Windows to Research: Increasing Children’s Self–Esteem Through Supportiveness Training for Youth Sport Coaches
Adapted from Smoll, F.L., Smith, R.E., Barnett,
N.P., & Everett, J,J. (1993). Enhancement of Children’s Self-Esteem Through
Social Support Training for Youth Sport Coaches.
Journal of Applied Psychology, 78, 602-610.
Historically, few aspects of personality
have received greater theoretical and research attention than self-esteem.
Although self-evaluations may relate to specific areas of experience,
such as social, academic, and athletic domains, most theorists also subscribe
to a more encompassing construct of global self-esteem that refers to one’s
generalized sense of self-worth. This
construct appears to be plausible on both theoretical and research-driven grounds.
Traditionally,
society has viewed self-esteem to be a product of social interaction. This focus can be seen in the vast amount of
previous research that has been conducted on this topic within the family and
school settings. However, a less studied
social setting that may also influence a child’s developing self-esteem is organized
sports.
In this study,
eight male coaches of Little League Baseball players were trained in a 2 1/2-hour
preseason workshop designed to increase their supportiveness and instructional
effectiveness when interacting with the players on their respective teams (click here for a more detailed
description of the workshop coaching guidelines). As a comparison group, ten coaches from two
other leagues were not given the workshop.
To assess the effectiveness of the coach training, boys (N=152) from
all three leagues were interviewed in the preseason and postseason. In the interview, each boy filled out a questionnaire
that measured his general self-esteem. There were 14 descriptive statements on the questionnaire, each
of which was rated on a 4-point scale. Six
of the statements referred to positive attributes, whereas eight statements
were negative self-evaluations. The
entire questionnaire provides a maximum range of scores from 14 to 56 (click
here to see the self-esteem questionnaire).
______________________________________________________________________________________
Analysis
of Group Equivalence
The
mean ages of the coaches and the boys and the mean number of years of total
coaching experience is listed in Table 1. Statistical
analyses revealed that there were no statistically significant group differences
between the experimental and control coaches for any of these three demographic
variables. Likewise, after collecting
pre and postseason data for 62 boys in the experimental group and 90 boys in
the control group (72% and 78% respectively), analyses revealed no statistically
significant difference in age between boys in the experimental and control groups.
Table 1. Demographics
of Age and Coaching Experience for Coaches and Age for Boys.
|
Overall Mean Ages Coaches (years) |
40.00 (SD=6.77) |
|
Overall Mean Coaching experience |
8.89 (SD=4.64) |
|
Overall Mean Ages Little League Boys (years) |
11.39 (SD=.81) |
______________________________________________________________________________________
Self-Esteem
Analyses
Analyses of the self-esteem
measures, both preseason and postseason, were conducted for 147 boys because
five boys in the original sample did not provide data at both times. The preseason self-esteem questionnaire yielded
nearly identical means of 47.80 (SD=4.73) for the experimental group (n=59)
and 47.76 (SD= 5.76) for the control groups (n=88). The postseason means for these two groups were
48.71 (SD=5.43) for the experimental group and 47.40 (SD=6.60) for the control
groups. Statistical analyses revealed
no significant differences between the two groups in postseason means.
In addition, when postseason means were adjusted using preseason scores
as a covariant (thereby removing all effects of any preseason differences),
the postseason differences remained statistically non-significant.
The mean preseason and postseason self-esteem scores for all samples
of boys are shown in Figure 1.
Figure 1.
Mean preseason and postseason self-esteem scores of all boys who played
for trained and control-group coaches. ______________________________________________________________________________________
Although,
it was possible that the coach training intervention had no real effect, previous
evidence led the researchers to hypothesize that low self-esteem children might
be more responsive to variations in coaching behaviors. The previous analyses
included boys of all varying levels of self-esteem.
The researchers conducted additional statistical analyses in search of
potential treatment effects for the 86 boys whose preseason self-esteem scores
were below the median value of 49. The
mean preseason and postseason self-esteem scores for this subsample of boys
are shown in Figure 2.
Figure
2. Mean preseason and postseason self-esteem scores
of boys with low self-esteem who played for trained and control-group coaches.
Similar to the larger sample of all boys, in this subsample there were no significant differences on the preseason self-esteem measure between the experimental group and the control group. However, when postseason means were again adjusted using preseason as a covariate, significantly higher level of self-esteem were revealed for the boys who had played for the trained coaches (F(1,84)= 4.56, p<.05). Further analysis of these two groups revealed that scores for the low self-esteem boys who had played for the trained coaches increased significantly. However, scores for the low self-esteem boys in the control group showed no significant change.
At the end of the season the researchers
also asked the athletes to evaluate their experiences in Little League. There
were no differences between the ratings for boys who played for trained and
control coaches for 1) liking for baseball or 2) their coach's baseball knowledge.
However, the boys who played for trained coaches rated items such as
1) their fun playing baseball, 2) their liking for their coach, 3) their coach's
teaching ability, and 4) how well they liked their teammates significantly higher
than did the boys who played for control coaches. Additionally, because the
two groups were from different leagues, there was no significant difference
in their average won/loss records.
The researchers concluded that the workshops were a successful
addition to the preseason preparation for Little League Baseball coaches.
1. What
type of experiment did the researchers conduct?
2. What was the dependent variable? The independent variables?
3. The researchers took advantage of intact groups - the different little leagues.
What steps did the researchers take to show that there were not any major differences
between these intact groups at the start of the experiment? Intact groups
have the disadvantage of building possible selection confounds into an experiment.
Are there other factors that might vary systematically with the different leagues
that could be possible confounding variables?
4. There were some advantages of using intact groups (separate
leagues) rather than randomly assigning coaches to the control group or experimental
group for this experiment. What were these advantages?
5. Suppose that you serve on the Board of Directors for your local Little League and would like to convince the other board members and the volunteer coaches that the Board should sponsor a preseason workshop designed to help coaches prepare to work effectively with young athletes. How you could take advantage of the results of this applied youth-sport study to support your position?