The final sample was divided depending on the travel motive, getting three sub-samples,
because the behaviour changes in each case, being necesary the estimation of a function of
demand for motive; the first sub-sample includes travels made by job motive, the second,
the travels made by study motives and the third one the displacements by shopping and
leisure, the rest of motives were rejected because they do not present homogeneous
behaviour.
Then with these facts we built the first sample called “non-restricted 1”, which includes 331
observations of which the modal distribution is of 37-63 between bus and car. Then we
chose the first of the travels for each person, to be only included once in the sample
independiently of the number of travels made by the same motive, because it means that
everyone has the same information, in this way, we built the sample called “non-restricted
2”, in this second sample, the sample size is of 262 observations, being the modal
distribution similar to the before one. From a theoretical point of view, to apply the model,
is necessary that all the individuals have access to all the alternatives, to make maxim his
utility in the choice of transport for this, we avoided all the individuals who had not
driving-licence and for this reason, they depended on the urban transport to get the
“restricted” sample that is formed by 192 observation of which 81% corresponds to travles
made in private transport and 19% to urban transport. This final sample, although is the
most correct theoretically, presents the inconvinient of the slant in the information about
urban transport, for this reason we decided to apply the model to the three samples.
4.- The model and its explicatives variables.
The explicative variables chosen to esteem the function of demand and that picks up as
attributes of the ways of transport as the individual socioeconomic caracteristics are these:
1.- OFFER VARIABLES
Cost. We consider that the individual chooses the transport looking at the relative price and
not the absolute price, for this reason, this variable is calculated by the difference of price
or cost between private and bus. To calculate this variable we esteemed firstly the whole
cost of each alternative, then for the car we included, facts like fuel, lubricants, tyre