Structural model estimation
To test the hypothesized relationships, a full structural model was estimated
using each scale item as an indicator of its associated latent variable
construct. Fan association and sponsorship recognition were modeled as
manifest variables. Table II presents the correlations between the
constructs. The causal paths between the latent variables were specified in
accordance with Figure 1 and the model was tested using the maximum
likelihood method of parameter estimation. We find strong support for
drawing the conclusion that the model fits the data well. The chi-square
statistic for the structural model is 809 with 222 degrees of freedom
(p < 0.0001). The GFI is 0.92 and the AGFI is 0.90, while the CFI and NNFI
are both 0.95. RMSEA estimate is equal to 0.0567. Table III provides a
summary of the standardized estimates and t-values for the seven
hypothesized paths. All seven of the hypothesized paths are supported
(p < 0.001).