Raimund K. Herz

The Use of the Bicycle

Homepage Manual Contents Navigation

CATEGORY ANALYSIS OF BICYCLE USE

Cluster analysis of individuals with respect to their daily activity programmes and travel behaviour¹¹ has shown that beyond half a dozen of clusters any further subdivision will only slightly reduce the percentage of behavioural variation explained by this subgrouping. The basic subgroups of the population are determined by the individual's stage in the life cycle and modified by the availability of a motor-car. From these empirical findings and compromising between homogeneous target group definition on the one side and the expected homogeneous behaviour on the other side, ten segments of the population were chosen as the basic analytical units.

They are confined to the German population over ten years of age. Soldiers and foreigners were excluded from the survey. The ten segments chosen for analysis are not exhaustive. The group of economically inactive persons under 60 years of ace who are neither housewives nor schoolchildren and apprentices are left out of this analysis, that is the group of students, unemployed and premature annuitants, approximately five percent of the sample. Seventeen percent fall into the category of schoolchildren and apprentices (from 10 to 18 years of age). 46 percent are economically active. 20 percent housewives and 12 percent elderly. i.e. older than 60 years and not economically active. Boys and girls are almost evenly distributed among the group of schoolchildren and apprentices, while the sex-proportion among economically active persons is 68 percent men and 32 percent women. There is only a neglegible number of men among the category housewife, and women also prevail with six to four among the elderly.

Adult persons were further subdivided according to car availability in the household. Eighty-seven percent of the, employed men and 84 percent of the employed women come from households with at least one car. The motorization of the housewives' household amounts to 79 percent. The lowest figure is 50 percent for the elderly. It is interesting to note that in Germany the number of bicycles always was and still is greater than the number of motor-cars. However. because there are not just one but on an average two bicycles in the household. the number of households without a bicycle is now larger than those without a car.

The overall data base of this analysis consists of almost 150,000 week-day trips of 44.000 persons throughout the year 1976.

The extent to which the bicycle is used by these subgroups of the population on week-days can be seen from Table I. For comparison, figures for the other modes of transport are given as well, although they cannot be analysed in detail within the scope of this paper. The percentages relate to the trip rate given in the last column. In order to calculate daily trip volumes, the number of persons who stay at home would have to be considered. The average week-day "at home ratio" is 10 percent of schoolchildren and apprentices (including holidays). 9 percent of economically active persons, 26 percent of housewives and 35 percent of the elderly.

The use of the bicycle ranges from less than 4 to over 33 percent of all trips made by the particular group. Of course, schoolchildren and apprentices rank highest. Very few of them are old enough to drive a car. The effect of car availability in the household on the use of the bicycle seems to be obvious, particularly for employed men. However, there are certainly other co-varying factors, especially for the relatively small group of employed persons who come from households without a car. The effect of car availability on bicycle use seems to be much smaller for housewives and indeed insignificant for the elderly.

TABLE I Modal structure of week-day trips by population subgroups

From looking at the figures for boys and girls as well as for employed men and women from households without car, it becomes apparent that emancipation has not reached bicycle use. Instead of cycling. women prefer to walk and to use public passenger transport. Is it still not ladylike to cycle?

Further insights may be gained by looking at the trip purposes. In Table II corresponding figures are given for duty and other trips. Duty-trips are defined as fixed in time and space and pursued with some regularity: school trips for schoolchildren and apprentices, work trips for the economically active, and, to a smaller degree, shopping and related trips for housewives and the elderly. Apparently the attitudes toward the bicycle differ between youngsters and adults. Schoolchildren and apprentices use their bicycles to a larger extent for leisure trips than for school trips. On the other side, adults seem to use the bicycle to a larger extent for obligatory trips than for others, which are more flexible in character.

However, there might be another, even stronger determinant for this behaviour, namely the distance between the places of activity. It is well known that bicycle trip lengths rarely (less than ten percent), exceed five kilometers, and only five percent are longer than seven kilometers. The reform of the German school system in the late sixties has led to larger units, which moved into greater distance from the pupils' homes. Especially in the rural areas they moved out of the reach of the bicycle. Public passenger transport and school buses had to bridge the gap. This effect can be nicely seen from the median trip length of school trips on Table III, which is probably the main reason for not using the bicycle more often for this purpose.

The median bicycle trip length for school and work is longer than for other purposes, and of course longer than for trips by foot. On an average, motorized trips are much longer, although one should not forget that roughly 20 percent of all trips by motor-car are shorter than two kilometers, 30 percent shorter than three kilometers and 40 percent shorter than four kilometers. Therefore, a considerable proportion of these trips is well within the range of the bicycle. To that extent they may be substituted by bike trips depends on many other factors, such as convenience and overall structure of the journey. Very often. shorter trips are links of complicated trip chains for which the mode of transport cannot be so easily changed, in particular not to the bicycle mode. Nevertheless, the fact that most car driver trips, except those of employed men, are shorter than five kilometers, can be regarded as a latent potential for bicycle use. Another potential for cycling is women. Their median bicycle trip lengths fall short at the 1.8 kilometers, which is 800 meters shorter than those of children and men, and even shorter than those of the elderly. This fact cannot be just the result of their activity patterns nor given by nature.

TABLE II Trip purpose and modal structure of week-day trips by population subgroups

TABLE III Median Trip length of work day trips by population subgroups for different modes of transport and trip purposes

How much of this potential may be realized will be discussed in the final part of this paper. This section shall be closed with another noteworthy phenomenon: shopping and related trips of housewives and the elderly made by bike are shorter than their other trips by bike. Maybe one reason for this is the inconvenience of transporting goods by bicycle, although this phenomenon holds for almost all modes of transport.

From previous analyses one would expect some variations of bicycle use along with personal characteristics of the individual such as age and education, his or her technical capability to move such as the effective availability of a motor-car, and the quality of the residential environment such as home tenure and size of the town, which is also related to the quality of the supply with public transport services. Furthermore, the topographical situation probably has a very strong effect on bicycle use, and cycling is certainly the mode of transport which is most seriously affected by weather conditions. So, subsequently the influence of each of these variables on bicycle use by the above population subgroups for duty and other trips will be analysed. Needless to say in this analysis factors other than those used for defining the subgroups are not controlled. Thus the usual caveat seems to be appropriate as to whether the ceteris paribus clause holds.

In this analysis, the effects on bicycle use are related to the travel behaviour irrespective of mode of transport. One possible kind of measurement would be Student's t-statistic for the difference of the percentages between bicycle trips and overall trips for a given attribute. However. for the sake of simplicity another measure is used here, which exhibits the same property of symmetry. albeit around the value of 1.0. A factor of 1.0 means, for example, that the percentage of bicycle trips in a given category, such as a village of less than 2000 inhabitants, is the same as the percentage of all trips there in relation to the rest of town size categories. A factor of plus 2.0 means that the share of bicycle trips is twice as much, and a factor of minus 2.0 means that it is only half of it, i.e. must be divided by 2.0. So, the minus sign actually stands for EXP(-1). Keeping this maybe a little unconventional rule of measurement in mind, let us look: through Tables IV and V in order to analyse the univariate effects of those variables on bicycle use for duty and other trips. For avoiding biases, factors for less than 20 entries. i.e. bicycle trips. have been removed from the tables.

TABLE IV Factors influencing the use of the bicycle for week-day daily Trips by population subgroups

TABLE V Factors influencing the use of bicycles for week-day non-duty trips by population subgroups

• AGE

  The young adults apparently do not prefer cycling. The use of the bicycle is slightly increasing in the middle age-group and decreasing toward the sixties, especially for women. This holds for duty and other trips as well.

• EDUCATION

  Individuals who have not received any formal education beyond elementary school, generally use the bicycle to a relatively large extent for all kinds of trips. This tendency is even stronger for persons who come from households where a car is available. On the other extreme, people with an O or A level standing are quite reluctant to cycle, even if there is no car available in the household. This tendency is most pronounced for the elderly.

• CAR AVAILABILITY

  From Table IV and V three degrees of car availability can be distinguished First those who own a car, second those who are entitled to drive a car and have a car in the household, and third those who may do so but there is no car in the household. Generally all persons who belong to these categories reveal no preferences for cycling. The reluctance to use the bicycle seems to be strongest for those who have a car at their disposal. Somewhat surprising may be the finding that employed men and women as well as housewives who are entitled to drive a car but have no car in their household, show a stronger dislike of cycling than the same category with a car available in the household. This holds for all trip purposes.

• HOME TENURE

  Whether a person lives in a flat at rent or in a self-contained house is not only a matter of income but also an indication for the quality of his or her living environment with respect to its suitability for cycling. Furthermore, these two types of dwelling are closely related to the size of the town. So we shouldn't fall into the trap of ecological fallacy! Fortunately though, in this case no analytical problems arise: the minus signs are all on the rent side, the plus signs on the own home side. From previous research we would have expected a positive effect of low income and education on bicycle use of persons who live in flats. However. the negative effects of the environment apparently prevail. On the other side, less densely populated residential areas are an environment favourable to cycling, in spite of the fact that people who live there generally would not reveal a particular preference for using the bicycle.

• SIZE OF TOWN

  From Table IV and V it becomes apparent that towns up to 100,000 inhabitants provide an environment which is relatively favourable to cycling. The negative effect of small towns with less than 5000 inhabitants on school and work trips by bicycle result from the fact that schools and working places are mostly beyond the town limits and beyond the reach of the bicycle. For non-duty trips and the shopping trips of housewives and the elderly, the bicycle gains a relatively large share here, particularly in small communities. The critical size for cycling seems to be a population of 100,000. The very high negative factors for cities with more than 500,000 inhabitants clearly indicate the deterrent influence of their traffic conditions on the use of bicycles. Women and the elderly seem to be most affected, especially for work and shopping trips.

• TOPOGRAPHY

  Cycling is very sensitive to topography. In a flat terrain bicycle trips increase by up to 50 percent. Under these optimal topographic conditions especially girls and adults from households with a car available shift to the bicycle. By favour of a plain topography particularly non-duty trips are made more often by bike. Under less favourable topographical conditions the same groups reduce cycling most significantly. In the most unfavourable situation. classified as mountainous by an experienced cyclist, bike trips drop to one half or almost one third of the share which would have been expected if cycling would be completely insensitive to topography. From the figures in Table IV and V it appears that boys and employed women from households without a car are least sensitive. as well as some elderly for leisure trips with free choice of route and destination.

• SEASON OF THE YEAR

  Although inveterate cyclists would hardly admit it, the use of the bicycle is quite sensitive with respect to weather conditions. In the winter of 1976, cycling dropped to almost 70 percent of the summer share of trips by all modes. While trips by car were also negatively affected (minus eight percent), trips by foot and public passenger transport increased their shares by 18 and 70 percent respectively.¹

  Unfortunately there is no information about the actual weather at the specific sampling day and location. Thus weather conditions are classified according to seasons of three months each, spring starting in March. Due to this smoothing effect and because it is easier to avoid bad weather, (by postponing a trip), than an unfavourable topographical situation by choosing another place of activity which. would be more convenient to reach by bike, we would expect a smaller variation of bicycle use with respect to seasons than with respect to topography, particularly for non-duty trips. The figures of Table IV and V reveal that the bicycle is indeed used more often during summer and less during winter. However, this effect seems to be stronger for leisure trips than for obligatory trips. Schoolchildren and apprentices reduce their bike shares among leisure trips during the bad weather season by 20 and 28 percent and increase them in the summer time by 8 and 33 percent respectively. Generally persons who come from car owning households reduce their cycling activities more than other persons. This phenomenon is most pronounced for employed men. In the summer. the same population subgroups appear to react most sensitively to the favourable weather conditions.