Measuring service experience: Applying the satisfaction with travel scale in public transport

How satisfying is the Scale for Travel Satisfaction?

“…Although our results show clear differences between transport modes in the structure of STS, and differ from previous studies analysing STS, all reliability tests (both in this study and in previous studies Olsson et al, 2012)) indicate that the adjective pairs confident/worried and especially alert/tired do not fully fit the affective dimension of STS, irrespective of whether this dimension is divided in two sub-dimensions or not. Although combining all six items related to emotions into a single dimension is a satisfactory option that produces high levels of internal consistency for our data, it could be interesting to replace alert/tired and confident/worried with alternative adjective pairs that are more successful in capturing the positive activation/negative deactivation and positive deactivation/negative activation dimensions proposed by the core affect approach (Russell, 1980(Russell, , 2003.…”

“…Previous studies on the STS also suggest that the alert/tired and confident/worried items do not relate to the STS's underlying dimensions as strongly as other affective items; however, the differences appear to be more pronounced for the Ghent data. Friman et al (2013) and Olsson et al (2012) use structural equation models (confirmatory factor analyses) to assess the psychometric properties of the STS. In most cases path coefficients from confident/worried and especially alert/tired to the accompanying affective STS dimensions are lower than path coefficients from the other items, even although the precise results differ between trips to and from work and according to the transport mode used and the city where data have been collected .…”

“…In most cases path coefficients from confident/worried and especially alert/tired to the accompanying affective STS dimensions are lower than path coefficients from the other items, even although the precise results differ between trips to and from work and according to the transport mode used and the city where data have been collected . In this context it should also be noted that Olsson et al (2012) deleted the items alert/tired and confident/worried from their analysis because these items had fairly high factor loadings (>0.3) on more than one of the three factors that capture the three dimensions of STS.…”

“…Recently, Ettema et al (2013) (using 256 Dutch car drivers), Friman et al (2013) (using 951 residents from Stockholm, Göteborg and Malmö (Sweden)) and Olsson et al (2012) (using 1000 public transport users from Karlstad and Göteborg (Sweden)) have tested the reliability of the STS with two affective and one cognitive dimension, using values of Cronbach's alphas and structural equation modelling (i.e., confirmatory factor analyses). These studies state that STS consists of three underlying dimensions.…”

“…These studies state that STS consists of three underlying dimensions. Although Ettema et al (2013) and Friman et al (2013) state that travel satisfaction is measured adequately by the aforementioned nine items, Olsson et al (2012) suggests excluding two of the nine items from STS. Friman et al (2013) have also analysed whether the structure of STS's underlying dimensions varies according to transport mode used, and their analysis on data from Stockholm, Göteborg and Malmö suggests this not to be the case.…”