Proportional Odds assumption χ2 = 7.95 p-value = .44
Measures of fit
|
LRT (df) a |
26.77 (4) |
p-value |
Log- | ||
|
.000 |
likelihood |
-789.8 | |||
|
AIC |
2.364 |
AIC*n |
1593.6 |
Number obs. |
674 |
|
BIC b |
-2764.73 |
BIC’ |
-0.721 | ||
a The likelihood ratio test is performed against the log-likelihood of the intercept-only model. In brackets are
degrees of freedom.
b The BIC criterion is more adequate for comparing models than just per se. The value of the statistic is reported
here for information purposes. It was used for comparison between models.
Enjoyment with participation, as a relevant predictor of the telephone-based channel
usage, reflects the pleasure that participating has on a self-service option usage. As a result,
H4 is supported. More specifically, the odds of ‘most days’ vs. less frequent usage are 20%
higher for those who experience a marginal change in their ‘enjoyment with participation’,
that is, for those who prefer to participate and enjoy being involved in the course of action.
Concerning cognitive influences on usage behavior, H7 was supported by the data. As
such, the odds of more frequent usage of this channel increase for a marginal increase of
perceived service quality as a whole, as well as increasing for marginal increases of perceived
service quality of technology-based channels and of the human-based channel. For the
telephone-based channel, perceived service quality of the human-based channel consisted of
items related to employees’ evaluation. This result can be explained when interpreting the
telephone-based channel as also comprising telephone calls requesting the service directly to
the account branch: this fact was acknowledged by comments made by telephone-based
channel users in their questionnaires. Perceived service quality of technology-based channels
also positively influenced the odds of a more frequent usage of the telephone-based channel:
this finding is in line with the fact that this channel may involve both person-to-person
(mediated by a technological device) and person-to-technology contacts. From the initial set
of hypotheses, enjoyment with technology, preference for technological interfaces, desire for
control, perceived communication efforts and perceived risk also emerged as factors although
not as relevant constructs for explaining usage frequency. As such, H1, H2, H3, H6 and H8
were not supported.
20