Step by Step: Sub-Goals as a Source of Motivation

Abstract

The present research explores the shifting impact of sub-goals on human motivation as individuals move closer to goal attainment, and attributes this shift to the changing source of motivation at different time points during the goal pursuit. In four lab and field experiments, we employed contexts such as exercising, business reviews, and work-for-pay jobs, and performed both within-subject and between-subject tests. We found that when individuals are initiating a goal and derive motivation primarily from the belief that the final goal state is attainable, the structure of sub-goals enhances the sense of attainability and therefore leads to greater motivation. Conversely, when people are completing a goal and the source of motivation centers primarily on the perception that their actions are of value, a focus on the overall goal (rather than sub-goals) heightens the perceived value of the goal-directed actions and leads to greater motivation.

Full text

SUB-GOALS AS A SOURCE OF MOTIVATION
Step by Step: Sub-Goals as a Source of Motivation
Szu-chi Huang
Liyin Jin
Ying Zhang
Forthcoming in Organizational Behavior and Human Decision Processes

SUB-GOALS AS A SOURCE OF MOTIVATION 2
Step by Step: Sub-Goals as a Source of Motivation
The present research explores the shifting impact of sub-goals on human
motivation as individuals move closer to goal attainment, and attributes this shift to the
changing source of motivation at different time points during the goal pursuit. In four lab and
field experiments, we employed contexts such as exercising, business reviews, and work-for-
pay jobs, and performed both within-subject and between-subject tests. We found that when
individuals are initiating a goal and derive motivation primarily from the belief that the final
goal state is attainable, the structure of sub-goals enhances the sense of attainability and
therefore leads to greater motivation. Conversely, when people are completing a goal and the
source of motivation centers primarily on the perception that their actions are of value, a
focus on the overall goal (rather than sub-goals) heightens the perceived value of the goal-
directed actions and leads to greater motivation.
Keywords: Sub-goal, goal structure, motivation, self-regulation, expectancy, value

SUB-GOALS AS A SOURCE OF MOTIVATION 3
An employee at a call center who aims to make 3,000 sales calls per month can frame
his/her goal in two ways: as one integrated goal of 3,000 sales calls or as an accumulation of
smaller, more manageable sub-goals, such as thirty sub-goals of 100 calls each. Setting sub-
goals thus creates an elaborated goal structure, delineating a set of successive approximations
and steps toward the achievement of the overall goal (Lewin, 1936; Murray, 1938;
Kruglanski et al., 2002). While some business practices emphasize the leaders’ role in
identifying and formalizing one overall goal for their employees (e.g., executives at Aetna
Inc. set and focus on a few annual companywide goals; Pratt, 2007), other industry guidelines
promote the idea of setting smaller (e.g., quarterly) sub-goals (Financial Planning, 2016;
Wilson, 2016).
Structuring the pursuit of an overall goal into a set of sub-goals has been shown to
reduce the difficulty of the pursuit and to provide positive reinforcements that lead to greater
motivation and persistence (Brunstein, 1993; Locke & Latham, 1990; Soman & Shi, 2003).
For example, the aforementioned employee at the call center might be more motivated to
work on the sales goal if it is divided into thirty sub-goals because the completion of 100
calls seems more easily achievable and motivating than that of 3,000 calls, which seems
excessively difficult and, hence, discourages goal engagement (Locke & Latham, 1990;
Pervin, 1989; Soman & Shi, 2003).
However, empirical evidence also suggests that focusing on sub-goals can conversely
interfere with the pursuit of the ultimate goal (Amar et al., 2011; Amir & Ariely, 2008;
Fishbach, Dhar, & Zhang, 2006; Newell & Simon, 1972). The achievement of sub-goals
could breed a sense of self-congratulation and encourage relaxation (e.g., taking a long lunch
break), thereby interfering with the progression toward and the attainment of the overall goal
(Fishbach, Dhar, & Zhang, 2006). Similarly, Amir and Ariely (2008) found that providing

SUB-GOALS AS A SOURCE OF MOTIVATION 4
discrete progress markers such as sub-goals hindered people’s performance in a spelling bee
when the task was already rich in progress information.
The diverging evidence on the effectiveness of sub-goals, coupled with mixed
business guidelines and principles, highlights the necessity of a closer examination of the
variables that determine the motivational consequences of this elaborated goal structure, and
this is precisely what we hope to achieve in the present research. In this research, we define
motivation as individuals’ tendencies to carry out goal-directed actions in order to reduce the
discrepancy between the current state and the ideal state (Carver & Scheier, 1990). We adopt
a longitudinal and dynamic view of the effects of sub-goals and aim to determine how,
compared with an exclusive focus on the overall goal, such an elaborated intermediate goal
structure influences motivation over the course of a pursuit. Drawing from the literature on
the various sources of human motivation (Liberman & Förster, 2008; Locke & Latham, 1990)
and the temporal variation of their impacts (Huang, Zhang, & Broniarczyk, 2012; Louro,
Pieters, & Zeelenberg, 2007), we propose the following hypothesis: Because individuals
derive motivation to persist on a goal from different sources as they progress toward the end
point (Liberman & Förster, 2008; Louro, Pieters, & Zeelenberg, 2007), framing the goal as
one integrated goal versus an accumulation of sub-goals may have a distinct influence on
motivation at different times.
Specifically, we build on the influential value-expectancy models (Atkinson, 1957;
Vroom, 1964) and propose that these two pillars of motivation have an interesting temporal
aspect to them: When people have accumulated only a low level of progress on the goal and
remain doubtful about the goal’s attainability, the information that signals the goal’s
attainability should be the primary determinant of their motivation (Zhang & Huang, 2010).
For example, researchers have recently found that having more variety within a set of means
to goal attainment increases motivation in the initial stage of the pursuit by reducing the

SUB-GOALS AS A SOURCE OF MOTIVATION 5
uncertainty associated with goal attainment (Etkin & Ratner, 2012). Because a sub-goal
structure fosters the sense of goal attainability (i.e., easier goal attainment) more than the
structure that has only one overall goal, we thus argue that a focus on the sub-goal should
elicit greater motivation when people first begin the pursuit, because in this early stage the
source of motivation lies critically in the belief of a goal’s attainability.
However, when people move into the advanced stages of the pursuit, the high level of
progress they have accumulated should alleviate the concern on whether the goal is attainable
(Liberman & Förster, 2008). At this stage, people instead focus on the reduction of the
discrepancy between their current position and the goal (Koo & Fishbach, 2008). Their
commitment to the goal and subsequent motivation therefore depend primarily on the extent
to which they value the goal (Koo & Fishbach, 2012). Importantly, because a focus on the
overall goal (instead of the next sub-goal) allows individuals to see their actions as directly
linked to the valuable outcome, we propose that focusing on the overall goal would elicit
greater motivation when people’s concern centers on value. Overall, depending on whether
people derive motivation from the perception of easy goal attainment or from the sense that
their actions are associated with a valuable outcome, the motivational consequences of a sub-
goal structure would change as people progress further in their pursuits.
This conceptualization reconciles conflicting findings in the sub-goal literature by
identifying the conditions under which sub-goals’ momentary impact on motivation shifts as
individuals move from the beginning of the pursuit to goal completion. Our critical
contribution lies in the finding that the motivational consequences of a sub-goal structure rely
heavily on individuals’ shifting concerns about the pursuit and in the delineation of how the
structure of sub-goals (vs. having only the overall goal) addresses these concerns. Our
findings suggest that a general statement on the effectiveness of sub-goals may be an
oversimplification; organizations and employers who wish to implement a sub-goal structure

SUB-GOALS AS A SOURCE OF MOTIVATION 6
to motivate employees, sales teams, or consumers should hence be mindful of this shifting
impact.
The Advantages and Disadvantages of Sub-Goals
The literature defines sub-goals as pre-established smaller steps toward the
achievement of an overarching goal (Borrelli & Mermelstein, 1994; Heath, Larrick, & Wu,
1999; Lewin, 1936). Because sub-goals are successive approximations toward an overall goal
(Murray, 1938), they are not ends in themselves. Instead, they exist only because of primary
goals (Kruglanski et al., 2002).
The use of sub-goals is associated with many benefits. Because sub-goals are
subordinate end points in the pursuit of an overall goal, they help to signify progress toward
the ultimate end goal, especially when the overall progress is uncertain (Amir & Ariely,
2008). In addition, sub-goals are easier and quicker to accomplish than the overall goal,
reducing the difficulty and complexity of the pursuit and providing a greater sense of
progress (Brunstein, 1993; Locke & Latham, 1990; Newell & Simon, 1972; Pervin, 1989;
Soman & Shi, 2003). As a result, the employment of sub-goals can help solve the “starting
problem” that arises when one confronts a difficult goal (Heath, Larrick, & Wu, 1999). The
achievement of sub-goals can enhance self-efficacy and competence, leading to greater
persistence and motivation (Bandura & Schunk, 1981; Stock & Cervone, 1990). In the
context of debt settlement, Gal and McShane (2012) found that as individuals paid off more
debt accounts (i.e., the more financial sub-goals they accomplished), their subsequent effort
in eliminating their overall debt increased. The actual dollar amount that was paid off did not
have such a motivational effect.
On the other hand, there are also costs associated with setting and accomplishing sub-
goals. Because sub-goals represent additional intermediate levels that individuals must work
toward, they may lead to motivational distraction and interfere with the ultimate goal (Heath

SUB-GOALS AS A SOURCE OF MOTIVATION 7
et al., 1999; Kruglanski et al., 2002; Newell & Simon, 1972). Sub-goals often replace the
overarching goal as the center of reference (Heath et al., 1999), and the sense of
accomplishment from completing individual sub-goals can cause complacency, leading to
lower motivation to continue working on the overall goal. For instance, Fishbach, Dhar, and
Zhang (2006) showed that when people considered their success on a subgoal, they would
view additional actions toward achieving the superordinate goal as substitutes and thus were
less likely to pursue these actions.
These conflicting findings suggest that sub-goals may not have either a universally
positive or negative impact on motivation, and the effectiveness of sub-goals calls for closer
examination. While various factors could change the impact of sub-goals (e.g., trait
procrastination, sub-goal alignment, expertise in goal pursuit), we are particularly interested
in the level of progress on the goal as the focal point of investigation for the following three
reasons: First, goal pursuit is a dynamic process that spans from initiation to completion, and
situations change from moment to moment during this process. For this reason, a longitudinal
perspective reveals much more information than the usual snapshot-like approach, as it
accounts for the influence of time/stage. Second, prior research has documented that
individuals actively monitor their progress (Carver & Scheier, 1998) and adjust efforts
accordingly (Kivetz, Urminsky, & Zheng, 2006; Nunes & Drèze, 2006). However, what these
findings did not address is how individuals’ relative position on a goal changes not only the
amount of effort they invest, but also the source of their effort and thus the way they interpret
goal-directed behaviors (Koo & Fishbach, 2012), both of which determine the impact of the
presence of a sub-goal structure. Third, in organizational settings, the goal structure often
remains static throughout the pursuit; for instance, once a sub-goal structure is employed in a
sales context, it would continue to be used throughout the fiscal year. It is thus especially
important to explore the dynamic impact of sub-goal structure across different stages of goal

SUB-GOALS AS A SOURCE OF MOTIVATION 8
pursuit to derive a goal structure and feedback system that maximizes individuals’ effort and
performance.
Sub-Goals as the Source of Motivation
On the conceptual level, goals function as reference points (Bonezzi, Brendl, & De
Angelis, 2011; Heath et al., 1999) and motivate people by creating a negative discrepancy
between a person’s desired state and his or her current state. What distinguishes goals with an
elaborated sub-goal structure from those without is that the latter focus people on a single
reference point whereas the former present two (or more) distinct and simultaneous reference
points. For example, an employee hoping to complete 3,000 sales calls might have only this
number in mind as the reference point. However, if the employee divides the goal into
consecutive 100-call sub-goals, he or she might simultaneously hold multiple numbers—100,
200, 300,…3,000—in mind, with both the most proximal sub-goal and the ultimate overall
goal as relevant reference points (Heath et al., 1999).
One critical characteristic of pursuing goals with a sub-goal structure is that when
both an overall goal and an immediate sub-goal are present, the immediate next sub-goal
functions as the primary reference and, therefore, the basis for motivation. Bandura and
Schunk (1981) found that a proximal reference point leads to greater motivation and
performance because it provides an immediate and achievable benchmark, whereas a distal
goal is ineffective in mobilizing or directing effort. Similarly, Hull’s (1932) maze-learning
experiments revealed that actions become progressively weaker as one moves further away
from a goal point. Therefore, when both reference points are present, people are likely to
focus on the proximal rather than the more distant reference point, anchoring on the
immediate sub-goal when deriving motivation for the pursuit.1
1 A pilot study in our lab confirmed that people focused on the sub-goal when this proximal reference
point was present. Participants (n = 201, 127 females) reviewed 10 restaurants for a $5 voucher, and

SUB-GOALS AS A SOURCE OF MOTIVATION 9
How, then, does an elaborated sub-goal structure function as the source of
motivation? Classic theories on motivation have largely focused on two primary sources of
motivation—goal expectancy and goal value—and have found that both variables contribute
to people’s commitment to a pursuit (Atkinson, 1957; Liberman & Förster, 2008; Locke &
Latham, 1990; Mischel, Cantor, & Feldman, 1996; Tolman, 1955; Vroom, 1964). Recent
findings suggest that these two sources may not play equal roles throughout the pursuit and
that peoples’ main concerns could change from one aspect to the other depending on the
stage of the pursuit they are in (Huang, Zhang, & Broniarczyk, 2012; Koo & Fishbach, 2012;
Louro, Pieters, & Zeelenberg, 2007). Specifically, when people first initiate the pursuit and
have only accumulated a low level of progress, their commitment to the goal depends heavily
on the belief that the goal is indeed attainable; people seek confirmation of the goal’s
attainability before investing further effort into the pursuit (Etkin & Ratner, 2012). Once the
uncertainty about the goal’s attainability abates, such as when people have made significant
progress on the goal and are approaching the end of the pursuit (Louro, Pieters, &
Zeelenberg, 2007), people shift their focus from the goal’s attainability to the reduction of the
remaining discrepancy between their current position and the final goal (Koo & Fishbach,
2008; 2012). Hence, in this advanced stage of goal pursuit, the source of motivation centers
on whether this final goal is worth the continued effort, that is, the value of the goal (Zhang &
Huang, 2010).
The potential impact of a sub-goal structure is particularly intriguing when viewed in
this context because it holds answers to both concerns, helping to address the question of
whether the goal is attainable, as well as whether it is worth pursuing. Consequently, whether
they received feedback based on either five sub-goals of two reviews each or one overall goal of 10
reviews. Both participants’ visual focus (captured by a heat-map measure) and self-reported focus
(captured by a continuous seven-point scale) showed a significant main effect of goal structure,
(F(1,197) = 68.02, p < .001, ηp2 = .26 and F(1,197) = 18.57, p < .001, ηp2 = .09, respectively) such that
people focused more on the next sub-goal than on the overall goal when sub-goals were present.

SUB-GOALS AS A SOURCE OF MOTIVATION 10
individuals focus on an immediate sub-goal or the overall goal should have important
implications for their answers to these questions and thus their subsequent motivation.
On the one hand, sub-goals represent smaller and more manageable steps forward
than the overall goal; thus focusing on these incremental steps leads to the perception that the
pursuit is easier and less complex than focusing on the overall goal (Brunstein, 1993; Soman
& Shi, 2003). This perception can be particularly useful when individuals remain doubtful
about the goal’s attainability and derive motivation primarily from the sense that they can
indeed attain the goal, at which time a focus on the next sub-goal (vs. the overall goal) should
lead to enhanced motivation (Heath et al., 1999; Sutton, 2010; Weick, 1984). This situation is
most likely to occur when people first initiate the pursuit and have made only a low level of
progress, because the substantial distance until completion casts doubts on the belief that the
goal is indeed achievable (Huang et al., 2012; Louro et al., 2007). For example, for a person
who has recently begun to repay a $40,000 student loan, a focus on the ultimate goal may
seem far and intimidating and may thus demotivate efforts. By contrast, a set of more budget-
friendly sub-goals (e.g., setting aside $400 in the next month) may seem more manageable
and make the overall target seem less out of reach, increasing people’s likelihood of engaging
in the goal-congruent behaviors.
On the other hand, although a focus on the more manageable sub-goals strengthens
the sense of attainability and enhances motivation early on, this perception ceases to be
instrumental when motivation is less dependent on the belief that the goal is attainable
(Garland, 1983; Liberman & Förster, 2008), such as when people have accumulated
sufficient progress and stop questioning whether the goal is a feasible target. For example,
when the debt-paying student in our previous example has passed the $35,000 mark, the goal
of $40,000 seems within reasonable reach and the concern about attainability dwindles. In
these situations, individuals’ tendencies to perform further goal-directed actions shift to the

SUB-GOALS AS A SOURCE OF MOTIVATION 11
second pillar of motivation, the perceived value of their actions. That is, people base their
tendency to act on the question of “Is this really worth doing?” (Bonezzi et al., 2011; Carver
& Scheier, 1998; Fishbach, Henderson, & Koo, 2011). Sub-goals, as the proximal and
primary point of reference, shape the answer to this question.
Goal-directed actions are valueless in themselves, and only become valuable by being
instrumental to the outcome that they help to achieve (Kruglanski et al., 2002). The value of
goal-directed actions should, in turn, be evaluated in the context of the outcome they serve
(Kruglanski, 1996; Kruglanski et al., 2002). When both reference points are present, any
goal-directed action may be represented in the context either of helping to achieve the next
sub-goal or of helping to attain the overall goal. By definition, sub-goals represent only small
steps toward the attainment of an overall goal and are associated with only a subset (if any at
all) of the benefits from the overall attainment. Therefore, whereas evaluating goal-directed
actions in both contexts would afford them value, doing so in the context of the overall
attainment would make the action seem more valuable than in the context of the sub-goal,
even though the actions and their effectiveness in helping to reach the end point remain
objectively identical.
For the student in our previous example, deciding whether to skip an expensive dinner
in order to save may be evaluated in the context of contributing to the monthly $400 target or
the context of helping to achieve the debt-free status. While the $400 is well aligned with the
overall goal of being debt free, representing the action as contributing to this sub-goal seems
less valuable and worthwhile than construing it as helping one achieve the debt-free status.
Therefore, when the motivation to act depends primarily on the perceived value of goal-
directed actions (Bonezzi et al., 2011; Levin, Schneider, & Gaeth, 1998), the focus on the
overall goal—rather than sub-goals—would yield a greater perceived value of these actions
and, thus, greater likelihood to act.

SUB-GOALS AS A SOURCE OF MOTIVATION 12
Based on the above reasoning, we propose the following formal hypotheses:
H1: The impact of a sub-goal structure on motivation depends on the stage of the
pursuit. The sub-goal structure (vs. an overall-goal–only structure) would lead to
greater motivation in the early stages of goal pursuit but lower motivation in the
advanced stages of goal pursuit.
H2: The positive impact of a sub-goal structure (vs. an overall-goal–only structure) on
motivation is mediated by perceived goal attainability; this pathway dominates in the
early (vs. advanced) stage of goal pursuit, because the goal’s attainability is the key
determinant of motivation in this stage.
H3: The negative impact of a sub-goal structure (vs. an overall-goal–only structure)
on motivation is mediated by the perceived value of goal-directed actions; this
pathway dominates in the advanced (vs. early) stage of goal pursuit, because the value
of goal-directed actions serves as the key determinant of motivation in this stage.
Three points merit further clarification. First, we focus on situations in which the
effort–performance relationship is strong: individuals know that by increasing their effort,
they will increase their chance of accomplishing the goal. This is important because when the
effort–performance relationship is weak or uncertain, focusing on the sub-goal may not
sufficiently enhance perceived goal attainability, which would inhibit the positive effect of a
sub-goal structure on effort investment in the early stage of goal pursuit.
Second, it is important to note that our theory does not predict an overall positive or
negative net effect of sub-goals on motivation. Instead, what we aim to explore is the relative
impact that focusing on the sub-goal (vs. the overall goal) has on individuals’ momentary
motivation at different points during the course of goal pursuit. By documenting the shifting
impact of sub-goals across different stages as well as the psychological mechanisms

SUB-GOALS AS A SOURCE OF MOTIVATION 13
underlying these effects, we can design better goal structures to help individuals stay
motivated.
Third, while previous literature emphasizes that individuals derive motivation from
the value of the ultimate goal (e.g., Liberman & Förster, 2008; Shah, Higgins, & Friedman,
1998), our theorizing fine-tunes existing theory by suggesting that what underlies the
motivation for goal-directed actions may be more than the value of the end goal itself; rather,
the perceived value of immediate actions may play an important role. The value of present
actions and the value of the ultimate goal are often highly correlated because the latter is the
key determinant of the former. However, at times when there is a separation between
individual actions and the ultimate outcome, such as when people focus on sub-goals and
associate actions with the sub-goals rather than with the overall goal, the distinction between
these two types of value is important. In these cases, the perceived value of the actions
relative to the salient reference point, rather than the value of the ultimate goal, determines
individuals’ tendency to carry out these actions and to work toward the goal. This is a critical
distinction because it emphasizes action-based value evaluation as an important determinant
of motivation.
In the next section, we report our empirical tests of the hypotheses. Study 1
demonstrated the motivational impact of sub-goals by testing our first hypothesis at two
extreme points of the pursuit—the beginning and the end—through a physical exercise in the
lab. Study 2 replicated the motivational effects in Study 1, directly assessed the underlying
psychological mechanisms at these two time points (H2 and H3), and tested them through the
path model. Study 3 captured the impact of sub-goal structure on people’s motivation and the
underlying psychological mechanisms (H1, H2, and H3) in a continuous manner using
repeated measures within-subject to document the natural change in the source of motivation.
Finally, based on our empirical findings in the first three studies, we tested a theory-driven

SUB-GOALS AS A SOURCE OF MOTIVATION 14
hybrid structure that set sub-goals in the initial stage and then removed them in the advanced
stage in an eight-day field experiment.
Study 1: Stepping Routine
In Study 1, we used an exercise task in which participants repeated a set of
“stepping routines” to burn calories. The exercise task was framed either as a holistic,
continuous task of burning 200 calories (overall goal condition) or as a task composed of four
sub-sessions that each required the participant to burn 50 calories (sub-goal condition). We
assessed the participants’ motivation by measuring their exercise intensity after they
accumulated either a low level of progress or a high level of progress (H1).
Method
We aimed to recruit 150 participants. A total of 134 undergraduate students (78
females, 56 males; average age = 20.78) from a public university completed the study and
constituted our final sample. The experiment used a 2 (goal structure: overall goal vs. sub-
goal) × 2 (progress level: 1/4 vs. 3/4 of the task) between-subject design.
The participants completed the study in individual experiment rooms. Each room
included one large monitor with a 30" × 30" square stepping pad in front of it. A laptop was
placed in the far corner of each room, and the participants entered their personal information
and received feedback on the computer. We also set up a camera in the back of the room to
record the experiment session and maintained the room temperature at 75 degrees Fahrenheit.
The cover story informed the participants that we were examining how the human
body functions by testing the effectiveness of a set of “stepping routines” to help burn
calories. The participants were informed that their goal for the session would be to burn 200
calories, and we emphasized that it was important that they reach this number because only
complete data would allow us to fully test the effectiveness of the exercise routine. Before
commencing the task, the participants were asked to measure their pulse rate and record this

SUB-GOALS AS A SOURCE OF MOTIVATION 15
information on the laptop along with other basic information such as gender, weight, age, and
exercise frequency.
In the overall goal condition, the participants were informed that there would be one
exercise session in which they would need to burn 200 calories. By contrast, the participants
in the sub-goal condition were informed that the exercise session consisted of four sub-
sessions (Stages 1–4) and that each set of 50 calories burned would complete one sub-
session, resulting in a total of 200 calories.
The participants then commenced the exercise session and followed the simple
sequences shown in the instructive video on the screen to move their feet to different squares
on the stepping pad. They were instructed to closely follow the rhythm of stepping in the
video to ensure that they did not miss a step. We allowed the participants to exercise for
either five minutes to accumulate a low (1/4) level of progress or 15 minutes to accumulate a
high (3/4) level of progress; the difference in the amount of time they were allotted to
exercise led to different levels of accumulated progress, mimicking real-life goal pursuit
situations. In addition, by fixing the frequency (i.e., rhythm) and the duration of exercising
within each progress level, we ensured that the participants invested roughly the same amount
of effort in the sub-goal and overall goal conditions (thereby holding the progress level
constant).
After the participants exercised for either 5 or 15 minutes, the video paused, and
they were asked to measure their pulse rate and report it on the laptop. After entering this
figure, the participants were told to wait while the computer processed the information.
Following a “Calculating …” page, the laptop provided the participants with feedback. In the
sub-goal condition, feedback was given on four horizontal progress bars that were arranged
parallel to one another and labeled Stages 1 through 4. By contrast, in the overall goal
condition, the feedback was given on a single long horizontal progress bar. The participants

SUB-GOALS AS A SOURCE OF MOTIVATION 16
in the 1/4 progress condition observed either that they had burned 50 calories and completed
Stage 1 (sub-goal condition) or that they had burned 50 calories toward the 200 calories
needed on the single progress bar (overall goal condition). In the 3/4 progress condition, the
feedback indicated that they had burned 50 calories for Stage 1, 50 calories for Stage 2, and
50 calories for Stage 3, thereby completing three sub-goals (sub-goal condition), or that they
had burned 150 of the 200 calories needed to complete the entire goal on the long progress
bar (overall goal condition).
After receiving the feedback, all participants were informed that they should resume
exercising. The instructions informed them that they would need to perform a slightly
different stepping routine for the remainder of the session. The monitor demonstrated a set of
simple stepping patterns and asked the participants to learn these steps and repeat them.
There was no fixed rhythm to follow, and the participants were informed that they should
repeat the patterns as quickly as possible to burn more calories. The participants began
exercising and were stopped by an on-screen instruction after five minutes. We measured the
intensity of stepping (i.e., the total number of steps during the five-minute period) as the
proxy for their motivation to burn more calories after receiving progress feedback. After the
exercise session, the participants measured and recorded their pulse rate one more time before
exiting the room for a full debriefing at the checkout desk.
Results and Discussion
Our main interest was participants’ motivation to burn more calories after receiving
feedback. We captured this variable by measuring their total number of steps during the five-
minute period. An ANOVA of this measure yielded a main effect of progress level (F(1, 130)
= 7.43, p < .01, ηp2 = .05) and, more importantly, a goal structure × progress level interaction
(F(1, 130) = 14.80, p < .01, ηp2 = .10). Further contrast analyses showed that when the
participants had completed 1/4 of the task, those who pursued sub-goals repeated the stepping

SUB-GOALS AS A SOURCE OF MOTIVATION 17
routine faster (M = 221.24 steps, SD = 28.34) than those who pursued the overall goal (M =
200.38 steps, SD = 31.08; t(66) = 2.89, p < .01). However, when the participants had
achieved 3/4 of the task, we observed the opposite pattern, i.e., those who pursued an overall
goal repeated the stepping routine faster (M = 233.03, SD = 28.17) than those who pursued
sub-goals (M = 215.67, SD = 27.20; t(64) = 2.55, p < .05; see Figure 1). Table 1 provides a
summary of the results of all of the studies.
Insert figure 1 about here
This study provided initial evidence for our first hypothesis, that while a sub-goal
structure leads to greater motivation in the early stages of goal pursuit, a focus on the overall
goal can be more motivating in the advanced stages. We hypothesize that this change in the
motivational impact of sub-goals occurs because of the shift of the source of motivation
during the course of goal pursuit—from the goal’s attainability (H2) to the value of goal-
directed actions (H3). We directly measured these underlying psychological mechanisms and
submitted the key variables to moderated mediation analyses in the next study.
Study 2: The Psychological Mechanisms
Participants in this experiment accumulated reward points by sharing their dining
experiences and completed the task in the context of a scoring system that involved either a
sub-goal structure or an overall goal structure. We asked participants to report their perceived
goal attainability and the perceived value of the next goal-directed action when they
completed either 30% or 70% of the task and captured their motivation at that time by
recording the number of words they shared to earn more points in the program.
Method
We aimed to recruit 150 participants. A total of 158 undergraduate students (84
females, 74 males; average age = 20.20) completed the study and constituted our final sample

SUB-GOALS AS A SOURCE OF MOTIVATION 18
(the last group of participants showed up at the same time, so we allowed all of them to
participate). The experiment used a 2 (goal structure: overall goal vs. sub-goal) × 2 (progress
level: 30% vs. 70%) between-subject design.
The cover story informed participants that this study was conducted in collaboration
with a social media website and that the researchers were interested in understanding how
individuals share their dining experiences with others through online review services. All
participants were informed that their task was to log in to a soon-to-be-launched website and
share their dining experiences at different restaurants, as they would do on sites such as Yelp
or Google. Similar to many real-world practices (e.g., earning reward points by writing
product reviews), we offered participants rewards for sharing their opinions and incentivized
them with a $10 restaurant voucher if they earned 100 points by the end of the task.
In the overall goal condition, the participants were informed that there would be one
session in which they would need to earn 100 points. By contrast, the participants in the sub-
goal condition were informed that the task consisted of five sections (1–5) and that they
needed to earn 20 points in each section by sharing dining experiences, totaling 100 points.
Depending on the condition, the feedback page appeared after the participants
earned either 30 points (30% progress) or 70 points (70% progress). In the 30% progress
condition, the feedback was displayed as 30 of 100 points for the overall goal condition or as
20 points in Section 1 and 10 points in Section 2 for the sub-goal condition. In the 70%
progress condition, the feedback was displayed as 70 of 100 for the overall goal condition or
as 20 points in Sections 1, 2, and 3 and 10 points in Section 4 for the sub-goal condition.
After providing progress feedback, we informed the participants that we would like to
ask about their experiences with the website before they continued the task. Among the filler
questions—such as questions about food preferences—we asked participants to report their
perceived goal attainability (“How likely do you think you are to complete the entire task for

SUB-GOALS AS A SOURCE OF MOTIVATION 19
the voucher?” [1 = Very unlikely, 9 = Very likely]) and their perceived value of the next
goal-directed action (“How much value do you see in sharing the next piece of dining
experience?” [1 = No value at all, 9 = Extremely high value]).
After the participants answered these questions, they resumed the main task. We
informed all participants that they could earn more points by sharing extensive information
regarding a dining experience; the more detailed the information they shared, the more points
they would earn. We measured how many words the participants typed to describe their
dining experience as the indicator of motivation. All participants were debriefed after they
completed the task and were entered into a lottery.
Results and Discussion
Motivation. We first performed an ANOVA on the number of words that participants
wrote to describe their dining experience after receiving the progress feedback. The analysis
yielded the hypothesized goal structure × progress level interaction (F(1, 154) = 11.39, p <
.01, ηp2 = .07); there were no main effects. When the participants completed only 30% of the
task, those who pursued a series of sub-goals wrote more words (M = 207.24 words, SD =
101.65) than those who pursued an overall goal (M = 156.73 words, SD = 63.56; t(77) = –
2.61, p = .01). However, we observed the opposite pattern after the participants completed
70% of the task. Specifically, those who pursued an overall goal wrote more words (M =
225.38 words, SD = 125.38) than those who pursued sub-goals (M = 175.88 words, SD =
65.39; t(77) = 2.19, p < .05; see Figure 2).2
2 We also analyzed actual effort expenditure (i.e., the number of words participants typed) prior to
receiving progress feedback to ensure that there was no significant difference in their internal effort
perception. The analysis revealed only a main effect of progress level (F(1,154) = 183.47, p < .001,
ηp2 = .54). Neither the main effect of goal structure (F(1,154) = 0.36, p = .55, ηp2 = .002), nor goal
structure × progress level interaction was significant (F(1,154) = 0.33, p = .57, ηp2 = .002).
Participants in the sub-goal condition did not invest greater effort than those in the overall goal
condition before receiving the 30% progress feedback (Msubgoal = 328.17, SD = 137.21 vs. Moverall goal =
329.43, SD = 119.23; t(77) = 0.04, ns), or before they received the 70% progress feedback (Msubgoal =
957.48, SD = 343.35 vs. Moverall goal = 1014.77, SD = 470.57; t(77) = 0.62, ns).

SUB-GOALS AS A SOURCE OF MOTIVATION 20
Insert figure 2 about here
Goal attainability and value of actions. We performed the same ANOVAs on the
two proposed mechanisms. The ANOVA on the perceived goal attainability yielded a main
effect of progress level (F(1,154) = 4.62, p < .05, ηp2 = .03), a main effect of goal structure
(F(1, 154) = 4.40, p < .05, ηp2 = .03), and a goal structure × progress level interaction (F(1,
154) = 5.64, p < .05, ηp2 = .04). Subsequent contrast analyses revealed that when the
participants had only completed 30% of the task, those who pursued a series of sub-goals
believed that they were more likely to attain the goal (M = 5.26, SD = 2.69) than those
pursuing an overall goal (M = 3.54, SD = 2.22; t(77) = –3.08, p < .01). However, after the
participants had completed 70% of the task, there was no significant difference in perceived
goal attainability between those pursuing sub-goals (M = 5.18, SD = 2.53) and those pursuing
an overall goal (M = 5.28, SD = 2.15; t(77) = .20, ns).
The ANOVA on the perceived value of goal-directed actions yielded a main effect of
progress level (F(1,154) = 9.49, p < .01, ηp2 = .06), a main effect of goal structure (F(1, 154)
= 12.04, p < .01, ηp2 = .07), and a goal structure × progress level interaction (F(1, 154) =
3.96, p < .05, ηp2 = .03). Further contrast analyses revealed that there was no difference in the
perceived value of actions between participants pursuing sub-goals and those pursuing an
overall goal when the participants had only completed 30% of the task and the end point was
still far (Msub-goal = 4.43, SD = 1.67 vs. Moverall goal = 4.86, SD = 1.80; t(77) = 1.12, ns).
However, after the participants had completed 70% of the task, those pursuing an overall goal
perceived their actions to be more valuable (M = 6.36, SD = 1.98) than those pursuing a
series of sub-goals (M = 4.75, SD = 1.95; t(77) = 3.64, p < .01).
From goal attainability and value of actions to motivation. To test the proposed
underlying mechanisms of perceived goal attainablity and perceived value of goal-directed
actions, we conducted a bias-corrected moderated mediation analysis with both variables

SUB-GOALS AS A SOURCE OF MOTIVATION 21
entered as simultaneous mediators (model 14, bootstrapping sample size = 5,000; Hayes,
2013; Preacher & Hayes, 2008). In this moderated mediation model, goal structure predicted
the perceived goal attainablity and the value of goal-directed action, and progress level
moderated the effects of these two proposed mechanisms on motivation.
The results supported our predictions (see Figure 3). A sub-goal structure enhanced
the perceived goal attainability (β = 0.79, t(158) = 1.99, p < .05); a sub-goal structure also
reduced the perceived value of the next goal-directed action (β = -1.05, t(158) = -3.40, p <
.001). However, which of these two mechanisms served as the dominant predictor of
participants’ motivation depended on their progress level, as shown by two significant
mechanism × progress level interactions (perceived attainability × progress level interaction:
β = -18.83, t(158) = -4.05, p < .001; perceived value of the next action × progress level
interaction: β = 25.20, t(158) = 4.20, p < .001). The conditional indirect effects showed that
goal attainability was the dominant predictor of motivation in the early stage (early stage: b =
17.03, 95% CI [1.34 to 35.35]; advanced stage: b = 2.24, 95% CI [-0.90 to 10.73]), whereas
the perceived value of the next goal-directed action served as the dominant predictor of
motivation in the advanced stage of goal pursuit (advanced stage: b = -35.82, 95% CI [-62.74
to -14.37]; early stage: b = -9.45, 95% CI [-26.96 to -1.13]). As people accumulated greater
progress in their pursuits, the driving force of their motivation shifted from goal attainabiltiy
to the value of their actions, and this change in the source of motivation hence determined the
impact (positive or negative) of sub-goal structure on their subsequent effort.
Insert figure 3 about here
Study 2 replicated the findings in Study 1 that the impact of sub-goals on motivation
indeed depended on the stage of the pursuit. In addition, this study provided direct evidence
for the underlying mechanisms (H2 and H3). A focus on the next sub-goal (vs. the overall

SUB-GOALS AS A SOURCE OF MOTIVATION 22
goal) led to the perception of greater goal attainability, which led to greater motivation early
in the pursuit when the goal’s attainability served as the source of motivation. Conversely, a
focus on the overall goal (vs. the next sub-goal) led to a higher perceived value of these goal-
directed actions, resulting in greater motivation when people were approaching the end and
were motivated by the sense that the next action that they performed toward the goal was
valuable and meaningful.
In the first two studies, we examined the dynamic impact of goal structure on
motivation at two time points of goal pursuit in a between-subject manner (1/4 vs. 3/4 or 30%
vs. 70% of the task). This between-subject paradigm helped to ensure that within each
progress level, the participants in the sub-goal and overall goal conditions experienced the
same amount of progress as well as similar senses of accomplishment and depletion
(participants in both the sub-goal and overall goal conditions were placed directly at 1/4, 3/4,
30%, or 70% of the task). Although this approach ensured the integrity of our experimental
manipulations, we were curious whether the same pattern would emerge for a natural,
continuous goal pursuit process. Therefore, we conducted the next study to test our three
hypotheses in continuous goal pursuit and used repeated measures to capture the trend of
individuals’ motivation (H1) and the underlying psychological mechanisms (H2 and H3).
In addition, the next study extended previous studies in two important ways. First, we
extended our test beyond physical goal pursuit and work for reward and moved to a more
typical situation in organizations: work-for-pay. Second, we expanded the goal type under
exploration from promotion-type goals (i.e., reaching the goal to earn the pay) to prevention-
type goals (i.e., losing the pay if one fails to reach the goal; adopted from Brockner &
Higgins, 2001; Higgins, 1997). This test further enhances the generalizability of our theory:
Both types of goals are widely used in organizations to motivate employees, and,

SUB-GOALS AS A SOURCE OF MOTIVATION 23
theoretically, prevention and promotion goals constitute the two major self-regulatory
systems that shape human motivation (Brockner & Higgins, 2001; Higgins, 1997).
Study 3: Work-for-Pay
Participants in Study 3 completed a transcription job for cash payment. We either
presented the job as a holistic, continuous task totaling 100 points (overall goal condition) or
divided the job into five 20-point consecutive sub-tasks (sub-goal condition). We measured
participants’ perceived goal attainability and perceived value of the next goal-directed action
at different time points in the pursuit and assessed their subsequent motivation.
Method
We aimed to recruit 150 participants. A total of 156 undergraduate students (87
females, 69 males; average age = 20.49) completed the study and all were included in the
final sample. The experiment used a 2 (goal structure: overall goal vs. sub-goal) × 2 (progress
level: 30% vs. 70%) mixed design; goal structure was manipulated between subjects, and
progress level was a within-subject variable.
The cover story told participants that they were hired to perform a transcription task in
which they needed to type, word for word, paragraphs of text in a foreign language from
JPEG files to a word processing program on a computer. The number of words for each piece
of transcription ranged from 50 to 100. Participants were told that 1) there was no time limit
for the task, 2) they would receive points based on the volume and precision of the
transcribed texts, and 3) the exact number of points awarded would be determined by a
computer program. To frame the job as a prevention goal, we prepaid all participants $5 prior
to the task and told them that they would need to meet the total number of points in the
transcription task in order to keep the payment (Brockner & Higgins, 2001; Higgins, 1997).
We structured the points in two different ways. In the overall goal condition,
participants were told that they needed to collect 100 points for the task, and we presented a

SUB-GOALS AS A SOURCE OF MOTIVATION 24
single progress bar at the top of the page. By contrast, participants in the sub-goal condition
were informed that the job consisted of five consecutive sections (1–5) and that they needed
to earn 20 points in each section (100 points in total). For this condition, five horizontal bars
running parallel to each other were shown on the screen. All participants were told that they
would not receive partial payment for an incomplete task.
Participants then began the transcription task. Importantly, there was no real-time
feedback on the goal; rather, participants were prompted to submit their answers and check
for progress twice during the task. The first prompt appeared approximately 10 minutes into
the task, and the computer displayed a “checking and calculating” page. Feedback then
appeared on the screen and informed participants that they had accumulated 31 points.
Participants were also presented with an updated progress bar(s). For those in the overall goal
condition, 31 points were shown on the single progress bar, and approximately 1/3 of the bar
changed color to indicate the progress. For those in the sub-goal condition, the 31 points were
represented by covering the entirety of the first small bar and approximately half of the
second bar. The second prompt for participants’ progress checking appeared another 20
minutes into the task. Following the same calculating page, participants learned that they had
earned 73 points. Similar to the previous round, those in the overall goal condition saw
approximately 3/4 of the single progress bar change in color to indicate their progress,
whereas those in the sub-goal condition saw three bars and 3/4 of the fourth short progress
bar change in color.
After participants viewed their progress via each of the two prompts, they were asked
to answer a few questions about their experiences before they could continue. Among these
questions, we assessed participants’ perceived goal attainability and perceived value of their
actions. Further expanding the measures in Study 2, we probed participants’ perceived
attainability through three questions: “How likely do you think you are to complete the entire

SUB-GOALS AS A SOURCE OF MOTIVATION 25
task to keep the pay? (1 = Very unlikely, 9 = Very likely)”; “How difficult do you think it is
for you to complete the entire task to keep the pay? (1 = Not difficult at all, 9 = Very
difficult)”; and “How attainable of a goal do you think it is for you to complete the entire task
to keep the pay? (1 = Not attainable at all, 9 = Very attainable)”. Similarly, we measured their
perceived value of the next goal-directed action through three additional questions: “How
much value do you see in transcribing the next piece to earn more points? (1= No value at all,
9 = Extremely high value)”; “How important is it for you to transcribe the next piece to earn
more points? (1= Not important at all, 9 = Extremely important)”; and “How crucial is it for
you to transcribe the next piece to earn more points? (1= Not crucial at all, 9 = Extremely
crucial)”. In each of the surveys, these six questions were embedded in 12 other filler
questions (e.g., “How often do you take part-time jobs?”; “What percentage of people around
you work part-time?”; and “What are the most enjoyable tasks that you worked on?”). And
other than these six core questions, all filler questions were different in the two instances in
order to minimize suspicion.
In both instances, participants resumed the task after completing the survey and were
immediately presented with a bonus piece to earn more points for this job. We told
participants that this round was an extra opportunity for them to earn additional points toward
the same goal and that the points they would earn in this round would be based on the volume
of accurate transcription. They were also told that they could work on this piece to earn
points for as long as they wanted and could click “Continue” at any time to exit. This design
allowed us to obtain a clean measure of their motivation by calculating how many words
participants accurately transcribed before they decided to return to the main task.
Furthermore, by measuring motivation after providing participants feedback in both early and
advanced stages of the task, we were able to assess how the motivational impact of an overall

SUB-GOALS AS A SOURCE OF MOTIVATION 26
goal structure versus a sub-goal structure changed over time. All participants were debriefed
after they completed the job and kept their compensation.
Results and Discussion
Motivation. We performed a repeated-measures ANOVA on the number of words
that participants accurately transcribed after receiving the progress feedback. The analysis
yielded a main effect of progress level (F(1, 154) = 4.97, p < .05, ηp2 = .03) and a goal
structure × progress level interaction (F(1, 154) = 24.47, p < .001, ηp2 = .14). After
participants had completed 30% of the job, those who pursued a series of sub-goals
transcribed more words (M = 558.53 words, SD = 413.45) than those who pursued an overall
goal (M = 375.12 words, SD = 434.28; t(154) = –2.70, p < .01). However, we observed the
opposite pattern after the participants had completed approximately 70% of the job: those
who pursued an overall goal transcribed more words (M = 645.20 words, SD = 480.51) than
those who pursued sub-goals (M = 456.23 words, SD = 327.23; t(154) = 2.88, p < .01).
Separate repeated-measures ANOVAs within the overall goal and sub-goal conditions
replicated the findings of first two studies. When the participants focused on the overall goal,
their motivation was low early in the pursuit but increased significantly as they approached
the end of the task (F(1, 75) = 23.65, p < .001, ηp2 = .24). By contrast, when participants
focused on the sub-goals, although they were highly motivated at the beginning of the task,
they showed lower motivation as they approached the end, even though they had exactly the
same amount of points as those in the overall goal condition (F(1, 79) = 4.03, p < .05, ηp2 =

SUB-GOALS AS A SOURCE OF MOTIVATION 27
.05; see Figure 4).34 This replicated the findings in the first two studies and provided
additional support for our first hypothesis in a work-for-pay, prevention-goal context using a
within-subject design.
Insert figure 4 about here
Goal attainability and value of actions. The three items used for each proposed
mechanism showed high reliability (Cronbach’s α for perceived attainability = .81;
Cronbach’s α for perceived value of action = .89); thus we averaged each set of three items to
form two composite indexes. A repeated-measures ANOVA on perceived goal attainability
yielded a main effect of progress level (F(1,154) = 31.85, p < .01, ηp2 = .17), a main effect of
goal structure (F(1, 154) = 6.65, p = .01, ηp2 = .04), and a goal structure × progress level
interaction (F(1, 154) = 9.88, p < .01, ηp2 = .06). Similar to Study 2, when participants had
completed only 30% of the task, those in the sub-goal condition believed that they were more
likely to attain the goal (M = 5.32, SD = 2.48) than those in the overall goal condition (M =
3.88, SD = 2.58; t(154) = –3.54, p = .01). However, when participants had completed
approximately 70% of the job, there was no significant difference in perceived goal
attainability between those in the sub-goal condition (M = 5.91, SD = 1.89) and those in the
overall goal condition (M = 5.96, SD = 1.88; t(154) = .17, ns).
3 We also analyzed the number of words participants had transcribed in total (i.e., their total
effort/performance). We found that the total effort/performance did not differ between the sub-goal
(M = 955.71, SD = 1239.64) and overall goal conditions (M = 1216.91, SD = 1410.90; F(1,154) =
1.51, p = .22, ηp2 = .10), indicating that although sub-goals (vs. overall goal) had momentary positive
or negative impact on motivation, they did not necessarily lead to a net positive or negative impact on
overall performance.
4 We also analyzed the number of words participants had transcribed before receiving progress
feedback to ensure that there was no significant difference in their internal effort perception. We
found that the actual effort expenditure prior to receiving progress feedback did not differ between the
sub-goal (M = 188.10, SD = 21.51) and overall goal conditions (M = 188.68, SD = 14.90; F(1,154) =
0.04, p = .85, ηp2 = .00). Participants in the sub-goal condition did not invest greater effort than those
in the overall goal condition prior to receiving progress feedback.

SUB-GOALS AS A SOURCE OF MOTIVATION 28
A repeated-measures ANOVA on the perceived value of action yielded a main effect
of goal structure (F(1, 154) = 6.12, p = .01, ηp2 = .04) and a goal structure × progress level
interaction (F(1, 154) = 4.39, p < .05, ηp2 = .03). Similar to Study 2, there was no difference
in the perceived value of actions when participants had only completed 30% of the task and
job completion was still far away (Msub-goal = 5.05, SD = 2.21 vs. Moverall goal = 5.32, SD =
2.47; t(154) = 0.71, ns). However, after the participants had completed 70% of the job, those
pursuing an overall goal perceived their actions to be more valuable (M = 6.11, SD = 1.94)
than those pursuing a series of sub-goals (M = 4.98, SD = 2.07; t(154) = 3.52, p = .001).
From goal attainability and value of actions to motivation. To test the underlying
mechanisms of perceived goal attainablity and perceived value of goal-directed actions, we
conducted bias-corrected mediation analyses with both factors entered as simultaneous
mediators (model 4, bootstrapping sample size = 5,000; Hayes, 2013; Preacher & Hayes,
2008). Unlike in Study 2, both the two mediators and the participants’ motivation were
measured twice (i.e., repeated measures). We thus conducted two separate mediation analyses
within each progress level (instead of entering progress level as a between-subject
moderator). We found that when the participants had completed only 30% of the job, the
influence of the structure of the goal (overall goal vs. sub-goal) on motivation was fully
mediated by perceived goal attainability (indirect effect: β = 141.32 [95% CI = 67.64 to
238.43]), whereas the indirect effect through the perceived value of actions was
nonsignificant in this early stage (indirect effect: β = –5.30 [95% CI = –31.21 to 6.44]). When
participants had reached 70% progress, the perceived value of actions became a significant
mediator (indirect effect: β = –127.64 [95% CI = –218.73 to –55.86]), such that focusing on
the sub-goal undermined the perceived value of the next goal-directed action, leading to
lower motivation. The indirect effect through perceived goal attainability was nonsignificant
in this late stage (indirect effect: β = –2.23 [95% CI = –35.37 to 22.44]).

SUB-GOALS AS A SOURCE OF MOTIVATION 29
The first three studies provided empirical evidence that supported our hypotheses in a
variety of contexts (e.g., exercising, work-for-pay), with both promotion and prevention
goals, and through between- and within-subject designs. Based on these findings and our
theorizing, the most productive goal structure for companies and employers to adopt would
be a hybrid one that sets sub-goals in the initial stage and then removes them in the advanced
stage. We tested the effectiveness of this hybrid goal structure in our final study in order to
gauge the external validity of our conceptual model as well as to provide a theory-driven
solution for organizations and managers. We tested this idea in an eight-day work-for-pay
field experiment.
In addition, an alternative explanation to our theory is that the participants provided
with a sub-goal structure might be better calibrated because they can distribute their effort
based on the milestones provided by sub-goals. Testing the hybrid goal structure would help
to rule out this possibility because, based on our theorizing, the proposed hybrid structure
should lead to the highest level of accumulated effort and performance by leveraging the
shifting driving forces in early and advanced stages throughout the course of goal pursuit.
Study 4: Sub-Goal First and Overall Goal Later
In Study 4, we collaborated with a crowdsourcing company and launched an eight-
day field experiment. Registered workers of the company completed a market intelligence
collection job for pay via the company’s mobile app. They were encouraged to either set a
series of sub-goals (sub-goal condition) or focus on the overall goal (overall goal condition).
We also included a “sub-goal first and overall goal later” hybrid condition, which presented
the sub-goals during the first four days of the job but changed the goal structure to the overall
goal for the latter four days. We provided progress feedback each day at approximately noon
and recorded participants’ performance at the end of each day.
Method

SUB-GOALS AS A SOURCE OF MOTIVATION 30
The study used a goal structure (overall goal vs. sub-goal vs. sub-goal first and overall
goal later) × progress level mixed design. The goal structure was manipulated as a between-
subject factor, and the progress level was a within-subject factor. We announced a job in
collaboration with an online crowdsourcing company to hire workers to collect up-to-date
information for books (e.g., shelf location, discounted price) at different bookstores. We
aimed to recruit 200 to 250 workers. A total of 207 workers (139 males, 68 females; average
age = 30.60) signed up. Prior to this experiment, these registered workers had completed, on
average, 32.09 work-for-pay tasks for this online crowdsourcing company since their
registration as part-time workers; the work history of the three groups did not differ (F(2,
204) = 1.22, ns).
The job required workers to collect up-to-date market information during an eight-day
period. The workers needed to collect sales information on books at different bookstores
across town by taking pictures and uploading them via the crowdsourcing company’s mobile
app. They would receive “work points” for each book they uploaded; the more books they
uploaded, the more points they would earn. If the workers accumulated a total of 80 work
points by the end of the eight-day period, they would be paid RMB 60 (approximately $10).
We sent daily reminders and feedback to workers through their smartphones at 12:00
p.m. each day during the course of the experiment. In the overall goal condition, the
reminders emphasized that the workers needed to earn a total of 80 work points by the end of
the task to receive the compensation. We illustrated the goal with a long progress bar
anchored by 80 points on the right end. In the sub-goal condition, the reminders encouraged
the workers to divide the 80 points into eight smaller sub-goals of earning 10 points each day.
To strengthen this manipulation, we illustrated eight short progress bars, each anchored by 10
points on the right end. We also included a hybrid condition in which we presented sub-goals
during the first four days but presented the overall goal during the latter four days. The

SUB-GOALS AS A SOURCE OF MOTIVATION 31
workers in this condition viewed their current number of points relative to their current sub-
goal (“4/10 points today”) from days 1 to 4 but received feedback relative to their overall
goal (“44/80 points in total”) from days 5 to 8. It is worth noting that while there was another
possible (reversed) hybrid condition that would present the overall goal during the first four
days and the sub-goals during the latter four days, we did not include this condition in the
field experiment because based on our theory, this reversed hybrid condition would only lead
to worse performance, and the director and managers at the collaborating company were
interested in maximizing the positive effect of goal structure rather than reducing it.
This company’s regular practice was to award one work point for each book upload.
To control for progress across three conditions and to ensure the credibility of our feedback
to participants, we told them that for this job, the points would be allocated based on the
number of books uploaded as well as the the quality of the information (e.g., preciseness,
informativeness, and uniqueness). We intentionally kept the quality standard vague such that
we could credibly and directly manipulate progress feedback. We sent out a progress report at
12:00 p.m. each day, and reported to each worker that he or she had progressed as follows: 4
points on day 1 (mid-day), 16 points on day 2, 23 points on day 3, 37 points on day 4, 44
points on day 5, 56 points on day 6, 63 points on day 7, and 77 points on day 8. The variation
of the points ensured steady progress toward the end point and the credibility of the feedback.
While the point feedback was manipulated and controlled, the platform recorded the actual
number of books the participants uploaded each day; the more motivated they were to
achieve the goal, the more books they would upload. This number served as our measure of
workers’ motivation and performance. All workers who uploaded 80 books were
compensated RMB 60 wage; those who uploaded fewer than 80 books were compensated in
proportion to their task completion level.
Results and Discussion

SUB-GOALS AS A SOURCE OF MOTIVATION 32
Total performance. We first analyzed total performance (total number of books
uploaded) across the three conditions. At the end of the eight-day period, the workers in the
overall goal condition (n = 68) uploaded 1,268 books in total, while those in the sub-goal
condition (n = 69) uploaded 1,392 books. The workers in the hybrid condition (sub-goal early
and overall goal later, n = 70) uploaded 1,906 books, generating the highest collective total
performance among the three goal structures. A Poisson regression of the total number of
books uploaded revealed a main effect of goal structure (Wald χ2 (2) = 130.51, p < .001),
indicating that the workers in the hybrid condition uploaded more books at the end of eight
days (M = 27.27, SD = 32.03) than those in the overall goal condition (M = 18.65, SD =
18.41; β = –0.38, Wald χ2 (1) = 109.13, p < .001) and those in the sub-goal condition (M =
20.19, SD = 19.67; β= –.30, Wald χ2 (1) = 72.34, p < .001). The people in the sub-goal
condition also performed slightly better than those in the overall goal condition (β = –
.079, Wald χ2 (1) = 4.11, p = .043), although we did not observe this net positive effect in our
prior within-subject study (Study 3) or the goal achievement dependent measure below.
Another measure of total performance is whether the participants actually reached the
goal based on the company’s regular practice—that is, whether they uploaded at least 80
books by the end of the job. A total of 90 workers uploaded 80 books or more, yielding a goal
achievement rate of 43.5%. There was a significant difference between the three groups (χ2
(2, N = 207) = 8.43, p < .05). The goal achievement rate was higher for the hybrid condition
(57.1%) compared to the sub-goal condition (39.1%; χ2 (1, N = 139) = 4.52, p < .05) and the
overall goal condition (33.8%; χ2 (1, N = 138) = 7.56, p < .01). The latter two groups did not
differ in their goal achievement rate (χ2 (1, N = 137) = .42, ns).
Motivation across eight days. Similar to Study 3, we conducted a repeated-measures
ANOVA on the number of books the workers uploaded each day, with goal structure as the
between-subject predictor and progress level as the within-subject factor. The analysis

SUB-GOALS AS A SOURCE OF MOTIVATION 33
yielded a main effect of progress level (F(7, 1428) = 9.29, p < .001, ηp2 = .044), qualified by
the predicted goal structure × progress level interaction (F(14, 1428) = 2.96, p < .001, ηp2 =
.028). Follow-up separate repeated-measures ANOVAs for each condition revealed that when
the workers focused on the overall goal of earning 80 work points, their motivation fit a
significant quadratic function (F(1, 67) = 20.54, p < .001, ηp2 = .24). Although they began the
task as motivated as the workers in the sub-goal condition, their motivation quickly decreased
after the first day, remained low during days 2 to 5, and increased on day 6. In comparison,
when the goal was structured as a set of sub-goals, the workers’ motivation fit a significant
negative linear function (F(1, 68) = 24.96, p < .001, ηp2 = .27). They began the work highly
motivated and then gradually reduced their effort as they inched forward in the task. The
motivation of workers in the hybrid condition also fit a quadratic function (F(1, 69) = 10.73,
p < .01, ηp2 = .14). During days 1 to 4, these workers’ motivation was higher than that of
those in the overall goal condition and was comparable to that of those in the sub-goal
condition. Importantly, these participants remained highly motivated during the latter half of
the task (days 5 to 8), which explained the higher total performance and goal achievement
rate in this condition (see Figure 5a). The pattern remained the same when we analyzed only
those who actually uploaded 80 books and completed the job according to the company’s
regular practice (goal structure × progress level interaction, F(14, 609) = 2.66, p = .001, ηp2 =
.058; see Figure 5b).
Insert Figures 5a and 5b about here
The results of Study 4 provide further evidence that although having sub-goals indeed
motivates greater effort at the beginning of the task, a focus on the overall goal elicits the
optimal amount of effort once people approach the end of the pursuit. Interestingly, the total
effort across the entire period differed slightly between the subgoal and overall goal

SUB-GOALS AS A SOURCE OF MOTIVATION 34
conditions, but not on participants’ goal achievement rate nor the total effort/performance
measure in Study 3 that also employed a within-subject design.
The most notable finding in this study was that while the total performance may not
reliably differ between the sub-goal and overall goal conditions, the workers in these two
conditions performed worse than those who had a hybrid structure of sub-goal first and
overall goal later. This finding provides a valuable solution for organizations and companies
that plan to use a sub-goal structure to motivate their employees.
General Discussion
Sub-goals, defined as the pre-established smaller steps toward an overarching goal
(Borrelli & Mermelstein, 1994; Lewin, 1936), bring both benefits and costs to individuals’
goal pursuit. Four studies that employed different contexts and different
structures/completion ratios of sub-goals in the lab and in the field provided supportive
evidence for our framework. In an exercise context, Study 1 showed that the impact of sub-
goals on motivation shifted depending on which stage people were in during goal pursuit.
Study 2 replicated this effect and directly tested the underlying mechanisms in a restaurant
review task; we found that a sub-goal structure (vs. an overall-goal–only structure) increased
motivation when progress was low because it enhanced perceived goal attainability, which
was the source of motivation at this stage. By contrast, a focus on the overall goal led to a
higher perceived value of goal-directed actions and thus greater motivation when people had
accumulated a high level of progress and the value of actions became their source of
motivation. Study 3 further tested the motivational impact of sub-goals and their underlying
psychological mechanisms with a wider set of measures in a continuous work-for-pay task; in
addition, it showed that the same motivational pattern could manifest for prevention-focused
goals. Finally, through an eight-day field experiment in collaboration with a croudsourcing
company, Study 4 showed that a hybrid structure of “sub-goal early and overall goal later”

SUB-GOALS AS A SOURCE OF MOTIVATION 35
elicited the greatest amount of total effort and sustained a high level of motivation throughout
the job.
Theoretical Implications
The present findings provide important insight into our understanding of the
variations in motivation as a person advances through a course of goal pursuit. The traditional
goal gradient effect proposes a general upward trend in motivation (Anderson, 1933; Hull,
1932). The current findings add a new layer of understanding and suggest that this effect
should be evident only when people view the entire goal pursuit as a holistic process, in
which case accumulated progress breeds increased motivation by making the goal-directed
actions appear more valuable. By contrast, the presence of sub-goals may alter this robust
effect (e.g., Studies 3 and 4), especially when people near the end of the pursuit.
A second key theoretical contribution of our findings is that we fine-tuned existing
theories that suggest that people are motivated by the value of and, thus, their commitment to
the ultimate goal (e.g., Liberman & Förster, 2008; Shah et al., 1998). Our findings support the
overall notion that greater goal value motivates more goal-directed actions. In addition, our
results aid in a more precise understanding that individuals’ perceived value of their present
actions—rather than the value of the overall goal—may have a more direct influence on
motivation. Therefore, it is possible that individuals experience different levels of motivation
even when both the action and the end goal remain unchanged.
This separation between action value and the value of the overall goal echoes
Higgins’ (2006) work on value and engagement. This line of research suggests that value is
not solely an experience of pleasure or pain. Value involves an experience of the intensity of
a motivational force, which could come from sources that are independent of the value of the
final goal, such as regulatory fit and the use of proper means in the pursuit (Higgins 2000;
2006). Our findings suggest that when sub-goals are the focus, people assess the value of

SUB-GOALS AS A SOURCE OF MOTIVATION 36
their goal-directed action based on the intermediate structure. Depending on how a sub-goal
relates to the overall goal, the same action might be experienced differently, adding to the
notion that there is value in how goals are pursued independent of the value of the ultimate
outcome.
Finally, we obtained evidence for the shift of individuals’ source of motivation,
adding a temporal dimension to the classic expectancy-value models (Atkinson, 1957;
Liberman & Förster, 2008; Locke & Latham, 1990; Mischel, Cantor, & Feldman, 1996;
Tolman, 1955; Vroom, 1964). We found that when progress is low, people’s source of
motivation lies in the perceived attainability of the goal; when the progress level increases
and the end point is within reasonable proximity, the source of motivation shifts to the value
people attach to their actions (e.g., Studies 2 and 3). As a result, the impact of a sub-goal
structure depends on the dominant source of motivation at a specific point of the pursuit. As
shown in Study 4, a hybrid structure that highlighted sub-goals early on and then removed the
sub-goal when employees reached the advanced stage of the task could help to leverage the
driving forces in both stages and maximize their total effort and performance.
Implications for Organizations and Managers
Although it is not surprising that sub-goals could facilitate goal pursuit, and many
organizations specifically set sub-goals for their employees (e.g., daily performance goals,
weekly sales goals, monthly fund-raising goals), our key contribution lies in demonstrating
how the motivational consequences of sub-goal structure may shift from stage to stage and
alterting people to the potential downside of such a structure. Importanlty, we by no means
suggest that sub-goals are not useful; they can be motivating indeed, as all of our studies have
shown. Rather, we examine the dynamic sources of motivation and explore the conditions
under which sub-goal structure may be the most beneficial.

SUB-GOALS AS A SOURCE OF MOTIVATION 37
For organizations that aim to motivate people (e.g., employees, sales teams, donors),
the implication of the current study is that the use of sub-goals should depend on the specific
situation. While sub-goal–based feedback could be very beneficial at the beginning of the
pursuit (e.g., the beginning of the day, week, or month), as people make progress toward the
end, providing feedback based on the overall goal (Study 4) can help ensure a high level of
motivation and performance until the ultimate goal is achieved.
What if organizations are not able to change the goal structure for employees once
they have embarked on the pursuits of their performance goals? One potential solution is
shifting individuals’ attention based on the most motivating component. We conducted
another experiment (N = 224, 118 females, 106 males; average age = 20.72) to explore a
potentially beneficial (yet less effective than the one tested in Study 4) hybrid structure, in
which we set sub-goals but experimentally focused the participants’ attention on the overall
goal (see Appendix for visual stimuli).
Participants rated news articles for a soon-to-be-launched news website for seven
days to gain 140 reviewer points; they were placed either in an overall-goal–only, sub-goals,
or hybrid structure. We measured these raters’ motivation on Day 2 (low progress) and Day 6
(high progress), and found the hypothesized goal structure × progress level interaction (F(2,
218) = 7.28, p < .01, ηp2 =.06): When the progress level was low, those in the sub-goal
condition were more motivated than those in the overall goal condition (t(81) = 3.63, p <
.01), and moderately more motivated than those in the sub-goal present/overall goal
highlighted condition (t(76) = 1.76, p = .08); there was no significant difference between the
latter two groups (t(75) = 1.35, ns). However, after raters accumulated a high level of
progress, those in the sub-goal condition were less motivated than those in the overall goal
condition (t(71) = –2.13, p < .05) or those in the sub-goal present/overall goal highlighted
condition (t(69) = –2.09, p < .05); the difference between the latter two groups was not

SUB-GOALS AS A SOURCE OF MOTIVATION 38
significant (t(64) = –.20, ns. It is interesting that visually highlighting the overall goal (while
keeping the structure of sub-goals) could help the raters elevate their focus to the final
destination when evaluating their actions, diminishing the negative impact of sub-goals in the
advanced stage of the task. While this attention-shifting solution was not as effective as the
hybrid goal structure tested in Study 4, it could serve as an executable alternative for
organizations and managers who are not able to change their goal structure and feedback
system/format halfway through the task. Future research is encouraged to explore different
hybrid goal structures that leverage the two dynamic driving forces across different stages of
goal pursuit.
Another inference that we can draw from our findings is that people who are new to a
pursuit (e.g., sales trainees, newly hired employees, first-time donors), as opposed to
experienced individuals, may need more help from organizations regarding the
design/structure of the goal because they tend to worry more about whether a goal is
attainable. Dividing an overall goal into manageable sub-units is thus crucial in sustaining
these individuals’ motivation. Although sub-goals are useful in initiating efforts, their use
should be carefully managed when people are deep into a pursuit and no longer need sub-
goals to ensure the goal’s attainability.
Limitations and Future Research
What happens when sub-goals do not dissect the overall goal in terms of quantity but,
rather, contribute to the overall goal in a qualitative manner? For instance, to achieve a sales
goal, an employee can divide this goal into weekly units of smaller quantities; he or she can
also set sub-goals such as taking shorter lunch breaks, reducing the use of social media at
work, and taking summary notes for the calls made at the end of each day. Although the latter
type of sub-goal does not quantitatively divide the overall goal, it specifies the steps one
could take to achieve this higher level goal (e.g., Fishbach et al., 2006; Kruglanski et al.,

SUB-GOALS AS A SOURCE OF MOTIVATION 39
2002). It is plausible that our theory could apply to this context as well, as an easier sub-goal
(e.g., taking a shorter lunch break today) could help to make the overall sales goal seem more
attainable, whereas a challenging sub-goal (e.g., acquiring a new skill or learning a new
software to generate sales calls; Kanfer & Ackerman, 1989) could make goal achievement
seem difficult and, therefore, affect effort when attainability is the source of motivation.
Similarly, if this sub-goal is associated with lower or even negative value (e.g., reducing the
use of social media enhances one’s productivity at work but hurts social relationships), it
could undermine the perceived value of actions and, thus, motivation. Explorations along this
line could not only provide valuable extensions of our findings, but also uncover important
insights for researchers and managers.
It is also important to note that we restricted our exploration to situations in which
sub-goals carried no value in themselves. In daily life, sub-goals often have benefits. For
example, reaching a weekly sales goal often comes with perks, albeit small, and the payment
of a sub-set of one’s student loan can help improve credit scores. We limited our explorations
to sub-goals with no benefits of their own, not only to simplify the study, but also to provide
a stronger test of how a sub-goal structure both increases perceived attainability and leads to
low perceived value of actions. If we lifted the restriction and allowed sub-goals to carry
benefits of their own, we may expect the present reversal to be even more pronounced.
Specifically, people could be more likely to associate their efforts with the benefits of these
sub-goals, further reducing their motivation toward the overall goal. Similarly, if
accomplishing a sub-goal is endowed with a high level of intrinsic reward, the sub-goal could
function in a manner that is similar to sub-goals carrying extrinsic benefits, and distract
efforts away from the overall goal. Future research shedding light on these possibilities would
be both interesting and instructive.

SUB-GOALS AS A SOURCE OF MOTIVATION 40
A careful examination of the manipulation in all of our studies reveals that the points
at which we assessed individuals’ motivation were uniformly set at the beginning, the middle,
or the end of a sub-goal. We purposefully conducted these studies in this manner to ensure
that we shed sufficient light on the process while keeping the scope of investigation
manageable. From a micro-perspective, each sub-goal pursuit is a goal; thus individuals
experience motivational variations within each of the sub-goals. To simplify the study, we
chose a specific point of sub-goals as a progress level in each study to ensure that the
motivational variation within a sub-goal would not complicate our investigation (Heath et al.,
1999). Although this simplified route satisfied our purpose of illustrating the general trend of
motivation when sub-goals are present versus absent, it left us with intriguing avenues for
future investigation. For example, how does a sub-goal gradient effect interact with the
gradient on the overall goal to influence motivation when a sub-goal structure is employed?
Does the sense of accomplishing a sub-goal interact with the amount of progress people have
accumulated on the overall goal to jointly determine motivation when people have just
completed a small milestone?
Finally, a critical assumption in our model is that goal value and attainability carry
different weights in shaping individuals’ behavior at different time points of goal pursuit. It is
worth noting that we are not making a case for the total separation of goal attainability from
its value, nor suggesting that attainability and value are entirely orthogonal constructs. What
we do believe in is that, while these constructs may not always be independent from each
other, their relative impact on momentary motivation changes when people move from one
stage to another, and the momentary motivation could be relatively more sensitive to either
consideration (goal attainability or goal value) depending on the stage that one is currently in.
The presence (and absence) of sub-goals, by addressing one or the other of these
considerations, thus has a shifting impact on momentary motivation. It is important to note

SUB-GOALS AS A SOURCE OF MOTIVATION 41
that in some situations, people may think about whether a goal is valuable at the initiation
stage and worry about attainability later on. For example, when people have multiple goals
and need to decide whether to keep pursuing the focal goal or abandon it (Unsworth, Yeo, &
Beck, 2014), it is possible that the value of the goal serves as the main driving force even in
the early stages of the pursuit. These possibilities provide interesting avenues for the test of
boundary conditions and are fruitful for future investigations.

SUB-GOALS AS A SOURCE OF MOTIVATION 42
References
Amar, M., Ariely, D., Ayal, S., Cryder, C. E., & Rick, S. I. (2011). Winning the battle but
losing the war: The psychology of debt management. Journal of Marketing Research,
48(SPL), S38–S50.
Amir, O., & Ariely, D. (2008). Resting on Laurels: The effects of discrete progress markers
as subgoals on task performance and preferences. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 34(5), 1158–1171. doi:10.1037/a0012857.
Anderson, A. C. (1933). Runway time and the goal gradient. Journal of Experimental
Psychology, 16(5), 423–428.
Atkinson, J. W. (1957). Motivational determinants of risk-taking behavior. Psychological
Review, 64(6).
Bandura, A., & Schunk, D. H. (1981). Cultivating competence, self-efficacy, and intrinsic
interest through proximal self-motivation. Journal of Personality and Social
Psychology, 41(3), 586–598. doi:10.1037/0022-3514.41.3.586.
Bonezzi, A., Brendl, C. M., & De Angelis, M. (2011). Stuck in the middle: The
psychophysics of goal pursuit. Psychological Science, 22 (5), 607–612.
doi:10.1177/0956797611404899.
Borrelli, B., & Mermelstein, R. (1994). Goal setting and behavior change in A Smoking
Cessation program. Cognitive Therapy and Research, 18(1), 69–83.
doi:10.1007/BF02359396.
Brockner, J., & Higgins, E. T. (2001). Regulatory focus theory: Implications for the study of
emotions at work. Organizational Behavior and Human Decision Processes, 86(1), 35–
66.
Brunstein, J. C. (1993). Personal goals and subjective well-being: A longitudinal study.
Journal of Personality and Social Psychology, 65(5), 1061–1070. doi:10.1037/0022-
3514.65.5.1061.
Carver, C. S., & Scheier, M. F. (1990). Principles of self-regulation: Action and emotion. In
E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and cognition:
Foundations of social behavior (Vol. 2, pp. 3–52). New York: Guilford Press.
Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. New York, NY:
Cambridge University Press.
Financial Planning. (2016). Perfecting performance reviews. Human Capital, 46(11), 27.

SUB-GOALS AS A SOURCE OF MOTIVATION 43
Etkin, J., & Ratner, R. K. (2012). The dynamic impact of variety among means on
motivation. Journal of Consumer Research, 38 (April), 1076-92.
Fishbach, A., Dhar, R., & Zhang, Y. (2006). Subgoals as substitutes or complements: The
role of goal accessibility. Journal of Personality and Social Psychology, 91(2), 232–
242. doi:10.1037/0022-3514.91.2.232.
Fishbach, A., Henderson, M. D., & Koo, M. (2011). Pursuing goals with others: Group
identification and motivation resulting from things done versus things left undone.
Journal of Experimental Psychology. General, 140(3), 520–534.
doi:10.1037/a0023907.
Gal, D., & McShane, B. B. (2012). Can small victories help win the war? Evidence from
consumer debt management. Journal of Marketing Research, 49(4), 487–501.
doi:10.1509/jmr.11.0272.
Garland, H. (1983). Influence of ability, assigned goals, and normative information on
personal goals and performance: A challenge to the goal attainability assumption.
Journal of Applied Psychology, 68(1), 20–30. doi:10.1037/0021-9010.68.1.20.
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. New York: Guilford Press.
Heath, C., Larrick, R. P., & Wu, G. (1999). Goals as reference points. Cognitive Psychology,
38(1), 79-109. doi:10.1006/cogp.1998.0708.
Higgins, E. T. (1997). Beyond pleasure and pain. American psychologist, 52(12), 1280–1300.
Higgins, E. T. (2000). Making a good decision: Value from fit. American Psychologist, 55,
1217–1230.
Higgins, E. T. (2006). Value from hedonic experience and engagement. Psychological
Review, 113(3), 439–460. doi:10.1037/0033-295X.113.3.439.
Huang, S., Zhang, Y., & Broniarczyk, S. M. (2012). So near and yet so far: the mental
representation of goal progress. Journal of Personality and Social Psychology, 103(2),
225–41. doi:10.1037/a0028443.
Hull, C. L. (1932). The goal-gradient hypothesis and maze learning. Psychological Review,
39(1), 25–43. doi:10.1037/h0072640.
Kanfer, R., & Ackerman, P. L. (1989). Motivation and cognitive abilities: An integrative/
aptitude-treatment interaction approach to skill acquisition. Journal of Applied
Psychology, 74(4), 657–690. doi: 10.1037/0021-9010.74.4.657.

SUB-GOALS AS A SOURCE OF MOTIVATION 44
Koo, M., & Fishbach, A. (2008). Dynamics of self-regulation: how (un)accomplished goal
actions affect motivation. Journal of Personality and Social Psychology, 4(2),183–195.
doi: 10.1037/0022-3514.94.2.183.
Koo, M., & Fishbach, A. (2012). The small-area hypothesis: effects of progress monitoring
on goal adherence. Journal of Consumer Research, 39(3), 493–509.
Kivetz, R., Urminsky, O., & Zheng Y. H. (2006). The goal-gradient hypothesis resurrected:
purchase acceleration, illusionary goal progress, and customer retention. Journal of
Marketing Research, 43(1), 39–58. doi: 10.1509/jmkr.43.1.39.
Kruglanski, A. W. (1996). Motivated social cognition: Principles of the interface. In E. T.
Higgins & A. W. Kruglanski (Eds.), Social psychology: handbook of basic principles,
(pp. 493-520). New York, NY: Guilford.
Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y., & Sleeth-Keppler,
D. (2002). A theory of goal-systems. In M. P. Zanna (Ed.), Advances in experimental
Social psychology (Vol. 34) (pp. 331–376). New York, NY: Academic Press.
Levin, I. P., Schneider, S. L., & Gaeth, G. J. (1998). All frames are not created equal: A
typology and critical analysis of framing effects. Organizational Behavior and Human
Decision Processes, 76(2), 149–188. doi:10.1006/obhd.1998.2804.
Lewin, K. (1936). A dynamic theory of personality. New York, NY: McGraw-Hill.
doi:10.1097/00005053-193611000-00051.
Liberman, N., & Förster, J. (2008). Expectancy, value and psychological distance: A New
look at goal gradients. Social Cognition, 26(5), 515–533.
doi:10.1521/soco.2008.26.5.515.
Locke, E. A., & Latham, G. P. (1990). A theory of goal setting and task performance.
Englewood Cliffs, NJ: Prentice Hall.
Louro, M. J., Pieters, R., & Zeelenberg, M. (2007). Dynamics of multiple-goal pursuit.
Journal of Personality and Social Psychology, 93(2), 174–193. doi:10.1037/0022-
3514.93.2.174.
Mischel, W., Cantor, N., & Feldman, S. (1996). Principles of self regulation: The nature of
willpower and self-control. In E. T. Higgins, & A. W. Kruglanski (Eds.), Social
psychology: Handbook of basic principles (pp. 329–360). New York, NY: Guilford
Press.
Murray, H. A. (1938). Explorations in personality. New York, NY: Oxford University Press.

SUB-GOALS AS A SOURCE OF MOTIVATION 45
Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, NJ: Prentice
Hall.
Nunes, J.C. & Drèze, X. (2006). The endowed progress effect: how artificial advancement
increases effort. Journal of Consumer Research, 32(4), 504–512.
Pervin, L. A. (1989). Goal concepts in personality and Social psychology: A historical
perspective. In A. P. Lawrence, & N. J. Hillsdale (Eds.), Goal concepts in personality
and Social psychology (pp. 1–17). Erlbaum.
Pratt, M.K. (2007). No more job reviews. Computerworld. Retrieved
from http://www.computerworld.com on November 4, 2016.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing
and comparing indirect effects in multiple mediator models. Behavior Research
Methods, 40(3), 879–891. doi: 10.3758/BRM.40.3.879.
Shah, J., & Higgins, E. T., & Friedman, R. S. (1998). Performance incentives and means:
How regulatory Focus influences goal attainment. Journal of Personality and Social
Psychology, 74(2), 285–293. doi:10.1037/0022-3514.74.2.285.
Soman, D., & Shi, M. (2003). Virtual progress: the effect of path characteristics on
perceptions of progress and choice. Management Science, 49(9), 1229–1250.
doi:10.1287/mnsc.49.9.1229.16574.
Stock, J., & Cervone, D. (1990). Proximal goal-setting and self-regulatory processes.
Cognitive Therapy and Research, 14(5), 483–498. doi:10.1007/BF01172969.
Sutton, R. I. (2010). Good Boss, bad boss: How to be the best ... and learn from the worst.
New York, NY: Business Plus.
Tolman, E. C. (1955). Principles of performance. Psychological Review, 62(5), 315–326.
Unsworth, K., Yeo, G., & Beck, J. (2014). Multiple goals: A review and derivation of general
principles. Journal of Organizational Behavior, 35(8), 1064–1078. Doi:
10.1002/job.1963.
Vroom, V. H. (1964). Work and motivation. Oxford England: Wiley.
Weick, K. E. (1984). Small wins: Redefining the scale of social problems. American
Psychologist, 39(1), 40–49. doi:10.1037/0003-066X.39.1.40.
Wilson, T. (2016). Creating co-accountability for workplace goals. Excellence
Essentials. Retrieved from http://hr.com on November 4, 2016.

SUB-GOALS AS A SOURCE OF MOTIVATION 46
Zhang, Y., & Huang, S. (2010). How endowed versus earned progress affects consumer goal
commitment and motivation. Journal of Consumer Research, 37(4), 641–654.
doi:10.1086/655417.

SUB-GOALS AS A SOURCE OF MOTIVATION
Appendix
Visual Stimuli
Overall goal in the low-progress condition:
Overall goal in the high-progress condition:
Sub-goal in the low-progress condition: Sub-goal in the high-progress condition:
Hybrid in the low-progress condition: Hybrid in the high-progress condition:

SUB-GOALS AS A SOURCE OF MOTIVATION
Table 1. Summary of the results of all studies
Sub-goal
Overall goal
Hybrid
Interaction statistics
Low progress
High progress
Low progress
High progress
Low progress
High progress
Study 1
Motivation
221.24(28.34)
215.67(27.20)
200.38(31.08)
233.03(28.17)
F = 14.80, p < .01, ηp
2 = .10
Study 2
Motivation
207.24(101.65)
175.88(65.39)
156.73(63.56)
225.38(125.38)
F = 11.39, p < .01, ηp
2 = .07
Attainability
5.26(2.69)
5.18(2.53)
3.54(2.22)
5.28(2.15)
F = 5.64, p < .05, ηp
2 = .04
Value of action
4.43(1.67)
4.75(1.95)
4.86(1.80)
6.36(1.98)
F = 3.96, p < .05, ηp
2 = .03
Study 3
Motivation
558.53(413.45)
456.23(327.23)
375.12(434.28)
645.20(480.51)
F = 24.47, p < .001, ηp
2 = .14
Motivation trend
F = 4.03, p < .05, ηp
2 = .05
F = 23.65, p < .001, ηp
2 = .24
F = 24.47, p < .001, ηp
2 = .14
Attainability
5.32(2.48)
5.91(1.89)
3.88(2.58)
5.96(1.88)
F = 9.88, p < .01, ηp
2 = .06
Value of action
5.05(2.21)
4.98(2.07)
5.32(2.47)
6.11(1.94)
F = 4.39, p < .05, ηp
2 = .03
Study 4
Total performance
20.19(19.67)
18.65 (18.41)
27.27(32.03)
χ2 (2, N = 207) = 130.51, p < .001
Goal achievement rate
39.1%
33.8%
57.1%
χ2 (2, N = 207) = 8.43, p < .05
Motivation trend
F = 24.96, p < .001, ηp
2 = .27
F = 20.54, p < .001, ηp
2 = .24
F = 10.73, p < .01, ηp
2 = .14
F = 2.96, p < .001, ηp
2 = .028
Notes: Means are presented in each cell. Standard deviations are presented in parentheses.

SUB-GOALS AS A SOURCE OF MOTIVATION
Figure 1. Total number of steps as a function of progress level and goal structure (Study 1).
150
170
190
210
230
250
1/4 of the task 3/4 of the task
Overall goal
Sub-goal
Total number of steps
Progress Level
Goal Structure

SUB-GOALS AS A SOURCE OF MOTIVATION 50
Figure 2. Number of words as a function of progress level and goal structure (Study 2).
100
120
140
160
180
200
220
240
260
30% 70%
Overall goal
Sub-goal
Number of words
Progress Level
Goal Structure

SUB-GOALS AS A SOURCE OF MOTIVATION 51
Figure 3. Moderated mediation model of the indirect effect of goal structure on motivation
through perceived goal attainability and perceived value of actions, while the impact of the
two mediators on motivation was moderated by progress level (Study 2).

SUB-GOALS AS A SOURCE OF MOTIVATION 52
Figure 4. Number of words participants accurately transcribed as a function of progress level
and goal structure (Study 3).
100
200
300
400
500
600
700
800
30% 70%
Overall goal
Sub-goal
Number of words participants transcribed
Progress Level

SUB-GOALS AS A SOURCE OF MOTIVATION 53
Figure 5a. Number of book pictures workers uploaded as a function of progress level and
goal structure (Study 4).
Figure 5b. Number of book pictures the workers who achieved the goal uploaded as a
function of progress level and goal structure (Study 4).
0
1
2
3
4
5
6
7
Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8
Number of book pictures uploaded
Progress Level (Days)
Overall goal
Sub-goal
Sub-goal first and
overall goal later
Goal structure
0
1
2
3
4
5
6
7
8
9
10
Day 1 Day 2 Day 3 Day 4 Day 5 Day 6 Day 7 Day 8
Number of book pictures uploaded
Progress Level (Days)
Overall goal
Sub-goal
Sub-goal first
and overall
goal later
Goal structure