Bazzini, McIntosh, Smith, Cook and Harris (1997) examined the 100 top-grossing movies from the 1940s to the 1980s and showed that older female characters were underrepresented and more negatively portrayed than their male counterparts. Similarly, Lauzen and Dozier (2005a) analyzed the 100 top-grossing domestic movies of 2002 and found a double standard of aging. Male characters clearly outnumbered female characters (73% vs. 27%), and the majority of male characters were in their 30s and 40s, while the majority of female characters were in their 20s and 30s. Older people of both genders were underrepresented compared to the demographic reality. For male characters, leadership and occupational power increased with age, while older female characters were less likely to have such power or goals. A more recent study (Smith et al., 2014) reported that in 11 different countries, there was a lack of women over 40 in movies compared to their share in society.

A study by Markson and Taylor (1993) analyzed the ages of 1,169 actresses and actors who were nominated in four award categories (best actress/actor and best supporting actress/actor) for the Academy Awards from 1927 until 1990. This research showed that people over 29 accounted for only 27% of winners for best actress and 67% for best actor. A similar study by Lincoln and Allen (2004) confirmed a significant negative effect of being female and of being older on the number of film roles received by actors—this effect is referred to as “double jeopardy” for older women. Markson and Taylor (2000) summarized their findings on movies in the following way: “Despite changing gender roles in later life since the 1930s and despite social and economic changes for older Americans (earlier retirement age and better health are but two examples), their film roles have remained remarkably static in age and gender stereotyping” (p. 137).

This is not to deny that there are cases of older women in movies. For example, in the Bond movie “Spectre” (2015) Monica Bellucci was 51 years old and was the oldest “Bond Girl” in any of the Bond movies. However, what some call a “revolutionary treatment of women” in that movie has to also be seen in perspective. On the one hand, 51 is actually not that old. Indeed, several articles mention that nobody ever wrote that Daniel Craig was an “older man” playing James Bond despite being a similar age. Besides that, her appearance in the movie is extremely short, so one might wonder if this was not primarily a publicity stunt without much meaning. In addition, Monica Bellucci represents the same stereotypical woman that has been used throughout the history of movies; she is a sex symbol and certainly not someone with whom the average older woman can identify. Last but not least, there have always been representations of older women throughout the history of television and movies (Lemish & Muhlbauer, 2012; Walsh, 1989). However, only the future will tell if these representations are exceptions or will become more the norm.

Similar results are found in fictional television content during primetime television. Women have generally been younger, and men have outnumbered women in the older age segment throughout the decades (Gerbner, 1993; Signorielli & Bacue, 1999). This pattern was true in the 1960s, 1970s, 1980s, 1990s, and 2000s (Aronoff, 1974; Gerbner, 1993; Glascock, 2001; Lauzen & Dozier, 2005b; J. D. Robinson & Skill, 1995; Signorielli & Bacue, 1999; Vernon, Williams, Phillips, & Wilson, 1990). For example, there are 3.35 male characters for each female character among middle-aged characters and 2.04 male for each female among older characters (65+) (Vernon et al., 1990). The underrepresentation of older people and particularly older women during primetime is also seen in other parts of the world, such as Germany and Taiwan (Kessler, Rakoczy, & Staudinger, 2004; Lien, Zhang, & Hummert, 2009). Gerbner (1993) explained that the “character population is structured to provide a relative abundance of younger women for older men, but no such abundance of younger men for older women” (p. 211). This difference was also recognized by focus group members in New Zealand, who argued that older men can have many young women in the media, but the reverse is not acceptable (Vares, 2009). The reason for the presence of young women and the absence of older women might be the continued power and privilege of men within media institutional structures, as suggested in a report in the UK (Dolan & Tincknell, 2013).

In addition to numerical gender differences on primetime television, previous research has also found that older men are shown to have more desirable characteristics (e.g., active, healthy, and socially involved), whereas older women are shown with more undesirable characteristics (e.g., low intelligence, socially isolated, and poor) (Vernon et al., 1990). Older female characters are shown as failing, while the same is true for male characters only when they are villains (Aronoff, 1974). Furthermore, leadership and occupational power are portrayed as increasing with age for men but not for women (Lauzen & Dozier, 2005b). Thus, older women receive less respect than older men in these programs (Signorielli & Bacue, 1999).

However, there are certainly also some incidences of older people being represented more positively on television. For example, Jane Fonda and Lily Tomlin were shown in the title roles of “Grace and Frankie” for Netflix. These changes might be connected with the new media developments mentioned later in the article which can target and segment audiences more effectively.

Similar results for television programs targeting a general audience have been found in studies on television programs targeting children. Götz and Lemish (2012) have found that older people in children’s television are highly underrepresented, making up only 4.1% of characters, and this underrepresentation is even stronger for older women. Another study (Rovner-Lev & Elias, 2020) analyzed episodes of the television show “Grandpa in My Pocket,” which depicts older people, particularly older women, in negative ways. Older women are shown as socially isolated, devalued, weird, bizarre, and mean and are often humiliated and intimidated in the television show.

Considering these findings, unsurprisingly, television audiences also take note of these representations. Focus groups among older people in the UK showed that older people thought that they were negatively stereotyped on television and that older women were invisible in this medium, since women need to be young and beautiful, and older women are assumed to be off-putting to a young audience (Healey & Ross, 2002).

Older people are portrayed similarly in advertising, television, and movies. Older people in general and particularly older women are underrepresented and portrayed as younger than men in advertisements across the globe (Prieler & Kohlbacher, 2016). This underrepresentation also occurs in North America (Atkins, Jenkins, & Perkins, 1990/1991; Baumann & de Laat, 2012; M. M. Lee, Carpenter, & Meyers, 2007; Roy & Harwood, 1997), Europe (Simcock & Sudbury, 2006), and Asia (B. Lee, Kim, & Han, 2006; Ong & Chang, 2009; Prieler et al., 2011; Prieler, Kohlbacher, Hagiwara, & Arima, 2017; Prieler, Ivanov, & Hagiwara, 2017; Singh & Cole, 1993). This observation is in line with an analysis of advertisements in the magazines Vogue and GQ from the United States, UK, South Korea and Japan, in which the mean age of models was 28.99 years for women and 39.79 years for men—nearly an 11-year age gap (Han & Rudd, 2014). In contrast to these findings, one study in the UK showed that magazines specifically targeting older consumers used older characters (50+) more often in their ads and depicted them in a more favorable way than other magazines (Carrigan & Szmigin, 1999). Gender differences in representations of older people, however, go beyond numerical differences; research in Canada has also shown that while older men are often shown as still working in jobs with authority, older women lack any clear role—whether occupational or familial—and are not associated with any socially valued schema (in contrast to younger women in domestic contexts) (Baumann & de Laat, 2012).

However, there have been examples breaking with these traditional depictions of older women and their number may have increased over the past years (Jerslev, 2018a). The most famous example in this context is Dove’s Pro-Age Campaign that highlighted older women in advertisements. This campaign depicted older women naked and added slogans such as “too old to be in an anti-aging ad,” “wrinkled? wonderful?,” “grey? gorgeous?,” or “beauty has no age limit.” This campaign clearly broke with what the audience was used to seeing in ads because even anti-aging ads generally use middle-aged women and are based on the anxiety of aging. Beauty and age is communicated as something that can be controlled and signs of aging are read as failure, and the primary responsibility lies on the individual to succeed or fail (Twigg, 2004). The inclusion of older women in advertising from which they were previously excluded might be a step forward in the representation of older women in the media. However, these representations of older women can also be criticized on several accounts. Some have argued that the Dove campaign commercializes feminism or is “consumer feminism (Johnston & Taylor, 2008). In other words, the power of feminism is used for commercial purposes (Murray, 2013) and feminism becomes a marketing strategy to attract attention. Despite ads like the ones from the Dove Pro-Age Campaign, the majority of ads (particularly anti-aging ads) still do not portray older women or portray them in highly stereotypical ways (Calasanti, Sorensen, & King, 2012).

Even though each medium has its own specific characteristics, these examples of movies, television, and advertising have used similar representations of older women over the years. The same is true for other types of media, such as fashion magazines, in which images of women over 40 are scarce (Lewis, Medvedev, & Seponski, 2011), children’s picture books in which 60% of older characters (55+) are male, and only 40% are female (Danowski & Robinson, 2012), and also newspapers in which older people are underrepresented with only 22.1% older characters, while 54% of the population in the Netherlands are 55 years or older (Lepianka, 2015).

Therefore, overall, gender representations of older people are rather similar across different media, leading to the question of how such representations affect the audience, which will be outlined in the following section.

As seen in the previous literature review, the majority of research on gender representations of older people has been on older women. While this area of research is important and there is definitely more to do in that area, future research should also focus more on representations and images of older men (Twigg, 2020), which is an important area of research, since studies show that older men are also struggling with their aging bodies (Calasanti, 2004; Thompson, 2006; Thompson & Langendoerfer, 2016) and the media can play a role in this process.

In addition to more research on older men, gender research in the media should also go beyond the binary concept of gender and capture the diversity of gender identities (Eisend, 2019). While there have been some studies on media representations of different gender identities (Gomillion & Guiliano, 2011; Poole, 2014), research on gender representations of older people generally use a binary concept of gender, and more research should go beyond this binary concept.

Finally, research should further consider intersectional research on gender representations of older people and how older men and women are represented differently in the media based on their social status (class), their race/ethnicity, or their disability. There have been only a few studies that have investigated several dimensions; for example, Rivadeneyra (2011) analyzed gender, race, and age. Generally, quantitative studies have faced problems using intersectional analysis examining several dimensions at the same time since this approach has led to very small sample sizes (e.g., older male black individuals versus older female black individuals), which made statistical analysis less powerful. Thus, in such instances, qualitative analysis methods might be more appropriate.

The term “older people” or similar terms, such as “mature” and “elderly,” have been used in the literature to describe people ranging in age from over 45 to over 65 (Prieler & Kohlbacher, 2016). In fact, marketing literature interested in target groups often starts the definition of “older people” with 50+; Peterson and Ross (1997) even start calling those aged 45+ “mature models” but explicitly state that the terms “mature,” “elderly,” and “older” are used interchangeably in their article. This age range is rather wide, which makes it difficult to summarize research findings adequately and to compare results. For example, the age range includes people who are still in the workforce and those who have left the workforce. Future research should be more careful in following similar definitions of who “older” people are, especially given the fact that increasing lifespan and different life expectancies in many parts of the world means the definitions of “older” people might differ and change over time. For example, while the United Nations and the World Health Organization (WHO) generally use a definition of 60+ for older people, the WHO used a definition of 50+ for a study on sub-Saharan Africa (Kowal, Wolfson, & Dowd, 2000).

In addition, it might be advisable to subdivide age groups. Many studies lump older people together in a rather large age span, but it can be assumed that these different age groups will be represented in rather different ways. This situation is problematic in terms of analysis; 65-year-olds might still be shown to be rather fit and active (Givskov & Petersen, 2018), which is not necessarily the case for people in their 80s. As a result, some researchers have further subdivided older people into smaller age ranges, e.g., 50–64 and 65+ (Prieler et al., 2011); young-old (60–69), old-old (70–79), and oldest-old (80+) (Garfein & Herzog, 1995); and youngest-old (65–74), middle-old (75–84), and oldest-old (85+) (S. Lee, Oh, Park, Choi, & Wee, 2018). Such divisions make particular sense in regard to media representations since previous research has shown that the oldest age segment among older people is virtually unrepresented in the media (Prieler et al., 2011). For example, Gerbner et al. (1980) have shown that older women and men over 75 are nearly nonexistent in primetime television, while there are at least some older people in the age segment between 65 and 75 years of age.

Since people within one age segment are highly different from each other based on their health, roles, and interests, researchers might also consider other categorizations, such as biological age, or a person’s physical and mental condition (Jarvik, 1975); social age, or the age of the person in terms of his/her social roles and habits (Birren & Renner, 1997); cognitive age, or the age people perceive themselves to be (Barak & Schiffman, 1981); or gerontographics, which focuses on the needs, attitudes, lifestyles and behaviors of older people (Moschis, 1996; Nimrod, 2013). Such age categorizations might often be more useful than chronological age, especially in regard to surveys or experiments.

As our literature review has shown, the majority of studies on media content highlight gender differences in terms of numerical representations, with little research going beyond that aspect. However, scholars should conduct additional analyses beyond the simple question of whether older women and men are underrepresented since numerical representation alone does not necessarily tell us whether representations are negative or positive and might even exacerbate any problematic effects of media use (Collins, 2011). Thus, research should be expanded to include the nature of the depictions, for which research is comparatively scant. While numerical representation of older men and women is a sign of their “recognition,” more studies should analyze what Signorielli and Bacue (1999) call “respect,” or how people are portrayed in the media, for example, in terms of occupation or age.

In addition to comparing representations of older men and women, researchers should be careful to compare media representations not only in terms of equal distributions of age groups but also of the actual realities of each population and its real-world situation, e.g., regarding areas such as occupation (Eisend, 2019). The media can improve the status quo if media portrayals are more positive than the reality; however, it is also important to understand whether the media is merely reflecting the status quo or might even lag behind it.

Research should not only analyze the existence or nonexistence of representations and stereotypes of older men and women but also try to measure the degree of such stereotypes (Eisend, 2019), which has rarely been done in research to date. An exception is Royo-Vela, Aldas-Manzano, Küster, and Vila (2008) who rated sexism on an ordinal scale from nonstereotypic behavior to putting someone down in a study of gender role portrayals in Spanish television advertisements. More research should also be undertaken in the future in the context of representations of older women and men and should analyze how different degrees of stereotyping lead to different effects on audience members.

Finally, research might try to differentiate between and analyze older people in other ways. For example, older people could be analyzed based on their roles, such as victims, grandparents, or sources of wisdom, among others. Some scholars (M. M. Lee et al., 2007; Miller, Leyell, & Mazachek, 2004) have used a similar approach by following a categorization of stereotypes based on Hummert, Garstka, Shaner, and Strahm (1994), which divides stereotypical traits of older people into a few different types. Williams, Wadleigh, and Ylänne (2010) have come up with several groupings: (1) frail and vulnerable; (2) happy and affluent; (3) mentors; and (4) active and leisure-oriented older adults. Other research has analyzed whether certain adjectives describe older people better, such as angry/disgruntled, active/healthy, happy/content, humorous/comical, sexy/macho, and sick/feeble (T. E. Robinson, 1998) and Langmeyer (1993) has used an even longer list of adjectives. Research might also differentiate representations based on the type of ageism, differentiating between compassionate, intergenerational, and new ageism (Marier & Revelli, 2017). Compassionate ageism considers older people as “deserving poor” since they are unable to take care of themselves, intergenerational ageism represents older people as strong, greedy, and part of an age war and who benefit from the social system, and new ageism refers to the overly positive ideals behind concepts, such as healthy or active aging, which are impossible for many older people to reach.

While there have been some studies on the effects of representations of gender or older people, of audience perceptions of such representations, and of media production culture (Donlon et al., 2005; Korzenny & Neuendorf, 1980; Oppliger, 2007; Smith & Granados, 2009), the majority of the literature has focused on media content. Thus, future research needs to more often address the question of the effects of these representations. In the context of gender content analysis, Collins (2011) correctly noted the following: “··· we are short on theory and empirical evidence regarding the consequences of consuming media in which one’s gender is chronically underrepresented. Are there likely to be effects, on whom, and through what processes?” (p. 295). The same can be said for research on the possible effects of gender representations of older people. While there is some evidence about some types of representations, we know very little about what effects the underrepresentation of older women and older men has on these individuals and other groups.

Another suggestion by Collins (2011) is also relevant for this research area, namely, linking content analysis with effect studies or, more generally, mixing methods. Unfortunately, very few studies to date have used more than one method to investigate gender representations of older people. Scholars could, for example, draw on the methods of Gerbner et al. (1980) in cultivation theory in which they combined content analysis and surveys. Similarly, stimuli in effects studies could be based on the results of content analyses. In short, in the future, more research should triangulate the research on gender representations of older people in the media and not depend solely on one method.

Finally, only a few studies have focused on the conditions of media production and how production culture might affect the gender representation of older people. However, one major obstacle to changing media representations of older people, particularly older women, is the production side, which includes very few women of any age. Only 7% of people working in television in the UK are women over 50 (MacLeod, 2015) and similar findings have been reported for other media and countries (Carrigan & Szmigin, 2000; Prieler & Kohlbacher, 2016). Studies have found a correlation between the number of female producers and writers and the number of female characters in television drama and movies (Glascock, 2001; Lauzen & Dozier, 1999). Similarly, members of focus groups thought that the main reason for the current portrayal of older people and especially older women is that the people working in television are young and that young writers have their own experiences and cannot convincingly write a part for an older person (Healey & Ross, 2002). Thus, research should also focus more on the production side.

Several studies on gender representations of older people have been published during the last few years. However, the majority of these studies still have focused on traditional media with very few on new media, a phenomenon that is still mostly true for content analysis of gender in general (Collins, 2011). However, such research is crucial since the new media environment is potentially changing the representation of older women and men. Research by Makita, Mas-Bleda, Stuart, and Thelwall (2019) has, however, found results quite consistent with those for traditional media. They analyzed 1,200 tweets around the terms “aging,” “old age,” “older people,” and “elderly” and found personal concerns and health care the predominant topics. The tweets mostly reinforced negative discourses about aging, such as aging as a process that needs to be slowed or resisted. However, changes might occur in other spheres of social media, such as personal social media, in which users have become increasingly powerful by adding self-representations to traditional media representations. In addition, in streaming services, such as the previously mentioned Netflix or Amazon Prime (Jerslev, 2018b), which accommodate individualized consumer interests, more and more television series are starring older characters (Givskov & Petersen, 2018). Therefore, these developments can potentially change the representations of older people, and thus, more research should be conducted on representations of older women and men in social media in addition to media content distributed by streaming services to add to the existing research.

Not only have there been extremely few cross-cultural studies on the gender representation of older people or even studies on older people also highlighting gender aspects (B. Lee et al., 2006; Marier & Revelli, 2017; Prieler, Ivanov, & Hagiwara, 2017; Smith et al., 2014), nearly all studies to date are single-country studies. While these studies are interesting and important, future research should in addition also focus on cross-cultural studies since single-country studies use different samples, and thus, we cannot compare the findings across countries. For example, if we see differences between two studies, we cannot be sure whether the differences are based on cultural differences or whether they are based on differences in terms of the time frame, code book, sampling, or other characteristics of an individual study (Matthes et al., 2016). Therefore, cross-cultural and comparative research is needed. Since single-country studies run the risk of ethnocentrism, only comparative research is capable of analyzing the significance of phenomena and establishing the generality of findings and the validity of interpretations derived from single-country studies. Comparative research is capable of letting us test theories across different settings and helping to develop universally applicable theories (for further discussion, see Esser & Hanitzsch, 2012). In conclusion, comparative and cross-cultural studies would definitely increase our knowledge and understanding of gender representations of older people in the media.

Finally, the findings of much of the previous research have provided few practical implications. Research should clearly go beyond the purely academic area and outline the implications of their findings for policy and industry. The responsibility of media producers should be further emphasized, and responsible gender representations of interest to older people and older audiences should be suggested (Eisend, 2019).

Some studies have suggested what should be changed. First, gender representations of older people, particularly of older women, should be increased and diversified. The media needs to help debunk myths about older people and older women in addition to depicting a more positive and diverse image of aging (Kjaersgaard, 2005) rather than just images of older women in the context of decline (Wallander, 2013). Second, there should be an increase in older people and women on the production side. Considering that only 7% of people working in television in the UK are women over 50 (MacLeod, 2015), more older women should be involved in media production as creators, writers, editors, and producers (Kjaersgaard, 2005). Similar findings have been reported for other media and countries (Carrigan & Szmigin, 2000; Prieler & Kohlbacher, 2016). Studies have found a correlation between the number of female producers and writers and the number of female characters (Glascock, 2001; Lauzen & Dozier, 1999), which even further proves the importance of such changes. Some researchers also advocate guidelines, codes of conduct, or even laws for the representation of older people and against ageism and stereotyping (Carrigan & Szmigin, 2000). Finally, perceptions about aging are not limited to the media but are developed across society (Kjaersgaard, 2005). Without trying to change the discourse about aging in society (in which media can play a role), there will be little change in the media in the long term.