“The Influence of Subjective Norm, Descriptive Norm and Counterfactual Thoughts on Young
Adults’ Sexual Risk-Taking”
Schwanen, J. 11145773
Inleverdatum: 25-07-2016 Begeleider: Dr. Jonas, K.
1 The Influence of Subjective Norm, Descriptive Norm and Counterfactual Thoughts on Young Adults’
Sexual Risk-Taking Justine Schwanen
University of Amsterdam, Amsterdam Abstract
Although the Netherlands is recognized for its “sex positive environment”, sexual risk-taking is still a problem, especially among young heterosexual men and women (<25 years old). Current interventions aimed at decreasing sexual risk taking in the Netherlands mainly focus on risk perception
and knowledge, whilst information about counterfactual thinking and goal attainment is ignored. The expected decrease in sexual risk taking in adolescents is currently not achieved. Perhaps the importance of counterfactual thinking and related processes (i.e. social norms) is overlooked. Therefore, the relationship between social influence (subdivided into subjective- and descriptive norms), counterfactual thinking and sexual risk taking are examined. 86 psychology students between
the ages of 18 and 25 years (M = 20.40, SD = 1.74) from the University of Amsterdam filled out online questionnaires regarding social influence, counterfactual thinking and sexual risk taking. Although no support for the expected relationships was found, it is important to further investigate the role of counterfactual thinking in young adults sexual taking, for it can be useful in decreasing sexual
risk-taking. Follow-up studies should focus on characteristics and processes (e.g. causal inference and/or contrast- and assimilation effects) of counterfactual thinking instead of the extent to which
counterfactual thoughts are present.
Key words: sexual risk-taking, counterfactual thinking, counterfactual thoughts, Functional Theory of Counterfactual Thinking, subjective norm, descriptive norm, the Netherlands, adolescents, young adults
2 The Influence of Subjective Norm, Descriptive Norm and Counterfactual Thoughts on Young Adults’
Sexual Risk-Taking
“If I didn’t went to bed so late, I would have been more productive today”, “If I enrolled for
dance academy instead of medicine, I might have been a dancer today”, “If we went shopping on a week-day instead of a weekend-day, it wouldn’t be so busy in town”. What might have happened if we
decided on some past choices differently? Human beings have a tendency to create alternative outcomes for events that already happened. Imagining alternative outcomes is called counterfactual thinking. Counterfactual thinking involves using mental simulation by “thinking about what might have been”, and can be conceptualized as the conditional proposition “if X, then Y”, in which X stands for an outcome-related antecedent and Y stands for the outcome (Epstude & Roese, 2008). Another example of a counterfactual thought would be: “If I had studied more (X), then I would have achieved a higher grade (Y).”
Counterfactual thoughts can be seen as expressions of unattained goals. are useful in that they explain past events and regulate emotions and behaviour change.
Many different theories explaining the functions and categorization of counterfactual thoughts have been proposed during the past few decades. One of the earliest attempts to explain counterfactual thinking was by using the norm theory. According to this theory, experiences are compared to norms, knowledge of what is considered to be “normal” (Kahneman & Miller, 1986). However, these norms were thought to be constructed in a bottom-up fashion: it was assumed that only after experiencing a certain outcome, a comparison with earlier similar circumstances was made. This would result in the current outcome begin termed as normal (similar to earlier experiences) or abnormal (different from earlier experiences) (Roese & Olson, What might have been: The social psychology of counterfactual thinking., 2014). Abnormal outcomes would provoke counterfactual thinking by cognitively altering aspects of antecedents of the outcome that are easy to alter
3 alter imaginatively, whilst altering “the quality of the attended lessons” probably is not (Roese & Olson, What might have been: The social psychology of counterfactual thinking., 2014).
Counterfactual thinking was, according to the norm theory, interpreted as a form of biased judgment, which lead to much subsequent research that was consistent with this interpretation (Epstude & Roese, 2008), e.g. the reflection-evaluation (REM) model (Markman, McMullen, & Elizaga, 2008; Markman & McMullen, 2003). However, current research shows that counterfactual thinking can also be beneficial, and thus refutes the previous mentioned theories partly (Epstude & Roese, 2011). Nonetheless, the main incentive driving counterfactual thinking was discovered: the motivation to alter unwanted or unexpected outcomes and therefore to regulate future behaviour (Roese & Olson, What might have been: The social psychology of counterfactual thinking., 2014). This is also the core of one of the most recent theories of counterfactual thinking: the Functional Theory of
Counterfactual Thinking (Epstude & Roese, 2008).
The Functional Theory of Counterfactual Thinking differs from earlier theories in that it focuses on top-down rather than bottom-up processes. Also, counterfactual thinking is seen as useful and beneficial instead of being biased an impediment (Epstude & Roese, 2008), for it provides a role in regulating behaviour in order to improve performance (Epstude & Roese, 2011).
According to the Functional Theory of Counterfactual Thinking, there are three main characteristics distinguishing which kind of counterfactual thoughts best stimulate behaviour regulation (Epstude & Roese, 2008) (see table 1). Firstly, the direction of a counterfactual thought specifies if the current situation is compared with a better evaluated outcome (upward counterfactual thinking) or a worse evaluated outcome (downward counterfactual thinking) (Markman & McMullen, 2003).
Secondly, the structure of a counterfactual thought indicates whether the counterfactual refers to a regret of inaction (additive counterfactual thinking) of a regret of action (subtractive counterfactual thinking) (Epstude & Roese, 2011). Additive counterfactual thoughts are more common (Savitsky, Medvec, & Gilovich, 1997) and have a higher chance on causing a change in behaviour than
4 subtractive counterfactual thoughts, because they are more precise in specifying which actions should have been taken to reach the desired outcome (Epstude & Roese, 2008). In other words, they have a promotion-focus. Subtractive counterfactual thoughts, on the other hand, have a prevention-focus, focusing on negated information, which requires more controlled mental operations and is therefore harder to process (Epstude & Roese, 2008).
Finally, counterfactual thoughts can be either focused on the self or on others. Counterfactual thoughts focusing on the self are more effective in stimulating a change in behaviour then
counterfactual thoughts focusing on others, because of their focus on self-improvement (Epstude & Roese, 2008); changing behaviour of others is harder than changing your own behaviour (Roese & Olson, What might have been: The social psychology of counterfactual thinking., 2014).
Table 1
Description of characteristics in counterfactual thoughts
Characteristic Description Example
Direction Direction is related to the outcome (Y) of the counterfactual thought, and indicates a comparison of the current situation with an evaluated better one (upward counterfactual thinking), or an evaluated worse one (downward counterfactual thinking).
“If I had studied harder, I would
have had a higher grade”.
“If I hadn’t studied as hard, I
would have had a lower grade”.
Structure Structure is related to the antecedent (X) of the counterfactual thought, and specifies the desire to add (additive counterfactual thinking) or subtract (subtractive counterfactual thinking) an aspect from the present state.
“If I had studied harder, I would
have had a higher grade”.
“If I hadn’t studied harder, I
would have had a lower grade”.
Self vs. other Self vs. other refers to the focus on one’s own
actions as opposed to the action of others. “If I was less distracted, I would have had a higher grade”.
“If you hadn’t distracted me, I
would have had a higher grade”.
Now that the basic characteristics of counterfactual thoughts are explained, we arrive at the core of the Functional Theory of Counterfactual Thinking, which distinguishes between content-specific and content-neutral pathways of counterfactual thinking. Both pathways are assumed to lead to an intention to change behaviour in order to achieve a desired outcome (Epstude & Roese, 2011; Smallman & Roese, 2009).
5 The content-specific pathway is activated by the occurrence and recognition of a problem or unsatisfying situation. In order to solve this problem, a desired outcome (Y) is connected to a
behaviour (X) (Epstude & Roese, 2011), using causal inference (Epstude & Roese, 2008). Causal inference identifies plausible causes for not reaching the desired outcome (Roese, 1997). For example “If I had studied more (X), I would have gotten a higher grade (Y)” is causally associated with studying harder, because the counterfactual thought implies that studying harder will cause higher grades. The information about the behaviour that should have been performed in order to achieve the desired outcome is then converted into behavioural intentions (Smallman & Roese, 2009), leading to (new) attempts of reaching the desired outcome (Epstude & Roese, 2011).
Within the content-specific pathway, upward counterfactual thoughts are more useful for behaviour regulation, because they are more precise in specifying what actions and strategies need to be utilized to reach the desired outcome than downward comparisons (Smallman & Roese, 2009). Also, according to the alternative outcome bias theory, upward counterfactual thoughts are expected to lead to more changes in behaviour than downward counterfactual thoughts, because they induce negative evaluations and feelings of regret regarding the current situation, which serve as a
motivation to achieve the desired outcome (Seta, Seta, Petrocelli, & McCormick, 2015). In the content-neutral pathway on the other hand, no specific causal connections or intentions are formed. The content-neutral pathway stimulates changes in behaviour through a general style of information processing or motivation that is not content related (Smallman & Roese, 2009). According to this pathway, change in behaviour is caused by negative affect, which stems from (1) a contrast effect in upward counterfactual thinking, or (2) an assimilation effect in downward counterfactual thinking (Epstude & Roese, 2008). Contrast effects occur when an imagined alternative outcome is used as a comparison standard to which the current outcome is compared. This
comparison results in more extreme evaluations of the current outcome (Epstude & Roese, 2011). For example: comparing a currently achieved grade B to an imagined grade A results in a more negative evaluation of the achieved grade. Assimilation effects occur when the resemblance between the
6 actual – and the alternative, imagined outcome is considered (Markman & McMullen, 2003), thereby temporarily experiencing the effects of the alternative outcome (e.g. achieving a higher grade) as if they were real. Thus, within the content-neutral pathway, upward and downward counterfactual thoughts as well as contrast and assimilation effects are lead to changes in behaviour intentions (Epstude & Roese, 2011).
Connection with norms
Counterfactual thoughts are not generated randomly: different people tend to generate similar counterfactual thoughts, which can be explained by counterfactual thoughts being connected to norms, as was already mentioned in one of the first theories of counterfactual thinking (e.g. norm theory, as described above). Which counterfactual thoughts are generated is thus guided by what is currently perceived to be normal (Landman & Manis, 1992). This is a functional process, because “what is perceived to be normal” represents scenario’s that are statistically frequent (and we are most likely to encounter in the future) as well as scenario’s that conform to prescriptive social norms. By aiming to comply to scenario’s that conform to prescriptive social norms, we make sure (the results of) our behaviour is socially accepted (Hitchcock & Knobe, 2009).
Connection with goals
Next to norms, the Functional Theory of Counterfactual Thinking connects to a great extent with goal cognition theories (Epstude & Roese, 2008), for they both assume behavioural intentions to have a main role in achieving desired outcomes (e.g. goals) (Smallman & Roese, 2009). In the
procedure of achieving goals, different processes are involved. When there is a discrepancy between the current state and the desired outcome, negative evaluations about the situation arise (Epstude & Roese, 2008; Förster, Libermand, & Higgins, 2005). To eliminate this tension as quickly as possible, a cognitive advantage for goals (and constructs associated with that goal) that have not yet been achieved is generated. In this way, unattained goals and closely related constructs stay cognitively more accessible, so that self-regulatory efforts towards reaching the goal are fostered (Förster, Libermand, & Higgins, 2005). This process is called the Zeigarnik effect (Zeigarnik, 1927), in which
7 counterfactual thoughts have the role of preparing for future action in order to be able to reach these alternative outcomes (Epstude & Roese, 2008). The cognitive advantage for goals and goal-related constructs is inhibited after the goal is attained, a process called post-fulfillment inhibition (Förster, Libermand, & Higgins, 2005).
The connection between counterfactual thinking and goals is an important finding, for all human behaviour is to a great extent driven by the desire to realize certain desired outcomes (e.g. goals) (Stroebe, 2011). Knowledge about how these goals are achieved may be used in intervention programs when attempts at goal attainment become problematic: Sometimes unhealthy or harmful behaviour is used to attain goals. This is the case in the goal of having sex, a general hedonistic goal in human beings, which is accompanied with taking sexual risks (Broek, van den, et al., 2010; Oeffelen, van, et al., 2015). Therefore, the goal of this study is to examine the relationships between
counterfactual thinking, goal theories, norms and sexual risk taking, in order to provide knowledge that is useful for the development of intervention programs aimed at reducing these risks.
Sexual risk-taking
Although the Netherlands is recognized for its “sex positive environment”, in which sexuality is accepted and teaching about sexual responsibility is part of children’s education (Ferguson,
Vanwesenbeeck, & Knijn, 2008), sexual risk-taking is still a problem. The amount of STI clinic attendees reporting not having used a condom during their last sexual intercourse, in addition to the average number of sex partners they had during the last six months (>three per person), as well as the number of STI diagnoses in the Netherlands keep climbing (Broek, van den, et al., 2010; Oeffelen, van, et al., 2015). Especially young heterosexual men and women (<25 years old) tend to engage in sexual risk behaviour; they belong to the largest risk group in the Netherlands (Rijksinstituut voor
Volkgsgezondheid en Milieu (RIVM), 2016).
The increase in sexual risk-taking is peculiar, considering the large amount of education and intervention programs aimed at reducing sexual risk-taking that are provided in the Netherlands (for example “Girls Talk”, “Boys R Us” and “Relations and sexuality”) (Mouthaan & Vlugt, van der, 2012).
8 However, the focus of these intervention programs appears to be mainly on increasing knowledge and risk perception (Mouthaan & Vlugt, van der, 2012; Steinberg, 2010), whilst information about
counterfactual thinking and goal attainment is ignored. Perhaps the importance of these processes is overlooked. Therefore, it is useful to examine the relationships between sexual risk-taking and counterfactual thinking/goal attainment, so that that these factors can be potentially implemented in intervention programs.
Sexual risk taking in relation to counterfactual thoughts and norms
As mentioned earlier, the Zeigarnik effect ensures goal-attainment by making goal-related constructs cognitively more accessible (Zeigarnik, 1927). Although the Zeigarnik effect is applicable to all goal-directed behaviour, specifically the goal of having sex contains a characteristic that makes it prone to provoke risk-behaviour. To wit, the desire to engage in sexual behaviour is a hedonistic goal with no specific end state.
Broad, hedonistic goals, in contradiction to concrete, cognitive goals such as achieving an exam, are never fully attained, because an end state is not specified, and therefore the goal is never met. Consequently, the Zeigarnik effect is continuously active (Jonas, 2016).The post-fulfillment effect mentioned earlier thus does not apply for hedonistic goals, resulting in adolescents continuously engaging in sexual behaviour (Oeffelen, van, et al., 2015). Although there is nothing wrong with engaging in sexual behaviour per se, it becomes problematic when sexual risks (such as having unprotected intercourse) are taken in order to reach these goals. Two processes might explain adolescents’ sexual risk behaviour.
Firstly, the risks taken during sexual behaviour can be explained by the cushion effect: Knowledge of being capable of performing sexual behaviour serves as a cushion, softening negative emotions associated with taking risks and stimulating risk-taking that wouldn’t otherwise be appealing (Jeffrey, Onay, & Larrick, 2010).
Secondly, another explanation for the risks taken during adolescent sexual intercourse is social influence, an important aspect that is also integrated in counterfactual thinking. Compared to other
9 life stages, adolescents tend to find social approval especially important. For example, adolescents engage in more risky behaviour when (overt or covert) watched by their peers than when they are alone (Chein, Albert, O'Brien, Uckert, & Steinberg, 2012; Widman, Choukas-Bradley, Helms, & Prinstein, 2016).
The important social influence described above can be operationalized as subjective norm: The perceived likelihood of what important others expect one to do multiplied with the motivation to comply to that expectation (Stroebe, 2011). When considering subjective norms, behaviour that is related to health-risk is cognitively more salient than behaviour that is health-promoting. Therefore, for instance, when deciding whether or not to use condoms, people are influenced more by images of role models not using condoms than by images of models that do use a condom (Blanton, et al., 2001). Indeed, research shows that subjective norms influence sexual risk-taking. For example, there is a strong association between subjective norms and condom use (Sheeran & Taylor, 1999; Gebhardt, Kuyper, & Greunsven, 2003).
However, when measuring subjective norm, responses on items often have low variability. Important others generally are perceived to approve of desirable behaviour and disapprove of undesirable behaviour, independently from how they act themselves (Ajzen, 2002). Subjective norm thus does not fully cover the range of social influences that work together to influence behaviour (Park, Klein, Smith & Martell, 2009). Therefore, when measuring subjective norm, it is advised to add items capturing descriptive norms (Rivis & Sheeran, 2003; Sheeran & Taylor, 1999; Park, Klein, Smith, & Martell, 2009; Ajzen, 2002). Descriptive norms can be operationalised as the person’s perception of how others act in comparable situations (Ajzen, 2002). In line with this, van Empelen and Kok (2008) showed that adolescent sexual risk-taking depends on the perceived “right thing to do” (subjective norm) but also on the person’s perception of how others act in comparable situations (descriptive norm) (Empelen & Kok, 2008).
10 The current study
Recapitulating, the goal of this study is to examine whether counterfactual thoughts might contribute to adolescents’ sexual risk behaviour. It is important to also involve social influence, as social influences determine how to act in certain situations and thus which counterfactual thoughts are generated (Hitchcock & Knobe, 2009). Examining the relationships between counterfactual thoughts, social influence and sexual risk-taking may provide new insights that are useful for the development of interventions, focusing on factors, other than risk perception and knowledge, that might ensure a decrease in sexual risk-taking (Oeffelen, van, et al., 2015; Steinberg, 2010).
Based on previously mentioned research several hypothesis are assumed. Firstly, it is hypothesized that adolescents are very susceptible to social influence. Therefore, experiencing high subjective and descriptive norms will increase sexual risk-taking (Gebhardt, Kuyper, & Greunsven, 2003; Rivis & Sheeran, 2003; Sheeran & Taylor, 1999).
Secondly, because of the desire to comply to mentally simulated preferred outcomes, counterfactual thoughts will lead to an increase in adolescent sexual risk-taking (Epstude & Roese, 2008; Seta, Seta, Petrocelli, & McCormick, 2015). Counterfactual thoughts that will lead to increased sexual risk-taking are expected to be upward instead of downward, because (1) they are more common, (2) they are more precise in operationalizing in what needs to be done to achieve a desired outcome, and (3) they lead to negative evaluations and feelings of regret about the current situation, which serves as a motivation to reach an alternative (better evaluated) outcome (Epstude & Roese, 2008; Seta, Seta, Petrocelli, & McCormick, 2015) and additive (focusing on a regret of inaction) instead of subtractive (focusing on a regret of action) (Epstude & Roese, 2008).
Thirdly, because mentally simulated alternatives are influenced by what is thought to be normal (Kahneman & Miller, 1986), subjective and descriptive norms are expected to affect the generation of counterfactual thoughts. Therefore, the presence of subjective and descriptive norms will result in the generation of more counterfactual thoughts.
11 Finally, the ways in which subjective norms and counterfactual thoughts interact will be analyzed. Subjective and descriptive norms are expected to moderate the relationship between counterfactual thoughts and adolescent sexual risk-taking.
Thus, the following hypotheses will be investigated: 1 A. Subjective norms increase sexual risk-taking in adolescents. 1 B. Descriptive norms increase sexual risk-taking in adolescents. 2. Counterfactual thoughts increase sexual risk-taking in adolescents.
3 A. Upward counterfactual thoughts induce more sexual risk-taking than downward counterfactual thoughts.
3 B. Additive counterfactual thoughts induce more sexual risk-taking than subtractive counterfactual thoughts.
3 C. There is a difference in sexual risk-taking based on direction and structure of the counterfactual. 4 A. Subjective norms increase the generation of counterfactual thoughts.
4 B. Descriptive norms increase the generation of counterfactual thoughts.
5 A. Subjective norm moderates the relationship between counterfactual thoughts and sexual risk-taking in adolescents.
5 B. Descriptive norm moderates the relationship between counterfactual thoughts and sexual risk-taking in adolescents.
Method Participants
Based on a power analysis (p=.05, 1-β=.8, E=0,3), it was determined that a minimum of 82 participants was required (Faul, Erdfelder, Lang, & Buchner, 2009). This criterion is met: the actual sample included 96 participants, of which ten were excluded due to the fact that their amount of missing items
exceeded 20%. The final sample included 86 psychology students between the ages of 18 and 25 years (M = 20.40, SD = 1.74) from the University of Amsterdam. Seventeen of them were male (19.8%) and 69 of them were female (80.2%). Duo to an malfunction in Qualtrics (the questionnaire programming
12 software that was used), two participants were unable to fill out the questions regarding sexual risk-taking. Missing cases will be deleted pairwise from calculations, thus sample sizes will differ according to the variables that are used in the analysis.
Sampling Procedure
The survey was created using Qualtrics, an online survey software tool (Qualtrics, 2016). An online link at the website of the university University of Amsterdam (https://www.lab.uva.nl/lab) was provided so that psychology students were able to enroll themselves in the study. Only participants that met the criteria of being sexually active and being between the ages of 18 and -25 years old were included. Informed consent was included and the participants’ anonymity was assured. Participation was rewarded with 0.5 research credits. Acquiring these credits is an obligatory part of the curriculum of psychology students.
Procedure
A cross-sectional, quasi-experimental study is used to provide an overview of the relationships and coherence between the independent variables subjective norm, descriptive norm and counterfactual thoughts and the dependent variable adolescents’ sexual risk-taking. Participants that enrolled themselves via lab.uva.nl are able to fill in the questionnaire online. Before starting the survey, information regarding the goal of the current study, anonymity and confidentiality, assurance and voluntariness is provided. Also, participants are asked to accept a declaration of consent, which is necessary for continuing in the survey to (see Appendix A). Subsequently, participants fill in questionnaires regarding (1) demographic variables, (2) subjective and descriptive norms, (3)
counterfactual thoughts, and (4) sexual risk-taking. Filling in the questionnaire takes approximately ten minutes.
Variables and measures
An overview of variables and measures is given in table 2. Subjective and descriptive norms are measured using 7-point Likert Scale questionnaires, composed of existing, valid, and reliable questionnaires (Ajzen, 2002; Francis, et al., 2004) (see Appendix B). Three items measure perceived
13 norm (for example “The people in my life whose opinions I value the most disapprove/approve of me using condoms every time I have sexual intercourse”). Three items measure motivation to comply (for example “Peers’ approval of my condom-use behaviour is important to me: not at all/very much”). Subjective norm is calculated by multiplying each perceived norm-items with the accompanying motivation to comply-item, and then adding them up (Ajzen, 2002; Francis, et al., 2004). Descriptive norm is defined as the person’s perception of how important others act in comparable situations (Ajzen, 2002), and will be measured by using two items (“The people in my life whose opinions I value use/do not use condoms every time they have sexual intercourse” and “Most people who are
important to me use condoms every time they have sexual intercourse”). A mean score of these two items is computed (Ajzen, 2002; Francis, et al., 2004). All items are based on recommendations from Ajzen (2002) and subsequently translated into Dutch.
Table 2
Description of measures used in the questionnaire
Variable Subject of items Description of items Nr. of
items Score-calculations Subjective
norm The perceived likelihood of what important others expect one to do (PN)
Likert scales ranging
from 1 to 7 3 (PN1 x MC1) + (PN2
x MC2) + (PN3 x
MC3)
Motivation to comply to
the perceived norm (MC) Likert scales ranging from 1 to 7 3 Descriptive
norm Perception of how important others act in comparable situations (DN)
Likert scales ranging
from 1 to 7 2 (DN
1 + DN2) / 2
Counterfactual thinking
Thoughts about “what might have been”
Likert scales ranging from 1 to 7
12 (DN1 + DN2 + DN3 +
… + DN12) / 12
Thoughts about “what might have been”, related to sexual behaviour
Participants write a short essay about counterfactual thoughts concerning their own sexual behaviour 1 Categorizing answers in direction (upward / downward) and structure (additive / subtractive) Sexual
risk-taking Sexual risk-taking among college students Sliding bar range from 0 to 100 14 (DN
1 + DN2 + DN3 +
14 The presence of counterfactual thoughts is measured using the Counterfactual Scale, which consists of 12 items covering the concept of “thinking of what might have been” (see Appendix C). Participants indicate the extent to which the items are applicable to them, with scores ranging from 1 (low presence of counterfactual thoughts) to 7 (high presence of counterfactual thoughts). The mean score over these 12 items is then calculated, representing the total counterfactual score (CFcon). Additionally, participants write a short essay about counterfactual thoughts concerning their own sexual behaviour. These essays are analysed, by categorizing them in direction (downward = 0, upward = 1) and structure (subtractive = 0, additive = 1) (CFdich), using the “Coding Scheme Counterfactual Thinking” (see appendix D). Counterfactual thoughts are thus represented by both a continuous and a dichotomous score.
For measuring sexual risk-taking, the Sexual Risk Survey (SRS), a questionnaire assessing sexual risk-taking among college students was used. The SRS has an overall internal consistency of .88 and a test-retest reliability of .93. It consists of 23-items, distributed over five scales (see Appendix E). The scales cover the concepts of sexual risk-taking with uncommitted partners, risky sex acts (e.g. not using condom), impulsive sexual behaviour, intention to engage in risky sexual behaviour, and risky anal sex acts (Turchik & Garske, 2008). However, the survey appeared to be too long and outdated (Jonas, 2016). Therefore, the survey is modified and reduced to fourteen questions. In addition, the survey is translated into Dutch (see Appendix F). Using a slide bar ranging from 0 to 100, participants are inquired to indicate the extent to which the items apply to their (sexual) behaviour in the last six months. To obtain the total SRS score, a mean score is calculated (Turchik & Garske, 2008) . During translations of the questionnaire on social influence as well as the SRS, conceptual rather than literal word-for-word translation was emphasized.
Analysis
A Pearson product-moment correlation coefficient is calculated for assessing H1, H2 and H4. A significant, positive, moderate to strong correlation is expected (with p level < .05 and r=.03-1) (Field, 2009).
15 For H3, a two-way factorial analysis of variance is performed, testing for main and interaction effects of direction and structure of counterfactual thoughts on sexual risk-taking. When the
interaction effect is significant, simple effect analyses need to be performed. Effect sizes are interpreted according to the following guidelines: small: = .01; medium: =.06; large: =.14 (Field, 2005).
Furthermore, there might be a moderation effect of social influence on the relationship between counterfactual thoughts and sexual risk-taking. To test these hypotheses (5A and 5B), the variables subjective norm, descriptive norm and counterfactual thoughts are centered, and interaction terms between the social influence variables and counterfactual thoughts are computed.
Subsequently, multiple regression analyses are carried out. A significance level of alpha = .05 is used. Results
Social influence and sexual risk-taking
Boxplot and histogram analyses revealed no significant outliers (see Appendix G). A Pearson product-moment correlation coefficient was computed to assess the relationship between the amount of social influence that the participant experiences and the amount of sexual risk-taking, as stated in hypothesis. No significant correlation appeared between descriptive norm and sexual risk-taking, r(84) = .085, p = .442. Therefore hypothesis 1A is rejected. Although the relationship between subjective norm and sexual risk-taking is borderline significant, r(84)= .21, p = .053, the strength of the relationship is weak (r < .3). Therefore hypothesis 1B is also rejected. Social influence and sexual risk-taking do not seem to be associated with each other.
Counterfactual thoughts and sexual risk-taking
According to hypothesis 2, the amount of counterfactual thoughts participants experience increases sexual risk-taking. Again, a Pearson product-moment correlation coefficient was computed to assess relationship. There was no significant correlation between the two variables, r(84) = .033, p = .768. Therefore, hypothesis 2 is rejected.
16 Furthermore, it was specified in hypothesis 3 that upward and additive counterfactual
thoughts induce increased sexual risk-taking. A two-way factorial analysis of variance was conducted to assess the influence of direction and structure of counterfactual thoughts on the amount of sexual risk-taking. Both independent variables consisted of two levels (direction: upward, downward; structure: additive, subtractive). A Levene’s test did not yield an indication for unequal variances, F (3,73) = .08, p = .973. The main effect for direction of counterfactual thoughts yielded an F-ratio of F( 1,73) = .10, p = .750, 2
p
η = .001, indicating that the difference in sexual risk-taking between
participants who wrote an example about upward counterfactual thoughts (M = 42.35, SD = 2.47) and participants who wrote an example about downward counterfactual thoughts (M = 44.72, SD = 5.60) is not significant. Therefore, hypothesis 3A is rejected.
The main effect for structure of counterfactual thoughts yielded an F-ratio of F(1,73) = 6.92, p = .010, 2
p
η = .087. Participants who wrote an example including additive counterfactual thoughts took significantly more sexual risks over the past six months (M = 50.97, SD = 4.98) than participants who wrote an example including subtractive counterfactual thoughts (M = 35.42, SD = 3.18). Therefore, hypothesis 3B is accepted.
There was a statistically significant interaction effect between direction and structure of counterfactual thoughts, F (1,73) = 6.09, p = .016, 2
p
η = .077, meaning that the effect of structure on sexual risk-taking is different for upward and downward counterfactual thoughts. An analysis of simple effects showed that the effect of structure is significant for participants who wrote an example using downward counterfactual thoughts, F (1,73) = 7.78, p = .007, 2
p
η = .096. From the participants that wrote an example of a downward counterfactual, those using additive counterfactual thoughts scored significantly higher on sexual risk-taking (M = 59.21, SE = 9.36) than those using subtractive
counterfactual thoughts (M = 29.07, SE = 5.41). When the counterfactual thoughts were upward, there was no significant difference between participants that wrote an example using an additive
17 counterfactual (M = 42.72, SE = 3.42) and those that used a subtractive counterfactual thoughts (M
=41.77, SE = 3.36), F(1,73) = .040, p = .842, 2
p
η = .001 (see table 1). Table 1
Mean differences in structure of counterfactual thoughts based on direction.
Direction Structure Mean Std. Error Sig.*
Upward Additive 42.73 3.42 .842
Subtractive 41.77 3.36
Downward Additive 59.21 9.36 .007
Subtractive 29.07 5.41
*The mean difference between additive and subtractive counterfactual thoughts at a .05 level.
The effect of direction was marginally significant for participants who wrote an example using a subtractive counterfactual, F(1,73) = 3.93, p = .050, 2
p
η = .052. Within the group of participants that wrote an example using subtractive counterfactual thoughts, those using upward counterfactual thoughts scored marginally significantly higher on sexual risk-taking (M = 41.77, SE = 3.36) than those using downward counterfactual thoughts (M = 29.07, SE = 5.41). When the counterfactual thoughts were additive, there was no significant difference between participants that wrote an example using upward counterfactual thoughts (M = 42.73, SE = 3.42) and those that wrote an example using downward counterfactual thoughts (M = 59.21, SE = 9.36), F(1,73) = 2.74, p = .102, 2
p
η = .036 (see table 2).
Table 2
Mean differences in direction of counterfactual thoughts based on structure.
Structure Direction Mean Std. Error Sig.*
Additive Upward 42.73 3.42 .102
Downward 59.21 9.36
Subtractive Upward 41.77 3.36 .050
Downward 29.07 5.41
18 Social influence and counterfactual thoughts
For testing H4, the relationship between social influence and counterfactual thoughts was examined using Pearson product-moment correlation analyses. No significant correlation appeared between subjective norm and counterfactual thoughts, r(86)= .081, p = .546, and descriptive norm and counterfactual thoughts, r(86) = -.046, p = .677. Therefore hypotheses 4A and 4B are rejected.
hypotheses 4A and 4B are rejected. Moderation effect of social influence
It was stated in H5 that social influence moderates the relationship between counterfactual thoughts and adolescent sexual risk-taking. Although previous analyses showed none of the expected relationships to be significant, multiple regression analyses were performed to confirm these results. Two separate multiple regression analyses were run to predict sexual risk-taking from counterfactual thoughts and social influence, using the enter method. All independent variables were centered first, and accompanying interaction terms were calculated. First, subjective norm was taken as a moderator (hypothesis 5A). The data met assumptions of collinearity (SN, Tolerance = .98; VIF = 1.02; CF,
Tolerance = .98, VIF = 1.02; Interaction term, Tolerance = .98, VIF = 1.02), independent errors (Durbin-Watson value = 2.13), normally distributed errors, homogeneity of variance, linearity and non-zero variances (SN, variance = 299.29; CF, variance = .51; Interaction term, variance = 162.65). The
regression analysis was not statistically significant, F (3,80) = 1.90, p = .136, 6.7% variance is explained by the model. Therefore, hypothesis 5A is rejected.
Subsequently, descriptive norm was taken as a moderator (hypothesis 5B). The data met assumptions of collinearity (DN, Tolerance = .98; VIF = 1.02; CF, Tolerance = 1.00, VIF = 1.00; Interaction term, Tolerance = .98, VIF = 1.02), independent errors (Durbin-Watson value = 2.13), normally distributed errors, homogeneity of variance, linearity and non-zero variances (DN, variance = 2.04; CF, variance = .51; Interaction term, variance = 1.04). The regression analysis was not statistically significant, F (3,80) = .25, p =.863. Only 0.9% variance can be explained by the model. Therefore, hypothesis 5B is also rejected.
19 Discussion
The goal of this study was to examine the relationships between subjective- and descriptive norms, counterfactual thoughts and sexual risk-taking, thereby contributing to the development of interventions focused on decreasing sexual risk-taking. Currently, information regarding the interplay between these variables is lacking, which makes this study of great value. Based on literature studies, several hypotheses were formulated.
Although these exists a large body of evidence for the relationship between social influence and sexual risk-taking (Azjen, 2002; van Empelen & Kok, 2008; Gebhardt, Kuyper & Greunsven, 2003; Rivis & Sheeran, 2003; Sheeran & Taylor, 1999), this finding could not be replicated by the current study (H1). There also seems to be no relationship between counterfactual thoughts and sexual risk-taking (H2) and social influence and counterfactual thoughts (H4). Subsequently, no moderation effect of social influence on the relationship between counterfactual thoughts and sexual risk-taking was found (H5).
There was a relationship between the two subdivisions of counterfactual thoughts (i.e. direction and structure) and sexual risk-taking (H3), but the results from the simple effects analyses contradicted with our expectations. Although additive, upward – rather than subtractive, downward counterfactual thoughts were expected to increase sexual risk-taking amongst adolescents, the current data showed increased sexual risk-taking for additive, downward counterfactual thoughts (for example “If I had given him a blowjob, as he pleased (additive), I would have felt submissive
(downward)”). This is a peculiar finding, because it states that downward counterfactual thoughts (thinking about a less desirable situation), induces sexual risk taking, which is in contradiction with the alternative outcome bias theory (Seta, Seta, Petrocelli, & McCormick, 2015) and the Functional Theory of Counterfactual thinking (Epstude & Roese, 2008). The alternative outcome bias theory states that upward counterfactual thoughts induce negative evaluations and feelings of regret regarding the current situation (Seta et al., 2015), which subsequently leads to changes in behaviour in order to reach a more desired situation (Jonas, 2016). The Functional Theory of Counterfactual Thinking states
20 that upward counterfactual thinking is more common and more precise in formulation behavioural intentions than downward counterfactual thinking (Epstude & Roese, 2008), and therefore leads to more changes in behaviour. However, these theories do not apply to the data.
Several factors might explain these unexpected results: Participants were asked to give only one example, which solely describes the way of thinking in that particular situation, and thus could be an exception on his or her general way of thinking in sexual situations. Secondly, only four
counterfactual thoughts were coded as downward as well as additive (see Table 3). This is not a representative sample for adolescents in the Netherlands, and therefore the data are not
generalizable. Thirdly, despite the neutral attitude of the researcher, some counterfactual thoughts are hard to categorize as downward or upward due to lack of context. For example “If I had not drank
as much, I would not have had sex” is rated as an upward counterfactual thought, but could also be
interpreted as a downward counterfactual thought. Furthermore, the structure and direction of counterfactual thoughts was rated by only one researcher, which makes data susceptible for bias and misinterpretations. For follow-up studies it is recommended to develop a more valid and reliable instrument for measuring the direction and structure of counterfactual thoughts. Finally, and perhaps most important, thinking about a less desirable outcome (for example “not having sex”) could be an incentive to avoid that outcome. This would partly explain the results, as it would promote sexual risk-taking in cases where the downward counterfactual states not having sex as a less desirable outcome. Although in downward counterfactual thoughts it is not specified exactly what needs to be done order to achieve the desired outcome, these counterfactual thoughts may induce a general motivation to achieve more sex through assimilation effects, as is stated in the content-neutral pathway of the Functional Theory of Counterfactual thinking (Epstude & Roese, 2008; Smallman & Roese, 2009). Also, perhaps the direction of a counterfactual thought per se is not determinative in increasing sexual risk taking, but the process that lies behind it is. Therefore, it might be more interesting to examine the roles that different processes (causal inference, contrast effects and/or assimilation effects) have in
21 sexual risk taking instead of looking solely at the direction (upward vs. downward) of counterfactual thoughts.
Characteristics of the sample
Furthermore, there are some characteristics of the sample that may have influenced the results. Firstly, there was a difference found in sexual risk taking between males and females, t(82) = 2.19, p = .031. Men scored significantly higher on sexual risk taking (M = 49.94, SD = 22.71) than women (M = 38.75, SD = 17.70), which is in accordance with previous research on sexual risk taking (Sheeran & Taylor, 1999). In addition, sexual risk taking differed between people with a different relationship status, t(82) = -5.03, p < .001). Single participants scored significantly higher on sexual risk taking (M = 50.97, SD = 16.65) than participants who were in a relationship (M = 32.39, SD = 17.11).
According to these results, future interventions should mainly focus on single, male
adolescents, because they scored significantly higher on sexual risk-taking than female adolescents and those in a relationship. It seems likely that single adolescents take more risks than those in a relationship, considering they are not expected to be monogamous. However, the current sample does not represent the gender ratio of the population, since only 19,8% (n=16) of the sample was male and 80.2% (n=96) was female. Therefore, conclusions based on gender cannot be drawn from this study. It was chosen not to split the sample when performing analyses, for the subsamples would have been too small, and would not represent the gender / relationship status differences in young heterosexual man and women in the Netherlands correctly.
Furthermore, the results are based on psychology students with an higher overall education level than the average adolescent or young adult in the Dutch population. In the current sample, 86,6% (n=74) achieved a VWO/gymnasium/athenaeum education level (the highest high school-leven in the Netherlands), whilst only approximately 17% of high school students achieve that education level in the Dutch population (Centraal Bureau voor de Statistiek, 2012)
22 Limitations
Information regarding the relationship between counterfactual thoughts and sexual risk-taking is currently not available. Therefore it is possible, although theorized otherwise, that truly no
relationship between the two variables exists. However, the relationship between social influence and sexual risk-taking is frequently confirmed in the literature (Blanton, et al., 2001; Empelen & Kok, 2008; Gebhardt, Kuyper, & Greunsven, 2003; Sheeran & Taylor, 1999), but not found in the current study. Therefore, limitations that may have influenced the results are discussed. First of all, participants consisted of psychology students of the University of Amsterdam (UvA) between ages of 18-25, for this time period represents adolescence (Fivush & Zaman, 2015), and the theoretical framework consisted mostly of adolescent-based research. However, the age range of adolescents depends on the definition that is being used. For example, the World Health Organization (2016) defines adolescence as the critical transition in biological processes and the start of puberty, ranging from ages 10 to 19. The study was represented on the UvA website as “Sexual behaviour among
adolescents”. It is possible that potential participants chose not to enroll in the study because they defined adolescence differently. To be able to generalize study results, differing definitions of adolescence need to be taken into consideration. When including participants from 18-25 years old, perhaps “young adults” might describe this age-range better than “adolescents”.
In addition, the questionnaires assessing social influence as well as the SRS were translated from English into Dutch, but due to time- and resource limitations they were not forward-back
translated, which is recommended for valid translations. It is recommended that follow-up studies first forward-back translate the questionnaires by translators with the to-be translated language as mother tongue, so that differences in results can be truly ascribed to characteristics of participants instead of inadequate translation (World Health Organization, 2016).
Although the sample size was sufficient according to the power analysis (a minimum of 82 participants was required) (Faul, Erdfelder, Lang, & Buchner, 2009), different analyses using sub-groups of the sample were performed. The sample sizes for comparing these sub-groups were small,
23 especially for downward counterfactual thoughts (see Table 3). These small sample sizes may increase type I errors, result in imprecise estimates of effects and do not correctly represent the adolescent (or young adult) population (Field, 2009). Also, the power that was calculated initially (1-β = .8) does not apply on these subgroups.
Table 3
Number of participants for comparing groups and interaction effects
Direction of counterfactual Upward Downward Total
Structure of counterfactual Additive 31 4 35
Subtractive 31 12 43
Total 62 16 78
Recommendations for future research
Although not represented by the results of the current study, a relationship between social norms and sexual risk taking is often confirmed in previous research (Azjen, 2002; van Empelen & Kok, 2008; Gebhardt, Kuyper & Greunsven, 2003; Rivis & Sheeran, 2003; Sheeran & Taylor, 1999), and therefore should be implemented in future interventions. It is important to bear in mind that
susceptibility to peer influence of social norms differs per person. For example, boys, especially those with a late pubertal development, seem to be more susceptible to social influences regarding sexual behaviour than girls (Widman, Choukas-Bradley, Helms, & Prinstein, 2016).
Due to the insufficient psychometric quality it is not possible to draw conclusions regarding the relationships between counterfactual thoughts and sexual risk taking. However, it is important to further investigate the role of counterfactual thoughts in adolescent sexual risk-taking, for it can be useful in decreasing sexual risk-taking. It is recommended that follow-up studies focus on
characteristics and processes of counterfactual thinking (CFdich), instead of the extent to which counterfactual thoughts are present (CFcon). However, a more extended method for measuring direction and structure of counterfactual thoughts should be developed. Firstly, the method should be more explicit in defining the tasks for the participants, since approximately half of the participants (51%) did not understand the question properly and wrote down an example of a counterfactual thought that was not sex-related. In addition, to better represent participants’ general way of thinking
24 during sexual situations, more items should be used, and processes that are present during
counterfactual thinking (i.e. causal inference and/or contrast- and assimilation effects) should be included. It may also be useful to instruct participants to write down the influence the counterfactual had on their sexual life, so that the way in which counterfactual thoughts operate to influence sexual risk-taking can be examined. Finally, in order to increase internal validity, ratings should be done by at least three researchers, using an improved version of the Coding scheme counterfactual thinking.
With the current research I hope to have indicated the importance of investigating the contribution of counterfactual thinking in young adults’ sexual risk taking. Hopefully future research will produce a valid and reliable method for measuring counterfactual thinking so that this relationship can be examined.
25 References
Ajzen, I. (2002, September). Constructing a TpB questionnaire: Conceptual and methodological considerations. Retrieved from
http://chuang.epage.au.edu.tw/ezfiles/168/1168/attach/20/pta_41176_7688352_57138.pdf Blanton, H., Vandeneijnden, R., Buunk, B., Gibbons, F., Gerrard, M., & Bakker, A. (2001). Accentuate
the negative: Social images in the prediction and promotion of condom use. Journal of Applied
Social Psychology, 31(2), 274-195.
Broek, van den, I. V., Verheij, R. A., Dijk, van, C. R., Koedijk, F. D., Sande, van der, M. A., & Bergen, J. E. (2010). Trends in sexually transmitted infections in the Netherlands, combining surveillance data grom gerenal practices and sexually transmitted infection centers. BMS Family Practice,
11(39), 1-8.
Centraal Bureau voor de Statistiek. (2012). Jaarboek onderwijs in cijfers 2012. Den Haag/Heerlen: Centraal Bureau voor de Statistiek.
Chein, J., Albert, D., O'Brien, L., Uckert, K., & Steinberg, L. (2012). Peers increase adolescent risk taking by enhancing activity in the brain's reward circuitry. Developmental Science, 14(2), f1-f10. Empelen, P., & Kok, G. (2008). Action-specific cognitions of planned and preparatory behaviors of
condome use among dutch adolescents. 37(4), 626-640.
Epstude, K., & Roese, N. J. (2008). The functional theory of counterfactual thinking. Personality and
Social Psychology Review, 12(2), 168-192.
Epstude, K., & Roese, N. J. (2011). When goal pursuit fails: The functions of counterfactual thoughts in intention formation. Social Psychology, 42(1), 19-27.
Faul, F., Erdfelder, E., Lang, A., & Buchner, A. (2009). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedival sciences. Behavior Research Methods, 39, 175-191.
Ferguson, R. M., Vanwesenbeeck, I., & Knijn, T. (2008). A matter of facts... and more: An exporatory analysis of the content of sexuality education in the Netherlands. Sex Education: Sexuality,
26 Field, A. (2009). Discovering statistics using spss (third edition ed.). Londen: Sage Publications.
Fivush, R., & Zaman, W. (2015). Debates: Identity development across the lifespan. In K. C. McLean,
The oxford handbook of identity development (p. 369).
Förster, J., Libermand, N., & Higgins, E. T. (2005). Accessibility from active and fulfilled goals. Journal of
Experimental Social Psychology, 41, 220-239.
Francis, J., Eccles, M., Johnston, M., Walker, A., Grimshaw, J., Foy, R., . . . Bonetti, D. (2004).
Constructing questionnaires based on the theory of planned behaviour. A manual for health services researchers. Newcastle upon Tuune: University of Newcastle.
Galinksy , A. D., & Moskowitz, G. B. (2000). Counterfactuals as behavioural primes: Priming the simulation heuristic and the consideration of alternatives. Journal of Experimental Social
Psychology, 36, 357-383.
Gebhardt, W., Kuyper, L., & Greunsven, G. (2003). Need for intimacy in relationships and motives for sex as determinants of adolescent condom use. Journal of Adolescen Health, 33(3), 154-164. Hitchcock, C., & Knobe, J. (2009). Cause and norm. The Journal of Philosophy, 106(11), 587-612. Jeffrey, S. A., Onay, S., & Larrick, R. (2010). Goal attainment as a resource: The cushion effect in risky
choice above a goal. Journal of Behavioral Decision Making, 23(2), 191-202.
Jonas, K. (2016). I should have done X instead of Y. The effect of counterfactual thinking in releven real
life contexts.
Kahneman, D., & Miller, D. T. (1986). Norm theory: Comparing reality to its alternatives. Psychological
Review, 93(2), 136-153.
Landman, J., & Manis, D. (1992). What might have been: Counterfactual thought concerning personal decisions. British Journal of Psychology, 83, 473-477.
Markman, K. D., & McMullen, M. N. (2003). A reflection and evaluation model of comparative thinking.
27 Markman, K. D., McMullen, M. N., & Elizaga, R. A. (2008). Counterfactual thinking, persistence, and
performance: A test of the relection and evaluation model. Journal of Experimental Social
Psychology, 44(1), 421-428.
Mouthaan, I., & Vlugt, van der, I. (2012). Seksuele gezondheidsbevordering van jongeren in Nederland.
Kind en Adolescent, 33(4), 279-288.
Oeffelen, van, A. A., Aar, van, F., Broek, van den, I., Coul, op de , E., Woestenberg, P., Heijne, J., . . . Benthem, van, B. (2015). Sexually transmitted infections, including HIV, in the Netherlands in
2014. Bilthoven: Rijksinstituut voor Volksgezondheid en Milieu RIVM.
Park, H. S., Klein, K., Smith, S., & Martell, D. (2009). Separating subjective norms, university descriptive and injunctive norms and U.A descriptive and injunctive norms for drinking behavior
intentions. Healt Communication, 24(8), 746-751.
Qualtrics. (2016, May). Qualtrics. Retrieved from http://www.qualtrics.com
Rijksinstituut voor Volkgsgezondheid en Milieu (RIVM). (2016). SOA, cijfers en context. RIVM. Rivis, A., & Sheeran, P. (2003). Descriptive norms as an additional predictor in the theory of planned
behaviour: A meta-analysis. Current Psychology, 121(1), 218-233.
Roese, N. J. (1997). Counterfactual thinking. Psychological Bulletin, 121(1), 133-148.
Roese, N. J., & Olson, K. M. (2014). What might have been: The social psychology of counterfactual
thinking. New York: Psychological Press.
Savitsky, K., Medvec, V., & Gilovich, T. (1997). Remembering and regretting: The zeigarnik effect and the cognitive availability of regrettable actoins and inactions. Psychological Bulletin, 25(3), 248-257.
Seta, C., Seta, J., Petrocelli, J., & McCormick, M. (2015). Even better than the real thing: Alternative outcome bias affects decision judgements and decision regret. Thinking & Reasoning, 21(4), 446-472.
28 Sheeran, P., & Taylor, S. (1999). Predicting intentions to use condoms: A meta-analysis and
comparison of the theories of reasoned action and planned behavior. Journal of Applied Social
Psychology, 29(8), 1624-1675.
Smallman, R., & Roese, N. J. (2009). Counterfactual thinking facilitates behavioural intentions. Journal
of Experimental Social Psychology, 45, 845-852.
Steinberg, L. (2010). A dual systems odel of adolescent risk-taking. Developmental Psychobiology, 52, 216-224.
Stroebe, W. (2011). Social Psychology and Health. New York: Open University Press.
Turchik, J., & Garske, J. (2008). Measurement of sexual risk-taking among college students. Archivves
of Sexual Behavior, 38(6), 936-948.
Widman, L., Choukas-Bradley, S., Helms, S., & Prinstein, M. (2016). Adolescent susceptibility to peer influence in sexual situations. Journal of Adolescent Health, 3223-329.
World Health Organization. (2016). Adolescent development. Retrieved from Maternal, newborn, child and adolescent health:
http://www.who.int/maternal_child_adolescent/topics/adolescence/dev/en/
World Health Organization. (2016). Process of translation and adaption of instruments. Retrieved from World Health Organization:
http://www.who.int/substance_abuse/research_tools/translation/en/
Zeigarnik, B. (1927). Das Behalten erledigter und unerledigter Handlungen [The memory of completed and uncompleted actions]. Psychologische Forschung, 9, 1-85.
29 Appendix A
Information letter and declaration of consent
Beste deelnemer,
Dankjewel voor je deelname aan dit onderzoek! Dit onderzoek is onderdeel van de these die ik schrijf voor de opleiding psychologie van gezondheidsgedrag aan de Universiteit van Amsterdam. Hiervoor zijn we op zoek naar mensen met een leeftijd van 18 t/m 25 jaar die in de afgelopen zes maanden seksueel actief zijn geweest in de vorm van
geslachtsgemeenschap of andere seksuele handelingen (orale stimulatie, stimulatie met de hand etc.). Indien je hier niet onder valt, kun je helaas niet aan het onderzoek meedoen. Deelname aan dit onderzoek duurt ongeveer 20-30 minuten en is volledig anoniem. Op de volgende pagina krijg je meer informatie over het onderzoek.
Informatie
Doel van het onderzoek
Het doel van dit onderzoek is om een duidelijker beeld te krijgen van seksuele activiteit onder adolescenten. Voor dit onderzoek zul je dus een aantal vragen invullen die gaan over je eigen seksuele activiteit en dat van je vrienden. Het invullen van deze vragen duurt ongeveer 20-30 minuten.
Anonimiteit
De ingevulde antwoorden worden volledig anoniem verwerkt en kunnen niet aan jou persoonlijk gekoppeld worden. De antwoorden op de vragen zullen alleen gebruikt worden voor het huidige onderzoek en zullen vertrouwelijk worden behandeld.
Verzekering
Omdat dit onderzoek geen risico’s voor je gezondheid of veiligheid met zich meebrengt, gelden de voorwaarden van de reguliere aansprakelijkheidsverzekering van de UvA.
Vrijwilligheid
Als u nu besluit af te zien van deelname aan dit experiment, zal dit op geen enkele wijze gevolgen voor u hebben. Als u tijdens het onderzoek zelf besluit uw medewerking te staken, zal dat eveneens op geen enkele wijze gevolg voor u hebben. Tevens kunt u 24 uur na dit onderzoek, alsnog uw toestemming om gebruik te maken van uw gegevens intrekken. U bent vrij om dit te doen zonder opgave van redenen. Mocht u uw medewerking staken, of achteraf, zij het binnen 24 uur, uw toestemming intrekken, dan zullen uw gegevens worden verwijderd uit onze bestanden en vernietigd.
Alvast bedankt voor je deelname en succes met het invullen van de vragenlijst.
Justine Schwanen (schwanen.justine@gmail.com) Masterstudent Psychologie van Gezondheidsgedrag Universiteit van Amsterdam
Toestemmingsverklaring
‘Ik verklaar hierbij op voor mij duidelijke wijze te zijn ingelicht over de aard en methode van het onderzoek. Mijn vragen zijn naar tevredenheid beantwoord’.
Vrijwillige deelname ‘Ik stem geheel vrijwillig in met deelname aan dit onderzoek. Ik behoud daarbij het recht deze instemming weer in te trekken zonder dat ik daarvoor een reden behoef op te geven en besef dat ik op elk moment mag stoppen met het experiment. Indien mijn onderzoeksresultaten gebruikt zullen worden in wetenschappelijke publicaties, dan wel op een andere manier openbaar worden gemaakt, zal dit volledig geanonimiseerd gebeuren. Mijn persoonsgegevens zullen niet door derden worden ingezien zonder mijn uitdrukkelijke toestemming’.
Gegevens voor informatie ‘Als ik nog verdere informatie over het onderzoek zou willen krijgen, nu of in de toekomst, kan ik me wenden tot de onderzoeker, Justine Schwanen (schwanen.justine@gmail.com), of haar begeleider, de heer dr. Kai J. Jonas (k.j.jonas@uva.nl; Nieuwe Achtergracht 129B, kamer G2.29; Postadres: Postbus 15900, 1001NK Amsterdam; t: +31 20 525 -6888 of -6890 (secr.); f: +31 20 525 -6585). Voor eventuele klachten over dit onderzoek kunt u zich wenden tot het lid van de Commissie Ethiek van de afdeling Psychologie van de Universiteit van Amsterdam, de heer dr. Mark Rotteveel (e-mail: m.rotteveel@uva.nl)’.
• Akkoord, ik wil deelnemen aan het onderzoek • Niet akkoord, ik wil niet deelnemen aan het onderzoek
30 Appendix B
Survey social influence
Er worden nu een aantal stellingen gegeven. Vul in hoeverre de stelling op jou van toepassing is. Bij sommige stellingen ontbreekt er een woord (zie "_____"). Geef dan aan in hoeverre je denkt dat het gegeven antwoord op de lijn past.
1. Vrienden vinden dat ik
wel :___:___:___:___:___:___:___: niet
condooms moet gebruiken iedere keer dat ik geslachtsgemeenschap heb. 2. Het wordt van mij verwacht dat ik condooms gebruik iedere keer dat ik geslachtsgemeenschap heb
heel onwaarschijnlijk :___:___:___:___:___:___:___: heel waarschijnlijk 3. De mensen in mijn leven waarvan ik de mening belangrijk vind keuren het af :___:___:___:___:___:___:___: keuren het goed
dat ik condooms gebruik iedere keer dat ik geslachtsgemeenschap heb.
4. De meeste mensen die belangrijk voor mij zijn gebruiken condooms iedere keer dat zij geslachtsgemeenschap hebben.
helemaal waar :___:___:___:___:___: ___:___: helemaal niet waar 5. The mensen in mijn leven waarvan ik de mening belangrijk vind gebruiken geen :___:___:___:___:___: ___:___: gebruiken wel
v.
6. Wat leeftijdsgenoten vinden dat ik moet doen is voor mij belangrijk helemaal niet :___:___:___:___:___: ___:___: heel erg
7. De goedkeuring van vrienden over mijn condoomgebruik is belangrijk voor mij Helemaal mee eens :___:___:___:___:___: ___:___: helemaal niet mee eens
8. Handelen in overeenstemming met wat mijn vrienden doen is helemaal niet :___:___:___:___:___: ___:___: heel erg
31 Appendix C
Survey counterfactual thoughts
We willen nu graag weten wat je over je eigen leven denkt en of en hoe je terugkijkt naar het
verleden. Daarover hebben we een aantal stellingen. We willen graag van jou weten of je het hiermee eens of oneens bent.
Zeer mee oneens (1) Mee oneens (2) Enigszin s mee oneens (3) Neutraa l (4) Enigszins mee eens (5) Mee
eens (6) Zeer mee eens (6) 1. Wat er in het
verleden is gebeurd, maakt me nu niet veel meer uit. o o o o o o o 2. Ik richt me vooral op dingen die er nu gebeuren. o o o o o o o 3. Ik richt me vooral op de toekomst. o o o o o o o
4. Ik denk er vaak over na wat ik in het verleden anders had kunnen doen.
o o o o o o o
5. Er zijn veel
gebeurtenissen in mijn verleden waarvan ik zou willen dat ze anders waren verlopen of een andere uitkomst hadden gehad.
o o o o o o o
6. Als ik een doel wil bereiken, denk ik er over na wat ik in het verleden anders had kunnen doen.
o o o o o o o
7. Ik denk dat wij ons gedrag steeds moeten aanpassen om onze doelen te bereiken.
32 We willen je nu vragen een seksueel-gerelateerde gebeurtenis uit jouw leven te omschrijven waarbij gedachten betrokken waren over hoe gebeurtenissen anders af had kunnen lopen. Het is dus denken over "hoe het had kunnen zijn". Zulke gedachten worden counterfactuele gedachten genoemd. Algemene voorbeelden van counterfactuele gedachten zijn:
"Als ik harder had gestudeerd, had ik het tentamen wel gehaald".
"Als ik mijn vlucht niet had gemist, had ik het vliegtuigongeluk meegemaakt".
We willen je vragen het door jou uitgekozen scenario zo gedetailleerd mogelijk te beschrijven. 8. Ik denk dat wij vooral
door het lot zijn gedreven.
o o o o o o o
9. Er zijn andere krachten dan ikzelf die bepalen wat er met mij gebeurt.
o o o o o o o
10. Ik denk vaak dat dingen in het verleden beter hadden kunnen gaan.
o o o o o o o
11. Ik denk er vaak over na wat ik in het
verleden had moeten doen.
o o o o o o o
12. Ik denk er vaak over na wat ik in het
verleden niet had moeten doen.
33 Appendix D
Coding scheme counterfactual thinking 1. Participant number (column A).
2. Direction of counterfactual (column B, C, and D).
This refers to whether the counterfactual specifies an alternative that is better (upward) of worse (downward) than what actually happened. Use column B for the total number of counterfactuals, then divide the total number into upward counterfactuals (column C), and downward counterfactuals (column D).
Examples:
“If only we had stayed together, I’d be much happier” (upward). “Good thing they both weren’t killed in the accident” (downward). “She almost made it to the airport on time” (upward).
“If I had picked door number two, I qould have lost all my money” (downward). 3. Structure of counterfactual (column E, and F).
This refers to whether the counterfactual adds something new that didn’t actually happen in reality (an additive counterfactual) or removes something that did happen (a subtractive counterfactual). Use columns E and F for this coding. E stands for number of additive counterfactuals and F stands for number of subtractive counterfactuals.
Examples:
“If only we had stayed together, I’d be much happier” (additive). “I wish I hadn’t yelled at her” (subtractive).
“Is only I had studied harder, I might have passed” (additive).
“He shouldn’t have bought that ring, it’s too expensive” (subtractive) 4. Comments.
If you are uncertain about certain statements, make some note in this column. Also if you notice that someone hasn’t worked on the task in an appropriate way.
34 Appendix E
Survey sexual risk taking
Instructions: Please read the following statements and record the number that is true for you over the past 6 months for each question on the blank. If you do not know for sure how many times a behavior took place, try to estimate the number as close as you can. Thinking about the average number of times the behavior happened per week or per month might make it easier to estimate an accurate number, especially if the behavior happened fairly regularly. If you’ve had multiple partners, try to think about how long you were with each partner, the number of sexual encounters you had with each, and try to get an accurate estimate of the total number of each behavior. If the question does not apply to you or you have never engaged in the behavior in the question, put a ‘‘0’’ on the blank. Please do not leave items blank. Remember that in the following questions ‘‘sex’’ includes oral, anal, and vaginal sex and that ‘‘sexual behavior’’ includes passionate kissing, making out, fondling, petting, oral-to-anal stimulation, and hand-to-genital stimulation. Refer to the Glossary for any words you are not sure about. Please consider only the last 6 months when answering and please be honest.
In the past six months:
1. How many partners have you engaged in sexual behaviour with but not had sex with? 2. How many times have you left a social event with someone you just met?
3. How many times have you ‘‘hooked up’’ but not had sex with someone you didn’t know or didn’t know well?
4. How many times have you gone out to bars/parties/social events with the intent of ‘‘hooking up’’ and engaging in sexual behavior but not having sex with someone?
5. How many times have you gone out to bars/parties/social events with the intent of ‘‘hooking up’’ and having sex with someone?
6. How many times have you had an unexpected and unanticipated sexual experience?
7. How many times have you had a sexual encounter you engaged in willingly but later regretted? For the next set of questions, follow the same direction as before. However, for questions 8–23, if you have never had sex (oral, anal or vaginal), please put a ‘‘0’’ on each blank.
8. How many partners have you had sex with?
9. How many times have you had vaginal intercourse without a latex or polyurethane condom? Note: Include times when you have used a lambskin or membrane condom.
10. How many times have you had vaginal intercourse without protection against pregnancy? 11. How many times have you given or received fellatio (oral sex on a man) without a condom?
35 12. How many times have you given or received cunnilingus (oral sex on a woman) without a dental dam or ‘‘adequate protection’’ (please see definition of dental dam for what is considered adequate protection)?
