Team Empowerment, Uncertainty, and Performance
In New Product Development
By
Jiyao Chen
Richard R. Reilly
and
Gary S. Lynn
Wesley J. Howe School of Technology Management
Stevens Institute of Technology
ISBM Report 14-2005
Institute for the Study of Business Markets
The Pennsylvania State University
402 Business Administration Building
University Park, PA 16802-3004
(814) 863-2782 or (814) 863-0413 Fax
Team Empowerment, Uncertainty, and Performance in New Product
Development
Jiyao Chen Richard R. Reilly and Gary S. Lynn
Jiyao Chen
Wesley J. Howe School of Technology Management
Stevens Institute of Technology
1 Castle Point on Hudson, Hoboken, NJ 07030 USA
Tel: 201-216-8978
{jchen1| rreilly| glynn}@stevens.edu
1
Jiyao Chen received the B.S. in Pharmacy at Tongji Medical University and the M. Phil. in
Management of Science and Technology at Central South University of Technology in China.
He is currently pursuing the Ph.D. degree in Technology Management at Stevens Institute of
Technology. He is a member of the Academy of Management Association and the Product
Development Management Association. His latest research appears at IEEE Transactions of
Engineering and Management. He has also served as an ad-hoc reviewer for this journal. His
interest areas include new product development and entrepreneurship, uncertainty and
strategy choice.
Richard R. Reilly holds the Ph.D. from the University of Tennessee and is a Research
Professor in the Howe School of Technology Management. Before joining Stevens, Dr. Reilly
was a research psychologist for Bell Laboratories, the Educational Testing Service and AT&T.
He is on the Editorial Board of Personnel Psychology and a Fellow of the American
Psychological Association and the American Psychological Society.
He has over 60
Publications related to Organizational Behavior and Team Performance. Recent publications
include Blockbusters: The Five Keys to Developing Great New Products, HarperCollins, 2002;
How to Build a Blockbuster, Harvard Business Review, 80(10), 10-19.
Dr. Gary Lynn is the Distinguished Service Professor and tenured Associate Professor at
Stevens Institute of Technology. He has authored or co-authored 4 books and over 60 articles
on creating new products and new businesses. Dr. Lynn was selected as being one of the most
active and prolific scholars in the field of Technology-Innovation Management by the
International Association for the Management of Technology. He won the Best Associate
Professor Award at Stevens; Best Teacher and Best Researcher in the School of Technology
Management at Stevens; won the Merritt Williamson research award given by the American
Society for Engineering Management; and received the first Innovation Fellowship given by
Rubbermaid, Inc. In 2002, Dr. Lynn was selected by Business 2.0 Magazine as one of the
nine leading management gurus in the country. His most recent book, Blockbusters, was
selected as one of the key management publications in the world by Manageris of France and
one of the 30 best management books published last year by SoundView. Dr. Lynn actively
works with venture capital firms and start-up companies helping them develop and launch
blockbuster products.
2
Team Empowerment, Uncertainty, and Performance in New Product
Development
ABSTRACT
A recent theoretical analysis argues that the effectiveness of empowerment is contingent
upon the degree of uncertainty in the environment (Wall, Cordery and Clegg 2002). In a study
of 212 New Product Development (NPD) projects we found that the relationship between
team empowerment and speed-to-market was moderated by uncertainty stemming from
technological turbulence but not by uncertainty related to technological novelty. Further, we
found that the relationship was a u-shaped function under conditions of low uncertainty
whether the source was novelty or turbulence. The findings imply that NPD teams should be
either not empowered or fully empowered under conditions of low uncertainty. In contrast,
when uncertainty was high the empowerment-speed relationship differed depending upon the
source of the uncertainty. Under conditions of high technological novelty a linear relationship
between empowerment and speed was found, but under conditions of high technological
turbulence the relationship was reverse u-shaped. Under the latter conditions a quadratic
function indicated that very high levels of empowerment do not result in increasing speed-tomarket. Thus, a contingent view should apply to choose and implement the empowerment
approach. Our findings also indicate that turbulence may be a more important source of
uncertainty than newness with different implications for empowerment.
Key words: Team empowerment; Speed-to-market; Technological turbulence; Technological
novelty
3
INTRODUCTION
Several recent papers support the positive effects of employee and team empowerment
(Wellins, Byham and Wilson 1991; Hyatt and Ruddy 1997; Kirkman and Rosen. 1997;
Kirkman and Rosen. 1999). Other literature argues that the effects of empowerment are
inconsistent or none (Ford and Fottler 1995; Wilkinson 1998; Forrester 2000; Kirkman, Rosen,
Tesluk and GibsonI 2004). Wall et al. (2002) argue that the effectiveness of empowerment is
contingent upon the degree of uncertainty operating in the environment and that this
proposition generalizes across both levels of analysis and areas of application. In sum,
although some theoretical and empirical research supports the importance of empowerment, it
remains an open question as to whether greater empowerment is always better, especially
under differing conditions of uncertainty.
Recently, an increasing number of scholars and practitioners are becoming interested in
exploring the effects of team empowerment (TEMPOWER). TEMPOWER is becoming a
new managerial tool in the area of new product development and there are a paucity of studies
involving the effectiveness of empowering NPD teams (McDonough and Barczak 1991;
McDonough 2000). This study examined the influence of TEMPOWER on speed-to-market
under conditions of uncertainty stemming from technological novelty and technological
turbulence. Our objectives were to gain a better understanding of when empowerment should
and should not be used. We also want to clarify turbulence and uncertainty which have been
blurred and used interchangeably in previous literature (Aram and Walochik 1996; Anderson
and Tushman 2001; Kamoche and Cunha 2001; Tatikonda and Montoya-Weiss 2001).
THEORETICAL BACKGROUND
Empowerment Definition Empowerment has been defined in one of two ways. The first
4
discusses empowerment as a set of power-sharing managerial strategies, practices and
techniques (Thomas and Velthouse 1990; Ford and Fottler 1995; Wilkinson 1998; Eylon and
Bamberger 2000; Forrester 2000; Wall, Cordery et al. 2002). Essentially, empowerment as
managerial practices represents the antithesis of traditional Tayloristic thinking (Wilkinson
1998) which emphasize the concentration of decision making in the upper organizational
hierarchy and provides low discretion to other jobs. In contrast, empowerment delegates
discretion to lower organizational levels and enables employees to control their own tasks.
And then team empowerment is defined as the range of decisions the team is authorized to
make in order to fulfill its mission (Gerwin 1999). The second definition derives from a
cognitive perspective and denotes the motivational state of employees (Thompson 1967;
Thomas and Velthouse 1990; Spreitzer 1995 and 1996; Kirkman and Rosen. 1997; Eylon and
Bamberger 2000; Kirkman, Rosen et al. 2004). Spreitzer (1995) (P. 1443) noted what really
matters is the “empowerment experience” on the part of individual job incumbent. Spreitzer
(1995 and 1996) conceptualized empowerment in terms of multidimensional constructs
capturing four cognitions of individual orientation, namely, meaningfulness, competence, selfdetermination, and impact, according to Thomas and Velthouse (1990). However, Forrester
(2000) contends that one of reasons of the failure of empowerment program is “over-reliance
on a narrow psychological concept of empowerment” (P68). He argued that the notion
creating psychological empowerment as an enduring psychological feature of employees is
unrealistic.
In contrast, empowerment as a managerial practice carries clear, practical
meaning for managers and employees.
Uncertainty and Structure in Organizational Theory Since the 1960s, organizational
structure has been regarded by many as an externally caused phenomenon, or an outcome
rather than the earlier view of structure as an artifact (Mohr 1982). This view maintains the
demands of technology, environment and size exert spontaneous impacts on organizational
5
structuring with direct human choice playing a minor role (Mohr 1982). Moreover, a widely
accepted view contends that uncertainty is the key ingredient influencing organizational
structure: the more uncertainty resulting from the three factors (technology, environment and
size), the more the organizational will compensate by departing from traditional bureaucratic
structure toward a looser and more decentralized mode of operation (Thompson 1967). Mohr
argues that the group or task level is the proper unit of analysis for exploring the relationship
between uncertainty and structure. Although decentralization of authority is accepted as the
best response to uncertainty, empirical studies have not produced consistent results (Mohr
1982). Mohr then argues that the fundamental theoretical idea does not concern average
uncertainty and average bureaucratization, but a particular occurrence of bureaucratization or
decentralization that is a response to one source of uncertainty. Two types of decentralization
are possible: operational authority—the delegated right to carry out a certain assignment
without close supervision but with rather detailed guidelines for action and the possibility of
intermittent effective oversight from above, vs. true authority that occurs to the extent that the
hierarchical official cedes or loses the capacity to exercise effective oversight, and Mohr
contends it is the latter type of decentralization that is relevant to uncertainty. Following this
argument we contend that empowerment, or having “true authority”, is the type of
decentralization that may produce different results depending upon the source and level of
uncertainty.
Empowerment: Embracing Uncertainty
A traditional view on the attitude to uncertainty might be a problem. “Uncertainty appears as
the fundamental problem for complex organizations and coping with uncertainty, as the
essence of the administrative process” (Thompson 1967) (P.159). Based on the notion of
organizational rationality, Thompson tries to reconcile a closed system to an open
6
environment and proposes that organizations seek to seal off their technical core from
environment fluctuations through buffering, leveling or forecasting uncertainty (Thompson
1967). Thus, according to the Thompson’ view, the key function of the administrative process
is to reduce uncertainty. In contrast, considering uncertainty as both threat and opportunity
(Schumpeter 1934), we advocate changing the attitude from on in which uncertainty should
be reduced to embracing uncertainty (Clampitt and Dekoch 2001). Embracing uncertainty has
a lot of benefits: fighting overconfidence, reducing frustration, fostering learning and
flexibility, properly framing information, encouraging thoughtful decision making, and
cultivating innovation (Clampitt and Dekoch 2001). One way to embrace uncertainty is
through empowerment. Empowered teams have the authority and responsibility to make fast
decisions based on local information and can then be flexible to deal with environmental
uncertainty. Therefore, we posit that empowerment is more suitable under conditions of high
uncertainty and that the relationship between empowerment and performance will be stronger
when uncertainty is high than when uncertainty is low.
Uncertainty, Newness and Turbulence Uncertainty is variously defined as unpredictability
of the environment, inability to predict the impact of environmental change, and the inability
to predict the consequence of a response choice (Milliken 1987). There is some confusion in
the literature regarding uncertainty vs. turbulence. For example, Kamoche and Cunha (2001)
use turbulence and uncertainty interchangeably in discussing improvisational approaches to
NPD. Also, Arma and Walochik (1996) perceive uncertainty and change as the same thing.
Anderson and Tushman (2001) also “substitute uncertainty for dynamism” (P. 682) because
dynamism refers to change that is hard to predict and uncertainty is a central control construct
in economics, strategy and organization theory. Tatikonda & Montoya-Weiss make the case
that uncertainty can arise from several sources that are quite different. They comment, “The
product development literature describes technological novelty in terms of the degree of
7
familiarity with the given technology or degree of change in the technologies relative to
products previously developed or manufactured by the company” (p. 157). We argue that
newness and turbulence represent different dimensions of unpredictability and therefore can
not be treated synonymously (Nunez 2004). Our emphasis is on studying empowerment in
relation to uncertainty deriving from two sources: turbulence and newness. Partitioning these
two distinct sources is a major departure from past research.
Consistent with Tatikonda and Montoya-Weiss we define the degree of technological novelty
(TNOV) in terms of newness with the given technology or
the extent to which new
technology or new manufacturing processes are used (Lynn and Akgun 1998); however, we
depart from their approach in our definition of turbulence (TTURB). We believe that TNOV
and TTURB stem from different sources and must be treated as separate constructs. We define
technological turbulence (TTURB) as the rate of change associated with technology used to
develop new products in an industry (Jaworski and Kohli 1993). We hypothesize that the level
of environment uncertainty resulting from TNOV or TTURB will differentially affect the
impact of empowerment on SPEED.
TEAM EMPOWERMENT AND NPD TEAM PERFORMANCE
Research has addressed the influence of empowerment job satisfaction (Spreitzer, Kizilos and
Nason 1997; Kirkman and Rosen. 1999; Eylon and Bamberger 2000), organizational and team
commitment (Kirkman and Rosen. 1999), speed (McDonough and Barczak 1991), and
productivity (Wellins, Byham et al. 1991). Empowered teams are allowed to make decisions
more quickly (Spreitzer 1995), are more involved and motivated (Kirkman and Rosen. 1997)
and are more likely to have a positive orientation to their jobs (Kirkman and Rosen. 1999).
Forrester (2000) concludes that more empowerment will lead to greater self-efficacy – the
belief that goals can be accomplished. McDonough and his colleagues found that the amount
8
of team freedom and responsibility were significantly associated to NPD speed (McDonough
and Barczak 1991) and empowering the team with the decision-making power is one of key
factors of new product success (McDonough 2000). Thus, we believe that TEMPOWER will
result in better team performance (see fig. I Conceptual Model):
H1: TEMPOWER is positively associated with team performance.
Moderating Effects of Uncertainty on TEMPOWER Empowerment does not work in all
situations. Yap and Souder (1994) studied twelve small high-technology electronics firms, and
found no relationship between empowerment and new product success under high technical
uncertainty. On the other hand, an empirical test of 45 projects from 12 firms in a variety of
industries suggested that the use of empowered teams is positively related to faster
completion time, better product quality, and achievement of sales objectives on highly
innovative NPD projects, but not on routine ones (Olson, Warker and Ruekert 1995). Souder
and Song (1998) found that decentralized decision-making has a pivotal effect on product
success under low market familiarity conditions, but not under high market familiarity
conditions for US companies. Reilly, et al. (2003) found that flexible responses of empowered
teams are important to deal with environmental uncertainty in developing new products.
Eisenhardt and Tabrizi (1995) found that having an empowered team leader and flexibility
was most effective in an uncertain environment. The research suggests that uncertainty will
moderate the relationship between empowerment and team performance. We posit the
following hypotheses:
H2a-b: TEMPOWER will have a stronger positive effect on team performance when (a)
TNOV or (b) TTURB is high than when TNOV or (b) TTURB is low.
Empowerment: A Matter of Degree More empowerment is not always better. Forrester
(2000) also suggests that piecemeal empowerment solutions will not be successful. Partial
9
empowerment may create ambiguity and confuse team members. Under conditions of low
uncertainty, either from TNOV or TTURB, the tasks are relatively routine and the
requirements for creativity are low. Minimal empowerment under these conditions might be
better than moderate empowerment. On the other hand, a highly empowered team can quickly
make decisions with little of the ambiguity or confusion that occurs when empowerment is
moderate. We expect that:
H3a-b: TEMPOWER will have a curvilinear relationship with EMPOWER when (a) TNOV
or (b) TTURB is low.
However, the pattern of team empowerment and performance might be different when
uncertainty increases from low to high. The advantages of empowered teams in mitigating the
effects of uncertainty stem from quicker decisions resulting from allocating power to the local
resource, reducing the decision chain, and motivating involvement. Hence, for a project with
high TNOV or TTURB, the benefits of empowerment would include quicker decisions and
greater motivation for creativity and innovation (Reilly, Bentley et al. 2003). Hence we posit
a linear, positive relationship between TEMPOWER and performance under conditions of
high uncertainty:
H4a-b: TEMPOWER has a linear relationship with team performance when (a) TNOV or (b)
TTURB is high.
SAMPLING AND MEASUREMENT
To test our hypotheses, a questionnaire was developed based on previous research (Jaworski
and Kohli 1993; Day 1994; Lynn and Akgun 1998; Kessler and Bierly 2002). We selected a
contact person in a variety of technology-based companies in the northeast region of the U.S.
The contact person selected primarily product/project managers (34.9%), team leaders
10
(26.4%), or senior project members as respondents (14.2%). The remainder included
presidents, VPs or others. Each respondent was asked to select a completed NPD project with
which they were familiar. Each project had one respondent describing the project. Lukas and
Ferrell (2000) found that managers rely on their own self-reports and provide reliable and
objective data. Huber and Power (1985) noted that simply averaging multi-sources is likely
to be less accurate than using a key informant. After the selection of the respondents, they
were informed that their responses would remain anonymous and that their responses would
not be linked to a company or product name. This increased the motivation of informants to
cooperate without fear of reprisals.
To improve the accuracy of retrospective reports, recent projects were selected to eliminate
the elapsed time between the events of interest and the collection of data. Of 261 “contact
people” asked to participate, 212 returned the questionnaires (81.3% response rate). Our
response rate was high because most of our data were collected as a part of an executive
graduate management program where the students were requested to identify a company
contact who was intimately involved in a new product that was launched into the marketplace.
The respondents represented the following industries: telecommunications (34 projects,
16.0%), machinery manufacturing (22 projects, 10.4%), equipment and materials (6 projects,
2.8%), chemical manufacturing (15 projects, 7.1%), food manufacturing (8 projects, 3.8%),
pharmaceuticals (5 projects, 2.4%), government or defense (48 projects, 22.6%), computer
software (12 projects, 5.7%), information services (56 projects, 26.4%), and consumer
products (4 projects, 1.9%), and pet care (2 projects, 0.9%). To test for between-group
differences, we conducted ANOVAs on all of the constructs using industry as the independent
variable. ANOVAS were significant for EMPOWER and technological turbulence. Post-hoc
Tukey multiple comparison tests showed that the average level of technological turbulence in
telecommunication and information service is higher than in food manufactory (p<.05). Post-
11
hoc Tukey tests showed no significant differences for empowerment.
209 returned questionnaires were usable for this study. Consistent with our expectations for
technology-based firms, 40.4 % of the sample involved a new technology, 36.4% of the
sample involved several new technologies, and 9.1% of the sample involved non-proved or
non-existing technologies. The median team size was 12 people (the average team size was 24
people and S.D. was 40, the mode is 8). Most projects were from large companies. 63.5% of
the projects were at companies earning annual income over 500 million dollars. 27.8% of the
projects were at companies employing 500-5000 people, and 48.8% projects were at
companies employing over 5000 people. As firm size was measured by three categories, tests
for between-group differences in all constructs in this study were taken. Separate analysis of
variance procedures and post-hoc Tukey multiple-comparison procedures revealed that the
average of TTURB is significant lower at small firms than at large firms (P<.001) and the
average of TNOV is significant lower at the small firms than at the medium firms (P<.05). We
created a dummy variable to measure firm size, “0” represented small firms with less than 500
employees; “1” represented firms with more than 500 employees.
Each construct was measured using multiple items and a Likert-type 0 to 10 scale (0 strongly disagree to 10 - strongly agree). A three-item scale was used to measure
TEMPOWER: The core team had the authority to make most of the decisions that impact this
project; The core team did not have to consult senior company management for most of the
decisions that had to be made; The core team was empowered to fulfill its mission. In this
study, we focus on speed-to-market (SPEED) as a key measure of team performance. SPEED
describes how quickly an idea moves from conception to its first commercialization or introduction
into the marketplace. SPEED measures the capability to move quickly from ideas to actual products in
the marketplace. Since we used a multi-company and multi-industry sample, we tried to control
differences in the nature of projects by using relative speed measures. SPEED was assessed
12
relative to pre-set schedules, company standards, and similar competitive projects. The
approach was similar to that of Kessler and Bierly (2002). TTURB was measured by two
items: the technology used in the product was rapidly changing; the technology in the
industry was changing rapidly (Jaworski and Kohli 1993). TNOV was also measured by two
items: the technology required to development the product was totally new to our company;
the manufacturing process used was totally new to our company (Lynn and Akgun 1998).
Team size and firm size were regarded as potential control variables. Team size was measured
by the number of core team members at pre-prototype and at launch. Because the distribution
of team size was skewed, team size was transformed to the log of team size.
ANALYSIS AND RESULTS
A confirmatory factor analysis was conducted to assess the measurement model. The fit
statistics produced suggested that a four-factor model (i.e., SPEED, TEMPOWER, TTURB,
and TNOV) fit the data well. The X2 value (X2 (29) =38.5, P= .118) was not significant. The
fit indices also provided supportive evidence (RMSEA = .039, NFI=.970, CFI=.993, and
AGFI=.923). The results indicated our measurement model has construct validity. Internal
consistency reliability (Cronbach's alpha) is shown on the diagonal of the correlation matrix
in Table I. Reliabilities ranged from 0.68 for TNOV to 0.86 for empowerment, indicating
acceptable levels of internal consistency as suggested by Nunnally (1978). Intercorrelations
were moderate or low. The correlation between TTURB and TNOV was 0.49. Our results
show that discriminant validity between constructs is acceptable.
Before hypotheses testing, we centered the values of TEMPOWER, TNOV, and TTURB in
order to avoid multicollinearity. The maximum (absolute value) correlation was .513 for
EMPOWER vs. the quadratic term of TEMPOWER, indicating no evidence of
multicollinearity.
Because cross-product terms might share substantial variance with
13
quadratic terms of the predictor, the findings testing by the traditional moderator analysis
might be spurious (Lubinski and Humphreys 1990). We used the following approach to test
our hypotheses. In step 1, the control variables, team size and firm size, TEMPOWER, and
either TNOV or TTURB, respectively, were entered, in step 2, the quadratic term of
TEMPOWER was entered, in step 3, the cross-product of TEMPOWER and the
corresponding uncertainty, and in step 4, the cross-product of the quadratic term of
TEMPOWER and the corresponding uncertainty was entered. Table II presents the results of
these tests.
A significant positive relationship between TEMPOWER and SPEED, supporting H1. The
results in step 1 also suggest that TTURB positively influences SPEED but TNOV does not,
indicating SPEED is faster under conditions of high TTURB. The R2 change in step 2 is not
significant, indicating that TEMPOWER does not have an overall curvilinear relationship
with SPEED. The R2 changes in step 3 are significant for TTURB (P<.05) and marginally
significantly for TNOV (P<.10). H2a was supported and H2b partially supported. Interestingly,
the R2 changes in step 4 are significant for both TTURB and TNOV (P<.01). The findings
indicate TEMPOWER does have a curvilinear relationship with speed in certain
circumstances. Fig. II shows graphs illustrating the relationship between TEMPOWER and
SPEED as uncertainty moves from low to high. In each case the graph shows the relationship
for 1 sd above the mean, at the mean and 1 sd below the mean for uncertainty. The graphs
clearly show that the TEMPOWER- SPEED relationship changes from a U-shaped
relationship to a straight line to a curvilinear relationship as uncertainty increases.
To further test H3-H4, the sample was divided into the top and bottom third based on
percentile scores for TNOV and TTURB. Polynomial regression analyses were then
conducted in each subgroup. H3a-b was supported (P< .05). The relationship between
14
TEMPOWER and SPEED is u-shaped when TNOV or TTURB is low. The B-weight of the
quadratic term of empowerment is .126 under conditions of low TNOV and .097 under
conditions of low TTURB. The relationship between TEMPOWER and SPEED is curvilinear
when TTURB is high but not when TNOV is high. The B-weight of the quadratic term for
empowerment is -.10 (P<.05). H4a was supported while H4b was not.
DISCUSSION AND IMPLICATIONS
Our findings indicate that TEMPOWER, overall, is positively related to speed. More
importantly, our results demonstrate that uncertainty has important effects on the
TEMPOWER- SPEED relationship, that is, the level and sources of uncertainty do not
influence the strength but also the form of the relationship between TEMPOWER and SPEED.
Generally speaking, TEMPOWER has stronger influence on SPEED under conditions of high
uncertainty than low uncertainty. This result gives a support that it is important to change our
attitude from reducing uncertainty (Thompson 1967) to embracing uncertainty in the current
fast-changing environment (Clampitt and Dekoch 2001) and TEMPOWER is one way to
embrace uncertainty. Our analysis suggests that under the conditions of low TNOV or
TTURB, TEMPOWER is less important and has a u-shaped relationship with SPEED as
expected. The implication is that when uncertainty is low management should either not
empower or fully empower the NPD team. Under low levels of technology uncertainty, NPD
teams are familiar with the technology employed and the technology environment is relative
stable, thus, the task of NPD is relatively routine. A clear distribution of power, either controloriented or empowerment-oriented will result in a more effective pattern in making decisions.
In contrast, moderate empowerment may be a compromise between the level of control
comfortable for both management and employees. On one hand, companies are “supposed” to
empower their employees, because it causes it employees to be more responsive, and
15
responsible. But on the other, senior executives are reluctant to empower their people because
of loss of control, etc., The compromise may result in role ambiguity for both parties leading
and is consistent with Fort and Fotter’s (1995) notion that top managers may be frustrated in
managing the delicate balance in empowering employees without at the same time losing
control over what employees do. The finding that empowerment is less important under
conditions of low uncertainty might also explain why higher levels of empowerment are not
associated with process improvement in virtual team with frequent face-to-face meetings
(Kirkman, Rosen et al. 2004).
Moreover, our results demonstrate that TNOV and TTURB are two distinct sources of
uncertainty and therefore have different effects. First, TTURB not only directly influences
SPEED (P<.05) but also moderates the relationship between TEMPOWER and SPEED
(P<.05); In contrast, TTNOV has neither direct nor strong moderating effects. Second, the
TEMPOWER–SPEED relationship varies from a u-shaped to a linear relationship when
TNOV increases from low to high; in contrast, the TEMPOWER–SPEED relationship varies
from a u-shaped to a reverse U-shaped when TTURB increases from low to high. Thus, the
results suggest that TTURB does differ from TNOV and might have a stronger influence than
TNOV. As a post-hoc test, we entered TTURB, the cross product between TTURB and
TEMPOWER or quadratic term of TEMPOWER after entering the terms related to TNOV in
previous steps (table II). The R2 change in this step is significant lending support to this
notion.
The finding that TTURB has a positive relationship with SPEED makes sense, since fastchanging technology exerts pressure to developing products fast. In contrast, TNOV has no
direct relationship with SPEED. A possible explanation is that TTURB is related to time while
TNOV is related to innovativeness by their definitions, and SPEED as an indicator of NPD
16
performance is time-oriented rather than innovation-oriented. Thus, it is reasonable that
TTURB will have a stronger effect on time-based strategy than uncertainty from TNOV.
Under conditions of Low TTURB or TNOV, projects are routine and easy to handle. Thus,
either full top management control or full empowerment works and moderate empowerment
may result in ambiguity and less effectiveness. Under conditions of high TNOV, technology
employed is new to the project team but is known or proved in the industry, thus, uncertainty
is moderate and the technology goal or vision is clear. The team may possess critical skills or
manage critical information that the executive cannot effectively control. Therefore, the more
empowerment, the better. Under high TTURB, technology in the industry changes rapidly and
may be new to everyone, thus, uncertainty is higher. Although some degree of empowerment
contributes to teams making decisions quickly and effectively, the requirement of dealing with
fast-changing environment may lie beyond the capability of the team, therefore, top
management involvement is required to clarify and maintain project vision and direction.
Empowered NPD Teams with some control in this situation might be more effective than
NPD teams with full empowerment. This again indicates that a central issue on empowerment
is how to balance between empowerment and control (1995).
Generally, our study confirms the notion that empowerment is important to maintain
competitiveness in fast-changing environment. However, it is also very important to analyze
the source and degree of uncertainty before implementing one-size-fits all empowerment
(Forrester 2000). When technology uncertainty is low, either from newness or pace of change,
empowerment will have a bipolar effect: top management control or full empowerment is
better than partial empowerment. NPD teams should pay more attention to find an appropriate
level of empowerment in stimulating SPEED. More importantly, under high levels of
technology uncertainty, managers need to take a contingent strategy. Under conditions of high
17
technological novelty managers can expect greater empowerment to lead to faster speed. But
when technological turbulence is high extremely high levels of empowerment might even end
up with slower NPD. Another point for practitioners is to be carefully balance empowerment
with some control, especially under conditions of extremely high TTURB.
LIMITATIONS AND CONCLUSION
One limitation of this study is the single source methodology. However, we have discussed
extensively the single source problem in a previous paper which applied several different
methods to examine potential bias (Chen, Reilly and Lynn 2005) and concluded that single
source bias exists but is not serious. The second limitation is the non-random sampling that
includes a high proportion of technology-based firms in the US Northeast. The third limitation
is that we only survey launched projects thereby restricting range of the dependent variable.
Finally, most projects in our study were at large companies. Although firm size did not
significantly influence the relationships studied, the results should be cautiously applied to
small firms.
This study contributes to our understanding of NPD in at least two ways. First, we confirmed
that change (turbulence) and newness should be regarded as distinctly different dimensions of
uncertainty. The findings in the study also indicate that turbulence is a more important source
of uncertainty than newness. Second, we also confirmed that uncertainty influences the
strength of the EPOWER-SPEED relationship as Wall et al. expected (Wall, Cordery et al.
2002) but also influences the form (linear vs. curvilinear) of the relationship. A contingent
view should be further investigated to better understand how uncertainty influences the
effectiveness of NPD practices. How to delicately balance empowerment and control also
needs further exploration.
18
References
Anderson, P. and M. L. Tushman (2001). "Organizational environments and industry exit: The effects of
uncertainty, munificence and complexity." Industrial and Corporate Change 10(3): 675-701.
Aram, J. D. and K. Walochik (1996). "Improvisation and the Spanish manager." International studies of
management and organization 26(4): 73-89.
Chen, J., R. R. Reilly and G. S. Lynn (2005). "The impacts of speed-to-market on new product success: The
moderating effects of uncertainty." IEEE Transactions on Engineering Management 52(2).
Clampitt, P. G. and R. J. Dekoch (2001). Embracing uncertainty: The essence of leadership. Armonk, New York,
M. E. Sharpe, Inc.
Day, G. S. (1994). "The Capabilities of Market-Driven Organizations." Journal of Marketing 58: 37-52.
Eisenhardt, K. M. and B. N. Tabrizi (1995). "Accelerating adaptive processes: product innovation in the global
computer industry." Administrative Science Quarterly 40: 84-110.
Eylon, D. and P. Bamberger (2000). "Empowerment cognitions and empowerment acts: Recognizing the
importance of gender." Group & Organization Management 25(4): 354-372.
Ford, R. C. and M. D. Fottler (1995). "Empowerment: A matter of degree." Academy of Management Executive
9(3): 21-31.
Forrester, R. (2000). "Empowerment: Rejuvenating a potent idea." Academy of Management Executive 14: 6780.
Gerwin, D. (1999). "Team Empowerment in NPD." Business Horizons: 29-36.
Huber, G. P. and D. J. Power (1985). "Research Notes and Communications Retrospective Reports of Strategiclevel Managers: Guidelines for Increasing their Accuracy." Strategic Management Journal: 171-180.
Hyatt, D. E. and T. M. Ruddy (1997). "An examination of the relationship between work group characteristics
and performance: Once more into the breech." Personnel Psychology 50(3): 553-585.
Jaworski, B. J. and A. K. Kohli (1993). "Market orientation: Antecedents and consequences." Journal of
Marketing 57(3): 53-70.
Kamoche, K. and M. P. E. Cunha (2001). "Minimal structure: from jazz improvisation to product innovation."
Organization Studies 22(5): 733-763.
Kessler, E. H. and P. E. Bierly, III (2002). "Is faster really better? An empirical test of the implication of
innovation speed." IEEE Transactions on Engineering Management 49(1): 2-12.
Kirkman, B. L., B. Rosen, P. E. Tesluk and C. B. GibsonI (2004). "The impact of team empowerment on virtual
team performance: The moderating role of fact-to-face interaction."
Kirkman, B. L. and B. Rosen. (1997). A model of work team empowerment. Research in organizational change
and development. R. W. Woodman and W. A. Pasmore. Greenwich, CT, JAI Press. 10: 131-167.
Kirkman, B. L. and B. Rosen. (1999). "Beyond self-management: Antecedents and consequences of team
empowerment." Academy of Management Journal 42(1): 58-74.
Lubinski, D. and L. G. Humphreys (1990). "Assessing spurious "moderator effects": Illustrated substantively
with the hypothesized ("synergistic" relationship between spatial and mathematical ability."
Psychological Bulletin 107(3): 385-393.
Lukas, B. A. and O. C. Ferrell (2000). "The effect of market orientation on product innovation." Journal of the
Academy of Marketing Science 28(2): 239-247.
Lynn, G. S. and A. E. Akgun (1998). "Innovation strategies under uncertainty: A contingency approach for new
product development." Engineering Management Journal 10(3): 11-18.
McDonough, E. F. I. (2000). "Investigating of factors contributing to the success of cross-functional teams."
Journal of Product Innovation Management 17: 211-235.
McDonough, E. F. I. and G. Barczak (1991). "Speeding up new product development: The effects of leadership
style and source of technology." Journal of Product Innovation Management 8(3): 203-211.
Milliken, F. J. (1987). "Three types of perceived uncertainty about the environment: State, effect, and response
uncertainty." Academy of Management Review 12(1): 133-143.
Mohr, L. B. (1982). Explaining organizational behavior. London, Jossey-Bass Publishers.
Nunez, E. (2004). Integrating new product development for turbulence and uncertainty environments. Howe
School of Technology Management. Hoboken, Stevens Institute of Technology.
Nunnally, J. C. (1978). Psychometric Theory. New York, McGraw-Hill.
Olson, E. M., O. C. Warker and R. W. Ruekert (1995). "Organizing for effective new product development: The
moderating role of product innovativeness." Journal of Marketing 59(1): 48.
Reilly, R. R., A. Bentley and G. S. Lynn (2003). Empowerment in new product teams: more is not always better.
Academy of Management Annual Conference 2003, Seattle, Washington.
Schumpeter, J. A. (1934). The Theory of Economic Development: An Inquiry into Profits, Capital, Credit,
19
Interest, and the Business Cycle. New York, Oxford University Press.
Souder, W. E. and X. M. Song (1998). "Analyses of U.S. and Japanese management processes associated with
new product success and failure in high and low familiarity markets." Journal of Product Innovation
Management 15: 208-223.
Spreitzer, G. M. (1995). "Psychological empowerment in the workplace: Dimensions, measurement, and
validation." Academy of Management Journal 38(5): 1442-1465.
Spreitzer, G. M. (1996). "Social structural characteristics of psychological empowerment." Academy of
Management Journal 39(2): 483-504.
Spreitzer, G. M., M. A. Kizilos and S. W. Nason (1997). "A dimensional analysis of the relationship between
psychological empowerment and effectiveness, satisfaction, and strain." Journal of Management 23(5):
679-704.
Tatikonda, M. V. and M. M. Montoya-Weiss (2001). "Integrating operations and marketing perspectives of
product innovation: The influence of organizational process factors and capabilities on development
performance." Management Science 47(1): 151-172.
Thomas, K. W. and B. A. Velthouse (1990). "Cognitive Elements of Empowerment: An 'Interpretive' Model of
intrinsic task motivation." Academy of Management Review 15(4): 666-681.
Thompson, J. D. (1967). Organization in action. New York, NY, McGraw-Hill.
Wall, T. D., J. L. Cordery and C. W. Clegg (2002). "Empowerment, performance, and operational uncertainty: A
theoretical integration." Applied Psychology: An International Review 51(1): 146-169.
Wellins, R. S., W. C. Byham and J. M. Wilson (1991). Empowered teams: Creating self-directed work groups
that improve quality, productivity, and participation. San Francisco, Jossey-Bass.
Wilkinson, A. (1998). "Empowerment: Theory and practice." Personnel Review 27(1): 40-.
Yap, C. M. and W. E. Souder (1994). "Factors influencing new product success and failure in small
entrepreneurial high-technology electronics firms." Journal of Product Innovation Management 11: 418432.
20
Appendix
Table I Descriptive statistics and Correlation Analysis Results
TEMPOWER SPEED TNOV
TTURB Team size Firm size
TEMPOWER (.86)
.441** -.050
.088
-.098
.054
SPEED
(.80)
.018
.165*
-.058
.059
(.68)
.490** .200**
.213**
(.85)
.247**
.217**
(.97)
.267**
TNOV
TTURB
Team size
Mean
6.84
5.53
4.52
5.31
24
S.D.
2.27
2.87
2.85
3.02
40
N#
212
209
211
211
203
205
** Correlation is significant at the 0.01 level; * at the 0.05 level (2-tailed).
Alpha coefficients are shown in parenthesis on diagonal.
21
Table II Results of Hierarchical Moderated Regression Analysis
TNOV
TTURB
Step1
Team size
-.010
-.043
Firm size
-.029
-.045
TEMPOWER
..443***
.425***
Uncertainty
.046
.148*
F
12.0***
13.4***
df
(4,196)
(4,196)
R2
.197
.214
Adjusted R2
.180
.198
Step 2
SQUEMPOWER
.079
.071
F
1.13
.092
R2
.201
.218
Adjusted R2
.181
.198
∆R2
.005
.005
STEP 3
TEMPOWER* Uncertainty
.108#
.138*
F
2.83#
4.4*
R2
.213
.235
Adjusted R2
.189
.212
∆R2
.011
.017
STEP 4
SQUEMPOWER*Uncertainty -.279**
-.246**
F
9.43**
6.86**
R2
.250
.262
Adjusted R2
.222
.235
2
∆R
.037
.026
STEP 5
TTURB
.249*
TEMPOWER*TTURB
.039
SQUEMPOWER*TTUTB
-.166#
F
2.87*
R2
.282
Adjusted R2
.244
2
∆R
.033
Standardized regression coefficients are showed.
# P<.1, *P<.05, ** P<.01, ***P<.001
22
Team empowerment
Team performance
Uncertainty
Technological Turbulence
Technological Novelty
Fig. I The conceptual Model
23
Mean TNOV
Low TTURB
High TNOV
9
8
8
7
7
6
6
5
Speed-to-market
Speed-to-market
Low TNOV
5
4
3
2
High TTURB
4
3
2
1
0
1
-1
0
-1
Mean TTURB
1
2
3
4
5
6
7
8
Team empowement
9
10 11
1
2
3
4
5
6
7
8
9
10 11
-2
Team empowerment
Fig. II The team empowerment and speed relationship under different levels of TNOV VS TTURB
24