Self-Regulation Differences during Athletic Practice by Experts, Non-Experts, and Novices

Article (PDF Available)inJournal of Applied Sport Psychology 13(2):185-206 · March 2001with 1,607 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
DOI: 10.1080/104132001753149883
Cite this publication
Basketball experts, non-experts, and novices were studied for differences in their self-regulatory forethought and self-reflection processes regarding their free-throw shooting. Forty-three adolescent boys participated individually in the study, which involved a practice session in a gymnasium. The subjects were queried regarding their forethought goals, strategy choice, self-efficacy as well as their self-reflection attributions and feelings of satisfaction as they practiced their shooting. Among the significant results, experts set more specific goals, selected more technique-oriented strategies, made more strategy attributions, and displayed higher levels of self-efficacy than non-experts and novices. Forethought phase processes intercorrelated significantly as did self-reflection phase processes. In addition, self-reflection attributions were predictive of forethought strategy selection during further efforts to learn. The results were discussed in terms of a social cognitive model of self-regulation.
Copyright © 2001 by the Association for Advancement of Applied Sport Psychology
1041-3200/01 $12.00 + .00
Manuscript received 19 October 1999; Revision submitted 13 June 2000.
The authors would like to thank Anastasia Kitsantas at Florida State University for her
helpful suggestions regarding an earlier draft of this manuscript.
Address correspondence to either author at the Doctoral Program in Educational Psy-
chology, Graduate School and University Center, City University of New York, 365 Fifth
Avenue, New York, New York 10016-4309. E-mail address is:
Self-Regulation Differences during Athletic Practice bySelf-Regulation Differences during Athletic Practice by
Self-Regulation Differences during Athletic Practice bySelf-Regulation Differences during Athletic Practice by
Self-Regulation Differences during Athletic Practice by
Experts, Non-Experts, and NovicesExperts, Non-Experts, and Novices
Experts, Non-Experts, and NovicesExperts, Non-Experts, and Novices
Experts, Non-Experts, and Novices
J. C
J. Z
Graduate School and University Center
City University of New York
Basketball experts, non-experts, and novices were studied for differences in their
self-regulatory forethought and self-reflection processes regarding their free-throw
shooting. Forty-three adolescent boys participated individually in the study, which
involved a practice session in a gymnasium. The subjects were queried regarding
their forethought goals, strategy choice, self-efficacy as well as their self-reflec-
tion attributions and feelings of satisfaction as they practiced their shooting. Among
the significant results, experts set more specific goals, selected more technique-
oriented strategies, made more strategy attributions, and displayed higher levels
of self-efficacy than non-experts and novices. Forethought phase processes
intercorrelated significantly as did self-reflection phase processes. In addition,
self-reflection attributions were predictive of forethought strategy selection dur-
ing further efforts to learn. The results were discussed in terms of a social cogni-
tive model of self-regulation.
Expert performance has been studied in many different domains such as
chess, mathematics, and music (Chase & Simon, 1973; Ericsson, Krampe,
& Tesch-Romer, 1993). Within the last few decades there has also been
much interest in analyzing expert performance within sports such as field
hockey, baseball, basketball, ice hockey, tennis, soccer, and football
(Starkes, 1987; Garland & Barry, 1990; McPherson, 1993; McPherson &
Thomas, 1989; Starkes, Allard, Lindley, & O’Reilly, 1994, Williams &
Davids, 1995). Regardless of the sport studied, researchers have attempted
to answer a similar question: Which variables reliably differentiate sport
experts and sport novices?
These studies have shown that experts have more sophisticated and elabo-
rate knowledge than novices. This knowledge has enabled experts to rec-
ognize (Allard & Starkes, 1980) and recall (Starkes et al., 1994; Williams,
Burwitz, & Williams, 1993) sport-specific information in a more efficient
manner than novices. In addition, studies employing anticipation and vi-
sual-occlusion tasks have shown that experts have a superior ability to
anticipate player movements and to predict shot placement in field hockey
(Starkes, 1987), tennis (Jones & Miles, 1978), ice hockey (Bard & Fleury,
1981), and soccer (Williams & Davids, 1995). This skill appears to be a
function of the expert’s efficient use and processing of relevant and mean-
ingful visual cues. A related line of research has shown that experts and
novices focus on quantitatively and qualitatively different visual cues when
engaged in a sport-specific task (Goulet, Bard & Fleury, 1989; McPherson,
1993). Goulet et al. (1989) recorded the eye movements and fixations of
expert and novice tennis players as they simulated returning the shot of a
server who was on a video screen. Experts showed a significantly greater
number of eye fixations than novices and focused the majority of their
attention on the server’s racquet and arm holding the racquet. Novices fo-
cused their attention on the tennis ball but neglected to attend to the server’s
arm or racquet movement. Baseball experts have also been shown to focus
on specific cues or stimuli when preparing to bat against pitchers
(McPherson, 1993).
These studies have provided much information about expert-novice dif-
ferences, but they have given little attention to how athletes develop into
experts. Although innate physical capabilities are necessary for the devel-
opment of expertise, they do not differentiate sport experts and novices
reliably (Thomas & Thomas, 1994). Rather, expert-novice differences ap-
pear to emerge from many years of deliberate practice. This type of prac-
tice refers to individualized training on tasks that are selected and highly
structured by knowledgeable teachers or coaches in order to provide “opti-
mal opportunities for learning and skill acquisition” (Ericsson & Charness,
1994, p. 739). Deliberate practice involves: (a) setting goals involving spe-
cific skills, (b) intense involvement in structured training sessions, (c) per-
forming tasks that are not inherently motivating and contain few external
rewards, and (d) self-monitoring performance outcomes and receiving feed-
back about current performance (Ericsson et al., 1993). These features of
deliberate practice appear to reflect teachers or coaches systematic at-
tempts to instill self-regulation in their pupils (Zimmerman, 1998). Like
students who are instructed to practice deliberately, self-regulated learners
structure their practice sessions by setting specific goals and self-monitor-
ing. Self-regulation is defined as self-generated thoughts, feelings, and
behaviors that are planned and cyclically adapted based on performance
feedback (Zimmerman, 1989, 1998)
To date, only a few retrospective studies have been performed to deter-
mine if sport expertise is related to the amount of deliberate practice by an
athlete. Ericsson et al. (1993) reviewed the expertise literature across sev-
eral domains (e.g., sport, music, chess) to determine if there was a relation-
ship between expertise and daily amounts of practice. They found that ex-
pert athletes spend a large amount of time (e.g., 25 to 30 hours per week)
practicing their sport. In addition, Ericsson and colleagues discovered that
expert athletes practice according to a highly structured routine, such as
daily practice sessions punctuated by regular breaks. More recently, Hodges
and Starkes (1996) performed a study with expert and non-expert wrestlers
in order to test Ericsson and colleagues hypotheses about deliberate prac-
tice. The wrestlers were asked to provide practice time estimates retro-
spectively as well as to rate different activities according to their relevance
in improving their skills and enjoyment level. Hodges and Starkes found
that when equated for years of wrestling experience, the experts (i.e., inter-
national level wrestlers) practiced their techniques significantly more of-
ten than non-experts (i.e., club wrestlers). Wrestlers who felt that training
was necessary for improving their skills also rated these experiences as
being highly enjoyable. Helson, Starkes, and Hodges (1998) extended these
findings by assessing the practice activities of athletes participating in team
sports, such as field hockey and soccer. In general, the results supported
Hodges and Starkes’ (1996) conclusion that experts practice significantly
more often than non-experts and rate relevant activities as being enjoy-
able, and the results underscored the importance of considering both indi-
vidual and team practice as part of deliberate practice sessions.
Although these retrospective studies are important, they gave little at-
tention to forethought and self-reflection processes of players as they en-
gaged in deliberate practice. Microanalytic studies are better able to focus
on these aspects of deliberate practice because they assess specific mental
and behavioral processes as they occur and change. For example, mi-
croanalyses can provide detailed information about the thoughts underly-
ing a players goal setting and intense involvement, a player’s motivation
to initiate and persist at tasks which are not intrinsically interesting, or a
player’s capability to self-correct mistakes or make adjustments when he
or she is not performing well during self-directed practice sessions. This
information is important because practicing by oneself (i.e., out of the pres-
ence of a coach) is a major form of learning for many athletes. Microanalyses
can also reveal shifts in source of regulation or “change of agency” during
practice from a coach to an athlete (Glaser, 1996) by including successive
measures of self-efficacy.
These cognitive and motivational issues have been the focus of much
self-regulation research. Self-regulation models describe how and why
certain processes influence an athletes motivation, ability to correct mis-
takes, and ability to self-manage behaviors, thoughts, and feelings. These
models have guided the study of a variety of self-regulatory processes,
such as goal setting and feedback (Carver, 1979), self-evaluation and self-
reward (Kanfer & Karoly, 1972), problem identification and environmen-
tal management (Kirschenbaum, 1984), and imagery and focusing strate-
gies (Singer, Flora, & Abourezk, 1989). A social cognitive model (Kitsantas
& Zimmerman, 1998; Zimmerman & Kitsantas, 1996, 1997) includes these
self-regulatory processes along with self-efficacy beliefs, attributions, and
self-satisfaction reactions. This model is particularly suitable for studying
the influence of self-regulatory processes during athletic practice because
it emphasizes three microanalytic phases: forethought (which precedes
motoric practice), performance control (which occurs during practice), and
self-reflection (which follows each practice effort). Self-reflection processes
are assumed to influence forethought cyclically regarding additional per-
formance efforts, especially self-efficacy beliefs (see Figure 1). This cy-
clical model hypothesizes close relations among processes within each of
the phases. For example, in terms of the forethought phase, an individual’s
goals will be related to his or her subsequent strategy choice and self-
motivational beliefs. Highly self-regulated learners will set specific pro-
cess and outcome goals, utilize technique-oriented strategies, and display
high levels of self-efficacy and intrinsic interest (Zimmerman, 1999). The
third phase in this model highlights the importance of self-reflection as
players make adjustments to improve future performances. It is hypoth-
esized that highly self-regulated learners will attribute outcomes to strat-
egy use, thus facilitating the selection of more useful or adaptive strategies
following failure. Previous research has shown that attributions of athletic
outcomes to trained strategies sustain high levels of motivation and skill
(Kitsantas & Zimmerman, 1998). This three phase cyclical model of self-
Figure 1. Cyclical phases and subprocesses of self-regulation.
regulation will be used to understand the cognitions and behaviors of ex-
pert and non-expert athletes during practice sessions.
The present study assesses differences among basketball experts, non-
experts, and novices in the quality and quantity of their self-regulation
during a practice session. A non-expert group was added to the classical
expert-novice paradigm to provide better control of a variety of background
variables, such as basketball playing experience and familiarity with the
game that novices typically lack. This microanalytic study will assess bas-
ketball players self-regulated processes as they practice their free throws.
On the basis of earlier experimental research on dart throwing (Kitsantas
& Zimmerman, 1998; Zimmerman & Kitsantas, 1996; 1997), we hypoth-
esized that expert basketball players would display more effective forms
of forethought and self-reflection by setting more specific goals (i.e., both
outcome and process), using more process-oriented strategies, making more
specific strategy attributions and displaying higher levels of self-efficacy,
and satisfaction than non-experts and novices. It was also hypothesized
that non-expert basketball players would display more effective forms of
self-regulation by setting more specific goals (i.e., both outcome and pro-
cess), using more process-oriented strategies, making more specific strat-
egy attributions and displaying higher levels of self-efficacy, and satisfac-
tion than novices. This study also investigated hypothesized links among
various processes within and between the forethought and self-reflection
phases of a cyclical model of self-regulation.
This study included 43 male students drawn from seven high schools in
a large eastern city. In terms of their ethnic composition, there were 26
Americans of European extraction, 14 African Americans, 2 Hispanic
Americans, and 1 Asian American. All boys voluntarily participated in the
study, and parental consent was obtained for each student. Three groups of
participants were recruited based on specific criteria. An expert group con-
sisted of 15 male varsity basketball players whose game free-throw per-
centage during the current season was 70% or higher. A non-expert group
was made up of 13 male varsity basketball players who made less than
55% of their game free throws during the current season. All players were
required to have shot a minimum of 20 free throws during the basketball
season during game competition. These students were recruited from the
varsity basketball teams of the schools by the primary investigator. All the
students who were contacted agreed to participate in the study. A novice
group was selected composed of 15 male high-school students who had
never played organized basketball higher than the seventh grade. These
individuals were recruited from physical education classes, with 30 stu-
dents volunteering to participate. Fifteen of the 30 novices who were clos-
est to the boys in the expert and non-expert groups in age and demographic
background were chosen. Age, number of free-throw practice hours per
week, prior playing experience (i.e., number of years playing organized
team basketball), and knowledge of free-throw techniques were computed
for each group (see Table 1). Students were asked to give their knowledge
of three essential parts of a free-throw shot, and their answers were scored
on the basis of any of the following parts that have been emphasized in the
literature: follow-through, bent knees, body position, arm extension, el-
bow against body, and use of fingertips. Thus, basketball free-throw knowl-
edge scores could range from zero to three. All groups were compared for
differences in age, practice, playing experience, and knowledge using one-
way analysis of variance procedures. Experts and non-experts were simi-
lar across all variables. As expected, the novice group differed from the
other two groups basketball players in terms of number of hours devoted to
practicing free-throws and playing experience. The age of novices was
significantly lower than non-experts but not experts.
Table 1
Demographic Information for each Expertise Group
Demographic Variables
Expertise Group Age Practice Playing experience Knowledge
Experts 16.3 1.92 5.7 1.53
Non-experts 16.5 2.09 7.9 1.54
Novices 15.5 0.05 1.5 1.13
Note: Age and playing experience are measured in years; practice is measured in hours/week.
Knowledge was measured by listing up to three essential parts of the free-throw shot.
Self-efficacy. A measure of self-efficacy during free-throw shooting was
developed according to the guidelines outlined by Bandura and Schunk
(1981). The measure consisted of three items assessing the participants
confidence in making two consecutive free throws. All items began with
the phrase “On a scale from 0 to 100 with 10 being not sure, 40 being
somewhat sure, 70 being pretty sure, and 100 being very sure, how sure are
you that you will make . . . ” This was followed by either the phrase (a) two
shots in a row or (b) the next two shots. The first question was asked before
the subjects began shooting free throws, and the second question was asked
following two consecutive misses and following two consecutive baskets.
The participants were asked to orally respond to each question based on a
100-point scale grouped in 10-unit intervals. They were given a cue card
depicting the scale as the examiner asked each question. The examiner
recorded the efficacy judgments. Participants scores for the three items
(i.e., before shooting, after two misses, and after two baskets) were also
averaged to yield an overall self-efficacy estimate. According to Cronbach’s
alpha, the inter-item reliability of this three-item scale was .95, surpassing
estimates from previous research (Kitsantas & Zimmerman, 1998).
Self-satisfaction.A single item measure was used to assess the boys’
satisfaction with their free-throw shooting performance during the prac-
tice session. After the participants finished the entire practice session, they
were asked, “On a scale from 0 to 100, with 10 being not satisfied, 40
being somewhat satisfied, 70 being pretty satisfied, and 100 being very
satisfied, how satisfied are you with your performance during the practice
session?” The boys were required to respond orally to the question while
viewing a cue card with a written description of the 100 point scale. The
examiner recorded the self-satisfaction judgments. This type of measure
has been in used prior experimental research studies and was found to be
predictive of other forms of self-regulation (Kitsantas & Zimmerman, 1998;
Zimmerman & Kitsantas, 1996).
Goal setting.Before attempting any free throws, each participant was
asked to answer the following question: “Do you have a goal when practic-
ing these free throws? If so, what is it?” The oral responses given by the
boys were recorded verbatim by the examiner. These responses were coded
independently by two individuals into one of nine categories: outcome-
specific, outcome-general, process-specific, process-general, focus-specific,
focus-general, rhythm, none, and other. An example of an outcome-spe-
cific goal is to make 10 out of 10 shots” whereas an example of an out-
come general goal isto make baskets.” An example of a process-specific
goal is “to bend your knees whereas an example of a process-general goal
is “to try harder to practice my shooting form.An example of focus-spe-
cific goal is “to keep your eye on the rim of the basket whereas an ex-
ample of a focus-general goal is to concentrate more.An example of a
rhythm goal is to improve my rhythm.” The category other” refers to
goals that do not fit within any of these categories, such as “to do my
thing” and when no goals were set, the response was classified as none
or “don’t know.” The coders coded the participantsresponses from unla-
beled protocols to prevent scoring bias. It is important to note that each
category was dichotomously coded with a 1 signifying present and a 0
“absent,” and categories were mutually exclusive. A kappa coefficient of
.95 for inter-rater agreement was obtained.
Strategy choice.The participants were asked to answer two questions
concerning the strategy that they planned to use to perform well. The first
question, “What do you need to do to accomplish that goal? was asked
immediately following the goal question. The second question What do
you need to do to make the next shot?” was designed to assess the strategy
that players use following missed free-throw attempts. This question was
asked following two sequential failures to make a free throw during the
practice session. The participants’ oral responses to both questions were
recorded verbatim by the examiner. The strategies were classified inde-
pendently by two coders into one of nine categories: specific technique,
general technique, visualization, specific focus, general focus, distractions,
rhythm, don’t know, and other. An example of a specific technique is “keep
my elbow in” while an example of a general technique is to do my normal
routine.” The visualization category includes responses such as I visual-
ize myself making every shotor I picture myself using the perfect tech-
nique.” A specific focus response includes to focus on the back of the
rim” whereas examples of general focus verbalizations are to focusand
“to concentrate. An example of a distraction response is “to block every-
thing out of my mind.” The rhythm category includes statements such as
“to relax and “to take my time.” The “don’t know” and other categories
for the strategic planning were defined similar to the “none and “other
categories used to classify goals. Each category was scored dichotomously
with a 1 representing present and a 0 signifying “absent. The coders
coded the participants responses from unlabeled protocols to prevent scor-
ing bias. A separate kappa coefficient was calculated separately for each
type of strategy (i.e., before shooting free throws and after two consecutive
misses). Kappa analyses revealed the inter-rater agreement for the coding
strategies reported before shooting was .91 and for coding strategies re-
ported following two consecutive misses was .88.
Attributions. The participants were asked questions about the reason
for their successful and unsuccessful free-throw attempts at junctures dur-
ing the practice session. Following two consecutive missed free-throw at-
tempts players were asked, Why do you think you missed those last two
shots?” Following two successful consecutive shots the subjects were asked,
“Why do you think that you made those last two shots?” The participants
verbal responses were recorded verbatim by the examiner and were cat-
egorized according to the reason for failure. These attributions were classi-
fied independently by two coders into one of 11 categories. Because indi-
viduals often make attributions based on their selected strategy, we at-
tempted to create parallel categories within the attribution and strategy
measures (i.e., specific and general technique, specific and general focus,
distractions, rhythm, dont know, and other categories). However, the fol-
lowing traditional attributions were also included as categories: confidence,
effort, and practice. An example of a confidence attribution is, I was not
sure that I could make them while an example of an effort attribution is,
“I did not try or “I let up a bit.” Finally, an example of a practice attribu-
tion is, “I seldom practice free-throw shooting.” Each attribution category
was coded as a dichotomous variable indicating the presence or absence of
the attribution. It should be noted that the attribution categories were mu-
tually exclusive. The coders coded the participants responses from unla-
beled protocols to prevent scoring bias. A kappa coefficient was calculated
for each type of attribution (i.e., after successful shots and after two con-
secutive misses). The inter-rater agreement for both attribution categories
was .88.
Shooting skill.The entry level of the boys free-throw shooting skill
was determined by the results of the first five shots. The experimenter
mentally noted the results of these efforts but did not record the informa-
tion until later to avoid introducing a competitive climate rather than one
of self-directed practice.
Design and Procedure
Participants (i.e., experts, non-experts, and novices) were asked to prac-
tice their free-throw shooting individually at a basket in the gymnasium of
their school outside the presence of other students. Before shooting, the
examiner told each player, “I would like you to go to the foul line and
practice your foul-shooting for about 10 minutes. It is important that you
practice your free throws like you would typically do when on your own. I
will stop you at a few points during the practice session in order to ask you
a few questions.” If the player understood the task, the examiner proceeded
to ask questions relating to the player’s goals, strategies for achieving goals,
and self-efficacy. Following these questions, the examiner allowed each
participant to begin shooting free throws. After the tenth free throw, the
examiner waited for the subject to either miss or make two consecutive
shots. When the player missed two consecutive shots, the examiner stopped
the practice session to ask the player questions concerning his attributions
for failure, strategy for making the next shot, and self-efficacy judgments.
If the participant made two consecutive shots, the examiner stopped the
session and asked questions about the player’s attributions for success and
self-efficacy judgments. The practice session continued until each partici-
pant missed and made two consecutive shots. After this criterion was met,
the participants were allowed to shoot five additional shots. These shots
were followed by questions assessing the player’s satisfaction, knowledge
of correct basketball technique, and number of hours per week devoted to
practicing free throws.
Free-Throw Shooting Skill
A single factor analysis of variance was used to confirm the presence of
differences in shooting skill among the three groups of boys and a priori
contrasts among experts, non-experts, and novices were conducted using t
tests. Recall that (a) experts were expected to surpass non-experts and (b)
Table 2
Means of Metric Dependent Variables for each Expertise Group
Expertise Group
Dependent measures Experts Non-experts Novices
Self-efficacy average 87.53 79.10 47.03
Self-efficacy before shooting 87.00 77.31 48.00
Self-efficacy after successful shots 89.13 78.85 49.33
Self-efficacy after two consecutive misses 86.47 78.08 43.33
Self-satisfaction 67.67 80.77 43.00
they would in turn surpass novices on all measures. Because specific hy-
potheses were advanced among experts, non-experts, and novices, a priori
contrasts were conducted using t tests as recommended by Kirk (1968).
The outcomes of the initial five practice attempts were compared among
the three expertise groups using analysis of variance procedures, and a
significant expertise effect was obtained, F(2,40) = 29.29, p < .01. A priori
contrasts revealed that the free-throw accuracy of the experts (M = 76 %)
was significantly (p <. 05) higher than that of the non-experts (M = 58 %),
which in turn was significantly (p <. 05) higher than that of the novices (M
= 42 %). These results confirmed the expertise designation that was used
to form the three experimental groups.
Group Differences in Self-Regulation
A multivariate analysis of variance was used to summarize differences
in average self-efficacy and in self-satisfaction among the expertise groups
(see Table 2). Univariate F tests were subsequently conducted on each of
the two dependent measures, and a priori contrasts among experts, non-
experts, and novices were conducted using t-tests. A significant multivari-
ate effect for expertise was obtained, F(4, 78) = 15.05, p < .01, Wilks’s
Lambda criterion. The standardized discriminant function coefficients for
the single canonical root that was extracted was .82 for average self-effi-
cacy and .46 for self-satisfaction. Univariate F tests for expertise were
significant for both average self-efficacy, F(2. 40) = 23.63, p < .01, and for
self-satisfaction, F(2. 40) = 11.67, p < .01. A priori contrasts revealed that
experts had significantly higher average self-efficacy judgments than non-
experts and non-experts had higher efficacy than novices (both ps < .05,
one-tailed). In terms of self-satisfaction, a priori contrasts revealed that
Table 3
Correlations among Metric Dependent Measures and Practice
Free-Throw Percentage
Variable 1 2 3 4 5
Self-efficacy before shooting 0.40**
Self-efficacy after misses 0.43** 0.85**
Self-efficacy after makes 0.48** 0.89** 0.90**
Practice free-throw % 0.51** 0.59** 0.70** 0.59**
** p < 0.01
experts and non-experts had significantly higher levels of satisfaction than
novice basketball players (both ps < .01), but the difference in self-satis-
faction between experts and non-experts was not significant (p > .05).
To examine the effect of expertise on the various types of self-efficacy,
analyses of variance were conducted. Significant effects were found for
self-efficacy beliefs before shooting any shots, F(2, 40) = 21.62, p < .01,
following successful shots, F(2,40) = 16.97, p < .01, and following misses,
F(2, 40) = 21.24, p <. 01. A priori contrasts revealed that experts had sig-
nificantly higher self-efficacy judgments than novices for all three mea-
sures (all ps < .05). In addition, experts surpassed non-experts in terms of
self-efficacy before shooting, and following two consecutive successful
shots (all ps < .05, one-tailed). The non-expert group had significantly
higher levels of self-efficacy before shooting, following two consecutive
successful shots, and two unsuccessful shots (all ps < .05, one-tailed) than
Correlation Analyses
Pearson coefficients among metric dependent variables and practice free-
throw statistics are presented in Table 3. All correlations were sizable,
positive in direction, and statistically significant.
Analyses of Attributions
A chi-square analysis was performed to assess group differences for
each of the 11 attribution categories following misses and successful shots.
Following two successful shots, there were no significant (p > .05) differ-
ences among expertise groups in their attributions. However, following
Table 4
Frequency of Attributions for each Expertise Group
after Two Unsuccessful Shots
Expertise Group
Attribution Experts Non-experts Novices
Specific technique 8 2 2
General technique 0 0 0
Confidence 0 0 0
Specific focus 0 1 0
General focus 2 2 0
Effort 2 0 0
Practice 0 0 3
Rhythm 1 4 2
Distractions 1 1 4
Don’t know 1 1 1
Other 0 2 3
two missed shots, there were significant differences among the three ex-
pertise groups for attributions to specific technique, chi-square (2) = 7.42,
p < .05, and attributions to practice, chi-square (2) = 6.02, p < .05 (see
Table 4). Experts attributed failure to specific techniques more often than
did non-experts or novices whereas no such difference was observed be-
tween the non-experts and novices. In addition, novices attributed their
free-throw misses to a lack of practice to a greater extent than experts or
Positive relationships were found between the boys’ type of attribution
for a missed shot and their choice of a corresponding strategy. For ex-
ample, attributions to a specific shooting technique were predictive (phi =
.38, p < .05) of choosing a specific technique strategy to improve the next
attempt, such as attributions of poor wrist follow-through would lead to a
wrist correction strategy. Similarly, attributions of misses to a lack of rhythm
were predictive (phi = .43, p < .05) of selecting rhythm enhancement strat-
egies, and attributions of misses to distractions were predictive (phi = .68,
p < .01) of selecting distraction reduction strategies. These correlations
accounted for 14%, 15%, and 45% of the variance, respectively. Since the
attribution question was asked prior to a follow-up strategy question dur-
ing free-throw shooting, the correlation coefficients can be interpreted as
attributions predicting strategy use.
Analyses of Goals
The frequencies of goals set before shooting the free-throws were ana-
lyzed using chi-square tests of independence across the three skill groups.
There were no significant (p > .05) differences between experts, non-ex-
perts, and novices when the goals were categorized individually. How-
ever, when the outcome-specific, process-specific, and focus-specific cat-
egories were collapsed into an inclusive specific goal category, there was a
significant difference among the three expertise groups, chi-square (2) =
7.05, p < .05. Experts (frequency = 11) set more specific goals than either
non-experts (frequency = 5) or novices (frequency = 4), but no differences
were observed between the latter two groups. Setting outcome-specific
goals was positively correlated with selecting a specific-technique oriented
strategy, phi = .44, p < .01. Conversely, setting general outcome goals is
predictive of selecting general technique oriented strategies, phi =.29, p =.
05. In addition, setting outcome-specific goals correlated significantly with
self-efficacy, point bierial = .48, p < .01
Analyses of Strategy Choice
The boys’ strategies before shooting free throws and following misses
are presented in Table 5. Chi-square tests assessed differences in strategy
use across the three expertise groups. The results showed that prior to shoot-
ing free-throws, being a member of a particular expertise group was re-
lated to choosing specific technique strategies, chi square (2) = 6.97, p <
.05. Experts chose more specific technique strategies than non-experts or
novices, but no differences were observed between non-expert and nov-
ices. In terms of strategies following two consecutive misses, the same
pattern held. Significant group differences were found in choice of spe-
cific technique strategies, chi-square (2) = 10.57, p < .01. Experts chose
more specific technique strategies than non-experts and novices, but the
latter two groups failed to differ significantly (p > .05). The results also
showed that novices differed significantly from both experts and non-
experts in terms of giving I don’t knowresponses (i.e., a non-strategic
approach), chi-square (2) = 6.02, p < 05; both of the latter groups reported
zeros for this non-strategic category.
There was a significant positive correlation between choice of specific-
technique strategies before shooting and self-efficacy before the task, point
biserial = .40, p < .01.
Table 5
Frequency of Strategies before Shooting Free Throws and after
Two Missed Free Throws for Each Expertise Group
Expertise Group
Experts Non-experts Novices
Strategies Before After Before After Before After
Free Missed Free Missed Free Missed
Throws Free Throws Free Throws Free
Throws Throws Throws
Specific technique 6 9 1 3 1 1
General technique 1 1 2 3 2 1
Visualization 1 0 0 0 0 1
Specific focus 1 1 1 0 0 1
General focus 4 3 3 4 4 4
Distractions 0 0 0 1 0 2
Rhythm 1 1 2 2 0 2
Don’t know 1 0 3 0 6 3
Other 0 1 1 0 2 1
Expert-Novice Differences
The present study compares basketball experts, non-experts, and nov-
ices for differences in self-regulation phase processes during self-directed
practice episodes. It does not address performance in group contexts, such
as trying out for a team or competing to make the starting lineup. Although
free throws are performed without interference or support from other play-
ers during both group and individual practice, they are undoubtedly more
difficult to regulate in group contexts. Competitive contexts introduce ad-
ditional variables that alter the forms of self-regulation that are needed,
and this precludes generalization of the present results to those settings.
In terms of forethought processes used during self-directed practice,
experts set significantly more specific free-throw goals than non-experts
and novices. Experts were more likely to state goals such as “to make 10
out of 10or “to keep my elbow in as I shoot” than the other two groups. In
addition, only 13% of the experts gave general outcome goals (i.e., “to
make them) or no goals; in comparison 53% of the non-experts gave gen-
eral outcome or no goals. A similar pattern emerged for strategy choice.
Experts selected more specific technique-related strategies to achieve their
goals than boys in the other two groups. Forty percent of the experts re-
ported technique-related strategies such as to bend my knees” and “to
follow through.” Only 8% of the non-experts chose a specific technique-
related strategy. In sum, experts appear to plan their practice sessions by
choosing more specific, technique-oriented processes than non-experts or
In terms of self-efficacy beliefs during forethought, experts exhibited
significantly higher self-efficacy perceptions at the outset of shooting free
throws and after successfully making these shots than non-experts or nov-
ices. That is, they felt more confident that they would be able to make two
consecutive shots in a row whether they were queried prior to or after suc-
cessfully making the shots. This makes sense because experts were de-
fined as individuals who shoot a higher free-throw percentage than non-
experts or novices.
Taken together, these findings suggest that non-experts’ lower self-effi-
cacy perceptions, nonspecific goals, and inefficient choice of learning strat-
egies impede their development as free-throw shooters. Experts appear to
be at a greater advantage for improving and sustaining high levels of skill
and motivation because their specific goals and strategies enable them to
focus on the essential form components. This is consistent with prior re-
search showing that when individuals set specific goals and select form-
related strategies, their level of performance and motivational processes
increased (Kitsantas & Zimmerman, 1998; Locke & Latham, 1990).
This study also assessed how the boys self-reflectively perceived and
reacted to their failed free-throw attempts and whether they attempted to
self-correct and adjust faulty processes. Following two consecutive misses,
a significantly greater number of experts than non-experts or novices at-
tributed their failure to faulty specific techniques (e.g., “I did not bend my
knees”). Fifty-three percent of the experts made this type of attribution, in
comparison to only 15% of the non-experts and 13% of novices. This attri-
bution pattern is highly beneficial because it reassures the learner that fu-
ture performances can be improved through the use of more appropriate
strategies. These attributional findings are consistent with McPherson’s
(1993) conclusion that learners’ choice of a strategy is dependent on their
level of declarative knowledge. Prior experimental research has shown that
making strategy attributions is significantly related to ones self-efficacy
level, level of satisfaction, and athletic skill (Kitsantas & Zimmerman,
It could be argued that differences in attribution as well as strategy choice
between three skill groups were due to experts’ greater knowledge of ver-
bal labels and skill in expressing basketball information than non-experts
or novices. It is well known (French, 1999) that experts have more do-
main-specific declarative knowledge than novices, as was found in the
present study. However, there are several findings that conflict with a ver-
bal knowledge interpretation of the present results. First, there were sig-
nificant differences in free-throw shooting knowledge between novices and
non-experts but no differences in verbal attributions to specific shooting
technique or in choice of specific shooting technique strategies (either be-
fore or after actual shooting). Second, there were differences between ex-
perts and non-experts on these same measures despite the absence of dif-
ferences in knowledge, and there were no differences between non-experts
and novices on these measures despite the presence of differences in knowl-
edge. Clearly, differences in basketball knowledge did not predict attribu-
tion or strategy verbalization patterns in the present study. Finally, it should
be noted that the verbal attribution and strategy choice differences among
the three expertise groups corresponded significantly to free-throw shoot-
ing skill during the study and during competitive play. These findings sug-
gest that qualitative differences in attributions and strategy choice were
predictive of free-throw performance.
Experts also adopted more specific, technique-oriented strategies fol-
lowing these two consecutive misses than boys in the other two groups.
Sixty percent of the experts indicated that they needed to focus on their
technique (i.e., “to keep my elbow in,” to follow through”) in order to
make the next shot. In contrast, only 20% of the non-experts and 7% of the
novices endorsed this type of strategy. Non-experts gave strategies related
to general technique (e.g., “to use good form) and general focus strategies
(e.g., “to concentrate) for a majority (i.e., 47%) of their responses. Such
strategies will not help an athlete to self-correct faulty techniques because
these strategies divert attention away from essential athletic form processes.
These differences between experts and non-experts in self-regulatory
processes are particularly informative because these two groups of boys
were similar across many background variables. The experts and non-ex-
perts were the same age, practiced the same number of hours per week,
played basketball for a similar number of years, and exhibited knowledge
of the same number and type of free-throw techniques. In terms of the
latter variable, the experts and non-experts reported that the most impor-
tant parts of a correct foul shot are the following: bending knees, pointing
feet towards the rim, following through, keeping elbow in toward the body,
and keeping the ball on fingertips. However, when asked how to achieve
one’s goals, why one missed a shot, or what one needs to do to improve,
very few non-experts utilized this knowledge base. On the other hand, ex-
perts were able to access and use this information to regulate their cogni-
tions and behaviors. These findings indicate that knowledge of essential
form components does not ensure that an athlete will use these techniques
in self-regulated manner. Historically, the advantage of experts has been
attributed to their sophisticated knowledge, but the inclusion of a non-ex-
pert group in the design shows that athletic knowledge is necessary but
insufficient to achieve the highest level of performance.
It could be asked whether significant differences in athletic knowledge
between the expert and non-experts may have emerged if a longer scale
were used. The measure of athletic knowledge involved naming three key
aspects of free-throw shooting. Although six possible answers were ac-
cepted, the scale was limited to four points (03) because of the instruc-
tions. However, there is little evidence in the present study to suggest that
increasing the length of the scale would have altered the conclusions. The
means for the expert (M =1.53) and non-expert group (M = 1.54) were
virtually identical and were located near the midpoint of the scale (1.50).
Clearly there was additional “ceilingspace available to distinguish these
two groups if true differences existed. This same four-point scale was shown
to be sufficiently sensitive to detect significant differences in athletic knowl-
edge between novices and both the experts and non-experts (M = 1.13).
Self-Regulation Model
Another important aspect of this study was the analysis of relations among
and between various self-regulatory processes within the forethought and
self-reflection phases. Within the forethought phase, all self-regulatory
processes were related significantly. For example, goal setting was corre-
lated with choice of strategy. Athletes who set outcome-specific goals (e.g.,
“to make 10 out of 10”) were more likely to select specific technique-
oriented strategies (e.g., “to follow through”), while those athletes setting
outcome-general goals (e.g., to make them”) were more likely to select
general technique strategies (e.g., “to concentrate on my form”). This im-
plies that training athletes to set specific goals can lead to their selection of
specific strategies designed to achieve those goals. Having specific goals
and strategies is essential when learning a skill because it directs the atten-
tion of the learner to basic elements of a task.
The goals and strategies that the players developed prior to shooting
free throws were also related to self-efficacy beliefs. For example, indi-
viduals who set specific outcome goals displayed higher levels of self-
efficacy. In addition, choice of technique-related strategies was predictive
of high levels of self-efficacy. Athletes who focus more on specific goals
and strategies displayed higher levels of confidence and enjoyment while
practicing the free-throws.
Turning to the self-reflection phase, we were interested in analyzing
how the boys’ attributions following two consecutive failed free-throw at-
tempts influenced their subsequent strategy choice. According to a cycli-
cal model of self-regulation, attributions are hypothesized to influence sub-
sequent strategy selection. In general, the results indicated that the attribu-
tions that participants made following their free throws were highly pre-
dictive of the strategy that they selected to improve future performances.
For example, the adolescent boys who attributed their failure to specific
techniques (i.e., I missed the last two shots because my elbow was going
to the left) were likely to select a specific technique oriented strategy to
improve their shooting accuracy (e.g., “I need to keep my elbow in”). Similar
attribution-strategy choice patterns were observed for rhythm (e.g., “I was
rushing my shots”) and distraction (e.g., “I was thinking too much) cat-
egories. These findings suggest that when practicing a motor skill, the boys’
attributions influenced their subsequent practice strategies. These findings
have implications for training athletes to practice motoric skills: Players
taught to attribute their failure to specific processes are more likely to fo-
cus on specific processes during subsequent practice sessions. Previous
training research has shown that focusing on form processes will lead to
more adaptive self-perceptions (self-efficacy, satisfaction) and higher lev-
els of motor skill (Kitsantas & Zimmerman, 1998). Novices and non-ex-
perts who have not mastered essential form components of specific sport
movements can clearly benefit from instruction regarding attributions for
the causes of their failure and self-corrective adjustments when they are
not performing well.
Unexpectedly, the measure of self-satisfaction was not related to any
type of attribution or strategy in the present study. This finding differs
from previous research (Kitsantas & Zimmerman, 1998) that reported strat-
egy attributions were positively related to satisfaction while ability and
effort attributions were negatively related to satisfaction. This finding may
be due to the more demanding self-evaluation standards used by experts
compared to non-experts (Zimmerman & Bandura, 1994). Although their
shooting accuracy was significantly lower, the non-experts were 80% sat-
isfied whereas experts were 68% satisfied with their performance. This
implies that self-evaluative standards should be assessed in conjunction
with self-sati