ArticlePDF Available

Abstract and Figures

On the basis of an extensive interdisciplinary literature review proactive decision-making (PDM) is conceptualised as a multidimensional concept. We conduct five studies with over 4,000 participants from various countries for developing and validating a theoretically consistent and psychometrically sound scale of PDM. The PDM concept is developed and appropriate items are derived from literature. Six dimensions are conceptualised: the four roactive cognitive skills ‘systematic identification of objectives’, ‘systematic search for information’, ‘systematic identification of alternatives’, and ‘using a ‘decision radar’’, and the two proactive personality traits ‘showing nitiative’ and ‘striving for improvement’. Using principal component factor analysis and subsequent item analysis as well as confirmatory factor analysis, six conceptually distinct dimensional factors are identified and tested cceptably reliable and valid. Our results are remarkably similar for individuals who are decision-makers, decision analysts, both or none of both with different levels of experience. There is strong evidence that individuals with high scores in a PDM factor, e.g. proactive cognitive skills or personality traits, show a significantly higher decision satisfaction. Thus, the PDM scale can be used in future research to analyse other concepts. Furthermore, the scale can be applied, e.g. by staff teams to work on OR problems effectively or to inform a decision analyst about the decision behaviour in an organisation.
Content may be subject to copyright.
European Journal of Operational Research 249 (2016) 864–877
Contents lists available at ScienceDirect
European Journal of Operational Research
journal homepage: www.elsevier.com/locate/ejor
Developing and validating the multidimensional proactive
decision-making scale
Johannes Sieberta,, Reinhard Kunzb
aOperations Management, Faculty of Law, Business and Economics, University of Bayreuth, Universitätsstr. 30, D-95440 Bayreuth, Germany
bMedia Management, Faculty of Law, Business and Economics, University of Bayreuth, Universitätsstr. 30, D-95440 Bayreuth, Germany
article info
Article history:
Received 28 July 2014
Accepted 28 June 2015
Availableonline16July2015
Keywo rds:
Behavioural OR
Decision analysis
Problem structuring
abstract
On the basis of an extensive interdisciplinary literature review proactive decision-making (PDM) is concep-
tualised as a multidimensional concept. We conduct five studies with over 4000 participants from various
countries for developing and validating a theoretically consistent and psychometrically sound scale of PDM.
The PDM concept is developed and appropriate items are derived from literature. Six dimensions are con-
ceptualised: the four proactive cognitive skills ‘systematic identification of objectives’, ‘systematic search for
information’, ‘systematic identification of alternatives’, and‘using a decision radar’, and the two proactive per-
sonality traits ‘showing initiative’ and ‘striving for improvement’. Using principal component factor analysis
and subsequent item analysis as well as confirmatory factor analysis, six conceptually distinct dimensional
factors are identified and tested acceptably reliable and valid. Our results are remarkably similar for indi-
viduals who are decision-makers, decision analysts, both or none of both with different levels of experience.
There is strong evidence that individuals with high scores in a PDM factor, e.g. proactive cognitive skills or
personality traits, show a significantly higher decision satisfaction. Thus, the PDM scale can be used in future
research to analyse other concepts. Furthermore, the scale can be applied, e.g. by staff teams to work on OR
problems effectively or to inform a decision analyst about the decision behaviour in an organisation.
© 2015 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
1. Introduction
“If I were given one hour to save the planet, I would spend
55 minutes defining the problem and five minutes resolving it.”
Albert Einstein
In the last decades, the methods in Operational Research (OR)
made substantial progress. Researchers developed methods, which
can be used to “solve problems” about which earlier generations had
dreamt. These OR methods have a great positive impact on the qual-
ity of individual and organisational decisions. In line with the fa-
mous quote from Albert Einstein it is important to spend effort in
defining a problem. The more appropriate the problem is defined and
structured, the greater the potential for positive impact of OR meth-
ods. This paper contributes to skilful problem structuring by pro-
viding a new concept concerning proactive decision-making and an
Corresponding author. Tel.: +49 921 556194.
E-mail addresses: Johannes.Siebert@uni-bayreuth.de (J. Siebert), reinhard.kunz@
uni-bayreuth.de (R. Kunz).
empirically validated scale that measures proactive cognitive skills
and personality traits to support making better decisions.
Woolley and Pidd (1981, p. 197) described problem structuring
as “the process by which the initially presented set of conditions is
translated into a set of problems, issues and questions sufficiently
well defined to allow specific research action.” In theory and prac-
tice, problem structuring methods have gained more and more atten-
tion (Franco & Montibeller, 2010; Mingers & Rosenhead, 2001; Tavella
& Papadopoulos, 2014). Problem structuring methods (PSM) are de-
scribed by Rosenhead (2013, p. 1162) as a “broad group of model-
based problem handling approaches whose purpose is to assist in
the structuring of problems rather than directly to derive a solution.”
These methods are most frequently applied by groups and are char-
acterised by participation and interactivity (Rosenhead, 2013).
Franco and Meadows (2007) indicated that McGrath’s (1984) cir-
cumplex is the most accepted framework for group decision sup-
port, theory, and research. McGrath identifies four basic actions that
need to be performed in a decision related meeting: generating,
choosing, negotiating, and executing decisions. The main tasks of
a group contain generating alternatives (i.e. ideas, plans, strategies,
etc.) and negotiating conflicting preferences. “Within the context of a
http://dx.doi.org/10.1016/j.ejor.2015.06.066
0377-2217/© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 865
PSM-supported process, groups will engage in information gather-
ing and the designing of strategic options (generate). They will also
structure and evaluate the relative advantages and disadvantages of
different strategic options before selecting a problem focus and/or a
course of action (negotiate)” (Franco & Meadows, 2007, p. 1624). Both
generating as well as negotiating are crucial for success. However, we
argue that a reasonable result in a generation phase is prerequisite
for an effective negotiation phase. Therefore, we are convinced that a
focus on the generating tasks is crucial to success.
Many studies recommend that cognitive styles and decision-
making styles do have an impact on individual decision-making (e.g.
Epstein, Pacini, Denes-Raj, & Heier, 1996; Novak & Hoffman, 2009;
Scott & Bruce, 1995), and on group decision-making processes (e.g.
Hough & Ogilvie, 2005; Schwenk, 1995). This applies also for the
design, use and acceptance of group decision support systems (e.g.
Benbasat & Dexter, 1982; Lu, Yu, & Lu, 2001; Taylor, 2004).
As a support for group decision-making Franco and Meadows
(2007) emphasise the importance of cognitive style in PSM research
and application. They pioneered in systematically analysing the im-
pact of Jung’s (1971) theory of psychological types in context of PSM
and derive logically from literature eight hypotheses, e.g. that “sens-
ing and intuitive individuals will play a lead role during option de-
signing tasks, in comparison to thinking and feeling individuals” (p.
1626). Garfield, Taylor, Dennis, and Satzinger (2001) identified em-
pirically that innovative, radical alternatives are created by intuitive
and feeling individuals more often than by sensing and thinking
individuals.
To the best of our knowledge, none of the existing psychological
tests and scales are suited for explaining the process of “generating”
comprehensively. In particular, it is of interest which skills individu-
als have in the generating phase and how much and why they take
initiative. Research on decision-making lacks a psychometrically re-
liable scale for measuring proactive decision-making. In this paper,
we develop a scale that distinguishes four cognitive skills and two
personality traits relevant to the generation phase in PSM. Our scale
measures proactive cognitive skills derived from value-focused think-
ing and proactive traits derived from proactive behaviour.
Bateman and Crant (1993) define proactive behaviour as the rela-
tively stable tendency to effect environmental change. The essential
characteristic of proactive behaviour is that “people can intentionally
and directly change their current circumstances, social or nonsocial,
including their physical environment” (Bateman & Crant, 1993,p.104,
referring to Buss, 1987). The prototypical proactive personality is rel-
atively unaffected by situational forces and interacts with its envi-
ronment actively. Individuals classified as reactive, by contrast, are
relatively passive and are rather shaped by their environment than
shaping it themselves (Parker, Bindl, & Strauss, 2010). Proactive in-
dividuals actively search for opportunities, take initiative, and pro-
ceed with their actions until they achieve their objectives (Bateman
& Crant, 1993). Schwarzer (1999) develops a scale to measure the
personality trait proactive attitude, which can affect motivations and
imply actions. Proactive individuals have a vision and are driven by
their values. They follow goals that they think are worth reaching for
(Parker et al., 2010; Schwarzer, 1999). Bateman and Crant’s proactive
behaviour and Schwarzer’s proactive attitude have in common that in-
dividuals show initiative and strive for improvements in their lives.
Individuals cannot change their personality traits related to decision-
making easily (VandenBos, 2007). However, Kirby, Kirby, and Lewis
(2002, p. 1542) find empirical evidence that proactivity can be trained
by the “development of context specific knowledge and skills”.
Making decisions, personal or work-related, is an essential part
of everyone’s life. However, not everyone and every organisation
make good decisions. As it has been postulated and verified em-
pirically, proactive personality traits (e.g. Seibert,Crant,&Kraimer,
1999; Thompson, 2005) as well as proactive cognitive skills (Keeney,
1992) can have positive impacts on an individual in decision situa-
tions. It can be presumed that PDM enables people to make better
decisions with results they are more satisfied with. Thus, we consider
PDM to be a relevant concept that is worth being looked at in more
detail.
Research in OR focuses on best practices or on developing and im-
proving highly sophisticated methods (e.g. Corbett, Overmeer, & Van
Wassenhove, 1995). Hämäläinen, Luoma, and Saarinen (2013 , p. 623)
indicate the importance of behavioural operational research (BOR) as
“the study of behavioural aspects related to the use of […] OR meth-
ods in modelling, problem solving and decision support”. Lu et al.
(2001) state that in OR applications, the personality as well as the
communication style of the decision analyst and the decision-maker
may have a huge impact. Appropriate tools and methods for elicit-
ing information about the decision analyst and the decision-maker
are still needed, in particular in problem structuring, since hardly any
behavioural research has been done “on the process itself and on the
role of the analyst and problem owner” (Hämäläinen et al., 2013,p.
623). These tools and methods have to be selected “on the basis of
the skills, knowledge, personal style and experience of the analyst”
(Hämäläinen et al., 2013, p. 624, referring to Ormerod, 2008). We de-
velop a scale to measure an individual’s behaviour in decision situa-
tions. Measurements on this scale can be used to select effective pro-
cedures using OR techniques in consideration of behavioural aspects
and to analyse behavioural facets in problem structuring.
In this interdisciplinary paper we explicate PDM as a multidimen-
sional concept that combines aspects of proactive personality traits
and proactive cognitive skills in decision situations. We develop a new
PDM scale and test it empirically in order to identify reliable and valid
measures. In particular, we pursue four objectives: firstly, PDM is con-
ceptualised, i.e. the concept is defined, clarified by its dimensions,
and differentiated from other constructs; secondly, the dimensions
of PDM are operationalised; thirdly, the multidimensional PDM scale
is empirically tested and validated in several studies; fourthly, deci-
sion satisfaction is explained by PDM.
The paper is organised as follows. In Section 2,wesum-
marise the theoretical foundation of proactive behaviour, decision-
making, value-focused thinking, and basic psychological concepts. In
Section 3, we conceptualise PDM and derive suitable dimensions
from literature. In Sections 4 and 5, we describe the operationalisa-
tion of constructs and the methodology. In Sections 6 and 7,wesum-
marise and discuss the results of our empirical studies. In Section 8,
we discuss implications for OR, limitations, and further research. In
Section 9, we draw our conclusions.
2. Theoretical foundation of proactive decision-making
PDM is based on different disciplines such as psychology, decision
theory, and behavioural OR. The term ‘proactive’ refers to personality
traits and cognitive skills. Therefore, PDM is framed by insights into
proactive personality traits and decision theories in general, value-
focused thinking as well as thinking and decision-making styles in
particular.
2.1. Proactive personality traits
Grant and Ashford (2008) point out that proactive behaviour in-
volves acting in advance of future situations. Individuals consider fu-
ture events in their current decisions with foresight, i.e. before they
occur. Researchers describe this characteristic using the adjectives
‘future-focused’, ‘anticipatory’, and ‘forward-looking’ (Frese, 2006;
Frese & Fay, 2001; Frese, Kring, Soose, & Zempel, 1996; Greenglass,
2002). Proactive behaviour is characterised by the intention of hav-
ing a “discernible effect on the self and/or the environment” (Grant &
Ashford, 2008, p. 9). Proactive individuals are change-oriented and
interested in creating a meaningful impact on their environment
(Buss, 1987; Diener, Larsen, & Emmons, 1984). Reactive individuals,
866 J. Siebert, R. Kunz / European Journal of Operational Research 249 (2016) 86 4–877
by contrast, are passive and react to, adapt to, and are constrained
by their environment instead of shaping it themselves (Bateman &
Crant, 1993). Proactive individuals have a vision and are guided by
their values. On the basis of a clear perception of what they want to
achieve they derive goals that comply with their vision. Schwarzer
(1999), who developed a scale to measure proactive attitudes, points
out that individuals’ purpose in life is defined by striving for ambi-
tious goals. Parker et al. (2010) emphasise the generation of goals as
well as the endeavours to achieve them. Proactive individuals take the
initiative in pursuing personal and organisational goals (Frese & Fay,
2001; Roberson, 1990). Proactive individuals “scan for opportunities,
show initiative, take action, and persevere until they reach closure
by bringing about change” (Bateman & Crant, 1993, p. 105). Proactive
behaviour is considered to be relatively stable and cannot be changed
easily (Bateman & Crant, 1993). The same is true for proactive atti-
tudes. Habitual behaviour and attitudes are actualised in many differ-
ent situations and therefore, are called personality traits or personal
characteristics. The VandenBos (2007, p. 950) defines a personality
trait as “a relatively stable, consistent, and enduring internal charac-
teristic that is inferred from a pattern of behaviours, attitudes, feel-
ings, and habits in the individual”. For our research we use the more
general term ‘personality traits’ to cover all of its subsumed aspects.
However, Kirby et al. (2002, p. 1542) find empirical evidence that
proactivity can be trained by the “development of context-specific
knowledge and skills”. Therefore, we consider personality traits as
well as cognitive skills to be relevant dimensions of PDM.
Referring to proactive behaviour, Seibert et al. (1999) analyse
the association with endogenous constructs. They find out that the
indicators of career success, i.e. the self-reported objective indica-
tor (salary and promotion) as well as a subjective indicator (ca-
reer satisfaction), correlate positively with proactive behaviour. Crant
and Bateman (2000) discover that employees who show proac-
tive behaviour are recognised as charismatic leaders. Furthermore,
Thompson (2005) links proactive personality to job performance con-
ceived by superiors. Thus, there is empirical evidence that proac-
tive behaviour may have positive impacts on individuals in different
contexts.
2.2. Decision-making and value-focused thinking
Decision-making is the process of making a choice between com-
peting courses of action (von Winterfeldt & Edwards, 1986). It is “a
dynamic process: a complex search for information, full of detours,
enriched by feedback from casting about in all directions, gather-
ing and discarding information, fuelled by fluctuating uncertainty, in-
distinct and conflicting concepts – some sharp, some hazy” (Zeleny,
198 2, p. 3). Furthermore, decision-making is regarded as a cogni-
tive process of choosing an alternative. Individuals and organisations
have only through decisions an impact on their situation (Keeney,
1992). On the basis of normative models of the decision theory, pro-
cesses can be derived as to how individuals should proceed in or-
der to achieve a maximum of their values, objectives, and well-being
(Bell, Raiffa, & Tversky, 1988; Howard, 1988). Furby and Beyth-Marom
(1992) summarise five steps of decision-making. First, possible alter-
natives have to be identified. Second, possible consequences of the
alternatives have to be identified. Third, the desirability of these con-
sequences has to be evaluated in terms of achieving one’s objectives.
Fourth, the likelihood of these consequences has to be assessed. Fifth,
the best alternative has to be identified among the generated set of
alternatives by combining preferences and uncertainty using certain
decision rules.
Howard (1988) emphasises the need to distinguish between deci-
sion and outcome. “A good outcome is a future state of the world that
we prize relative to other possibilities. A good decision is an action we
take that is logically consistent with the alternatives we perceive, the
information we have, and the preferences we feel” (Howard, 1988,
p. 682). However, because one never perceives all possible alterna-
tives and because uncertainty is associated with our knowledge de-
rived of generally incomplete information, a bad decision can result in
a good outcome and vice versa.von Winterfeldt and Edwards (1986)
conclude that a decision cannot be judged by its outcome. Usually, the
outcome is determined by several factors that cannot be controlled by
the decision-maker. Thus, the quality of a decision can only be evalu-
ated by the process in the course of which it was made and it has to be
evaluated by looking at the stages of the decision process before the
outcome occurs (von Winterfeldt & Edwards, 1986). However, across
the life span, decision-making skills are related to obtaining good de-
cision outcomes (Bruine de Bruin, Parker, & Fischhoff, 2012).
Furthermore, Howard (1988) introduces seven elements of deci-
sion quality. ‘Decision framing’ deals with the question whether the
real problem is addressed and whether it is framed appropriately. ‘In-
formational excellence’ refers to the cost-effectiveness of information
sources and the gathering of meaningful and reliable information.
‘Creativity – significantly different alternatives’ implies appropriate-
ness of the effort put into the search for and identification of creative
and doable alternatives. ‘Clear values’ refers to the process of gaining
clarity about values and trade-offs. ‘Integration and evaluation with
logic’ requires the use of logically correct reasoning. ‘Balance of basis’
makes the optimal balance of efforts a subject of discussion, e.g. the
efforts put into clarifying values, creating and evaluating alternatives.
‘Commitment to action’ deals with the clarity and straightforward-
ness that is necessary to communicate and execute a decision.
Keeney (1992) introduces a normative approach to PDM. The
paradigm of this way of thinking is that values provide the basis of
interest in a decision problem. In this context, values refer to what
someone hopes to achieve by decision-making, given a certain (set of)
alternative(s). Therefore, values should guide the effort made to solve
the problem. Values are explicated through an individual’s goals and
objectives, which serve as a starting point for the decision-making
process. Instead of following the traditional process of identifying a
problem, generating alternatives, and defining criteria for the evalua-
tion of outcomes before making a choice, Keeney (1992) suggests that
decision-makers seek out decision opportunities proactively. Such
decision opportunities could be created either by broadening an ex-
isting decision context or by recognising individual objectives that
might be improved.
Value-focused thinking provides numerous guidelines using val-
ues to support the search for more and better alternatives. The prin-
ciple is to create alternatives taking account of at least one of the val-
ues specified for the decision situation. In other words, people think
first of what they desire (themselves) and then of possible alterna-
tives that are useful for reaching the desired objective (Keeney, 1992).
Siebert and Keeney (forthcoming) identify that decision-makers are
able to list only one third of their potentially relevant alternatives, but
significantly more and better when being stimulated with objectives.
2.3. Thinking and decision-making styles
Other relevant approaches to PDM are thinking and decision-
making styles. Scott and Bruce (1995, p. 820, based on Driver, 1979;
Driver, Brousseau, & Hunsaker, 1990; Harren, 1979) speak of “a habit
based propensity to react in a certain way in a specific decision con-
text.” Thus, styles are neither of habitual behaviour nor of attitude.
The decision-making style here is, on the one hand, influenced by the
specific situation and, on the other hand, by a more general term in
form of a habitual reaction. Apart from this understanding, we find
many other constructs; however, they are not fully compatible with
each other. Some deal with observable behaviour, while others refer
to internal processes. Some are derived from personality traits (e.g.
Epstein, 1973), while others are derived from situational aspects (e.g.
Novak & Hoffman, 2009).
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 867
Epstein (1973, 1983, 1985, 1994, 2003) introduce the concept
of thinking styles, providing also implications for decision-making.
They find out that people use in general two different thinking
styles: experiential and rational thinking. Experiential thinking is as-
sociative, emotional, of little effort, rapidly implemented but slowly
changed, parallel, immediate, outcome-oriented, holistic, precon-
scious, and passively experienced with the process being opaque to
the individual. Rational thinking, by contrast, is logical, based on
the cause-and-effect rule, hierarchical, sequential, process-oriented,
slowly implemented but quickly changed, of much effort, oriented
towards delayed action, conscious, and actively experienced with
the individual being aware of and in control of the process (Epstein,
1994, 2003; Kahnemann & Frederick, 2002; Sloman, 1996; Smith &
DeCoster, 2000). The Rational–Experiential Inventory (REI) devel-
oped by Epstein et al. (1996) has been used to measure individual
differences in dispositional tendencies to adopt rational and experi-
ential thinking styles.
In contrast to these dispositional thinking styles, Novak and
Hoffman (2009) define a situation specific thinking style as the partic-
ular thinking style or momentary thinking orientation adopted by a
consumer in a specific (decision) situation. As Scott and Bruce (1995)
do, they emphasise the situational factors. In recent studies, Novak
and Hoffman (2009) have considered situations in order to incorpo-
rate the different tasks or activities that consumers may undertake
as well as different motivations or orientations that consumers may
reveal when they undertake a specific task or activity. The situation
specific thinking style may be influenced by the task itself or by the
consumer’s underlying motive for performing a given task, indepen-
dently of the task itself (Novak & Hoffman, 2009).
An individual with a rational thinking style will probably use cog-
nitive skills for gaining and processing information. This can be done
without being noticed by others; it is also possible that this cogni-
tive approach precipitates observable behaviour. The use of cogni-
tive skills characterises how an individual approaches decision sit-
uations. These cognitive skills can be acquired through training and
practice (VandenBos, 2007). In this paper, we concretise aspects of ra-
tional thinking in decision situations, in particular problem structur-
ing, using four cognitive skills derived from the proactive paradigm of
decision-making in general and value-focused thinking in particular.
2.4. Theoretical considerations
Proactivity can be manifested in decision-making by proactive
personality traits and a proactive way of thinking. While proactive
personality traits are regarded as a stable disposition in decision sit-
uations, the proactive way of thinking can be changed more easily by
training the related cognitive skills (VandenBos, 2007). The distinc-
tion between personality traits and cognitive skills can help explain
the results of Kirby et al. (2002)), who have found empirical evidence
that proactivity is trainable since cognitive skills can be trained.
It has been verified empirically that proactive personality traits
can have positive impacts on an individual in decision situations
(Seibert et al., 1999). Although Keeney (1992) postulates positive im-
pacts of a proactive way of thinking, these have not been verified
empirically yet. Furthermore, we assume that proactive personality
traits and proactive cognitive skills complement each other, for exam-
ple, reactive individuals who neither take initiative nor strive for im-
provement will not be motivated to apply their cognitive skills effec-
tively. Similarly, proactive individuals who take initiative and strive
for improvement but have low proactive cognitive skills will not be
effective in decision-making either. Proactive decision-makers who
are characterised by proactive personality traits and high proactive
cognitive skills should be most effective in their decision-making. To
analyse such hypotheses it is necessary to conceptualise PDM and to
create a scale that covers proactive personality traits as well as proac-
tive cognitive skills.
3. Conceptualisation of proactive decision-making
At first, the meaning of PDM is specified. Our working definition
evolved over time on the basis of an extensive and thorough litera-
ture review and several group discussions with experts from different
academic disciplines, and it is to be seen as a consensus. We generally
speak of decision-making as being proactive if it is value-/objective-
oriented and self-initiated by a foresighted individual who strives for
improvement. Alternatives are created systematically. Information on
opportunities and threats are gathered systematically, and potential
outcomes of an individual’s actions are anticipated. In our context,
PDM can be characterised by (proactive)personality traits and the use
of (proactive)cognitive skills in decision situations (see Fig. 1).
In this section, we derive the four dimensions ‘systematic iden-
tification of objectives’, ‘systematic search for information’, ‘systematic
identification of alternatives’, and “using a ‘decision radar’” – which
represent proactive cognitive skills in decision situations – from
decision-making, in particular from the first four of the elements
of decision quality (Howard, 1988), and the value-focused thinking
framework (Keeney, 1992). Other aspects of cognitive skills, such as
the evaluation of alternatives, the final decision, or the implementa-
tion of a decision, are not postulated as dimensions since there are
no fundamental differences between reactive and proactive decision-
making. Individuals can acquire proactive cognitive skills intuitively
or by learning methods taught in courses, books, and papers on
decision-making. However, even if individuals are experts at using
these skills, they still have to apply them. The dimensions taking ini-
tiative’and‘striving for improvement’ are related to proactive person-
ality traits. Individuals who do not strive for improvement will have
no reason to apply their skills. In addition, even if there is a reason to
apply these skills, this person needs to take initiative.
3.1. Dimensions of proactive decision-making
According to Schwarzer (1999) and Keeney (1992), proactive indi-
viduals have a vision and are guided by values. On the basis of a clear
perception of what they want to achieve they derive goals that are in
line with their vision. Their purpose in life is generated by striving for
ambitious goals (Schwarzer, 1999). The anticipation and imagination
of objectives encourage individuals to pursue these objectives (Locke
& Latham, 2002). Objectives are crucial for PDM. They are the basis for
systematically creating alternatives, guiding the methodical search
for information, and planning decisions (Siebert & Keeney, forthcom-
ing). If individuals are not aware of their objectives in a specific de-
cision situation, they generally cannot be proactive in their decision-
making. Therefore, the proactive cognitive skill of ‘systematic identifi-
cation of objectives’(Objectives) is assumed to be a dimension of PDM.
Reactive and passive individuals do not strive for more ambitious
goals or values, but they simply accept the alternatives given in a spe-
cific decision situation, even if they are not content with them. Proac-
tive individuals, by contrast, try to create more and better alternatives
(Keeney, 1992). As previous research has shown, using objectives for
the systematic identification of alternatives results in more and bet-
ter alternatives (Butler & Scherer, 1997; Gettys, Pliske, Manning, &
Casey, 1987; Jungermann, Ulardt, & Hausmann, 1983; Pitz, Sachs, &
Heerboth, 1980; Siebert & Keeney, forthcoming). Furthermore, using
objectives for the identification of alternatives increases the likeli-
hood that individuals will actually achieve their objectives (Grant &
Ashford, 2008, referring to Gollwitzer, 1999; Gollwitzer & Brandstät-
ter, 1997; Gollwitzer & Oettingen, 1998). Training in the use of ob-
jectives for the creation of alternatives enhances proactive cognitive
skills in creating alternatives. If individuals are not aware of their al-
ternatives in a specific decision situation, they cannot be proactive in
decision-making. Therefore, the proactive cognitive skill ‘systematic
identification of alternatives’(Alternatives) that takes objectives into
account is considered to be a dimension of PDM.
868 J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877
Fig. 1. Dimensions of proactive decision-making (PDM).
The relevant values guide the collection of information in a de-
cision situation (Keeney, 1992). A proactive individual searches ac-
tively and purposefully for such information that helps to evaluate
alternatives in terms of achieving the relevant values. Reactive in-
dividuals only use information that is available or easily accessible
and do not gather information systematically (Keeney, 1992). If in-
dividuals do not systematically search for relevant information in
a specific decision situation, they cannot be proactive in decision-
making. Therefore, the proactive cognitive skill ‘systematic search for
relevant information’(Information) is regarded as a dimension of
PDM.
Proactive individuals are characterised as future-oriented (Frese
et al., 1996; Frese & Fay, 2001, 2006; Greenglass, 2002). They act in
advance of future situations. This includes not only the anticipation
of future events but also the prevention of future problems and ac-
tive creation of decision opportunities (Frese & Fay, 2001; Weick &
Roberts, 1993; Weick, Sutcliffe, & Obstfeld, 1999). The use of objec-
tives for creating decision opportunities oneself may help to gener-
ate different and better decision situations. Furthermore, decisions
can be planned in a broader context and it can be ensured that the
right problems are dealt with (Howard, 1988). The planning of de-
cisions implies continuous involvement with decisions. Individuals
have influence on their decisions when they thoroughly consider
which decisions are to be made at what time. Instead of dealing with
any challenge in life operatively, isolated from other decisions, reac-
tively, and with limited sight from a worm’s-eye view, individuals can
frame their decisions strategically, taking account of other decisions,
proactively, and in a future-oriented manner the bird’s-eye view. If
individuals are not continuously involved in current and future deci-
sions, they cannot be proactive in their decision-making. Therefore,
the proactive cognitive skill ‘using a decision radar’’ (Decision Radar)
is considered to be dimension of PDM.
Proactive individuals take initiative in decision situations
(Bateman & Crant, 1993; Greenglass, 2002). They want to shape their
environment actively (Ashford & Black, 1996; Kim, Cable, & Kim,
2005; Saks & Ashford, 1996). If individuals do not take initiative in de-
cision situations, they cannot be proactive in their decision-making.
Therefore, the personality trait of ‘taking initiative’(Initiative)isas-
sumed to be a dimension of PDM.
Proactive individuals strive for a “discernible effect on the self
and/or the environment” (Grant & Ashford, 2008). They are inter-
ested in creating a meaningful impact (Buss, 1987; Diener et al., 1984;
Grant & Ashford, 2008). Proactive individuals strive for improvement
in decision situations (Iyengar, Wells, & Schwartz, 2006; Parker et al.,
2010). If an individual does not strive for improvement in decision sit-
uations, they cannot be proactive in their decision-making. Therefore,
another dimension of PDM deals with the personality trait of ‘striving
for improvement’(Improvement). The personality traits Initiative and
Improvement can be regarded as different facets of the commitment
to proactive action in decision processes and they result in different
degrees of effectiveness of decision processes. As long as individuals
do not take initiative, even if they strive for improvement intensely,
this will not lead to proactive action in decision processes. Individuals
who take high initiative without striving for improvement will not be
effective in decision processes either.
3.2. Proactive decision-making in the context of other constructs
On the basis of the working definition indicated above, the con-
cept is concretised by its six dimensions. In this section, these six di-
mensions relevant to PDM are differentiated from related concepts.
Firstly, proactive cognitive skills are contrasted with vigilance (Mann,
Burnett, Radford, & Ford, 1997), rational decision-making (Scott &
Bruce, 1995), and rational thinking (Novak & Hoffman, 2009). Sec-
ondly, proactive traits are differentiated from buck-passing and pro-
crastination (Mann et al., 1997) and maximisation (Schwartz et al.,
2002).
Using a conflict model, Mann et al. (1997) have found empir-
ical evidence for four basic patterns of coping with stress gener-
ated by difficult, potentially threatening decisions: vigilance, buck-
passing, procrastination, and hyper-vigilance. Vigilance “posits a
decision maker who clarifies objectives, considers alternatives, eval-
uates consequences, and thinks through how to implement chosen
options” (Mann et al., 1997, p. 5). On the basis of this definition, the
concept seems closely related to PDM. To illustrate the concept of vig-
ilance in the context of decisions in more detail, we discuss the six
items Mann et al. (1997) used to measure vigilance. Some of these
items are relatively similar to our proactive cognitive skills. The item
I try to be clear about my objectives before choosing” corresponds to
Objectives and the item “I consider how best to carry out a decision
corresponds to aspects of Decision Radar. Since these two vigilance
items may represent two different dimensions of PDM, they appear
suitable for our studies. However, regarding the other vigilance items,
there are huge differences in orientation and extent. The items “Itry
to find out the disadvantages of all alternatives”and“When making de-
cisions I like to collect a lot of information” correspond to Alterna-
tives and Information. But, by using the terms “all” or “a lot of”, they
put emphasis on the quantity of the information sought instead of
its quality, i.e. whether the information is decision-relevant. A fifth
item “I like to consider all of the alternatives” implies that also bad al-
ternatives shall be considered in a decision situation. According to
Keeney (1992), this cannot be effective. It is important to systemat-
ically create a broad set of good alternatives. The last item “I take
a lot of care before choosing” seems quite general. In our study, tak-
ing intensive care before making a decision is implied by systemati-
cally searching for information (Information), identifying objectives
(Objectives), and creating alternatives (Alternatives). In summary,
it can be concluded that vigilance addresses all of the four proactive
cognitive skills. However, the cognitive skills in PDM are rather char-
acterised by purposefulness than the amount of implementation.
Scott and Bruce (1995) identify four decision-making styles
that represent the way how individuals usually react in decision
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 869
situations: a rational decision-making style, an intuitive decision-
making style, a dependent decision-making style and an avoidant
decision-making style. Rational decision-making basically refers to
cognitive skills. It is measured by four items. The first two items
I make decisions in a logical and systematic way”and“My decision
making requires careful thought” generally describe the approach to
decision-making. The other two items describe the specific actions
that have to be taken to be rational. The item “I double check my infor-
mation sources to be sure I have the right facts before making decisions
corresponds to our dimension Information and the item “When mak-
ing a decision, I consider various options in terms of a specific goal” cor-
responds to our dimension Alternatives.Eventhoughtherational
decision-making style covers aspects of proactive cognitive skills, it
remains at a level that is too general to be applicable to our purpose
of measuring PDM.
Novak and Hoffman (2009) develop a scale for rational and ex-
perimental thinking. Rational thinking is characterised by a thorough
search for and logical evaluation of alternatives. This style is mea-
sured by items such as “I reasoned things out carefully”, “Itackledthis
task systematically”, “I figured things out logically”, and “Iwasvery
aware of my thinking process”. These items show that the concept re-
mains abstract and general. The items of the four cognitive skills of
PDM, by contrast, provide insight into the specific actions, which have
to be taken so that decision-making is considered to be proactive.
Mann et al. (1997) find empirical evidence for procrastination and
buck-passing as two aspects of defensive avoidance. The decision-
makers try to reduce stress by avoiding conflict through procrasti-
nation or transferring responsibility to others. Scott and Bruce (1995)
describe the same phenomenon regarding the avoidant and depen-
dent decision-making styles. Frost and Shows (1993) develop a scale
to measure compulsive indecisiveness. These constructs provide dif-
ferent explanations as to the question why an individual takes ini-
tiative in decision situations. Concerning the concept of PDM, by con-
trast, it is of importance whether, not why, individuals show initiative
in decision-making. Thus, these constructs are exogenous to PDM, es-
pecially to the Initiative dimension, and may serve as an explanation
for the endogenous construct.
Schwartz et al. (2002) distinguish between maximiser and satis-
ficer referring to the degree to which people select ideal alternatives
instead of acceptable alternatives. While maximisers try to find the
best alternative and never settle for second best, satisficers tend to
choose the first acceptable alternative. The Maximisation Scale fo-
cuses on the behaviour in a decision situation. The PDM dimension
Improvement, by contrast, refers to striving for improvement in both,
a specific and general decision situation. It deals with the question
whether individuals are motivated to put effort into decision-making
in order to improve their entire current situation.
As mentioned above, proactive personality traits may have pos-
itive impacts on individuals in different contexts. Burnett, Mann,
and Beswick (1989) identify a modest but significant correlation
between dimensions of Flinder’s Decision-Making Questionnaire
(Mann, 1982), which uses, for example, vigilance, and course satis-
faction of students. Since PDM is either based on these constructs
or at least related to them, we assume a positive impact of PDM
for the user. This assumption is supported by Keeney (1992) who
postulates that using value-focused thinking, what implies the use
of proactive cognitive skills, improves an individual’s quality of life.
For proving this assumption, we analyse a consequence construct of
PDM and have decided to use the adapted Decision Satisfaction Scale
(Fitzsimons, 2000).
4. Operationalisation of constructs
PDM is theoretically presumed to be a (multidimensional) con-
struct, which is not directly observable (DeVellis, 1991). The purpose
of this paper was to develop a valid scale being able to measure PDM.
Therefore, using the definition of the concept and its dimensions as
a basis, indicators had to be found and examined in order to enable
indirect self-report measurements.
Since PDM was an innovative construct, operationalisation had to
consider the adaption of reliable and valid items from previous (re-
lated) concepts that had already been tested in other studies, which
were as closely related as possible, however in different contexts; fur-
thermore, it had to consider the creation of completely new items
that represent the content of the presumed dimensions (Netemeyer,
Bearden, & Sharma, 2003; Hair, Anderson, Tatham, & Black, 2010).
Original items that did not fit perfectly were adapted to our context.
Concerning dimensions for which we did not find an item in litera-
ture, we derived new items from theory. Those items were the re-
sult of several revisions by individual members of our research team
and group discussions that reflect the agreement of the researchers
and experts involved. Each item was translated by native speakers
from English into German and vice versa to ensure content validity.
The original, adapted, and newly created items are listed in Table A1
(Electronic Appendix). All derivations are documented.
To ensure consistency, reflective specified multiple-item-
measures were operationalised for each construct dimension
(Bollen, 1989). In terms of wording, although some similarity among
items was intended to tap the domain, items were phrased sim-
ple, straightforward, and non-redundant, i.e. with a certain degree
of variety to express the amplitude of each construct dimension
(Netemeyer et al., 2003, p. 57). However, this might interfere with
some reliability and validity criteria, since part of the variation is due
to the wider formulation of the items. After analysing the data of the
pre-study, some items had to be eliminated for reliability reasons
and new items were added.
Since we were interested in the relationship between PDM
and other constructs, Decision Satisfaction was operationalised by
choosing appropriate items from a short version of the Decision Sat-
isfaction Scale (Fitzsimons, 2000)thatweadapted.
5. Methodology
At the exploratory stage, the construction of an experimental item
pool was useful in order to explore the actual dimensions (pre-study)
and to find reliable and valid measures. Multiple phases of data col-
lections were processed at the confirmatory stage (main studies) in-
cluding additional items for refinement and validations of the PDM
construct and its dimensions.
A questionnaire was designed and five surveys with indepen-
dent samples were conducted. The questionnaire was piloted and
discussed with experts before data collection in the pre-study. The
model’s and item’s goodness (of fit) need to be evaluated (Churchill,
1979). The questionnaire was modified for the main studies. More
items were tested and retested within this scale development pro-
cess (cf. Netemeyer et al., 2003).
The participants were asked to evaluate PDM statements on a
7-point Likert scale ranging from (1) ‘disagree very strongly’ to (7)
‘agree very strongly’, also providing an ‘I don’t know’-option and to
evaluate their satisfaction on an intensity scale from 0 (extremely un-
satisfied) to 10 (extremely satisfied). The first two surveys were con-
ducted in German and data were collected by paper-and-pencil. Al-
most all students participated at the beginning of a lecture. The third,
fourth, and fifth survey were conducted in English and data were col-
lected by establishing an online survey in NovoEd, an online educa-
tion platform collaborating with well-reputed universities in the USA.
The response rates were 28 percent, 27 percent, and 22 percent.
The data analysis during the scale development was guided by
classical test theory. Factor analyses were employed to assess the
dimensionality, reliability, and validity of the PDM instrument. Ex-
ploratory factor analyses (EFA, principal component factoring in
IBM SPSS with varimax rotation) were used in both, the pre-study
870 J. Siebert, R. Kunz / European Journal of Operational Research 249 (2016) 864–877
and main study I (initial testing of scale, exploring of multidimen-
sionality and judging of measurement items). As recommended by
Netemeyer et al. (2003, p. 149), we employed “EFA and item analyses
to trim/retain items for the final form of a scale.” The factor struc-
ture (components) was proven to be highly reliable at the end of
the item selection process (eigenvalues, scree tests, total variance ex-
plained, communalities, Cronbach’s alphas, and item-to-total correla-
tions are used to explore the factors; according to Fabrigar, Wegener,
MacCallum, & Strahan, 1999). In addition, a covariance-based con-
firmatory factor analysis (CFA) in IBM AMOS, which was part of the
main studies (I and II a–c), served for confirming dimensionality and
reliability as well as for further validation of the previously identi-
fied construct structure and measures. The goodness-of-fit was tested
by the Root Mean Square Error of Approximation (RMSEA .08), chi
square degrees of freedom ratio (X2/d.f.), Standardised Root Mean
Residual (SRMR .08), Normed Fit Index (NFI .9), Tucker Lewis
Index (TLI .9), Comparative Fit Index (CFI .9), and Akaike’s In-
formation Criterion (AIC). Fit statistics were evaluated according to
the thresholds recommended by Akaike (1987), Browne and Cudeck
(1993), Homburg and Baumgartner (1995), Hu and Bentler (1999),
and Arbuckle (2008).
In main study I (refining the scale), reliability as well as construct
validity were tested by composite reliability (Nunnally, 1978), cross-
loadings between different construct dimensions and chi-square dif-
ference tests (Anderson & Gerbing, 1988; Bagozzi & Phillips, 1982;
Jöreskog, 1971). In main studies II (finalising the scale), reliability as
well as convergent and discriminant validity were gauged by compos-
ite reliability (CR .6, Bagozzi & Yi, 1988), average variance extracted
(AVE) and the criterion established by Fornell and Larcker (1981).
Thus, the squared correlations between constructs – expressed by the
maximum shared variance (MSV) – should be lower than their cor-
responding AVE. Regarding the latter “[f]or newly developed scales,
values near the .5 threshold (>.45) seem reasonable” (Netemeyer et
al., 2003, p. 153).
Finally, structural equations modelling (main study II a–c) was
used to explain Decision Satisfaction (endogenous construct) by the
dimensions of PDM (exogenous construct) and gauge the nomological
validity. The inclusion of Decision Satisfaction (DSA) in our study as
a relevant outcome of PDM helped to better understand the influence
of each of the dimensions. It provided further information on the po-
tential effects of PDM and hypotheses for future research can be de-
rived. Applying these analyses multicollinearity among the PDM di-
mensions was tested by detecting variance inflation factors (VIF, Hair
et al., 2010; O’Brien, 2007). Furthermore, independent two-sided t-
tests were applied to prove the significance of differences between
decision-makers’ degree of proactivity regarding their DSA.
All the PDM construct dimensions as well as the DSA construct
were measured at the same time and must be measured appropri-
ately by the subjects’ self-evaluations, since solely each individual
decision-maker can assess and report their own cognitive skills and
personality traits as well as their satisfaction with decisions, i.e. ob-
tainment of data from different sources or measurement in differ-
ent contexts was not applicable in our surveys. As a consequence,
common method bias may be an issue (Podsakoff, MacKenzie, Lee, &
Podsakoff, 2003; Podsakoff & Organ, 1986). This was considered a pri-
ori during the design of our studies and data collections (MacKenzie
& Podsakoff, 2012), e.g. by separating predictor (PDM) and criterion
(DSA) variables in different blocks of the questionnaire, relatively low
complexity and short length of the surveys, emphasis that therewere
no right or wrong answers, provision of an “I don’t know”-option,
limited reverse scoring, application of different scale types for de-
pendent and independent variables, voluntary participation, and pro-
tecting respondent anonymity. Common method bias was further ad-
dressed post hoc by applying Harman’s single factor test (Podsakoff
et al., 2003).
6. Pre-study
The pre-study pursued our third objective of empirically testing
the proactive decision-making (PDM) scale for the first time. This in-
volved the empirical exploration of the construct’s dimensionality as
well as the examination of its measures’ (new and adapted items) re-
liability and validity. In this first study, the six potential dimensions
of PDM were tested: systematic identification of objectives,systematic
search for information,systematic identification of alternatives, using a
decision radar, taking initiative,andstriving for improvement.
Data were collected in November 2013. The participants were full-
time master students enrolled in business-related courses at a Ger-
man university. Overall, 188 participants, 97 males and 73 females
on average 23.2 years old, completed the questionnaire and needed
an average time of approximately 10 minutes. The responses of those
18 participants who were not German native speakers were excluded
from the results, since their responses showed many inconsistencies
in the data.
An exploratory factor analysis with 19 items that represent the
definition of the PDM was conducted. The principal component anal-
ysis extracted five factors with eigenvalues over 1. The rotated five-
factor solution explained 61.7 percent of the total variance. We iter-
atively removed items depending on their communalities and factor
loadings and evaluated the consistency of each factor. In particular,
items Ini_2 and Alt_6 were deleted because of communalities below
.5. Items Rad_1, Rad_2, and Rad_3 were eliminated due to low factor
loadings.
As a result, the reduced 16-item-scale consisted of reliable and
valid measures. The five factors explained 66.1 percent of the total
variance. The Kaiser–Meyer–Olkin value amounted to .759 and was
considered to be middling, i.e. it served as an indication that the fac-
tor analysis was useful for these variables. Factor loadings varied be-
tween .581 and .836 (see Table 1). No item loaded more than .4 on
another factor. Thus, the analysis of the factor loadings revealed that
all items loaded high on the five dimensions.
Each factor represents one dimension of PDM. The explored fac-
tors can be interpreted as the following dimensions: Objectives,
Alternatives,Information,Initiative, and Improvement. Conse-
quently, five out of six postulated dimensions were identified through
explorative data analyses. Since all of the items that correspond to
using a ‘decision radar’ were eliminated, we could not verify the as-
sumed sixth dimension of PDM. For this reason, ‘decision radar’ as a
potential dimension was operationalised by new items. Although the
factor loadings of Information were acceptable, the consistency of
the factor (Cronbach’s alpha) was very low. Thus, we decided to use
different indicators in the main studies.
7. Main studies
The main studies aimed at empirically testing the further devel-
oped and optimised Proactive Decision-Making (PDM) scale (third
objective) by means of larger samples and at explaining Decision Sat-
isfaction (DSA) by the dimensions of PDM (fourth objective).
7.1. Main study I
The participants in main study I were full-time bachelor students
enrolled in business-related courses at a German university. Overall,
567 participants with an average age of 20.53 years completed the
study within approximately 10 minutes. Three hundred eleven male
and 243 female students participated. Thirteen participants did not
reveal their gender.
The EFA extracted six factors (components) with eigenvalues over
1. Again, Cronbach’s alpha and item-to-scale correlations were exam-
ined. Unreliable items were iteratively eliminated and the scale was
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 871
Tabl e 1
Factor loadings and reliability measures (pre-study).
Factors/constructs/dimensions Indicators Reliability analysis
Factor loadings Cronbach’s alpha (corrected) Item-to-scale correlation
Objectives Obj_1 .691 .766 .550
Obj_2 .836 .635
Obj_3 .806 .614
Information Inf_1 .759 .325 .387
Inf_2 .619 .326
Inf_3 .581 .082
Alternatives Alt_1 .655 .714 .498
Alt_2 .716 .366
Alt_3 .720 .585
Alt_4 .705 .579
Initiative Ini_1 .775 .762 .574
Ini_3 .759 .641
Ini_4 .772 .594
Improvement Imp_1 .693 .796 .624
Imp_2 .805 .656
Imp_3 .784 .639
reduced until it proved acceptably valid. During this process, items
Imp_4, Alt_5, Ini_4, Inf_5, Obj_4, Ini_2, and Rad_1 were removed be-
cause of low communalities or low factor loadings. Concerning the
dimensions Initiative (α=.625), Objectives (α=.714), and Alter-
natives (α=.759) Cronbach’s alpha could not be improved by ex-
tending or eliminating one of the items. We eliminated item Inf_1
of dimension Information because of its low item-to-scale corre-
lation to increase Cronbach’s alpha from .704 to .735. We included
item Rad_1 of dimension Decision Radar because Cronbach’s alpha
increased from .706 to .742. As to the dimension Improvement the
elimination of item Imp_1 increased Cronbach’s alpha from .800 to
.840. However, Imp_1 remained because each factor should be mea-
sured by at least three items (Bollen, 1989) and the item tapped the
construct domain and was non-redundant.
Again, the remaining 21 items yielded six factors (components)
with eigenvalues over 1. The Kaiser–Meyer–Olkin value amounted to
.853 and was considered meritorious, i.e. principal component fac-
tor analysis was applicable to these variables. The six-factor solution
using 21 items explained 62.6 percent of the total variance. The ro-
tated component matrix is illustrated in Table 2. Only Obj_2, Alt_4,
and Inf_3 (among proactive cognitive skills) loaded more than .3 on
a second factor. However, no item loaded more than .4 on a second
factor.
The factor Objectives implies the active and systematic identi-
fication of relevant objectives in a decision situation resulting in
awareness of these. Information deals with the active and system-
atic search for information. Alternatives refers to the systematic and
purposeful identification of promising alternatives using objectives.
Decision Radar implies future-orientation and purposeful planning
of decisions. Initiative describes whether individuals take initiative
and change their environment. It has to be noted that concerning
the factor Initiative only reversely formulated, reactive items re-
mained. Thus, the items correlated negatively with their factor. How-
ever, scores were recoded for further analyses to simplify interpre-
tations of the six factors. Improvements deals with an individual’s
inherent desire to improve their situation.
The identified factors were confirmed in a CFA, in which items
loaded significantly and highest on their corresponding constructs
(dimensional factors). Due to relatively low factor loadings and relia-
bilities, items Alt_1, Rad_1 and Rad_4 were removed from the model.
The factor loadings of the remaining 18 items were predominantly
within the desired range of .6 and .9 (Netemeyer et al., 2003, p. 153).
Although indicator reliabilities of Alt_2, Rad_5 and Ini_3 were a bit
low, all values of composite reliabilities were acceptable and above
the threshold of .6 (Bagozzi & Yi, 1988). A certain number of tested
items remained, reflecting each dimension appropriately.
The four cognitive skills related factors correlated with .476 to
.644 among each other and the two personality traits related factors
with .307. Intercorrelations among all six PDM factors across cogni-
tive skills and personality traits ranged between .259 and .465. Dis-
criminant validity was further tested by single degree of freedom
tests. The chi-square differences between one- and two-factor mod-
els for each possible pair of measures as well as the entire model
(PDM) were all significantly above the threshold of 3.841 (Jöreskog,
1971; Anderson & Gerbing, 1988). Thus, support for a sufficient level
of discriminant validity was provided in this study.
Overall, our PDM model showed a very good fit of the data to the
theoretically derived model structure (Table 2). The global fit indices
(RMSEA, X2/d.f., TLI, CFI, SRMR) were within their thresholds; only
the NFI was marginally lower than the recommended criterion, but
rounded up to .9. Thus, the 6-factor model structure and its measures
fit well and were used to further evaluate the PDM scale. Construct
validity was assessed using multiple sources of data in main study II.
7.2. Main study II
The participants in main study II (a), (b) and (c) were individ-
uals who have enrolled in one of NovoEd’s online courses DQ 101:
Introduction to Decision Quality by Carl Spetzler. We had indepen-
dent samples of three of these courses in 2014. The PDM surveys
were administered and promoted by NovoEd before each course
started. Overall, 3872 participants, one third female and two thirds
male, completed the surveys. The participants came from 132 coun-
tries, the majority originated in the USA. They mainly worked in
industries such as banking/financial services/insurance, education,
energy/utilities/chemical, manufacturing, technology, or transporta-
tion. Different groups of age were well represented: 39 percent were
between 18 and 30 years old, 31 percent were between 31 and 40
years old, and 30 percent were older than 40 years. After data screen-
ing, preparation and elimination of cases with missing values a to-
tal of 3307 cases remained; 1245 in sample (a), 1071 in sample (b),
and 991 in sample (c). Most of the participants had substantial ex-
perience in dealing with decision-making professionally: 31 percent
more than 10 years, 18 percent between 5 and 10 years, 32 percent
between 1 and 5 years, and only 19 percent less than 1 year. Further-
more, 48 percent were both decision-maker and decision analyst, 15
percent were decision-maker, 23 percent were decision analyst, and
only14percentwerenoneofboth.
We used CFA to further validate the hypothesised model structure
of the multidimensional construct as well as measures. Due to low
factor loadings and high crossloadings with other factors two items
Alt_4 and Rad_2 were removed and 19 items remained for further
872 J. Siebert, R. Kunz/European Journal of Operational Research 249 (2016) 864–877
Tabl e 2
Factor loadings and reliability measures (main study I).
Factors/constructs/
dimensions
Exploratory factor analysis Reliability analysis Confirmatory factor analysis
Indicators Factor
loadings
Cronbach’s
alpha
(corrected) Item-to-scale
correlation
Standardised
factor loading
Significance
factor loading
Indicator
reliability
Composite
reliability
Objectives Obj_1 .770 .714 .529 .688 ∗∗∗ .473 .717
Obj_2 .723 .571 .676 ∗∗∗ .457
Obj_3 .703 .500 .667 ∗∗∗ .445
Information Inf_2 .761 .735 .504 .619 ∗∗∗ .383 .751
Inf_3 .739 .615 .842 ∗∗∗ .709
Inf_4 .756 .576 .653 ∗∗∗ .426
Alternatives Alt_1 .730 .759 .467 .752
Alt_2 .588 .566 .617 ∗∗∗ .381
Alt_3 .713 .603 .663 ∗∗∗ .440
Alt_4 .510 .596 .837 ∗∗∗ .701
Decision Radar Rad_1 .603 .742 .496 .668
RAd_2 .660 .539 .649 ∗∗∗ .421
Rad_3 .659 .510 .673 ∗∗∗ .453
Rad_4 .644 .426 –
Rad_5 .741 .580 .578 ∗∗∗ .334
Initiative Ini_1 .782 .625 .464 .679 ∗∗∗ .461 .666
Ini_3 .764 .422 .580 ∗∗∗ .336
Ini_5 .697 .418 .634 ∗∗∗ .402
Improvement Imp_1 .731 .800 .535 .625 ∗∗∗ .391 .825
Imp_2 .872 .717 .877 ∗∗∗ .769
Imp_3 .835 .692 .828 ∗∗∗ .686
Overall model fit RMSEA =.050, X2/d.f. =2.035, NFI =.899,
TLI =.930, CFI =.945, SRMR =.045
∗∗∗ significance level: .001.
Tabl e 3
Factor loadings, reliability and validity measures (main study II).
Factors/constructs/
dimensions
Indicators CFA main study II (a), N=1245 CFA ma in stu dy II ( b), N=1071 CFA m ain s tud y II (c ), N=991
Standardised
factor loading
CR AVE MSV Standardised
factor loading
CR AVE MSV Standardised
factor loading
CR AVE MSV
Objectives Obj_1 .804 .835 .628 .697 .836 .847 .649 .661 .824 .838 .634 .635
Obj_2 .813 .823 .823
Obj_3 .760 .755 .738
Information Inf_2 .737 .824 .610 .697 .698 .787 .552 .716 .708 .820 .604 .704
Inf_3 .837 .789 .847
Inf_4 .765 .739 .770
Alternatives Alt_1 .782 .846 .646 .605 .748 .845 .646 .694 .765 .852 .658 .637
Alt_2 .831 .849 .858
Alt_3 .798 .811 .808
Decision Radar Rad_1 .710 .824 .540 .689 .712 .822 .537 .716 .732 .837 .563 .704
Rad_3 .704 .674 .700
Rad_4 .756 .774 .796
Rad_5 .766 .766 .770
Initiative Ini_1 .803 .671 .509 .240 .773 .641 .476 .085 .813 .683 .523 .171
Ini_3 .611 .595 .621
Ini_5 .719 .657 .674
Improvement Imp_1 .790 .838 .634 .328 .807 .832 .625 .359 .783 .831 .623 .295
Imp_2 .877 .867 .863
Imp_3 .713 .688 .715
Overall model fit RMSEA =.068, X2/d.f. =6.690, NFI =.928,
TLI =.923, CFI =.938, SRMR =.057
RMSEA =.070, X2/d.f. =6.680, NFI =.919,
TLI =.914, CFI =.931, SRMR =.058
RMSEA =.062, X2/d.f. =4.745, NFI =.936,
TLI =.936, CFI =.948, SRMR =.051
analyses. Except for Ini_3 (.595) in study (b) all factor loadings were
above .6 and below .9 (see Table 3).
In all three studies the items loaded significantly on the intended
factors. Across the three samples, the composite reliabilities (CR)
were good; compared to the other factors the reliability of Initia-
tive was lower but acceptable. Evaluated on basis of AVE greater than
.45 for newly developed scales (Netemeyer et al., 2003), evidence for
convergent validity existed for all six PDM factors. However, the four
factors related to cognitive skills, i.e. Objectives, Information, Alter-
natives, and Radar, faced discriminant validity issues that had to be
considered in further analyses. They had relatively high intercorrela-
tions (between .732 and .830 on average across the three samples)
resulting in high MSV values that in main part were above their AVE
values. Intercorrelations among Initiative and Improvement (.260 on
average) as well as between cognitive skills factors and personality
traits factors (between .281 and .572 on average) supported a suf-
ficient distinctiveness. Indeed, the average shared variance of each
factor was below the AVE and the chi-square differences were all sig-
nificant as well. Regarding the overall goodness, adequate levels of
model fit for RMSEA, NFI, TLI, CFI, and SRMR were found. Only the
X2/d.f. ratios were relatively high due to the large sample sizes.
The six dimensions represent different facets of PDM. Objectives,
Information, Alternatives, and Radar all measure cognitive skills.
Since proactive decision-makers who are cognitively skilled usually
define and follow their objectives, search for information, evaluate al-
ternatives and monitor their ‘decision radar’, it is apparent that these
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 873
Tabl e 4
Goodness-of-fit indices for sub groups (main study II).
Groups NRMSEA X2/d.f. NFI TLI CFI SRMR
Neither decision-maker, nor decision analyst 448 .064 2.815 .916 .930 .944 .050
Decision-maker 513 .072 3.688 .917 .923 .938 .065
Decision analyst 774 .062 3.935 .924 .927 .942 .057
Decision-maker and decision analyst 1572 .069 8.425 .931 .913 .931 .059
Experience in decision-making less1 year 631 .0 61 3.366 .931 .938 .950 .052
Experience between 1 and 5 years 1062 .0 68 5.263 .932 .930 .944 .051
Experience between 5 and 10 years 576 .074 4.123 .902 .924 .923 .074
Experience more than 10 years 1038 .071 4.471 .918 .912 .930 .060
Tabl e 5
Goodness-of-fit indices and squared multiple correlations for the four substantive models (main study II).
Model RMSEA X2/d.f. NFI TLI CFI SRMR AIC R2
Model 1 Six 1st order factors: objectives, information, alternatives, radar, initiative, and
improvement
.059 12.481 .944 .937 .948 .050 2476 .305
Model 2 One 2nd order factor (cognitive skills): objectives, information, alternatives, and
radar; Two 1st order factors: initiative and improvement
.060 12.868 .939 .934 .944 .054 2669 .483
Model 3 One 2nd order factor (PDM): objectives, information, alternatives, radar,
initiative and improvement
.060 13.050 .937 .933 .942 .056 2738 .479
Model 4 One 3rd order factor (PDM): cognitive skills, initiative and improvement.One
2nd order factor (cognitive skills): objectives, information, alternatives, and radar
.061 13.109 .937 .933 .942 .056 2739 .466
Tabl e 6
Fornell–Larcker criterion (AVE on diagonal, correlations and squared correlations between constructs above
and below diagonal), composite reliability and standardised path coefficients for model 1 (main study II).
Obj Inf Alt Rad Ini Imp DSA CR Path coefficients
Objectives .636 .816 .733 .809 .343 .572 .605 .840 .215∗∗∗
Information .666 .590 .791 .830 .295 .536 .575 .811 .057
Alternatives .537 .626 .649 .784 .405 .538 .655 .847 .375∗∗∗
Radar .654 .689 .615 .546 .281 .514 .642 .828 .304∗∗∗
Initiative .118 . 087 .164 .0 79 .499 .260 .352 .662 .145∗∗∗
Improvement .327 .287 .289 .264 .068 .628 .349 .834 .044
DSA .366 .331 .429 .412 .124 .122 .775 .912
significance level: .05.
∗∗∗ significance level: .001.
dimensions can be highly intercorrelated. Proactive cognitive skills
may account for their substantial intercorrelations. Both Initiative
and Improvement represent personality traits. However, even less
initiative decision-makers may strive for improvement and vice versa.
An intercorrelation of these two (separate) dimensions may only be
low. The same is true for intercorrelations across skills and traits.
A comparison of the goodness-of-fit statistics of different groups
of professionals and experience levels revealed that the model struc-
ture was relatively stable across these groups showing similar results
(see Table 4). However, the model fit better for decision analysts and
less experienced subjects. The pattern of factor loadings was very
similar across all groups (see Electronic Appendix, Tables A2 and A3).
Thus, measurement and factorial invariance was found.
7.3. Explanation of decision satisfaction
By analysing the structural relationships between PDM and other
concepts the nomological validity was tested. PDM was considered
as an antecedent of decision satisfaction (DSA). Since DSA was mea-
sured with an adapted short scale introduced by Fitzsimons (2000),
we ensured validity and reliability by another factor analysis. By ap-
plying the same procedure as described above, one factor was formed
with high factor loadings among three items (Dsa_p, Dsa_d, Dsa_s).
In terms of the relationships between the six PDM dimensions,
one has to consider PDM as being a higher order construct (Hair
et al., 2010), i.e. PDM as a second order or even a third order con-
struct comprising cognitive skills and personality traits as second or-
der constructs. Thus, in the context of other concepts one has to de-
cide which PDM dimensions should be included in which order. We
postulated four substantive first and higher order models and
examined their impact on DSA in a structural equations model (SEM).
The SEM was carried out on the total data of main study II (N=3307).
The goodness-of-fit indices for the four models and the squared mul-
tiple correlations coefficients (R2) of decision satisfaction are pre-
sented in Table 5.
Although the four competing models all fit adequately to the em-
pirical data (again X2/d.f. ratios were high due to the large sample
size), model 1 had the best overall fit (RMSEA =.059), followed by
model 2 and 3 (RMSEA =.060). A comparison on basis of AIC sugges-
ted that model 1 should be the preferred model, followed by model 2.
Multicollinearity should not be an issue when explaining DSA
since the PDM factors had initially been extracted in an EFA, the reli-
ability of the measures as well as the explanatory capacity were high
and sample size was large (Grewal, Cote, & Baumgartner, 2004). But to
ensure low multicollinearity, it was tested in linear regression analy-
sis explicitly. The VIF values ranged from 1.121 to 2.553, i.e. all values
were close to 1 and clearly below the threshold of 3 (Hair et al., 2010;
O’Brien, 2007).
The six PDM factors and the DSA factor were intercorrelated (see
Table 6 for model 1). However, with a composite reliability of .912, an
AVE of .775 and a MSV of .429 the DSA construct was reliable, conver-
gent valid and discriminated sufficiently from PDM, since support for
the Fornell–Larcker criterion was provided for DSA in relation to the
six PDM factors.
In particular, the four cognitive skills factors were highly intercor-
related, which was considered in model 2 and model 4, but not in
model 3. However, model 4 is very abstract and its interpretation in
the context of other concepts is complex. Regarding the presented
higher order models 2–4 model 2 provides the lowest AIC value and
might be an applicable variant for explaining DSA or other concepts,
874 J. Siebert, R. Kunz/ European Journal of Operational Research 249 (2016) 864–877
Tabl e 7
Fornell-Larcker criterion (AVE on diagonal, correlations and squared correlations between constructs above
and below the diagonal), composite reliability and standardised path coefficients for model 2 (main study II).
Cognitive Skills Initiative Improvement DSA CR Path coefficients
Skills .798 .369 .602 .694 .922 .679∗∗∗
Initiative .136 .490 .258 .352 .740 .144∗∗∗
Improvement .362 .067 .628 .347 .834 .033
DSA .4 82 .124 .12 0 .776 .912
significance level: .05.
∗∗∗ significance level: .001.
when a conceptually distinct scale with acceptable internal consis-
tency is needed. In the following we highlight model 2. Regarding re-
liability and validity of Cognitive Skills as a second order construct
as well as of Initiative, Improvement and Decision Satisfaction as
first order constructs, high levels of composite reliability as well as
convergent and discriminant validity were achieved (see Table 7;for
detailed results on model 2 regarding different groups of profession-
als and experience levels see Electronic Appendix, Tables A5 and A6).
Whereas model 1 explained about 30.5 percent of the DSA con-
struct’s variance, model 2 explained about 48.3 percent of DSA.
Both models were supported by theory as was conceptualised in
Section 3. However, model 1 allows for more detailed analyses of the
influences on DSA. Overall, regarding model 1 the standardised path
coefficients of all six factors were significant. Alternatives (γ=.375)
and Radar (γ=.304) had the strongest influence on DSA. Informa-
tion and Improvement had only weak impacts. Regarding model 2
Cognitive Skills as a second order construct had the largest impact
on DSA (γ=.679). The influences of Initiative and Improvement
were similar as in model 1.
The explained variances of 30.5 percent (model 1) and 48.3 per-
cent (model 2) respectively were moderate but acceptable since DSA
was only explained by one concept in this case, i.e. PDM. Consider-
ing the uncertainty in decision-making and the distinction between
a decision and its outcome (Howard, 1988) as well as numerous pos-
sible antecedents of satisfaction with one’s decision in addition to
PDM, such as anticipated regret, evaluation costs, choice confidence
(Heitmann, Lehmann, & Herrmann, 2007), etc., and the distorted per-
ception of a decision, PDM represented a relevant predictor with sig-
nificant explanatory power.
The average DSA of 814 participants with high Improvement
(mean: 6.584, standard deviation 1.484) was significantly better than
of 2493 participants with low Improvement (5.682, 1.569). The av-
erage DSA of 1032 participants with high Initiative (6.910, 1.440)
was significantly better than of 2275 participants with low Initia-
tive (6.114, 1.541). The average DSA of 1890 participants with high
Cognitive Skills (7.015, 1.280) was significantly better than of 1417
participants with low cognitive skills (5.492, 1.459). All differences
were significant on a .001-level (t-statistic). We concluded that Ini-
tiative,Improvement, and proactive Cognitive Skills were neces-
sary to achieve a high Decision Satisfaction.
Although procedural controls of method bias had been imple-
mented ad hoc, common method variance was gauged statistically
post hoc. By extracting a single factor in an unrotated EFA including
all manifest variables of PDM and DSA, in both main studies consid-
erably less than 50 percent of the variance was explained by a single
factor. Thus, common method bias had been sufficiently controlled
and did not appear to be problematic in these studies (Podsakoff
et al., 2003). In main study II the common method variance was a bit
higher than in main study I. The reason could be that people respond
differently online than in a paper-and-pencil survey.
8. Implications for OR, limitations, and further research
The construct structure of PDM is composed of six dimen-
sions/factors (components). Acceptably reliable items ensuring valid
measurements were created for the entire construct and each di-
mension/factor. The created and tested items serve as proxies for
the construct and its dimensions and are well applicable in research
and daily business. Nevertheless, the contributions include that more
measurement work is needed on the relatively low AVE measures.
Achieving high levels of convergent and discriminant validity among
dimensions of a multidimensional construct is a long term objective.
Each dimension of the PDM construct should be re-evaluated con-
stantly and could be highlighted in separate studies in order to gen-
erate a high number of consistent multi-item-measures as well as
single-item-measures that discriminate adequately from other PDM
dimensions as well as different constructs. Hierarchical confirmatory
factor analysis multi-trait multi-method approach could be used to
explore the validity of the PDM as higher order construct in future
studies (Guo, Aveyard, Fielding, & Sutton, 2008).
We suggested two related models, a first order only model, treat-
ing each of the six PDM factors individually, and a second order
model, comprising the highly intercorrelated cognitive skills factors
in addition to the two separate personality traits factors. Since both
models exhibit a similar goodness-of-fit they can be applied depend-
ing on the specific research context. Our results are remarkably sim-
ilar for individuals who are decision-makers, decision analysts, both
or none of both with different levels of experience. However, our scale
needs to be further tested for different populations.
The PDM scale, as any other self-report scale, has the limitation
of self-evaluations. Some people may argue that decision processes
are mainly inner psychological processes. However, some parts of
the process are often observable. Therefore, it may be interesting to
analyse if the results of self-estimation by means of PDM and ob-
servations by other people (observers, superiors, or colleagues) are
similar. Here, we can expect a high correlation. Thus, the multidi-
mensional PDM scale could be further validated by applying multi-
ple methods regarding the correlation between the decision-makers’
self-evaluations of multi-item measures on the one hand and third
party observations, responses, or ratings on single-item measures us-
ing a different scale on the other hand (Netemeyer et al., 2003, p. 80).
Involving various data sources as well as data collections in multiple
points in time also contributes to control common method bias.
The level of PDM can be used to group individuals. Individuals can
be classified and described as being proactive or reactive decision-
makers. In particular, PDM can be used for explanation and predic-
tion purposes in studies dealing with individuals’ satisfaction with
their decision-making or actual decisions. Especially satisfaction is a
multi-attributive construct, which is determined by a variety of dif-
ferent factors. Many conceptual papers, empirical research studies,
and experiments regard satisfaction as a mediator variable in cause-
and-effect modelling. In our study, we found empirical evidence that
PDM has a significant influence on decision satisfaction. Although de-
cision satisfaction is already explained moderately by PDM, further
research is required to identify other concepts and to analyse their in-
fluence on decision satisfaction in combination with PDM. In future,
the consideration of PDM will enable academic, business, OR and
marketing researchers as well as psychologists to analyse individuals’
satisfaction as a predictor of behaviour in more detail. PDM should
be validated as a focal construct in a nomological network with
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 875
different antecedents and consequences as well as potential moder-
ators and mediators (Edwards, 2001; MacKenzie, Podsakoff, & Pod-
sakoff, 2011). We follow Calder, Phillips, and Tybout (1983 ,p.113)ar-
gument that “[t]heory must be the driving force in designing theory-
testing research” and thus a theory can only be supported by con-
vergence, discriminant and external validity. However, the latter
is deemed as less important. Apart from concepts that are poten-
tially influenced by PDM (directly and indirectly), future research
should also discuss and analyse the antecedents of PDM, especially
of each dimension. In this regard, research questions may focus on
the reasons why individuals are proactive in their decision-making.
As to that, psychological questions could be analysed. Each indi-
vidual has a long learning history (e.g. operant conditioning pro-
cesses and learning by models). Some of these results will influence
their decision-making. In accordance with previous literature (e.g.
VandenBos, 2007), we suggest that the cognitive skills can be trained
or learned more easily compared to personality traits. This presump-
tion should be verified empirically in a study in which participants
are given the PDM questionnaire twice, i.e. before and after a course
on decision-making. On average, the scores in cognitive skills should
increase while the scores in personality traits should remain stable.
The results could be used to optimise the course on decision-making
in respect to increasing the participants’ cognitive skills even more.
Franco and Meadows (2007) emphasise the importance of cog-
nitive style in PSM research and application. The PDM scale can be
applied in all research questions in which other scales are used to
gain more insights about the cognitive skills and personality traits in
decision-making and problem structuring. Experiments on hypothe-
ses derived by Franco and Meadows (2007) and empirical findings
(Garfield et al., 2001) regarding the impact of Jung’s theory of psy-
chological types in context of PSM could be complemented by a scale
that covers in particular cognitive skills and personality traits that
are relevant in the generating phase in problem structuring. For ex-
ample, Garfield et al. (2001) found empirical evidence that innova-
tive, radical alternatives are created more often by intuitive and feel-
ing individuals than by sensing and thinking individuals. The results
of Siebert and Keeney (2013) indicate that using objectives enhances
the quality and quantity of created alternatives. Proactive individuals
use objectives to create alternatives (Keeney, 1992). The PDM scale al-
lows identifying proactive individuals. Therefore, this scale may help
to understand why intuitive and feeling individuals are more produc-
tive in creating alternatives. Another example is that the PDM scale
may be useful to explain individual differences and conceptual mod-
elling task performance (Dhillon & Dasgupta, 2011).
We have already discussed the aspect that PDM may be linked to
procrastination and buck-passing (defensive avoidance, Mann et al.,
1997; avoidant and dependent decision-making styles, Scott & Bruce,
1995) as well as compulsive indecisiveness (Frost & Shows, 1993).
We expect that those individuals who buck-pass or procrastinate will
have lower scores in ‘initiative’ and that maximisers will tend to have
higher scores in ‘improvement’ than satisficers. In our conceptuali-
sation Section 3, we discuss the similarities and differences between
proactive cognitive skills and vigilance (Mann et al., 1997), the ratio-
nal thinking style (Epstein, 1973, 1983, 1985, 1994,and2003), and
the situation specific thinking style (Novak & Hoffman, 2009). We are
convinced that proactive cognitive skills will be useful to substantiate
these rational thinking styles. It should be further analysed whether
rational and vigilant individuals tend to have higher scores in proac-
tive cognitive skills compared to non-rational and non-vigilant indi-
viduals.
According to Murphy and Davidshofer (1998) there is a huge need
for psychological tests in practice and they defined three major fields
of application for psychological testing, namely educational, person-
nel, and clinical testing.For the application of the PDM scale mainly
the first two areas are of importance. The purpose of these tests is to
classify and assign the subjects to different categories, which serve as
source of information and therefore as basis for a specific decision for
the observer. Furthermore, the results of the tests provide informa-
tion on certain characteristics of the subject (Murphy & Davidshofer,
199 8).
The PDM scale could be used in assessment centres to evaluate
applicants regarding their cognitive skills and personality traits rel-
evant in phases of problem structuring and decision-making, similar
to scales like the Myers-Briggs Type Indicator to ascertain a person’s
basic preferences. Applicants with low scores in proactive cognitive
skills and proactive personality traits (28.6 percent in main study II)
could be regarded as less suitable for certain positions. The PDM scale
could also be applied to personnel. Especially, at the stage of problem
structuring, which is crucial for every OR application, it is highly im-
portant to have individuals with high proactive cognitive skills in a
team since even the best method or algorithm will be ineffective as
long as the problem is framed wrongly. In almost all OR applications
it could be useful to have a team that consists of individuals with high
proactive personality traits and individuals with high proactive cog-
nitive skills who complement each other.
Furthermore,thePDMscalecanbeusedtocapturebehavioural
aspects in OR procedures. Companies often ask analysts for support
in important and complex decision situations (e.g. building a new
power plant, outsourcing of production, adopting new technologies,
etc.). For offering tailor-made support, the decision analyst needs to
understand the decision-making in a company with regard to the or-
ganisation itself and its stakeholders, i.e. managers and personnel.
Completing the PDM scale a week before the kick-off meeting is a
cost-effective option to measure the PDM of the individuals and the
entire company. In contrast to individual conversations, this would
save time and in contrast to group discussions, the use of the PDM
questionnaire would prevent single opinion leaders from dominating
the results, since certain individuals often lead decisions in compa-
nies (Crant, 1996).
9. Conclusion
This paper aims at developing a theoretically sound and empiri-
cally tested proactive decision-making (PDM) scale. Therefore, PDM is
conceptualised, operationalised, measured, validated, and modelled
with regard to explaining decision satisfaction.
We derive the concept from previous literature and identify six
dimensions that describe necessary proactive cognitive skills (1–4)
and proactive personality traits (5, 6) of individuals: (1) a system-
atic and active search for objectives (Objectives),(2)apurpose-
ful and active search for information (Information), (3) a purpose-
ful and systematic identification of alternatives (Alternatives), (4)
a future-oriented and purposeful planning of decisions (Decision
radar), (5) taking initiative (Initiative), and (6) the inherent desire
to improve one’s situation by striving for improvement (Improve-
ment). By means of five studies, these six factors are identified of
being of importance in PDM. On the basis of theoretical considera-
tions and empirical results, PDM can therefore be defined briefly as
follows:
Proactive decision-making summarises the purposeful use of cog-
nitive skills and certain foresighted personality traits of the
decision-maker.
We argue that both, proactive cognitive skills and proactive per-
sonality traits, are relevant aspects of PDM. Furthermore, we as-
sume that both aspects complement each other. The comparison of
the decision satisfaction of participants with low and high scores in
proactive cognitive skills and proactive personality traits respectively
empirically verify this assumption. Since proactive personality traits
are characterised to be relatively stable, individuals should enhance
their cognitive skills in decision-making to achieve a higher decision
satisfaction.
876 J. Siebert, R. Kunz/ European Journal of Operational Research 249 (2016) 864–877
Acknowledgements
The authors thank the editors, the anonymous reviewers, Ralph
L. Keeney, Richard John, Jörg Schlüchtermann, Herbert Woratschek,
Maresa Heisig, Hannah Weber and Bea Semba for their valuable
comments.
Supplementary Materials
Supplementary material associated with this article can be found,
in the online version, at doi:10.1016/j.ejor.2015.06.066.
References
Akaike, H. (1987). Factor analysis and AIC. Psychometrika, 52, 317–332.
Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A
review and recommended two-step approach. Psychological Bulletin, 103, 411–423.
Arbuckle, J. (2008). Amos 17.0 user’s guide. Chicago, Ill: SPSS Inc..
Ashford, S. J., & Black, J. S. (1996). Proactivity during organizational entry: The role of
desire for control. Journal of Applied Psychology, 81, 199–214.
Bagozzi, R. P., & Phillips, L. W. (1982). Representing and testing organizationaltheories:
A holistic construal. Administrative Science Quarterly, 27, 459–489.
Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal
of the Academy of Marketing Science, 16, 74–94.
Bateman, T. S., & Crant, J. M. (1993). The proactive component of organizational behav-
ior: A measure and correlates. Journal of Organizational Behavior, 14, 103–118.
Bell, D. E., Raiffa, H., & Tversky, A. (1988). Decision making. Descriptive, normative, and
prescriptive interactions. Cambridge: Cambridge University Press.
Benbasat, I., & Dexter, A. S. (1982). Individual differences in the use of decision support
aids. Journal of Accounting Research, 20, 1–11.
Bollen, K. A. (1989). Structural equations with latent variables.NewYork:Wiley.
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In
K. A. Bollen, & J. S. Long (Eds.), Testing structural equation models (pp. 136–162).
Beverly Hills: SAGE.
Bruine de Bruin, W., Parker, A. M., & Fischhoff, B. (2012). Explaining adult age differ-
ences in decision-making competence. Journal of Behavioral Decision Making, 25,
352–360.
Burnett, P. C., Mann, L., & Beswick, G. (1989). Validation of the flinders decision making
questionnaire on course decision making by students. Australian Psychologist, 24,
285–292.
Buss, D. M. (1987). Selection, evocation, and manipulation. Journal of Personality and
Social Psychology, 53, 1214–1221.
Butler, A. B., & Scherer, L. L. (1997). The effects of elicitation aids, knowledge, and prob-
lem content on option quantity and quality. Organizational Behavior and Human
Decision Processes, 72, 184–202.
Calder, B. J., Phillips, L. W., & Tybout, A. M. (1983). Beyond external validity. Journal of
Consumer Research, 10, 112–114.
Churchill, G. A., Jr. (1979). A paradigm for developing better measures of marketing
constructs. Journal of Marketing Research, 16, 64–73.
Corbett, C. J., Overmeer, W. J., & Van Wassenhove, L. N. (1995). Strands of practice in
OR (the practitioner’s dilemma). European Journal of Operational Research, 87, 484–
499.
Crant, J. M. (1996). The proactive personality scale as a predictor of entrepreneurial
intentions. Journal of Small Business Management, 34, 42–45.
Crant, J. M., & Bateman, T. S. (2000). Charismatic leadership viewed from the impact of
proactive personality. Journal of Organizational Behavior, 21, 63–75.
DeVellis, R. F. (1991). Scale development. Theory and applications. Newbury Park, Califor-
nia: SAGE.
Dhillon, M. K., Dasgupta, S., et al. (2011). Individual differences and conceptual mod-
eling task performance: Examining the effects of cognitive style, self-efficacy,
and application domain knowledge. In W. van der Aalst, J. Mylopoulos, M. Rose-
mann, M. J. Shaw, C. Szyperski, & T. Halpin, et al. (Eds.), Enterprise, business-process
and information systems modeling, Lecture notes in business information processing
(pp. 483–496). Berlin, Heidelberg: Springer Berlin Heidelberg.
Diener, E., Larsen, R. J., & Emmons, R. A. (1984). Person situation interactions: Choice of
situations and congruence response models. Journal of Personality and Social Psy-
chology, 47, 580–592.
Driver, M. J. (1979). Individual decision making and creativity. In S. Kerr (Ed.), Organi-
zational behavior. Columbus: Grid Publishing.
Driver, M. J., Brousseau, K. R., & Hunsaker, P. L. (1990). The dynamic decisionmaker. Five
decision styles for executive and business success.NewYork:Harper&Row.
Edwards, J. R. (2001). Multidimensional constructs in organizational behaviorresearch:
An integrative analytical framework. Organizational Research Methods, 4, 144–192.
Epstein, S. (1973). The self-concept revisited: Or a theory of a theory. American Psychol-
ogist, 28, 404–416.
Epstein, S. (1983).The unconscious, the preconscious, and the self-concept.In J. M. Suls,
&A.G.Greenwald(Eds.).Psychological perspectives on the self: 2. Psychology Press.
Epstein, S. (1985). The implications of cognitive-experiential self-theory for research
in social psychology and personality 1. Journal for the Theory of Social Behaviour, 15,
283–310.
Epstein, S. (1994). Integration of the cognitive and the psychodynamic unconscious.
American Psychologist, 49, 709–724.
Epstein, S. (2003). Cognitive-experiential self-theory. In T. Millon, M. J. Lerner, &
I. B. Weiner (Eds.). Comprehensive handbook of psychology, personality and social
psychology:5.NewYork:JohnWiley&Sons,Inc..
Epstein, S., Pacini, R., Denes-Raj, V., & Heier, H. (1996). Individual differences in
intuitive–experiential and analytical–rational thinking styles. Journal of Personal-
ity and Social Psychology, 71, 390–405.
Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the
use of exploratory factor analysis in psychological research. Psychological Methods,
4, 272–299.
Fitzsimons, G. J. (2000). Consumer response to stockouts. Journal of Consumer Research,
27, 249–266.
Fornell, C., & Larcker, D. F. (1981). Evaluation structural equation models with unob-
servable variables and measurement error. Journal of Marketing Research, 18, 39–
50.
Franco, L. A., & Meadows, M. (2007). Exploring new directions for research in prob-
lem structuring methods: On the role of cognitive style. Journal of the Operational
Research Society, 58, 1621–1629.
Franco, L. A., & Montibeller, G. (2010). Facilitated modelling in operational research.
European Journal of Operational Research, 205, 489–500.
Frese, M. (2006). Grand theories and mid-range theories: Cultural effects on theoriz-
ing and the attempt to understand active approaches to work. In K. G. Smith, &
M. A. Hitt (Eds.), Great minds in management. The process of theory development
(pp. 84–108). Oxford: Oxford University Press.
Frese, M., & Fay, D. (2001). Personal initiative: An active performance concept for work
in the 21st century. Research in Organizational Behavior, 23, 133–187.
Frese, M., Kring, W., Soose, A., & Zempel, J. (1996). Personal initiative at work: Differ-
ences between East and West Germany. Academy of Management Journal, 39,37
63.
Frost, R. O., & Shows, D. L. (1993). The nature and measurement of compulsive indeci-
siveness. Behaviour Research and Therapy, 31, 683–692.
Furby, L., & Beyth-Marom, R. (1992). Risk taking in adolescence: A decision-making
perspective. Developmental Review, 12, 1–44.
Garfield, M., Taylor, N., Dennis, A., & Satzinger, J. (2001). Research report: Modifying
paradigms-individual differences, creativity techniques, and exposure to ideas in
group idea generation. Information Systems Research, 12, 322–333.
Gettys, C. F., Pliske, R. M., Manning, C., & Casey, J. T. (1987). An evaluation of human act
generation performance. Organizational Behavior and Human Decision Processes, 39,
23–51.
Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans.
American Psychologist, 54, 493–503.
Gollwitzer, P. M., & Brandstätter, V. (1997). Implementation intentions and effective
goal pursuit. Journal of Personality and Social Psychology, 73, 186–199.
Gollwitzer, P. M., & Oettingen, G. (1998). The emergence and implementation of health
goals. Psychology & Health, 13, 687–715.
Grant, A. M., & Ashford, S. J. (2008). The dynamics of proactivity at work. Research in
Organizational Behavior, 28, 3–34.
Greenglass, E. (2002). Proactive coping. In E. Frydenberg (Ed.), Beyond coping. Meet-
ing goals, visions, and challenges (pp. 37–62). Oxford, New York: Oxford University
Press.
Grewal, R., Cote, J. A., & Baumgartner, H. (2004). Multicollinearity and measurement
error in structural equation models: Implications for theory testing. Marketing Sci-
ence, 23, 519–529.
Guo, B., Aveyard,P., Fielding, A., & Sutton, S. (2008). Testing the convergentand discrim-
inant validity of the decisional balance scale of the transtheoretical model using
the multi-trait multi-method approach. Psychology of Addictive Behaviors: Journal
of the Society of Psychologists in Addictive Behaviors, 22, 288–294.
Hair,J.F.,Anderson,R.E.,Tatham,R.L.,&Black,W.C.(2010).Multivariate data analysis.
New Jersey: Prentice Hall.
Hämäläinen, R. P., Luoma, J., & Saarinen, E. (2013). On the importance of behavioral op-
erational research: The case of understanding and communicating about dynamic
systems. European Journal of Operational Research, 228, 623–634.
Harren, V. A. (1979). A model of career decision making for college students. Journal of
Vocational Behavior, 14, 119–133.
Heitmann, M., Lehmann, D. R., & Herrmann, A. (2007). Choice goal attainment and
decision and consumption satisfaction. Journal of Marketing Research, 44,234
250.
Homburg, C., & Baumgartner, H. (1995). Beurteilung von kausalmodellen: Bestandsauf-
nahme und anwendungsempfehlungen. Marketing ZFP – Journal of Research and
Management, 17, 162–176.
Hough, J. R., & Ogilvie, D. (2005). An empirical test of cognitive style and strategic de-
cision outcomes.Journal of Management Studies, 42, 417–448.
Howard, R. A. (1988). Decision analysis: Practice and promise. Management Science, 34,
679–695.
Hu, L.-T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation Model-
ing, 6, 1–55.
Iyengar, S. S., Wells, R. E., & Schwartz, B. (2006). Doing better but feeling worse. Looking
for the “best” job undermines satisfaction. Psychological Science, 17, 143–150.
Jöreskog, K. G. (1971). Simultaneous factor analysis in several populations. Psychome-
trika, 36, 409–426.
Jung, C. G. (1971). The collected works of C.G. Jung. Vol. 6 Psychological Types. Princeton,
NJ: Princeton University Press.
Jungermann, H., Ulardt, I. V., & Hausmann, L. (1983). The role of the goal for generating
actions. In P. Humphreys, O. Svenson, & A. Vári (Eds.), Analysing and aiding decision
processes, advances in psychology:14 (pp. 223–236). North Holland: Elsevier Science
Ltd.
J. Siebert, R. Kunz /European Journal of Operational Research 249 (2016) 864–877 877
Kahnemann, D., & Frederick, S. (2002). Representativeness revisited: Attribute sub-
stitution in intuitive judgment. In T. Gilovich, D. Griffin, & D. Kahnemann (Eds.),
Heuristics of intuitive judgment: Extensions and applications (pp. 49–81). New York:
Cambridge University Press.
Keeney, R. L. (1992). Value-focused thinking. A path to creative decision making. Cam-
bridge, Mass., London: Harvard University Press.
Kim, T., Cable, D. M., & Kim, S. (2005). Socialization tactics, employee proactivity, and
person-organization fit. The Journal of Applied Psychology, 90, 232–241.
Kirby, E. G., Kirby, S. L., & Lewis, M. A. (2002). A study of the effectiveness of training
proactive thinking. Journal of Applied Social Psychology, 32, 1538–1549.
Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal set-
ting and task motivation: A 35-year odyssey. American Psychologist, 57,705
717.
Lu, H., Yu, H., & Lu, S. S. (2001). The effects of cognitive style and model type on DSS
acceptance: An empirical study. EuropeanJournal o f Operational Research, 131,649–
663.
MacKenzie, S. B., & Podsakoff, P. M. (2012). Common method bias in marketing: Causes,
mechanisms, and procedural remedies. Journal of Retailing, 88, 542–555.
MacKenzie, S. B., Podsakoff, P. M., & Podsakoff, N. P. (2011). Construct measurement
and validation procedures in MIS and behavioral research: Integrating new and
existing techniques. MIS Quarterly, 35, 293–334.
Mann, L. (1982). Flinders decision making questionnaire I and II. Unpublished question-
naires. The Flinders University of South Australia, School of Social Sciences.
Mann, L., Burnett, P., Radford, M., & Ford, S. (1997). The Melbourne decision making
questionnaire: An instrument for measuring patterns for coping with decisional
conflict. Journal of Behavioral Decision Making, 10, 1–19.
McGrath, J. E. (1984). Groups, interaction and performance. Englewood Cliffs, N.J.:
Prentice-Hall.
Mingers, J., & Rosenhead, J. (2001). Rational analysis for a problematic world revis-
ited. Problem structuring methods for complexity,uncertainty and conflict.Chichester,
New York: Wiley.
Murphy, K. R., & Davidshofer, C. O. (1998). Psychological testing. Principles and applica-
tions. Upper Saddle River, N.J .: Prentice Hall.
Netemeyer, R. G., Bearden, W. O., & Sharma, S. (2003). Scaling procedures. Issues and
applications. Thousand Oaks, C A: Sage Publications.
Novak, T. P., & Hoffman, D. L. (20 09). The fit of thinking style and situation: New mea-
sures of situation-specific experiential and rational cognition. Journal of Consumer
Research, 36, 56–72.
Nunnally, J. C. (1978). Psychometric theory.NewYork:McGraw-Hill.
O’Brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors.
Quality & Quantity, 41, 673–690.
Ormerod, R. J. (2008). The transformation competence perspective. Journal of the Oper-
ational Research Society, 59, 1435–1448.
Parker, S. K., Bindl, U. K., & Strauss, K. (2010). Making things happen: A model of proac-
tive motivation. Journal of Management, 36, 827–856.
Pitz, G. F., Sachs, N. J., & Heerboth, J. (1980). Procedures for eliciting choices in the anal-
ysis of individual decisions. Organizational Behavior and Human Performance, 26,
396–408.
Podsakoff, P. M., MacKenzie, S. B., Lee, J., & Podsakoff, N. P. (2003). Common method
biases in behavioral research. Journal of Applied Psychology, 88, 879–903.
Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Prob-
lems and prospects. Journal of Management, 12, 69–82.
Roberson, L. (1990). Prediction of job satisfaction from characteristics of personal work
goals. Journal of Organizational Behavior, 11, 29–41.
Rosenhead, J. (2013). Problem structuring methods. In S. I. Gass, & M. C. Fu (Eds.), En-
cyclopedia of operations research and management science (pp. 1162–1172). Boston,
MA: Springer US.
Saks, A. M., & Ashforth, B. E. (1996). Proactive socialization and behavioral self-
management. Journal of Vocational Behavior, 48, 301–323.
Schwartz, B., Ward, A., Monterosso, J., Lyubomirsky, S., White, K., & Lehman, D. R.
(2002). Maximizing versus satisficing: Happiness is a matter of choice. Journal of
Personality and Social Psychology, 83, 1178–1197.
Schwarzer, R. (1999). The proactive attitude scale (PA scale). http://userpage.fu-
berlin.de/health/proactive.htm.
Schwenk, C. R. (1995). Strategic decision making. Journal of Management, 21, 471–493.
Scott, S. G., & Bruce, R. A. (1995). Decision-making style: The development and assess-
ment of a new measure. Educational and Psychological Measurement, 55, 818–831.
Seibert, S. E., Crant, J. M., & Kraimer, M. L. (1999). Proactive personality and career suc-
cess. Journal of Applied Psychology, 84, 416–427.
Siebert, J., & Keeney, R. L. Creating more and better decisions using objectives. Opera-
tions Research, Forthcoming.
Sloman, S. A. (1996). The empirical case for two systems of reasoning. Psychological
Bulletin, 119, 3–22.
Smith, E. R., & DeCoster, J. (2000). Dual-process models in social and cognitive psychol-
ogy: Conceptual integration and links to underlying memory systems. Personality
and Social Psychology Review, 4, 108–131.
Tavella, E., & Papadopoulos,T. (2014). Expert and novice facilitated modelling: A case of
a viable system model workshop in a local food network. Journal of the Operational
Research Society, 66, 247–264.
Taylor, W. A. (2004). Computer-mediated knowledge sharing and individual user dif-
ferences: An exploratory study. European Journal of Information Systems, 13, 52–64.
Thompson, J. A. (2005). Proactive personality and job performance: A social capital
perspective. The Journal of Applied Psychology, 90, 1011–1017.
VandenBos, G. R. (2007). APA dictionary of psychology. Washington: American Psycho-
logical Association.
Weick, K. E., & Roberts, K. H. (1993). Collective mind in organizations: Heedful interre-
lating on flight decks. Administrative Science Quarterly, 38, 357–381.
Weick, K. E., Sutcliffe, K. M., & Obstfeld, D. (1999). Organizing for high reliability: Pro-
cesses of collective mindfulness. In R. I. Sutton, & B. M. Staw (Eds.), Research in
organizational behavior. An annual series of analytical essays and critical reviews: 21
(pp. 81–123). Greenwich, Conn .: JAI Press.
von Winterfeldt, D., & Edwards, W. (1986). Decision analysis and behavioral research.
Cambridge [Cambridgeshire], New York: Cambridge University Press.
Woolley, R. N., & Pidd, M. (1981). Problem structuring – A literature review. Journal of
the Operational Research Society, 32, 197–206.
Zeleny, M. (1982). Multiple criteria decision making. New York: McGraw-Hill.
1
Electronic Appendix
Item
proactive/
reactive
English item
German item
Scale, Authors, Source
INI_1
r
I usually wait for something to happen rather
than taking the initiative myself.
Ich warte lieber darauf, dass etwas passiert, anstatt selbst die
Initiative zu ergreifen.
Proactive Scale (Schwarzer 1997)
INI_2
p
I love to challenge the status quo.
Ich hinterfrage gerne den Status Quo.
Proactive Personality Scale (Bateman and
Crant 1993)
INI_3
r
I tend to adapt to given circumstances rather
than changing them.
Ich passe mich eher an die gegebenen Umstände an als diese
zu ändern.
Based on Crant (2000), p. 439
INI_4
p
I take initiative to shape my environment.
Ich ergreife die Initiative, um meine Umwelt zu gestalten.
Based on Crant (2000), p. 441
INI_5
r
I do not make decisions unless I really have to.
Ich treffe keine Entscheidungen, außer wenn ich wirklich
muss.
Regret Experience Measure (Creyer and
Ross 1999)
IMP_1
p
I am always looking for better ways to do things.
Ich suche immer nach besseren Wegen, Dinge zu tun.
Proactive Personality Scale (Bateman and
Crant 1993)
IMP_2
p
I am constantly on the lookout for new ways to
improve my life.
Ich bin beständig auf der Suche nach neuen Möglichkeiten,
mein Leben zu verbessern.
Proactive Personality Scale (Bateman and
Crant 1993)
IMP_3
p
I continually try to improve my current situation.
Ich versuche ständig, meine aktuelle Situation zu verbessern.
Based on Crant (2000), p. 436
IMP_4
p
I never settle for the second best solution.
Ich begnüge mich nicht mit der zweitbesten Lösung.
Maximisation Scale (Schwartz et al. 2002)
OBJ_1
p
I try to be clear about my objectives before
choosing.
Ich versuche, mir über meine Ziele klar zu sein, bevor ich
mich entscheide.
Melbourne Decision- Making
Questionnaire (Mann et al. 1997)
OBJ_2
p
In general, I am aware of my objectives in a
decision situation.
In der Regel bin ich mir in Entscheidungssituationen meiner
Ziele bewusst.
Melbourne Decision-Making
Questionnaire (Mann et al. 1997)
OBJ_3
p
For important decisions, I engage in systematic
reflection, what I wish to achieve.
Bei einer wichtigen Entscheidung überlege ich mir
systematisch, was ich erreichen möchte.
Based on Keeney (1992)
OBJ_4
r
Often I do not know what I want in a decision
situation.
Häufig weiß ich in einer Entscheidungssituation nicht, was ich
eigentlich möchte.
Indecisiveness Scale (Frost and Shows
1999)
INF_1
r
I arrived at my answers by assessing only the
information in front of me.
Bei einer Entscheidung begnüge ich mich mit den mir
vorliegenden Informationen.
Situation-Specific Thinking Style (Novak
and Hoffman 2009)
INF_2
p
I seek actively information to improve my
decision making.
Ich suche aktiv nach Informationen, um besser entscheiden zu
können.
Based on Crant (2000), p. 437, Mann et al
(1997), p. 12, Novak and Hoffmann
(2009) p. 60
INF_3
p
I systematically collect the decision-relevant
information.
Ich trage systematisch die entscheidungsrelevanten
Informationen zusammen.
Based on Crant (2000), p. 437, Mann et al
(1997), p. 12, Novak and Hoffmann
(2009) p. 60
INF_4
p
I double check my information sources to be sure
to have the right facts before making decisions.
Bevor ich eine Entscheidung treffe, stelle ich durch eine
gründliche Überprüfung der Informationsquellen sicher, die
richtigen Informationen vorliegen zu haben.
General Decision-Making Style (Scott and
Bruce 1995)
INF_5
r
I often forget or overlook important information
about choice alternatives.
Ich vergesse oder übersehe häufig wichtige Informationen
bezüglich meiner Wahlmöglichkeiten.
Revers item. See above.
ALT_1
p
I excel at identifying opportunities.
Ich bin gut darin, aussichtsreiche Alternativen zu
identifizieren.
Proactive Personality Scale (Bateman and
Crant 1993)
ALT_2
p
I systematically use my objectives to create
Ich nutze meine Ziele systematisch, um Handlungsalternativen
Based on Keeney (1992), p. 4, 202
2
alternatives.
zu entwickeln.
ALT_3
p
I am good at finding ways to achieve my
objectives.
Ich bin gut darin, Wege zu finden, meine Ziele zu erreichen.
Proactive Personality Scale (Bateman and
Crant 1993)
ALT_4
p
I think twice how I can achieve my objectives.
Ich überlege mir gut, wie ich meine Ziele erreichen kann.
Based on Keeney (1992), p. 4, 202
ALT_5
p
Whenever I`m face with a choice, I try to
imagine what all the other possibilities are, even
ones that aren`t present at the moment.
Immer wenn ich mit einer Entscheidung konfrontiert bin,
versuche ich, mir alle möglichen Handlungsalternativen
vorzustellen, auch wenn diese eventuell derzeit nicht zur
Verfügung stehen.
Maximisation Scale, Schwartz et al.
(2002)
RAD_1
p
I thoroughly think about when I make which
decision.
Ich mache mir intensiv Gedanken, welche Entscheidung ich
wann treffe.
Based on Keeney (1992)
RAD_2
p
I spend a lot of time identifying long-range goals
for myself.
Ich verbringe viel Zeit damit, meine langfristigen Ziele zu
identifizieren.
Proactive Scale (Schwarzer 1997)
RAD_3
p
I consider future events in my current decisions.
Ich berücksichtige zukünftige Ereignisse und beziehe sie in
meine derzeitigen Entscheidungen ein.
Based on Grant and Ashford (2008), p. 10;
Nenkov et al. 2008), p. 129
RAD_4
p
I am very aware of my thinking process in a
decision situation.
Ich bin mir meiner Denkprozesse bei Entscheidungen bewusst.
Situation-Specific Thinking Style (Novak
and Hoffman 2009)
RAD_5
p
I thoroughly consider how best to carry out a
decision.
Ich überlege mir gründlich, wie ich Entscheidungen am besten
treffe.
Melbourne Decision-Making
Questionnaire (Mann et al. 1997)
DSA_1
I find the process of deciding frustrating.
Ich finde Entscheidungsprozesse frustrierend.
Decision Satisfaction (Fitzsimons 2000)
DSA_2
I think my choice selections are good.
Ich denke, dass meine Auswahlentscheidungen gut sind.
Decision Satisfaction (Fitzsimons 2000)
DSA_3
I find the process of deciding interesting.
Ich finde Entscheidungsprozesse interessant.
Decision Satisfaction (Fitzsimons 2000)
DSA_4
Normally I am satisfied with the alternatives
ultimately available for my decisions.
Ich bin in der Regel zufrieden mit den letztlich zur Verfügung
stehenden Wahlmöglichkeiten.
Decision Satisfaction (Fitzsimons 2000)
DSA_s
How satisfied are you with the final set of
alternatives among which are to be chosen?
Using a scale from 0 (extremely unsatisfied) to
10 (extremely satisfied):
Wie zufrieden sind Sie mit den letztlich bei einer
Entscheidung zur Wahl stehenden Alternativen? Auf einer
Skala von 0 (extrem unzufrieden) bis 10 (extrem zufrieden):
Decision Satisfaction (Fitzsimons 2000)
DSA_p
How satisfied are you with your decision
processes? Using a scale from 0 (extremely
unsatisfied) to 10 (extremely satisfied):
Wie zufrieden sind Sie mit Ihren Entscheidungsprozessen?
Auf einer Skala von 0 (extrem unzufrieden) bis 10 (extrem
zufrieden):
Decision Satisfaction (Fitzsimons 2000)
DSA_d
How satisfied are you with your decisions?
Using a scale from 0 (extremely unsatisfied) to
10 (extremely satisfied):
Wie zufrieden sind Sie mit Ihren Auswahlentscheidungen?
Auf einer Skala von 0 (extrem unzufrieden) bis 10 (extrem
zufrieden):
Decision Satisfaction (Fitzsimons 2000)
Table A1: Items used in our studies of proactive decision-making. Adaptions of original items are printed in bold. Newly created items are written in italics.
3
CFA Main Study II
(Decision-Maker)
N = 513
CFA Main Study II
(Decision Analyst)
N = 774
CFA Main Study II
(Decision-Maker and
Decision Analyst)
N = 1,572
CFA Main Study II
(neither Decision-Maker
nor Decision Analyst)
N = 448
Factors/
Constructs/
Dimensions
Indi-
cators
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
OBJECTIVES
OBJ_1
.844
.877
.703
.767
.793
.828
.616
.679
.818
.824
.611
.615
.813
.820
.603
.691
OBJ_2
.850
.805
.808
.797
OBJ_3
.821
.755
.714
.717
INFORMA-
TION
INF_2
.784
.851
.655
.767
.728
.784
.549
.679
.673
.795
.565
.697
.698
.824
.611
.661
INF_3
.839
.783
.822
.855
INF_4
.804
.709
.752
.784
ALTERNA-
TIVES
ALT_1
.765
.851
.656
.674
.739
.841
.639
.643
.773
.842
.641
.634
.734
.831
.622
.575
ALT_2
.856
.827
.846
.840
ALT_3
.806
.829
.780
.788
DECISION
RADAR
RAD_1
.767
.846
.579
.714
.697
.816
.526
.642
.702
.820
.533
.697
.729
.824
.539
.691
RAD_3
.707
.672
.692
.695
RAD_4
.780
.770
.763
.774
RAD_5
.786
.757
.761
.737
INITIATIVE
INI_1
.799
.685
.524
.031
.749
.622
.455
.089
.802
.663
.501
.212
.805
.655
.493
.214
INI_3
.640
.590
.599
.582
INI_5
.716
.670
.646
.689
IMPROVE-
MENT
IMP_1
.877
.865
.683
.352
.752
.813
.595
.320
.763
.812
.592
.286
.803
.867
.687
.392
IMP_2
.895
.864
.860
.906
IMP_3
.692
.687
.674
.771
Overall Model Fit
RMSEA=.034, X2/d.f.= 4.716, NFI=.921, TLI=.92, CFI=.936, SRMR=.050 (for detailed results see Table A2).
Table A2: Factor loadings and reliability measures for Decision-Makers and Decision Analysts (main study II).
4
CFA Main Study II
(less than 1 year experience)
N = 631
CFA Main Study II
(between 1 and 5 years
experience)
N = 1,062
CFA Main Study II
(between 5 and 10 years
experience)
N = 576
CFA Main Study II
(more than 10 years
experience)
N = 1,038
Factors/
Constructs/
Dimensions
Indi-
cators
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
OBJECTIVES
OBJ_1
.819
.831
.622
.667
.831
.855
.663
.694
.815
.840
.637
.626
.802
.818
.600
.709
OBJ_2
.799
.844
.826
.790
OBJ_3
.746
.765
.751
.730
INFORMA-
TION
INF_2
.756
.826
.614
.656
.718
.796
.565
.707
.663
.797
.569
.712
.693
.818
.601
.691
INF_3
.838
.800
.816
.847
INF_4
.754
.735
.776
.778
ALTERNA-
TIVES
ALT_1
.786
.857
.666
.582
.764
.844
.644
.650
.744
.854
.662
.689
.75
.832
.623
.618
ALT_2
.835
.84
.87
.846
ALT_3
.826
.801
.822
.768
DECISION
RADAR
RAD_1
.712
.835
.559
.667
.723
.824
.540
.707
.722
.820
.533
.712
.728
.830
.550
.709
RAD_3
.720
.693
.659
.692
RAD_4
.782
.751
.782
.782
RAD_5
.774
.770
.751
.761
INITIATIVE
INI_1
.789
.666
.503
.070
.740
.608
.440
.130
.850
.665
.510
.288
.828
.697
.539
.166
INI_3
.619
.577
.545
.627
INI_5
.686
.663
.574
.705
IMPROVE-
MENT
IMP_1
.842
.863
.678
.406
.765
.833
.625
.346
.758
.807
.588
.277
.801
.829
.620
.282
IMP_2
.879
.860
.901
.871
IMP_3
.744
.741
.614
.679
Overall Model Fit
RMSEA=.034, X2/d.f.= 4.743, NFI=.922, TLI=.921, CFI=.937, SRMR=.052 (for detailed results see Table A4).
Table A3. Factor loadings and reliability measures for different levels of experience (main study II).
5
Groups
N
RMSEA
X2/d.f.
NFI
TLI
CFI
SRMR
R2
Neither decision-maker, nor decision analyst
448
.060
2.609
.915
.933
.946
.049
.300
Decision-maker
513
.066
3.203
.920
.930
.943
.061
.287
Decision analyst
774
.055
3.304
.932
.940
.951
.053
.287
Decision-maker and decision analyst
1,572
.062
7.114
.930
.925
.939
.055
.315
Experience in decision-making less1 year
631
.057
3.037
.931
.942
.952
.049
.281
Experience between 1 and 5 years
1,062
.057
4.481
.937
.939
.950
.049
.339
Experience between 5 and 10 years
576
.065
3.423
.913
.922
.937
.068
.317
Experience more than 10 years
1,038
.066
5.519
.922
.920
.935
.057
.293
Table A4: Goodness-of-fit indices and squared multiple correlations considering PDM and DSA for sub groups (main
study II).
6
CFA Main Study II
(Decision-Maker)
N = 513
CFA Main Study II
(Decision Analyst)
N = 774
CFA Main Study II
(Decision-Maker and
Decision Analyst)
N = 1,572
CFA Main Study II
(neither Decision-Maker
nor Decision Analyst)
N = 448
Factors/
Constructs/
Dimensions
Indicators
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
Stand-
ardised
factor
loading
CR
AVE
MSV
COGNITIVE
SKILLS
OBJECTIVES
.927
.926
.808
.375
.897
.914
.779
.331
.851
.923
.800
.319
.877
.910
.771
.399
INFORMATION
.951
.915
.909
.877
ALTERNATIVES
.860
.850
.864
.840
DECISION
RADAR
.883
.882
.910
.916
INITIATIVE
INI_1
.793
.764
.520
.013
.750
.710
.452
.088
.803
.726
.474
.181
.819
.736
.487
.181
I