On Creepiness 1 Running Head: ON CREEPINESS (On the Nature of) CREEPINESS *Francis T. McAndrew and Sara S. Koehnke Knox College Basis of a poster presented at the annual meeting of the Society for Personality and Social Psychology, New Orleans, January, 2013. *Correspondence concerning this article should be addressed to Francis T. McAndrew, Department of Psychology, Knox College, 2 East South Street, Galesburg, IL 614014999. Phone: +1-309-341-7525; E-mail: fmcandre@knox.edu On Creepiness 2 Abstract Given how frequently the concept of “creepiness” is invoked in everyday life to describe the relationships and encounters that we have with others, it is surprising that it has not been studied in a formal way. This study attempted to uncover the cues that are used to label someone as “Creepy” and to identify the basic elements of creepiness. An international sample of 1,341 individuals (1029 females, 312 males; ages 18-77, M(SD) = 28.97 (11.34)) responded to an online survey about creepiness. The results revealed that males are perceived as creepier than females and that females are more likely to associate sexual threat with creepiness. Nonverbal behaviors and characteristics associated with unpredictability are also predictors of creepiness, as are some occupations and hobbies. The results are consistent with the hypothesis that being “creeped out” is an evolved adaptive emotional response to ambiguity about the presence of threat that enables us to maintain vigilance during times of uncertainty. Keywords: creepiness, nonverbal behavior, emotion, person perception, threat perception, evolutionary psychology On Creepiness 3 On the Nature of Creepiness Most people have probably used the concept of “creepiness” to describe reactions to the behaviors of individuals that they have encountered in real life or in the movies, and an initial perception of an individual as “creepy” undoubtedly creates an impediment to comfortable future social interactions or relationships with that person. The “creepy” psychological reaction is both unpleasant and confusing, and it may be accompanied by physical symptoms such as feeling cold or chilly (Leander, Chartrand, & Bargh 2012). Although our evidence is completely anecdotal, we believe that creepiness is a universal human experience that is a by-product of evolved human psychology. In conversations with people from a wide array of cultural backgrounds, the individuals we questioned instantly understood what we were asking about and were largely in agreement about the things that make some individuals come across as creepy. Thus, given its pervasiveness in everyday human social life, it is very surprising that no one has studied it in a scientific way. The only research that is even close is the aforementioned study by Leander and colleagues who discovered that interacting with individuals displaying inappropriate levels of nonverbal mimicry during social interaction produces an actual physical sensation of feeling cold. Their explanation for the phenomenon is that such nonnormative nonverbal behaviors signal a social mismatch and put us on our guard against a cold and potentially untrustworthy interaction partner. The fact that social exclusion and other types of social threat produce similar feelings of “getting the chills” is consistent with the idea that our “creepiness detector” is in fact a defense against some sort of threat (Knight & Borden 1979; Zhong & Leonardelli 2008). On Creepiness 4 But what exactly is it that our creepiness detector is warning us about? It cannot just be a clear warning of physical or social harm. A mugger who points a gun in your face and demands money is certainly threatening and terrifying, and a rival who threatens to destroy your reputation by revealing secret information about you fills you with dread. Yet, most people would probably not use the word “creepy” to describe these situations. It is our belief that creepiness is anxiety aroused by the ambiguity of whether there is something to fear or not and/or by the ambiguity of the precise nature of the threat (e.g., sexual, physical violence, contamination, etc) that might be present. Such uncertainty results in a paralysis as to how one should respond. In the mugging and reputationsavaging situations, there is no ambiguity about the presence or nature of threat. It may be that it is only when we are confronted with uncertainty about threat that we get “creeped out,” which could be adaptive if it facilitates our ability to maintain vigilance during periods of uncertainty. Creepiness may be related to the “agency-detection” mechanisms proposed by evolutionary psychologists (Atran 2002; Barrett 2005). To oversimplify a bit, these mechanisms have evolved as adaptations to protect us from harm at the hands of predators and enemies. If you are walking down a dark city street and hear the sound of something moving in the dark alley to your right, you will respond with a heightened level of arousal and sharply focused attention and behave as if there is a willful “agent” present who is about to do you harm. If it turns out that it is just the wind or a stray cat, you have lost little by over-reacting, but if you fail to activate the alarm response when there is in fact a threat present, the cost of your miscalculation may be quite high. Thus, humans have evolved to err on the side of detecting threats in such ambiguous situations. On Creepiness 5 Consequently, people become uneasy in environments that are dark and/or offer a lot of hiding places for potential predators and also lack clear, unobstructed views of the landscape. These environmental qualities have been called “prospect” and refuge” by the British geographer Jay Appleton (1975, 1984). Fear of crime and a pervasive sense of unease are experienced in environments with less than optimal combinations of prospect and refuge (Fisher & Nasar 1992). So, it is not the clear presence of danger that makes us feel creepy, but the uncertainty of whether danger is present or not. It is the goal of this study to uncover the cues that we use to label other people as creepy and to identify the building blocks of this thing we call “creepiness.” Szczurek, Monin, & Gross (2012) have found that we wish to keep greater social distance between ourselves and individuals who display inappropriate or non-normative expression of emotion, and Leander et al (2012) indicated that inappropriate nonverbal behaviors may serve as creepiness cues, but surely there must be other things. Are particular physical characteristics or types of people considered creepy? Do certain occupations or hobbies also cause us to perceive others as creepy? Is creepiness a characteristic of humans alone, or can places, things, and animals be thought of as creepy too? At this time, we simply do not know the answers to these questions. Since there is no previous body of research and theory to build upon directly, this study is unavoidably exploratory in nature. However, there are a few hypotheses that can be tested. 1) If creepiness communicates potential threat, males should be more likely to be perceived as creepy than females, since males are simply more violent and physically threatening to more people (McAndrew 2009). On Creepiness 6 2) Related to the first prediction, females should be more likely than males to perceive some sort of sexual threat from a creepy person. 3) Occupations that signal a fascination with threatening stimuli (e.g, death or “non-normative” sex) may attract individuals that would be comfortable in such a work environment. Hence, some occupations should be perceived as creepier than other occupations. 4) Since we hypothesize that creepiness is a function of uncertainty about threat, non-normative nonverbal behaviors and behaviors or characteristics associated with unpredictability will be positively associated with perceptions of creepiness. Method Participants A snowball sampling technique was employed to recruit participants. People were recruited through invitations to Facebook events that were created by the researchers, through campus-wide emails distributed to students, faculty, and staff at a liberal arts college in the American Midwest, and through the “Social Psychology Network” website. Volunteers were encouraged to forward the link to the online survey to their friends and acquaintances. A brief description of the study and a link to the survey were posted on the invitation page. This resulted in a final sample of 1,341 individuals (1,029 females, 312 males) ranging in age from 18 to 77 with a mean age of 28.97 (SD = 11.34). We did not ask participants to report their country of origin, but in an unrelated study using an On Creepiness 7 identical recruitment strategy, respondents from 54 different nations were acquired. Thus, although our sample was primarily American, we are confident that there was significant international representation. Participants had to check a box confirming that they were at least 18 years of age before they could access the survey. Procedure and Materials An online survey was created using Google Documents. Participants began the survey by reporting their sex and age and by responding to a forced choice question that asked them to choose whether they thought that a creepy person was more likely to be a male or a female. They then proceeded to a survey divided into four sections. In the first section of the survey, participants considered the following scenario: Imagine a close friend of yours whose judgment you trust. Now imagine that this friend tells you that she or he just met someone for the first time and tells you that the person was “creepy.” After reading this scenario, the participants rated the likelihood that the creepy person exhibited 44 different behaviors (e.g., the person never looked your friend in the eye) or physical characteristics (e.g., this person had visible tattoos) on a “1” (very unlikely) to “5” (very likely) scale. In the second section of the survey, participants rated the creepiness of 21 different occupations on a “1” (not at all creepy) to “5” (very creepy) scale. In the third section of the survey, participants simply listed two hobbies (via free response) that they thought were creepy. On Creepiness 8 In the fourth and final section of the survey, participants expressed their degree of agreement with 15 statements about the nature of creepy people on a “1” (strongly disagree) to “5” (strongly agree) scale. Examples of these statements include the following: “I am uncomfortable because I cannot predict how he or she will behave.” “I think that the person has a sexual interest in me.” “People are creepier online than when I meet them face-to-face.” There was one final question on the survey. Participants chose a response of “yes,” “no,” or “unsure” to the question “Do most creepy people know that they are creepy?” Results Tests of Hypotheses. The first prediction was that creepy individuals would be expected to be males more often than females. This prediction was assessed directly via the question that asked people to choose whether a creepy person was more likely to be a male or a female. 95.3% of our respondents thought that creepy people were much more likely to be males than females, a finding that was highly significant, Χ2 (1, N = 1,341) = 1100.84, p. < .00001. This perception was equally likely to be held by male participants (95.5% vs. 4.5%) and female participants (95.2% vs. 4.8%). Thus, our first prediction was supported: males are creepier than females. The second prediction was that females are more likely to perceive a sexual threat from a creepy person than are males. This hypothesis was tested with a-priori t tests comparing male and female responses to two items: The degree to which steering a On Creepiness 9 conversation toward sex was perceived as a probable characteristic of a creepy person and the degree to which the respondent agreed with the statement that the creepy person “has a sexual interest in me.” The prediction was supported by both of these items. Females were more likely than males to think that steering a conversation toward sex was characteristic of a creepy person, t (1339) = 5.46, p. < .0001, Means (SD) = 4.23 (.930) vs. 3.90 (1.03), and they were also more likely to think that the creepy person had a sexual interest in them, t (1339) = 7.63, p. < .0001, Means (SD) = 3.51 (1.02) vs. 2.99 (1.15). The third prediction was that occupations would differ in their level of creepiness according to how threatening or strange the “subject matter” of the occupation is. The means and standard deviations of the creepiness ratings for the 21 stimulus occupations are displayed in Table 1. A repeated measured ANOVA using a Greenhouse-Geisser adjustment revealed that the differences in how occupations were rated was highly significant, F (13.636, 18271.956) = 734.29, p. < .00001, ηp2 = .354. A Tukey test (HSD = .01) indicated that all of the occupations except two (construction workers and computer software engineers) were significantly different from each other. However, one-sample t tests revealed that only four occupations were judged to be significantly higher than the neutral value of “3” on the creepiness rating scale: Clowns, t (1340) = 21.14, p. < .0001, Taxidermists, t (1340) = 21.46, p. < .0001, Sex Shop Owners, t (1340) = 9.09, p. < .0001, and Funeral Directors, t (1340) = 6.58, p. < .0001. Therefore, it appears that occupations associated with death (taxidermy and funeral directors) or reflective of a fascination with sex (sex shop owners) are perceived as creepy; clowns were the creepiest of all. On Creepiness 10 The fourth prediction was that things that make a person unpredictable also predict creepiness. One item among the ratings of creepy individuals (“I am uncomfortable because I cannot predict how he or she will behave”) and one item among the items assessing beliefs about creepy people (“Even though someone may seem creepy, I usually think that I understand his or her intentions”) allowed a direct test of this prediction. A one-sample t test revealed that the mean rating for being uncomfortable because of an inability to predict behavior (4.33 on a 5 point scale, SD = .815) was significantly above the neutral point of 3.0, t (1340) = 59.96, p. < .00001 and therefore highly likely to be characteristic of creepy individuals. The mean for the item about understanding the intentions of a creepy person (2.96 on a 5 point scale, SD = .966) was just below and not significantly different from the neutral point of “3”, meaning that believing that one understands the intentions of an individual makes them less creepy, t (1340) = 1.67, p. = 0.096. Collectively, the results of the analyses of these two items indicate that unpredictability is indeed an important component of creepy behavior. Data Reduction and Exploratory Analyses. The many items in our survey afford ample opportunities for exploration of the elements of creepiness. Our first step in this direction was to combine items that seemed to be measuring the same thing within the two longest sections of our questionnaire. The first section contained 44 items assessing the likelihood that a creepy person described by one’s trusted friend would display a particular behavior or possess a particular physical characteristic. In an attempt to reduce the number of “dependent” variables to be analyzed, these 44 items were subjected to a principal components factor analysis using varimax rotation. Only items with factor loadings exceeding .50 on a common factor would be combined into a single On Creepiness 11 composite variable for further analysis. The factor analysis was able to identify only one factor that connected multiple variables. This factor included 15 of the 44 items, all of which reflected a nonverbal behavior or physical characteristic of creepy people. A new variable called Appearance/NVB was calculated by computing a mean based upon the scores of each individual on these 15 items. The 15 items that comprised this new variable are as follows. The factor loading for each item is given in parentheses. The person stood too close to your friend (.509) The person had greasy hair (.582) The person had a peculiar smile (.546) The person had bulging eyes (.563) The person had long fingers (.503) The person had unkempt hair (.609) The person had very pale skin (.566) The person had bags under his or her eyes (.599) The person was dressed oddly (.601) The person licked his or her lips frequently (.580) The person was wearing dirty clothes (.571) The person laughed at unpredictable times (.546) The person made it nearly impossible for your friend to leave the conversation without appearing rude (.500) The person relentlessly steered the conversation toward one topic (.519) This new composite Appearance/NVB variable along with the remaining 29 items from the first portion of the questionnaire were analyzed via one-sample t tests to determine which of these characteristics was significantly above the neutral point of “3,” On Creepiness 12 and therefore very likely to be a characteristic of a creepy person. The means, standard deviations, and results of the t tests are presented in Table 2. Given the large number of comparisons that were made and the exploratory nature of these comparisons, a Bonferroni correction suggested that a more conservative p-value of .002 should be the guide for determining which differences are least likely to have been due to chance. An examination of Table 2 reveals that the following elements were thought to be very likely to be found in a creepy person: The appearance and nonverbal behavior items in the composite variable (Appearance/NVB), being of the opposite sex (probably due to the predominantly female sample in our study), being extremely thin, not looking the interaction partner in the eye, asking to take a picture of the interaction partner, watching people before interacting with them, asking about details of one’s personal life, having a mental illness, talking about his/her own personal life, displaying too much or too little emotion, being older, and steering the conversation toward sex. Similarly, the section of the questionnaire consisting of 15 items that reflected beliefs about the nature of creepy people was subjected to a principal components factor analysis using varimax rotation. Only items with factor loadings exceeding .50 on a common factor would be combined into a single composite variable for further analysis. The analysis yielded four factors on which at least two items loaded. The first factor tapped into how fearful or anxious the person felt while interacting with a creepy person, and it included the following items, with factor loadings in parentheses. Each statement began with the expression “When I meet someone that seems creepy . . . I am sure that the person intends to harm me (.691) I am uncomfortable because I cannot predict how he or she will behave (.718) On Creepiness 13 I feel anxious (.756) I believe that he or she is intentionally hiding something from me (.509) The second factor reflected how intimately involved one would be with a creepy person, and it consisted of two items: People are less creepy if I know I won’t have to speak to them ever again (.553) People are creepier when I meet them online compared to face-to-face (.526) The third factor measured the extent to which creepiness is an inherent part of the individual, and it consisted of two items: Some people can do the exact same behavior as someone else and one person can be perceived as creepy while the other person is not (.629) Behaviors often admired in “bad guys” in movies and TV shows are actually really creepy if done in real life (.586) The fourth factor reflected the extent to which people willfully deviate from social norms, and it consisted of two items: When I meet someone who seems creepy, I expect him or her to follow the usual rules for socially acceptable behavior (.634) People choose to act in a creepy manner (.532) New composite variables labeled “fearfulness,” “proximity,” “individual creepiness,” and “non-normativity,” were calculated by computing a mean of the items that loaded on each factor. These four composite variables along with the remaining 5 items from the last portion of the questionnaire were analyzed via one-sample t tests to determine which of these characteristics was significantly different from neutral point of “3,” and therefore strongly believed to be characteristics of a creepy person. The means, standard deviations, and results of the t tests are presented in Table 3. Given the large number of On Creepiness 14 comparisons that were made and the exploratory nature of these comparisons, a Bonferroni correction suggested that a more conservative p-value of .005 should be the guide for determining which differences are least likely to have been due to chance. An examination of Table 3 reveals that the following things were believed to be true of a creepy person: They make us fear fearful/anxious (composite fearfulness variable) Creepiness resides in the individual more than in his/her behavior (composite individual creepiness variable) We think they may have a sexual interest in us They are creepy when they exhibit multiple “symptoms” of creepiness rather than just one The expected intimacy and frequency of interaction with the person moderates perceptions of creepiness Creepy people are unable to change, but they do not necessarily have bad intentions People who follow social rules of behavior are not perceived as creepy There was also one final item in which participants chose among “yes,” “no,” and “unsure” in response to the question “Do most creepy people know that they are creepy?” The responses were 115 “yes” (8.6%), 797 “no” (59.4%), and 429 “unsure” (32%), indicating that our participants did not believe that most creepy people know that they are creepy, Χ2 (2, N = 1,341) = 401.02.84, p. < .0001. Correlations with Age. There were many significant correlations between the age of the participant and his/her responses to the items in the survey. Given the exploratory nature and large number of these correlation coefficients, we will not discuss them in any detail here. However, the general finding of interest was that older people On Creepiness 15 seemed to be less alarmed by creepy people than are younger people, being less likely to perceive sexual threat, r (1341) = -0.21, p. < .0001, or intended harm, r (1341) = -0.11, p. < .0001. They also expressed less anxiety at the prospect of interacting with a creepy person, r (1341) = -0.13, p. < .0001. Creepiness of Hobbies. Just for fun, we asked our participants to list two hobbies that they thought of as creepy. Easily, the most frequently mentioned creepy hobbies involved collecting things (listed by 341 of our participants). Collecting dolls, insects, reptiles, or body parts such as teeth, bones, or fingernails was considered especially creepy. The second most frequently mentioned creepy hobby (listed by 108 participants) involved some variation of “watching.” Watching, following, or taking pictures of people (especially children) was thought to be creepy by many of our participants, and bird watchers were considered creepy by many as well. A fascination with pornography or exotic sexual activity and taxidermy were also frequently mentioned. Discussion Everything that we found in this study is consistent with the notion that the perception of creepiness is a response to the ambiguity of threat. Males are more physically threatening to people of both sexes than are females (McAndrew 2009), and they were more likely to be perceived as creepy by males and females alike. The link made by females between sexual threat and creepiness is also consistent with the fact that females are simply at greater risk of sexual assault and have potentially greater costs associated with it than males. We are placed on our guard by people who touch us or exhibit non-normative nonverbal behavior, or those who are drawn to occupations that On Creepiness 16 may reflect a fascination with death or unusual sexual behavior. People who have hobbies that involve collecting things that we are predisposed as a species to fear such as spiders and snakes (Öhman, Flykt, & Esteves 2001; Rakison 2009) or things that can only be acquired after something has died (e.g., skulls or bodies to be stuffed) seem creepy to us as well. We are also wary of individuals who have a preoccupation with monitoring the behavior of others. While they may not be overtly threatening, individuals who display unusual nonverbal behaviors (Leander et al 2012), odd emotional behavior (Szczurek et al 2012), or highly distinctive physical characteristics are outside of the norm, and by definition unpredictable. This activates our “creepiness detector” and increases our vigilance as we try to discern if there is in fact something to fear or not from the person in question. Interestingly, our results indicate that we do not necessarily assume ill intentions from people who are creepy, although we may still worry that they are dangerous. Most of our subjects believed that creepy people cannot change, and only a small minority of our subjects (8.6%) believed that creepy people are aware that they are creepy. As always, after the fact we can think of things that should have been done differently. We are assuming significant international representation in our sample, but in hindsight it would have been useful to have data on the nationalities of our participants so that cross-cultural comparisons could have been made. We also wish that we had specifically identified the hypothetical friend interacting with a creepy person in the first part of our questionnaire as a same-sex friend, as this would have allowed a more nuanced examination of sex differences in the perception of creepy individuals. It looks as if most of our participants were thinking of the scenario in this way, but there is no On Creepiness 17 way that we can be sure. It might also have been enlightening to ask individuals to rate themselves on creepiness on the chance that this may have been a good predictor of something else. Finally, we must also acknowledge the limitations of self-selection that occur in any study in which people voluntarily spend time filling out an online survey, especially when the sample is drawn primarily from individuals who were recruited by way of Facebook pages. In spite of these limitations, we believe that our research is a good first step in looking at a topic that has not been studied before, and we see nothing in our data to discourage us from pursuing the idea that creepiness is an adaptive human response to the ambiguity of threat from others. In other words, creepy individuals provide us with the social equivalent of the less than optimal “prospect and refuge” found in the physical settings that make us uneasy (Fisher & Nasar 1992). Consequently, we would like to extend this line of research in future studies by looking at responses to creepy places (e.g., haunted houses) as well as to creepy people to determine if our creepiness detectors are attuned specifically to social interaction, or if they function in response to the ambiguity of threat in general. On Creepiness 18 References Appleton, Jay. 1975. “The experience of landscape.” London: John Wiley & Sons. Appleton, Jay. 1984. “Prospects and refuges revisited.” Landscape Journal, 8: 91-103. Atran, Scott. 2002. “In gods we trust: The evolutionary landscape of religion.” New York: Oxford University Press. Barrett, H. Clark. 2005. “Adaptations to predators and prey.” In D. M. Buss (Ed.), The handbook of evolutionary psychology (pp. 200-223). Hoboken, NJ: John Wiley & Sons. Fisher, Bonnie S., & Nasar, Jack L. 1992. “Fear of crime in relation to three exterior site features: Prospect, refuge, and escape.” Environment and Behavior, 24: 35-65. Knight, Martha L., & Borden, Richard J. 1979. “Autonomic and affective reactions of high and low socially-anxious individuals awaiting public performance.” Psychophysiology, 16: 209-213. Leander, N. Pontus, Chartrand, Tanya L., & Bargh, John A. 2012. “You give me the chills: Embodied reactions to inappropriate amounts of behavioral mimicry.” Psychological Science, 23: 772-779. DOI: 10.1177/0956797611434535 McAndrew, Francis T. 2009. “The interacting roles of testosterone and challenges to status in human male aggression.” Aggression and Violent Behavior, 14: 330335. DOI: 10.1010/j.avb.2009.04.006 Öhman, Arne, Flykt, Anders, & Esteves, Francisco. 2001. “Emotion drives attention: Detecting the snake in the grass.” Journal of Experimental Psychology: General, 130: 466-478. Rakison, David H. 2009. “Does women’s greater fear of spiders and snakes originate in On Creepiness 19 infancy?” Evolution and Human Behavior, 3: 438-444. Szczurek, Lauren, Monin, Benoît, & Gross, James J. 2012. “The Stranger effect: The rejection of affective deviants.” Psychological Science, 23: available online preprint, doi: 10.1177/0956797612445314. Zhong, Chen-Bo & Leonardelli, Geoffrey J. 2008. “Cold and lonely: Does social exclusion literally feel cold?” Psychological Science, 19: 838-842. On Creepiness 20 Table 1 Creepiness Ratings of Occupations Occupation Clown Taxidermist Sex Shop Owner Funeral Director Taxi Driver Unemployed Clergy Janitor Garbage Collector Guard Writer Actor Construction Worker* Computer Software Engineer* Cafeteria Worker Financial Adviser Doctor/Physician College Professor Farmer Teacher Meteorologist Mean 3.71 3.69 3.32 3.22 2.86 2.83 2.57 2.51 2.25 2.18 2.14 2.13 2.09 2.09 2.08 1.78 1.77 1.67 1.65 1.57 1.53 SD 1.24 1.19 1.30 1.23 1.19 1.29 1.28 1.19 1.12 1.08 1.08 1.02 1.09 1.11 1.06 0.98 0.96 0.86 0.90 0.82 0.83 Note: Occupations marked with an asterisk are not significantly different from each other (Tukey HSD = .01). Ratings were made on a “1” (not very creepy) to “5” (Very creepy) scale. On Creepiness 21 Table 2 One Sample t-test Results for Ratings of Probable Characteristics of a Hypothetical Creepy Person Interacting with Friend of Participant Variable/Questionnaire Item Mean (SD) Appearance/NVB (Composite) 3.87 (0.54) Talked a lot about clothes 1.91 (0.91) Extremely thin 3.18 (0.90) Dressed too formally for situation 2.64 (1.13) Never looked friend in the eye 3.74 (1.23) Opposite sex of friend 4.01 (1.09) Muscular 2.41 (0.93) Asked to take picture of friend 4.11 (1.03) Watched friend before interacting 4.55 (0.67) Asked for personal details of friend’s family 4.09 (0.94) Tall 3.08 (0.91) Greasy Hair 3.90 (0.91) Same sex as friend 2.25 (0.91) Smiled a lot 2.82 (1.07) Had mental illness 3.45 (1.06) Talked a lot about personal life 3.41 (1.15) Touched friend frequently 4.24 (0.92) Was a child 1.67 (0.89) Significantly older than friend 3.72 (1.03) Displayed a lot of emotion 3.15 (1.12) Had facial hair 2.89 (0.97) Crossed arms 2.61 (0.97) Obese 2.63 (0.93) Steered conversation toward sex 4.16 (0.96) Dressed too casually for situation 2.89 (1.04) Fashionably Dressed 1.92 (0.92) Frequently played with hair 2.57 (0.96) Wore revealing clothing 2.57 (0.96) Showed little emotional expression 3.62 (1.07) Nodded frequently 2.82 (0.98) t value 59.69 44.13 7.45 11.73 22.20 33.99 23.18 39.55 84.66 42.70 3.02 36.43 30.35 6.26 15.57 13.03 49.55 54.53 25.73 5.04 4.29 14.65 14.45 43.89 3.71 43.19 16.49 16.65 21.46 6.61 p.< .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0003 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 .0001 Note: All degrees of freedom (df) = 1340. Ratings are on a “1” (very unlikely that creepy person displayed this characteristic/behavior) to “5” (very likely that creepy person displayed this characteristic/behavior) scale. On Creepiness Table 3 One Sample t-test Results for Beliefs about the Qualities of Creepy People Variable/Questionnaire Item Fearfulness (Composite) Proximity (Composite) Individual Creepiness (Composite) Non-normativity (Composite) Has bad intentions Has sexual interest Intentions are understood Creepier with multiple characteristics Not possible for creepy person to change Mean (SD) 3.79 (0.65) 2.78 (0.91) 4.18 (0.69) 2.56 (0.87) 2.74 (0.97) 3.39 (1.08) 2.96 (0.97) 4.35 (0.82) 2.66 (1.14) t value 44.63 8.91 62.24 19.54 9.87 13.16 1.67 60.13 10.93 p.< .0001 .0001 .0001 .0001 .0001 .0001 .096 .0001 .0001 Note: All degrees of freedom (df) = 1340. Ratings are on a “1” (strongly disagree with statement about creepy person) to “5” (strongly agree with statement about creepy person) scale. 22