integrative literature review’,
Human Resource Development Review, 6, 3, 263–96.
Caputi, P., Chan, A. and Jayasuriya, R. (2011), ‘How helpful are error management and
counterfactual thinking instructions to inexperienced spreadsheet users’ training
task performance?’, British Journal of Educational Technology, 42, 4, 592–7. https://doi.
org/10.1111/j.1467-8535.2011.01177.x.
Carter, M. and Beier, M. E. (2010), ‘The effectiveness of error management train-
ing with working-aged adults’, Personnel Psychology, 63, 3, 641–75. https://doi.
org/10.1111/j.1744-6570.2010.01183.x.
Cullen, M., Muros, J., Rasch, R. and Sackett, P. (2013), ‘Individual differences in the effectiveness
of error management training for developing negotiation skills’, International Journal of
Selection and Assessment, 21, 1, 1–21.
Devine, D. J. and Kozlowski, S. W. J. (1995), ‘Expertise and task characteristics in decision mak-
ing’, Organizational Behavior and Human Decision Processes, 64, 3, 294–306.
Dimitrova, N. G., van Dyck, C., van Hooft, E. A. J. and Groenewegen, P. (2015), ‘Don’t fuss, focus:
the mediating effect of on-task thoughts on the relationship between error approach in-
structions and task performance’, Applied Psychology, 64, 3, 599–624.
Dunaway, G. (2005), ‘Adaption of team learning to an introductory graduate pharmacology
course’, Teach Learn Med, 17, 1, 56–62.

150    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
Emke, A., Butler, A. and Larsen, D. (2016), ‘Effects of Team-Based Learning on short-term and
long-term retention of factual knowledge’, Medical Teacher, 38, 3, 306–11.
Fitzpatrick, R. (2001), ‘The strange case of the transfer of training estimate’, The Industrial-
Organizational Psychologist, 39, 2, 18–9.
Freifeld, L. (2014), ‘The future of SAP training’, Training, 51, 5, 7.
Fund for the Improvement of Postsecondary Education [FIPSE] (2003), Baylor University: eval-
uation of team learning in health-sciences education Available at http://www.fipse.aed.org/
grantshow.cfm?grantNumber=P116B000884.
Grohmann, A., Beller, J. and Kauffeld, S. (2014), ‘Exploring the critical role of motivation to
transfer in the training transfer process’, International Journal of Training and Development,
18, 2, 84–103.
Grossman, R. and Salas, E. (2011), ‘The transfer of training: what really matters’, International
Journal of Training and Development, 15, 2, 103–20.
Gully, S. M., Payne, S. C., Koles, K. L. K. and Whiteman, J. K. (2002), ‘The impact of error training
and individual differences on training outcomes: an attribute–treatment interaction per-
spective’, Journal of Applied Psychology, 87, 1, 143–55.
Haidet, P. and Fecile, M. (2006), ‘Team-based learning: a promising strategy to foster active
learning in cancer education’, Journal Cancer Education, 21, 3, 125–8.
Haidet, P., Kubitz, K. and McCormack, W. T. (2014), ‘Analysis of the team-based learning litera-
ture: TBL comes of age’, Journal on Excellence in College Teaching, 25, 3–4, 303–33.
Heimbeck, D., Frese, M., Sonnentag, S. and Keith, N. (2003), ‘Integrating errors into the training
process: the function of error management instructions and the role of goal orientation’,
Personnel Psychology, 56, 2, 333–61.
Holton, E. F. III, Chen, H. and Naquin, S. S. (2003), ‘An examination of learning transfer system
characteristics across organizational settings’, Human Resource Development Quarterly, 14,
4, 459–82.
Huang, C. K. and Lin, C. Y. (2017), ‘Flipping business education: transformative use of team-
based learning in Human Resource Management classrooms’, Journal of Educational
Technology & Society, 20, 1, 323–36.
Hunt, D. P., Haidet, P., Coverdale, J. H. and Richards, B. F. (2003), ‘The effect of using team learn-
ing in an evidence-based medicine course for medical students’, Teaching and Learning in
Medicine, 15, 2, 131–9.
Keith, N. and Frese, M. (2005), ‘Self-regulation in error management training: emotion control
and metacognition as mediators of performance effects’, Journal of Applied Psychology, 90,
4, 677–91.
Keith, N. and Frese, M. (2008), ‘Effectiveness of error management training: a meta-analysis’,
Journal of Applied Psychology, 93, 1, 59–69.
Keith, N., Richter, T. and Naumann, J. (2010), ‘Active/exploratory training promotes transfer
even in learners with low motivation and cognitive ability’, Applied Psychology, 59, 1, 97–123.
Kelly, P., Haidet, P., Schneider, V., Searle, N., Seidel, C. and Richards, B. (2005), ‘A comparison of
in-class learner engagement across lecture, problem based learning, and team learning
using the STROBE classroom observation tool’, Teach Learn Med, 17, 2, 112–8.
Kibble, J., Bellew, C., Asmar, A. and Barkley, L. (2016), ‘Team-based learning in large enrollment
classes’, Advances in Physiology Education, 40, 4, 435–42.
Koles, P., Nelson, S., Stolfi, A., Parmelee, D. and DeStephen, D. (2005), ‘Active learning in a year
2 pathology curriculum’, Medical Education, 39, 10, 1045–55.
Lave, J. (1991), ‘Situating Learning in Communities of Practice’, in L. Resnick, J. M. Levine and
S. D. Teasley (eds), Perspectives on Socially Shared Cognition, (Washington, DC: American
Psychological Association), pp. 63–82.
Leimbach, M. P. (2014). ‘Innovative Approaches to Improving the Transfer and Sustainability
of Learning’, in F. Soliman (ed.), Learning Models for Innovation in Organizations: Examining
Roles of Knowledge Transfer and Human Resources Management (Hershey, PA: IGI Global), pp.
111–26. https://doi.org/10.4018/978-1-4666-4884-5.ch006.
Lim, D. H. and Johnson, S. D. (2002), ‘Trainee perceptions of factors that influence learning
transfer’, International Journal of Training and Development, 6, 1, 36–48.
Loftin, C. and West, H. (2017), ‘Evaluating self-efficacy after a team-based learning activity’,
Journal of Physician Assistant Education, 28, 2, 96–102.
Marler, J. H. and Liang, X. (2012), ‘Information technology change, work complexity and service
jobs: a contingent perspective’, New Technology: Work and Employment, 27, 2, 133–46.
Marler, J. H., Liang, X. and Dulebohn, J. H. (2006), ‘Training and effective employee information
technology use’, Journal of Management, 32, 5, 721–43.
Michaelsen, L. and Sweet, M. (2008), ‘The essential elements of team-based learning’, New
Directions for Teaching and Learning, 2008, 116, 7–27.

The power of the team  151

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt
Nordstrom, C., Wendland, D. and Williams, K. (1998), ‘“To err is human”: an examination of
the effectiveness of error management training’, Journal of Business and Psychology, 12, 3,
269–82.
O’Malley, K., Moran, B., Haidet, P., Schneider, V., Morgan, R., Kelly, P. and Richards, B. (2003),
‘Validation of an observation instrument for measuring student engagement in health
professions settings’, Evaluation and the Health Professions, 26, 1, 86–103.
Perino, C. and Michael, T. (2013), ‘SAP training trends’, Training, 50, 5, 12–4.
Reisslein, J., Atkinson, R., Seeling, P. and Reisslein, M. (2006), ‘Encountering the expertise re-
versal effect with a computer-based learning environment on electrical circuit analysis’,
Learning and Instruction, 16, 2, 92–103.
Saks, A. and Belcourt, M. (2006), ‘An investigation of training activities and transfer of training
in organizations’, Human Resource Management, 45, 4, 629–48.
Salas, E., Tannenbaum, S., Kraiger, K. and Smith-Jentsch, K. (2012), ‘The science of training and
development in organizations’, Psychological Science in the Public Interest, 13, 2, 74–101.
Sharma, R. and Yetton, P. (2007), ‘The contingent effects of training, technical complexity, and
task interdependence on successful information systems implementation’, MIS Quarterly,
31, 2, 219–38.
Sternberg, R. J. and Frensch, P. A. (1992), ‘On Being an Expert: A Cost–Benefit Analysis’, in R. R.
Hoffman (ed), The Psychology of Expertise (New York: Springer-Verlag), pp. 191–203.
Valentine, J. C. and Cooper, H. (2003), Effect Size Substantive Interpretation Guidelines: Issues in the
Interpretation of Effect Sizes (Washington, DC: What Works Clearinghouse).
Wegner, E. (2000), ‘Communities of practice and social learning systems’, Organization, 7, 2,
225–46.

152    International Journal of Training and Development

© 201 9 Brian Towers (BRITOW) and John Wiley & Sons Lt