Gender roles are changing. As more women enter positions of social power, global assessments of gender equality reflect greater access and participation of women in the domains of education, healthcare, and politics. As of 2016, Canada and the United States had closed over 70% of the gender gap in economic participation, educational attainment, physical health, and political empowerment (Leopold et al. 2016). Personal attitudes reflect these social changes, becoming increasingly more positive toward women and more egalitarian over time (e.g., Twenge 1997a). But this change seems slow, if not nonexistent in the domain of romantic relationships (see England 2010). In the intimate and sexual relationships between men and women, traditional gender roles appear to be resistant to the evolving social beliefs outside the bedroom. In these relationships, women traditionally hold less power (Felmlee 1994; Sprecher 1985) and are expected to enact a reactive role to complement the more active role of their male partners (Rose and Frieze 1989).

As the springboard for all future romantic interactions between women and men, research has focused a great deal of attention on gender role expectations on a first date (for a review see Eaton and Rose 2011). We add to the existing literature by collecting the first date scripts of a recent sample of university students and community adults to investigate the influence of gender, age, and relational context before the first date. Moreover, we are the first known to quantitatively compare a recent sample to a past sample (collected by Laner and Ventrone 2000) to explore whether hypothetical first date scripts are changing to reflect growing egalitarian attitudes and a shift in women’s social roles.

First Date Scripts

Scripts are socially constructed expectations or norms for the behaviors that should be enacted in certain contexts and as such, they serve to ease the flow of social interaction by providing clear guidelines for actions (Ginsberg 1988). First date scripts, a type of relationship script, prescribes the actions and events that are expected to occur on a first date (Eaton and Rose 2011). As such, first date scripts similarly allow script enactors to assuage their own uncertainty by providing a pre-scripted set of behaviors to follow (Eaton and Rose 2011; Laner and Ventrone 1998). Although first dates may mark a successful overture (i.e., an invitation was made and accepted), they are best described as an initial encounter that reflects the potential transition from a platonic relationship to a romantic one (Baxter and Bullis 1986). Thus, first dates are rich with both hope for the formation of a romantic (and/or sexual) relationship and fear of rejection and disappointment (see also Cameron et al. 2010). Logically, uncertainty reduction is a common and primary concern for most people on a first date (Mongeau et al. 2007), and adhering to social scripts may fulfill that need (Eaton and Rose 2011; Laner and Ventrone 1998).

Scripts not only serve to reduce the uncertainty of one’s own actions but also provide clear rules on how to evaluate the actions of others. People ascribe more positive evaluations to those who adhere to the first date script (Eaton and Rose 2011; Green and Sandos 1983; Laner and Ventrone 1998; McCarty and Kelly 2015). Moreover, people use scripts not only as evaluative benchmarks when reading hypothetical scenarios as typical in research on this topic, but also when evaluating their own relationships (see Holmberg and MacKenzie 2002). Although scripts may facilitate social interaction and judgment, they also restrict behavior by limiting the range of what is considered socially acceptable.

In first date scripts, behavior is most noticeably restricted by gender. First date scripts follow traditional gender stereotypes whereby men are assigned active and dominant behaviors (e.g., asking for the date) and women are prescribed reactive behaviors (e.g., waiting to be asked; Cate and Lloyd 1992; Rose and Frieze 1989). Thus, in the expectations that people hold for how first dates should unfold, women are given the less powerful, submissive role of reacting to the actions and decisions of men. As such, even in the most intimate of interactions, men are provided the privileged and powerful role of actor (see also Sassler and Miller 2011). This adherence to traditional gender roles is evident not only in the types of behaviors ascribed to women and men but also in the sheer number of behaviors. Compared to the actions assigned to women, men are assigned a greater number reflecting the expectation that men are the lead actor in the first date script (Rose and Frieze 1989). Not only is it considered normative that men are expected to take the initiator role and women the role of waiting (Schleicher and Gilbert 2005) but these roles are also more likely to be enacted when initiating relationships (Clark et al. 1999; Rose and Frieze 1993). Moreover, these roles transfer to expectations for the resulting romantic relationship. For example, men are expected to take control over and initiate all later movement toward deepening commitment (Sassler and Miller 2011), including more formal rituals such as marriage proposals (Robnett and Leaper 2013). First date scripts, then, seem to lay the foundation for a consistent expectancy of adherence to traditional gender roles in romantic relationships (see Cate and Lloyd 1992).

In addition to extending across stages of a relationship, the consistency of gender prescribed roles extend across time. In a recent selective review of the literature, Eaton and Rose (2011) noted that the gender-prescribed behaviors for first date scripts has remained consistent as evidenced in over more than two decades of research and more than 60 years of published books on dating advice. Previous researchers report remarkable consistency in the presence of traditional gender roles in dating advice books from the 1950s to the 1980s (Rose and Frieze 1989) and into the 1990’s (Laner and Ventrone 2000), extending to the bestselling dating advice books of 2010 (Eaton and Rose 2011). Moreover, research on both hypothetical and actual first date scripts finds that behaviors on first dates are largely scripted in accordance with traditional gender stereotypes. These observations have led to the conclusion that not much has changed in gender role prescriptions on a first date.

First date scripts are also highly coherent across a diverse array of samples. Both men and women describe hypothetical first dates, typical dates, good dates, and actual dates in a remarkably consistent manner, supporting the strength of the cultural script for what events should occur on a first date (Alksnis et al. 1996; Bartoli and Clark 2006; Morr Serewicz and Gale 2008; Pryor and Merluzzi 1985; Rose and Frieze 1989, 1993). And although the majority of research on first date scripts has been conducted with predominantly White samples collected in the United States (Rose and Frieze 1989), Hispanic participants produce similar actual first date scripts (Eaton et al. 2016). First date scripts are also consistent across age groups, from early adolescence (Robnett and Leaper 2013) to undergraduate samples (Laner and Ventrone 2000) and college-educated community-recruited adult samples (Lamont 2014). First date scripts are similarly immune to experience, being similarly constructed among those with and without previous dating experience (Pryor and Merluzzi 1985). The consistency of the gendered nature of first date scripts is also evidence across both cognitive script (Morr Serewicz and Gale 2008; Rose and Frieze 1989, 1993) and checklist (Laner and Ventrone 2000) methods. Overall, the gender-typed structure of first date scripts appears to be robust and culturally entrenched.

Although the literature overwhelmingly suggests that both hypothetical and actual first date scripts adhere to traditional gender stereotypes, no known research to date has quantitatively contrasted an earlier sample to a more recent sample using a comparable methodology. Even though Eaton and Rose (2011) provide a selective and systematic review of the literature on first date scripts, such reviews are limited to conclusions from previous work. Thus, systematic reviews can readily identify consistency in the literature but any nuanced changes, such as diminishing effect sizes, could easily be missed. Thus, we quantitatively compared a recently collected sample in 2015 to a previously collected sample (Laner and Ventrone 2000) using parallel methodologies to allow for direct comparisons. By doing so, we were afforded an opportunity to quantify the stability of gender-typed hypothetical first date scripts while also allowing for the possibility of observing any hint that gender-typed hypothetical first date scripts might be weakening over time.

Will First Date Scripts Change?

Despite their adherence to traditional gender roles in first date scripts, gender roles and gender-based attitudes, especially toward women, are changing in North America. Although gender equality has not been achieved, women now hold more positions of power and influence than before. For example, compared to decades past, more Canadian women are employed, well-educated, and in leadership roles (Statistics Canada 2017). According to social role theory, changes in social roles will alter attitudes and stereotypes to eventually align with new social roles (Eagly and Wood 2012). Thus, as more women take the agentic roles traditionally assigned to men in work and educational spheres, attitudes toward women should similarly change, seeing women as more agentic and competent in general. In support of the theory, empirical evidence has demonstrated that when social roles change, stereotypes also change (Koenig and Eagly 2014). Indeed, social trends in gender-based attitudes support this proposition. For example, in the United States, people are more accepting of women in leadership roles, have more positive attitudes about women’s competence, and are increasingly seeing women and men as equals (Eagly et al. 2003; Twenge 1997a, 1997b).

According to social role theory, women’s changing social roles should create changes in stereotypes about women which should trickle down to how stereotypes are enacted in various interpersonal scenarios such as hypothetical first date scripts. The question at hand, however, is whether we can see this change beginning or whether this change is yet to come. On the one hand, longitudinal research has demonstrated attitudes toward women have been changing for some time (Twenge 1997a) and, thus, we may expect to see changes in first date scripts. On the other hand, the cultural lag hypothesis suggests that gender-based attitudes will change more slowly than will social roles (Diekman et al. 2010) and that gender stereotypes are robust (see Haines et al. 2016). Thus, it may simply be too early to see changes in first date scripts. Indeed, changes in first date scripts and other romantic relationship scripts have been particularly slow (England 2010).

First dates might also be particularly immune to societal changes in gender roles. Due to the evaluative nature and inherent uncertainly of a first date context, people may be particularly likely to take solace in traditional gender scripts (Eaton and Rose 2011). In other words, people might endorse and adhere to traditional gender-typed scripts in an attempt to ensure acceptance. There is evidence that acting in accordance with traditional gender scripts generally makes a positive impression and violating gender norms in this context results in a negative impression (Richardson et al. 1980). This broader trend is also evident for relationship initiation: Women who initiate dates are seen more negatively and as less attractive than women who do not initiate dates (Finkel and Eastwick 2009; Green and Sandos 1983). And despite reporting positive attitudes toward female-initiated dates, men are less likely to accept and continue relationships with women who initiate (Kelley et al. 1981).

In addition to risking rejection for failing to follow dating scripts, there are other personal costs to violating gender norms for a first date. First, people feel guilt and shame when they violate gender norms. For example, men report these negative emotions as a consequence of not paying for a first date (Lever et al. 2015). Second, women may actively avoid initiating a first date because doing so may create an undesirable impression. Men evaluate women who initiate dates as more sexually active and interested (Muehlenhard and Scardino 1985) and thus expect more sexual contact on first dates initiated by women (Mongeau and Carey 1996). Given men’s higher expectations for sexual activities on a first-date (Morr and Mongeau 2004), women may be especially cautious initiating dates to avoid creating such expectations that might ultimately lead to a date’s disappointment or even aggression. Because violating the script comes at a cost, people may not only be more likely to enact their prescribed gender role on first dates but to endorse the hypothetical first date script. Indeed, hypothetical first date scripts are more heavily influenced by gender stereotypes than actual first dates are (Rose and Frieze 1993), and, thus, may be even more resistant to evolving gender roles.

Taken together, there are strong social forces pushing for the maintenance of gender-typed first date scripts. Thus, we expected that first date scripts will still represent traditional gender stereotypes. However, given changing roles and attitudes toward women’s competence as agents and leaders, we predicted to see relatively weaker endorsement of gender stereotypic roles in first date scripts compared to data collected in the past. Because attitudes toward men’s gender role is slower to change than women’s, we suspected that any differences in first date scripts over time would likely involve seeing women as more likely and capable of engaging in behaviors traditionally prescribed to men.

To make these temporal cohort comparisons, we opted to compare our recently collected sample (in 2015) to that reported 15 years earlier by Laner and Ventrone (2000). We reasoned that the 15 years between these two samples would be enough time to observe change, even if it was relatively small or nuanced. Given that attitudes toward women are becoming more egalitarian since the 1970s (Twenge 1997a) and that there continues to be growing positive attitudes toward women’s social role, even between 2000 and 2010 (Donnelly et al. 2016), we expected to see a shift in the egalitarian nature of hypothetical first date scripts between 2000 and 2015, in line with social role theory (Eagly and Wood 2012).

The Role of Context in First Date Scripts

In addition to changing gender stereotypes and gender roles, the changing context of dating may make first date scripts particularly malleable. To understand how context matters, we must first review how first date scripts are typically assessed. In the standard paradigm for collecting hypothetical first date scripts, participants are asked to imagine a date between two people. The majority of research on first date scripts requires participants to imagine the first date between a man and a woman (for an exception, see Klinkenberg and Rose 1994). In some cases, no other information is provided (e.g., Laner and Ventrone 2000) and in others, participants are asked about a first date with someone new (e.g., Rose and Frieze 1989, 1993). Thus, the majority of research on first date scripts has focused on a first date between relative strangers.

A potentially important feature of the first date script, then, might be the context in which the imagined daters actually met: Are they strangers? Friends? Did they meet online? Context is particularly relevant for the study of first date scripts because dating has changed with the emergence and growth of online dating (see Aron 2012) and the increased frequency of platonic friendship between men and women (Monsour 2002). If new dating scripts are being written to accommodate these new contexts, they might be constructed to align with more egalitarian ideals wherein women and men are prescribed more equally active roles. Alternatively, new dating experiences may be less prone to diverge from gender-typed first date scripts because of their novelty and thus their greater uncertainty. People might be especially likely to adhere to gender scripts when the situation is more uncertain (Laner and Ventrone 1998), such as a first date resulting from an online dating service. If this is the case, when familiarity is low and, thus, uncertainty high, individuals should more strongly endorse gender-typed first date scripts. We reasoned that uncertainty should be at its peak when first dates involve two strangers, such as those who had met online. On the converse, uncertainty should be at its lowest when first dates involve a man and a woman who had been friends prior to the date wherein familiarity between daters is high, despite the notion of friends dating being a relatively new cultural phenomenon.

Currently, only two studies on hypothetical first date scripts have manipulated dating context. Both Laner and Ventrone (1998) and Morr Serewicz and Gale (2008) reported that first date scripts were remarkably consistent regardless of whether the daters were only acquaintances or had known each other well, either through being work colleagues or friends first. Neither study, however, pitted these familiar contexts with the context of online dating, which should feel particularly uncertain. Research on actual first date scripts might also support our expectation that hypothetical first dates among those who are already friends would be less gender-stereotyped. Although recent research on actual first dates has not directly studied familiarity before the first date, there is notable work comparing the scripts of first dates to hangouts and hookups (Eaton and Rose 2011; Eaton et al. 2016). All three types of scripts express traditional gender roles, although these gender stereotypes are strongest for dates. Presumably, hangouts and hookups are more likely to occur between individuals who are already familiar with each other, thus, this work also suggests that hypothetical first dates between friends will be more egalitarian than the typical first date script.

Gender, Gender-Based Attitudes, and First Date Scripts

If changing gender roles ultimately change gender-based stereotypes and scripts within a culture (see Eagly and Wood 2012) at the group level, we may also observe this change at the individual level. In other words, people who endorse more egalitarian beliefs (or at least less sexist beliefs) may also endorse more egalitarian (or at least less gender-typed) first date scripts. Relatedly, women and men, who traditionally differ in their endorsement of egalitarian and sexist beliefs (Glick et al. 2000; Twenge 1997a), may differentially endorse egalitarian first date scripts.

To capture these differences we measured individual differences in sexist beliefs as captured by attitudes toward women (Spence et al. 1973) and ambivalent sexism (Glick and Fiske 1996). Positive attitudes toward women represents more liberal and feminist ideologies whereas negative (i.e., low) attitudes toward women represent more conservative and sexist ideologies. Ambivalent sexism, however, comprises two forms of sexism: hostile and benevolent (Glick and Fiske 1996). Hostile sexism is closer to traditional measures of sexism, assessing attitudes of mistrust toward women and a desire to punish women who violate their assigned gender role. Benevolent sexism, on the surface, appears to benefit women by placing them on a pedestal and requiring men to behave in a chivalrous manner but ultimately these attitudes undermine women’s competence and agency. These two apparently distinct beliefs are strongly correlated (Glick et al. 2000), and together they serve to reinforce men’s social power (Lee et al. 2010). Ultimately, all of these beliefs support an ideology wherein women are deprived of social power and influence and thus are assigned a more passive role in romantic interactions and beyond.

Previous research on sexist beliefs reveals that these belief systems are indeed related to expectations and evaluations of dating practices. For example, people who endorse benevolent sexist beliefs are more likely to endorse double standards, such as men initiating and paying for dates (Paynter and Leaper 2016) and proposing marriage (Robnett and Leaper 2013). Indeed, the role of being a reactor to men’s advances in itself represents a translation of benevolent sexism (Paynter and Leaper 2016). People with more sexist attitudes judge individuals who violate gender norms more harshly (Richardson et al. 1980). Similarly, when men are portrayed as not adhering to their traditional gender role on a date, those higher in ambivalent sexism evaluate them more negatively (McCarty and Kelly 2015). Thus, we predicted that those participants higher in negative attitudes toward women and both hostile and benevolent sexism would endorse less egalitarian first date scripts.

The Present Research

We had four primary goals in the present research. First, we examined whether women and men would differentially endorse egalitarian female-male first-date scripts. We predicted that women would provide more egalitarian first date scripts than men would (Hypothesis 1) under the rationale that women generally have more positive attitudes toward women (Twenge 1997a) and hold less sexist attitudes (see Glick et al. 2000), although previous research on first date scripts exhibits mixed gender effects. If women more strongly endorse egalitarian first-date scripts than men do, then a man and a woman on a first date likely hold diverging beliefs about how the date should proceed. Such divergences will likely result in disappointment from women who expect more control over the date itself or frustration from men who expect to enjoy the privilege of control.

Second, we tested whether different relational contexts provide affordances for more egalitarian first date scripts. To accomplish this goal, we followed Morr Serewicz and Gale’s (2008) initiative by systematically varying the context of the first date. Thus, in both between-subjects studies, participants were randomly assigned to provide first date scripts for one of three contexts. One context described a man and a woman who had met online. A second context described a man and a woman who had been friends first. A third context followed a typical first date script procedure wherein no information was provided about how dating partners had met (Laner and Ventrone 2000). Despite the historical novelty of first dates between friends and those who meet online, we predicted that contexts that afforded greater familiarity and lower uncertainty between hypothetical daters (i.e., friends first) would result in more egalitarian first date scripts compared to contexts where familiarity was low and uncertainty high (i.e., online; Hypothesis 2).

Although previous research on hypothetical dates have investigated familiarity before the first date in terms of friendship or being work colleagues (e.g., Morr Serewicz and Gale) and work on actual initial romantic encounters have investigated different types of first engagement (e.g., hangouts, Eaton et al. 2016), we are the first known to systematically contrast meeting online compared to being friends prior to a hypothetical first date. As such, our work should illuminate the role of gender stereotypes in these two historically novel paths to a first date and begin to address whether friendship scripts provide a path to egalitarian romantic relationships between men and women as Eaton and Rose (2011) suggested.

Third, we directly examined whether individual differences in gender-based attitudes influence hypothetical first date scripts. Thus, we measured gender-based attitudes via assessments of attitudes toward women and ambivalent sexism. We predicted that people who endorsed more positive attitudes toward women (Hypothesis 3a) and less sexist beliefs (Hypothesis 3b) would also endorse more egalitarian first date scripts. Because some research finds that benevolent sexism is more predictive than hostile sexism for the types of chivalrous behaviors linked to men’s prescribed first date behaviors (e.g., opening the door for a date; paying the bill; Viki et al. 2003), we analyzed hostile and benevolent sexism separately. Although previous work has investigated specific dating practices (e.g., paying on a first date; Paynter and Leaper 2016), we are the first known to include measures of sexism and attitudes toward women in a study of hypothetical first date scripts with the intention of investigating how these attitudes influence overall expectations for such initial encounters.

Fourth, we addressed the degree to which contemporary first date scripts still adhere to traditional gender stereotypes. Although a recent selective review of the literature concluded that first date scripts have been remarkably stable over the last 25 years (see Eaton and Rose 2011), no known study to date has quantitatively compared hypothetical first date scripts across temporal cohorts using identical methodologies. Such an approach should be able to decipher the degree of gender-typed stability and to detect whether any change, even nuanced change, has occurred within hypothetical first date scripts.

To address this gap in the literature, we closely followed the methodology used by Laner and Ventrone (2000). We selected this particular study for three reasons. First, Laner and Ventrone used a checklist procedure, and the use of a checklist procedure allowed us to directly compare our data to their data. Second, because this procedure was fully articulated in the published article, we were confident we were correctly utilizing their procedures and thus would be able to compare our recently collected younger adult sample to the younger adult sample they reported in 2000. Third, there was at least 15 years between our recent sample and their published sample, which we reasoned was a large enough time period to observe some change, even if such change was relatively small or nuanced.

Previous research has reported the endurance and stability of first date scripts (see Eaton and Rose 2011), yet changing gender roles and more positive attitudes toward women (Donnelly et al. 2016) may be reflected in more recent first date scripts. Thus, we expected to see both stability and change. Specifically, we expected to see traditional gender stereotypes in the hypothetical heterosexual first date scripts collected in both 2000 and 2015, but we also predicted somewhat less endorsement of gender stereotypic roles in more recently collected first scripts (Hypothesis 4). By making such direct comparisons, we hope that our research addresses Risman’s (2009) call to researchers to study areas where traditional gender stereotypes might be unraveling. Even if such unraveling is only represented by the loosening of a few threads, such small effects might foretell larger future changes in the social norms regarding expectations for the first dates between men and women.

Method

Participants

We recruited two separate samples to address our research questions. One sample (younger adult sample) consisted of Canadian university students and the second sample (adult sample) consisted of older adults recruited through a crowdsourcing company. By doing so we hoped to capture diverse samples currently lacking in the study of first date scripts, thus answering the call to diversify this research topic (Mongeau et al. 2007).

Younger Adult Sample

Fully 809 introductory psychology students from a university located on the Canadian prairies participated in exchange for research credit toward their final course grade. Participants were required to be fluent in English and had to be living in North America for at least 2 years. In total, 53 participants were excluded from the analyses. The majority were excluded due to significant missing data (n = 20) or violating the requirement of living in North America for at least the last two years (n = 13). To ensure this sample was as similar in age range and life stage to Laner and Ventrone’s (2000) original sample, we also excluded participants over the age of 40 years (n = 14). Because binary gender as it relates to existing gender roles was a key component of our planned analyses, we opted to exclude the six participants who provided a non-binary gender (e.g., genderqueer). Thus, for all analyses, the sample consisted of 756 adults (372, 49%, men; 384, 51%, women), with an average age of 19.70 years (SD = 1.51, mdn = 18.00, range = 17–40). The sample was predominantly Canadian born (n = 565, 74.7%), with some racial/ethnic diversity: White (n = 418, 55.3%), Asian, (n = 168, 22%), Black (n = 46, 6.1%)). Most participants were single (n = 396, 52.4%) at the time of the survey, with the remaining participants reporting being exclusively dating (n = 189, 25.0%), casually dating (n = 76, 10.1%), cohabitating (n = 38, 5.0%), married (n = 21, 2.8%), engaged (n = 8, 1.1%), dating multiple people (n = 2, .3%), or separated (n = 3, .4%).

Older Adult Sample

Fully 860 adults were recruited through Crowdflower™ and received a small monetary compensation (40 cents USD) in appreciation of their time. Participants had to be fluent in English and living in North America for at least 2 years to ensure exposure to cultural scripts. In total, 39 participants were excluded from the initial analysis (12 for significant missing data; 8 for violating the requirement of living in North America for at least the last 2 years; 3 for non-binary gender; 16 for not indicating gender). Thus, our final sample comprised 821 adults (400, 49%, men; 421, 51%, women) with an average age of 34.70 years (SD = 12.72, mdn = 31, range = 18–80). The sample was predominantly U.S.-born (n = 612, 74.5%); Canada-born (n = 182, 22.2%), with greater racial ethnic diversity: White (n = 581, 70.8%), Asian, (n = 83, 10.1%), Hispanic (n = 69, 8.4%), Black (n = 48, 5.8%). Less than a third of the sample was single (n = 236, 28.7%) with the majority in a romantic relationship (n = 580, 70.6%) at the time of the survey. These participants reported being married (n = 285, 34.7%), cohabitating (n = 88, 10.7%), exclusively dating (n = 65, 7.9%), casually dating (n = 62, 7.6%), engaged (n = 54, 6.6%), dating multiple people (n = 7, .9%), divorced (n = 5, .6%) or widowed (n = 1, .1%).

Procedure and Measures

Participants in both samples completed an online survey about the expected behavior of men and women on first dates. Participants were randomly assigned to one of three first date scenario conditions: (a) unspecified, (b) online, or (c) friends-first. All participants were prompted with “Below is a list of behaviors that may or may not occur on a first date between a man and a woman.” However, the descriptions of the context varied by condition. Those in the unspecified condition were provided with no further information regarding the first date context, following the typical paradigm in previous research and, importantly, the same instructions as those provided to participants in Laner and Ventrone (2000). Participants in the online condition were asked to answer questions about a “first date between a man and a woman who met on an online dating site.” Those in the friends-first condition were asked to answer questions about a “first date between a man and a woman who have been good friends for some time before going on a romantic date.” All participants then completed the first date scripts questionnaire, followed by demographic questions. After concluding the survey, participants received compensation and a fuller description of the research hypotheses.

First Date Scripts Questionnaire

To measure participants’ first-date scripts, we used the list of first-date behaviors constructed by Laner and Ventrone (2000) based upon their reexamination of open-ended descriptions of first date scripts (Laner and Ventrone 1998). The list of 34 behaviors follows the typical temporal progression of a first date, starting with “ask someone for a date” and ending with “call friend to discuss date.” For each behavior, participants indicated who they believed would engage in that behavior using one of the four options (man, woman, either or both, or neither). Participants received the following instructions following the context prompt and prior to the list of 34 behaviors:

The behaviors are listed in a temporal order of events (i.e., the first thing that happens on this date is listed as #1, and the second as #2, and so on). Please indicate whether you think that each behavior would typically be something that the man, something that the woman, or something either the man or the woman would do during a heterosexual first date. If you feel that the behavior would not occur on the first date, select “neither.”

Thus, selecting “either or both” represents an egalitarian attitude, seeing the behavior as equally available to women and men. Selecting “man” or “woman” represents that the individual believes that the script is gendered whereby certain behaviors are only expected of men or women. Finally, selecting “neither” indicates that the individual believes the behavior does not belong in the first date script.

Demographics and Individual Differences

All participants completed a demographic questionnaire including gender (man, woman, or open-ended option), age, country of birth, ethnicity, language proficiency, and relationship status. The procedure for the two samples was identical, except participants in the younger adult sample completed two additional measures following the first date scripts questionnaire.

Ambivalent Sexism Inventory

To measure sexism, participants completed the 12-item short-form Ambivalent Sexism Inventory (ASI; Glick and Fiske 1996; see also Rollero et al. 2014). Participants indicated how much they agreed or disagreed using a 6-point scale from 0 (disagree strongly) to 5 (agree strongly) to items representing hostile sexism (e.g., “Once a woman gets a man to commit to her, she usually tries to put him on a tight leash”) and benevolent sexism (e.g., “Women should be cherished and protected by men”). Higher averaged scores represent greater sexist attitudes toward women, both benevolent (α = .80) and hostile (α = .87).

Attitudes toward Women Scale

To further measure attitudes toward women, participants completed the 25-item Attitudes Toward Women Scale (ATW; Spence et al. 1973). Participants indicated how much they agreed or disagreed with each statement (e.g., “Women should be given equal opportunity with men for apprenticeship in the various trades”) using a 6-point scale from 1 (agree strongly) to 6 (disagree strongly). Higher averaged scores indicate more positive attitudes toward women (α = .90).

Results

Demographic Comparisons between Samples

Our first step was to compare the demographic characteristics of the younger adult sample to the older adult sample. As we expected, the younger adult sample was younger (M = 19.70, SD = 3.39) than the older adult sample (M = 34.70, SD = 12.72), t(1545) = 31.19, p < .001, d = 1.61. Compared to the older adult sample, the younger adult sample also included more Canadians compared to those born in the United States or other countries, χ2(2) = 928.70, p < .001, φc = .77, as well as individuals who self-identified as White, χ2(1) = 40.60, p < .001, φc = .16, and Asian, χ2(1) = 43.15, p < .001, φc = .17. Lastly, a greater proportion of participants reported being single at the time of the survey in the younger adult sample compared to the older adult sample, χ2(1) = 89.84, p < .001, φc = .24. However, the proportion of men and women in each sample was similar, χ2(1) = .04, p = .847, φc = .02. Thus, any resulting differences in first date scripts between samples may also reflect a difference in age, citizenship, ethnicity, and current relationship status.

We next compared women and men within each sample on each demographic characteristic measured. Within the younger adult sample, men and women were similar in age, t(745) = .07, p = .942, d = .005, ethnic diversity, χ2(7) = 10.09, p = .184, φc = .12, and relationship status for those currently in a relationship, χ2(7) = 12.23, p = .093. φc = .13. However, women were less likely to be single (n = 185, 48.2%) than men (n = 211, 56.7%), χ2(1) = 5.53, p = .019, φc = .09. Within the older adult sample, men and women were only similar in terms of engagement in a relationship (i.e., single vs. in a relationship), χ2(1) = 3.14, p = .079, φc = .06. However, women were older (M = 37.45, SD = 13.27) than men (M = 31.79, SD = 11.43), t(798) = 6.45, p < .001, d = .46. Women were more likely to be White (n = 320, 76.0%) than men (n = 261, 65.6%), χ2(1) = 10.80, p = .001, φc = 11., and men more likely to be Native American or Hispanic (n = 19, 4.8% and n = 48, 12.1%, respectively) than women (n = 2, .5% and n = 21, 5.0%, respectively): χ2(1) = 15.13, p < .001, φc = .14 and χ2(1) = 13.26, p < .001, φc = .13, respectively. Among those currently in a romantic relationship, women were more likely to be married (n = 167, 53.9%) than men (n = 118, 43.2%), χ2(1) = 8.89, p = .003, φc = .10, but less likely to be engaged (n = 16, 5.1%) than men (n = 38, 13.9%), χ2(1) = 11.09, p = .001, φc = .12.

Testing Hypotheses 1–3

The Egalitarian Nature of Current First Date Scripts

To test our first two predictions (Hypothesis 1 and Hypothesis 2), we first created an egalitarianism score for each participant. To calculate the egalitarianism score, we simply counted the number of behaviors to which the participant had responded with “either or both.” Thus, possible egalitarianism scores could range from zero (minimum), meaning that the participant believed that all behaviors in the list were gendered or irrelevant, to 34 (maximum), meaning that the participant believed that all behaviors in the list were non-gendered and thus could be exhibited by either the woman and or the man in the presented scenario. We then conducted a three-way Analysis of Variance (ANOVA) with Condition (unspecified vs. online vs. friends-first), Participants’ Gender (men vs. women), and Sample (Younger Adult vs. Older Adult) as between-subjects factors on egalitarianism scores. The variable, Sample, was included for exploratory purposes.

A main effect of participants’ gender would support our first prediction that women would endorse more egalitarian dating scripts (Hypothesis 1). Results revealed the expected pattern: Women’s first date scripts were more egalitarian (M = 17.25, SE = .25) than men’s (M = 15.71, SE = .26), F(1, 1565) = 18.33, p < .001, ηp2 = .012. However, this main effect of gender was qualified by an interaction between Gender and Sample, F(1, 1565) = 7.50, p = .006, ηp2 = .006. Decomposing the interaction revealed that in the older adult sample, women provided more egalitarian first date scripts (M = 17.82, SE = .35) compared to men (M = 15.29, SE = .36), F(1, 1565) = 25.18, p < .001, ηp2 = .025, whereas there was no difference in the egalitarian nature of the first date scripts endorsed by women (M = 16.68, SE = .37) and men (M = 16.13, SE = .37) in the younger adult sample, F (1, 1565) = 1.03,p = .310 ηp2 = .002. Thus Hypothesis 1 was supported for the older adult sample but not for the younger adult sample.

The results revealed a main effect of Condition, F(2, 1565) = 7.48, p = .001, ηp2 = .009. As expected, participants in the online condition (M = 15.65, SE = .31) and those in the unspecified condition (M = 16.42, SE = .31) produced similarly egalitarian scripts (p = .081, d = .11). However, participants in the friends-first condition (M = 17.36, SE = .47) produced more egalitarian scripts than those in both the online (p < .001, d = .24) and the unspecified conditions (p = .034, d = .13), supporting Hypothesis 2. Thus, friends-first dating scripts were more egalitarian than scripts for dating pairs who had met online and when no context was provided (unspecified condition). No other effects or interactions emerged, suggesting that date context was not moderated by the sample nor participants’ gender.

Note that controlling for the variables that differed between the younger adult and older adult samples (i.e., citizenship, White and Asian ethnicity, and relationship engagement) and controlling for the demographic variables that differed between men and women in the younger adult (i.e., relationship engagement) and adult (i.e., age, White ethnicity, Native American/Indigenous ethnicity, Hispanic ethnicity, and married and engaged relationship status) samples did not change the pattern of significant effects. See the online supplemental file for further details.

We also conducted a cognitive script analysis of the 34 behaviors divided by gender, sample, and dating context. Across participants’ gender, sample, and date context, participants endorsed all 34 script actions as part of all first date scripts (using the 50% endorsement criteria for checklist methodologies; Eagly and Wood 2012). Thus, despite contextual differences in overall egalitarianism, the basic content of a hypothetical first date script is the same. (See Tables 1s, 2s, and 3s in the online supplement.)

Table 1 Percentage of participants indicating “man” for Men’s prescribed behaviors and “woman” for Women’s prescribed behaviors in Laner and Ventrone’s 2000 (younger adult sample) and our 2015 (younger adult sample – unspecified condition)
Full size table
Table 2 Response frequency for men’s and women’s prescribed behaviors in 2000 and 2015 (younger adult sample – unspecified condition)
Full size table
Table 3 Frequency of responses for Men’s and Women’s prescribed behaviors in 2000 and 2015 (younger adult sample – Unspecified condition) by participants’ gender
Full size table

The Influence of Attitudes on First Date Scripts (Younger Adult Sample Only)

To test our hypotheses that younger adults holding more positive attitudes toward women (Hypothesis 3a) and less sexist attitudes (Hypothesis 3b) would endorse more egalitarian first date scripts, we conducted a hierarchical multiple regression entering all three mean-centered predictors (i.e., attitudes toward women, hostile sexism, benevolent sexism) in the first step to predict egalitarian scores for our younger adult sample. To explore whether gender and date context moderated the influence of these individual differences on egalitarian scores, we had also included gender (0 = women; 1 = men) and date context (−1 = friends first; 0 = unspecified; 1 = online) in the first step. We subsequently included all two-way interactions between gender and each of the three individual differences and between date context and each of the three individual differences in the second step. In the third step, we included all possible three-way interactions among gender, date context, and each of the three individual differences.

Results revealed support for our predictions in the first step (R2 = .37), F(5, 741) = 23.00, p < .001. In addition to a date context main effect in line with the previously reported main effect (β = −.08, t = −2.25, p = .025), people who endorsed more positive attitudes toward women (Hypothesis 3a) (β = .14, t = 3.46, p = .001), less hostilely sexist beliefs (Hypothesis 3b) (β = −.09, t = −2.32, p = .020), and less benevolently sexist beliefs (Hypothesis 3b) (β = −.22, t = −5.59, p < .001) thought that more of the first date behaviors were equally likely to be enacted by a woman or a man. Gender did not predict egalitarian scores nor did gender or date context moderate the association between any of the individual differences and the egalitarian nature of first date scripts (ts < 1.5, ps > .070).

Testing Hypotheses 4 (Younger Adult Sample)

Comparing Younger Adult Samples

Before testing our fourth prediction regarding cross-sectional differences across our recent cohort (2015) and the 2000 cohort, we first compared the demographic characteristics of our younger adult sample (2015; unspecified condition) to the younger adult sample used in Laner and Ventrone (2000). Because our comparisons for cross-sectional differences focus only on the unspecified condition in our younger adult sample (n = 247), we focus only on this condition in our demographic contrasts between the 2015 and 2000 samples. Our younger adult sample (unspecified condition) was younger (M = 19.74 years, SD = 3.49) than the sample described in Laner and Ventrone (M = 23.19 years), t(436) = −10.48, p < .001. This result is perhaps not surprising given that our sample was recruited from a first-year cohort and Laner and Ventrone’s was recruited from upper-level courses. Our parallel sample also comprised greater ethnic diversity with only 55% of our sample identifying as White whereas 90% of Laner and Ventrone’s sample identified as White (z = 8.40, p < .001). Of course, our parallel sample was predominantly Canadian-born (73.7%) whereas we presume that Laner and Ventrone’s sample, recruited from a large U.S. Southwestern university, was predominantly American-born. Thus, any resulting differences for first date scripts between samples might not only reflect cohort or temporal differences, but also differences in country of origin, ethnicity and age (see online supplemental file for further discussion and analysis).

Gender Prescriptions in Hypothetical First Date Scripts in 2000 and 2015

To quantify the degree of stability and to test our hypothesis that hypothetical first date scripts are less gender stereotyped in 2015 (Hypothesis 4), we compared our younger adult sample (unspecified condition only) to the data reported in Laner and Ventrone (2000). To accomplish this comparison, we calculated two sets of scores in our own dataset and from the data reported in Laner and Ventrone. Because we did not have Laner and Ventrone’s dataset and relied on the frequency table reported in their 2000 article, we calculated all scores from the overall frequencies of endorsement of each of the four possible answers (man, woman, either or both, neither) to each of the 34 script actions, both collapsing across men and women and then again separately for women and men. In other words, we first tested whether the 2015 younger adult sample was more egalitarian than the 2000 sample regardless of gender. We then tested whether both men and women were more egalitarian in 2015 compared to the sample reported in 2000 and whether any gender differences evident in 2000 persisted in 2015.

Endorsing Traditionally Gender-Prescribed Behavior

To calculate endorsement of men’s gender roles in first date scripts (men’s prescribed behavior), we first identified the first date behaviors traditionally prescribed to men or the behaviors for which at least 50% of participants (collapsed across gender) in Laner and Ventrone’s (2000) original study responded with “man.” This resulted in a list of 14 behaviors (e.g., ask someone for a date, pay the bill). As illustrated in Table 1, in our own sample, 10 of these behaviors were prescribed to men, meaning that 50% or more of our recent younger adult sample indicated that 10 of the original 14 behaviors were still a man’s responsibility. Dividing by participants’ gender, men and women similarly identified 11 and 10 behaviors, respectively, as currently prescribed to men.

We followed the same procedure for identifying behaviors traditionally prescribed to women (women’s prescribed behavior), whereby over 50% of participants in Laner and Ventrone’s (2000) original sample responded with “woman.” This resulted in a list of seven behaviors (e.g., wait to be asked). Comparatively, in our own sample, for six of these behaviors, 50% or more of our recent sample indicated that six of the original seven behaviors were still a woman’s responsibility (see Table 1). Dividing by participants’ gender, men identified only half of the behaviors (n = 3) that women identified as currently prescribed to women (n = 6).

Overall, compared to Laner and Ventrone’s (2000) younger adult sample, the majority of behaviors previously prescribed to men are still prescribed to men (71%) and the majority of behaviors traditionally prescribed to women are still prescribed to women (86%). However, there appears to be a weakening of the traditional gender stereotypes in hypothetical first date scripts for the behaviors traditionally identified as the man’s responsibility.

Indeed, contrasting the average frequency of endorsement of “man” for behaviors traditionally prescribed to men and of “woman” for behaviors traditionally prescribed to women further revealed a difference across temporal cohorts, collapsed across participants’ gender, in hypothetical first date scripts (see Table 1). Only four of the 14 actions prescribed to men (29%) were similarly endorsed by the 2000 and 2015 younger adult samples. Indeed, for the majority of actions (71%), the 2015 younger adult sample was less likely to select “man” as the one responsible for these actions traditionally prescribed to men. A similar pattern emerged for behaviors traditionally prescribed to women, although not as pronounced. Here, three of the seven actions traditionally prescribed to women (43%) were similarly endorsed as the woman’s responsibility in 2000 and 2015. However, for a slim majority of actions (57%), the 2015 younger adult sample was less likely to select “woman” as the one responsible for these actions. So far, our analysis supports both the persistence of gender stereotypes in hypothetical first date scripts and a shift in the endorsement of these stereotypical gender roles in the more recent sample.

Response Option Frequencies for Prescribed Behavior

Next, we averaged the total frequencies within each of the four response options (man, woman, either or both, neither) across the 14 items for men’s prescribed behavior and separately for the seven items reflecting women’s prescribed behavior. This process resulted in two sets of scores reflecting the average frequency for each response option for men’s and women’s prescribed behavior from Laner and Ventrone’s (2000) data and our own younger adult sample. We then calculated an overall score across all 34 behaviors representing the average frequency of the “either or both” response for all items to directly test whether participants saw all first date behaviors as egalitarian for both genders. These scores allowed us to conduct more nuanced tests regarding the degree to which each younger cohort (2000 and 2015) supported gender-typed versus egalitarian prescriptions for the hypothetical first date across all three divisions of behaviors.

To statistically test whether the two cohorts differed in egalitarian responses (Hypothesis 4), we utilized Chi-square tests. As a first step, we focused on identifying the most frequent response for men’s and women’s prescribed behavior. In Laner and Ventrone’s original 2000 younger adult sample, most participants indicated that men should engage in men’s prescribed behavior, χ2(3) = 127.76, p < .001, φc = .79 (see Table 2a), and that women should engage in women’s prescribed behavior, χ2(3) = 128.96, p < .001, φc = .80 (see Table 2b). Moreover, most participants in our own 2015 young adult unspecified condition also indicated that men should engage in men’s prescribed behavior, χ2(3) = 89.23, p < .001, φc = .60 (see Table 2a), and that women should engage in women’s prescribed behavior, χ2(3) = 74.41, p < .001, φc = .55 (see Table 2b). Thus, even in 2015, first date scripts align with traditional gender stereotypes. However, the strength of these effects appear stronger for Laner and Ventrone’s sample compared to our own recent younger adult sample, suggesting that despite the remarkable stability in the first date script, traditional gender stereotypes might be starting to weaken within these hypothetical first date scripts.

As a second step in testing Hypothesis 4, we directly compared our recent younger adult sample to Laner and Ventrone’s (2000) using z-statistics for two population proportions. Targeted analysis comparing response options across the two cohorts revealed that our participants were less likely to assign men to behaviors traditionally prescribed to men (z = −2.76, p = .006) and were instead more likely to select the egalitarian option for the set of behaviors traditionally prescribed to men (z = −3.51, p < .001) (see Table 2a for frequencies). For behaviors traditionally prescribed to women, our recent sample was again less likely to assign women to these behaviors (z = −3.55, p < .001), and more likely to select the egalitarian option (z = −3.28, p = .001) compared to the 2000 cohort (see Table 2b).

Investigating the overall average frequencies across all 34 behaviors, we found a similar pattern as illustrated in Table 2c. The most frequent response across all first date behaviors was “man” followed by “either or both” in Laner and Ventrone’s (2000) study, χ2(3) = 26.64, p < .001, φc = .36. The pattern was markedly different in our recent sample whereby the most frequent response was the egalitarian option followed by “man,” χ2(3) = 41.07, p < .001, φc = 41. Direct comparisons between the 2000 and 2015 younger adult samples revealed that the recent sample was indeed more egalitarian than the 2000 sample (z = −2.85, p = .004). As predicted, despite our own evidence of the enduring longevity of gender stereotypes in first date scripts, gender scripts for hypothetical first dates are indeed less gender-stereotyped in the 2015 cohort compared to the 2000 cohort, supporting Hypothesis 4. Of course, these differences do not reflect radical departures from traditional gender roles, but they do reveal a detectable shift in the expectations for women and men on a hypothetical first date.

Differences across Time

Are the hypothetical scripts provided by both the men and women in our 2015 sample more egalitarian (or at least less gender stereotyped) than those provided by the men and women in 2000? Yes, when collapsing across the behaviors traditionally prescribed to either women or men. Both women (z = 2.37, p = .018) and men (z = 2.68, p = .007) in our 2015 younger adult sample were less likely to indicate that women’s prescribed behavior should be enacted by the woman than were participants the 2000 sample (see Table 3b). However, only men were less likely to endorse that men’s prescribed behavior should be the responsibility of the man in our 2015 younger adult sample compared to the 2000 younger adult sample (z = 2.01, p = .044). Women were equally likely to prescribe the total of these behaviors to men in 2000 as in 2015 (z = 1.90, p = .057) (see Table 3a). Although these contrasts were less powerful due to splitting the sample size in half, both men and women seem to be shifting toward endorsing less gender-stereotypical roles in hypothetical first date scripts.

We next explored each of the individual behaviors traditionally prescribed to either men or women because some behaviors are more important in advancing the date (e.g., “ask someone for a date”) and more likely to create conflict through a sense of entitlement or obligation (e.g., “pay the [first] bill”). As illustrated in Table 4a, women in 2015 were less likely to select “man” for behaviors prescribed to men compared the women in Laner and Ventrone’s (2000) sample for nine of the 14 behaviors (64%). Similarly, men in 2015 were also less likely to select “man” for these behaviors for seven of the 14 behaviors (50%). Thus, although the responses of both men and women in 2015 were very similar to those in 2000 for some script behaviors (e.g., “pick up date”), they were different for other behaviors believed to be essential to the power structure of a first date (e.g., “decide on plans by yourself”; “pay the [first] bill”; “make affectionate move”). As depicted in Table 4b, the behaviors traditionally prescribed to women revealed a similar pattern: Both men and women in 2015 were less likely to select “woman” for four of these seven behaviors (57%), and in addition, men in 2015 were less likely to prescribe “go to the bathroom to primp” to women. Thus, for behaviors prescribed to men and women, we see both women and men providing less gender-typed hypothetical scripts.

Table 4 Percentage of participants indicating (a) “man” for men’s prescribed behaviors and (b) “woman” for women’s prescribed behaviors in Laner and Ventrone’s 2000 (younger adult sample) and 2015 (younger adult sample – unspecified condition) by participants’ gender
Full size table

Do the Gender Differences Evident in 2000 Exist in our 2015 Younger Adult Sample?

To address this question, we first had to test for gender differences in the Laner and Ventrone (2000) sample in addition to testing for gender differences in our own parallel 2015 sample. Again, we collapsed across the behaviors prescribed to men or women in the traditional hypothetical first date script. As depicted in Table 3, when collapsing across gender-typed behaviors, men and women in both 2000 and 2015 did not differ in their likelihood of prescribing women and men to gender-typed behaviors.

Next, we addressed this question for individual behaviors because previous research for certain script components (e.g., “paying the bill”; Lever et al. 2015) exhibits gender differences. Here, we contrasted the proportion of men and women who endorsed “man” for each of the men’s prescribed behaviors and then the proportion of men and women who endorsed “woman” for each of women’s prescribed behaviors, separately for Laner and Ventrone’s (2000) younger adult sample and our own younger adult sample (unspecified condition). As illustrated in Table 4a, most of the gender differences persisted for men’s prescribed behavior across the two temporal cohorts. In Laner and Ventrone’s sample, compared to women, men were more likely to endorse “man” for seven of the behaviors traditionally prescribed to men (50%). However, for our recent younger adult sample, this gender difference disappeared for two of these seven behaviors, including the essential script component of “asking someone for a date.” For behaviors traditionally prescribed to women, men and women did not differ in their stereotypical endorsement of “woman” in Laner and Ventrone’s sample (see Table 4b). Unexpectedly, in our own younger adult sample, compared to men, women were more likely to endorse “woman” for three of the behaviors (buy new clothes; go to the bathroom to primp; call a friend to discuss date). Because all of these behaviors can occur without the awareness of the date, such endorsements might not only reflect adherence to gender roles but egocentric biases. In other words, women may not be aware of men’s increased participation in these traditionally feminine aspects of the first date script among our recent younger sample.

Together, our analyses suggest that despite enduring gender differences in the endorsement of traditional gender roles in hypothetical male/female first date scripts, both men and women are more egalitarian (or at least less gender-stereotyped) in their endorsement of specific behaviors in our 2015 sample. For instance, although men are more likely in 2000 and in 2015 to assign “man” to “decide on plans by yourself” and “pay the [first] bill,” both women and men are less likely to assign men to these tasks in 2015 than in 2000. Thus, despite the persistence of traditional gender roles, their endurance in hypothetical first date scripts seem to be weakening.

Discussion

First date scripts are one of the many ways traditional gender roles are expressed within a given society. A cognitive scripts analysis of our two samples revealed that all of the hypothetical first date script actions identified over 15 years ago are still strong components of the first date script, even among those who met online and those who were friends first. Despite the consensus in the content of a first date between a woman and a man, participants’ gender, sample age, and relational context prior to the hypothetical first date influenced notions of what men and women would do on a first date. Compared to men, women endorsed more egalitarian first date scripts in our older adult sample but not in our younger adult sample. Although one might be tempted to interpret this effect, or lack thereof, as indicating that our younger sample holds more egalitarian attitudes about dating than our older sample does, this was not the case. Although previous research has suggested that single adults place more emphasis on traditional components of the first date script than younger samples (Mongeau et al. 2007) and that more experienced daters endorse more traditional gender roles on first dates (Rose and Frieze 1989), our younger and older samples were similarly egalitarian in their first date scripts when collapsing across date contexts.

We hope that the lack of a gender difference in the younger sample reflects an equalizing of gender-based attitudes among men and women. However, such effects may simply reflect inconsistencies already present in the first date script literature. Some studies report that women endorse more egalitarian scripts (e.g., Eaton and Rose 2011), others find the reverse (e.g., Lever et al. 2015; Paynter and Leaper 2016), and still others find no gender differences (e.g., Rose and Frieze 1993). However, the previously reported diverse findings could be driven by differences in the influence of relational contexts prior to a first date. First dates between a woman and a man who had been friends first were seen as more egalitarian than dates between those who had met online and when dating context was left unspecified. Echoing previous research, when initial romantic interactions deviate from the traditional first date, people are less apt to rely on traditional gender norms to script behavior (see Eaton et al. 2016). Our results suggest that familiarity between daters can loosen traditional gender stereotypes, at least within the context of a hypothetical first date.

In the present research, gender-based attitudes also influenced the endorsement of egalitarian first date scripts. People who held more positive attitudes toward women and less sexist attitudes (both benevolent and hostile) were more likely to see first date behavior as equally likely to be expressed by men or women. Thus, gender-based attitudes are associated with expectations for how interactions between men and women should play out, not only in ongoing romantic relationships between women and men (see Hammond and Overall 2017) but also in the very first date.

Did hypothetical first date scripts among younger adults differ between the cohort collected in 2000 by Laner and Ventrone and the one we collected in 2015? Yes and no. In support of the stability of the hypothetical first date script, all components of the first date script endured in the more recent sample. Further, participants in both cohorts were more likely to ascribe more behaviors to men than to women and, specifically, to ascribe men to the more direct behavior and to ascribe women to the more passive behaviors in both samples. Hence, our younger adult sample still endorsed a gendered perspective of the behaviors appropriate for men and women on a first date, with certain behaviors similarly endorsed as men’s responsibility (e.g., “pick up date”, “open doors”) and as women’s responsibility (e.g., “wait to be asked”, “go to the bathroom to primp”).

However, there also was evidence for changing endorsement of the traditional first date script. In support of change, or at least the weakening of traditional gender stereotypes in hypothetical first date scripts, fewer behaviors were prescribed solely to men; fewer participants in our 2015 younger sample endorsed gender-typed hypothetical script; and there was a greater reliance on indicating that either men or women could enact a greater variety of behaviors, especially ones traditionally ascribed to men. For example, fewer young adults in 2015 supported the notion that men should be responsible for asking for the first date, deciding on the plans, and making an affectionate move. Thus, despite persistent gender roles, the expectations for the behaviors of women and men on a hypothetical first date are becoming more egalitarian than 15 years ago. Although women were less likely to provide gender-stereotyped first date scripts for several script behaviors, both men and women in our parallel young adult 2015 sample were less gender-typed than the men and women in 2000 were. Granted, this difference is not radical, and by no means do our results indicate that traditional gender roles are a “thing of the past” in first date scripts. However, our results suggest that the endorsement of traditional gender roles in the hypothetical first date scripts are weakening.

A close inspection of our data suggests that this difference in first date scripts across time is being driven by women’s changing role. Compared to the past, women perceive themselves as more agentic (Twenge 1997b), women assume more roles traditionally reserved for men (Leopold et al. 2016), and attitudes toward women are becoming more egalitarian (Donnelly et al. 2016; Twenge 1997a). In hypothetical first date scripts, we similarly see that behaviors traditionally prescribed to men are increasingly being seen as appropriate for both men and women (e.g., “make affectionate move”) and fewer behaviors are prescribed solely to men than in the past, especially among first dates that occur between friends. However, there are fewer differences in behaviors traditionally prescribed to women and, thus, the increasing egalitarianism in first date scripts appears lopsided.

Strengths, Limitations, and Future Research Directions

Our research boasts many strengths. First, we are the first known to directly compare a recent sample with a past sample using the same hypothetical first date script measures and metrics. In doing so, we are the first known to quantitatively address whether first date scripts differ across temporal cohorts by using a methodology able to detect relative temporal shifts. Second, although others have examined the link between sexist beliefs and specific behaviors within the script (e.g., paying for the date; Paynter and Leaper 2016), we are the first known to investigate the association between sexist beliefs and the whole first date script. As such, our work adds to the growing body of research examining how sexist attitudes influence intimate relationships between men and women (see Hammond and Overall 2017). Third, we are the first known to contrast hypothetical first date scripts for daters meeting online to those who were friends first. As such, our work contributes information on first date expectations to the growing literature on different types of actual first initial romantic encounters (Eaton and Rose 2011; Eaton et al. 2016). Fourth, we recruited two large samples through two different means, allowing us to compare and generalize our results to a larger population. Indeed, our use of a crowdsourced sample likely produced a more diverse sample than typically recruited through undergraduate populations (see Buhrmester et al. 2011).

Despite our study’s many strengths, some of our design choices might be construed as limitations. Our large and diverse samples might also be perceived as a weakness. Specifically, by collecting our samples from different locations and confounding location with age, we were unable to decipher whether location (Canada vs. United States) or age/cohort was the important factor behind the differing influence of gender observed between the two samples. Relatedly, our two samples were not perfect comparison points to the sample collected by Laner and Ventrone (2000). First, our adult sample was roughly 15 years older than the average person in Laner and Ventrone’s sample collected 15 years earlier, meaning that these two samples represented the same generation. Thus, we opted not to compare the adult sample to Laner and Ventrone’s sample because we could not be confident any resulting differences would reflect differences across cohorts (i.e., changing social roles) or simply maturation. Second, our younger adult sample was predominantly Canadian, more ethnically diverse, and younger than the sample used in Laner and Ventrone’s study. Thus, any resulting differences across cohorts might reflect not only temporal differences, but also differences in country of origin, ethnicity, and age. We encourage future researchers to return to this research question in another 15 years with more comparable samples.

In addition to our choice of sample, our choice of measures might also be seen as a weakness. We opted to compare our more recently collected samples to those of Laner and Ventrone (2000) not only because it reflected a comprehensive investigation of hypothetical first date scripts 15 years earlier but because the authors had clearly reported a closed-ended measure that we could easily adopt and then directly compare both datasets. However, in making such a choice we did not collect open-ended data which may have provided a different window into first date scripts. Future research should revisit the components of the hypothetical first date script by collecting open-ended data.

Finally, our analysis focused on hypothetical first date scripts and, as such, we cannot speak to how gender, attitudes, and context influences actual dating behavior. Despite evidence of the weakening of the traditional gender stereotype in hypothetical first date scripts, actual first date behavior might be as gender-typed as ever. Our work on hypothetical first date scripts suggests that sexism and attitudes toward women influence how the first date is expected to unfold, and as such future researchers may be wise to investigate the role of these beliefs during actual relationship initiation. Not only would investigating whether sexism is associated with actual relationship initiation or first date behavior be empirically interesting but it also might be practically important for those actively dating.

Practice Implications

Although changes in social roles lead to changes in gender stereotypes that eventually trickle down to changes in interpersonal scripts (social role theory; Eagly and Wood 2012), our results suggest that this change is slow for hypothetical first date scripts. Despite a loosening of the grip of traditional gender roles on expected first date behaviors, these scripts are still dominated by traditional gender roles. Men are still expected to do the lion’s share of the active role whereas women are given the less powerful role of “reactor” to men’s “moves.” Although resistance to change in first date scripts has sometimes been interpreted as women trying to maintain “female privilege” (Lever et al. 2015, p. 11), being assigned the passive role means that women will experience little control over the events of a first date. Not only might such roles force individuals to act contrary to their own dispositions (i.e., be inauthentic), but also the enactment of traditional gender roles in these initial romantic experiences might set an unwelcome foundation for any unfolding romantic relationship and subsequent future interactions (see also Laner and Ventrone 2000). Reassuringly, interventions that enhance women’s personal control encourage women to take the active, instead of reactive, role during relationship initiation (MacGregor and Cavallo 2011) and, thus, may encourage a more egalitarian structure in any developing relationship.

Potential negative consequences of gendered first date scripts may also play out in different expectations among men and women. Men were more likely to endorse traditional gender roles in first date scripts than were women in our older adult sample and were more likely to endorse traditional gender roles for certain behaviors traditionally prescribed to men (e.g., “get money”; “collect keys”) in our younger adult sample (unspecified condition). If men are more likely to expect, and presumably then adopt, the dominant role, then women may be forced into a more submissive role. Although men and women in our younger sample endorsed egalitarianism to the same degree when looking at the whole first date script collapsed across date contexts, there were some notable differences on individual behaviors. Specifically, among our younger adult sample (collapsed across date contexts), women were more likely to endorse an egalitarian response for “pay bill” than were men, replicating other recent research (Lever et al. 2015). If men are expected to pay for the date, some men may interpret such payment as entitlement for sexual acts and some women may feel indebted to provide sexual favors (see also Eaton et al. 2016). Such expectations not only may create tension between men and women but also may exacerbate anxiety, especially among those already fearful of relationship initiation.

Practitioners and first daters should also consider the importance of individual differences in expectations for the first date. In our younger adult sample, those who endorsed more sexist beliefs also more strongly endorsed the traditional first date script. Presumably such beliefs would continue to influence any developing romantic relationship (see Hammond and Overall 2017). First daters who hold more egalitarian or feminist ideals may be wary of individuals who wish to strictly follow traditional gender roles on first dates because such behavior may betray their sexist inclinations. Women and men who hold more egalitarian ideals may wish to actively reject the traditional first date script not only to convey their belief system but also to set a foundation of egalitarianism for any developing relationship. Our data on hypothetical first date scripts suggests that forming romantic relationships from friendships might be one way to encourage a more egalitarian first date (see Eaton and Rose 2011, for a similar suggestion).

Conclusion

At both societal and personal levels, sexist ideologies and adherence to traditional gender stereotypes influence hypothetical first date scripts between women and men. Among our older sample, men were less likely to endorse egalitarian roles in their expectations for a first date compared to women. Furthermore, those who adhered to more sexist beliefs, and thus undermined the agency of women, were more likely to endorse first date scripts wherein men took the active role and women the reactive role. Despite the remarkable endurance of traditional gender roles, comparison of our diverse younger adult sample from 2015 to the younger adult sample reported by Laner and Ventrone in 2000 revealed weaker gender prescriptions in the hypothetical first date scripts reported in 2015. In line with social role theory (Eagly and Wood 2012), as gender roles evolve, first date scripts between women and men may also be rewritten and, hopefully, provide a more egalitarian foundation for developing romantic relationships. Indeed, relational context prior to the first date might reveal the most readily egalitarian path to romantic relationship initiation. Hypothetical first dates among individuals who were originally friends and, thus, where familiarity is high were the most egalitarian and, thus, friendship may deliver the elusive egalitarian foundation for romantic relations between women and men.