12
Wallhead and Ntoumanis
Table 2 Cronbach’s Alpha Coefficients (M ± SD) for Pre- and
Postintervention Measured Dependent Variables for Both Programs
Sport Education Traditional
|
Dependent |
(n = 26) |
(n = 25) |
a | |||
|
M |
SD |
M |
SD | |||
|
Enjoyment |
Pre |
4.51 |
1.21 |
5.25 |
0.93 |
0.73 |
|
Post |
5.60** |
0.75 |
5.13 |
0.89 |
0.81 | |
|
Effort |
Pre |
5.14 |
0.94 |
5.62 |
0.74 |
0.76 |
|
Post |
5.73** |
1.03 |
5.45 |
1.03 |
0.70 | |
|
Perceived competence |
Pre |
4.82 |
0.84 |
4.76 |
1.30 |
0.70 |
|
Post |
5.45 |
1.04 |
4.81 |
1.15 |
0.79 | |
|
Task goal orientation |
Pre |
3.72 |
0.53 |
4.09 |
0.56 |
0.73 |
|
Post |
4.11 |
0.61 |
3.96 |
0.59 |
0.82 | |
|
Ego goal orientation |
Pre |
2.84 |
0.70 |
2.57 |
1.04 |
0.84 |
|
Post |
2.50 |
0.79 |
2.31 |
0.82 |
0.88 | |
|
Task climate |
Pre |
3.74 |
0.51 |
3.97 |
0.62 |
0.72 |
|
Post |
3.91 |
0.47 |
3.71 |
0.68 |
0.70 | |
|
Ego climate |
Pre |
2.98 |
0.46 |
3.01 |
0.64 |
0.61 |
|
Post |
2.98 |
0.44 |
2.87 |
0.41 |
0.65 | |
|
Perceived autonomy |
Pre |
6.93 |
3.08 |
4.91 |
7.44 |
0.81 |
|
Post |
7.63 |
2.92 |
4.63 |
5.74 |
0.73 | |
** p < .006.
autonomy were each entered into a regression model in the first block of the analy-
sis to control for the initial level of these variables. In the second block,
postintervention goal orientations, perception of the motivational climate, and per-
ceived autonomy were entered as independent variables into the model.
Results
Preliminary Analyses
Cronbach alpha coefficients, means, and standard deviations for all mea-
sures are displayed in Table 2. The alpha coefficients for all measures were deemed
acceptable based on Nunnally’s (1978) cutoff criterion of .70 for the psychological
domain with the exception of ego-involving climate (pre α = .61, post α = .65).
However, due to the importance of this measure for this study, the subscale was
retained.
According to Kenny and La Voie (1985), in order to determine whether the
individual or the group should be used as the unit of analysis, a test of non-
independence of individual observations must be performed. The calculation of
an intraclass correlation coefficient (ICC) provides evidence of the degree of vari-
ance in responses among members of a group in relation to the responses of
nongroup members (Zhang, Hausenblas, Barkouras, & Pease, 2002). Intraclass