A 42 item Internet questionnaire was completed by 268 undergraduates at a large southeastern university to assess the frequency, motivations, and outcome of snooping in romantic relationships. Almost two thirds (66%) reported that they had engaged in snooping behavior, most often when the partner was taking a shower. Primary motives were "curiosity" and "suspicion" that the partner was cheating. Being female, jealous, and having been cheated on were associated with higher frequencies of snooping behavior. Uncertainty reduction theory was used to explain snooping behavior. Implications of the data suggest that the decision to snoop should be considered with great caution as almost twice as many relationships were worse (28%) as improved (18) as a result of the snooping. Limitations of the data are identified.
Snooping, also known as covert intrusive behavior, is defined as investigating (without the partner's knowledge or permission) a romantic partner's private communication (e.g. text messages, cell phone and email). Snooping behavior emerges in a context of mistrust and is fed by the perception that the partner is withholding information (not disclosing). Vinkers et al. (2011) noted that snooping is functional for the snooper since it provides information that is supposed to be a secret and gives power over the partner who cannot deny certain facts (e.g. "I know you sent a text message to your ex saying that you missed her because I read your text"). Snooping typically has negative consequences in that it is associated with increased conflict, decreased trust, and strained interaction.
High levels of personal disclosure are expected in close relationships. When an individual in a romantic relationship feels that his or her partner is not open/disclosing/secretive, he or she feels hurt and devalued. Individuals who feel anxious and uncertain about the partner's lack of disclosure are motivated to gain increased information which sometime results in snooping. Hence, low disclosure creates the context in which persons may snoop to gain increased information about the partner's behavior to correct the behavior if it is a threat to the relationship.
Vinkers et al. (2011) studied 188 couples married an average of two years and found that lower levels of perceived partner disclosure were associated with higher levels of intrusive behavior (snooping). Second, perceived disclosure was negatively associated with intrusive behavior at lower levels of trust in one's partner, but not at higher levels of trust. Hence, if one partner did not trust another, even though there was high disclosure, intrusive behavior still occurred. And, if the trust level was high, even though disclosure was low, snooping was less likely. Also, couples in the Vinkers et al. (2011) study believed that wives engaged in more snooping (intrusive behavior) than husbands because women have a higher need for emotional involvement and men have a greater need to control their privacy than women. The researchers did not find an actual gender difference in snooping behavior.
A 42 item questionnaire (approved by the Institutional Review Board) on "Snooping-A Survey on Attitudes and Behaviors) was posted on the Internet from March 15 to April 15, 2011. Students in five lower division sociology classes at a large southeastern university were emailed the link and asked to complete the survey. Demographic questions regarding gender, year in school, relationship status, religion, number of serious relationships, etc. preceded being asked questions about snooping.
A total of 268 respondents completed the survey. The majority of respondents (approximately 84%) were female, white (approximately 74%), and heterosexual (approximately 91%). Over half of the respondents (over 54%) described their relationship status as emotionally involved with one person, but not engaged or married. Nearly 29% were not dating and not involved with anyone; about 10% were dating different people. The mean age of the respondents was approximately 20 years old.
In addition to identifying the frequency of snooping behavior, the following research questions were explored:
1. By what means do college students snoop in romantic relationships?
2. When do college students in romantic relationships snoop?
3. Are there gender differences in snooping behavior?
In order to answer the first (how snooping occurs) and second (when snooping occurs) research questions, we descriptively explored responses to the survey questions where respondents revealed the various ways they snooped and the time at which they did so, respectively. To answer the third (gender differences) research question, regression models were created with two different dependent variables using the following survey questions:
1. With snooping is defined as investigating (without the partner's knowledge) a romantic partner's private communication (such as text messages, email, and cell phone use), I have engaged in snooping behavior with a current or past partner.
2. How many times have you snooped?
Logistic regression analysis in SPSS was used to explore whether snooping has occurred and OLS regression was utilized to explore how many times.
Almost two thirds (66%) of the respondents admitted to snooping on a romantic partner. Respondents reported having snooped a mean of approximately 3 times (with a standard deviation of 5 and a range of 0-50). Over 80 percent (81%) reported that they had a friend who told them they had snooped on a romantic partner.
Why Respondents Snooped
Table 1 reveals the reasons respondents gave for engaging in snooping behavior.
The data in Table 1 reflect "curiosity" as the top reason for engaging in snooping behavior. Whether such curiosity was innate to the snooper or triggered by the partner is unknown. Suspicion of cheating was the second greatest explanation for snooping. We discuss suspicion in detail in the section on theoretical explanation for cheating. Always being a snoop and revenge for the partner's snooping were identified by only three percent of the respondents. A fourth noted no reason for snooping since they had never snooped.
How Respondent Snooped
Table 2 reveals the various ways the respondents identified that they engaged in snooping behavior.
The data in Table 2 reflect that the text messages the partner had sent was the primary target for snooping behavior. Almost 90 percent (88%) percent reported going through the partner's text messages. The second most frequent target was the partner's cell phone with almost
two-thirds (65 %) reporting snooping through the partner's cell phone history. Not far behind was snooping the Internet history of the partner (42%). Less often the target of snooping was the partner's email (25%), room (19%), desk (13%) or clothes 4%.
When Respondents Snooped
Table 3 reveals the various times and percentages the respondents reported engaging in snooping behavior.
The data in Table 3 reflect that snooping occurred most frequently when the partner was taking a shower. Almost forty percent selected this option. The second most frequent time for snooping was when the partner was not at home. Snooping when the partner was on the phone or on a date accounted for only a small percentage (3 and 2 percent). As noted, 27% reported that they never engaged in snooping behavior.
Demographics of Respondents who Snooped
In regard to the final research question, "Who snoops?", we ran logistic and OLS regression models. In the full model, the following independent/control variables were included for each: sex, race, age, sexual orientation, religiosity (level of and religious preference), self concept, jealousy, prior cheating and being cheated on. Sex was coded as male/female (male-reference category), race was coded as white/black/other (white-reference category), sexual orientation was coded as heterosexual/other (other-reference category), level of religiosity was coded as very religious/somewhat religious/not at all religious (not at all religious-reference category), religious preference was coded as Christian/other/no religious association (no religious association-reference category). In addition, four variables were response categories on a 5-point Likert scale assessing agreement with each of the following statements: I have a positive self-concept; I am a jealous person; I have cheated on a partner I was involved with; A partner I was involved with cheated on me.
Logistic regression was performed for the dependent variable of whether or not respondents had ever snooped, with categories of "yes" and "no." Findings indicate that four of the independent variables were significant at the p < .05 level in predicting odds of snooping. Being female, being older, being a jealous person, and having a partner who cheated on the respondent were all associated with engaging in snooping behavior. Race, sexual orientation, self concept, and having cheated were not associated with snooping behavior. Table 4 summarizes the full model, which resulted in a Chi-Square value of 43.446 and a Nagelkerke R-Squared value of .245. This R-squared value indicates that still over 75% of the DV is left to be explained after accounting for these variables.
To assess associations between the number of times a respondent snooped (dependent variable) and various independent variable factors, linear (OLS) regression was performed. Findings revealed that four of the independent variables are significant at the p < .05 level in predicting the odds of snooping. Being female, being religious, identifying with Christianity (in contrast to no religion) and being jealous were associated with higher numbers of times the respondent reported engaging in snooping behavior. Race, sexual orientation, and having been cheated on were not associated with higher numbers of snooping behavior. Table 5 summarizes the full model, which resulted in an R-Squared value of .087. This is a very low R-squared, indicating that this model can explain less than 9% of the variance in the dependent variable.
1. "You only snoop on the one you love." The data revealed that persons in new relationships thought snooping was inappropriate. But if the couple had been seeing each other and were involved, snooping was OK. Only six percent of the respondents agreed that snooping was appropriate if the partners had just begun seeing each other. But one fourth agreed that if the partners are in a mutually agreed monogamous relationship, snooping is justified. And, if there was a hint of suspicion, the percent who felt that snooping was justified jumped to almost 60 percent (59%).
2. "One should feel guilty for snooping." While snooping seems justified if there is reason to be suspicious, where there is no reason to doubt the partner, there is the feeling that snooping is not right and that the snooper should feel guilty. Almost 70% of the respondents (68%) agreed that "a person should feel guilty for snooping without a reason." Forty percent had no reason- they say that they just snooped out of curiosity. About a third (32%) said that their partner drove them snoop. Almost two thirds of the respondents (65%) reported that they would be angry if their partner was snooping through their messages, cell phone, email, etc. One in five partners of snoopers did not know they were being snooped on.
4. "Confrontation sometimes occurs." Almost a third (31%) of the respondents reported that they confronted the partner with what they found snooping (21% said they discovered their partner had cheated).
5. "Outcome of snooping." Over a third (36%) said that they found what they were snooping for. About a third (33%) said that their snooping confirmed their doubts and fears but one fourth said that snooping reduced their doubts. About one in five (18%) said that their relationship was stronger because they snooped but almost 30% (28%) reported that their relationship was worse. In effect, while snooping helped to reduce the uncertainty the snooper was feeling in terms of information, the outcome for the relationship had the potential to create damage.
Theoretical Explanation for Snooping
Uncertainty Reduction Theory (URT) provides a way to understand relationship ((Knobloch et al. 2010; Dawkins, 2010) and snooping behavior (Afifi and Burgoon. 1998). Uncertainty reduction theory argues that the primary goal of individuals in relationships is to reduce uncertainty and increase the ability to predict the behavior of others. Defined by Berger and Calabrese (1975), uncertainty reduction is a primary motivating factor involved in interpersonal communication. As such, it can be both proactive and retroactive. Proactive mechanisms used to get information/reduce uncertainty include Googling as an aid to predict the partner's future behavior. Retroactive mechanisms include asking friends/peers to provide explanations so as to give meaning/understanding for the events that have already occurred. In both cases the goal is to reduce uncertainty and increase predictability.
Berger (1979) argues that certain conditions exist in all situations involving uncertainty reduction. The first condition is the potential of the other person to reward or punish. For example, if Erin is a very popular person on campus, then John may see her attention paid to him as a reward. Likewise, John might experience a rejection from Erin as punishment.
Another condition is when the other person's behavior is contrary to expectations. In the case of Erin and John, let us assume that John expects Erin to always be interested in hanging out with him. But if she suddenly says she has other plans, John's expectations would be violated and his desire to reduce his perception of her uncertainty would increase. In reference to snooping, individuals in a relationship value each other and are reinforcing to each other. When one begins to show the desire for distance in the relationship, uncertainty is created and reasons for the uncertainty are sought. If direct questioning (e.g. "Why don't you want to hang out with me?") does not provide the desired answer, or asking friends for an interpretation, the person may snoop for answers. Berger (1986) refers to this as using various strategies to reduce uncertainty interactive (e.g. ask the person a direct question) or active (ask friends/snoop).
Related to looking for signs of commitment to resolve feelings of uncertainty about the relationship is the fear that there may be violations in expectations the partners have of each other which may increase feelings of uncertainty (Afifi and Metts, 1998). For example, if there is suspicion that one's partner is lying/cheating, snooping provides a means of discovery to remove the uncertainty. Such snooping is further indicated when the partner is perceived as not providing full disclosure and a norm may have developed in the relationship not to discuss the perception that one is not being open. Baxter and Wilmot (1985) noted that taboo topics emerge in all relationships and one's level of openness is one of them. Hence, snooping replaces non disclosure.
Other forms of uncertainty are about what the partner is thinking, curiosity about the partner's life, and jealousy. Afifi and Reichert (1966) emphasized the value of uncertainty theory for understanding jealousy. Feeney (1994) noted the hurt partners feel when they have been betrayed. Indeed, a third of the respondents in the current study reported that their partner drove them to snoop. The level of uncertainty was so high the individual had to act ... to do something ... to snoop ... to reduce the anxiety.
URT has also been criticized as limited in terms of its assumptions and validity. Sunnafrank (1986) argued that "maximization of relational outcomes," and not uncertainty, is the primary goal of initial encounters. He called for a different framework based on predicted outcome values (POV). Drawing on our Erin and John examples, Sunnafrank argues that John will be more concerned with maximizing rewards in a potential relationship with Erin than figuring out her actions and motives. Berger responds to this critique by concluding that Sunnafrank has extended the scope of URT rather than offered an alternative to it. Other areas of critique are: that uncertainty exists beyond initial encounters (Parks & Adelman, 1983); that many times we communicate to reduce uncertainty only because we care about and/or are interested in the other (Kellerman & Reynolds, 1990).
In reference to the current study, snooping as a mechanism to reduce uncertainty seemed to work. Over a third (36%) said that they found what they were snooping for. Another third (33%) reported that their snooping confirmed their doubts and fears and a fourth said that snooping reduced their doubts. Hence, the uncertainty was relieved by the snooping either by confirming one's fears or confirming there was no basis for one's fears.
Snooping was normative in this sample of undergraduate relationships. About two thirds of the respondents reported that they had snooped on their romantic partner. Over 80% reported that they had a friend who had engaged in snooping behavior. While most snooped out of curiosity, others snooped because they were suspicious that their partner was cheating. Uncertainty reduction theory suggests that relationship uncertainty (e.g. how committed is the partner?, is the partner cheating?) drives the person to seek information to resolve the uncertainty.
Being female, being jealous, and having been cheated on were associated with a higher number of times the respondent reported engaging in snooping behavior. Previous research has documented that women are more likely to be invested in romantic relationships (Knox et al., 1998; Bell and Buerkel-Rothfuss, 1990) and are more attentive to when their partner is lying/cheating (Abowitz et al. 2009). Snooping is another expression of concern about the relationship and an attempt to keep the partner and the relationship on track. It comes without surprise that feeling jealous was associated with snooping behavior. In addition, having cheated on a partner creates the belief that no one can be trusted, not even ones self.
Most snooped when their partner was taking a shower and looked at the partner's text messages, cell phone or Internet history. Such snooping occurred after the partners were in an established monogamous relationship and included some guilt-almost seventy percent said one should feel guilty if snooping without a reason.
Snooping had a variable outcome. About a third (33%) said that their snooping confirmed their doubts and fears but one fourth said that snooping reduced their doubts. About one in five (18%) said that their relationship was stronger because they snooped; but 28% said the relationship got worse.
There are three implications of the data. First, snooping in romantic relationships is normative- two thirds of the undergraduates in this study reported having snooped.
Second, snooping may have negative consequences for the relationship. Almost twice as many relationships were worse (28%) as improved (18) as a result of the snooping. Vinkers et al. (2011) confirmed that snooping typically has negative consequences in that it is associated with increased conflict, decreased trust, and strained interaction.
Third, snooping on one's romantic partner should be engaged in with some hesitation/reservation. One must weigh the potential benefits (e. g. finding out if the partner is cheating) against the costs (e. g. the partner may not like it). A quarter of the respondents said their partner would be outraged if they were aware of the snooping.
There are several limitations of this study. First, the data should be interpreted cautiously in that they are skewed toward white females (84%) as well as freshmen/sophomores (62%). Second, the convenience sample of 268 undergraduates is hardly representative of the 19.8 million college students throughout the United States (Statistical Abstract of the United States: 2012). Third, the data are quantitative with no qualitative interviews to provide insights on the raw statistics. Subsequent research might include interviews with students who have engaged in snooping behavior to assess their feelings in doing so and the consequences for them, their partner and the relationship. Fourth, there were numerous low r-squared values which explained very little variation in the dependent variables. Five, some inconsistencies in the data emerged. For example, 66 percent of the respondents reported that they had snooped. But when asked "When did you snoop?" 27% reported that they had never snooped.
Abowitz, D. A., D. Knox, M. Zusman, and A. McNeely. 2009. Beliefs about romantic relationships: Gender differences among undergraduates. College Student Journal 43: 276-284.
Afifi, W. A. and J. K. Burgoon. 1998. "We never talk about that": A comparison of cross-sex friendships and dating relationships on uncertainty and topic avoidance. Personal Relationships 5: 255-272.
Afifi, W. A. and S. Metts. 1998. Characteristics and consequences of expectation violations in close relationships. Journal of Social and Personal Relationships. 15: 365-392.
Afifi, W. A. and T. Reichert 1996. Understanding the role of uncertainty in jealousy experience and expression. Communication Reports. 9: 93-103.
Arriaga, X. B., J. T. Reed, W. Goodfriend and C. R. Agnew. 2006. Relationship perceptions and persistence: Do fluctuations in perceived partner commitment undermine dating relationships? Journal of Personality and Social Psychology. 92: 1045-1065.
Baxter, L. A. and W. W. Wilmot. 1985. Taboo topics in close relationships. Journal of Social and Personal Relationships. 2: 253-269.
Bell, R. A. and N. L. Buerkel-Rothfuss. (1990). S(he) loves me, s(he) loves me not: Predictors of relational information-seeking in courtship and beyond. Communication Quarterly. 38: 64-82.
Berger, C. R. 1986. Uncertain outcome values in predicted relationships: Uncertainty reduction theory then and now. Human Communication Research, 13: 34-38.
Berger, C. R. 1979. Beyond initial interaction: Uncertainty, understanding, and the development of interpersonal relationships". In Language and social psychology, edited by: Giles, H. and St. Clair, R. 122-144. Oxford: Blackwell.
Berger, C. R. and R. J. Calabrese 1975. Some explorations in initial interaction and beyond: Toward a developmental theory of interpersonal communication. Human Communication Research 1: 99-112.
Dawkins, M. A. 2010 How it's done: Using Hitch as a guide to uncertainty reduction Theory. Communication Teacher, 24:136-141
Feeney, J. A. 2004. Hurt feelings in couple relationships: Towards integrative models of the negative effects of hurtful events. Journal of Social and Personal Relationships. 21: 487-508
Kellerman, K. and Reynolds, R. 1990. When ignorance is bliss: The role of motivation to reduce uncertainty in uncertainty reduction theory. Human Communication Research 17: 5-35.
Knobloch, L. K., K. L. Satterlee and S. M. DiDomenico. 2010 Relational uncertainty predicting appraisals of face threat in courtship: Integrating uncertainty reduction theory and politeness theory. Communication Research 37: 303-334.
Knox, D., S. Hatfield, and M. E. Zusman. 1998. College student discussion of relationship problems. College Student Journal 32:19-21.
Parks, M. R. and Adelman, M. B. 1983. Communication networks and the development of romantic relationships: An expansion of uncertainty reduction theory. Human Communication Research 10: 55-79.
Statistical Abstract of the United States: 2012. 131st Ed. Washington, D.C.: U.S. Bureau of the Census
Sunnafrank, M. 1986. Predicted outcome value during initial interactions: A reformulation of uncertainty reduction theory. Human Communication Research 13: 191-210.
Vinkers, C. D. W., C. Finkenauer and S. T. Hawk. 2011. Why do close partners snoop? Predictors of intrusive behavior in newlywed couples. Personal Relationships, 18: 110-124.
DAVID KNOX *
East Carolina University
University of Tennessee at Knoxville
Table 1 Why Undergraduates Snooped N = 268 Reason Percent "I am just curious" 40% "I suspected my partner was cheating" 29% "I have always been a snoop." 3% "My partner snooped on me." 3% "I have never snooped." 25% Table 2 Most Frequent Ways Undergraduates Snooped N = 268 Snooping Behavior Percent Text messages 88% Cell phone 65% Internet history 42% E mail 25% Searched room 19% Searched desk 13% Searched clothes 4% Other 18% Table 3 When the Respondent Snooped N=268 Time of Snooping Behavior Percent When partner in shower 39% When partner was not at home 17% When partner was sleeping 13% When partner was on phone 3% While on a date 2% Never engaged in snooping 27% Table 4 B (Exp) B Sig. Female 1.575 4.883 .001 Race-Black .162 1.175 .741 Race-Other -.170 .844 .747 Very Religious -.663 .531 .383 Somewhat Relig -1.017 .362 .103 Christian 1.274 3.575 .075 Other Religion .876 2.402 .298 Heterosexual .410 1.506 .516 Age .246 1.279 .026 Pos. Self-concept .213 1.238 .293 Jealous .623 1.864 .001 Cheated .114 1.121 .353 Been cheated on .242 1.274 .037 (Constant) -9.982 .000 .000 Table 5 B Beta Sig. (Constant) -13.771 .014 Female 2.531 .153 .036 Race-Black -.358 -.021 .775 Race-Other -.740 -.041 .582 Very Religious -4.516 -.307 .009 Somewhat Relig -3.958 -.337 .007 Christian 4.898 .349 .004 Other Religion 2.699 .130 .201 Heterosexual 1.730 .073 .324 Age .365 .133 .079 Pos. Self-concept .299 .044 .557 Jealous 1.037 .176 .017 Cheated .583 .139 .061 Been cheated on -.083 -.022 .771