PERSONALITY TRAITS AND WOMEN’S ALCOHOL
CONSUMPTION
Michelle Andersson & Therese Ohlsson
Fall semester 2019 Honor thesis
School of Law, Psychology and Social Work Advisor: Catherine Tuvblad
Abstract
Women are a particularly vulnerable group when it comes to victimization while intoxicated. Personality traits and characteristics have been shown to have an impact on alcohol consumption. Previous research regarding the Big Five Personality Traits indicates higher drinking levels if a person scores high within “Extraversion” and “Neuroticism”, or scores low within the traits of “Agreeableness” and “Conscientiousness”. Concerning women, “Openness” seems to be the only personality related to alcohol consumption, but previous research is scarce. The main focus of previous research has been on the Big Five Personality Traits, people in general and alcohol consumption. Our focus was Swedish women, age range 18-50 years. We proposed that personality traits could affect alcohol consumption and because of the lack of research we specifically wanted to examine women. The results showed that personality traits have an impact on women’s alcohol consumption. In conclusion, women tend to drink more alcohol if they score high on “Extraversion” and “Neuroticism”, and, if they scored low on “Agreeableness” and “Conscientiousness”. This could mean that personality traits serves as risk factors for alcohol consumption. “Openness” and alcohol consumption showed no association. Future research should investigate how other factors combined with personality traits might affect alcohol consumption.
Key words: Personality traits, women, alcohol consumption, The Big
Personlighetstyper och kvinnors alkoholkonsumtion Sammanfattning
Kvinnor är en extra utsatt grupp när det kommer till viktimisering i samband med alkoholkonsumtion. Fortsättningsvis, vissa personlighetsdrag och karaktärsdrag har visats påverka alkoholkonsumtion. Tidigare forskning gällande Big-Five-teorin och dess personlighetsdrag indikerar vanligtvis högre nivåer av alkoholkonsumtion om en person skattar högre inom dragen “Utåtriktning” och “Känslomässig instabilitet”, eller skattar lägre inom dragen “Vänlighet” och “Målmedvetenhet”. Gällande kvinnor, draget “Öppenhet” verkar vara det enda som är kopplat till alkoholkonsumtion, men tidigare forskning är begränsad. Huvudfokuset inom tidigare forskning har varit på Big-Five-teorin, människor överlag och alkoholkonsumtion, vi ville därmed specifikt undersöka kvinnliga vanor. Vårt fokus har legat på svenska kvinnor i åldrarna 18- 50 år. Vi menade att personlighetsdrag kunde ha en påverkan på alkoholkonsumtion och ville undersöka specifikt kvinnor vad gäller detta då befintlig forskning ansågs bristfällig. Resultatet visade att olika personlighetsdrag påverkar alkoholkonsumtionen hos kvinnor. Sammanfattningsvis, kvinnor tenderar att dricka mer alkohol om de skattar högt inom “Utåtriktning” och “Känslomässig instabilitet”, och om de skattar lågt inom “Vänlighet” och “Målmedvetenhet”, vilket skulle kunna innebära att olika personlighetsdrag fungerar som riskfaktorer för alkoholkonsumtion. Dock visade “Öppenhet” inte på något samband med alkoholkonsumtion. Framtida forskning bör undersöka hur eventuellt andra faktorer kombinerat med personlighetsdrag kan ha en effekt på alkoholkonsumtion.
Nyckelord: Personlighetstyper, kvinnor, alkoholkonsumtion,
Index
Personality Traits and Women’s Alcohol Consumption………..……..………....1
Theoretical framework.………...…….1
Big Five Personality Traits………...……..…..1
Previous research………...………..…….2
Limitations with previous research………...……....…...4
The present study………...………..…….4
Method………...……….……….5 Participants………..………...………..……5 Material………...……….………..…...6 Measures………...……….…...6 Personality traits………...………..…….6 Alcohol consumption………...…………..……...8 Procedure………...……….……….………….9 Ethics………...……….………..11 Statistical analyses………...………..………...12 Results………...……….………13 Discussion………...……….………..16 References………...……….……….…23
Personality Traits and Women's Alcohol Consumption
Women are one of the most victimized groups in our society. Research reveals that both alcohol consumption (Testa & Livingston, 2009; McKinney, Caetano, Rodriguez & Ngozi, 2010) and personality traits (Van de Weijer & Leukfeldt, 2017) can increase the risk of victimization. Also, certain personality traits can be seen as risk factors for alcohol
related problems (Eshetu & Book, 2016; Kotov, Gamez, Schmidt & Watson, 2010; Malouff, Thorsteinsson, Rooke & Schutte, 2007; Martin, Groth, Longo, Rocha & Martens, 2015). Furthermore, recent reports from the Public Health Agency of Sweden shows that women’s alcohol consumption is increasing (Folkhälsomyndigheten, 2019). Therefore, it is
imperative to gather further knowledge regarding the connection between personality traits and alcohol consumption, particularly in regards to women. Women as a group have not been researched in the same extent as men (Hughes, 2008), and this is one area we aimed to address with this study.
Hence, the purpose of this study was to investigate how different personality traits might have an effect on women’s alcohol consumption and therefore examine the
personality traits individually and as a group to see a potential correlation and/or explained variance within alcohol consumption.
Theoretical framework
Big Five Personality Traits. Researchers within psychology have after decades of research found a general understanding and common ground in the integrative personality dimension framework called Big Five Personality Traits (John & Srivastava, 1999). Big Five Personality Traits is a theory discussing personality traits and how a person scores on the spectrum of a specific personality trait (John & Srivastava, 1999). Either you score low or high on the spectrum, but all people will be placed somewhere in the spectrum. All the different scores help understand differences within and between individuals and their characteristics (John & Srivastava, 1999). This theory will be used as part of the foundation for this study. The personality traits will portray the study’s independent variables. The five personality traits included in the theory are “Extraversion”, “Openness”,
“Conscientiousness”, “Agreeableness” and “Neuroticism”. In Sweden, research shows that women tend to score higher in “Conscientiousness”, “Agreeableness” and “Openness”, compared to what they score in “Extraversion” and “Neuroticism” (Zakrisson, 2010). A person that scores high on Extraversion is usually described as talkative, social and prefers interaction with other people (Costa & McCrae, 1990). If a person were to score low on the
spectrum of “Extraversion” they are described as more independent people who do not prefer social interactions. These people are often more quiet than their antipole and are called Introverts (Costa & McCrae, 1990). The personality trait “Openness” refers to a person's willingness to open up towards other people, for example, personal values and new thoughts and ideas (Costa & McCrae, 1990). Furthermore, a person who score high within the spectrum of “Openness” values new experiences and is not to close minded to change his or hers personal beliefs and values. If one person were to score low within the spectrum of “Openness” it would be described as a person who don't appreciate change, has no intentions of change, and would be called conservative (Costa & McCrae, 1990). When it comes to the personality trait “Conscientiousness” it refers to a person with high work ethics, a person who is organized and reliable (Costa & McCrae, 1990). On the other hand, the ones who score low on the spectrum of “Conscientiousness” could be perceived as the opposite, with low work ethics and rather lazy. People who receive a high score on the spectrum for the personality trait “Agreeableness” is perceived as helpful and friendly, also forgiving and easy to deal with (Costa & McCrae, 1990). Additionally, the ones who receive a low score within the spectrum of “Agreeableness” would be described as manipulative, even suspicious and rude. Finally, there is the spectrum of “Neuroticism” which is about emotional instability (John & Srivastava, 1999). Neurotic people are often both vulnerable, and emotional, often combined with self-critical tendencies. On the other side of the spectrum there is emotional stability (John & Srivastava, 1999).
In this study we will include the Big Five Personality Traits theory as a model of explanation since according to John and Srivastava (1999), this theory is the most widely accepted personality theory among researchers within the field. We will include this as a model of explanation to try to understand the variation in female characteristics and behavior, and what this potentially could increase the risk of, i.e. consumption of alcohol. Previous research
Certain personality traits are more eminent in women than men. Women are for example found to be more neurotic (Zakrisson, 2010), this might be because of their tendency to be socialized to pay emotional attention, more than men(Donges, Jachmann Kersting, Egloff & Suslow, 2015). Also, women are found to be more conscientious and more agreeable than men (Rubinstein, 2005; Zakrisson, 2010). The higher scores in “Conscientiousness” could potentially be explained because of women redirecting their traditional sense of adherence to their family values instead of work engagements (George, Helson & John, 2011). That women are more conscientious when it comes to taking care of
their families(George, Helson & John, 2011). The higher scores in “Agreeableness” could potentially be explained because many of the different characteristics that are measured in the Big Five Inventory, that is connected to the personality trait of “Agreeableness”, are considered to be more feminine (Magnusson, 2002, referred to in Zakrisson, 2010).
Alcohol related problems has been proven to be connected to personality traits in the past. Certain personality traits has been identified as risk factors for alcohol related
problems (Eshetu & Book, 2016; Kotov et al., 2010; Malouff et al., 2007; Martin et al., 2015). Research that has been made regarding both the Big Five Personality Traits and addiction show that people in general who receive a high score in the spectrum for
“Neuroticism”, and people who receive a low score in the spectrum for “Conscientiousness” are the ones most linked to addiction, such as alcohol problems (Kotov et al., 2010;
Malouff, Thorsteinsson & Schutte 2005; Stewart, Zvolensky & Eifert 2001). Further research show that people who score low within the spectrum of “Agreeableness”, low in “Conscientiousness” and high on the spectrum of “Extraversion” is found to be more related to alcohol problems (Kotov et al., 2010; Malouff et al., 2007). Although, when examining women alone, the only personality trait shown to be related to high alcohol consumption was “Openness” (Martin et al., 2015). In summary, personality traits is shown to be a risk factor for alcohol related problems.
Women’s drinking habits can increase the risk of victimization. Research show that intoxicated women are more likely to be victimized (McKinney et al., 2010; Testa & Livingston, 2009). For example, 50-70 % of sexual assaults involve alcohol consumption, and happens to approximately 1 in 20 women each year (Abbey, Zawacki, Clinton & McAuslan, 2004; Center for Disease Control and Prevention, 2016). According to the Swedish National Council for Crime Prevention, intoxication has a strong connection to different types of violent crimes and other types of crimes, both in regards to victims but also offenders (Brottsförebyggande rådet [BRÅ], 2015). Therefore, the Swedish National Council for Crime Prevention highlights the importance of investigating alcohol as a factor (BRÅ, 2015). In order to gather further knowledge of the possible consequences alcohol consumption might have in the lives of women, understanding their drinking habits is imperative.
In summary, the studies of women is an urgent matter, partly because of the increase in women's drinking habits, partly because of the connection with personality traits and how different traits might have an effect on alcohol consumption. But also, because of the severe lack of studies involving women (Hughes, 2008).
Limitations with previous research. Based on the literature review we found limitations with previous research. Mainly, the lack of research in regards to the habits of women when it comes to alcohol consumption and how personality traits might have an effect on this. Previous research mentions people in general quite a lot but focus on women as a group is rare. The same goes for previous research of alcohol consumption, there is a lot of research regarding addiction or dangerous habits in general, but not really anything specific on alcohol consumption in connection to women and personality traits.
We now know, based on previous studies, that some personality traits are more prominent in women than men. Also, that some personality traits in fact are connected to alcohol consumption. We can also deduct that the connection between personality traits in women and their alcohol consumption, needs to be further investigated. This because of the serious absence of research on females, for example, within the field of criminology
(Hughes, 2008). Also, because research within the field of criminology reveals that alcohol as a factor plays a large role in many different crime related questions (BRÅ, 2015).
Further research on alcohol consumption, personality traits and the differences within women, is imperative for the general understanding in human variation. The present study
The research method for our study was a quantitative approach with a cross-sectional design. It is an appropriate design for this study’s purpose, since we only want to investigate the variables at one occasion. The participants will be gathered by a convenience sample from various Facebook groups for women. To obtain data, the participants will answer a survey.
In this study, our purpose was to investigate how different personality traits might have an effect on women’s alcohol consumption and therefore examine the personality traits individually and as a group to see a potential correlation and/or explained variance within alcohol consumption.
Our research questions was as follows:
1. Does women’s personality traits affect their alcohol consumption? If so, which personality trait, and how?
a) Is there an association between personality traits and alcohol consumption? If so, how strong and in what direction?
Because of previous research regarding people in general, alcohol consumption and personality traits (Kotov et al., 2010; Malouff et al., 2005; Malouff et al., 2007; Martin et al., 2015; Stewart et al., 2001), we hypothesize that women who score high on the
personality trait “Neuroticism” will have higher alcohol consumption. We hypothesize that women who score low on the personality trait “Conscientiousness” will have higher alcohol consumption. We hypothesize that women who score low on the personality trait
“Agreeableness” will have higher alcohol consumption. We hypothesize that women who score high on the personality trait “Extraversion” will have higher alcohol consumption. Finally, we hypothesize that women who score high on the personality trait “Openness” will have higher alcohol consumption.
Method
This study was conducted with a quantitative method through a cross-sectional design. It was considered the most suitable method, since we only wanted to investigate our variables at one occasion. The population in this study was women, 18+ years of age, who were residents in Sweden and were members in various Swedish Facebook groups for
women. Since Facebook is the largest social media platform (Clement, 2019), it gives a solid arena to investigate female habits on. The selection method was convenience sampling, since the participants were conveniently selected and reached out to because of their availability to take part in the study. The participants had the opportunity to respond to the questions of the data collection method, the survey, which was created in Google Forms. Participants
The population size for this study was 321 679 members of different Swedish Facebook groups for women. The criteria to be a part of this study was to be a woman defined by legal gender, a minimum of 18 years of age, to mainly reside in Sweden, and to be a member of at least one Swedish Facebook group for women. All participants were asked to give their consent to participate in the study. The participants in our sample
population consisted of 677 persons. After 26 participants were removed due to the fact that they did not consent to the participation research information of the study, were not a female by legal gender or where not mainly residing in Sweden, 651 women remained as the
study’s sample (n = 651). Their age varied from 18 to 50 with an average age of 25.82 (M = 25.82) years (SD = 6.41). The sampling method of the study was a convenience sampling method due to the fact that the survey was posted in the different Swedish Facebook groups
for women, where the women were conveniently available. Each member of these groups had an equal chance to answer the survey if they fulfilled the criteria for the study. No reward was given for the participation of the study.
Material
The survey used for this study was combined by two, well used and tested, measuring instruments. All questions in the survey was mandatory and the survey was divided into 5 different sections. All questions in every section had to be answered for the participants to move on to the next section. In the first section the participants had to read through a missive letter with information about the study and the participants rights. The second section was about the participants giving their consent or not, after which some general questions were asked such as gender, age, and if they mainly resided in Sweden or not. When that part was completed the participant were directed into the third section which was a personality test translated into Swedish by Claesson, Persson & Akrami (2001) referred to in Zakrisson (2010). The original version of the test is called The Big Five
Inventory and is created by John & Srivastava (1999). Following the Big Five Inventory was the fourth and last section where the participants responded to a test to measure their alcohol consumption. This test is called Alcohol Use Disorder Identification Test, also known as AUDIT, translated into Swedish by Bergman, Källmén, Rydberg & Sandahl (1994). We chose the translated version instead of the original version by Babor, Higgins- Biddle, Aaunders & Monteiro (2001) since we wanted the participants’ mother tongue language to minimize the risk of misunderstandings.
Measures
We have measured 2 different variables, personality traits and alcohol consumption. Cronbach’s Alpha is a value that measures items reliability on a scale, and if the items measure the same thing in an adequate matter the value will be .70 or higher, which
indicates good measurement (Bryman, 2011). The measurements are valid and reliable and their Cronbach’s Alpha value will be presented alongside with descriptions of the items and questions to exemplify their content.
Personality traits. The Big Five Inventory was used to measure personality traits among the women participating in this study. The Big Five Inventory is a measuring
instrument containing 44 statements regarding personal characteristics which is connected to the 5 personality traits, “Extraversion”, “Openness”, “Conscientiousness”, “Agreeableness” and “Neuroticism”, Costa and McCrae (1990) acknowledges in The Big Five Personality Traits theory. There were 8 statements connected to the personality trait “Extraversion”. An
example of a statement connected to “Extraversion” is “I see myself as someone who is talkative”. On the opposite side of “Extraversion”, on the same spectrum, is “Introversion” (John & Srivastava, 1999). There were 10 statements connected to the personality trait “Openness”. An example of a statement connected to “Openness” is “I see myself as someone who has an active imagination”. On the opposite side of “Openness”, on the same spectrum, is “Closeness to experience” (John & Srivastava, 1999). There were 9 statements connected to the personality trait “Conscientiousness”. An example of a statement connected to “Conscientiousness” is “I see myself as someone who is a reliable worker”. On the
opposite side of “Conscientiousness”, on the same spectrum, is “Lack of Direction” (John & Srivastava, 1999). There were 9 statements connected to the personality trait
“Agreeableness”. An example of a statement connected to “Agreeableness” is “I see myself as someone who has a forgiving nature”. On the opposite side of “Agreeableness”, on the same spectrum, is “Antagonism” (John & Srivastava, 1999). Finally, there were 8 statements connected to the personality trait “Neuroticism”. An example of a statement connected to “Neuroticism” is “I see myself as someone who worries a lot”. On the opposite side of “Neuroticism”, on the same spectrum, is “Emotional Stability” (John & Srivastava, 1999). The participant got to read the statements and thereafter answer on a Likert scale ranging from 1 to 5. The response options were “Disagree strongly”, “Disagree a little”, “Neither disagree nor agree”, “Agree little”, and “Agree strongly”. Even if the Likert scale usually is considered ordinal, several researchers have concluded that with 5 points or more, this scale can be treated as continuous without any misleading results after a performed analysis (Johnson & Creech, 1983; Norman, 2010; Sullivan & Artino, 2013; Zumbo & Zimmerman, 1993). These variables will be seen as ordinal approximation of a continuous variable.
The Big Five Inventory had a Cronbach’s Alpha value of .71. The five different personality traits also have their individual Cronbach’s Alpha values. The Cronbach’s Alpha value for “Extraversion” is .85. The Cronbach’s Alpha value for “Openness” is .76. The Cronbach’s Alpha value for “Conscientiousness” is .78. The Cronbach’s Alpha value for “Agreeableness” is .73. The Cronbach’s Alpha value for “Neuroticism” is .82.
To interpret the Big Five Inventory we used the original scoring system. If a
participant where to answer “disagree strongly” on one of the statements that would equal 1 point, if the participant instead answered “disagree a little” that would generate 2 points, “neither agree nor disagree” equals 3 point, “agree a little” generated 4 points and the answer “agree strongly” generated 5 points (Zakrisson, 2010). The maximum score one could receive per statement was 5 points. Thereafter, a summation of every personality trait
for each participant was made, every participant received one total score per personality trait. From that total score on every personality trait, one could see how a person scores within every spectrum connected to personality traits and se which personality traits that is most eminent within that participant.
Alcohol consumption. AUDIT was used to measure the participants’ alcohol consumption. It is a measuring instrument containing 10 questions regarding alcohol
consumption with questions such as “How often do you drink alcohol?”, “How often during the last year have you needed a ‘drink’ in the morning after a lot of drinking the day
before?”, and “How often during the last year have you had feelings of guilt or remorse because of your drinking?”. Question number 1 has response alternatives on a 5 point Likert scale which are “Never”, “1 time a month or more rarely”, “2-4 times a month”, “2-3 times a week” and “4 times a week or more”. Question number 2 has response alternatives on a 6 point Likert scale which are “0 st”, “1-2 st”, “3-4 st”, “5-6 st”, “7-9 st” and “10 st or more” (where “st” represents “the amount of”). The response alternatives for question 3 to 8 are also on a 5 point Likert scale. The alternatives are “Never”, “More rarely than once a month”, “Every month”, “Every week” and “Daily or almost every day”. The final 3 questions have 3 response alternatives which are “No”, “Yes, but not during the past year” and “Yes, during the past year”. AUDIT had a Cronbach's Alpha value of .84.
To interpret AUDIT we used the original scoring system. If a participant responds in a denying matter on the questions, they received a low score which indicates low risk of alcohol problems (Babor et al., 2001). The response alternatives for question 1 range from “Never” to “4 times a week or more” and the score range from 0 to 4 points. The response alternatives for question 2 range from “0 st” to 10 st or more” and the scores range from 0 to 4 points. Question 3 to 8’s response alternatives range from “Never” to “Daily or almost every day” and the scores range from 0 to 4 points. Finally, question 9 and 10’s response alternatives range from 0 to 4 points. The maximum score a participant could receive on AUDIT is 40 points.
According to World Health Organisation, who are the founders of AUDIT, the general breaking point for hazardous use, heavy consumption and addiction lies at 8 points or more (Babor et al., 2001). Since the alcohol culture varies between countries, this was taken into consideration when deciding to use the AUDIT- scoring system used in Sweden to interpret Swedish alcohol habits. The breaking point for women are a bit lower, than the general breaking point set by Babor et al. (2001), and lies at 6 points (Riddargatan 1, Mottagningen för alkohol och hälsa, n.d.). Furthermore, the Swedish interpretation of the
scoring system for women according to this country’s alcohol culture is as follows; 0 to 5 points as a complete test score, indicates non-existing or low levels of alcohol related problems (Riddargatan 1, Mottagningen för alkohol och hälsa, n.d.). 6 to 13 points as a complete test score, indicates a medium level of alcohol related problems (Riddargatan 1, Mottagningen för alkohol och hälsa, n.d.). Meanwhile, 14 to 17 points as a complete test score indicates a high level of alcohol related problems and a potential addiction diagnosis might be necessary (Riddargatan 1, Mottagningen för alkohol och hälsa, n.d.). If a
participant would receive more than 18 points, as a total test score, this indicates the
existence of a very high level of alcohol related problems and there should be a presence of a strong addiction, and most likely an addiction diagnosis (Riddargatan 1, Mottagningen för alkohol och hälsa, n.d.).
Procedure
To obtain an extensive platform of previous research many different databases on internet was searched through, for example Primo and SAGE Journals. The specified criteria that were used during the searches was only peer reviewed scientific articles and the search words used was for example: Women, Facebook, Alcohol, Drinking, Personality traits, Victimization and Big five.
After the compilation of the previous research we chose to conduct a survey study to answer the research questions. In the making of the survey, a survey program called Google Forms was used. All questions were made obligatory and if any question were to be leaved unanswered, the participant received a message that they had to attend to the unanswered question to be able to continue. The survey was also programmed in a manner where some questions, answered in a certain way, would lead to the end of the survey and the participant was not allowed to continue with the survey. These particular questions regarded the study's’ sample criteria. First of all the study only intended to investigate females with the legal gender female, therefore the gender question was programmed to send all participants with different answers than “female” to the end of the survey. Since the study also intended to investigate females no younger than 18 years of age, and mainly Swedish residents, the study was also programmed to send all other participants, those under the age of 18 and those not mainly residing in Sweden, to the end of the survey. The rest of the questions or statements throughout the survey were open to all answering alternatives and the
participants could continue answering the survey regardless of their answer.
When the participants were to answer the survey, they first got to read through a missive letter that described what the study aimed to investigate, how different personality
traits might have an effect on alcohol consumption. Also, information of the sample criteria or the participants the study was looking for was described. That is, women in different Facebook groups, women with the legal gender “female”, who at least were 18 years of age, mainly resided in Sweden and who consented to participation in the study.
Thereafter information followed regarding the participants rights revolving their participation in the study, that their participation was completely anonymous, that their participation was voluntary, that they could end their participation at any time, in what purpose the collected data would be used, who would have access to the data, how the information they revealed about themselves would be stored and also later destroyed. After the missive letter there was a consent form which the participants had to fill out. The consent form contained of questions whether or not the participant gave their consent to part take in the study and if they gave consent to the way their information would be handled, in line with current law. If the participants chose to answer “no” in the consent form, they were directly directed to the end of the survey and they were not allowed to participate.
The survey then started with part two which contained a small number of
background questions like age, their legal gender and whether they are mainly residing in Sweden or not. Thereafter, part three started, which contained the test Big Five Inventory, that is, statements about personal qualities for the participants to answer. After that part four started, which contained questions about the participants own experiences regarding their alcohol consumption. In this part the survey used AUDIT, a reliable tool to measure alcohol habits.
When the survey was finished it was published in different Facebook groups for women. The total member count was 321 679 persons (Facebook, 2019). At the same time the survey was being published, a descriptive post was also published, in which the study’s aim was described with a specification that the study only was interested in female
participants with a minimum age of 18 years.
The survey were available in the groups for precisely 120 hours (5 days). The members of the groups were also reminded twice, in the same manner as the original published post. The survey was taken down and made unavailable when the sample size reached 677 participants. This because it was considered sufficient since 384 answers on a survey, roughly, is a sufficient number for a population that is 100 000+ if you want a confidence interval at 95 % (SurveyMonkey, n.d.).
Furthermore, the answers to the survey were downloaded to an Excel file to be transformed from letters into numbers. When this was completed, the answers were
transferred to the data analysis program Statistical Packages for the Social Sciences version 26 (henceforth named SPSS). Some of the answers to the personality trait statements needed to be reversed in the scales, to represent the same direction as the majority of the questions. Afterwards, the analysis of the collected data were made.
Ethics. The missive letter for this study was written according to the research ethical principles. An awareness regarding the possibility that the section concerning alcohol consumption could be a sensitive topic existed. To handle this the missive letter first described the aim of the study and highlighted that there would be questions regarding alcohol consumption that could be interpreted as sensitive depending on what the
participants relationship to alcohol was. If participants were to have unwanted or unpleasant feelings it was described that they could cancel their participation of the study whenever they wanted, without any explanation. This also applied to those who did not experience anything unpleasant but simply just changed their minds about participating in the study. This in accordance with existing information requirements (Vetenskapsrådet, 2002). If one were to have strong unpleasant and/or unwanted feelings, a website called “Alkohollinjen” was linked, this is a website who supports people that want to change their alcohol
consumption, if one have questions regarding their own or someone else's alcohol
consumption. A phone number to their support line was also included (Alkohollinjen, n.d.). Another way to handle the sensitivity of the questions regarding alcohol consumption was that a well-used and tested measuring instrument that other authorities in Sweden use was the primary measuring instrument for this subject. Furthermore, regarding the existing information requirements was that, the whole aim of the study was described, what special characteristics specifically was wanted for this study and how long time the survey was expected to take.
The survey was made in Google Forms and because of this the participants
anonymity could be preserved due to the fact that no e-mail addresses was saved, no login was demanded and no names was required to participate in this study. This in order to limit the amount of personal information that would be saved, and therefore no answers could be connected to one specific participant, this in accordance with the requirements of
confidentiality (Vetenskapsrådet, 2002) and the General Data Protection Regulation (GDPR) (Datainspektionen, n.d.). The answers was saved in a password protected computer which no outsiders had access to, only the authors for this study. When the Excel file had been transmitted to SPSS version 26, the Excel file and the responses from Google Forms were deleted. When this study is completed, the SPSS file will also be deleted.
Before the participants started answering the survey, they were to read a consent form, which came below the missive letter. Therefore, the participants had already been informed about the details regarding the study. Thereafter, they first had to consent to participate in the study and thereafter consent to that personal information would be treated the way it were described in the missive letter. This is accordance with existing requirements of consent (Vetenskapsrådet, 2002) and GDPR (Datainspektionen, n.d.). If one were to not leave their consent they would no longer have the option to participate. Finally, in
accordance with the requirements regarding usage of participants data, it was described that collected information would only be used for the intended aim of the study and would not be further distributed (Vetenskapsrådet, 2002).
Questions that was asked in this study were relevant to the purpose of the study, no other irrelevant questions were asked. We remained as neutral and objective as possible throughout the course of the study.
Statistical analyses
The analyzing strategies used to analyze the data was found in SPSS version 26. One strategy that was used to compile descriptive statistics concerning the data, such as mean age of participants, standard deviation, range of age and total amount of participants, was the feature of Descriptives, which gave us these statistics.
Some of the statements regarding personality traits had to be reversed in order to receive the same direction in the answers as the other statements. Thereafter the statements regarding each personality trait was combined by the feature Compute variable and by that 5 different variables was made, one for each personality trait. Furthermore, all of the questions regarding alcohol consumption was also combined by the feature Compute variable and one new alcohol consumption variable was made.
Another strategy used to see whether the data was normally distributed or skewed was a test named Explore, which showed that the data was normally distributed through Normal Q-Q plots for each personality trait. Since we had reasonably straight lines in our large data sample, we can assume a normally distributed data set (Pallant, 2016). Also, through Histograms we could observe a normal distribution for each personality trait. Through Explore, we could observe a trimmed mean in the descriptive table and a regular mean for alcohol consumption, indicating a normal distribution according to Pallant (2016).
To investigate whether there was a linear relationship between alcohol consumption and the different personality traits a Pearson Product-moment Correlations analysis was performed. This was performed in purpose of examine the strength and the direction of the
potential relationship between the variables (Pallant, 2016).
In order to examine the study’s dependent variable “Alcohol consumption” in relation to the study’s different independent variables “Extraversion”, “Openness”, “Agreeableness”, “Conscientiousness” and “Neuroticism”, a Hierarchical Multiple Regression model was performed. This because we wanted to investigate how much of a variance, in alcohol consumption, the different personality traits would explain as a group or each of them individually. Therefore, according to Pallant (2016), the choice of performing a Hierarchical Multiple Regression model was appropriate. The order in which the variables was entered into the model was because of their individual mean values. Based on previous research we could see that each trait had an impact on alcohol consumption (Eshetu & Book, 2016; Kotov et al., 2010; Malouff et al., 2005; Malouff et al., 2007; Martin et al., 2015; Stewart et al., 2001), and therefore we entered them into the model based on how eminent they were in our sample. So, the more eminent one trait was it got to represent the
importance of that personality trait in predicting the outcome. Which would be an appropriate way of entering the variables according to Field (2013).
Results
The purpose of the study was to investigate how different personality traits might have an effect on women’s alcohol consumption and therefore examine the personality traits individually and as a group to see a potential correlation and/or explained variance within alcohol consumption. The sample consisted of 677 (N = 677) people who answered the survey. We had a total loss of 26 participants, 24 of these were internal losses and 2 of these were external losses. After the loss of participants the study’s sample consisted of 651 women (n = 651, Mage = 25.82, SD = 6.41). When examining the means of our different personality trait variables we could see that their values differed, ranging from the highest value which was found in “Agreeableness” (M = 34.47, SD = 4.79), “Openness” (M = 33.70,
SD = 6.19) and “Conscientiousness” (M = 32.53, SD = 5.38), to the lowest values that was
found in “Extraversion” (M = 26.51, SD = 5.90) and “Neuroticism” (M = 25.22, SD = 5.75). From this we can extract that women in our sample tend to score higher within the
personality traits “Agreeableness”, “Openness” and “Conscientiousness”, than the rest of the personality traits.
After conducting tests to control the data’s normality and distribution, linearity and multicollinearity, we continued performing a Pearson Product-moment Correlations analysis
to investigate our research question “Is there an association between personality traits and alcohol consumption? If so, how strong and in what direction?” . The Pearson Correlation showed that some of the personality traits had a statistically significant linear relationship with alcohol consumption. See Table 1.
Table 1. Pearson Product-moment Correlations Between Alcohol Consumption and the Big Five Personality Traits.
Variable 1. 2. 3. 4. 5. 6. 1. Alcohol - 2. Agreeableness -.188*** - 3. Openness .062 .117** - 4. Conscientiousness -.295*** .337*** -.46 - 5. Extraversion .081* .238*** .183*** .250*** - 6. Neuroticism .194*** -.321*** .049 -.425*** -.379*** -
Note. *p < .05. **p < .01. ***p < .001. Alcohol = Alcohol consumption.
Statistical significances were found between alcohol consumption and 4 out of 5 personality traits. A statistical significant negative correlation between “Agreeableness” and alcohol consumption was found, which could mean that the direction of the variables
indicates that if women have less characteristics of “Agreeableness”, they could potentially consume more alcohol. A statistical significant negative correlation was found between “Conscientiousness” and alcohol consumption, which could mean that the direction of the variables indicated that if women have less characteristics of “Conscientiousness”, they could potentially consume more alcohol. A statistical significant positive correlation was found between “Extraversion” and alcohol consumption, which could mean that the direction of the variables indicates that if women have more characteristics of
“Extraversion”, they could potentially consume more alcohol. A statistical significant positive correlation was found between “Neuroticism” and alcohol consumption, which could mean that the direction of the variables indicates that if women have more
characteristics of “Neuroticism”, they could potentially consume more alcohol. No
statistical significant correlation was found between “Openness” and alcohol consumption. In the investigation of how much variance each personality trait explained within alcohol consumption we performed a five stage Hierarchical Multiple Regression model. At Step 1, the personality trait “Agreeableness” was entered. At Step 2, the personality trait “Openness” was entered. At Step 3, the personality trait “Conscientiousness” was entered, thereafter at Step 4, the personality trait “Extraversion” was entered. In the final stage, Step
5, the personality trait “Neuroticism” was entered. The analysis showed significant results regarding the explained variance in some of the personality traits. See Table 2.
Table 2. Hierarchical Regression Model Predicting Alcohol Consumption
Step 1 Step 2 Step 3 Step 4 Step 5
Agreeableness -.188*** -.198*** -.109** -.136** -.114** Openness .085* .063 .032 .017 Conscientiousness -.255*** -.292*** -.254*** Extraversion .181*** .218*** Neuroticism .132** t -4.87*** 2.20* -6.41*** 4.62*** 3.04** R .188*** .206* .316*** .358*** .375** R2 .035*** .042* .100*** .128*** .141** R2Change - .007* .058*** .028*** .013** Note. *p < .05. **p < .01. ***p < .001.
By first adding the personality trait “Agreeableness” to the model in Step 1, it explained 3.5 % of the variance within alcohol consumption, F (1, 649) = 23.73, p < .001. At Step 2, adding the personality trait “Openness” to the model, it explained an additional 0.7 % of the variance within alcohol consumption, after controlling for “Agreeableness”. Combined with the previous model, “Agreeableness” and “Openness” explained a total of 3.9 %, F (2, 648) = 14.36, p < .001. At Step 3, adding the personality trait
“Conscientiousness” to the model, it explained additional 5.7 % of the variance within alcohol consumption, after controlling for “Agreeableness” and “Openness”. Combined with the previous models, “Agreeableness”, “Openness” and “Conscientiousness” explained a total of 10 %, F (3, 647) = 23.84, p < .001. At Step 4, adding the personality trait
“Extraversion” to the model, it explained an additional 2.9 % of the variance within alcohol consumption, after controlling for “Agreeableness”, “Openness” och “Conscientiousness”. Combined with the previous models, “Agreeableness”, “Openness”, “Conscientiousness” and “Extraversion” explained a total of 12.8 %, F (4, 646) = 23.77, p < .001. Lastly, at Step 5, adding the personality trait “Neuroticism” to the model, it explained 1.2 % of the variance in alcohol consumption, after controlling for “Agreeableness”, “Openness”,
“Conscientiousness” and “Extraversion”. After the final personality trait was entered the total variance within alcohol consumption explained by the model as a whole was 14.1 %, F (5, 645) = 21.10, p <.001.
In the final model, 4 out of 5 personality traits in the analysis showed a statistical significant impact on the variance of alcohol consumption. “Conscientiousness” reported the
highest beta value of -.254 (p < .001), which could mean that the less characteristics of “Conscientiousness” women hold, the more likely they are to drink more. “Extraversion” in turn, reported a beta value of .218 (p < .001), which could mean that the more
characteristics of “Extraversion” women hold, the more likely they are to drink more. “Neuroticism” in turn, reported a beta value of .132 (p < .001.), which could mean that the more characteristics of “Neuroticism” women hold, the more likely they are to drink more. “Agreeableness” in turn, reported a beta value of -.114 (p < .001), which could mean that the less characteristics of “Agreeableness” women hold, the more likely they are to drink more. Lastly, the personality trait “Openness” showed no statistical significant impact in the final model.
Discussion
The purpose of the study was to investigate how different personality traits might have an effect on women’s alcohol consumption and therefore examine the personality traits individually and as a group to see a potential correlation and/or explained variance within alcohol consumption. After controlling for all the personality traits of the Big Five
Personality Traits theory in the different steps of the Hierarchical Regression Model, we found connections between four out of five personality traits in women and alcohol consumption. These connections indicated an independent association of each trait on alcohol consumption. Implying that the women in our sample who tended to be more extraverted than introverted, and the women in our sample who tended to be more neurotic than emotionally stable, consumed more alcohol in general. The connections also showed that the women in our sample who tended to be less agreeable and less conscientious also consumed more alcohol in general. Since women’s personality traits seems to be related to different levels of alcohol consumption (Eshetu & Book, 2016; Kotov et al., 2010; Malouff et al., 2005; Malouff et al., 2007; Martin et al., 2015; Stewart et al., 2001), our findings suggests that women who are more extraverted than introverted, more neurotic than emotionally stable, less agreeable and less conscientious, have an increased risk of consuming more alcohol compared to the women who scores in the opposite direction.
Our contribution to the literature is that we focused solely on personality traits and alcohol consumption in women. Previous research have primarily focused on people in general. Furthermore, previous studies have emphasized the importance of future research examining different types of women on social media (Pruchniewska, 2019). Therefore, our
sample consisted of specifically women, who were found on social media. This could be seen as an additional contribution to the literature, the sample of women from social media, since previous studies have wished for it.
The score our sample received on “Neuroticism” seems to be equivalent to what 71-80 % women scores in general in Sweden (Zakrisson, 2010), and the score our sample got on “Extraversion” and “Openness” seems to be equivalent to what 41-50 % women scores in general in Sweden (Zakrisson, 2010). These numbers may suggest that the women in our sample could possibly resemble the women in society regarding these three personality traits. Furthermore, the score our sample received within “Agreeableness” seems to be equivalent to what 31-40 % women scores in general in Sweden (Zakrisson, 2010). Finally, the score our sample got on “Conscientiousness” seems to be equivalent to what 21-30 % women scores in general in Sweden (Zakrisson, 2010). In general, “Agreeableness” and “Conscientiousness” in our study seems to differ compared to what Zakrisson (2010) found regarding women in Sweden. In our study, fewer women seem to possess the characteristics of i.e., long-term planning skills and friendliness. Which can be found within the traits of “Conscientiousness” (Bogg & Roberts, 2004, referred to in Hakulinen et al., 2015) and “Agreeableness” (Costa & McCrae, 1990). This could potentially be explained by
differences in characteristics between the women in our sample and women in Zakrisson’s (2010) sample. For instance, Zakrisson (2010) divided her participants into different subcategories of i.e. students. Two of these groups were “economics students” and “psychology students”. In our study, we did not control for the women’s occupation or choice in university majors. Perhaps, women who are drawn to certain types of subjects, interests in life or occupational choices, are more or less likely to possess a personality trait. Maybe we received different results than Zakrisson (2010) because of our potential broad range of different fields of interests among the women in our sample. Since Zakrisson (2010) used a snowball sampling method, the participants had to know or know of each other in some way (Bhat, n.d.), which could increase the chance of sharing interests (Misswebweaver, 2014). Furthermore, because of Zakrisson’s (2010) sampling method it could have decreased the range of participant reach for the study, based on how willingly the participants were to pass the survey along to others. In our study we used a fairly large population in which we published the survey, by that, hundred of thousands (Facebook, 2019) potential participants had the opportunity to answer the survey.
According to Zakrisson (2010), the most prominent personality traits for women in Sweden, the ones they scored the highest on, was “Conscientiousness”, ”Agreeableness”
and “Openness”. In our study, although in a different order, we also found the most
prominent traits to be “Agreeableness”, ”Openness” and “Conscientiousness”. Furthermore, according to Zakrisson (2010) the least prominent personality traits for women in Sweden, the ones they scored the lowest on, was “Extraversion” and “Neuroticism”. In our study, we also found the least prominent traits to be “Extraversion” and “Neuroticism”.
Regarding personality traits and alcohol consumption, previous research have shown that those who score high on the spectrum of the personality trait “Neuroticism” and
“Extraversion” are more likely to suffer from alcohol related problems (Kotov et al., 2010; Malouff et al., 2005; Stewart et al., 2001), in line with that, we found similar results that supported what we hypothesized. Women in our study who tended to be more neurotic and more extraverted were connected to consuming more alcohol. Previous research indicates that people in general who score low on the spectrums of “Conscientiousness” and,
“Agreeableness” are related to alcohol problems (Kotov et al., 2010; Malouff et al., 2007). In our study of women, we could see that the women who seemed to be less agreeable and less conscientious could be more prone to drink alcohol in a higher degree than women who seemed to be more agreeable and conscientious, in line with what we hypothesized. This could potentially be explained by the certain characteristics these women have. Since a person who scores high on “Conscientiousness” often have a capacity for long-term planning and a good self-control (Bogg & Roberts, 2004, referred to in Hakulinen et al., 2015), those who seems to be less conscientious might have a harder time resisting alcohol because of their lack of self-control (Gullo, Dawe, Kambouropoulos, Staiger, & Jackson, 2010). Women who received a high score on “Neuroticism” was more prone to drink than women who scored low. This could potentially be explained by certain characteristics within neurotic women, such as they can often be self-critical, feel vulnerable and display a lack of control over their emotions (John & Srivastava, 1999). Drinking could therefore be explained as an aid in handling negative emotions, as a form of self-medication (Sayette, 2017).Furthermore, women who scored low on “Agreeableness” were more prone to drink than women who scored high within that trait. This could potentially be explained by the characteristics found on the opposite side of “Agreeableness”, that is, that they are often less sociable, and not abiding by social norms (Heinström, 2010). Because of these
characteristics, they might be more prone to deviant behavior and more problematic alcohol use (Finn, Sharkansky, Brandt & Turcotte, 2000, referred to in Ibáñez et al., 2015). Perhaps this could be a potential explanation to why people who score low on “Agreeableness” drink more than people who score high within that trait.Lastly, women who scored high on
“Extraversion” are more prone to drink than women who scored low within that trait. This could potentially be explained by the fact that women who are high on “Extraversion” often are talkative and prefer social interactions (John & Srivastava, 1999), they tend to drink more in social situations, such as parties (Fairbairn et al., 2015). Women who scored low within that trait could be described as women who are independent and do not prefer social interactions (John & Srivastava, 1999), and because of that they might not be exposed to places where alcohol is as frequent as the social contexts the extraverted women prefer. We hypothesized that women who score high on the personality trait “Openness” will have higher alcohol consumption in line with the findings of Martin et al. (2015). Although, unlike their findings we found no connection between “Openness” and alcohol
consumption. This might be explained by other factors, not included in this study, that could affect a persons alcohol use. Potentially, different health problems, self-esteem, cultural aspects or factors of heritage (Burke, 2017). Further investigations are needed to draw any firm conclusions.
We found that “Conscientiousness”, in our study, could explain the most variance in alcohol consumption, more so than any other personality trait. But, when all personality traits were combined, after the independent connections was investigated, they could explain much more of the variance in alcohol consumption than any of the personality traits alone. So many different factors might affect alcohol consumption (Burke, 2017), but nevertheless, personality traits showed to explain more than14.1 %.Meaning that, in our sample, approximately one-seventh of women's alcohol consumption could be explained by personality traits.
A limitation with our study could be that the women who participated might not have answered the survey in a truthful manner. This is called the social desirability bias (Fisher, 1993), there is always a risk of the participant answering in a way they think the researchers want to hear, or, what might be acceptable for others to read. At the same time, we did ensure their anonymity with our online survey, also, the participants had the “safety” of answering behind a computer screen (SurveyMethods, 2011). Another limitation with this study could be that we used a cross-sectional design, which does not help in determining cause and effect that we can generalize on to a greater population (Rivers, n.d.). This design limits us in a way that does not allow us to investigate women’s behavior over a period of time, perhaps the specific timing (Rivers, n.d.) of our study gave us misleading results. Another limitation with the present study could be that the different personality traits actually can affect a person’s tendency to participate in surveys in general. This is called
inclusive bias, that because of our convenience sample, it addresses volunteers to participate rather than a set selection of participants (Shuttleworth, n.d.). Perhaps that have had an effect on our data.
Distinct advantages with the present study has been the measures we’ve used. That all measurements had individual and common Cronbach’s Alpha values over .70 which would indicate good reliability (Bryman, 2011). In addition, we had a fairly large sample of participants according to SurveyMonkey (n.d.), compared to other similar studies (Donges et al., 2015; Gullo et al., 2010; Ibáñez et al., 2015 & Martin et al.,2015). Therefore, based on our sample, we might have received results with good reliability. This because of our
sample size, and how the size allowed us to gather more information about the data, increase the certainty of it and therefore reduce the uncertainty of it (Littler, n.d.). Another strength with this study is that it has been very cost-effective in its procedure and since we could use digitized instruments to reach our participants it was also very time-effective (Sincreo, n.d.), which made room for a lot more time to try to ensure that we used the proper analyses in examining the data to provide us with as accurate results as possible. In addition, using these digitized instruments kept us from falling into the in-group bias (Dwyer, 2018), which secured the detachment from our participants and no unfair special treatments. Furthermore, to our knowledge, there has not been any research conducted regarding our specific
population on the connection between specifically women’s personality traits and women’s alcohol consumption. The aspect of contributing new material to the field could be seen as another strength of this study (Sumeracki, n.d.).
Our findings imply that there indeed is a relevant field of interest in understanding women’s drinking habits and their personalities. From our study we can see that women on social media, with certain types of personality traits, tend to consume more alcohol than women with other types of personality traits. According to previous research, the risk of being victimized increases if you are a women (Von Hentig, 1948). In addition, Hans von Hentig’s typology of victimization could also be interesting to highlight. He strived to find characteristics that could explain an increasing risk of victimization, and believed that because of their characteristics, victims might provoke their own victimization, but not necessarily by their own fault or intent. He believed you could place all victims in one of thirteen categories: 1) young, 2) women, 3) old, 4) immigrants, 5) depressed, 6) mentally defective/deranged, 7) the acquisitive, 8) dull normal, 9) minorities, 10) wanton, 11) the lonesome and heartbroken, 12) tormentor and 13) the blocked, exempted and fighting (Von Hentig, 1948). Furthermore, he describes the victim category of “mentally
defective/deranged” as people who might not react or respond in an appropriate manner in certain situations because they do not recognize the threat or environment they are in. This mental impairment could also be because of alcohol use (Von Hentig, 1948). Examples Von Hentig uses to explain why women might be in a higher risk of victimization is because they can be easier targets because of potential risk taking behavior or ignorance to certain
situations and because they might be taken advantage of, for example in cases of sexual assault (Von Hentig, 1948). In relation to the findings in our study, it is relevant because two of these categories of victim, “women” and “mentally defective/deranged”, are connected to the variables examined in our study.
Since alcohol, according to previous research (BRÅ, 2015), plays such a large role in understanding both why some people commit crimes and why others become victims to crime, increasing the understanding of female drinking habits could potentially work as a good foundation for working with crime prevention (BRÅ, 2015). Specifically, we could reach out to women, and focus the information towards the women with the personality traits that potentially could be linked to a higher alcohol consumption. We could inform them of these risks and focus on prevention for alcohol use in various social media groups were women gather. Or, at least provide information of how to address the question of drinking among one's peers, or just advice the women on how to reach out for help if they or someone in their surrounding use alcohol in a hazardous manner. Using social media as a platform, we could reach hundreds of thousands of women in a matter of seconds, to work for greater awareness in handling the risk of women's increased alcohol intake connected to their personality traits.
Thus, since previous research implies that there is a connection between women and victimization, and between alcohol and victimization, to continue future research within this area could aid in a broader and more up-to-date understanding within these topics. Also, to potentially shed new light on different ways to work with crime prevention.Future research could focus on adding variables that measures how specific characteristics within the Big Five Personality Traits affect women’s alcohol consumption and also control for
victimization as a variable. This in order to see how certain personality traits could affect women’s victimization when intoxicated. Furthermore, examining other samples, such as women who does not use social media, or women who have children versus women who does not have children, to see if there are any differences could be very interesting. Also, investigating other variables such as aggression, genetics, different types of substance use, demographics and socio economic status, etcetera, in relation to different types of crimes in
connection to women and their personality traits, could be fascinating. The possibilities are many and the future has a lot of work ahead.
To expand the knowledge of women and further investigate female habits should be a priority in today’s society. Especially considering the increased rate of certain types of crimes against women (BRÅ, 2008).
In conclusion, some of the personality traits found within the Big Five Personality Traits have an effect on women’s alcohol consumption. The trait of “Conscientiousness” was found to be the most eminent trait. The connection between “Conscientiousness” and alcohol consumption was negative, which indicates that women who are for example less ambitious tend to drink in a higher degree than women on the opposite side of that trait. All of the traits, except for “Openness”, showed an impact on alcohol consumption, that one will drink more or less depending on what personality trait or characteristics one possesses.
We suggest that the platform of social media could be advantageously used to reach out to women, due to the easy accessibility and high number of members. Hundreds of thousands of women could be reached in a very time-effective and cost-effective way to raise awareness of how one’s personality traits might increase the risk of consumption of alcohol, which in turn could have very negative consequences.
References
Alkohollinjen. (n.d.). Alkohollinjen [Information site]. Retrieved 2019-12-02 from
https://www.alkohollinjen.se
Babor, T., Higgins-Biddle, J. C., Aaunders, JB., & Monteiro, MG. (2001). Audit. The
alcohol disorder Identification Test. Guidelines for Use in Primary Care (2nd ed.).
Geneva: World Health Organization.
Bergman, H., Källmén, H., Rydberg, U. & Sandahl, C. (2019, 03). AUDIT (Alcohol Use Identification Test) [Information site]. Retrived 2019-11-28 from
https://www.socialstyrelsen.se/utveckla-verksamhet/evidensbaserad-praktik/metodguiden/audit-alcohol-use-disorders-identification-test/
Bhat, A. (n.d.). Snowball Sampling: Definition [Information site]. Retrieved 2020-01-17 from https://www.questionpro.com/blog/snowball-sampling/
Brottsförebyggande rådet. (2015). Alkohol- och drogpåverkan vid misshandel, hot,
personrån och sexualbrott. Stockholm: Brottsförebyggande rådet. Retrieved from
https://www.bra.se/download/18.779f51ff14b839896444f1/1424700908919/2015_Kor tanalys_Alkohol_och_drogpåverkan.pdf
Brottsförebyggande rådet. (2008). Brottsutvecklingen i Sverige fram till år 2007 (978-91-86027-15-5). Stockholm: Brottsförebyggande rådet. Retrieved from
https://www.bra.se/download/18.cba82f7130f475a2f180006972/2008_23_brottsuveck lingen.pdf
Bryman, A. (2011) Samhällsvetenskapliga metoder. Malmö: Liber.
Burke, D. (2017, 08). What is alcoholism, or alcohol use disorder? [Information site]. Retrieved 2020-01-16 from https://www.healthline.com/health/alcoholism/basics
Centers for Disease Control and Prevention. (2016). Excessive Alcohol Use and Risk to Women’s Health. Washington D.C.: U.S. Department of Health & Human Services. Retrived from https://www.cdc.gov/alcohol/fact-sheets/womens-health.htm
Clement, J. (2019, 09). Global social networks ranked by number of users 2019 [Article]. Retrieved 2019-09-19 from https://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-of-users/
Costa, P. T. & McCrae, R. R. (1990). Personality Disorders and The Five-Factor Model of personality. Journal of Personality Disorders, 4 (4th ed.), 362-371.
DOI:10.1521/pedi.1990.4.4.362
Datainspektionen. (n.d.). Dataskyddsförordningen (GDPR) [Information site]. Retrieved 2019-12-02 from
https://www.datainspektionen.se/lagar--regler/dataskyddsforordningen/
Donges, U-A., Jachmann, A., Kersting, A., Egloff, B. & Suslow, T. (2015). Attachment anxiety and implicit self-concept of neuroticism: Associations in women but not men.
Personality and Individual Differences, 72, 208-213. DOI: 10.1016/j.paid.2014.08.043
Dwyer, C. (09, 2018). 12 Common Biases That Affect How We Make Everyday Decisions [Information site]. Retrived 2020-01-17 from
https://www.psychologytoday.com/us/blog/thoughts-thinking/201809/12-common-biases-affect-how-we-make-everyday-decisions
Eshetu, Y., & Book, A. (2016). The Relationship between Personality Traits and Victimization. Personality and Individual Differences, 101, 476-476. DOI: 10.1016/j.paid.2016.05.129
Facebook. (2019, 11). Facebook groups [Internet community]. Retrieved 2019-12-02 from
https://www.facebook.com
Fairbairn, E. C., Sayette, A. M., Wright, C. G. A., Levine, M. J., Cohn, F. J. & Creswell, G. K. (2015). Extraversion and the Rewarding Effects of Alcohol in a Social Context. Journal of Abnormal Psychology, 124 (3). DOI: 10.1037/abn0000024
Field, A. (2013). Discovering Statistics Using IBM SPSS Statistics (4th ed.) London: Sage Publications.
Fisher, J. R. (1993). Social Desirability Bias and the Validity of Indirect Questioning. Journal of Consumer Research, 20 (2), 303-315. DOI: 10.1086/209351
Folkhälsomyndigheten. (2019, 07). Alkoholkonsumtion i befolkningen [Information site]. Retrieved 2019-11-28 from https://www.folkhalsomyndigheten.se/livsvillkor-
levnadsvanor/alkohol-narkotika-dopning-tobak-och-spel-andts/alkohol/alkoholkonsumtion/
George, G. L., Helson, R. & John, P. O. (2011). The “CEO” of Women’s Work Lives: How Big Conscientiousness, Extraversion, and Openness Predict 50 Years of Work
Experience in a Changing Sociocultural Context. Journal of Personality and Social
Psychology, 101 (4), 812-830. DOI: 10.1037/a0024290
Gullo, J. M., Dawe, S., Kambouropoulos, N., Staiger, K. P. & Jackson, J. C. (2010). Alcohol Expectancies and Drinking Refusal Self-Efficacy Mediate the Association of
Impulsivity With Alcohol Misuse. Alcoholism: Clinical and Experimental Research,
34 (8), 1386–1399. DOI:10.1111/j.1530-0277.2010.01222.x
Hakulinen, C., Elovainio, M., Batty, D. G., Virtanen, M., Kivimäki, M. & Jokela, M. (2015). Personality and alcohol consumption: Pooled analysis of 72,949 adults from eight cohort studies. Drug and Alcohol Dependence, 151, 110-114. DOI:
10.1016/j.drugalcdep.2015.03.008
Heinström, J. (2010). From Fear to Flow. Cambridge: Chandos Publishing.
Hughes, A. L. (2008). The Representation of Females in Criminological Research. Women
& Criminal Justice, 16 (1-2), 1-28. DOI:10.1300/J012v16n01_01
Ibáñez, I. M., Camacho, L., Mezequita, L., Villa, H., Moya-Higueras, J. & Ortet, G. (2015). Alcohol Expectancies Mediate and Moderate the Associations between Big Five Personality Traits and Adolescent Alcohol Consumption and Alcohol-Related Problems. Frontiers in Psychology. DOI:10.3389/fp-syg.2015.01838
John, O.P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In L.A. Pervin, & O.P. John (Ed.), Handbook of
Johnson, D. R. & Creech, J. C. (1983). Ordinal measures in multiple indicator models: A simulation study of categorization error. American Sociological Review, 48 (3), 398-407. DOI: 10.2307/2095231
Kotov, R., Gamez, W., Schmidt, F. & Watson, D. (2010). Linking “big” personality traits to anxiety, depressive, and substance Use disorders: a meta-analysis. Psychological
Bulletin, 136, 768-821. DOI:10.1037/a0020327
Littler, S. (n.d). The Importance and and Effect of Sample Size [blogpost]. Retrieved 2020-01-03 from https://select-statistics.co.uk/blog/importance-effect-sample-size/
Malouff, J., Thorsteinsson, E., Rooke, S. & Schutte, N. (2007). Alcohol Involvement and the Five-Factor Model of Personality: A Meta-Analysis. Journal of Drug Education. 37, 277–294. DOI: 10.2190/DE.37.3.d
Malouff, J. M., Thorsteinsson, E. B. & Schutte, N. S. (2005). The Relationship Between the Five-Factor Model of Personality and Symptoms of Clinical Disorders: A Meta-Analysis. Journal of Psychopathology and Behavioral Assessment, 27 (2), 101-114. Martin, L. J., Groth, G., Longo, L., Rocha, L. T. & Martens P. M. (2015). Disordered eating
and alcohol use among college women: Association with race and big five traits. Eating Behaviors, 17, 149-152. DOI: 10.1016/j.eatbeh.2015.02.002
McKinney, M. C., Caetano, R., Rodriguez, A. L. & Ngozi, O. (2010). Does Alcohol
Involvement Increase the Severity of Intimate Partner Violence? Alcoholism: Clinical
and Experimental Research, 34 (4), 655-658. DOI:
10.1111/j.1530-0277.2009.01134.x
Misswebweaver. (02, 2014). Friends - Common Interest [Information site]. Retrieved 2020-01-17 from https://possibilitiesunlimited.com/friends-common-interest/
Norman, G. (2010). Likert scales, levels of measurement and the “laws” of statistics.
Advances in Health Sciences Education, 15(5), 625-632. DOI:
10.1007/s10459-010-9222-y
