The importance of timing for breaking commuters’ car driving habits

Full text

Alan Warde & Dale Southerton (eds.) 2012
The Habits of Consumption
Studies across Disciplines in the Humanities and Social Sciences 12.
Helsinki: Helsinki Collegium for Advanced Studies. 130–140.
The Importance of Timing for Breaking
Commuters’ Car Driving Habits1
John Thøgersen
Aarhus University
A large sample of Copenhagen car drivers were randomly assigned to either receive
a free month travel card for public transportation or serve as a control group. As
predicted, the free travel card neutralized the negative effect of car driving habits and
made the use of public transportation more consistent with the traveller’s conscious
intentions. However, the behavioural effects of the free travel card appeared only
among individuals who had recently relocated residence or workplace, prior to the
intervention. This suggests that timing is essential when designing interventions to
promote alternatives to car-driving.
Introduction
The growing number of cars contributes to serious problems all over the World,
including congestion, air pollution and noise at the local level and climate changes
at the global level. In their struggle to solve these problems, governments and
NGOs promote alternative means of transportation, including walking, cycling and
using public transport, but mostly with little success. This paper reports from a
study designed to test an innovative way to assist consumers who want to change
commuting mode from car to public transportation.
Mounting evidence suggests that at least part of the reason why it is so difcult
to get drivers out of their cars and use other travel modes is that the choice of travel
mode tends to become habitual (e.g., Gärling, Boe, & Fujii 2001; Thøgersen & Møller
2008; Verplanken, Aarts, Knippenberg, & Knippenberg 1994; Verplanken, Aarts,
Knippenberg, & Moonen 1998). This has profound inuence on the effectiveness of
interventions that might be considered for inuencing travel-mode choice (Assael
1987; Ronis, Yates, & Kirscht 1989; Aarts, Verplanken, & Knippenberg 1997).
1 A previous version of this paper was published as Thøgersen, J. 2009. Seize the opportunity:
The importance of timing for breaking commuters’ car driving habits. In A. Klein & V. W. Thoresen
(eds.) Making a Difference: Putting Consumer Citizenship into Action. Hedmark: Høgskolen i
Hedm a r k , 35 – 4 7.

Thøgersen
131
Especially, it means that drivers are unlikely to search for – or even contemplate
– new information before choosing a travel mode (Verplanken, Aarts, & Van
Knippenberg 1997). Hence, in order to be noticed at all, persuasive information
needs to be obtrusive and to be perceived as personally relevant (Dahlstrand &
Biel 1997; Hoyer & MacInnis 2006).
According to psychological theory, individuals form a habit when a behaviour is
repeated frequently in a stable context and leads to rewarding outcomes (Ouellette
& Wood 1998), something which is true for most everyday travel mode choices
(Thøgersen 2006). For example, the daily commute is usually performed frequently
and extensively in stable surroundings, and the car-commuter usually reaches the
destination in a timely and comfortable manner.
Following Verplanken and Aarts (1999), a habit is dened as a learned
sequence of acts that has become an automatic response to specic cues and is
functional in obtaining certain goals or end-states. Frequent repetition in a stable
context facilitates the learning of a habit. The instigation of a new behaviour is
usually volitional and intentional, but subsequent repetitions may eventually be
performed in an unintentional, habitual way (Ouellette & Wood 1998; Verplanken &
Aarts 1999). Before a habit is formed (i.e., learned) several repetitions are usually
required, however. With repetition, each step in the execution of the behaviour
can be done with less effort and less conscious awareness (Ouellette & Wood
1998; Thøgersen & Ölander 2006). Eventually, even the initiation of the behaviour
may become automatic, triggered by the stimulus cues that normally precede it
(Bargh & Barndollar 1996). Because of its obvious advantages, most frequently
repeated behaviours are habitual. Hence, when people are in situations they have
encountered and acted in many times before, their actions tend to be automatic
repetitions of their previously repeated behaviours.
It follows from this that people’s stated intentions are good predictors of their
behaviour only under conditions of weak habits while intentions are a bad predictor
of behaviour when habits are strong (Triandis 1977). Hence, habits moderate the
inuence of behavioural intentions on behaviour. For example, in the eld of travel-
mode choice, Verplanken et al. (1994) found that the correlation between the
attitude towards using a specic travel-mode option and travel-mode choice (for
shopping trips to either of two cities located approximately 5 miles away and where
a realistic public transport option existed) was signicantly weaker for strong than
for weak habit individuals.
As strongly emphasized by Bargh and Barndollar (1996), a habit is not a static
behavioural response, however. A habit is a mental system that interacts with
environmental information and which requires input from the environment in order
to operate. Many established behavioural routines in daily life, such as car use,

The Habits of Consumption
132
have both volitional (e.g., planning to go somewhere) and automatic elements (e.g.,
picking the car, driving). Hence, Bargh and Barndollar (1996) consider the learning
of habits an automated strategy for dealing with the environment to affect desired
goals.
A habit that was functional in obtaining some goal(s) at the time when it was
formed may lose its functionality if the goal(s) change at a later point in time,
however (Verplanken & Aarts 1999). In such cases, the habit may become counter-
intentional. Counter-intentional habits are particularly prevalent when the behaviour
is based on short-term, hedonistic motives at the expense of long-term goals
(Verplanken & Faess 1999).
For example, habitual travel mode choices are often found to deviate from the
person’s expressed intention. In practice, the deviation is usually in the direction of
a higher-than-intended use of private cars and a lower use of public transportation,
bicycling and walking (Møller & Thøgersen 2008; Verplanken et al., 1994;
Verplanken, et al. 1998; Aarts, Verplanken, & Knippenberg 1998). This means that
it is possible to achieve a more desirable modal split (from a societal point of view)
by just helping individual travellers to act according to their expressed intentions.
In general, people reserve deliberation and conscious decision-making for novel
situations and for when new problems arise in old situations, such as situational
barriers (e.g., a freeway closure blocking one’s usual commuting route; cf. Fujii,
Gärling, & Kitamura 2001) or major life changes (e.g., residential relocation; cf.
Bamberg 2006; Verplanken, Walker, Davis, & Jurasek 2008). Hence, the key to
changing habitual behaviour is to create conditions that, for some reason or other,
make the automatic execution of the habit impossible or at least unattractive (Ronis
et al. 1989) and which give individuals sufcient motivation and ability to make a
deliberate choice (Fazio 1990). The challenge is to design interventions that are
effective in producing this outcome, yet politically and individually acceptable.
In this connection, change agents may take advantage of people’s habitual
patterns being more vulnerable to inuence attempts when major changes happen
in their lives (Andreasen 1984). Specically with regard to travel mode choice, it
has been suggested that people’s car-driving habits can more easily be inuenced
when they have recently changed residence (Bamberg 2006).
The approach
The approach proposed here to make car-drivers voluntarily change their everyday
travel mode choices is similar to the way newspapers, telecommunication services,
and other goods and services that are bought on a subscription basis are often

Thøgersen
133
marketed to new customers: by means of a temporary promotion offer, typically
including a trial period for free or at a substantially reduced price. The promotion
is intended to create sufcient initial interest to entice new customers to try the
product or service, and it is hoped that the experience creates a positive attitude and
perhaps a new habit that secures repeated purchase after the trial period. In order
to avoid misuse, it is customary to restrict the promotion offer to people who have
not been subscribers to the product or service for some time. To my knowledge,
there have as yet been only sporadic attempts to promote public transportation this
way, in spite of the obvious similarities between the areas (Thøgersen 2007).
It is an important assumption behind this kind of intervention that at least some of
the receivers of the promotion will continue using the service more than before the
promotion period, even though they have to pay full fare again. In the travel-mode
choice case, there are at least two reasons for expecting such a long-term effect.
One reason is that some car-drivers may hold unjustied negative expectations
about public transport. Hence, trial-based experience resulting from the promotion
period would result in more favourable attitudes towards using public transport
(Bamberg & Schmidt, 1999). Another possible reason is that some car-drivers
have a vague (but not necessarily negative) perception about how it would be to
use public transport. Using their car works for them, so they have not bothered
to seriously consider alternatives. Also, even a relatively small cost in terms of
time and effort needed to investigate and possibly test alternatives seems to be
an insurmountable barrier. Hence, any means that could make them try public
transport would increase the quality of their knowledge of this alternative and some
would realize that for them using public transport is actually preferable to using the
car, at least for some purposes.
Method
We tested the price promotion in a eld experiment with car-drivers in the Greater
Copenhagen area fullling certain screening criteria. The basic idea was to make
car-drivers, many of which were assumed to choose travel mode habitually, “an
offer they could not resist,” and thereby motivate them to deliberate about their
travel mode choices, and in fact try public transport.
Participants
Data were collected by means of telephone interviews carried out in October and
November 2002 and April 2003.2 Subjects were a random sample of car-owners
2 The data were collected by TNS Gallup.

The Habits of Consumption
134
in the Greater Copenhagen area fullling the following screening criteria: have a
driver’s license and a car at their disposal, commute to job or study at least once
a week, and not having been a monthly travel card holder for mass transit in the
Greater Copenhagen area for at least a year. Also, traveling salesmen and others
that are dependent on a private car for their job were excluded. If more than one
person in the household fullled the criteria, the “next birthday” method was used
to pick the participant for the study.
Of those meeting the screening criteria, 1071 agreed to participate in the
rst wave, resulting in a response rate of 75 percent of those qualifying. Thirty
individuals were excluded because letters with experimental treatment material
were returned by post due to incorrect addresses, because they claimed that
they had not received the experimental treatment material, or because of errors
in the administration of experimental treatments during the telephone interview.3
The screening criteria did not take into account that some live so close to their
workplace that they have no need of motorized transport for commuting. As a crude
measure it was judged that everyone who in Wave 1 reported having commuted by
foot more than once or by bicycle more than four times out of the last ten times fell
into this category and they were excluded from the study.
The allocation to experiment and control groups followed a somewhat complex
design: First, participants were randomly assigned to either experimental treatment
(70%) or control group (30%). In the experiment group, subjects were then assigned
to one of several treatments.
Those expressing any intention to use mass transit in the near future were
randomly assigned to one of two treatments: (a) a planning exercise alone or (b) a
planning exercise plus a free month travel card. The planning exercise consisted
in asking subjects to plan their next trip by mass transit (when exactly they would
go, from where to where, using which bus or train connection, see Bamberg 2002).
Those expressing no intention to use mass transit in the near future were
randomly assigned to one of three treatments: (a) a customized timetable alone,
(b) a customized timetable plus a free month travel card, or (c) a free month travel
card alone. The customized timetable treatment consisted in sending subjects a
customized timetable for his or her daily commute based on information about
home and work given during the rst interview. Free month travel cards and
customized timetables were sent to participants by ordinary mail immediately after
the rst interview.
3 The excluded subjects did not differ signicantly (the 5 % level) from included subjects on any of
the target variables in the rst wave.

Thøgersen
135
Neither the planning exercise nor the customized timetable produced an increase
in commuting by public transport over and above the control group (Thøgersen &
Møller, 2008). For this reason, and because the focus here is on the effects of
the price promotion, experimental subjects not receiving a free travel card were
excluded from this study. Hence, the nal sample consisted of 597 car-owners
living in the greater Copenhagen area and being in employment or under education,
who (apparently) did not live too close to work to need motorized transportation for
commuting, serving either as experimental subjects (373, receiving a free month
travel card) or as members of the control group (224).
The gender distribution of the participants was 56/44 percent males/females.
The average age was 43 and the age range 18 to 71. Seventy-ve percent were
living with at least one other adult and 43 percent had children under the age
of 18 in the household. Forty-seven percent had a college or university degree.
Forty-nine percent lived in a house, 49 percent in an apartment and 2 percent in
other types of homes. None of these descriptors differed signicantly between the
experiment group and the control group.
Measures
In all three waves, answers were obtained to questions about travel behaviour and
a number of beliefs and psychological constructs regarding traveling. Participants
were also asked questions about major changes in their lives in the last three
months before lling out the rst questionnaire, including whether they had changed
residence and/or workplace. For the analysis presented in this paper, this latter
information was used to classify participants. Besides this, only the frequency of
using public transportation is used. The wording and scale of this measure are
explained in the note to Figure 1.
Previously reported results
Results of the wider study have previously been reported in several published
papers:
In Møller and Thøgersen (2008), the implications of car use habits for drivers’
use of public transportation is analysed. A relatively low percentage of the drivers
in this study (10-20%) considered commuting by public transportation in the near
future. A hierarchical analysis, where use of public transportation was regressed
onto intentions to do so, car use habit, and the interaction between the two,
conrmed the theory-derived hypothesis that car use habits act as a moderator of
the intention-behaviour relationship for public transportation. In other words, car

The Habits of Consumption
136
use habits are an obstacle to the transformation of intentions to commute by public
transportation into action.
Thøgersen and Møller (2008) extended these results by the eld experiment
where a free month travel card was tested as a tool to persuade drivers to skip the
habitual choice of the car and consider using – and to try – public transport instead.
As predicted, the free month travel card had a signicant impact on drivers’ use
of public transport and it also neutralized the impact of car-driving habits on the
intention-behaviour relationship for public transportation. However, according to
the calculations reported in this article, in the longer run (i.e., four months after the
experiment) experimental subjects did not use public transport more than control
subjects.
In Thøgersen (2009), these data were reanalysed. After excluding participants
that had no need of motorized transportation for commuting, based on their
pattern of walking or bicycling to work or study, a signicant long-term effect of the
free month travel card was revealed. Four months after the free travel card had
expired, those that had received it and actually needed motorized transportation
for commuting still used public transportation signicantly and substantially (40%)
more than at baseline.
In sum, we have shown that:
• strong car-driving habits are an obstacle for converting intentions to use
public transportation into action,
• a free month travel card can remove this obstacle, and
• among drivers with a need for motorized transportation, there is still a
signicant and substantial effect on their use of public transportation four
months later.
This study
In this paper, we study whether the effects of the intervention are contingent on
specic context factors. Specically, we study whether people’s habitual travel-
mode choices are more vulnerable to inuence attempts when major changes
happen in their lives (Andreasen 1984). Major life changes that may have
implications for travel mode choice include change of residence and change of
workplace (Bamberg 2006; Verplanken, et al. 2008). Hence, I test the following
hypothesis:
Hypothesis: The effect of receiving a free month travel card (i.e., the experimental
treatment) on the use of public transportation for commuting is stronger for people

Thøgersen
137
who have recently changed residence or workplace than for who people who have
not.
Operationally, “recently” is dened as the last three months before the
intervention. Since relocation is assumed to inuence people’s receptivity to
inuence attempts, the effect should show up in the near term. Hence, the
hypothesis is tested by comparing the pattern of travel mode choice reported at
baseline (i.e., before the intervention) and at the second interview (i.e., during the
intervention).
Results
I use a 2 (free card vs. control group) x 2 (relocation or not) x 2 (Wave 1 vs. Wave
2) mixed between and within subjects design to analyse the impacts of the free
travel card and relocation on participants’ use of public transportation. The means
are reported in Figure 1.
According to the GLM analysis, there was no direct effect of the time of the
interview (p = .28). However, there was a signicant two-way interaction between
experimental condition (free card vs. control) and time (p = .001) and a signicant
three-way interaction between experimental condition (free card vs. control),
relocation and time (p = .02). The two-way interaction was produced by the use of
public transport increasing more in the experiment group (0.5 trips out of 10) than
in the control group (-0.05 trips out of 10), as expected. The three-way interaction
was the outcome of this difference in the increase in the use of public transportation
between the experiment group and the control group being especially pronounced
among those that had changed residence or workplace in the last three months
(difference between experimental groups: 1.34 trips out of 10, p < .001), whereas
the difference in the change over time between experimental groups was small
among those that were staying put (0.27 trips out of 10, n.s.). This is consistent
with the hypothesis.
Figure 1. Use of public transport by experimental treatment and relocation of residence or
workplace at Time 1 (baseline) and Time 2 (intervention period). Means.
Note. Behaviour frequency was measured on a scale from 0 to 10 with the item: “How many of
the last 10 times did you use public transport for the trip between home and work/educational
institution?” N = 536.

The Habits of Consumption
138
Discussion
Previous studies have shown that some drivers would actually like to use public
transportation more, but do not, mainly because of old habits. A price promotion
in the form of a free month travel card was offered to a random sample of
Copenhagen car drivers. The results were encouraging. The price promotion led
to a doubling of the use of public transportation in the experiment group and a
signicant effect remained four months after the intervention (Thøgersen 2009). As
reported in Thøgersen and Møller (2008), the success of the price promotion was
partly due to the fact that it neutralized the negative inuence of car-driving habits
on transforming intentions to use public transportation into behaviour.
However, the results reported in this paper show that the effect of the free
travel card was limited to an identiable sub-set of car-drivers: people who had
recently (i.e., within the last three months) changed either residence or workplace.
Among those that had not experienced a change in any of these locations recently,
the behavioural impact of the free travel card was not signicant. This result is
consistent with previous research suggesting that individuals’ habitual behaviours
are more easily inuenced under circumstances where they experience major life
changes.
Commuters who have not recently experienced a relocation of residence
or workplace make their commuting-mode choices in a situation they have
encountered and acted in many times before. Hence, their choices tend to be
automatic repetitions of their previously repeated choices. They lack the motivation
to consider or even pay attention to alternative options. Even a free month travel
card is not sufcient to change that. However, the relocation of residence or
workplace is usually a sufciently big life change to motivate people to deliberate
and consciously consider their commuting options. Hence, car drivers are more
likely to develop an intention to try public transportation for commuting in connection
with changes of residence or workplace than under stable conditions. Hence, it is
under these circumstances that a discrepancy between intentions and behaviour
are most likely to appear. Further, our study shows that drivers who in this situation
get the chance to try public transport for free for a month are more likely to increase
their use of public transport. A nding that we did not predict is that people who do
not receive a free month pass when they change residence or workplace tend to
decrease their already low use of public transport.
Results that were not predicted should be interpreted with care until replicated in
other studies. However, one may speculate that during relocation, people often feel
that the uncertainties of the new situation are uncomfortable. One way of coping
with this uncertainty would be to stick to general habits that are exible enough to

Thøgersen
139
accommodate the new situation. It has been suggested that the habit of using the
car for commuting is one such exible habit (cf., Verplanken, et al. 1994).
The results presented in this paper show that the cost effectiveness of a price
promotion in the form of a free month travel card can be improved by targeting
it to consumer segments whose lives are undergoing changes that make them
more open to reconsider their travel options. Hence, timing is essential when
designing interventions to promote alternatives to car driving. This insight adds
further renements to our knowledge about price promotions as a method to make
habitual car drivers consider and try public transportation.
After a successful intervention that moves drivers from a habitual to a deliberate
mode of decision-making and try, for example, public transportation, a long-term
change in the way they commute may result. However, it is an obvious prerequisite
that the commuter is satised with the tested alternative to the private car. Hence,
interventions to break car-driving habits, such as the one discussed in this paper,
make little sense unless there are indeed acceptable alternative transport options
available.
References
Aarts, H. & B. Verplanken 1999. Habit, attitude, and planned behaviour: Is habit an empty
construct or an interesting case of goal-directed automaticity? European Review of
Social Psychology 10, 101–134.
Aarts, H., van Knippenberg, A., van Knippenberg, C & B. Verplanken 1994. Attitude versus
general habit: Antecedents of travel mode choice. Journal of Applied Social Psychology
24, 285–300.
Aarts, H., van Knippenberg, A., Moonen, A. & B. Verplanken 1998. Habit versus planned
behavior: A eld experiment. British Journal of Social Psychology 37, 111–128.
Aarts, H., van Knippenberg, A. & B. Verplanken 1997. Habit, information acquisition, and the
process of making travel mode choices. European Journal of Social Psychology 27,
539–560.
Aarts, H., van Knippenberg, A. & B. Verplanken 1997. Habit and information use in travel mode
choices. Acta Psychologica 96, 1–14.
Aarts, H., van Knippenberg, A. & B. Verplanken 1998. Predicting behavior from actions in
the past: Repeated decision making or a matter of habit. Journal of Applied Social
Psychology 28, 1355–1374.
Andreasen, A. R. 1984. Life status changes and changes in consumer preferences and
satisfaction. Journal of Consumer Research 11, 784 –7 9 4.
Assael, H. 1987. Consumer behavior and marketing action. Boston: Kent Publishing Co.
Bamberg, S. 2002. Effects of implementation intentions on the actual performance of new
environmentally friendly behaviours. Results of two eld experiments. Journal of
Environmental Psychology 22, 399 – 411.

The Habits of Consumption
140
Bamberg, S. 2006. Is a residential relocation a good opportunity to change people’s travel
behavior? Results from a theory-driven intervention study. Environment & Behavior
38(6), 820–840.
Bamberg, S., & P. Schmidt 1999. Regulating transpor t: Behavioral changes in the eld. Journal
of Consumer Policy 22, 479–509.
Bargh, J. A., & K. Barndollar 1996. Automaticity in action: The unconscious as repository of
chronic goals and motives. In J. A. Bargh & P. Golwitzer (eds.) Psychology of action.
New York: Guilford, 457–481.
Biel, A. & U. Dahlstrand 1997. Pro-environmental habits: Propensity levels in behavioral
change. Journal of Applied Social Psychology 27, 588–602.
Davis, A., Jurasek, M., Verplanken, B. & I. Walker 2008. Context change and travel mode
choice: Combining the habit discontinuity and self-activation hypotheses. Journal of
Environmental Psychology 28(2), 121–127.
Faess, S. & B. Verplanken 1999. Good intentions, bad habits, and effects of forming
implementation intentions on healthy eating. European Journal of Social Psychology
29, 591–604.
Fazio, R. H. 1990. Multiple processes by which attitudes guide behavior: The MODE model
as an integrative framework. In M. P. Zanna (ed.) Advances in experimental social
psychology. Vol. 23. New York: Academic, 75–109.
Fujii, S., Gärling, T., & R. Kitamura 2001. Change in drivers’ perceptions and use of public
transport during a freeway closure: Effects of temporary structural change on cooperation
in a real-life social dilemma. Environment and Behavior 33(6), 796 – 808.
Boe, O., Fujii, S. & T. Gärling 2001. Empirical tests of a model of determinants of script based
driving choice. Transportation Research Part F: Trafc Psychology and Behaviour 4,
89 –102.
Hoyer, W. D., & D. J. MacInnis 2006. Consumer behavior. 4th edition. Boston: Houghton Mifin.
Møller, B. & J. Thøgersen 2008. Breaking car-use habits: The effectiveness of a free month
travel card. Transportation 35, 329–345.
Møller, B. & J. Thøgersen 2008. Car-use habits: An obstacle to the use of public transportation?
In C. Jensen-Butler, B. Madsen, O. A. Nielsen & B. Sloth (eds.) Road pricing, the
economy, and the environment. Amsterdam: Elsevier.
Ouellette, J. A., & W. Wood 1998. Habit and intention in everyday life: The multiple processes
by which past behavior predicts future behavior. Psychological Bulletin 124, 54–74.
Kirscht, J. P., Ronis, D. L. & J. F. Yates 1989. Attitudes, decisions, and habits as determinants
of repeated behavior. In A. R. Pratkanis, S. J. Breckler & A. G. Greenwald (eds.) Attitude
structure and function. Hillsdale, N.J.: Lawrence Erlbaum, 213–239.
Thøgersen, J. 2006. Understanding repetitive travel mode choices in a stable context: A panel
study approach. Transportation Research Part A: Policy and Practice 40, 621–638.
Thøgersen, J. 2007. Social marketing of alternative transportation modes. In T. Gärling & L.
Steg (eds.) Threats to the Quality of Urban Life from Car Trafc: Problems, Causes, and
Solutions. Oxford: Elsevier, 367–381.
Thøgersen, J. 2009. Promoting public transport as a subscription service: Effects of a free
month travel card. Transport Policy 16, 335–343.
Thøgersen, J., & F. Ölander 2006. The dynamic interaction of personal norms and environment-
friendly buying behavior: A panel study. Journal of Applied Social Psychology 36, 1758 –
1780.
Triandis, H. C. (1977). Interpersonal behavior. Monterey: Books/Cole.