Content uploaded by Tabea Augner
Author content
All content in this area was uploaded by Tabea Augner on Jan 09, 2024
Content may be subject to copyright.
Working Out Loud: an
intervention study to test
an agile learning method
Tabea Augner and Carsten C. Schermuly
Department of Business Psychology, SRH Berlin University of Applied Sciences,
Berlin, Germany, and
Franziska Jungmann
Department of Business Psychology, International School of Management,
Berlin, Germany
Abstract
Purpose –Today’s unpredictable and fast-changing work environment challenges researchers and
organizations to rethink learning. In contrast to traditional learning designs, new learning frameworks such
as agile learning are more learner centered, integrated into the workplace and socially shaped. The purpose of
this study is to examine Working Out Loud (WOL) as an agile learning method.
Design/methodology/approach –This intervention study used a pre–post and six-month follow-up
design (N¼507) to evaluate the effects of WOL on learners’vigor (affective outcome), WOL behavior
(behavioral outcome) and psychological empowerment (cognitive outcome) at work.
Findings –The authors compared the three longitudinal measurements using multilevel modeling. Results
revealed that WOL could significantly increase learners’WOL behavior and psychological empowerment at
work in the post and six-month follow-up measurements. No effect was found on learners’vigor at work.
Originality/value –This study highlights the need for research on new, more agile learning frameworks
and discusses their relevance to the literature. Agile learning frameworks enable learners to be more
autonomous and flexible, allowing them to better adapt to changing environmental demands.
Keywords Intervention study, Agile learning, New learning, Working Out Loud
Paper type Research paper
Introduction
Today’s rapidly changing business environment challenges modern organizations to
constantly adapt and remain flexible to keep pace with international competition (Decius
et al.,2022). Employees must be willing to change jobs flexibly and acquire new skills as
needed (van Laar et al., 2020). To survive and thrive in such a world, leading organizations
need to focus on the continuous development of their employees (Muzam et al.,2023).
Employees’continuous and self-directed development has consequently become a key
competence (Kortsch et al., 2019). This places a new focus on learning within organizations.
Learning is defined as an engagement in mental processes that leads to acquiring and
retaining skills, knowledge and affect over time (Kraiger and Ford, 2021).
Three primary work-related learning contexts can be distinguished: “on the job,”as
informal learning behavior that occurs more casually in the work process; “near the job,”
which leads to learner-planned, self-directed learning; and “off the job,”as a more structured
or formal method of learning (Decius et al., 2022). Organizations, as well as researchers, tend
to focus on “off the job”learning, through formal development programs such as training or
Agile learning
method
Received 9 May 2023
Revised 8 September2023
Accepted 12 October2023
Journal of Workplace Learning
© Emerald Publishing Limited
1366-5626
DOI 10.1108/JWL-04-2023-0067
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/1366-5626.htm
seminars –but rapidly changing work demands or restricted time commitments hinder
companies from offering and employees from participating in these programs (Noe et al.,
2014). Furthermore, in today’s dynamic and volatile environment, organizations often have a
limited understanding of what learning content is required for their employees and when
and where it is best for them to learn (Decius et al., 2022). Only limited empirical research has
examined alternative approaches to these static programs (Armanious and Padgett, 2021).
While traditional “off the job”learning designs regard learners as passive actors acquiring
knowledge and skills, new learning frameworks are more learner driven, occur naturally in
the workplace and are socially influenced (Noe et al.,2014).
As a result, a new learning framework has been proposed, known as agile learning
(Deery and Deery, 2014), which refers to “the process through which individuals learn
following the agile principles”(Noguera et al., 2018, p. 112). With its origins in software
development, this new framework helps learners adapt more easily to rapid change (Deery
and Deery, 2014) and thus be better prepared to meet the challenges of modern workplaces.
The educational literature has already addressed agile learning in school and university
environments, with the agile manifesto in higher education, for example (Kamat, 2012).
However, no unified classification of agile learning in the work environment has been
established, although similar elements have been outlined in the research.
Accordingly, we reviewed relevant literature and identified four common elements of
agile learning. First, as the agile manifesto highlights the human factor by focusing on
teamwork, human behavior and continuous development (Beck et al.,2001), agile learning
places learners at the center by granting them a more self-directed role in the learning
process (Deery and Deery, 2014;Longmuß and Höhne, 2017;Noguera et al., 2018). Second,
most researchers emphasize the iterative design of agile learning (Deery and Deery, 2014;
Longmuß and Höhne, 2017;Noguera et al.,2018). As stated in the agile manifesto, the project
team works in short iterations, so-called sprints, delivering valuable software (Beck et al.,
2001). This iterative approach is also reflected in agile learning, where learning is gradually
divided and integrated into the workplace. The team thereby learns in alternating phases of
working and learning (Longmuß and Höhne, 2017). Third, collaboration and interaction
between team members, managers and customers are inherent parts of the agile manifesto
(Beck et al., 2001). As such, agile learning is also based on collaboration (Deery and Deery,
2014;Longmuß and Höhne, 2017;Noguera et al., 2018). Through social exchange, learners
create content and develop skills in a collaborative but competent environment (Lazorenko
and Krasnenko, 2020). Fourth, agility originates in the world of technology; the agile
manifesto was created by a small group of software industry leaders (Beck et al.,2001).
Accordingly, technology is also crucial in agile learning (Deery and Deery, 2014;Longmuß
and Höhne, 2017;Noguera et al.,2018). A suitable digital infrastructure supports the learners
in their learning processes, enabling them to share their knowledge and collaborate closely
(Longmuß and Höhne, 2017;Noguera et al., 2018).
Based on this previous research, we define four meta-principles characterizing agile
learning: self-direction, iteration, collaboration and technology. In terms of the three contexts
of work-related learning (Decius et al.,2022), we commonly classify agile learning methods
as “near the job,”first, because they lead learners to a more active, self-directed
understanding of their role in the learning process, and second, because the iterative
approach provides a semi-formal setting with working and learning phases that are
integrated into the workplace but still leave space for learning.
In this article, we aim to test this agile learning framework by investigating its impact on
learners. We operationalize agile learning using Working out loud (WOL), a learning method
developed by John Stepper (Stepper, 2020). WOL is a self-organized 12-week program that
JWL
fits the characteristics of agile learning, as it is self-directed, iterative, focuses on collaborative
learning and uses technology. Therefore, WOL seems to be a suitable instrumentalization to
test the agile learning framework. Furthermore, the high prevalence of WOL indicates the
critical need for more scientific research. A German study reported that approximately 20% of
the surveyed companies already used WOL within their organizations (Schermuly and Meifert,
2022), and a growing number of organizations are incorporating WOL as part of organizational
change programs (Stepper, 2020). WOL may thus represent a valuable contribution to
professional learning that should be further explored and understood.
We aim to contribute to research in several ways. First, we introduce WOL as an agile
learning method in the literature. On the one hand, the complex work environment forces
organizations to continuously develop their employees (Muzam et al., 2023); on the other, the
currently predominant “off the job”learning methods do not meet the demands of today’s
work context (Noe et al., 2014). We present WOL as a valuable approach to address these
current shortcomings –as an agile learning method, WOL is self-directed, can be integrated
into the workplace and builds on the fundamental concept of social exchange in learning.
Second, we test the effectiveness of WOL by evaluating its impact on three learning
outcomes at work (ABC –affective, behavioral and cognitive outcomes) in a pre–post design
(N¼507) and assessing its long-term effects in a six-month follow-up survey. To examine
the impact of WOL in the work context, we investigate its effects on three work-related
constructs: vigor (affective outcome), WOL behavior (behavioral outcome) and psychological
empowerment (cognitive outcome). Third, we provide practical contributions for companies
already using or planning to use WOL in the future.
The conceptual understanding of Working Out Loud
Understanding the construct requires differentiating between first, the behavior and,
second, the learning method to practice and develop this behavior. Although we focus
primarily on the WOL learning method –simplified as WOL –it is important to understand
its origins at the behavioral level.
The original term “Working Out Loud”was first identified by Williams (2010),who
described it as a behavior with two key elements: first, narrating work in blog posts or status
updates; and second, performing work transparently for others to see, follow and contribute to.
When people “work out loud,”they share how they approach their tasks, ask questions and
deliver results as they are being produced rather than waiting until a final result is ready to be
revealed. In 2014, John Stepper extended this understanding by establishing five WOL
principles: relationships, generosity, visible work, purposeful discovery and a growth mindset.
These principles shift the focus to people and relationships, clarifying that a purposeful
network can improve effectiveness and provide access to new opportunities. Thereby, Stepper
deepens the fundamental understanding and notes that making work visible is only one type of
contribution people can make to build trust and relatedness with other people (Stepper, 2020).
To train and learn this behavior, Stepper developed the WOL learning method, which
guides learners through a 12-week self-organized program. The method is captured in 12
weekly WOL guides, which provide orientation and guidance. In small groups, called “WOL
circles,”four to five circle members meet for 1 h per week over 12weeks. In these meetings,
circle members share their individually set goals and support each other in achieving their
diverse goals. Through various exercises, they build and structure a network outside the
circle that supports them. Each week, circle members learn to make different contributions
to the people in their network. Within small exercises to foster appreciation, attention,
visible work and vulnerability, they learn to cultivate trusting relationships that enhance
cooperation and access to resources and opportunities (Stepper, 2020).
Agile learning
method
Working Out Loud as an agile learning method
WOL demonstrates distinct characteristics of an agile learning method by addressing all
four above-mentioned meta-principles.
(1) Agile learning emphasizes the learner’s role by demonstrating a high degree of self-
direction (Deery and Deery, 2014;Longmuß and Höhne, 2017;Noguera et al.,2018).
WOL provides sufficient space for self-directed learning, as circle members choose
their learning goals and guide themselves independently –within their circles –
through the 12-week process. Simultaneously, however, they receive a systematic
framework and order for orientation through the WOL guides.
(2) Agile learning is iterative and divides the learning process into incremental steps that
can be embedded in the daily work context (Longmuß and Höhne, 2017). This allows
participants to learn from prior iterations and improve for upcoming ones (Noguera
et al., 2018). Likewise, WOL divides the learning process into 12 incremental weekly
steps. Circle members are encouraged to apply their learning between sessions by
contributing to their networks and deepening their connections over time.
(3) Agile learning is based on collaborative exchange in a competent environment
(Deery and Deery, 2014;Longmuß and Höhne, 2017;Noguera et al., 2018).
Consistent with the social learning approach within WOL, participants learn from
and with each other within their circles and networks and build meaningful
connections and support each other in the process.
(4) Agile learning aims to integrate technology into the learning process (Deery and
Deery, 2014;Longmuß and Höhne, 2017;Noguera et al., 2018). This is consistent
with WOL, as one type of contribution is sharing knowledge in internal and
external social networks. Additionally, virtual collaboration tools enable locally
distributed WOL circles to meet across organizational boundaries.
Overall, WOL fits the characteristics of agile learning and might therefore provide a fruitful path to
assess its contribution to professional learning. A graphic overview is presented in Figure 1.
Figure 1.
WOL characterized
as an agile learning
method
JWL
Testing the effects of Working Out Loud on various learning outcomes
Learning not only involves learners being able to do something they could not accomplish
previously (Kraiger et al., 1993); changes in affective and cognitive states are equally important
(Ford et al., 2009). To measure the effects of WOL, we used a classification scheme by Kraiger
et al. (1993) that differentiates between three broad categories: affective, behavioral and cognitive
learning outcomes (the ABCs of learning; Kraiger, 2002). We evaluated changes in all categories
by comparing measures before the completion of WOL, immediately after and six months later.
As WOL is a “near the job”learning method, we measured its effects at work rather than during
WOL itself. We thereby addressed Baldwin and Ford’s (1988) widely recognized “transfer
problem,”the gap between learning and sustained performance on the job. We further assumed
that, according to the transfer literature (Wexley and Latham, 2002), a higher immediate effect
occurs directly after completing WOL, which decreases again slightly over time:
Affective outcome: Affective outcomes include attitudes and motivations (Ford et al., 2009),
which are essential in the learning process, as they might determine learners’behavior and
performance (Gagn
e, 1984). We chose to measure the effect of WOL on circle members’
vigor at work. The literature has classified vigor as an important dimension of affective
experience, as vigor increases employees’work-role effectiveness through motivational
processes at work (Kanfer and Kantrowitz, 2002). Vigor is expressed in high levels of
energy and mental resilience; a willingness to exert effort; and perseverance, even when
facing difficulties (Schaufeli et al., 2006). As a context-specific construct, vigor results from
individuals’evaluations of events, objects and situations (Shirom, 2011).
We argue that WOL increases circle members’vigor at work via a two-step process. In the
first step, we assume that WOL increases circle members’vigor levels during WOL. Based
on the conservation of resources theory (Hobfoll, 2011), positive relationships enhance the
experience of vigor, as they create a positive gain spiral in which feelings of vigor and peer
social support synergize and reinforce each other (Shirom, 2011). WOL leads circle
members to build high-quality relationships within their circles and networks. The circle
members make valuable contributions to their network by sharing resources, such as
empathetic listening, appreciation and attention, or by reinforcing each other’s self-esteem
as they learn from and with each other (Stepper, 2020). These are all essential components
of high-quality interpersonal interactions (Carmeli et al.,2009) and comprise contributions
that can be made in person or with the help of technology, via email, intranet or social
networks (Stepper, 2020). These interpersonal interactions should occur over time (Bakker
and Xanthopoulou, 2009), which is given through the iterative approach of WOL over
12weeks. In the second step, we assume that these positive effects spill over into the
work context. According to the spillover theory, employees’experiences in the
workplace can extend to the non-work domain, and vice versa (Staines, 1980). Due
to the close connection between WOL and work –through self-directed, work-
related goals and relationships –we propose that WOL increases participants’
vigor at work. Given the two-step process outlined, we hypothesize the following:
H1. Circle members’vigor at work will significantly increase immediately after
completing WOL (post) and slightly decrease again in the long term (follow-up).
Behavioral outcome: Behavioral change comprises the performance of a behavior
that the learner has not previously exhibited or has exhibited ineffectively (Kraiger,
2002). As the WOL method aims to develop behavior, we compared the pre, post and
six-month follow-up measurements of WOL behavior at work. WOL behavior
Agile learning
method
comprises observable work performance alongside the creation of meaningful
connections in a supportive network (Stepper, 2020;Williams, 2010).
According to Bandura’s social learning theory, people learn through observing role models,
paying attention, retaining observed information and reproducing observed behavior
(Bandura, 1962). In WOL, circle members learn in a collaborative environment. They
observe and copy the behavior of other circle members and the people in their network
(Stepper, 2020). On this basis, we assume that learners should build or maintain lasting
mental models through WOL, depending on whether they fit information into existing
mental models and confirm them (mental model maintenance) or modify and restructure
their mental models to accommodate new information (mental model building;
Vandenbosch and Higgins, 1996). The learning literature has already highlighted the
importance of mental models, as they guide human behavior and reduce the menatl
workload in planning fure actions (Norman, 1983). We argue that during WOL, circle
members develop or extend specific mental models to perform WOL behaviors as they
become more experienced and continuously apply their newly learned behaviors in
different exercises. These mental models guide learners in their future behaviors and
further mature over the 12 weeks via the iterative approach of WOL. Thus, learners should
build and internalize WOL behavior, which should be maintained even after WOL has
been completed. Therefore, we propose the following:
H2.Circlemembers’WOL behavior at work will significantly increase immediately after
completing WOL (post) and slightly decrease again in the long term (follow-up).
Cognitive outcome: Cognitive outcomes encompass beliefs, knowledge structures
and thoughts (Breckler, 1984). In terms of cognitive change, we measured differences in
circle members’psychological empowerment at work. Previous research has demonstrated
positive associations between psychological empowerment and various employee outcomes,
such as higher job satisfaction or lower turnover intention (see Seibert et al., 2011, for meta-
analytic findings). Psychological empowerment comprises four cognitions: meaning,
competence, self-determination and impact (Spreitzer, 1995). Compared to a personality trait
that can be generalized across different situations, psychological empowerment represents a
cognition shaped by the environment (Thomas and Velthouse, 1990). Organizational
conditions, such as decision-making autonomy and responsibility, help employees feel more
appreciation and thus experience a sense of empowerment (Safari et al., 2011).
We propose that WOL acts as a launching point for developing psychological
empowerment by triggering self-initiated changes at work. WOL drives individual
development. This newly developed skillset might initiate a change in the work
context, increasing the perception of psychological empowerment at work. The
underlying mechanism could be explained in terms of the four dimensions of
psychological empowerment. First, WOL enables circle members to choose their own
learning goals. A close connection between work and the chosen learning goals in
WOL could trigger participants to engage with and question their work differently.
This reflection on their position can increase the sense of meaning in the work context.
Second, WOL is a method to learn a new topic or skill; this new knowledge can be
applied in the work context, increasing the experience of competence in the workplace.
Third, WOL provides a high degree of autonomy. Through the self-organization of
WOL, the circle members learn to organize, motivate and discipline themselves. This
newly learned skillset can be applied in the work context, triggering change and
JWL
thereby strengthening the sense of self-determination. Fourth, circle members learn to
make their work visible through various exercises while receiving feedback and
commenting on others’work. This approach can be applied in the work context and
helps individuals recognize their impact by allowing them to regularly evaluate the
immediate results of their work (Kraimer et al.,1999). Following the arguments
presented, we hypothesize the following:
H3.Circlemembers’psychological empowerment at work –comprising the four
cognitions of meaning, competence, self-determination and impact –will significantly
increase immediately after completing WOL (post) and slightly decrease again in the
long term (follow-up).
Method
Design
The study was a single-arm trial to explore the effects of WOL on different learning
outcomes in a pre–post and six-month follow-up design. The pre-measurement occurred
within the first week of WOL, the post-measurement within the last week and the follow-up
measurement sixmonths later. WOLwas part of a campaign to promote women launched in
Germany in January 2021. Although the campaign was mainly directed toward women,
anyone could participate. It was primarily promoted via the professional networking
platform LinkedIn. Participation was free; in return, attendees were asked to complete self-
reported evaluation questionnaires via the platform SoSci Survey.
Sample
Within the first week of WOL, 1,354 participants completed the first survey. We excluded
107 cases with incomplete data sets. The resulting 1,247 pre-measurement cases were
matched to the post and follow-up measurements by individualized codes. To measure the
effects of WOL with at least two measurement points –one before and one after WOL –we
excluded 608 participants who did not answer either the post or six-month follow-up
measurement (dropout rate: 48.76%). Of the 639 resulting cases, 92 showed incomplete data
sets and were excluded. Because WOL comprises 12 weekly sessions, we also excluded 40
participants who reported attending fewer than one-quarter of the total sessions. The final
data set comprised 507 participants (97% women; age M¼41.53 years, SD ¼9.0) who
completed the pre-measurement and at least one other measurement: post or follow-up. Most
participants were employed (83%) and well educated (85%) with a university degree or
higher.
Measures
Vigor. To measure vigor, we used the vigor subdimension of the Utrecht Work Engagement
Scale developed by Schaufeli and colleagues (2006). Vigor describes the level of energy,
resilience and perseverance when facing difficulty. The short version comprises three items.
Participants indicate agreement on a five-point scale ranging from 1 (strongly disagree)to5
(strongly agree) to items such as “At my work, I feel bursting with energy.”Cronbach’s alpha
was between 0.91 and 0.93 for the three measurements.
WOL behavior. WOL behavior was measured using a subscale of the instrument
developed by Pearce (2014), who developed a scale to measure WOL behavior with two
dimensions: individual and group WOL. Because we focused on individuals, we only used
the subscale to measure WOL behavior on the individual level with three items, including “I
Agile learning
method
share my thoughts and ideas on social platforms with others beyond my immediate co-
workers.”The items focus on specific actions rather than feelings or opinions to capture
actual WOL behavior and not merely associated attitudes. Agreement was indicated on a
five-point scale ranging from 1 (strongly disagree)to5(strongly agree). Cronbach’s alpha was
between 0.85 and 0.92 for thethree measurements.
Psychological empowerment. Psychological empowerment was measured using a 12-item
questionnaire created by Spreitzer (1995). The scale comprises four dimensions: meaning,
self-determination, competence and impact. Example items include “The work I do is
meaningful to me”and “I am confident about my ability to do my job.”All items were scored
on a seven-point scale ranging from 1 (strongly disagree)to7(strongly agree). For the three
measurements, Cronbach’s alpha for overall empowerment score was between 0.90 and 0.92.
Data analysis
We ran three separate multilevel models to evaluate the development of vigor, WOL behavior
and psychological empowerment over time. The three models were estimated using a
multilevel modeling approach with the lmer function in the lme4 package in R (version 4.2.2;
R Core Team, 2022). The mixed-effects models comprised two levels, with repeated
measurements of vigor, WOL behavior and psychological empowerment (Level 1 ¼time)
nested within the participants (Level 2 ¼individual). Nesting the data in WOL circles, as a
third level, was considered but deemed unsuitable due to an insufficient amount of
participants per circle.
We examined different nested models with increasing levels of complexity per outcome
variable. Starting with a “null model,”we estimated the sources of variance in outcome
variables at the occasion level (within participants) and participant level (between participants)
and used this baseline model to determine whether the model’sfit to the data improved. We
then included time as a Level 1 predictor and the control variables age and intensity, as the
number of attended WOL sessions. We selected the best-fitting model by successively
comparing the Akaike information criterion (AIC) and Bayesian information criterion (BIC)
values of the competing models. To test our hypotheses, we performed post-hoc multiple
comparison tests using the multcomp package with Tukey contrasts in R (Hothorn et al., 2008).
Results
Descriptive statistics and correlations. Table 1 presents descriptive statistics and correlations
for the main study variables. The stability correlations for vigor, WOL behavior and
Table 1.
Means, standard
deviations and
correlations
Variable MSD 12345678
1. Vigor (pre) 3.72 0.88
2. Vigor (post) 3.76 0.87 0.78**
3. Vigor (follow-up) 3.52 0.91 0.46** 0.50**
4. WOL behavior (pre) 2.28 1.08 0.28** 0.23** 0.19*
5. WOL behavior (post) 2.74 1.00 0.25** 0.29** 0.20* 0.72**
6. WOL behavior (follow-up) 2.52 1.16 0.24* 0.24* 0.37** 0.70** 0.80**
7. Psych. empowerment (pre) 5.69 0.89 0.65** 0.56** 0.27** 0.30** 0.26** 0.21
8. Psych. empowerment (post) 5.83 0.85 0.62** 0.69** 0.39** 0.27** 0.33** 0.19 0.76**
9. Psych. empowerment (follow-up) 5.84 0.89 0.49** 0.50** 0.54** 0.25** 0.33** 0.38** 0.58** 0.63**
Notes: M and SD are used to represent mean and standard deviation. * p<0.05; ** p<0.01
Source: Authors’own work
JWL
psychological empowerment across the three measurements were moderate to high (Cohen,
1977).
Multilevel modeling and post hoc comparison on vigor. To test whether WOL could
increase circle members’vigor at work (H1), we started our analysis with the unconditional
means, or “null model.”A model with no predictor was used to examine the relative amounts
of within- and between-person variance in vigor (Model 1), AIC ¼2,543.4, BIC ¼2,558.5.
The intraclass correlation was 0.69, indicating that 69% of the variance in vigor was due to
differences between people, while the other 31% was due to differences within people. We
then added time as a predictor (Model 2), AIC ¼2,527.0, BIC ¼2,552.2, which improved the
model fit, as indicated by lower AIC andBIC values. Finally, we added age (Model 3), AIC ¼
2,505.9, BIC ¼2,536.2 and intensity (Model 4), AIC ¼2,507.4, BIC ¼2,542.7, as control
variables. While age improved the model fit, no improvement was found for intensity, so our
final model included only age as a control variable(Model 3).
Regarding vigor, a post hoc comparison between the three measurements revealed that
participants’vigor did not significantly increase after completing WOL (pre–post: z¼1.173,
p¼0.460), and that the vigor levels even decreased compared to follow-up measurement
(pre-follow-up: z¼–4.608, p<0.001; post-follow-up: z¼–3.931, p<0.001; see Table 2 for
more details). The visual inspection, shown in Figure 2, indicates that vigor at work
remained stable from pre- to post-measurement but decreased at follow up. Thus, we found
no support for H1. A significant effect of age (b¼0.02, p<0.001) revealed that older
participants scored higher on vigor than younger participants.
Multilevel modeling and post hoc comparison on WOL behavior. In terms of an increase in
circle members’WOL behavior (H2), the null model with no predictor in WOL behavior
(Model 1), AIC ¼2,923.1, BIC ¼2,938.1, revealed an intraclass correlation of 65%. Adding
time as a predictor (Model 2), AIC ¼2,776.4, BIC ¼2,801.4, improved the model fit, as did
controlling for age (Model 3), AIC ¼2,758.3, BIC ¼2,788.3. However, no improvement was
found by including intensity in the model (Model 4), AIC ¼2,756.5, BIC ¼2,791.5. Therefore,
we performed the post hoc comparison with Model 3.
Regarding WOL behavior, the post hoc comparison revealed a significant increase after
completing WOL in the short (pre–post: z¼13.073, p<0.001) and long term (pre-follow-up:
z¼3.921, p<0.001). The decrease from post-measurement to follow-up was not significant
Table 2.
Summary of the post
hoc multiple
comparisons between
the three
measurements
Estimate SE z p
Vigor
Pre–post 0.036 0.031 1.173 0.460
Pre–follow–up –0.207 0.053 –3.931 <0.001
Post–follow–up –0.243 0.053 –4.608 <0.001
WOL behavior
Pre–post 0.453 0.035 13.073 <0.001
Pre–follow–up 0.288 0.073 3.921 <0.001
Post–follow–up –0.166 0.073 –2.266 0.056
Psychological empowerment
Pre–post 0.138 0.029 4.707 <0.001
Pre–follow–up 0.121 0.050 2.416 0.039
Post–follow–up –0.017 0.050 –0.344 0.935
Note: Tukey post hoc comparison
Source: Authors’own work
Agile learning
method
(post-follow-up: z¼–2.266, p¼0.056; see Table 2 for more details). As the visualization in
Figure 2 demonstrates, participants’WOL behavior scores increased from pre- to post-
measurement and decreased at follow up. Thus, we found support for H2. The significant
effect of age (b¼0.02, p<0.001) revealed that older participants scored higher on WOL
behavior than younger participants.
Multilevel modeling and post hoc comparison on psychological empowerment. Concerning
the development of psychological empowerment through WOL (H3), the null model with no
predictors (Model 1), AIC ¼2,477.9, BIC ¼2,493.0, revealed an intraclass correlation of 71%.
The model fit improved when adding time (Model 2), AIC ¼2,458.8, BIC ¼2,484.0; and age
(Model 3), AIC ¼2,430.1, BIC ¼2,460.3, as predictors. As with vigor and WOL behavior,
including intensity (Model 4), AIC ¼2,429.7, BIC ¼2,465.0, did not improve the model fit.
We used Model 3 to perform a post hoc comparison.
Regarding psychological empowerment, a post hoc comparison revealed significantly
higher levels of psychological empowerment after completing WOL –with higher scores in
the short (pre-post: z¼4.707, p<0.001) and long term (pre-follow-up: z¼2.416, p<0.05).
No significant differences were found between post-measurement and follow up (post-
follow-up: z¼–0.344, p¼0.935; see Table 2 for more details). The visual inspection shown
in Figure 2 confirms that psychological empowerment scores increased from pre- to post-
measurement and remained stable at follow up. In H3, we assumed higher levels in post-
measurement, but a slight decrease at follow up. As the effects remained stable in the follow
up, H3 was only partially supported. The significant effect of age (b¼0.02, p<0.001)
suggests that older participants report higher psychological empowerment scores than
younger participants.
Discussion
Our study aimed to introduce WOL as an agile learning method and measured its impact on
three learning outcomes at work: learners’vigor (affective outcome), WOL behavior
(behavioral outcome) and psychological empowerment (cognitive outcome). In a single-arm
trial, we compared the pre, post and six-month follow-up measurements of 507 WOL
participants. The results indicated that WOL significantly increased participants’WOL
behavior and psychological empowerment at work, with higher levels immediately after
Figure 2.
Intervention effects of
WOL on vigor, WOL
behavior and
psychological
empowerment
JWL
completing WOL (pre–post). While the effect remained stable for psychological
empowerment in the follow-up measurement (post-follow-up), WOL behavior slightly
decreased again in the long term (post-follow-up) but remained significantly higher
compared to the pre-measurement (pre-follow-up). Contrary to our assumptions, we found
no significant effect of WOL on learners’vigor at work.
Theoretical implications
The findings of our study expand upon prior research in several ways.
First, we propose an alternative approach to traditional static training by introducing WOL
as an agile learning method. Thereby, we address shortcomings in current static “off the
job”training, which does not fit modern corporate learning needs. The agile learning
framework offers a fruitful contribution to professional learning, as it is learner-driven,
flexible and fosters social learning within learning communities (Deery and Deery, 2014). As
WOL as a learning method has emerged and evolved from practice, its theoretical
foundation is limited. However, this shortcoming was addressed by highlighting the
thematic proximity between WOL and agility. WOL emphasizes the importance of human
connections (Stepper, 2020), just as agility emphasizes the human factor in software
development (Beck et al.,2001). WOL’s alignment with the four characteristics of agile
learning made it an appropriate operationalization for testing the agile learning framework.
Future research on WOL should assess its impact on other constructs, such as job
satisfaction or commitment, and consider collecting data weekly through diary studies to
identify its dynamic consequences. To get a more comprehensive picture of agile learning,
other learning methodologies should be examined alongside WOL, and research from work
and organizational psychology should be incorporated to better understand how individuals
learn in the context of agile learning.
Second, we examined the effects of WOL on learners’vigor, WOL behavior and
psychological empowerment at work. As we measured the three constructs in the work
environment, we placed WOL in the context of workplace behavior and tested its
effectiveness as an agile learning method. In the following, we discuss our results regarding
the three learningoutcomes and provide implications for future research.
Regarding learners’vigor at work, our results showed no effect, with even lower scores in
the follow-up measurement. As we did not find any differences in the pre–post comparison, we
assume that the lower scores six months after the completion of WOL (follow-up) were not due
to WOL. A recent long-term study found that participants’vigor levels declined during the
COVID-19 pandemic (Richardson et al.,2022). Accordingly, a statistic from a German health
insurance company showed an increase in sick days in October and November 2021 compared
to January and April 2021 (AOK, 2022). Future research could examine whether WOL impacts
vigor when circle members work in the same organization and thus can meet face to face. In
this study, WOL was part of a nationwide campaign to promote women. Therefore, the circles
consisted of participants who were locally distributed and could only interact through virtual
collaboration tools. This might have influenced the development of vigor.
Concerning learners’WOL behavior, we found a positive trend, with higher levels
immediately after WOL (post) and a slight decline six months later (follow-up). This aligns
with the transfer literature (Wexley and Latham, 2002). New learnings are applied
immediately following the training, leading to stronger effects in post-measurement.
However, these effects decrease over time if learners are unable or less motivated to recall
and apply the new learnings (Velada et al., 2007). Because WOL is a method to develop WOL
behavior, the strongest effect occurred for this construct. Future research could examine the
Agile learning
method
impact on similar constructs, such as knowledge sharing, and use more objective measures
such as ratings from teammates or managers in addition to self-reported questionnaires.
In terms of learners’psychological empowerment, we found positive effects of WOL in the
short term (post), which remained stable in the long term (follow-up). The significant increase in
learners’empowerment is rather small. However, compared to targeted empowerment
programs, WOL is not inherently designed to increase empowerment, and the participants
already had relatively high empowerment scores before WOL (M¼5.69; scale 1–7). Previous
research has found mixed results regarding the effectiveness of empowerment initiatives.
While one study on nurses reported an increase in psychological empowerment scores of
approximately 20% (Özbasand Tel, 2016), another clinical study found no effect on patients’
empowerment after attending an empowerment training (Alegría et al., 2008). This indicates
that further research is needed to more accurately interpret effect sizes and understand the
critical aspects that make empowerment modifiable. Furthermore, future research could test
the influence of moderators, such as organizational environment or culture.
Regarding all three learning outcomes, adding intensity –as the number of attended
WOL sessions –did not improve the model fit, presumably because we excluded
participants who attended less than one-quarter of the 12 sessions, and the majority of the
remaining participants attended between 10 and all 12 sessions (M¼11.36).
Practical implications
Previous research has shown that modern professionals are more likely to remain with
companies that offer various learning opportunities (D’Amato and Herzfeldt, 2008), turning
workplace learning into a useful instrument for talent retention (Muzam et al.,2023). This
has practical implications for human resources and organizational development. As
employees only learn when it is relevant and appealing, organizations need to constantly
adapt their learning capabilities (Muzam et al., 2023). The flexibility of implementing WOL
could offer a viable alternative to static “off the job”learning. The hours spent on WOL (12 h
plus preparation time) are spread over three months and can easily be integrated into their
daily work routines by the employees themselves. In addition, WOL is highly scalable. Due
to WOL’s self-organization, employees can independently guide themselves through the
12 weeks. Human resources departments do not have to organize a trainer or coach, nor do
they have to cover hotel and travel expenses. Furthermore, WOL could be integrated into
existing organizational processes, such as onboarding or change programs. Our findings
regarding the positive impacts on WOL behavior and psychological empowerment also
suggest that WOL could be used as part of knowledge sharing programs or empowerment
initiatives.
Limitations
While the results of our study are promising, limitations should not be dismissed: first, our
study design included no control condition. Although we had a relatively large sample size
(N¼507) and chose a reasonable period of six months for the follow-up measurement, the
results should be interpreted cautiously and not generalized. In future studies, randomized
controlled designs are needed to evaluate the efficacy of WOL more generally. Additionally,
diverse samples with an equal distribution of men and women should be used to identify
gender differences.
Second, we did not nest our multilevel model on a third level, the WOL circles. As agile
learning is generally based on collaboration, WOL relies on collaboration within the circle. Circle
members learn from and with each other; therefore, the circle could influence the development of
the learning outcome. During data collection, each circle received a unique number for
JWL
identification. However, this identification number was requested voluntarily and therefore not
provided by all participants. Nesting the data in WOL circles as a third level was deemed
unsuitable because the remaining participants who provided their identification number were
mainly in different circles, leaving an insufficient amount of participants per circle. Future
research should therefore nest the data within WOL circles for a more holistic picture and
investigate the influence of group variables such as psychological safety or trust on the
development of participants’learning outcomes.
References
Alegría, M., Polo, A., Gao, S., Santana, L., Rothstein, D., Jimenez, A., Hunter, M.L., et al. (2008),
“Evaluation of a patient activation and empowerment intervention in mental health care”,
Medical Care, Vol. 46 No. 3,pp. 247-256, doi: 10.1097/MLR.0b013e318158af52.
AOK (2022), “Krankenstand in der COVID-19-Pandemie”, available at: www.aok-bv.de/presse/
medienservice/politik/index_25983.html
Armanious, M. and Padgett, J.D. (2021), “Agile learning strategies to compete in an uncertain business
environment”,Journal of Workplace Learning, Vol. 33 No. 8, pp. 635-647, doi: 10.1108/JWL-11-
2020-0181.
Bakker, A.B. and Xanthopoulou, D. (2009), “The crossover of daily work engagement: test of an actor–
partner interdependence model”,Journal of Applied Psychology, Vol. 94 No. 6, pp. 1562-1571, doi:
10.1037/a0017525.
Baldwin, T.T. and Ford, K.J. (1988), “Transfer of training: a review and directions for future research”,
Personnel Psychology, Vol. 41 No. 1, pp. 63-105,doi: 10.1111/j.1744-6570.1988.tb00632.x.
Bandura, A. (1962), “Social learning through imitation”, in Jones, M.R. (Ed.), NE Symposium on
Motivation, 1962, University of NE Press, Oxford, England, pp. 211-274.
Beck, K., Beedle, M., Van Bennekum, A., Cockburn, A., Cunningham, W., Fowler, M., Grenning, J., et al.
(2001), “Manifesto for agile software development”, available at: https://agilemanifesto.org/
Breckler, S.J. (1984), “Empirical validation of affect, behavior, and cognition as distinct components of
attitude”,Journal of Personality and Social Psychology, Vol. 47 No. 6, pp. 1191-1205, doi: 10.1037/
0022-3514.47.6.1191.
Carmeli, A., Ben-Hador, B., Waldman, D.A. and Rupp, D.E. (2009), “How leaders cultivate social capital
and nurture employee vigor: implications for job performance”,Journal of Applied Psychology,
Vol. 94 No. 6, pp. 1553-1561, doi: 10.1037/a0016429.
Cohen, J. (1977), Statistical Power Analysis for the Behavioral Sciences, Academic Press, New York, NY,
doi: 10.1016/C2013-0-10517-X.
D’Amato, A. and Herzfeldt, R. (2008), “Learning orientation, organizational commitment and talent
retention across generations: a study of European managers”,Journal of Managerial Psychology,
Vol. 23 No. 8, pp. 929-953, doi: 10.1108/02683940810904402.
Decius, J., Kortsch, T., Paulsen, H. and Schmitz, A. (2022), “Learning what you really, really want:
towards a conceptual framework of new learning in the digital work environment”,proceedings
of the Annual HI International Conference on System Sciences, pp. 5231-5240, doi: 10.24251/
hicss.2022.637.
Deery, K. and Deery, D. (2014), “The agile learner –using new technologies and social networks to
make learning a lifestyle, not an event”,Proceedings of the European Distance and E-
Learning Network 2014 Annual Conference,Zagreb, pp. 487-492, doi: 10.38069/edenconf-
2014-ac-0057 THE.
Ford, J.K., Kraiger, K. and Merritt, S.M. (2009), “An updated review of the multidimensionality
of training outcomes: New directions for training evaluation research”,inKozlowski,S.
W.J. and Salas, E. (Eds), Learning, Training, and Development in Organizations,
Agile learning
method
Routledge/Taylor and Francis Group, New York, NY, pp. 135-165, doi: 10.4324/
9780203878385.
Gagn
e, R.M. (1984), “Learning outcomes and their effects: useful categories of human
performance”,American Psychologist, Vol. 39 No. 4, pp. 377-385, doi: 10.1037/0003-
066X.39.4.377.
Hobfoll, S.E. (2011), “Conservation of resource caravans and engaged settings”,Journal of
Occupational and Organizational Psychology, Vol. 84 No. 1, pp. 116-122, doi: 10.1111/j.2044-
8325.2010.02016.x.
Hothorn, T., Bretz, F. and Westfall, P. (2008), “Simultaneous inference in general parametric models”,
Biometrical Journal, Vol. 50 No. 3, pp. 346-363, doi: 10.1002/bimj.200810425.
Kamat, V. (2012), “Agile manifesto in higher education”,2012 IEEE 4th International Conference on
Technology for Education, pp. 231-232, doi: 10.1109/T4E.2012.49.
Kanfer, R. and Kantrowitz, T.M. (2002), “Emotion regulation: command and controlof emotion in work
life”, in Lord, R.G., Klimoski, R.J. and Kanfer, R. (Eds), Emotions in the Workplace:
Understanding the Structure and Role of Emotions in Organizational Behavior, Jossey-Bass,
San Francisco, Californien, pp. 433-472.
Kortsch, T., Schulte, E.M. and Kauffeld, S. (2019), “Learning@work: informal learning strategies of
German craft workers”,European Journal of Training and Development, Vol. 43 Nos 5/6,
pp. 418-434, doi: 10.1108/EJTD-06-2018-0052.
Kraiger, K. (2002), “Decision-based evaluation”, in Kraiger, K. (Ed.), Creating, Implementing, and
Managing Effective Training and Development: State-of-the-Art Lessons for Practice, Jossey-
Bass, San Francisco, Californien, pp. 331-375.
Kraiger, K. and Ford, J.K. (2021), “The science of workplace instruction: Learning and development
applied to work”,Annual Review of Organizational Psychology and Organizational Behavior,
Vol. 8 No. 1, pp. 45-72, doi: 10.1146/annurev-orgpsych-012420-060109.
Kraiger, K., Ford, J.K. and Salas, E. (1993), “Application of cognitive, skill-based, and affective theories
of learning outcomes to new methods of training evaluation”,Journal of Applied Psychology,
Vol. 78 No. 2, pp. 311-328, doi: 10.1037/0021-9010.78.2.311.
Kraimer, M.L., Seibert, S.E. and Liden,R.C. (1999), “Psychological empowerment as a multidimensional
construct: a test of construct validity”,Educational and Psychological Measurement, Vol. 59
No. 1, pp. 127-142, doi: 10.1177/00131649921969785.
Lazorenko, L. and Krasnenko, O. (2020), “Applying agile learning to teaching English for specific
purposes”,International Journal of Learning, Teaching and Educational Research, Vol. 19 No. 9,
pp. 238-258, doi: 10.26803/ijlter.19.9.13.
Longmuß, J. and Höhne, B.P. (2017), “Agile learning for vocationally trained expert workers: expanding
workplace-based learning one sprint at a time”,Procedia Manufacturing, Vol. 9, pp. 262-268, doi:
10.1016/j.promfg.2017.04.003.
Muzam, J., Bendkowski, J., Mah, P.M. and Mudoh, P. (2023), “The state-of-the-art of modern
workplace learning: an applied assessment method of a pre-train deep learning on modern
learning tools”,The Learning Organization, Vol. 30 No. 5, pp. 607-629, doi: 10.1108/TLO-01-
2023-0019.
Noe, R.A., Clarke, A.D.M. and Klein, H.J. (2014), “Learning in the twenty-first-century workplace”,
Annual Review of Organizational Psychology and Organizational Behavior, Vol. 1 No. 1,
pp. 245-275, doi: 10.1146/annurev-orgpsych-031413-091321.
Noguera, I., Guerrero-Rold
an, A.E. and Mas
o, R. (2018), “Collaborative agile learning in online
environments: strategies for improving team regulation and project management”,Computers
and Education, Vol. 116, pp. 110-129, doi: 10.1016/j.compedu.2017.09.008.
Norman, D.A. (1983), “Some observations on mental models”, in Gentner, D. and Stevens, A.L. (Eds),
Mental Models, Lawrence Erlbaum Associates, Hillsdale, NJ, pp. 7-14.
JWL
Özbas, A.A. and Tel, H. (2016), “The effect of a psychological empowerment program based on
psychodrama on empowerment perception and burnout levels in oncology nurses: psychological
empowerment in oncology nurses”,Palliative and Supportive Care, Vol. 14 No.4, pp. 393-401, doi:
10.1017/S1478951515001121.
Pearce, D.E. (2014), Developing a Method for Measuring “Working out Loud”,ProQuest Dissertations
and Theses, doi: 10.13140/RG.2.2.10355.94243.
R Core Team (2022), “R: a language and environment for statistical computing”, R Projects, Vienna.
Richardson, D.L., Tallis, J., Duncan, M.J., Clarke, N.D. and Myers, T.D. (2022), “The ongoing effects of
the covid-19 pandemic on perceived physical activity, physical function and mood of older
adults in the U.K: a follow-up study”,Experimental Gerontology, Vol. 165, pp. 1-10, doi: 10.1016/j.
exger.2022.111838.
Safari, K., Haghighi, A.S., Rastegar, A. and Jamshidi, A. (2011), “The relationship between
psychological empowerment and organizational learning”,Procedia –Social and
Behavioral Sciences, Vol. 30, pp. 1147-1152, doi: 10.1016/j.sbspro.2011.10.224.
Schaufeli, W.B., Bakker, A.B. and Salanova, M. (2006), Utrecht Work Engagement Scale-9 (UWES-9),
[Database Record]. APA PsycTests., pp. 1-60, doi: 10.1037/t05561-000.
Schermuly, C.C. and Meifert, M. (2022), “Auf dem weg ins postagile zeitalter?”,Personalmagazin,
Vol. 9/21, pp. 24-30.
Seibert, S.E., Wang, G. and Courtright, S.H. (2011), “Antecedents and consequences of psychological
and team empowerment in organizations: a meta-analytic review”,Journal of Applied Psychology,
Vol. 96 No. 5, pp. 981-1003, doi: 10.1037/a0022676.
Shirom, A. (2011), “Vigor as a positive affect at work: conceptualizing vigor, its relations with related
constructs, and its antecedents and consequences”,Review of General Psychology, Vol. 15 No. 1,
pp. 50-64, doi: 10.1037/a0021853.
Spreitzer, G.M. (1995), “Psychological empowerment in the workplace: dimensions, measurement
and validation”,Academy of Management Journal, Vol. 38 No. 5, pp. 1442-1465, doi:
10.2307/256865.
Staines, G.L. (1980), “Spillover versus compensation: a review of the literature on the relationship
between work and nonwork”,Human Relations, Vol. 33 No. 2, pp. 111-129, doi: 10.1177/
001872678003300203.
Stepper, J. (2020), Working out Loud: A 12-Week Method to Build New Connections, a Better Career,
and a More Fulfilling Life, Page Two Books, Vancouver.
Thomas, K.W. and Velthouse, B.A. (1990), “Cognitive elements ofempowerment: an‘interpretive’model
of intrinsic task motivation”,The Academy of Management Review, Vol. 15 No. 4, pp. 666-681,
doi: 10.2307/258687.
van Laar, E., van Deursen, A.J.A.M., van Dijk, J.A.G.M. and de Haan, J. (2020), “Determinants of 21st-
century skills and 21st-century digital skills for workers: a systematic literature review”,SAGE
Open, Vol. 10 No. 1, pp. 1-14, doi: 10.1177/2158244019900176.
Vandenbosch, B. and Higgins, C. (1996), “Information acquisition and mental models: an investigation
into the relationship between behaviour and learning”,Information Systems Research, Vol. 7
No. 2, pp. 198-214, doi: 10.1287/isre.7.2.198.
Velada, R., Caetano, A., Michel, J.W., Lyons, B.D. and Kavanagh, M.J. (2007), “The effects of training
design, individual characteristics and work environment on transfer of training”,International
Journal of Training and Development, Vol. 11 No. 4, pp. 282-294, doi: 10.1111/j.1468-
2419.2007.00286.x.
Wexley, K.N. and Latham, G.P. (2002), Developing and Training Human Resources in Organizations,
Prentice Hall, Englewood Cliffs, NJ.
Williams, B. (2010), “When will we work out loud? Soon!”, The Bryce Write, available at: https://
thebryceswrite.com/2010/11/29/when-will-we-work-out-loud-soon/
Agile learning
method
About the authors
Tabea Augner has an MSc in Business Psychology and is currently a PhD candidate in cooperation
with the SRH Berlin University of Applied Sciences. Her research focuses on agile teams,
psychological empowerment and new learning approaches in modern work environments. Tabea
Augner is the corresponding author and can be contacted at: augnerta@srhk.srh.de
Carsten C. Schermuly is a Professor for business psychology and vice president for research and
transfer at the SRH Berlin University of Applied Sciences. He is the Director of the Institute for New
Work and Coaching (INWOC). His main research focuses on psychological empowerment, diversity
in teams and the efficacy of coaching processes.
Franziska Jungmann is a Professor for Business Psychology at the ISM International School of
Management Berlin. Her main research focuses on healthy aging at work, leadership of diverse teams
and design and evaluation of intervention studies.
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
JWL
