research.edgehill.ac.uk et al.e… · web viewallocating under the influence: effects of alcohol...

Running head: ALCOHOL INTOXICATION AND GROUP BIAS 1

Allocating under the influence: Effects of alcohol intoxication and social identification on in-

group favouritism

Jin Zhoua,1, PhD

Derek Heima, PhD

Rebecca Monka, PhD

Andrew Levya, PhD

Paul Pollardb, PhD

aEdge Hill University, Ormskirk, Lancashire, United Kingdom

bUniversity of Central Lancashire, Preston, Lancashire, United Kingdom

1Corresponding author: Jin Zhou (present address), Centre for Education Statistics and

Evaluation, Department of Education, NSW 2000, Australia, [email protected],

+61(0)295641024

Tables: 4

Figures: 1

mailto:[email protected]

ALCOHOL INTOXICATION AND GROUP BIAS 2

Abstract

The ‘social lubrication’ function of alcohol during interpersonal interactions is well

documented. However, less is known about the effects of alcohol consumption on group-level

behaviour. Empirical findings from social psychological literature suggest that individuals

tend to favour those who are considered as members of their own social group. Not yet

evaluated is how alcohol intoxication interacts with this group-level bias. Therefore, the

current study examined experimentally the effects of intoxication on group bias. Ninety-four

individuals (Mage = 20.18, SD = 2.36, female = 55) were randomly assigned to consume an

alcoholic (n = 48) or a placebo (n = 46) drink before completing manipulated allocation

matrices, a task which measured the distribution of hypothetical monetary awards based on

social groups. Results point to an interaction between drink condition and social

identification, whereby social identification was significantly associated with in-group

favouritism among intoxicated individuals only. Following alcohol consumption, participants

with higher identification with their social group were more likely to demonstrate allocation

strategies that favoured their own group members. However, non-significant effects were

observed for those in the placebo condition. The findings highlight how alcohol intoxication

may facilitate group bias that results from social group identification.

Keywords: alcohol intoxication; social identification; in-group favouritism; group bias


Public significance statement

This study provides laboratory evidence that alcohol intoxication may magnify group biases

that are based on our social group affiliation. When it comes to managing drinking

environments and drinkers’ behaviours, it is important to consider how intoxication affects

group-level processes that may contribute to discriminatory intergroup behaviour.


Disclosures and Acknowledgements

This work was financially supported by an Alcohol Research UK PhD Studentship grant

awarded to the lead author.

All authors have contributed significantly to the writing, analyses, and the interpretations

provided in the paper and have read and approved the final manuscript.

No conflicts of interest are reported.

We would like to thank Maddison Barnes, Lydia Suffling, and Keelan Donohue from Edge

Hill University for their research assistance during this study.

Author Note

This study forms part of a doctoral programme of research and is included in an archived

thesis at the lead author’s institution.


Allocating under the influence: Effects of alcohol intoxication and social identification on in-

group favouritism

Alcohol consumption is often a social activity (Dietler, 2006; Douglas, 1987), and

researchers have begun to examine the mechanisms that may help explain its social

lubricating effects (Frings, Hopthrow, Abrams, Hulbert, & Gutierrez, 2008; Kirchner,

Sayette, Cohn, Moreland, & Levine, 2006; Sayette et al., 2012). However, although alcohol

can prompt social bonding and improve interpersonal interactions (de Visser, Wheeler,

Abraham, & Smith, 2013; Fairbairn, Sayette, Aalen, & Frigessi, 2015; Monahan & Lannutti,

2000), concurrently, its consumption is also associated with interpersonal and intergroup

hostility and violence (Graham et al., 1998; Hunt & Laidler, 2001; Levine, Lowe, Best, &

Heim, 2012). In this respect, remarkably little is known about how alcohol consumption

affects the psychological processes that may underpin adverse group-level behaviours. To

examine this, the present research utilises a classic social psychological experimental

paradigm to examine the psychopharmacological effects of intoxication on group bias.

A considerable amount of research has revealed that alcohol consumption can

powerfully shape a range of social behaviours such as flirtation (Monahan & Lannutti, 2000),

group formation (Kirchner et al., 2006), and even business negotiations (Au & Zhang, 2016).

Researchers have observed groups of unacquainted participants to display an increased

amount of positive behaviours, such as more expressive speech patterns and smiling,

following a moderate dose of alcohol (Sayette et al., 2012). It was also found that smiles were

more likely to be “caught” and reciprocated when the group had consumed alcohol, with this

mutual smiling more pronounced among heavier drinking participants (Fairbairn et al., 2015).

Together, such work indicates that alcohol consumption may increase people's expression of

and response to socially directed behaviours.


Of note from the literature is that acute intoxication can lead to different outcomes

depending on determinants related to both the individual and the contextual cues. Alcohol’s

“myopic” effects are posited to be the result of impaired attentional capacity due to

intoxication (Steele & Josephs, 1990). This alcohol-induced impairment creates a narrowing

focus of attention to the most salient cues in one’s environment. As a consequence, the acute

effects of alcohol may result in, for example, increased cooperative or competitive behaviours

(Hopthrow, Abrams, Frings, & Hulbert, 2007) or increased or decreased aggression

(Giancola, Duke, & Ritz, 2011), depending on the cues in the drinker’s environment. From

this perspective, it is not unreasoned to suggest that the social attributes of alcohol along with

the myopic effects of intoxication can lead a drinker’s attention to the salient intergroup cues

in one’s environment.

The acute effects of alcohol on intergroup responses have been principally examined

in studying racial bias. Some studies suggest that alcohol increases the expression of

prejudice and discrimination based on racial categories (Bartholow, Dickter, & Sestir, 2006;

Reeves & Nagoshi, 1993; Schlauch, Lang, Plant, Christensen, & Donohue, 2009; Schofield,

Unkelbach, & Denson, 2015). For example, Schofield and colleagues (2015) reported

increased participants’ bias to shoot Middle Eastern targets during a computerised shooter

task following alcohol consumption, compared to participants who received a placebo dose.

Relatedly, Mitchell et al. (2015) found that alcohol consumption led to an increase in ingroup

ratings, whereby Caucasian participants were significantly more likely to rate White-faced

stimuli as more attractive and physically healthy after drinking alcohol compared to sober

participants. However, the authors noted no differences in ratings for Black-face stimuli

between sober individuals or following alcohol, suggesting that the effects of alcohol on

intergroup bias in this study may be driven by increases in ingroup liking following

consumption (as opposed to outgroup derogation). With such work, researchers have begun


to emphasise the effect of alcohol in amplifying people’s predispositions of group-related

attitudes and evaluations, rather than universally arousing hostility (Loersch, Bartholow,

Manning, Calanchini, & Sherman, 2015). However, it is increasingly apparent that there is a

paucity of research that is dissociated from the domain of race bias and relations. A crucial

gap remains in how we may apply social psychological theory to better understand

generalised social behaviours when under the influence of alcohol.

One of the most established findings in social psychological literature highlights how

the formation of individuals into groups can prompt the pursuit of in-group identity

enhancement and belonging through actions that favour one’s own group vis-à-vis others

(Diehl, 1990; Hogg & Abrams, 1990; Tajfel, 1982). Tajfel and colleagues first revealed that,

when faced with the task of distributing rewards to anonymous members from two groups,

participants tended to favour the members belonging to their own group, despite this

distribution strategy providing no personal advantage (Tajfel, Billig, Bundy, & Flament,

1971; Tajfel & Turner, 1979).

Since its conception, a substantial bank of research supports the notion that, when

given the opportunity, those with greater identification with their group enact behaviours that

differentiate between their own (in-group) and other groups (out-group) by displaying in-

group favouritism (Gagnon & Bourhis, 1996; Jetten, Spears, & Manstead, 1996; Sidanius,

Pratto, & Mitchell, 1994). This display of bias behaviour is theorised to arise from social

identification, the process whereby people believe they share a group membership with

others, and this has consequences for their interactions within their social environment

(Turner, 1985). In application, Reicher’s analysis of crowd behaviour provides a framework

for understanding the dynamics of unfolding intergroup hostilities through the theoretical lens

of social identity processes (Reicher, 1984, 1996).


The social identity approach has been applied to describe and explain the mechanisms

underpinning a range of behaviours (Haslam, 2014). Most pertinent to the present topic is the

empirical research highlighting the moderating influence of individuals’ social identities on

their alcohol behaviours. For instance, studies find that the extent to which an individual feel

similar and close to their social group(s) affects how strongly referent normative beliefs

around drinking predicts actual personal consumption (Neighbors et al., 2010; Reed, Lange,

Ketchie, & Clapp, 2007). Interventions, such as reducing alcohol intake and drinking

intentions, are found to be more effective when social identities are taken into consideration

(Berger & Rand, 2008; Livingstone & McCafferty, 2015; Tarrant & Butler, 2011). Primarily,

what the findings demonstrate is that behavioural outcomes can be influenced by social

identification. Low identifiers are less likely to respond to group-relevant information, while

high identifiers are more likely to express attitude and behaviour alignment when presented

with the same information (Livingstone, Haslam, Postmes, & Jetten, 2011).

When considering the findings from the social identity and alcohol literature in

combination, it seems likely that alcohol has the potential to magnify group differentiating

behaviours that arise due to social identity mechanisms, as the evidence suggests that

intoxication facilitates people’s automatic or predisposed cognitive processes (Bartholow et

al., 2006; Hopthrow et al., 2007; Loersch et al., 2015). Previous alcohol administration and

group processes research has examined intoxicated social behaviours through a variety of

paradigms (see Frings et al., 2008; Hopthrow et al., 2007; Loersch et al., 2015). However, no

research to date has examined how alcohol affects the underlying psychological processes

related to group bias behaviour that is substantiated in the social identity literature.

The aim of the current study was to examine experimentally the effects of alcohol

intoxication on group bias. Here, we adopt the Tajfel matrices as an operationalised measure

of discrimination in order to examine how alcohol affects intergroup bias behaviour (Bourhis,


Sachdev, & Gagnon, 1994; Tajfel et al., 1971; Turner, Brown, & Tajfel, 1979). We

hypothesised that, relative to those receiving a placebo, intoxicated individuals who identify

with their social group would be more likely to demonstrate group bias in the form of in-

group favouritism.

To test this notion, we recruited participants from the general undergraduate student

cohort and, in addition, purposefully recruited students who were members of a university-

affiliated sports team. Emergent theoretical perspectives suggest that identity processes are

particularly defined and salient for those participating in sport (Rees, Haslam, Coffee, &

Lavallee, 2015; Zhou & Heim, 2014). For instance, sportspeople typically report a strong

identification with their sporting group, and this connectedness is important for both sporting

and social elements of sports involvement (Miller, 2009) and can drive their alcohol

behaviours (Zhou, Heim, & Levy, 2016). Recruiting from a highly identified group such as

those involved in sport was therefore deemed to be a useful strategy for examining the

interaction between intoxication and social group identification, in comparison with a

(hypothetically) less defined group members (undergraduate coursemates).

Methods

Participants

Ethical guidelines ensured that alcohol administration procedures were in line with

recommended practice (NIAAA, 2004) and institutional approval was obtained. Study

recruitment attempted to recruit as many participants as possible within a three-month period

(February to April 2015) and sample sizes were not predetermined. Opportunity sampling

and recruitment yielded 121 individuals interested in participating in the study. The Alcohol

Use Disorders Identification Test (AUDIT; Saunders, Aasland, Babor, & Grant, 1993) was

used to screen the alcohol behaviours of potential participants. The process excluded


respondents with scores of 20 or more (indicating a clinical dependency score of problematic

drinking; Babor, Higgins-Biddle, Saunders, & Monteiro, 2001) and less than three (indicating

non-regular drinkers). Female participants confirmed that they were not pregnant (with

pregnancy tests made available). All participants completed a medical questionnaire to

identify contraindications for alcohol consumption and medications known to react

negatively with alcohol that would warrant medical exclusion.

During participant screening procedures, 15 participants were deemed ineligible due

to medical factors and nine were excluded due to resulting AUDIT scores of 20 and above. A

total of 97 participants were admitted to the study. No statistical analyses were conducted

before the conclusion of data collection. Preliminary data cleaning removed three participants

due to incomplete or incorrect task responses. The final sample used for the main analyses

included 94 participants (55 female, 39 male; 91.6% White British), ranging in age between

18 and 32 years (Mage = 20.18, SD = 2.36). Among the total cohort, 40 were individuals who

indicated currently participating in sport and belonging to a sports club; 54 were non-sporting

undergraduate students. All participants were recruited from the same university.

Design and Materials

Drink conditions

The study implemented a 2 (Group: sports, non-sports) x 2 (Drink: alcohol, placebo)

between-participants design. Random allocation assigned 21 sports participants to the alcohol

condition (placebo = 19). The same procedure assigned 27 non-sports participants to the

alcohol condition (placebo = 27).

In the alcohol condition, the drink mixture was calculated for vodka 37.5% ABV at

0.6 g/kg for males and 0.5 g/kg for females, mixed with equal parts orange juice and tonic


water (methodological replication, see Hopthrow et al., 2007; Rose & Grunsell, 2008). This

measurement of alcohol allows for intoxication at the average level of 0.35 Breath Alcohol

Concentration (BrAC): This converts to 0.08% Blood Alcohol Concentration (BAC), the UK

drink-and-drive limit. The placebo drink was a mixture of orange juice, tonic water, and 2ml

of surface vodka with a vodka ‘mist’ sprayed across the glass to support the placebo

manipulation. To further support the alcohol-placebo manipulation, all participants ingested a

strong mint lozenge prior to drinks consumption in order to disguise the flavour of the

beverages (e.g., Frings et al., 2008; Hopthrow et al., 2007). The mixture was divided into two

drinks of equal quantities, and participants were instructed to consume each drink spaced

across 3-4 minutes to help ensure a consistent drinking pace across all participants.

Task materials

The task booklet presented to participants contained a series of allocation matrices.

Each matrix consisted of 13 boxes containing pairs of numbers, displaying the choices of

possible payoffs. The receiver of the pay-offs, identified only by their group membership,

was either an in-group member (‘a fellow team- or course-mate’) or an out-group member (‘a

member from another institution’). For each matrix, participants chose a distribution pairing

that best determines how they wished to allocate the monetary payoffs. Hence, the matrices

provided a situation where moving from one end to the other offered a range of strategies that

required participants to decide to what extent they wished to maximise and compromise on

allocation amounts. Three types of matrices were used to pit different strategies of

distribution against each other to measure the strength, or ‘pull’, of specific strategies (see

Online Supplementary Material for illustrated examples; see also Bourhis et al., for a

comprehensive discussion of matrix construction and pull scores calculation):


(a) In-group Favouritism (FAV = MD + MIP) versus Maximum Joint Profit (MJP) – distinguishes

between allocations that reflects in-group favouritism in terms of allocating the most amount

available for in-group profit while maximising the difference between the amount allocated to the

out-group, against allocating the maximum amount of total points for both groups when combined.

(b) Maximising Difference in favour of in-group (MD) versus Maximum In-group Profit (MIP) –

distinguishes between allocations that maximise the quantity difference between groups in a way

that favours the in-group, against awarding the highest amount available for the in-group overall.

(c) Parity (P) versus In-group Favouritism (FAV) – distinguishes between allocations that give

numerically equal amounts to each group, against in-group favouritism in terms of providing

maximum amount available for in-group profit while maximising the difference between the

amounts allocated to the out-group.

We presented each of the aforementioned matrices twice using the same number sequencing

as Turner and Tajfel (1979) and as exampled in Bourhis et al. (1994), with the labels of “in-

group” and “out-group” inverted on the second presentation for counter-balancing. In

addition, we included a direct measure of in/out-group favouritism: (d) Direct FAV, with a

matrix that moved from out-group favouritism, through to parity, through to in-group

favouritism (Moghaddam & Stringer, 1986). Finally, each matrix type was repeated with a

different sequence of numbers (although pairing strategies were unchanged). Thus, in total,

each participant completed 14 matrices. We report all measures, manipulations, and

exclusions in this study.

Procedure

There was no mention of the group bias interests of the research during the

recruitment process and throughout the experiment to ensure that participants were unaware

of the intent of the matrices. Instead, to explain the use of the allocation matrices, the study

protocol informed participants that the experiment sought to explore the effects of alcohol on

economic-related decision making.


Recruited respondents completed the medical screening to determine their eligibility

for the study. Once confirmed, participants were requested to abstain from consuming alcohol

for 12 hours and to refrain from eating for three hours prior to attending their designated

laboratory session. Due to the consumption requirements, participants were also required to

consent to abstain from driving or exercising following study participation. Upon arrival,

participants completed consent forms and demographics questionnaire and were weighed

(kg). Using the Lion SD400 Alcometer®, participants provided a breathalyser reading to

check for baseline BrAC and confirm none had consumed alcohol before the study (all study

participants scored 0).

Following this, participants completed a questionnaire that contained items asking

them to confirm their group membership. The sports group members indicated the sport they

played and the sports club to which they belonged. The non-sports group members identified

the degree/course in which they were currently enrolled. Social identification was assessed

across three items using a seven-point Likert scale (1=strongly disagree, to 7=strongly agree):

“I have a lot in common with other members of my sports club/course”, “I feel strong ties to

other members of my sports club/course”, and “In general, being a sportsperson/student is an

important part of my self-image” (items taken from Cameron, 2004). The three items were

averaged to provide a single index of social identification (current α = .72).

The consumption phase lasted approximately 10 minutes. During the subsequent

absorption phase, participants left alone to watch an American sitcom television show for 20

minutes as a distraction activity. Following this, a further breathalysation to check for alcohol

intoxication was performed. The results of these readings were purposely masked from the

participants across all the conditions to uphold the placebo condition and so that none knew

their current BrAC levels prior to the commencement of the task.


We framed the task as an economic-related exercise that gave participants the

responsibility of apportioning a hypothetical fund. Participants were instructed to use the

matrices to divide funding points between two individuals, only identified by their

designation as a fellow sports teammate /coursemate or someone from an external institution.

We instructed the participants to convert the points into monetary terms, whereby one point

was valued at £10 [approximately $15]. Thus, instead of distributing one and seven points,

for example, the ‘real’ amounts were £10 and £701. Instructions also stressed that these

allocations would be anonymous, and participants could not award these points to

themselves. No time constraints were in place and participants were left alone to complete the

task (approximated average task time to be five minutes). Only after completing the task

booklet were participants asked to provide a subjective intoxication rating via the scale “how

intoxicated do you feel right now?” (1=not at all, to 10=extremely), to check for the alcohol

manipulation.

On completion of the task, participants were fully debriefed on the true aims of the

study and compensated for the hour session (£6 [approximately $8] or undergraduate course

credits). In line with standard practice (NIAAA, 2004), those in the alcohol condition were

breathalysed and permitted to leave the laboratory when their BrAC scored below 0.14

(converts to a BAC level of 0.028%). Entertainment and refreshments were made available

while the participants waited for their BrAC to fall: disclaimers were signed if participants

wished to remove themselves from the study area before they had reached this recommended

level.

Task analysis

1 This methodological step was included in order to address previous concerns of using point allocations for distribution versus monetary amounts (Gaertner & Insko, 2001). Specifically, it has been argued that allocations represented by arbitrary points are relatively valued only against a quantity possessed by either group (Rabbie & Schot, 1990; Rabbie, Schot, & Visser, 1989). However, the allocation of money represents absolute terms where its value does not depreciate in relation to distribution choices. In an effort to address these methodological considerations, the current study instructed participants to consider both representations during the decision-making task.


Each matrix choice (the paired allocations) was scored in terms of ranks from zero to

12; the ‘pull’ of one strategy will receive the highest, with the competing strategy receiving

the lowest score (Bourhis et al., 1994; Tajfel & Turner, 1979). When presented with the

labels of “in-group” and “out-group” inverted, the matrix score calculations ranged between -

12 and zero. Therefore, each strategy score (excluding Direct FAV) had a theoretical range of

-12 to 12, where zero epitomised parity (equal distribution), positive scores represented a

greater exhibition of the strategy, and negative scores characterised a greater intent to avoid

the strategy. For matrix type (d) Direct FAV, the intent was to directly measure in-/out-group

favouritism without the presence of a competing strategy, i.e., it was impossible to maximise

joint profits (Moghaddam & Stringer, 1986). Here, scores ranged from zero to 12, where zero

represented extreme out-group favouritism and 12 represented extreme in-group favouritism,

with parity scored at six. Responses from this matrix were analysed separately in the

manipulation checks.

Analytic strategy

Preliminary univariate tests compared group and drink conditions across the study

variables. Missing values were handled as listwise deletions. A within-treatment analysis of

the matrix types was conducted to determine if the ‘pull’ score from participants were

significantly different from 0 on the theoretical range of -12 to 12 (as described in Task

Analysis above). That is, the analysis determined whether the participant adopted a strategy

that pulled them away from parity (i.e., a score of 0) and engaged with the differentiation

strategies presented within each matrix. Prior to the main analyses, an examination of social

identification z-scores revealed one participant scoring more than three standard deviations

below the mean. The subsequent factor analysis and regression modelling was performed on

the full cohort and again with the outlier removed. No differences were observed across

outcomes and therefore, for completeness, the results with the full sample are presented.


An exploratory factor analysis was used to reduce the seven allocation strategies into

latent factors in order to simplify analysis and interpretation (e.g., Jetten et al., 1996). For the

main analyses, interaction terms were calculated with mean centred continuous variables and

dummy-coded categorical variables. Main and interaction effects were entered into a

regression model examining the association between the study variables on the extracted

latent allocation strategies identified in the factor analysis. All multivariate analyses were

conducted with bootstrapping to fit significant models onto 10,000-sampled population to

adjust for unequal samples and non-normal distributions within our dependent measures.

Results

For those in the alcohol condition, BrAC readings taken after the absorption phrase

indicated that the intoxication level for the current study protocol was reached (M = .33, SD =

.09). BrAC readings confirmed no detectable intoxication levels among those in the placebo

condition. Those in the alcohol condition (M = 5.19, SD = 1.94) perceived themselves to be

significantly more intoxicated than the placebo group (M = 2.02, SD = 1.89), F(1,90) = 59.60,

p < .001, ηp2 = .398. No differences were found on subjective intoxication ratings between

cohort groups.

Two-way ANOVAs were performed to assess cohort and condition differences on

study dependent variables (see Table 1). The sports participants (M = 6.04, SD = .80) scored

significantly higher on social identification than the non-sports cohort (M = 5.30, SD = 1.04),

F(1,87) = 12.77, p = .001, ηp2 = .128. No differences were found on social identification

scores between drink conditions. A check for gender differences revealed no significant

differences among our study variables (inter-variable correlations can be found in the

supplementary material).

[TABLE 1 HERE]


Strategy analyses within each treatment condition

A Wilcoxon matched pairs test analysed the differences among the ranked paired

allocations across the three matrix types (see Table 2). The pull of parity (P on FAV) was

statistically significant across all conditions, with the pull of fairness over in-group

favouritism representing the most utilised strategy for all groups. This suggests that when

faced with the choice of distributing in an equal manner or a differentiating manner, most

participants tended to opt for fairness. There was also a significant pull of in-group

favouritism across all groups when pitted against maximum joint profit (FAV on MJP),

suggesting that when presented with the choice of favouring one’s in-group at the expense of

the out-group, most participants would demonstrate some pull towards in-group favouritism.

[TABLE 2 HERE]

Factor analysis

Exploratory factor analysis was performed on the seven allocation strategies to

identify the underlying latent relationships representing group bias. Maximum likelihood

factor analysis performed with varimax rotation opted for an orthogonal rotation in

anticipation of emergent factors (in-group vs. out-group favouritism) to be uncorrelated.

Kaiser-Meyer-Olkin statistic reflected a good sample size adequate for factor analysis (KMO

= .76) and Bartlett’s test of sphericity indicated that the correlations between items were

sufficiently large for exploratory factor analysis, χ2 (21) = 303.83, p < .001.

The analysis revealed a two-factor solution, which accounted for 69.60% of the initial

variance and 56.97% of the extracted variance. The clustered strategies suggest that the first

factor contained four items that measured the pull of in-group favouritism strategies (see


Table 3). This factor accounted for 48.98% of the initial variance and 44.39% of the extracted

variance. The second factor contained two items that measured the pull of overall profit gains

(profit maximising), accounting for an additional 20.62% of the initial variance and 12.58%

of the extracted variance. As the two identified factors were theoretically divergent strategies,

the Anderson-Rubin method calculated factor scores representing each participant’s

placement on the factors identified in the extraction (Anderson & Rubin, 1956).

[TABLE 3 HERE]

Interactions between drink condition and social identification

A regression model was computed for each extracted factor. Interaction terms were

derived from mean-centred social identification scores multiplied with dummy-coded groups

(control = 0, sports = 1; placebo = 0, alcohol = 1). Step 1 involved the main effects (group,

condition), Step 2 added social identification scores, and their interactions terms were

included in Step 3.

A significant regression model (see Table 4) was found for in-group favouritism,

F(5,84) = 12.53, p = .028, R2 = 13.8% (see Table 4). In the final step, only the interaction

between drink condition and social identification was significant, b = .42, p = .028, CI

[.08, .83]. Examination of the interaction slopes revealed that social identification was

associated with increased in-group favouritism for those in the alcohol condition only (b

= .56, p < .002, CI [.21, 390]. Non-significant effects were observed for those in the placebo

condition (b = .14, p = .211, CI [-.08, .36]. A check of simple contrasts revealed non-

significant differences on in-group favouritism between those in the alcohol versus the

placebo condition at above (+1 SD; p = .135) and below (-1 SD; p = .095) the mean value of

social identification. Posthoc power analysis was conducted using G*Power 3.1.9.2 on the


significant effects revealed. This found that the power (1 – β error probability) was .71 –

acceptably close to the traditional standard of 0.8.

A separate regression analysis performed on profit maximising resulted in a non-

significant model and bared no significant mains or interaction effects in the final model.

Finally, posthoc probes were conducted to examine differences due to gender or group

membership. No main (p = .775) or two- or three-interaction effects (all p’s > .10) were

found for gender. Three-way interaction checks for drink condition x group x social

identification effects also yielded non-significant findings (p = .345).

[TABLE 4 HERE]

[FIGURE 1 HERE]

Discussion

The current study implemented an experimental design to examine the effects of

intoxication on group bias behaviour. The significant interaction between drink condition and

social identification demonstrated that participants who reported higher identification with

their social group were significantly more likely to engage in distribution strategies that

favoured their in-group members following alcohol consumption. However, this pattern was

not observed among those in the placebo condition. Results also indicate that there are lower

levels of in-group favouritism among those less identified with their group following alcohol

intoxication. In this regard, low identifiers may be insufficiently interested in or aware of

their group membership to engage in group discriminating strategies. The social identity

literature does suggest that significant in-group devaluation can occur when individuals do

not feel strong ties to their social group (Karasawa, 1991). If we consider the myopic effects

of alcohol, the group-level differentiation strategies required during the study’s task may


have accentuated low identifiers’ lack of group identification and resulted in a reduced

tendency to engage in group bias behaviour. In-line with the alcohol myopic model (Steele &

Josephs, 1990), the findings of the present study indicate that the psychopharmacological

effects of alcohol may perhaps facilitate drinkers’ preferences (or lack of) for their social

group members.

While previous research outlines how alcohol consumption can increase positive

perceptions of group relations (Kirchner et al., 2006; Sayette et al., 2012), the current study

contributes to this body of literature by suggesting how group bias behaviour may also occur

following intoxication. These results may help to provide a theoretically grounded

understanding of intoxicated social behaviours. Favouring members from one’s own group

over out-group members may provide individuals with a greater sense of in-group identity

and positive self-esteem (Amiot, Sansfaçon, & Louis, 2014; Hogg & Abrams, 1988; Lemyre

& Smith, 1985; Oakes & Turner, 1980). However, these processes and motivations also

underpin the literature that describes the antecedents of intergroup discrimination and conflict

(Tajfel, 1982). The current findings suggest that such group-level processes may be

heightened following intoxication and, when taken together with theoretical applications and

evidence from social psychological literature, begin to identify the mechanisms through

which intoxicated social behaviours can precipitate intergroup hostilities (Ostrowsky, 2014;

Schofield et al., 2015). By indicating that intoxication can heighten (or disinhibit) intergroup

bias, particularly among those with strong and salient group identities, our findings suggest

that strategies aimed at minimising alcohol-fuelled intergroup hostilities may be usefully

informed by considering these group-level cognitions and responses. Delineating how alcohol

consumption may facilitate an ‘us’ versus ‘them’ mentality (Tajfel, 1982) potentially offers a

complementary approach to interventions aimed at tackling alcohol-related violence which

hitherto typically target individual drinkers (Plant, Plant, & Thornton, 2002).


It should be noted that, in the present study, there was no inclusion of a non-alcoholic

drink control condition. To clarify, the current study specifically sought to examine how

alcohol consumption affects group-level behaviour. With this research aim in mind, it was

deemed appropriate to use the placebo manipulation to compare with the

psychopharmacological impact of intoxication on participants’ task responses. The current

study employed a number methodological steps to ensure an effective placebo manipulation

and, at the debrief, participants in the placebo condition were asked whether they suspected

the deception with none disclosing any strong suspicions. However, the lack of the expected

behavioural responses on the task for those in the placebo condition suggest that there may be

some degree of compensatory response where our participants’ expectancies around the acute

effects of alcohol consumption lead to behavioural monitoring to counteract such effects

(Marczinski & Fillmore, 2005; Schlauch et al., 2010). Given the small sample, and the

methodological obscuring of the veritable alcohol effects on group bias in the present study,

further research incorporating a non-alcohol condition is necessary in order to interpret fully

the psychopharmacological effects of intoxication from the expectancy-based alcohol effects

(Testa et al., 2006) in this paradigm.

Although the present findings indicate that intoxicated group members were more

likely to exhibit in-group favouring behaviour, we also acknowledge that behaviours are

often influenced by social (Abrams, Hopthrow, Hulbert, & Frings, 2006) and contextual

(Monk & Heim, 2013) cues. This interaction between alcohol and social processes requires

further interrogation away from a laboratory-based setting in order to evaluate how groups,

and individuals in groups, react to interpersonal interactions and the social context

(Hopthrow, Randsley de Moura, Meleady, Abrams, & Swift, 2014; Monk & Heim, 2014;

Sayette et al., 2012). Relatedly, there is existing commentary on the individual difference

factors that may affect the individual’s behaviours and/or their response to intoxication itself


(see Sher & Wood, 2005). Previous research notes that, when in a similar economic-decision

making paradigm, individuals who demonstrate psychopathic traits, such as recklessness and

impulsivity, offer more money to their ingroup (members affiliated with their university),

relative to an outgroup (Gillespie, Mitchell, Johnson, Dawson, & Beech, 2013). Additionally,

aggression following alcohol is more likely among participants who have high levels of

dispositional aggressivity (Giancola, 2002). These form alternative perspectives on how

individual differences may factor in how intergroup behaviours are evaluated and expressed.

Finally, while the social identification items in the present study were selected to

represent the proposed three dimensions of social identification (Cameron, 2004), it should be

noted that there are short validated scales that have since been developed to assess social

identification (Postmes, Haslam, & Jans, 2013; Reysen, Katzarska‐Miller, Nesbit, & Pierce,

2013). Future research could take advantage of the brevity of these validated scales when

examining these socio-cognitive processes alongside alcohol intoxication in the field.

To conclude, the current study found that in-group bias behaviours increased among

intoxicated individuals who identified with their social group. The current research presents a

first look at the interplay between the psychopharmacological state of intoxication and social

identity mechanisms, and explores a novel yet fundamental social psychological perspective

of the behavioural consequences associated with alcohol. Replication and further

experimental research are necessary to investigate further how various contexts and alcohol

states interact with social identity processes to qualify the findings of this preliminary study.


References

Abrams, D., Hopthrow, T., Hulbert, L., & Frings, D. (2006). ‘Groupdrink’? The effect of alcohol on

risk attraction among groups versus individuals. Journal of Studies on Alcohol, 67(4), 628–

636.

Amiot, C. E., Sansfaçon, S., & Louis, W. R. (2014). How normative and social identification

processes predict self‐determination to engage in derogatory behaviours against outgroup

hockey fans. European Journal of Social Psychology, 44(3), 216–230.

https://doi.org/10.1002/ejsp.2006

Anderson, T. W., & Rubin, H. (1956). Statistical inference in factor analysis (Vol. 5, p. 1). Presented

at the Proceedings of the third Berkeley symposium on mathematical statistics and

probability.

Au, P. H., & Zhang, J. (2016). Deal or no deal? The effect of alcohol drinking on bargaining. Journal

of Economic Behavior & Organization, 127, 70–86.

Babor, T. F., Higgins-Biddle, J. C., Saunders, J. B., & Monteiro, M. G. (2001). The Alcohol Use

Disorders Identification Test (AUDIT): Guidelines for use in primary care. Geneva: World

Health Organisation.

Bartholow, B. D., Dickter, C. L., & Sestir, M. A. (2006). Stereotype activation and control of race

bias: cognitive control of inhibition and its impairment by alcohol. Journal of Personality and

Social Psychology, 90(2), 272–287.

Berger, J., & Rand, L. (2008). Shifting signals to help health: Using identity signaling to reduce risky

health behaviors. Journal of Consumer Research, 35(3), 509–518.

https://doi.org/10.1086/587632

Bourhis, R. Y., Sachdev, I., & Gagnon, A. (1994). Intergroup research with the Tajfel matrices:

Methodological notes (Vol. 7, pp. 209–232). Presented at the The psychology of prejudice:

The Ontario symposium, Erlbaum Mahwah.

Cameron, J. E. (2004). A three-factor model of social identity. Self and Identity, 3(3), 239–262.


de Visser, R. O., Wheeler, Z., Abraham, C., & Smith, J. A. (2013). ‘Drinking is our modern way of

bonding’: Young people’s beliefs about interventions to encourage moderate drinking.

Psychology & Health, 28(12), 1460–1480. https://doi.org/10.1080/08870446.2013.828293

Diehl, M. (1990). The minimal group paradigm: Theoretical explanations and empirical findings.

European Review of Social Psychology, 1(1), 263–292.

Dietler, M. (2006). Alcohol: anthropological/archaeological perspectives. Annu. Rev. Anthropol., 35,

229–249.

Douglas, M. (1987). Constructive Drinking: Perspectives on Drink from Anthropology. Cambridge:

Cambridge University Press.

Fairbairn, C. E., Sayette, M. A., Aalen, O. O., & Frigessi, A. (2015). Alcohol and Emotional

Contagion An Examination of the Spreading of Smiles in Male and Female Drinking Groups.

Clinical Psychological Science, 3(5), 686–701. https://doi.org/10.1177/2167702614548892

Frings, D., Hopthrow, T., Abrams, D., Hulbert, L., & Gutierrez, R. (2008). Groupdrink: The effects of

alcohol and group process on vigilance errors. Group Dynamics: Theory, Research, and

Practice, 12(3), 179.

Gaertner, L., & Insko, C. A. (2001). On the measurement of social orientations in the minimal group

paradigm: Norms as moderators of the expression of intergroup bias. European Journal of


Gagnon, A., & Bourhis, R. Y. (1996). Discrimination in the minimal group paradigm: Social identity

or self-interest? Personality and Social Psychology Bulletin, 22(12), 1289–1301.

Giancola, P. R. (2002). Alcohol-related aggression in men and women: the influence of dispositional

aggressivity. Journal of Studies on Alcohol, 63(6), 696–708.

Giancola, P. R., Duke, A. A., & Ritz, K. Z. (2011). Alcohol, violence, and the alcohol myopia model:

Preliminary findings and implications for prevention. Addictive Behaviors, 36(10), 1019–

1022.

Gillespie, S. M., Mitchell, I. J., Johnson, I., Dawson, E., & Beech, A. R. (2013). Exaggerated

intergroup bias in economical decision making games: differential effects of primary and

secondary psychopathic traits. PloS One, 8(8), e69565.


Graham, K., Leonard, K. E., Room, R., Wild, T. C., Pihl, R. O., Bois, C., & Single, E. (1998). Current

directions in research on understanding and preventing intoxicated aggression. Addiction,

93(5), 659–676.

Haslam, S. A. (2014). Making good theory practical: Five lessons for an Applied Social Identity

Approach to challenges of organizational, health, and clinical psychology. British Journal of


Hogg, M. A., & Abrams, D. (1988). Social identifications: A social psychology of intergroup

relations and group processes. London, UK: Taylor & Frances/Routledge.

Hogg, M. A., & Abrams, D. (1990). Social motivation, self-esteem and social identity. In Social

Identity Theory: Constructive and Critical Advances (pp. 28–47). New York: Harvester

Wheatsheaf.

Hopthrow, T., Abrams, D., Frings, D., & Hulbert, L. G. (2007). Groupdrink: The effects of alcohol on

intergroup competitiveness. Psychology of Addictive Behaviors, 21(2), 272–276.

Hopthrow, T., Randsley de Moura, G., Meleady, R., Abrams, D., & Swift, H. J. (2014). Drinking in

social groups. Does ‘groupdrink’provide safety in numbers when deciding about risk?

Addiction, 109(6), 913–921.

Hunt, G. P., & Laidler, K. J. (2001). Alcohol and violence in the lives of gang members. Alcohol

Research and Health, 25(1), 66–71.

Jetten, J., Spears, R., & Manstead, A. S. (1996). Intergroup norms and intergroup discrimination:

distinctive self-categorization and social identity effects. Journal of Personality and Social

Psychology, 71(6), 1222–1233.

Karasawa, M. (1991). Toward an assessment of social identity: The structure of group identification

and its effects on in‐group evaluations. British Journal of Social Psychology, 30(4), 293–307.

Kirchner, T. R., Sayette, M. A., Cohn, J. F., Moreland, R. L., & Levine, J. M. (2006). Effects of

alcohol on group formation among male social drinkers. Journal of Studies on Alcohol and

Drugs, 67(5), 785–793.

Lemyre, L., & Smith, P. M. (1985). Intergroup discrimination and self-esteem in the minimal group

paradigm. Journal of Personality and Social Psychology, 49(3), 660–670.


Levine, M., Lowe, R., Best, R., & Heim, D. (2012). ‘We police it ourselves’: Group processes in the

escalation and regulation of violence in the night‐time economy. European Journal of Social

Psychology, 42(7), 924–932. https://doi.org/10.1002/ejsp.1905

Livingstone, A. G., Haslam, S. A., Postmes, T., & Jetten, J. (2011). ‘We Are, Therefore We Should’:

Evidence That In‐Group Identification Mediates the Acquisition of In‐Group Norms1.

Journal of Applied Social Psychology, 41(8), 1857–1876.

Livingstone, A. G., & McCafferty, S. (2015). Explaining reactions to normative information about

alcohol consumption: A test of an extended social identity model. International Journal of

Drug Policy, 26(4), 388–395.

Loersch, C., Bartholow, B. D., Manning, M., Calanchini, J., & Sherman, J. W. (2015). Intoxicated

prejudice: The impact of alcohol consumption on implicitly and explicitly measured racial

attitudes. Group Processes & Intergroup Relations, 18(2), 256–268.

https://doi.org/10.1177/1368430214561693

Marczinski, C. A., & Fillmore, M. T. (2005). Compensating for alcohol-induced impairment of

control: effects on inhibition and activation of behavior. Psychopharmacology, 181(2), 337–

346.

Miller, K. E. (2009). Sport-related identities and the ‘toxic jock’. Journal of Sport Behavior, 32(1),

69–91.

Mitchell, I. J., Gillespie, S. M., Leverton, M., Llewellyn, V., Neale, E., & Stevenson, I. (2015). Acute

alcohol consumption and secondary psychopathic traits increase ratings of the attractiveness

and health of ethnic ingroup faces but not outgroup faces. Frontiers in Psychiatry, 6, 25.

Moghaddam, F. M., & Stringer, P. (1986). Trivial and important criteria for social categorization in

the minimal group paradigm. The Journal of Social Psychology, 126(3), 345–354.

Monahan, J. L., & Lannutti, P. J. (2000). Alcohol as social lubricant. Human Communication

Research, 26(2), 175–202.

Monk, R. L., & Heim, D. (2013). Environmental context effects on alcohol-related outcome

expectancies, efficacy, and norms: a field study. Psychology of Addictive Behaviors, 27(3),

814–818. https://doi.org/10.1037/a0033948


Monk, R. L., & Heim, D. (2014). A systematic review of the Alcohol norms literature: A focus on

context. Drugs: Education, Prevention and Policy, 21(4), 263–282.

Neighbors, C., LaBrie, J. W., Hummer, J. F., Lewis, M. A., Lee, C. M., Desai, S., … Larimer, M. E.

(2010). Group identification as a moderator of the relationship between perceived social

norms and alcohol consumption. Psychology of Addictive Behaviors, 24(3), 522.

Oakes, P. J., & Turner, J. C. (1980). Social categorization and intergroup behaviour: Does minimal

intergroup discrimination make social identity more positive? European Journal of Social

Psychology, 10(3), 295–301.

Ostrowsky, M. K. (2014). The social psychology of alcohol use and violent behavior among sports

spectators. Aggression and Violent Behavior, 19(4), 303–310.

Plant, M., Plant, M., & Thornton, C. (2002). People and places: some factors in the alcohol-violence

link. Journal of Substance Use, 7(4), 207–213.

Postmes, T., Haslam, S. A., & Jans, L. (2013). A single‐item measure of social identification:

Reliability, validity, and utility. British Journal of Social Psychology, 52(4), 597–617.

Rabbie, J. M., & Schot, J. C. (1990). Group behavior in the minimal group paradigm: Fact or fiction?

In European Perspectives in Psychology: Work and Organizational, Social and Economic,

Cross-cultural (Vol. 3, pp. 251–263). Oxford, England: John Wiley & Sons.

Rabbie, J. M., Schot, J. C., & Visser, L. (1989). Social identity theory: A conceptual and empirical

critique from the perspective of a behavioural interaction model. European Journal of Social

Psychology, 19(3), 171–202.

Reed, M. B., Lange, J. E., Ketchie, J. M., & Clapp, J. D. (2007). The relationship between social

identity, normative information, and college student drinking. Social Influence, 2(4), 269–

294. https://doi.org/10.1080/15534510701476617

Rees, T., Haslam, S. A., Coffee, P., & Lavallee, D. (2015). A social identity approach to sport

psychology: Principles, practice, and prospects. Sports Medicine, 45(8), 1083–1096.

https://doi.org/10.1007/s40279-015-0345-4

Reeves, S. B., & Nagoshi, C. T. (1993). Effects of alcohol administration on the disinhibition of racial

prejudice. Alcoholism: Clinical and Experimental Research, 17(5), 1066–1071.


Reicher, S. (1984). The St. Pauls’ riot: An explanation of the limits of crowd action in terms of a

social identity model. European Journal of Social Psychology, 14(1), 1–21.

Reicher, S. (1996). ‘The Battle of Westminster’: developing the social identity model of crowd

behaviour in order to explain the initiation and development of collective conflict. European

Journal of Social Psychology, 26(1), 115–134.

Reysen, S., Katzarska‐Miller, I., Nesbit, S. M., & Pierce, L. (2013). Further validation of a single‐item

measure of social identification. European Journal of Social Psychology, 43(6), 463–470.

Rose, A. K., & Grunsell, L. (2008). The subjective, rather than the disinhibiting, effects of alcohol are

related to binge drinking. Alcoholism: Clinical and Experimental Research, 32(6), 1096–

1104.

Saunders, J. B., Aasland, O. G., Babor, T. F., & Grant, M. (1993). Development of the alcohol use

disorders identification test (AUDIT): WHO collaborative project on early detection of

persons with harmful alcohol consumption‐II. Addiction, 88(6), 791–804.

Sayette, M. A., Creswell, K. G., Dimoff, J. D., Fairbairn, C. E., Cohn, J. F., Heckman, B. W., …

Moreland, R. L. (2012). Alcohol and Group Formation: A Multimodal Investigation of the

Effects of Alcohol on Emotion and Social Bonding. Psychological Science, 23(8), 869–878.

Schlauch, R. C., Lang, A. R., Plant, E. A., Christensen, R., & Donohue, K. F. (2009). Effect of

alcohol on race-biased responding: The moderating role of internal and external motivations

to respond without prejudice. Journal of Studies on Alcohol and Drugs, 70(3), 328–336.

Schlauch, R. C., Waesche, M. C., Riccardi, C. J., Donohue, K. F., Blagg, C. O., Christensen, R. L., &

Lang, A. R. (2010). A meta-analysis of the effectiveness of placebo manipulations in alcohol-

challenge studies. Psychology of Addictive Behaviors, 24(2), 239–253.

Schofield, T. P., Unkelbach, C., & Denson, T. F. (2015). Alcohol consumption increases bias to shoot

at Middle Eastern but not White targets. Group Processes & Intergroup Relations, Early

Online Print. https://doi.org/10.1177/1368430215603461

Sher, K. J., & Wood, M. D. (2005). Subjective effects of alcohol II: Individual differences. In Mind

altering Drugs: Scientific Evidence for Subjective Experience. M. Earleywine (Ed.). (pp. 135–

153). New York: Oxford Press.


Sidanius, J., Pratto, F., & Mitchell, M. (1994). In-group identification, social dominance orientation,

and differential intergroup social allocation. The Journal of Social Psychology, 134(2), 151–

167.

Steele, C. M., & Josephs, R. A. (1990). Alcohol myopia: Its prized and dangerous effects. American

Psychologist, 45(8), 921–933.

Tajfel, H. (1982). Social psychology of intergroup relations. Annual Review of Psychology, 33(1), 1–

39.

Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social categorization and intergroup

behaviour. European Journal of Social Psychology, 1(2), 149–178.

Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In The Social

Psychology of Intergroup Relations (pp. 33–47). Monterey, CA: Brooks/Cole.

Tarrant, M., & Butler, K. (2011). Effects of self‐categorization on orientation towards health. British

Journal of Social Psychology, 50(1), 121–139. https://doi.org/10.1348/014466610X511645

Testa, M., Fillmore, M. T., Norris, J., Abbey, A., Curtin, J. J., Leonard, K. E., … George, W. H.

(2006). Understanding alcohol expectancy effects: Revisiting the placebo condition.

Alcoholism: Clinical and Experimental Research, 30(2), 339–348.

Turner, J. C. (1985). Social categorization and the self-concept: A social cognitive theory of group

behavior. Advances in Group Processes, 2, 77–122.

Turner, J. C., Brown, R. J., & Tajfel, H. (1979). Social comparison and group interest in ingroup

favouritism. European Journal of Social Psychology, 9(2), 187–204.

Zhou, J., & Heim, D. (2014). Sports and Spirits: A Systematic Qualitative Review of Emergent

Theories for Student-Athlete Drinking. Alcohol and Alcoholism, 49(6), 604–617.

Zhou, J., Heim, D., & Levy, A. (2016). Sports Participation and Alcohol Use: Associations With

Sports-Related Identities and Well-Being. Journal of Studies on Alcohol and Drugs, 77(1),

170–179. https://doi.org/10.15288/jsad.2016.77.170


Table 1.Participant demographics and scores across study variables.

Study variables Conditions

Non-sports Sports Placebo Alcohol

Age, mean (SD) 20.22 (2.87) 20.13 (1.45) 20.46 (2.90) 19.92 (1.68)

Gender, female (%) 38 (70.4%) 17 (42.5%) 26 (56.5%) 29 (60.4%)

Social identification, mean (SD) 5.30 (1.04) 6.04 (.80) 5.52 (1.19) 5.69 (.82)

BrAC, mean (SD) 0.34 (.09) 0.33 (.08) 0.00 0.33 (.09)

Subjective intoxication, mean (SD) 2.04 (1.89) 5.19 (1.94)

Non-sports - - 2.00 (1.66) 5.52 (2.12)

Sports - - 2.11 (2.21) 4.72 (1.64)

Note. BrAC = Breath Alcohol Concentration taken after absorption phase.


Table 2.Means pull scores of participants' matrix strategies as a function of condition (placebo vs. alcohol) and group (non-sports vs. sports participants).

Study conditions

Allocation strategy Non-sports Sports Placebo Alcohol

FAV on MJPMJP on FAV

1.92 (3.12)*0.32 (1.57)

3.56 (3.92)**-0.14 (1.80)

2.30 (3.07)**0.13 (2.07)

2.92 (3.98)**0.13 (1.20)

MD on MIPMIP on MD

1.32 (2.76)1.17 (3.83)

2.74 (3.84)*-0.19 (4.15)

1.71 (3.13)0.76 (4.85)

2.11 (3.50)*0.45 (3.01)

FAV on PP on FAV

1.72 (3.75) 8.15 (3.51)**

2.38 (3.89) 8.76 (3.77)**

1.86 (2.93) 9.07 (3.10)**

2.14 (4.55) 7.76 (4.00)**

Direct FAV₸ 7.98 (1.92) 8.75 (2.01) 8.31 (1.78) 8.30 (2.19)

NOTE: Possible pull scores for each strategy range from -12 to 12. FAV = in-group favouritism; MJP = maximum joint profit; MD = maximum difference; MIP = maximum in-group profit; P = parity.

₸ Separate direct measure of favouritism. Scores range from 0 to 12.* p < .01; **p < .001; Wilcoxon matched-pairs test (two-tailed)


Table 3.Rotated factor matrix (maximum likelihood) with orthogonal component loading (varimax with Kaiser Normalization) for seven distribution strategies (N = 94).

Factor

(1) In-group favouritism (2) Profit maximising

Fav on MJP .95 -.11

Direct FAV .84 -.24

FAV on P .80 .09

P on FAV -.65 -.15

MD on MP .59 -.17

MP on MD .12 .65

MJP on FAV -.21 .61

Note. Values in boldface type indicate the item’s primary factor loading.


Table 4.Hierarchical regression model to compare mains effect of group and drink condition, and interaction effects on in-group favouritism, bias-corrected 95% confidence intervals indicated in brackets.

(1) In-group favouritism

b [BCa 95% CI]

Step 1

Group .31 [-0.08-0.70]

Drink condition .07 [-0.30-0.45]

Step statistics F(2,88) = 1.27, p = .286, R2 = .028

Step 2

Group .12 [-0.29-0.57]


Social identification .24 [0.08-0.44]**

Step statistics F(1,87)=5.75, p = .019, ΔR2 = .06

Step 3

Group -.01 [-0.40-0.43]


Social identification .08 [-0.17-0.28]

Group x social identification .14 [-0.27-0.56]

Drink x social identification .42 [0.08-0.83]*

Step statistics F(2,85)=2.48, p = .09, ΔR2 = .05

*p < .05, **p < .01


Figure 1.Graphed interaction between social identification and alcohol/placebo drink condition.

-1-.5

0.5

1In

-gro

up F

avou

ritis

m

-1 1Social Identification (-/+SD)

placebo alcohol

research.edgehill.ac.uk et al.e… · web viewallocating under the influence: effects of alcohol...

Documents