ERIC EJ750841: Behavioral Interventions to Improve Performance in Collegiate Football

2006·0.19 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2006, 39, 385–391 NUMBER3 (FALL2006)
BEHAVIORAL INTERVENTIONS TO IMPROVE PERFORMANCE
IN COLLEGIATE FOOTBALL
SHANNON L. SMITH
UNIVERSITYOFNEBRASKA
AND
PHILLIP WARD
THEOHIOSTATEUNIVERSITY
Usingamultitreatmentwithdrawaldesign,thisstudyevaluatedthedifferentialeffectsofpublicly
posted plus verbal feedback, goal setting plus verbal feedback, and publicly posted feedback,
verbalfeedback,andgoalsettingtogetherontheperformanceof3collegiatefootballplayersin
practice scrimmages. Also assessed was whether the changes in practice behavior generalized to
games. The dependent variables were performances on three wide receiver skills. The results
show that public posting with verbal feedback, goal setting, and public posting with verbal
feedbackandgoalsettingwereeffectiveinimprovingplayerperformancetoa90%criterionlevel
during practice,andthese changes generalized togame performance.
DESCRIPTORS: coachinginterventions, sports
_______________________________________________________________________________
Two commonplace interventions in sports privately; (b) when it is tied to comparison of
are feedback and goal setting (Martin, Thomp- an individual’s previous performance rather
son, & Regehr, 2004). Feedback occurs when than to another person’s performance; and
an individual (e.g., self-charting), others (e.g., (c) when it is compared to a standard (Alvero
peers and coaches), or equipment (e.g., heart- etal.).Goalsettingisstrengthenedwhen(a)the
rate monitor, photoelectric beam) provide goals are made public, and (b) the goals
information about some aspect of performance emphasize short-term immediate outcomes
(Lee, Nyity, & McGill, 1993). Goal setting (Locke & Latham).
occurs when a performance standard is estab- Todate,therehasbeennodirectcomparison
lished by the athlete or by a coach (Locke & of these components reported in sport settings.
Latham, 1990). The purpose of this investigation was twofold:
Reviews of feedback (Alvero, Bucklin, & first, to examine the relative effects of three
Austin, 2001; Balacazar, Hopkins, & Suarez, interventions (i.e., goal setting plus verbal
1986)andgoalsetting(Locke&Latham,1990) feedback; public posting plus verbal feedback;
concludethat there is considerable variability in and goal setting, public posting, and verbal
the efficacy of each when used alone. These feedback together) on the practice performance
same reviews note that there are stronger effects of collegiate football players, and second, to
when feedback and goal setting are combined determinetheextenttowhichtheseeffectsshow
with other strategies. Feedback appears to be generality to game settings.
most effective (a) when it is presented both
publicly and privately, rather than publicly or
METHOD
Address correspondence to Phillip Ward, The Ohio Participants and Setting
State University, Sport and Exercise Education, 305 The participants were three wide receivers of
Pomerene Hall, 1760 Neil Ave., Columbus, Ohio
a National Association of Intercollegiate Ath-
43210(e-mail: [email protected]).
doi:10.1901/jaba.2006.5-06 leticsDivision2footballteamata4-yearliberal
385
386 SHANNON L. SMITH and PHILLIP WARD
arts college in the midwest. At the beginning of sessions, the first 10 releases, blocks, and routes
the season, the team was ranked third of the performed by each player were recorded for
seven teams in their conference. The wide analysis.Datawerecollectedbyvideotape.Each
receivers’ coach selected players that he rated as performance of a release, block, or a route was
frequently demonstrating poor execution of coded as either correct or incorrect, and the
wide receiver skills during practice and game number of correct of trials performed was
play. Dave, Alex, and Mike consistently played converted to a percentage.
in games throughout the season, but were not
classified as starters at the beginning of the Experimental Design and Procedure
season (i.e., did not start play every game). An ABACABC multitreatment withdrawal
Practice sessions were held 3 days per week for design wasusedto assess theeffectivenessof the
2 hr per day. During that time, approximately interventions.
45 min were spent on the skills analyzed in this Baseline. During baseline the players met
study. The practice sessions were conducted on with the coach, reviewed expectations for each
the college football field. Games against other skill, and then proceeded to practice. While
teams in the same division were held weekly. practicing, players received verbal feedback and
error correction from the coach.
Data Collection Public posting plus verbal feedback. In this
Three dependent variables were measured. condition, players were first informed of their
First, the percentage of correct blocks was used mean percentage correct performance for each
to determine whether the wide receiver blocked skill during the previous baseline session and
effectively. A block was defined as correct (i.e., how the percentage was calculated. They were
effective) if the wide receiver checked the informed that the results of the coming day’s
progress of a defending player using legal practice and each successive practice would be
techniques (e.g., not holding or pushing the posted on a daily performance chart. The chart
defender’s back). Second, the percentage of was located on the door leading to the locker
correct routes run was used to determine room. Data were posted on the chart prior to
whether the wide receiver ran a predetermined the next practice session. During practice
pathduring anoffensive play.The routerun by sessions, players continued to receive feedback
the player was compared to the coach’s and error correction from the coach. The chart
playbook to determine if it was correctly was removed at the end of this condition.
performed. The playbook described set routes Goal setting plus verbal feedback. The condi-
tobeperformedbythewidereceivercontingent tion began with the lead investigator reviewing
on the defense used by the defending team. the mean percentage correct performance for
Thus, the route was defined as correct if the each skill during the previous baseline session.
wide receiver ran the route as described in the Because the dependent variables were skills that
playbook for the correct defense. Third, the the coaches expected these players to have in
percentage of correct releases from the line of their repertoire (as opposed to learning new
scrimmage was used to determine whether the skills), there was an expectation that players
wide receiver released freely from the line of would meet a 90% correct performance in
scrimmage. If the receiver avoided contact practice scrimmage. Previous studies in similar
initiated by a defender and began running the settings have shown that 90% correct perfor-
predetermined route, the release was coded as mance is a reasonable expectation (Ward &
correct or free. Because of variability in the Carnes, 2002; Ward, Smith, & Sharpe, 1997).
number of opportunities that were afforded the Prior to each session, the investigator met with
different players during practice and games each player individually and informed him of
COLLEGIATE FOOTBALL PERFORMANCE 387
his performance during the previous session. Interobserver agreement was distributed
He also reminded each player of the 90% equally across phases and assessed during 50%
criterion expectation and either encouraged of practice sessions and 67% of games using
him to meet it or congratulated him on having point-by-point agreement. Percentage agree-
met the criterion in the previous session. ment was calculated by dividing the total
During practice sessions, players continued to number of agreements by the total number of
receive feedback and error correction from the agreements plus disagreements and multiplying
coach. by 100%. The mean agreement was 96%. The
Public posting plus verbal feedback plus means and ranges were, for releases, 97%
goal setting. This condition combined the (range, 92% to 100%); for blocks, 95% (range,
procedures of the two previous interventions. 93% to 100%); and for routes, 97% (range,
Thechartwasreintroduced,andtheexpectation 90% to 100%).
of meeting or exceeding the 90% criterion was
discussed individually prior to each practice Social Validity
session. As before, during practice sessions Following the conclusion of the study, each
players continued to receive feedback and error of the participants and the head coach were
correction from the coach. asked to complete an open-ended question-
The team coaches were not informed of the naire. Feedback was sought from the players to
nature of the conditions and baselines, or when allow researchers (a) to assess the acceptability
they would be phased into the practice of the three interventions and their effects,
schedule. Coaches did see the chart during (b) to detect other effects not discernible from
public posting plus verbal feedback. They did thedata(includingundesiredeffects),and(c)to
not have access to the performance data at any adjust the procedures in future studies. The
othertime. To control for variance in feedback, questionnaire was completed anonymously.
we randomly tracked six sessions to see if
coaches spent more time with players in one
RESULTS AND DISCUSSION
condition or another. There were minor
deviations from condition to condition, but Figures 1 through 3 show the percentage of
most often these had to do with nontarget correct performances in practices and games for
behavior issues (e.g., learning a new play, the three dependent variables for each player.
injuries); they occurred in a similar manner The data are remarkably similar among the
across all players. players across skills. During baseline, perfor-
mances ranged between 50% and 80% correct,
Interobserver Agreement except for Dave’s release performance for Day
An independent observer recorded each 10, which was 90%. Correct performance
player’s performance of releases, blocks, and during games ranged between 60% and 80%.
routes to determine the reliability of the During public posting plus verbal feedback,
observational data. The observer, an experi- correct performance ranged between 80% and
enced football coach, was not directly involved 100%, and correct performance during games
with the study’s implementation or coaching ranged between 90% and 100%. Baseline levels
the wide receivers. The reliability observer had wererecoveredforallbehaviorsduringpractices
beentrainedpriortothestartofthestudyusing and games.
directobservationandvideorecordings.Hemet Performances were better during goal setting
or exceeded a criterion of 95% agreement with plus verbal feedback than during public posting
the primary investigator on three 45-min with verbal feedback for at least one of the
videotapes. dependentmeasuresofeachparticipant,andthe
388 SHANNON L. SMITH and PHILLIP WARD
Figure 1. The percentages ofblocks, routes,andreleases during practice andgamesessions for Dave.
resultsweresimilarforthetwoconditionsonthe part because all participants had high initial
other dependent measures. Baseline levels were baselines and because the order of treatment
recovered for all behaviors during practices; introduction was the same across all conditions.
however, the game data remained at least 10% There was also some evidence for an improving
higherthanbaselinelevels.Duringthecombined trend during public posting plus verbal feed-
phase levels, all practice performances were back for some of the data paths. The findings
consistently between 90% and 100% correct do show, however, that each treatment was
and all game performances were 100% correct. better than baseline, and this finding supports
It is not possible to draw conclusions about the use of goal setting and public posting with
the relative effectiveness of the treatments, in verbal feedback in sport. The social validation
COLLEGIATE FOOTBALL PERFORMANCE 389
Figure 2. The percentages ofblocks, routes,andreleases during practice andgamesessions for Mike.
questionnaires indicated that players and coa- tervention procedures but also the effects of the
chespreferred thecombinedinterventionto the intervention.
other two interventions and to baseline. Players A third contribution of this study is that
reported that individual goal setting was the dependent measures were defined in terms of
least preferred intervention because of the outcomes rather than topography alone. The
absence of visual feedback. A second contribu- topography of the performance was determined
tion of this study to the literature is the either by comparing it to the playbook (e.g.,
demonstration of generalization from practice running a pattern in response in a specific
settings to game settings. The strength of this offense or defense) or according to the legal
transfer serves to validate not only the in- rules of play (e.g., tackling). In addition, the
390 SHANNON L. SMITH and PHILLIP WARD
Figure 3. The percentages ofblocks, routes,andreleases during practice andgamesessions for Alex.
successful outcome of the performance was have suggested that publicly posted feedback
included in the definition. In sporting contexts, may have multiple functions (e.g., discrimina-
both outcomes and topography are important. tive, reinforcing, and punishing) relative to the
However, the outcome is also a function of the particular context (Alvero et al., 2001; Duncan
play selection by the coach or quarterback and & Bruwelheide, 1986; Van Houten, 1980).
performances of the other team. Thus, care Giventheimmediacyofthechangebetweenthe
mustbetakentodefinetheoutcomeintermsof interventions and baseline, the results are
the initial success. unlikely to be explained in terms of learning.
The specific behavioral processes that oper- It seems more likely that the results were
ated in this study are unclear. Some authors influenced by motivational factors. In particu-
COLLEGIATE FOOTBALL PERFORMANCE 391
lar, we hypothesize that these are demonstra- Duncan, P., & Bruwelheide, L. (1986). Feedback: Use
andpossiblebehaviorfunctions.JournalofOrganiza-
tions of motivating operations (Laraway, Sny-
tional BehaviorManagement,7,91–114.
cerski, Michael, & Poling, 2003). However,
Laraway, S., Snycerski, S., Michael, J., & Poling, A.
more tightly controlled research will be neces- (2003). Motivating operations and terms to describe
sary to test this hypothesis. them: Some further refinements. Journal of Applied
Behavior Analysis,36,407–414.
The problem addressed in this study of
Lee, A., Nyity, C., & McGill, R. (1993). Instructional
assuring consistent high-quality performances effects of teacher feedback in physical education.
over the course of a season is a common JournalofTeachinginPhysicalEducation,3,228–243.
challenge to coaches and athletes. Although the Locke, E. A., & Latham, G. P. (1990). A theory of goal
setting and task performance. Englewood Cliffs, NJ:
gains reported in this and similar studies (e.g.,
Prentice Hall.
Ward & Carnes, 2002; Ward et al., 1997) are Martin,G.,Thompson,K.,&Regehr,K.(2004).Studies
small relative to baseline measures of 70% to using single-subject designs in sport psychology:
30 years of research. The Behavior Analyst, 27,
80%correct,theconsequencesofperformingat
263–280.
levels below the treatment criterion of 90%
Van Houten, R. (1980). Learning through feedback: A
include missing an opportunity to start on the systematic approach for improving academic perfor-
team or to receive scholarships for student mance.New York:Human Sciences Press.
Ward,P.,&Carnes,M.(2002).Effectsofpostingself-set
athletes.
goals on collegiate football players’ skill execution
during practices and games. Journal of Applied
Behavior Analysis,35,1–12.
REFERENCES
Ward,P.,Smith,S.,&Sharpe,T.(1997).The effectsof
Alvero,A.,Bucklin,B.,&Austin,J.(2001).Anobjective accountability on task accomplishment in collegiate
reviewoftheeffectivenessandessentialcharacteristics football.JournalofTeachinginPhysicalEducation,17,
of performance feedback in organizational settings 40–51.
(1985–1998). Journal of Organizational Behavior
Management,21,3–30.
Balcazar,F.,Hopkins,B.,&Suarez,Y.(1986).Acritical Received January 25,2006
objective review of performance feedback. Journal of Final acceptanceMarch 30, 2006
Organizational Behavior Management,7, 65–89. Action Editor,Ron VanHouten