Addictive Behaviors 36 (2011) 502–507

Contents lists available at ScienceDirect

Addictive Behaviors

The theory of planned behavior and binge drinking among undergraduate students:
Assessing the impact of habit strength
Paul Norman ⁎
Department of Psychology, University of Sheffield, Sheffield, S10 2TP, UK

a r t i c l e i n f o a b s t r a c t

Article history: The present study sought to apply the theory of planned behavior (TPB) to the prediction of binge drinking
Received 5 August 2010 intentions and behavior among undergraduate students and to test whether habit strength explains
Received in revised form 15 December 2010 additional variance in binge drinking behavior. Undergraduate students (N = 137) completed measures of the
Accepted 17 January 2011
TPB (i.e., attitude, subjective norm, self-efficacy, perceived control, and intention) and habit strength (Self-
Report Habit Index) in relation to binge drinking. Frequency of binge drinking was assessed one month later
Keywords:
Theory of planned behavior
(n = 109). The TPB explained 75% of the variance in binge drinking intentions, with attitude and self-efficacy
Habit making significant contributions, and 35% of the variance in binge drinking behavior at one-month follow-up,
Binge drinking with only intention making a significant contribution. Habit strength explained additional variance in binge
Alcohol drinking behavior (ΔR2 = .06), although intention remained as a significant predictor. The results suggest that
Students binge drinking among undergraduate students is under the control of both intentional and habitual processes.
Interventions to reduce binge drinking should therefore focus on the motivational determinants (e.g.,
perceived positive and negative consequences) of binge drinking as well as the environmental factors (i.e.,
contextual cues) that promote binge drinking.
© 2011 Elsevier Ltd. All rights reserved.

1. Introduction Reducing binge drinking is a key element of the government's alcohol

harm reduction strategy in England (The Cabinet Office, 2004). In
There are over 800,000 alcohol-related hospital admissions in order to develop effective interventions, research is needed to identify
England each year, costing the NHS approximately £2.7 billion (NHS, the key psychosocial determinants of binge drinking.
2009). Excessive alcohol use over a prolonged period may lead to The present study draws upon the theory of planned behavior
serious health problems, such as cirrhosis of the liver, and premature (TPB) (Ajzen, 1988) to examine binge drinking intentions and
death (NHS, 2009). There are also more immediate negative social and behavior among undergraduate students. According to the TPB, the
health consequences that may follow excessive drinking in a single proximal determinant of behavior is intention which represents the
session, including accidents, injuries, public disorder, crime, physical person's motivation or decision to act. Intention, in turn, is
violence and sexual assault (Miller, Plant, & Plant, 2005; The Cabinet determined by three factors. First, is the person's attitude towards
Office, 2004). The cost of alcohol-related crime and anti-social performing the behavior which reflects an overall positive or negative
behavior has been estimated to be over £7 billion a year (NHS, evaluation of performing the behavior. Second, is subjective norm
2009). Binge drinking is a common behavior in the UK, particularly which reflects the person's perception of whether or not important
among young people (Kuntsche, Rehm, & Gmel, 2004). In England, others would want him or her to engage in the behavior. Third, is
32% of men and 25% of women aged between 16 and 24 reported perceived behavioral control which reflects the person's perception of
having engaged in binge drinking, defined as consuming more than 8/ the extent to which internal and external factors may facilitate or
6 (male/female) units of alcohol in a single session, in the previous hinder behavioral performance. To the extent that people are accurate
week (NHS, 2009). The prevalence of binge drinking has been noted to in their perceptions of control, perceived behavioral control is also
be higher among university students than among their non-student expected to have a direct effect on behavior. A distinction is
peers (Gill, 2002; Webb, Ashton, Kelly, & Kamali, 1996). Norman and commonly made between self-efficacy which refers to the person's
Conner (2006) reported that 67% of students in their sample had confidence in their ability to perform the behavior, and perceived
engaged in binge drinking in the previous week. Similarly high figures control which refers to the person's perception of the amount of
have been noted in other studies (Cooke, Sniehotta, & Schüz, 2007; control they have over performing the behavior (e.g., Armitage,
Jamison & Myers, 2008; Norman, Armitage, & Quigley, 2007). Conner, Loach, & Willetts, 1999; Norman & Hoyle, 2004; Terry &
O'Leary, 1995).
⁎ Tel.: + 44 114 2226505; fax: + 44 114 2766515. The TPB has been applied to the prediction of various health-
E-mail address: p.norman@sheffield.ac.uk. related behaviors (see Conner & Sparks, 2005, for a review), and meta-

0306-4603/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.addbeh.2011.01.025
 P. Norman / Addictive Behaviors 36 (2011) 502–507 503

analytical reviews attest to its predictive utility (e.g., Armitage & found to correlate with a response-frequency measure of habit
Conner, 2001; Cooke & French, 2008; Hausenblas, Carron, & Mack, strength and with the frequency of past behavior (Verplanken &
1997). For example, Armitage and Conner (2001) reported that the Orbell, 2003). In addition, the SRHI has been found to have direct
TPB explains an average of 39% of the variance in intention and 27% of effects on future behavior over and above the effects of social
the variance in behavior. The TPB has been applied to the prediction of cognitive variables (e.g., Burg, de Vet, de Nooijer, & Verplanken, 2006;
various alcohol-related behaviors (e.g., Armitage et al., 1999; Conner, Gardner, 2009; Verplanken, 2006) although, to date, no studies have
Warren, Close, & Sparks, 1999; McMillan & Conner, 2003; Schlegel, tested whether habit strength explains additional variance in binge
d'Avernas, Zanna, & DeCourville, 1992) including binge drinking drinking behavior.
among undergraduate students (e.g., Cooke et al., 2007; Hagger, A number of studies have examined the interaction between habit
Anderson, Kyriakaki, & Darkings, 2007; Johnston & White, 2003; strength and intention when predicting health-related behavior.
Norman, Armitage, & Quigley, 2007; Norman & Conner, 2006). For According to Triandis (1977), novel behaviors are likely to be
example, Johnston and White (2003) reported that the TPB explained determined by intention, whereas repeated behaviors are likely to
69% of the variance in binge drinking intentions, with attitude, be determined by habit strength. An interaction is therefore expected
subjective norm and self-efficacy emerging as significant predictors, between intention and habit strength when predicting future
and 51% of the variance in binge drinking behavior at two-week behavior; habit strength should moderate the intention–behavior
follow-up, although only intention was a significant predictor. relationship such that intention becomes a weaker predictor of future
Despite the ability of the TPB to explain a significant amount of behavior as habit strength increases. A handful of studies have
variance in behavior, a common criticism of the model is its inability provided support for this moderation hypothesis using the SRHI as a
to account for the strong relationship between past behavior and measure of habit strength (De Bruijn, 2010; De Bruijn, Kremers, Singh,
future behavior. Past behavior is often found to have a direct effect on van den Putte, & van Mechelen, 2009; De Bruijn, Kroeze, Oenema, &
future behavior that is not mediated by the TPB. Conner and Armitage Brug, 2008; De Bruijn et al., 2007; Gardner, 2009; Kremers & Brug,
(1998) estimated that past behavior explains, on average, an 2008), although Norman and Cooper (in press) failed to find evidence
additional 13% of the variance in future behavior over and above the for such an interaction. However, relatively few studies have
influence of the TPB. Past behavior has also been found to explain employed prospective designs when testing the moderation hypoth-
additional variance in TPB studies on binge drinking (e.g., Cooke et al., esis (De Bruijn et al., 2007; Gardner, 2009; Norman & Cooper, in
2007; Norman & Conner, 2006; Norman et al., 2007). When past press), which represents an important limitation of previous research.
behavior is found to have a direct effect on future behavior, it is Moreover, no studies have tested the moderation hypothesis in
usually taken to reflect the operation of habits—i.e., “learned relation to binge drinking.
sequences of acts that have become automatic responses to specific Binge drinking among undergraduate students has many of the
cues” (Verplanken & Aarts, 1999, p. 104). For example, Ouellette and defining features of a habitual behavior (Verplanken & Orbell, 2003)
Wood (1998) argue that when behaviors are performed repeatedly and, as a result, is likely to be predicted by habit strength. First, binge
over time in stable contexts they are likely to come under the control drinking at university is a frequently performed behavior. Norman,
of habitual, as opposed to intentional, processes. Bennett, and Lewis (1998) found that 46.3% of their sample of
However, there are a number of problems with this account of the undergraduate students reported engaging in binge drinking, on
past behavior–future behavior relationship (Ajzen, 2002; Norman & average, at least once a week. Similarly high figures have been
Conner, 2005). First, the direct effect of past behavior on future reported in other studies of UK undergraduate students (Cooke et al.,
behavior may simply indicate that the TPB is not sufficient and that 2007; Jamison & Myers, 2008; Norman & Conner, 2006; Norman et al.,
other social cognitive variables need to be considered. Second, 2007). Second, students often report engaging in binge drinking in
measures of past and future behavior typically share method variance, response to environmental cues. Norman et al. (1998) found that
not shared by measures of the TPB constructs, which may account for students who engaged in binge drinking were more likely to cite a
the strong past behavior–future behavior relationship. Third, inferring range of external factors (e.g., being at a party) as facilitating factors.
the existence of habit from a strong correlation between past behavior Similarly, Carpenter et al. (2007) reported that students reported a
and future behavior, and then using this concept to explain the strong range of environmental factors that encouraged binge drinking at
correlation involves circular reasoning. From a theoretical perspec- university such as the promotion of cheap drinks at “student nights”.
tive, past behavior is an empty construct that adds little to our Alcohol is a prominent part of university culture and is present at most
understanding of the determinants of behavior. Moreover, from a social functions (Thombs, 1999). As a result, students are likely to
practical viewpoint, knowing that past behavior is a strong predictor encounter many cues to engage in binge drinking. Third, binge
of future behavior has no implications for interventions given that it is drinking is closely linked to students' sense of identity. Students
not possible to change past behavior. report that binge drinking is an expected and integral part of their
To examine the role of habitual processes in behaviors such as student identity, and represents a phase which is likely to stop or
binge drinking, it is necessary to have a direct measure of habit decline after leaving university (Carpenter et al., 2007; Colby, Colby, &
strength. Triandis (1977) suggested that the frequency of past Raymond, 2009). In addition, Norman et al. (2007) found that
behavior could be used to measure habit strength. However, as Mittal students who perceived themselves to be similar to the “typical
(1988, p. 997) argues “repeated occurrence is necessary for the binge drinker” were more likely to engage in binge drinking, reflecting
formation of habit, but is not habit itself”. Using frequency of past the importance of self-presentation processes (Schlenker, 1980).
behavior as a measure of habit strength fails to capture all of the The present study sought to apply the TPB to the prediction of
defining features of a habitual response (Ajzen, 2002). Habitual binge drinking intentions and behavior and to test whether habit
behaviors are performed frequently (i.e., have a history of repetition), strength explains additional variance in binge drinking behavior and
but they are also performed automatically (i.e., efficiently) in response moderates the relationship between binge drinking intentions and
to stable environmental cues (Verplanken & Aarts, 1999). In addition, behavior. It was predicted that the TPB would explain significant
given that habits are an integral part of how people organize their amounts of variance in both binge drinking intentions and behavior
everyday lives, they may also reflect a person's sense of identity but that habit strength would explain additional variance. It was
(Trafimow & Wyer, 1993). Verplanken and Orbell (2003) developed further predicted that habit strength would moderate the relationship
the Self-Report Habit Index (SRHI) to assess meta-cognitions between binge drinking intentions and behavior such that the
regarding the key features of the habit construct; namely, a history strength of the intention–behavior relationship would become
of repetition, automaticity, and expressing identity. The SRHI has been weaker as habit strength increased.
504 P. Norman / Addictive Behaviors 36 (2011) 502–507

2. Method varimax rotation was then performed. The three self-efficacy items all
loaded highly on the first factor (.93, .95, and .89), whereas the three
2.1. Respondents and procedure perceived control items all loaded highly on the second factor (.89,
.89, and .65).
Undergraduate students (N = 137) completed measures of the TPB
and habit strength in relation to binge drinking (age M = 19.12, 2.2.2. Habit strength
SD = 1.85, male n = 25, female n = 112). At one-month follow-up, 109 Habit strength was assessed using the Self-Report Habit Index
of these respondents (79.6% response rate) completed a second (SRHI; Verplanken & Orbell, 2003). Participants were presented with
questionnaire on their binge drinking behavior over the previous the stem “Binge drinking is something…” followed by 12 items (e.g., “I
month. Both questionnaires were distributed, and completed, in an do frequently”, “I do automatically”, and “that's typically “me””)
introductory Psychology lecture. Students were under no obligation to (α = .94).
participate in the study, although those who completed both
questionnaires received a course credit. Ethical approval for the 2.2.3. Future behavior
study was obtained from the Department of Psychology's Research The frequency of binge drinking at one-month follow-up was
Ethics Committee. assessed by asking respondents to indicate how many times they had
engaged in binge drinking over the previous month. The reliability of
2.2. Measures this measure was assessed by asking participants to also write down
what they had typically drunk on each day of the week over the
The time 1 questionnaire included direct measures of the main previous month. The number of days in a typical week over the
constructs of the TPB constructed in line with current recommenda- previous month that respondents had engaged in binge drinking was
tions (Conner & Sparks, 2005) and based on items used in previous found to be highly correlated with the frequency of binge drinking
TPB-binge drinking studies (e.g., Norman & Conner, 2006; Norman et measure, r(107) = .83, p b .001.
al., 2007). The questionnaire also included a measure of habit strength
(SRHI; Verplanken & Orbell, 2003). All items were measured on 7- 3. Results
point response scales and coded so that high values indicated high
levels on the variable of interest. The time 2 questionnaire assessed 3.1. Descriptive findings
frequency of binge drinking over the previous month. Binge drinking
was defined at the beginning of each questionnaire for men as In order to assess attrition biases, the time 1 questionnaire
“consuming at least 8 units of alcohol in a single session” which responses of respondents who completed both questionnaires were
“approximately equates to 4 pints of beer/larger (or 8 shorts/small compared with those who only completed the first questionnaire. No
glasses of wine)”, and for women as “consuming at least 6 units of significant differences were found between these two groups on the
alcohol in a single session” which “approximately equates to 3 pints of time 1 questionnaire measures, F(6,130) = 0.27, p = .95.
beer/larger (or 6 shorts/small glasses of wine)” (NHS, 2009). Intercorrelations between the main measures are presented in
Table 1 along with means and standard deviations. At time 2,
2.2.1. TPB measures respondents reported having engaged in binge drinking an average
Four items assessed intention to engage in binge drinking over the of 6.00 times in the previous month, with 75 (68.8%) respondents
next month (i.e., “I intend to engage in binge drinking over the next reporting having engaged in binge drinking at least 4 times over the
month”, “Do you intend to engage in binge drinking over the next previous month. All of the TPB variables (with the exception of
month?”, “How likely is it that you will engage in binge drinking over perceived control) and habit strength were found to have significant
the next month?”, “I expect that I will engage in binge drinking over positive correlations with both binge drinking intentions and
the next month”) (α = .95). Attitude towards engaging in binge behavior. Associations were also examined between age and gender
drinking over the next month was measured using five semantic and binge drinking intentions and behavior; however, none of the
differential scales (i.e., “Engaging in binge drinking over the next associations were found to be significant.
month would be… “good–bad”, “beneficial–harmful”, “wise–foolish”,
“enjoyable–unenjoyable”, “pleasant–unpleasant””) (α = .88). Subjec- 3.2. Predicting binge drinking intentions
tive norm was measured using three items (i.e., “People who are
important to me think I should/should not engage in binge drinking A hierarchical regression analysis was conducted to predict
over the next month”, “People who are important to me would intention to engage in binge drinking over the next month (see
approve/disapprove of me engaging in binge drinking over the next Table 2). The independent variables were entered in two blocks:
month”, and “I feel under social pressure to engage in binge drinking (i) attitude, subjective norm, self-efficacy and perceived control, and
over the next month”) (α = .61). The third item was subsequently (ii) habit strength. The TPB variables explained 75% of the variance in
dropped in order to increase the internal reliability of the subjective
norm measure. The remaining two items were highly correlated, r Table 1
(135) = .60, p b .001. Three items were used to measure self-efficacy Descriptive statistics and intercorrelations between the study variables (N = 137).
(i.e., “If I wanted to, I could easily engage in binge drinking over the 2. 3. 4. 5. 6. 7.a Mean SD
next month”, “If I wanted to, engaging in binge drinking over the next
1. Intention .82*** .44*** .64*** .02 .70*** .59*** 4.99 1.96
month would be… easy-difficult”, and “How confident are you that
2. Attitude .60*** .48*** −.01 .63*** .51*** 3.89 1.35
you could engage in binge drinking over the next month?”) (α = .92) 3. Subjective .23** .01 .33*** .21* 3.57 1.41
and three items were used to measure perceived control (e.g., “I feel in norm
complete control over whether or not I engage in binge drinking over 4. Self-efficacy .18* .44*** .37*** 6.10 1.43
5. Perceived −.15 −.05 6.20 0.96
the next month”, “How much control do you have over whether or not
control
you engage in binge drinking over the next month?”, and “It is up to 6. Habit strength .60*** 3.13 1.39
me whether or not I engage in binge drinking over the next month”) 7. Time 2 binge 6.00 4.90
(α = .76). These items were also subjected to a principal components drinking
analysis which extracted two factors, with eigen values greater than 1, frequency

that explained 45.05% and 32.16% of the variance in item scores. A Note. * p b .05. ** p b .01. *** p b .001. a n = 109.
 P. Norman / Addictive Behaviors 36 (2011) 502–507 505

Table 2 Table 3
Summary of hierarchical regression analysis for variables predicting intention to Summary of hierarchical regression analyses for variables predicting time 2 binge
engage in binge drinking over the next month (N = 137). drinking frequency (N = 109).

Step Variable B SE β Step Variable B SE β

1. Attitude 1.00 .09 .69⁎⁎⁎ 1. Intention 1.44 .26 .58⁎⁎⁎

Subjective norm − 0.07 .08 −.05 Self-efficacy 0.05 .35 .02
Self-efficacy 0.45 .07 .33⁎⁎⁎ Perceived control − 0.35 .40 −.07
Perceived control − 0.06 .09 −.03 2. Intention 0.78 .31 .31⁎
2. Attitude 0.81 .09 .56⁎⁎⁎ Self-efficacy 0.04 .34 .01
Subjective norm − 0.06 .07 −.04 Perceived control − 0.02 .40 −.004
Self-efficacy 0.37 .07 .27⁎⁎⁎ Habit strength 1.34 .40 .37⁎⁎
Perceived control 0.03 .09 .02 3. Intention 1.11 .37 .45⁎⁎
Habit strength 0.34 .08 .24⁎⁎⁎ Self-efficacy 0.03 .33 .01
Perceived control − 0.22 .41 −.04
Note. R2 = .75 for Step 1, p b .001; ΔR2 = .03 for Step 2, p b .001. ⁎⁎⁎ p b .001.
Habit strength 1.10 .42 .31⁎⁎
Intention × habit strength 0.29 .17 .16
binge drinking intentions, R2 = .75, F(4,132) = 99.32, p b .001. Only
Note. R2 = .35 for Step 1, p b .001; ΔR2 = .06 for Step 2, p = .001; ΔR2 = .01 for Step 3,
attitude and self-efficacy made significant independent contributions ⁎ p b .05. p = .10. ⁎⁎ p b .01. ⁎⁎⁎ p b .001.
to the regression equation. The addition of habit strength at step 2
produced a significant increment in the amount of variance explained,
ΔR2 = .03, ΔF(1,131) = 18.79, p b .001. The final regression equation The TPB explained 37% of the variance in the frequency of binge
explained 78% of the variance in binge drinking intentions, R2 = .78, drinking at one-month follow-up, although only intention emerged as
F(5,131) = 93.92, p b .001, with attitude, self-efficacy and habit a significant predictor. The present results are broadly in line with
strength making significant contributions.1,2 previous studies which have found the TPB to be predictive of binge
drinking intentions and behavior. Thus, both attitude and self-efficacy
3.3. Predicting time 2 binge drinking behavior (or perceived behavioral control) are usually found to be predictive of
binge drinking intentions (e.g., Cooke et al., 2007; Hagger et al., 2007;
A hierarchical regression analysis was conducted to predict the Johnston & White, 2003; Norman & Conner, 2006; Norman et al.,
frequency of binge drinking at one-month follow-up (see Table 3). 2007). In contrast, subjective norm has only been found to be a
The independent variables were entered in three blocks: (i) intention, significant predictor in a couple of studies (e.g., Hagger et al., 2007;
self-efficacy and perceived control, and (ii) habit strength, and Johnston & White, 2003). Intention is usually found to be predictive of
intention × habit strength. Intention and habit strength were mean binge drinking behavior (e.g., Cooke et al., 2007; Hagger et al., 2007;
centered prior to the construction of the interaction term in order to Johnston & White, 2003; Norman & Conner, 2006; Norman et al.,
minimize any problems of multi-collinearity and to aid interpretation 2007), although in some studies self-efficacy (or perceived behavioral
of the results (Aiken & West, 1991). The variables entered at step 1 control) has also emerged as a significant predictor (e.g., Cooke et al.,
explained 35% of the variance in the frequency of binge drinking at 2007; Hagger et al., 2007; Norman & Conner, 2006).
time 2, R2 = .35, F(3,105) = 18.65, p b .001, with only intention In line with other TPB studies that have reported that habit
emerging as a significant independent predictor. The addition of strength has a direct effect on future behavior (e.g., Burg, de Vet, de
habit strength at step 2 explained an additional 6% of the variance in Nooijer, & Verplanken, 2006; Gardner, 2009; Verplanken, 2006), the
time 2 binge drinking behavior, ΔR2 = .06, ΔF(1,104) = 11.08, present study found that habit strength explained an additional 6% of
p = .001. The addition of the intention × habit strength interaction the variance in binge drinking behavior at one-month follow-up.
term failed to produce a significant increment in the amount of However, it is noteworthy that both intention and habit strength
variance explained, ΔR2 = .01, ΔF(1,103) = 2.81, p = .10. The final emerged as significant predictors, suggesting that binge drinking
regression equation explained 42% of the variance in the frequency of behavior is under the control of both intentional and habitual
binge drinking at time 2, R2 = .42, F(5,103) = 15.29, p b .001, with processes. This suggests that rather than being in competition, as
intention and habit strength making significant contributions.3 implied by the Triandis (1977) model, intentional and habitual
processes may complement each other. Thus, both strong intentions
4. Discussion and strong habits may be important in determining binge drinking.
For example, respondents who have strong intentions may seek out
The present study examined the ability of the TPB to explain binge opportunities to engage in binge drinking which they are likely to
drinking intentions and behavior in a sample of UK undergraduates. encounter given that alcohol is a prominent part of university culture
The TPB explained 75% of the variance in binge drinking intentions (Thombs, 1999), and those who have strong habits are likely to
with attitude and self-efficacy emerging as significant predictors. readily respond to such contextual cues.
Thus, those students who had a positive attitude towards binge The hypothesized interaction between intention and habit
drinking and felt that it would be relatively easy to perform, had strength was found to be non-significant. The present findings are
stronger intentions to engage in binge drinking over the next month. therefore in contrast to other studies which have found that the
intention–behavior relationship becomes weaker as habit strength
1
Collinearity statistics (i.e., tolerance, variance inflation factors) and collinearity increases (e.g., De Bruijn et al., 2007, 2008, 2009; Gardner, 2009). The
diagnostics (i.e., condition index, variance proportions) were computed for the present results may be unique to binge drinking, although Norman
regression analyses, in order to assess the potential impact of multicollinearity among and Cooper (in press) also failed to find evidence for the moderation
the independent variables. These tests indicated that multicollinearity was not a cause
hypothesis. However, on theoretical level, it is not clear why
for concern.
2
The regression analyses were repeated with age and gender included as intentions should lose their predictive validity as habit strength
independent variables; however, this had no impact on the pattern or significance increases (Ajzen, 2002); people may continue to hold positive
of the results. intentions even if the behavior is under the control of habitual
3
The regression analysis predicting time 2 binge drinking behavior was repeated with processes, as evidenced by the strong correlation (r = .70) found
only intention and habit strength entered as independent variables. Together, they
between intention and habit strength in the present study. Moreover,
explained 41% of the variance in time 2 binge drinking behavior, R2 = .41, F(2,106) =36.88,
pb .001, with both intention, B=0.79, SE =0.27, β=.32, p=.004, and habit strength, Ouellette and Wood (1998) reported that intentions were predictive
B= 1.34, SE =0.39, β=.37, pb .001, emerging as significant independent predictors. of behaviors performed frequently in stable contexts (r = .59); the
506 P. Norman / Addictive Behaviors 36 (2011) 502–507

very conditions under which the influence of habitual processes alcohol (Toomey, Lenk, & Wagenaar, 2007). An alternative approach
should be strongest. Ajzen (2002) has further argued that intentions would be to change how students react to the contextual cues that
might themselves be automatically activated and used to guide encourage binge drinking. One way in which this may be achieved is
behavior outside of conscious awareness. through the making of implementation intentions (i.e., if-then plans)
The present study has a number of limitations that should be specifying when, where and how a goal intention (e.g., to avoid binge
noted. First, the internal reliability of the subjective norm measure drinking) will be enacted (Gollwitzer, 1993). For example, Murgraff,
was found to be low and, as a result, one item had to be dropped. White, and Phillips (1996) reported that instructing students to specify
Previous reviews have commented on the poor reliability of when, where and how they would refuse a drink led to a reduction in the
subjective norm measures which may attenuate the strength of frequency of binge drinking.
correlations with intention (Armitage & Conner, 2001; Conner & In conclusion, the present findings concur with previous studies
Sparks, 2005). Nonetheless, in the present study, the two-item that have applied the TPB in relation to binge drinking among
measure of subjective norms was found to correlate significantly undergraduate students. In particular, students' attitudes towards
with intention and was of a similar strength to that reported in binge drinking and their feelings of self-efficacy were found to be
previous TPB-binge drinking studies (e.g., Norman & Conner, 2006; important determinants of binge drinking intentions which, in turn,
Norman et al., 2007). Second, a self-report measure of binge drinking predicted binge drinking behavior. However, a measure of habit
behavior was employed which may have inflated the size of strength was found to explain additional variance in binge drinking
correlations with the TPB variables. Armitage and Conner (2001) behavior, consistent with the idea that binge drinking is a behavior
reported that the TPB provides stronger predictions of self-report that is performed frequently in response to strong contextual cues
behaviors than observed behaviors. Alternative measures of alcohol- (e.g., student nights and “happy hours”). The present findings suggest
related behavior exist (e.g., nominated peers) although these also that binge drinking among undergraduate students is under the
have their own limitations (Gill, 2002). Third, the present study control of both intentional and habitual processes. Interventions
assessed binge drinking over a relatively short time period (i.e., one should therefore focus on the motivational as well as the environ-
month), although previous TPB studies on binge drinking among mental factors that promote binge drinking.
students have been limited to one- (e.g., Cooke et al., 2007; Norman &
Conner, 2006; Norman et al., 2007) or two-week (e.g., Hagger et al., Role of Funding Sources
2007; Johnston & White, 2003) follow-up periods. Fourth, similar to The research had no specific funding.
previous TPB studies on binge drinking among students (e.g., Johnston
& White, 2003, Norman & Conner, 2006; Norman et al., 2007), the Contributors
respondents comprised a convenience sample of predominantly Sole author.
female students, and as a result, the generalizability of the results
can be questioned.4
Conflict of Interest
Notwithstanding the above limitations, there are a number of None.
practical implications that can be drawn from the present findings. Both
intention and habit strength were predictive of binge drinking behavior
Acknowledgement
suggesting that interventions should target both the motivational and I would like to thank Emily Hill and Gemma Williams for their help with collecting
habitual determinants of binge drinking among undergraduate students and entering the data.
(cf., Verplanken & Wood, 2006). In relation to the motivational
determinants of binge drinking, interventions need to target attitudes
References
and feelings of self-efficacy. To change attitudes towards binge drinking,
interventions should highlight the negative consequences associated Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions.
with binge drinking (e.g., greater risk of assault) and challenge the Newbury Park: Sage.
Ajzen, I. (1988). Attitudes, personality and behavior. Milton Keynes: Open University
perceived positive consequences of binge drinking (e.g., reduced
Press.
inhibition). Alternatively, the positive consequences of light drinking Ajzen, I. (2002). Residual effects of past on later behavior: Habituation and reasoned
could be emphasized so that students are encouraged to enjoy action perspectives. Personality and Social Psychology Review, 6, 107−122.
Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A
themselves without binge drinking. Encouragingly, recent reviews
meta-analytic review. British Journal of Social Psychology, 40, 471−499.
have highlighted the efficacy of individual-level interventions that Armitage, C. J., Conner, M., Loach, J., & Willetts, D. (1999). Different perceptions of
target alcohol-related beliefs and expectancies through techniques such control: Applying an extended theory of planned behavior to legal and illegal drug
as motivational interviewing and the use of decisional balance exercises use. Basic and Applied Social Psychology, 21, 301−316.
Burg, J., de Vet, E., de Nooijer, J., & Verplanken, B. (2006). Predicting fruit consumption:
(Carey, Scott-Sheldon, Carey, & Demartini, 2007; Larimer & Cronce, Cognitions, intention, and habits. Journal of Nutrition Education and Behavior, 38,
2007). Self-efficacy expectations might be reduced through changes in 73−81.
the environment that make binge drinking more difficult or by Carey, K. B., Scott-Sheldon, L. A. J., Carey, M. P., & Demartini, K. S. (2007). Individual-
level interventions to reduce college student drinking: A meta-analytic review.
highlighting the lack of opportunities if other goals are prioritized Addictive Behaviors, 32, 2469−2494.
(e.g., lecture attendance). Alternatively, refusal self-efficacy could be Carpenter, R., Fishlock, A., Mulroy, A., Oxley, B., Russell, K., Salter, C., et al. (2007). After
enhanced by providing students with specific examples of strategies for ‘Unit 1421’: An exploratory study into female students' attitudes and behaviours
towards binge drinking at Leeds University. Journal of Public Health, 30, 8−13.
refusing drinks (Murgraff, Abraham, & McDermott, 2007). In relation to Colby, S. M., Colby, J. J., & Raymond, G. A. (2009). College versus the real world: Student
the habitual determinants of binge drinking, interventions need to perceptions and implications for understanding heavy drinking among college
target the contextual cues that are likely to promote binge drinking students. Addictive Behaviors, 34, 17−27.
Conner, M., & Armitage, C. (1998). Extending the theory of planned behavior: A review
(Kuo, Wechsler, Greenberg, & Lee, 2003). For example, the promotion of
and avenues for further research. Journal of Applied Social Psychology, 28,
cheap drinks at certain times (e.g., “happy hours”) or in specific contexts 1429−1464.
(e.g., student nights in clubs) could be restricted. A recent review of Conner, M., & Sparks, P. (2005). The theory of planned behaviour and health behaviour.
In M. Conner, & P. Norman (Eds.), Predicting health behaviour (pp. 170−222).
environmental approaches to reduce student drinking has highlighted
Buckingham, UK: Open University Press.
the potential efficacy of strategies such as restricting the availability of Conner, M., Warren, R., Close, S., & Sparks, P. (1999). Alcohol consumption and the
theory of planned behaviour: An examination of the cognitive mediation of past
behavior. Journal of Applied Social Psychology, 29, 1676−1704.
4
The regression analyses were repeated for females participants only. The Cooke, R., & French, D. P. (2008). How well do the theory of reasoned action and the
significant predictors of binge drinking intentions and time 2 binge drinking behavior theory of planned behaviour predict intentions and attendance at screening
were the same as for the full sample. programmes? A meta-analysis. Psychology and Health, 23, 745−765.
 P. Norman / Addictive Behaviors 36 (2011) 502–507 507

Cooke, R., Sniehotta, F., & Schüz, B. (2007). Predicting binge-drinking behaviour using NHS (2009). Statistics on Alcohol, England 2009. Retrieved 4 February, 2010, from.
an extended TPB: Examining the impact of anticipated regret and descriptive http://www.ic.nhs.uk/pubs/alcohol09
norms. Alcohol and Alcoholism, 42, 84−91. Norman, P., Armitage, C. J., & Quigley, C. (2007). The theory of planned behavior and
De Bruijn, G. J. (2010). Understanding college students' fruit consumption: Integrating binge drinking: Assessing the impact of binge drinker prototypes. Addictive
habit strength in the theory of planned behaviour. Appetite, 54, 16−22. Behaviors, 32, 1753−1768.
De Bruijn, G. J., Kremers, S. P. J., de Vet, E., de Nooijer, J., van Mechelen, W., & Brug, J. Norman, P., Bennett, P., & Lewis, H. (1998). Understanding binge drinking among young
(2007). Does habit strength moderate the intention–behaviour relationship in the people: An application of the theory of planned behaviour. Health Education Research,
theory of planned behaviour? Psychology and Health, 22, 899−916. 13, 163−169.
De Bruijn, G. J., Kremers, S. P. J., Singh, A., van den Putte, B., & van Mechelen, W. (2009). Norman, P., & Conner, M. (2005). Predicting and changing health behaviour: Future
Adult active transportation: Adding habit strength to the theory of planned directions. In M. Conner, & P. Norman (Eds.), Predicting health behaviour
behavior. American Journal of Preventive Medicine, 36, 189−194. (pp. 324−371). Buckingham, UK: Open University Press.
De Bruijn, G. J., Kroeze, W., Oenema, A., & Brug, J. (2008). Saturated fat and the theory of Norman, P., & Conner, M. (2006). The theory of planned behaviour and binge drinking:
planned behaviour: Exploring additive and interactive effects of habit strength. Assessing the moderating role of past behaviour within the theory of planned
Appetite, 51, 318−323. behaviour. British Journal of Health Psychology, 11, 55−70.
Gardner, B. (2009). Modelling motivation and habit in stable travel mode contexts. Norman, P., & Hoyle, S. (2004). The theory of planned behavior and breast self-
Transportation Research Part F: Traffic Psychology and Behaviour, 12, 68−76. examination: Distinguishing between perceived control and self-efficacy. Journal of
Gill, J. S. (2002). Reported levels of alcohol consumption and binge drinking within the Applied Social Psychology, 34, 694−708.
UK undergraduate student population over the last 25 years. Alcohol and Norman, P., & Cooper, Y. (in press). The theory of planned behaviour and breast self-
Alcoholism, 37, 109−120. examination: Assessing the impact of past behaviour, context stability and habit
Gollwitzer, P. M. (1993). Goal achievement: The role of intentions. European Review of strength. Psychology and Health.
Social Psychology, 4, 141−185. Ouellette, J., & Wood, W. (1998). Habit and intention in everyday life: The multiple
Hagger, M., Anderson, M., Kyriakaki, M., & Darkings, S. (2007). Aspects of identity and processes by which past behavior predicts future behavior. Psychological Bulletin, 124,
their influence on intentional behavior: Comparing effects for three health 54−74.
behaviors. Personality and Individual Differences, 42, 355−367. Schlegel, R. P., d'Avernas, J. R., Zanna, M. P., & DeCourville, N. H. (1992). Problem
Hausenblas, H. A., Carron, A. V., & Mack, D. E. (1997). Application of the theories of drinking: A problem for the theory of reasoned action? Journal of Applied Social
reasoned action and planned behavior to exercise behaviour: A meta-analysis. Psychology, 22, 358−385.
Journal of Sport and Exercise Psychology, 19, 36−51. Schlenker, B. R. (1980). Impression management: The self-concept, social identity and
Jamison, J., & Myers, L. B. (2008). Peer-group and price influence students drinking interpersonal relations. Monterey: Brooks-Cole.
along with planned behaviour. Alcohol and Alcoholism, 43, 492−497. Terry, D. J., & O'Leary, J. E. (1995). The theory of planned behaviour: The effects of
Johnston, K. L., & White, K. M. (2003). Binge drinking: A test of the role of group norms perceived behavioural control and self-efficacy. British Journal of Social Psychology,
in the theory of planned behaviour. Psychology and Health, 18, 63−77. 34, 199−220.
Kremers, S. P. J., & Brug, J. (2008). Habit strength of physical activity and sedentary The Cabinet Office (2004). Alcohol harm reduction strategy for England. Retrieved 4
behaviour among children and adolescents. Pediatric Exercise Science, 20, 5−14. February, 2010, from. http://www.cabinetoffice.gov.uk/media/cabinetoffice/strategy/
Kuntsche, E., Rehm, J., & Gmel, G. (2004). Characteristics of binge drinkers in Europe. assets/caboffce%20alcoholhar.pdf
Social Science and Medicine, 59, 113−127. Thombs, D. L. (1999). An introduction to addictive behaviors. New York: Guilford Press.
Kuo, M., Wechsler, H., Greenberg, P., & Lee, P. (2003). The marketing of alcohol to Toomey, T., Lenk, K. M., & Wagenaar, A. C. (2007). Environmental policies to reduce
college students: The role of low prices and special promotions. American Journal of college drinking: An update of research findings. Journal of Studies on Alcohol and
Preventive Medicine, 25, 204−211. Drugs, 68, 208−219.
Larimer, M. E., & Cronce, J. M. (2007). Identification, prevention, and treatment Trafimow, D., & Wyer, R. S., Jr. (1993). Cognitive representation of mundane social
revisited: Individual-focused college drinking prevention strategies 1999–2006. events. Journal of Personality and Social Psychology, 64, 365−376.
Addictive Behaviors, 32, 2439−2468. Triandis, H. C. (1977). Interpersonal behavior. Monterey, CA: Brooks-Cole.
McMillan, B., & Conner, M. (2003). Using the theory of planned behaviour to Verplanken, B. (2006). Beyond frequency: Habit as a mental construct. British Journal of
understand alcohol and tobacco use in students. Psychology, Health and Medicine, Social Psychology, 45, 639−656.
8, 317−328. Verplanken, B., & Aarts, H. (1999). Habit, attitude and planned behaviour: Is habit an
Miller, P., Plant, M., & Plant, M. (2005). Spreading out or concentrating weekly empty construct or an interesting case of automaticity? European Review of Social
consumption: Alcohol problems and other consequences within a UK population Psychology, 10, 101−134.
sample. Alcohol and Alcoholism, 40, 461−468. Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: A self-report index of
Mittal, B. (1988). Achieving higher seat belt use: The role of habit in bridging the habit strength. Journal of Applied Social Psychology, 33, 1313−1330.
attitude–behaviour gap. Journal of Applied Social Psychology, 18, 993−1016. Verplanken, B., & Wood, W. (2006). Interventions to break and create consumer habits.
Murgraff, V., Abraham, C., & McDermott, M. (2007). Reducing Friday alcohol Journal of Public Policy and Marketing, 25, 90−103.
consumption among moderate, women drinkers: Evaluation of a brief evidence- Webb, E., Ashton, C. H., Kelly, P., & Kamali, F. (1996). Alcohol and drug use in UK
based intervention. Alcohol and Alcoholism, 42, 37−41. university students. Lancet, 348, 922−925.
Murgraff, V., White, D., & Phillips, K. (1996). Moderating binge drinking: It is possible to
change behaviour if you plan it in advance. Alcohol and Alcoholism, 6, 577−582.