Home / Articles / Volume 17 (2019) / Audience Address on Dating Profiles: My Desired-Other and Second Person Reference
Document Actions



Dating sites are goal-oriented spaces where users make profiles to form relationships. These users know that the actual audience reading their profile is large, containing both the desired-others (auditor-addressees) and undesired-others (auditor-overhearers). This study explores how profile makers on OkCupid reveal the desired-others in their imagined audiences through direct address via the profile affordance: “You should message me if…” This direct audience address is achieved by you-statements, which are phrases about or directed towards the desired-other using second person reference. 300 OkCupid dating profiles were examined for the frequency and variety of you-statements through content analysis. The analysis demonstrates that you-statements are a common feature on OkCupid and proposes a robust list of 14 functions. Findings suggest that you-statements can be used to reveal the profile maker’s desired-others in their imagined audiences through broadening and narrowing moves. Finally, you-statements both describe the desired-others, who are unique to each profile maker, and detail important aspects about the profile makers themselves.


Dating sites are a rich context in which to examine how website affordances, linguistic choices, and conceptions of the imagined audience interact in a real-stakes, goal-oriented space. As Ellison, Heino, and Gibbs (2006) observe, “the online dating arena represents an opportunity to document changing cultural norms surrounding technology-mediated relationships and to gain insights into important aspects of online behavior” (p. 415). This study focuses on how and why profile makers address their audience and include the profile reader via audience address.

Audience Design During Online Dating

In face-to-face communication, speakers typically know who the members of their actual audience are and are sensitive to their needs in terms of word choice, style of speech, and topic (Bell, 1984). Bell (1984) describes four categories of listeners: the addressees (known, ratified, and addressed), the auditors (known and ratified, but not addressed), overhearers (known, but neither ratified nor addressed), and eavesdroppers (completely unknown to the speaker). Speakers attend first to their addressee. In face-to-face contexts, they alter their performance in real time to match the needs of their addressee. Communicating online via profile text can be more challenging for users, because language choices cannot be altered in real time and lack many of the non-verbal clues present in face-to-face contexts. Online users can adapt to a restricted medium (text only) and learn to use it effectively to develop close relationships, but it is a skill that must be developed (Walther, 1992).

Knowing who is in the audience becomes more difficult online, particularly on dating sites. Bell’s categories blur; for example, online dating sites often keep track of visitors and then alert users when someone visits their profiles, which collapses the categories of eavesdroppers and overhearers together. Furthermore, all profile readers are in a way auditors, as the profile makers know they are there, but they are not necessarily being addressed. Among the profile readers, the people whom the profile maker would like to communicate with are the intended addressees; at the same time, they are unknown, which also makes them auditors. This study will refer to this group of online daters, who are both auditors and addressees, as the desired-others. Among all the profile readers, the people whom the profile maker would not like to communicate with are overhearers and auditors. This study will refer to this group as the undesired-others. The status of who is a desired-other and who is an undesired-other is unique to each profile maker.

Moreover, online daters make their profiles available to a large online community, knowing that they have little or no control over who looks at their profiles. The audience members of dating profiles have diverse backgrounds, interests, and purposes, which complicates the performance of the profile maker. This blurring is further complicated by context collapse: “the lack of spatial, social, and temporal boundaries” online (boyd, 2010, p. 10). More concretely, context collapse is the bringing together of diverse audiences into one networked audience, the members of which would not typically be addressed simultaneously (Androutsopoulos, 2014). People navigating social media must contend with context collapse daily, where it was once a relatively rare occurrence typical of large social functions, like weddings (Androutsopoulos, 2014).

This merging of diverse audiences can cause problems on social networks. For example, Facebook users must exercise caution, since one message meant to entertain friends could be highly pragmatically inappropriate for professional colleagues. Similarly, online daters cannot design their profiles to fit all possible audiences. The audience of online dating profiles is large and contains both desired-others (auditor-addressees) and undesired-others (auditor-overhearers). Online daters likely care about successful communication with their desired-others and care much less about adapting their language choices to their undesired-others.

Online dating profiles are examples of one-to-many communication, where the profile maker generates one profile to be viewed by an unknown number of other users. In a study of a group of French online daters, van Compernolle (2008a) found evidence that they envisioned communicating to a “real, definite ‘other,’” which “indicated an underlying or inferred one-to-one communication” (p. 2073) revealed through their use of the singular address tu (informal)/vous (formal). The structure of the dating ads had statements involving the reader as follows: greeting (36%), partner’s description (28.5%), and invitation to contact (62.5%) (van Compernolle, 2008b). These dating ad makers preferred the singular address tu (informal)/vous (formal) rather than the plural vous in 150 of the 200 examined ads (van Compernolle, 2008a). The singular address in French, particular the use of tu, marks a more intimate and interpersonal style of address (van Compernolle, 2008a). van Compernolle (2008 a, b) argued that this preference for a more personal address shows that most profile makers envisioned their conversations as one-to-one and interpersonal, particularly when the second person reference was used for descriptions of the desired-other. The following two excerpts are examples of dating ad makers describing their desired-other:

(1) Daniel Boon, où es-tu? Il ne manque que toi à mon bonheur.

Daniel Boon, where are you[T]? You’re[T] the only thing missing from my happiness.

(2) tu n’est pas compliqué, tu aimes la vie, tes simple tout comme moi.

you[T] aren’t complicated, you[T] love life, you’re[T] easy going just like me.

(van Compernolle, 2008a, p. 2070, numbering and emphasis in original)

In these examples and others like them, van Compernolle states that the advertiser is directly addressing her reader. Although the existence of her ideal partner is not known, the use of tu focuses the description on the referent, “putting the reader in the place of the desired other” (van Compernolle, 2008a, p. 2070). This envisioning of talking to a definite other adds another level of complexity on who the addressee of a dating profile is. The addressee is definite (members of my imagined audience), but non-referential (meaning that I do not have a particular person in mind, but rather a class of people).

Building on van Compernolle’s work, this article uses the term I-statements to refer to statements of self-description or self-preference (sentences using first person reference, such as I or my) made by the profile makers about themselves. The term you-statements is used to refer to statements made by the profile makers about, addressing, or implicating their desired-others using second person reference, such as you. You-statements are further described in the Data Analysis section under Methods.

It should be noted that this study’s you-statements and van Compernolle’s tu and vous statements are a good, but not a perfect, comparison. The second person pronouns tu and vous in French can distinguish levels of politeness and formality and whether a singular or plural audience is being addressed. The English pronoun you by itself does not distinguish levels of politeness or singular versus plural, but rather relies on surrounding context to make these distinctions. On a dating site, however, English you-statements can be understood by the profile reader (who is always an auditor) to imply an interpersonal communication, because the reader understands that someone definite is being addressed through the use of the pronoun you.

Sanford, Garrod, Lucas, and Henderson (1983) argued that when pronouns are used without a clear, direct antecedent in the communication, those involved use their shared knowledge to identify the antecedent (the referent) of the pronoun. On dating profiles, when you-statements are employed, online daters know whom the pronoun you refers to because of the expectation and expressed purpose of the place. Profile readers can use their shared knowledge of the dating site to realize that the statement is potentially about them, and they can read themselves into the statements or not (Sanford et al., 1983), similarly to what van Compernolle (2008 a, b) showed with the use of tu and vous singular.

Dating Profiles and the Imagined Audience

When a user makes an online personal profile, they are crafting an online artifact, or a contained and recorded performance, which can be accessed countless times asynchronously through websites by unknown users, and as artifacts can act as proxies for the profile makers themselves (Hogan, 2010; Reed, 2005). As Ellison et al. (2006) observed, a profile is “a crucial self-presentation tool because it is the first and primary means of expressing one’s self during the early stages of a correspondence and can therefore foreclose or create relationship opportunities” (p. 423). Users of dating sites such as OkCupid usually do not know other users before coming into contact. Indeed, the purpose of dating profiles is to connect people who are strangers for potentially personal and intimate relationships.

Dating profiles as artifacts are tools to separate the desired-others (auditor-addressees) from the undesired-others (auditor-overhearers). For writers, Ede and Lunsford (1984) make the distinction between the actual audience (the real readers) and the audience invoked (the envisioned group the writer constructs and writes for). On dating profiles, the actual audience is all the readers of the profile. The audience invoked is only the desired others, because profiles are written for the intended purpose of selling the profile makers to and attracting their desired-others (Ellison et al., 2006; Fiore, Taylor, & Mendelsohn, 2008; Jagger, 1998).

The imagined audience is a person’s “mental conceptualization of the people with whom he or she is communicating” (Litt, 2012, p. 330). People alter their communication to match and communicate effectively with their imagined audience, because they are their addressees (Bell, 1984), their desired-others (auditor-addressees). Dating profiles are premade artifacts rather than live performance, and they cannot be altered in real time. As artifacts, they are meant to help profile readers to self-select or self-deselect, as neither the profile readers nor makers want to date just anyone.

Ong (1975) observed that writers need to keep in mind their target audience in order to appropriately compose about their topic through using specific registers and styles of text. Moreover, there is a symbiotic relationship between online places and the textual interactions that happen in them (King, 2011). The place where interaction happens matters and helps shape the actions in it, while simultaneously the actions shape the place. For example, King (2011) described how a chatroom “emerge[d] as queer through identities and identifications of the participants” through their textual register, topics, and word choice (p. 21).

Profile makers, similarly, must through their writing style, topic selection, word choice, and register create an artifact that will attract and be judged positively by their imagined audience of desired-others. For example, if a profile maker employs emojis, memes, and textspeak, they are likely trying to display themselves as casual, young, or cool, because they want to attract people who would judge those qualities positively. In this case, the emojis, memes, and textspeak are some of the textual cues of the profile artifact that indicate who the desired-other is for this profile maker. They are signals for the actual audience to start self-determining if they can read themselves into the addressee role, into the imagined audience of the profile maker. Ideally, profile readers would then message the profile makers, believing themselves likely to be someone who would like and be liked based on evidence from the profile. In other words, profile readers identify themselves as legitimate members of the desired-others in the imagined audience by correctly reading themselves as the referent of the you-statements and positively judging the I-statements.

The Imagined Audience Online and Online Dating Profiles

Recent research (e.g., Litt, 2012; Marwick & boyd, 2010) has suggested that the idea of the imagined audience should be applied to social network sites such as Twitter and Facebook. These distinct online communities are constituted by registered users who have various tools to interact and share content (Androutsopoulos, 2014). As discussed above, along with context collapse, there are often negative repercussions for online textual posts because of a mismatch between the actual audience and the imagined audience. For instance, the tweet of a picture of a young person playing a drinking game in college in 2010 could lead to him not being hired in 2018, along with other negative effects (e.g., Sun, 2016). This negative tension on social media arises because online users make public posts only thinking about their (current) imagined audience of auditor-addressees (e.g., their college friends); they neglect to consider that on social media there are many current (and future) auditor-overhearers (e.g., family members) and/or auditor-eavesdroppers (e.g., future employers) (e.g., Barnes, 2006; Viégas, 2005) who have the potential to cause them social or professional harm.

Furthermore, Litt (2012) has highlighted a need to study the imagined audience across multiple online platforms because of the different affordances they offer. An affordance is the “relationship between an organism and a particular feature of its environment” (van Lier, 2000, p. 252). Affordances represent opportunities for action; they do not compel action, but rather they are subject to what is done with them (van Lier, 2000). The affordances, or the functionalities, of social network sites, such as friending, liking, or posting a video, impact how and when people communicate with their imagined audiences. Every website has different affordances, and the particular suite of affordances for each website significantly affects how users interact (boyd, 2006, 2010; Hogan, 2010; Litt, 2012).

The imagined audience on online dating sites and the dating profile affordance have been studied tangentially before (e.g., Ellison et al., 2006; van Compernolle, 2008a). However, they have not been examined directly on dating profiles in general or on OkCupid in particular, as far as this researcher was able to determine. Actually, the tension between the actual audience and the imagined audience on dating sites may have a positive strategic effect. Profiles are affordances meant to distinguish between desired-others and undesired-others, so the more quickly and accurately they can get undesired users to self-deselect, the better. When a profile maker composes their profile, they may benefit the most from writing in a way that pleases and helps their desired-others read themselves into the role of addressee. Thus profiles are affected by the tension of the imagined audience, but in a way that is distinct from other social network sites.

The present study is focused on OkCupid, a popular free English-language online dating platform, as it includes an affordance (a direct prompt) for users to include you-statements in their profile artifact. In previous research, online daters have been found to use certain language in their profiles in attempts to get some users (the undesired-others) to “weed themselves out” or self-deselect (Ellison et al., 2006, p.424). For instance, one male user included sexually explicit wording in order to signal to the profile readers who is, and who is not, his desired-other:

The reason I put [the language] in there is because I had some experiences where I got together [with someone], we both really liked each other, and it turned out that I was somebody who really liked sex and she was somebody that could take it or leave it. So I put that in there to sort of weed those people out. (Ellison et al., 2006, p. 424)

van Compernolle (2008a) found that often the use of you-statements “invites the reader – who is potentially the desired other – to measure himself or herself against the description of this hypothetical person” (p. 2070). The present study is concerned with the use of you-statements on OkCupid, as they represent the most explicit device for profile makers to reach their target users and reveal their imagined audience, through describing who is and who is not wanted.

Research Questions

This study responds to Litt’s (2012) call to consider the imagined audience across as many platforms as possible and focuses particularly on OkCupid because of its explicit affordance inviting you-statements. The study’s goals are to further understanding of the variety and frequency of you-statements, contribute to the literature on the language used by online daters, and show how online daters reveal their imagined audience. These aims are pursued by addressing two research questions:

RQ1: What are the functions of you-statements, how frequently does each function occur, and how common are you-statements in general, regardless of function?

RQ2: How are you-statements used to reveal the desired-others of the imagined audience?



OkCupid is an online platform where adults over the age of 18 make profiles about themselves primarily to pursue social, romantic, or sexual relationships with other adults they have not met before. A PEW study found that most US adults (59%) consider online dating a good way to meet people (Attitudes, 2015). In 2016, OkCupid had 10.15 million unique visitors a month (eBizMBA, 2016). More recently, OkCupid reports that on its platform 50,000 dates are made every week, and over 91 million connections are made every year (OkCupid, 2019).

The OkCupid profile offers several affordances that allow users to self-report and edit personal details. Users can add characteristic and demographic information, for example sexual orientation, gender, relationship status (e.g., single or seeing someone), and physical/personal characteristics. Minimally, as of February 2017, users must provide one picture and information on gender, orientation, and relationship status, and must select at least one goal: new friends, short-term dating, long-term dating, or casual sex. The profile also provides eight writing prompts (see Appendix). The prompts provide a structure to the profile, but users do not have to answer all prompts. The first seven prompts invite I-statements, and the eighth prompt invites you-statements. This study examines responses to the eighth prompt, “You should message me if…,” on the OkCupid profile.


Initially, information from 300 profiles (150 male and 150 female) was collected. These users were single, between the ages of 25 and 35, lived within 250 miles of the same large southern city, and were looking for long-term or short-term dating. They were online during February 2017. Single is a self-reported status that means not currently in a relationship: unmarried, widowed, or divorced. According to a 2015 PEW study, among American adults, people between the ages of 25 to 35 make up the largest group (22%) participating in online dating (Online, 2015).

Data Collection

Utilizing a dummy profile, the researcher employed the browse function to control for gender, age range, location, and stated goals (seeking long-term or short-term dating) in order to collect usernames. Due to the lack of an automatized randomization method on the site, every fifth profile was sampled. The usernames’ information was collected by a computer routine. The routine recorded biographical information and the responses to the writing prompts in an Excel sheet. Profiles that had not answered at least two of the writing prompts were considered unused and were deemed ineligible for this study. Following this criterion, 24 female and 20 male profiles were removed, which left 126 female and 130 male profiles to be analyzed, for a total of 256 profiles.

Anonymity of the Profiles

Because privacy is especially important to online daters, multiple steps were taken to insure the anonymity of the collected profiles. All the usernames needed for the original collection of information were immediately deleted and replaced with anonymous labels, such as Male Profile 1 or Female Profile 45. No pictures or exact locations were recorded. Further, the researcher tested the excerpt samples used in this article with various search engines to see if using the excerpts could lead back to the original profiles. The researcher experimented with combinations of keywords, exact phrasings, long and short excerpts, and the inclusion of the site name. The searches never led back to the profiles. In addition, the researcher used the data excerpts to search using the OkCupid internal search function and other general information (e.g., gender and age range) and could not locate the profiles. It was also decided not to reveal the name of the city that served as the epicenter of the search radius in order to provide an additional layer of privacy. Furthermore, the method by which these data were gathered in February 2017 no longer functions, as OkCupid changed how their platform works in early 2018.

Data Analysis

This study investigates the frequency and variety of you-statements within the context of OkCupid. The dual focus led to a mixed-method research approach (MMR). MMR has become increasingly popular with researchers over the years, because of the wide array of tools offered by the combination of quantitative and qualitative inquiry (Nunan & Bailey, 2009). The quantitative and qualitative data in MMR act as complementary elements, adding elaboration, clarity, and completeness to each other (Greene, Caracelli, & Graham, 1989).

Following data collection, the profiles were reviewed for tokens of you-statements. A token is a single occurrence of the target feature (Ellison et al., 2006). In this case, a token was one occurrence of a you-statement. You-statements are phrases that address or involve the profile reader. “Unitization [or the chunking of information for tokens] was flexible to capture complete thought units” (Ellison et al, 2006, p. 423). As necessary, surrounding I-statements were included in the token if the you-statement required context for clarity. You-statements are sentences referring to the second person. First, there were the sentences using the pronoun you. Each excerpt was treated as one token (i.e., one occurrence of a you-statement), for example:

  • You actually have the intention of meeting up in the near future.

  • A woman of confidence, ambition and drive impresses you.

Second, there were sentences using the implied you within imperatives. Each excerpt was treated as one token, for example:

  • Message Me me, im like a wal mart. I want your business :)

  • I’m an adult. Let's be adults and keep it classy.

Third, in rare cases, there were sentences using the pronouns we and us inclusively of the profile reader. In those few cases, they represented a combination of referencing the first person (i.e., the profile maker) and the second person (i.e., the profile reader). For example, the following excerpt was treated as one token:

  • Its gonna be us against the world – I’m waiting to meet

Thematic analysis was employed using a grounded theory approach (Creswell, 2009; Creswell & Clark, 2007; Strauss & Corbin, 1998), which allowed the generation of themes to emerge from the data. This approach was appropriate as one aim was to establish categories for you-statements where none existed before, as far as this researcher was able to find. Each token was examined by the researcher and assigned a function; the functions were iteratively tested and refined. More specifically, the first 100 tokens were examined for what functions they could serve. For example, some functions were valuing honesty or being serious. After the function list was generated, the researcher went back to the start of the list and began classifying each token again according to the list. If a token did not fit well into one of the functions, a new function was added to the list and the researcher went back to the start of the list to check the classification of each token again. Each token was treated as independent and could only be classified in one function.

After all tokens were classified, a list of 20 functions had been created. Some functions with tokens less than 2% of the total were reclassified into larger functions, as in the case of valuing honesty, which was moved into Profile Reader Personality. The function list was thereby reduced to 14 functions with a minimum frequency of at least 10 tokens each (2% or more of all tokens). Following this, the researcher revisited each token again to classify it according to the finalized list of functions. In addition, the frequency of the profile makers who used you-statement at all was noted. Finally, the frequency of each function was calculated, and the accompanying confidence ranges were calculated.

A second coder categorized the tokens into the 14 functions to determine the interrater agreement coefficient. The researcher trained the second coder on a different portion of the data, and 10% of the data was coded jointly. The second coder worked blind to the researcher’s coding, and each token was independent, counting for only one function. Using Cohen’s Kappa, the agreement was found to be moderate (κ = 0.73) between the researcher and the second coder.

Findings and Discussion

This section reviews the frequency and functions of you-statements in detail. In addition, how you-statements are used to depict, signal, and shape the imagined audience is discussed.

Frequency and Functions of You-Statements

Each of the 265 profiles was analyzed for the presence of you-statement tokens. A total of 202 profiles makers (76.3%) chose to use at least one kind of you-statement when prompted by OkCupid’s interface. Thus, prompted You-statements may be said to be a fairly common occurrence on OkCupid profiles. This finding is aligned with the notion that affordances affect the behavior of users by providing a structure (boyd, 2006, 2010; Hogan, 2010; Litt, 2012). However, it is worth noting that 23.7% of profile users chose not to make use of the eighth prompt inviting you-statements, which affirms van Lier’s (2000) observation that affordances offer but do not compel action.

The 202 profile makers together made 496 you-statements. As seen in Table 1, 14 functions or themes were identified during the thematic content analysis. Each theme is listed with its associated token count and percentage.



N          %



N          %

Physical Qualities of Profile Reader



Don’t Message



Friendship First












Physical Qualities of Profile Maker



We Match



Message Me



Maker’s Personality



Being Serious or Real






Meet in Person



Reader’s Personality



Table 1. Themes ordered by frequency with tokens count and percentage

Example tokens from each theme are given below in Table 2. No one theme accounted for more than about 13% of the tokens, and many users employed multiple kinds of you-statements on their profiles. The most frequently-occurring themes were Interests (12.3%) and Reader’s Personality (13.3%), but information about the Maker’s Personality (10.5%) followed closely behind. How a section ostensibly used to describe only the profile reader can also be used to talk about the profile maker will be discussed further on.




Example Token (Profile, Age)

Physical Qualities of Profile Reader



You have eyes like Sherlock

(Female Profile 9, 25 years old)

Friendship First



If you're looking for a new friend.

(Male Profile 39, 26 years old)




If you're witty and can keep up

(Female Profile 110, 30 years old)

Physical Qualities of Profile Maker



If you like negro noses and afros. You think I
am Cute
(Male Profile 16, 28 years old)

Message Me



Message Me me, im like a wal mart. I want your business :) (Male Profile 128, 31 years old)

Being Serious or Real



You're not playing any games and know what

you want (Male Profile 142, 31 years old)

Meet in Person



You actually have the intention of meeting up in the near future. (Male Profile 135, 30 years old)

Don’t Message



You don't have human kids

(Female Profile 150, 32 years old)




You like to laugh and can make me laugh

(Female Profile 31, 26 years old)




You want to have a good conversation over coffee, dinner, or drinks. (Male Profile 99, 29 years old)

We Match



After reading all this you think you might be the right fit for me. (Female Profile 70, 28 years old)

Attitude/Outlook/Attributes of Profile Maker



A woman of confidence, ambition and drive

impresses you. (Female Profile 20, 25 years old)




You love baseball (points for being a Giants fan) (Female Profile 56, 27 years old)

Attitude/Outlook/Attributes of Profile Reader



You’re 420 [weed] friendly

(Female Profile 23, 25 years old)

Table 2. Themes ordered by frequency with example tokens

As seen in Table 1, bodily characteristics are discussed in two themes, Physical Qualities of the Profile Reader and of the Profile Maker; the two themes account for 29 tokens, 10 for Profile Reader and 19 for Profile Maker. When profile makers did write about physical qualities, they were more often interested in wanting their own evaluated well, as seen in Table 2, e.g., If you like negro noses and afros. You think I am Cute (Male Profile 16, 28 years old). Personality characteristics were discussed in eight themes with a combined total of 328 tokens: Friendship First, Intelligence/Witty, Being Serious or Real, Humor, Conversation, Interests, Profile Maker Personality, and Profile Reader Personality. The comparatively smaller number of occurrences for physical characteristics may lend support to Jagger’s (1998) finding that daters promote their own personality and are interested in others’ personalities or life style choices more than physical traits in the text of dating profiles.

Friendship First, Intelligence/Witty, Humor, and Conversation themes marked potentially important cues for early relationship building, such as first having a solid conversation and working to build a friendship foundation. All four themes were either about wanting this quality valued in themselves, as in “I made you laugh” (Female Profile 8, 24 years old), or wanting it in their desired other, as in “You can hold a somewhat intelligent conversation” (Female Profile 73, 28 years old). Some you-statements reveal that OkCupid users show preferences for what they are looking for in their imagined audience, and these tokens may suggest areas commonly valued among some online daters. Interest tokens tended to describe things people feel strongly about, for example, “You wanna tell me how beautiful my dog is” (Female Profile 113, 30 years old), or suggestions for conversation starters, such as “you like deftones [a music group]” (Male Profile 101, 30 years old).

A major trend among the themes is the anticipation and encouragement of future interaction accomplished through the use of the use of broadening or narrowing moves. Broadening statements are those meant to expand the pool of potential desired-others. The most obvious broadening theme is Message Me, where profile makers directly request that profile readers message them: “Just message me me, im like a wal mart. I want your business :)” (Male Profile 128, 31 years old). Narrowing statements are commonly seen in the Being Serious or Real, Meet in Person, and Don’t Message themes. Narrowing statements attempt to constrict or set up minimum guidelines for contacting the profile maker. For example, the Being Serious or Real theme is used to narrow the field of potential valid matches with statements such as “You're not playing any games and know what you want” (Male Profile 142, 31 years old). This profile maker is likely trying to narrow his audience by actively discouraging other users whose interest is only casual from messaging.

You-statements in the theme We Match can direct profile readers to contact if they consider themselves a match for the profile maker. Consider, for example, the following profile creator’s answer to prompt eight (the you-statement is set off by [ ] and includes a superscript number):

[If you think you actually fit what I'm looking for and are actually interested in a relationship]1: I am looking for someone who can enjoy life, who loves to explore, willing to try new things and see new places, and make life an adventure (Female Profile 128, 31 years old)

This token invites the profile readers to measure themselves against the parameters that the profile maker has outlined (i.e., to judge whether they are members of the imagined audience described by the profile maker). Tokens like this example support van Compernolle’s (2008a) finding that audience address can in some cases invite “the reader – who is potentially the desired other – to measure himself or herself against the description of this hypothetical person” (p. 2070).

Another common pattern among all themes is a tendency for the profile makers to talk about themselves as opposed to the profile readers. The Maker’s Personality is in the top three most common themes (see Table 1) and almost exclusively concerned with describing parts of the profile makers’ characteristics that they want judged positively. For instance, the token “A woman of confidence, ambition and drive impresses you” (Female Profile 20, 25 years old) can act as a litmus test for what kind of person would be the desired-other of this profile maker, as she has described what to her are likely important and essential parts of her character.

The Conversation theme contains many examples of both kinds of statements. For instance, “If you like intelligent conversations and critical thinking” (Female Profile 59, 27 years old) and “If you just want to chat, I'm a talker and like meeting people” (Male Profile 166, 33 years old) are both statements about valuing or wanting conversation, but the former provides much stricter guidelines for the kind of conversation, whereas the latter opens the door for any topic. The first token also uses a more formal register with words like conversation and critical thinking, compared to the latter, which uses more casual words, like chat and talker. Broadening and narrowing can thus be accomplished through register, word choice, and degree of hedging.

Using You-Statements to Reveal the Desired-Others of the Imagined Audience

One of the profile maker’s goals for the text of the profile is to get the desired-others to initiate contact and undesired-others to self-deselect (Ellison et al., 2006). Online daters know that the actual audience (those who read their profile) is large and contains both the desired-others (auditor-addressees) and undesired-others (auditor-overhearers). You-statements in general can be used as affordances to distinguish between the addressees and overhearers in the actual audience.

This section examines how you-statements can assist in distinguishing between desired-others and undesired-others. As noted before, many profile makers include several kinds of you-statements in the eighth writing prompt (You should message me if…), which can send an overall mixed message, including both broadening and narrowing sentiments. For example, Female Profile 51’s eighth prompt contains four you-statements tokens:

[you're actually serious about making a connection with someone, and not just for "fun".]1 [I'll save you the trouble of messaging me by letting you know you're barking up the wrong tree if that's all you're after.]2 Also a good sense of humor and the ability to care about other people is pretty sweet, too [so message me]3 [if you've got that going on. ;)]4 (Female Profile 51, 26 years old)

The first token (“you're actually serious about making a connection with someone, and not just for “fun””) is a fairly strong limiting condition as a Being Serious or Real token. The imagined audience of desired others for this profile maker includes people who are seeking a serious relationship. By extension, the undesired-others for this profile maker include those seeking only casual relationships. The profile maker would like any undesired-others (in this case online daters looking for something “just for "fun"”) to self-deselect by not messaging. The second token (“I'll save you the trouble of messaging me by letting you know you're barking up the wrong tree if that's all you're after.”) intensifies the sincerity of the first token by using the phrase you’re barking up the wrong tree and ends with a second negative evaluation of casual daters, “if that's all you're after.”

The third and fourth tokens are interrelated. The third token is a Message Me invitation (“so message me”), which is conditional upon the fourth token (“if you've got that [a good sense of humor and the ability to care] going on. ;)”) being true. The fourth token is a Maker Personality one that sets up a broadening affect. First, the qualities of humor and care are evaluated in a casual and friendly way as “pretty sweet” in the previous I-statement. Second, the use an emoticon winking face adds an air of playfulness and openness to the Message Me token. This example shows how the mixing of you-statement themes can create a multifaceted message for profile readers. While not a guarantee of success, a profile reader who is looking for a serious relationship and is an honest and humorous person can tell that they meet some minimum set of conditions for this profile maker, which have been set out through the explicit shaping of these you-statements.

Some mixtures of you-statements create a message that describes both the profile reader and maker. For example, Male Profile 21’s eighth prompt contains four you-statements tokens:

[you want to meet up with someone to practice dating]1, [want a fluffy [heavy] guy to go workout with you at the gym]2, or [just want someone to go see that awesome new movie with you.]3 Random meetups for deep conversation is always a good night in my book. [You should know, I've only ever "dated" people with whom I was in a long-term relationship, so it would be a new practice for me to go on a date with someone I don't already know.]4 But that is what makes it fun, no? (Male Profile 21, 25 years old)

The first two tokens contain important narrowing caveats: First, the profile reader learns that this profile maker is somehow inexperienced with dating as he is seeking “someone to practice dating” with, and that he is a heavy guy, a “fluffy [heavy] guy.” The profile reader is asked to evaluate these two factors positively with the phrase “you want.” The message becomes more layered with the ending of the phrase “want a fluffy [heavy] guy to go workout with you at the gym.” This statement sets up a situation to let the profile reader know he is interested in getting fit, would likely prefer a partner who is interested in getting fit/is fit, and anticipates a possible future activity. These sentiments have the pronoun you as a subject, but they describe the maker’s personality and physical qualities primarily and the reader secondarily.

Next there is an I-statement buried in this text block, detailing an interest: “Random meetups for deep conversation is always a good night in my book.” This prompt answer ends with one more Maker Personality token stressing that he is new to dating unfamiliar people. The main focus of this message block seems to be a direct attempt to shape the desired audience by creating a litmus test for the profile readers to judge themselves by. That is, this profile maker is looking for someone who would like and be accepting of someone who is a somewhat inexperienced dater, heavy but looking to get in shape, and who likes deep conversations. The detailing of what is wanted by the profile maker also simultaneously implies what is not wanted, which in this case is someone not alright with inexperienced daters or heavy guys.


This study set out to explore the frequency and types of you-statements on OkCupid dating profiles. Speaking broadly, you-statements are a common feature on OkCupid. Over three-quarters of profiles in this study used one or more you-statements in response to OkCupid’s eighth prompt. Although dating profiles are actually one-to-many communication, You-statements were imperatives, descriptions, invitations, and questions seemingly directed at a single, unnamed subject. This supports the idea that OkCupid users treat the communication as interpersonal and confirms van Compernolle’s (2008a) finding that daters treat textual dating ads as a form of one-to-one, interpersonal communication to a real, if unknown, other.

Dating profile creators make intentional choices with their language, using you-statements to shape who may or may not read themselves into their imagined audience of desired-others (Ellison et al., 2006; van Compernolle, 2008a). You-statements are often used to describe the desired-other’s qualities. They also describe important aspects of the profile maker (e.g., not wanting children or being heavy) that are meant to be judged positively by the desired-others (i.e., the members of this profile maker’s imagined audience). This judgment may act as a litmus test for desired-other status.

The majority of users reacted to the eighth prompt (You should message me if…) by including you-statements, although many also included I-statements. Furthermore, while working with the data, this researcher observed that you-statements were also made when not prompted, that is, in the first seven prompts when only I-statements were anticipated. While users are affected by affordances and the structure they provide, as observed in this study (boyd, 2006, 2010; Hogan, 2010; Litt, 2012), they can take initiative and co-opt affordances for their own purposes.

The number of you-statements that are about the profile maker coupled with the number of I-statements in response to a prompt nominally for you-statements provided the foundation for the following observation. On online dating profiles, as I talk about ‘the you’ (whom I want), I talk about myself. In the same way, as I talk about myself, I talk about ‘the you’ (who would want me). Moreover, profile readers most likely treat a profile as a singular artifact, the words of each section affecting the overall impression a profile reader has of the profile maker. Thus on dating sites, the entire profile can potentially be said to differentiate between desired and undesired others.

When writing and representing one’s self online, users are typically more reliant on their imagined audience (or the audience invoked) than in face-to-face communication (Ong, 1975), due to the reduced number of verbal and nonverbal cues online (Walther, 1992). A tension exists between the actual audience and the imagined audience on social media sites, for “while we are dependent on the imagination as a guide during social media use, it is the actual audience on the other side of the screen reacting and judging the performance” (Litt, 2012, p. 303). Litt (2012) noted that this tension has had documented negative effects: professional, educational, and social fall out, such as getting fired or being publicly embarrassed (e.g., Sun, 2016). However, the success of online daters may be related to how well they utilize this tension between the audiences. Rather than being a negative factor for online daters, this tension may facilitate getting members of the actual audience to self-select as desired-others (auditor-addressees) or self-deselect as undesired others (auditor-overhearers). It may be the case that the more honest and specific profile makers are about who they are and who they want, the more likely their desired-others will self-select and message them. This suggests a special relevance for the imagined audience concept on dating sites as compared to other social networks.

Every statement, even ones describing the profile maker, has (hidden) audience selection. For example, consider the statement “I love Nascar racing!”; this phrase potentially narrows the field to those who are friendly to Nascar racing or at least not averse to it. What distinguishes you-statements from I-statements is that they are more explicit in their audience selection by directly addressing the audience. They contain invitations for interaction, sketch out what kind of people the profile maker is looking for, and detail potential deal breakers for the profile makers. The information encoded in you-statements is important to successful initial communication. This is meaningful, because interaction on dating sites has real stakes in the formation of interpersonal relationships.

You-statements have the potential to be a crucial affordance for online dating communication. Increasingly, dating apps (e.g., Tinder, Bumble, OkCupid) are moving towards providing browsing affordances that prioritize pictures and short textual information, which encourage users to swipe quickly right to keep and left to pass. If the text of dating profiles (particularly you-statements) carries critical information that leads to more successful interactions, this has several implications for both online daters and online dating website and app developers. First, readers of profiles should pay special attention to what is addressed to the reader. Second, profile makers should consider addressing the audience at least to some extent, regardless of whether or not there is an affordance for this. When addressing the audience, profile makers should privilege information important to themselves. The more personalized and honest the information, the better the litmus tests will be for readers to self-select as desired others or self-deselect as undesired others. Finally, designers of online dating websites and apps should consider including an affordance that prompts for you-statements because of their potential communicative power.

Some limitations of this study should be noted. First, the profiles collected do not necessarily represent all users of OkCupid or all people who participate in online dating. For example, all examined profiles were from users between the ages of 25 and 35; it is unclear if younger or older users behave the same way. Future research could expand the study to focus on different age groups, sexual orientations, or gender identities. A second limitation is that this study only examined responses to the eighth prompt. A deeper understanding of how the imagined audience is revealed could be achieved by analyzing profiles as whole units for how I-statements and you-statements are used to narrow and broaden. Third, this study did not systemically examine the linguistic structure of you-statements, such as comparing imperatives verses declaratives; this could add further nuance to understandings of language use in dating profiles. Finally, the validity of future studies on dating profiles could be increased by having follow-up interviews with profile makers to talk about their language choices and experiences with getting their desired-others to message them. Finally, future research could compare dating sites to other social media environments where the content producers are trying to get members of the actual audience to self-select or deselect.


I would like to thank Dr. Ran Wei from Rutgers University for his help in designing the routine to collect the user information. I would also like to extend gratitude to Kristin Quach from the Middlebury Institute of International Studies for her help as my second data coder. Thanks are also due to my anonymous reviewers for Language@Internet for their helpful comments and recommendations.


Androutsopoulos, J. (2014). Languaging when context collapse: Audience design. Discourse Context and Media, 4(5), 62-72.

Attitudes toward online dating grow more positive. (2015). [Graph illustrating the differences in attitudes between 2005 and 2015 Bar Graph June 12, 2015]. The PEW Research Center. Available from www.pewresearch.org/fact-tank/2016/02/29/5-facts-about-online-dating/ft_16-02-29_onlinedating_attitudes/

Barnes, S. B. (2006). A privacy paradox: Social networking in the United States. First Monday, 11(9), 4-9.

Bell, A. (1984). Language style as audience design. Language in Society, 13(2), 145-199.

boyd, d. (2006). Friends, friendsters, and myspace top 8: Writing community into being on social network sites. First Monday, 11(12). Available from https://firstmonday.org/article/view/1418/1336

boyd, d. (2010). Social network sites as networked publics: Affordances, dynamics, and implications. In Z. Papacharissi (Ed.), Networked self: Identity, community, and culture on social network sites (pp. 39-58). New York, NY: Routledge.

Creswell, J. W. (2009). Qualitative procedures. In J. W. Creswell (Ed.), Research design: Qualitative, quantitative, and mixed methods approaches (pp. 173-202). Thousand Oaks, CA: Sage.

Creswell, J. W., & Clark, V. L. P. (2007). Designing and conducting mixed methods research. Thousand Oaks, CA: Sage.

Ede, L., & Lunsford, A. (1984). Audience addressed/audience invoked: The role of audience in composition theory and pedagogy. College Composition and Communication, 35(2), 155-171.

Ellison, N., Heino, R., & Gibbs, J. (2006). Managing impression online: Self-presentation processes in the online dating environment. Journal of Computer-Mediated Communication, 11(2), 415-441.

eBizMBA. (2016). Leading dating websites in the United States in November 2016, based on visitor numbers (in millions). In Statista. Available from https://www.statista.com/statistics/274144/most-popular-us-dating-websites-ranked-by-monthly-visitors/

Fiore, A. T., Taylor, L. S., Mendelsohn, G. A., & Hearst, M. (2008). Assessing attractiveness in online dating profiles. Proceedings of Computer-Human Interaction, April 2008, 797-806.

Greene, J. C., Caracelli, V. J., & Graham, W. F. (1989). Toward a conceptual framework for mixed-method evaluation designs. Educational Evaluation and Policy Analysis11(3), 255-274.

Hogan, B. (2010). The presentation of self in the age of social media: Distinguishing performances and exhibitions online. Bulletin of Science, Technology & Society, 30(6), 377-386.

Jagger, E. (1998). Marketing the self, buying an other: Dating in a post modern, consumer society. Sociology, 32(4), 795-814.

King, B. W. (2011). Language, sexuality and place: The view from cyberspace. Gender and Language, 5(1), 1-30.

Litt, E. (2012). Knock, knock. Who's there? The imagined audience. Journal of Broadcasting & Electronic Media, 56(3), 330-345.

Marwick, A., & boyd, d. (2010). I tweet honestly, I tweet passionately: Twitter users, context collapse, and the imagined audience. New Media & Society, 13(1), 114-133.

Nunan, D., & Bailey, K. M. (2009). Exploring second language classroom research: A comprehensive guide. Boston, MA: Cengage Learning.

OkCupid The Proof. (2019). [Graphic listed on the OkCupid About Us Page]. Okcupid. Available from https://www.okcupid.com/about

Ong, W. (1975). The writer's audience is always a fiction. PMLA, 90(1), 9-21.

Online daters, by age group. (2015). [Graph illustrating the age brackets of online daters by percentage] April 2015. The PEW Research Center. Available from http://www.pewresearch.org/fact-tank/2016/02/29/5-facts-about-online-dating/

Reed, A. (2005). My blog is me: Texts and perons in UK online journal culture (and anthropology). Ethonos, 70, 220-242.

Sanford, A. J., Garrod, S., Lucas, A, & Henderson, R. (1983). Pronouns without explicit antecedents. Journal of Semantics, 2(3-4), 303-318.

Strauss, A., & Corbin, J. (1998). Grounded theory methodology: an overview. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of qualitative research (pp. 158-183). Thousand Oaks, CA: Sage.

Sun, C. (2016). These social media fails got people fired. Entrepreneur [website]. Available from https://www.entrepreneur.com/article/271823

van Compernolle, R. A. (2008a). Second-person pronoun use and address strategies in on-line personal ads from Quebec. Journal of Pragmatics, 40(12), 2062-2076.

van Compernolle, R. A. (2008b). Sociopragmatic norms in Quebecois on-line personal advertisements. In M. C. Koop (Ed.), Le Québec à l'aube du nouveau millénaire: Entre tradition et modernité (pp. 139-147). Québec, Canada: Presses de l'Université du Québec.

van Lier, L. (2000). From input to affordance: Social-interactive learning from an ecological perspective. In J. P. Lantolf (Ed.), Sociocultural theory and second language learning (pp. 245-259). Oxford, UK: Oxford University Press.

Viégas, F. B. (2005). Bloggers’ expectations of privacy and accountability: An initial study. Journal of Computer-Mediated Communication, 10(3). doi: 10.1111/j.1083-6101.2005.tb00260.x

Walther, J. B. (1992). Interpersonal effects in computer-mediated interaction: A relational perspective. Communication Research, 19(1), 52–90.


Biographical Note

Amy Russo [amy.russo@sjsu.edu] is the Coordinator of Multilingual Writing Support Services in the Writing Center at San Jose State University in San Jose, California. Her research interests include online language use, academic discourse, audience design, and peer writing tutor training.


Any party may pass on this Work by electronic means and make it available for download under the terms and conditions of the Digital Peer Publishing License. The text of the license may be accessed and retrieved at http://www.dipp.nrw.de/lizenzen/dppl/dppl/DPPL_v2_en_06-2004.html.