Abstract
Ageism was found to be prevalent across various life domains, but not in the music domain. This study aimed to examine whether negative ageist attitudes toward older adults are associated with music preferences across life. Subjects (n = 214, age range 30–92, M = 56.14, SD = 16.39) were exposed to old (1910–1946), intermediate (1979–1987), and new (2014–2018) songs, chosen by a focus group
• The study demonstrates that ageism may be connected to liking for songs. • The study demonstrates that the oldness of songs is perceived relatively to the chronological age of their listeners along adulthood and old age. • The study is suggesting that the perception of oldness in songs may be interpreted by the out-group homogeneity effect, as the association between ageism and reduced liking for old songs was only found among the out age-group (i.e., younger adults).
• The awareness of ageist perceptions in music can pave the way to reduce them. • The study of levels of liking for songs in different age groups can be the first step in bridging intergenerational gaps. • It can be followed by an intergenerational intervention program that includes songs, which can trigger a dialogue between generations about ageist perceptions.What this paper adds
Applications of study findings
Introduction
Music is a significant part of human beings’ daily lives. It enables non-verbal communication between people and has the power to unite them (Bodner & Gilboa, 2009; de Nora, 2000; Sacks, 2006). Nevertheless, music can also signify a cultural conflict, which is reflected in in-group favoritism and in prejudice against out-groups (Bergh & Sloboda, 2010; Bodner & Bergman, 2017). Moreover, people tend to attribute personality traits to other people they do not know based on their music preferences (Neto et al., 2016; Tarrant et al., 2002). In this regard, Reyna et al. (2009) found that individuals who reported dislike toward rap music demonstrated more discrimination against African Americans. In other words, when a given out-group is perceived negatively, music which seemingly defines this group, will share a similar fate.
While previous research has demonstrated that music can define in-group boundaries in political, ethnic, and socio-demographic contexts (Bodner & Bergman, 2017), much less is known about age groups boundaries that music can enhance. The aim of the present study is to examine if younger, middle-aged and older adults demonstrate different associations between ageist attitudes and liking for songs who are old, intermediate and new.
Although the love of music is shared among people of all ages, people of different ages treat music differently. For example, younger individuals demonstrate preference for contemporary music (Bonneville-Roussy et al., 2017), whereas older adults prefer musical styles, which are familiar to them from an early age (Hallam et al., 2016). In this regard, it was found that music experienced during adolescence has a special importance for the rest of one’s life (Gabrielsson, 2002). Therefore, it is no wonder that older adults who were young when Europe was in a war, preferred singing anthems and patriotic songs over singing folk songs (Moore et al., 1992). These tendencies were also evident in the preference of older adults to music in their national language over songs in foreign languages (Hallam et al., 2016).
A negative attitude toward old songs may not always be expressed explicitly, perhaps because expressing such an opinion may seem impolite. Nevertheless, old songs can be considered as outdated and anachronistic, and sometimes the attitude to old songs can be quite mocking (Eliram, 2006). In this regard, there is evidence for bias in the music industry, reflected by reduced public media play of old songs and older singers, initiating public performances mostly for younger audiences, signing contracts mainly with younger artists, and negative attitudes toward older artists, who are often regarded as if they have finished their careers (Donzelli, 2020; Katz, 2014). Even when older adults express a preference for old music, they may seem to others as incapable of making sense of the contemporary pop music (Forman, 2012).
These abovementioned behaviors seem to be expressions of ageism as it has been defined by Butler (1969). Butler defined ageism as negative stereotypes, prejudice, or discrimination against people because of their age, and as directed mainly toward older adults (Butler, 1969). Studies have found that ageism adversely affects the health of older people and their subjective well-being (Gvili & Bodner, 2021; Marquet et al., 2019). Previous studies have shown ageism to be prevalent in different life domains, such as in organizational regulations (Spencer, 2013), at work (Stypińska & Nikander, 2018), in doctors’ offices (Wyman et al., 2018), and in media (Loos & Ivan, 2018). However, as previously described, ageism may also be reflected in people’s attitudes toward the songs that older people like. Therefore, the current work argues that ageist attitudes can be associated with reduced liking for old songs, as such songs may signify oldness.
In order to explain why old songs may signify oldness, the concept of age should be defined. Age is an objective number that means the number of years that have passed since the year of birth until today. However, age is also a cultural and social structure linked with an arsenal of specific cultural beliefs and norms about its meaning. While these cultural beliefs and age norms help in adjusting to reality, they also limit humans thinking (Ojala et al., 2016; Zebrowitz & Montepare, 2000). In fact, age norms can be expressed through the smallest details, and dictate how people in different cultures perceive the age of themselves and others (Rothermund et al., 2021). For example, a study that discussed the clothing practices of older women, found that different outfits and even different clothing stores are coded in society by age, and are perceived by people of different ages as more or less appropriate for specific age groups (Krekula, 2009; Twigg, 2012). Studies have shown that people of different ages perceive differently what old age is (Barrett & Toothman, 2016). Accordingly, we suggest that people of different ages would also treat the concept of oldness in music differently. Therefore, it might be expected that from the perspective of the age-group, middle-aged and younger people would associate old songs with oldness, whereas older adults would not do so. In contradistinction, younger people may also associate intermediate songs with oldness, whereas older and middle-aged adults would not do so.
As a theoretical framework for this idea, we suggest the Social Identity Theory (SIT; Tajfel & Turner, 1986). SIT stipulates that individuals’ personal identity is determined, to a large extent, by their social identities, that is, by the extent to which they perceive themselves as part of several social groups. According to SIT, humans categorize themselves and others in social groups and favor members of their in-group over out-group members, due to a psychological need for positive social distinctiveness. SIT has been used for explaining the tendency to prefer the in-group over the out-group, and thereby to explain the positive attitudes young people have toward their young peers, as opposed to their negative attitudes toward older adults—their “out-group.” However, contrary to the tendency to prefer the in-group over the out-group, social groups who are considered somewhat inferior (in this case—older adults) often internalize a negative evaluation of themselves and are positively oriented toward out-groups (Lev et al., 2018; Tajfel & Turner, 1979), as it also may happen in the case of older adults (Bodner, 2009; Kite et al., 2002). Following this line of thought, it was found that older adults often identify with younger age groups, as a way to maintain their social value (Weiss & Freund, 2012; Weiss & Lang, 2012). This tendency to prefer the younger out-group over the in-group may lead older adults to attribute negative attitudes toward their own age-group, a phenomenon which has been described as “self-ageism” (Ayalon & Tesch-Römer, 2017; Bodner, 2009; Schuurman et al., 2022).
Following SIT, we suggest that in-group favoritism among people of different ages can be detected by an examination of the association between ageist attitudes and liking for songs characterized by three levels of oldness (i.e., old songs, in between songs, that is—intermediate songs, and new songs). We defined the oldness of the songs as follows: old songs are songs created in the years 1910–1946, intermediate songs are songs created in the years 1979–1987, and new songs are songs created in the years 2014–2018. We suggest that for young adults, a negative association will be found between their level of ageism and their level of liking for old songs, meaning that high levels of ageist attitudes will be associate with negative attitudes towards the music of their “out-group”—the older adults. Moreover, we expect that similar to young adults, middle-aged people, for whom these old songs were not part of their youth, will present the same negative association between ageist attitudes and liking for old songs. However, older adults, for whom the same old songs are part of their years as adolescents, will not associate ageist attitudes with liking for these old songs. Accordingly, even if older participants may find some old songs unpleasant, their liking for these songs will not be associated with a general stereotypical ageist perception of older adults.
Moreover, following the same reasoning, while we expect young adults to treat intermediate songs in a similar manner as they treat old songs (i.e., as reflecting their age out-group), for middle-aged people, who also perceive intermediate songs as part of their years of adolescence, we do not expect an association between ageist attitudes and liking for intermediate songs. Therefore, we hypothesize that like older adults, middle-aged adults will not present the same negative association between ageist attitudes and liking for intermediate songs. As for the relation of older adults to intermediate songs and to new songs, we expect that these songs would signify modern time for older adults, and following their tendency to prefer the younger out-group over their in-group (Kite et al., 2002) they will demonstrate a positive association between ageist attitudes and liking for intermediate songs and new songs. That is, the higher the ageist attitudes of older people, the more they will like intermediate songs and new songs (representing their preference for the out-group).
Additionally, and in a similar manner, we expect a positive association between ageist attitudes and liking for new songs among middle-aged people as these songs can signify modern time to them and because of the tendency of an inferior out-group to prefer the younger out-group over their in-group (Kite et al., 2002). Finally, we expect that the younger age-group will not present such positive association between ageist attitudes and liking for new songs. This is because in similar to older and middle-aged adults who perceive old and intermediate songs, (respectively), as an integral part of their years of adolescence, young people relate these new songs to their youth. Therefore, their liking for these songs will not be associated with ageist attitudes.
Our hypotheses are: (1) the association between ageist attitudes and liking for old songs would be negative and more pronounced among young and middle-aged people (the age out-group for such songs), in comparison with older adults; (2) the association between ageist attitudes and liking for intermediate songs would be negative among young adults, and positive among older adults (the two age out-groups for such songs) in comparison with middle-aged people; (3) the association between ageist attitudes and liking for new songs would be positive and more pronounced among older adults and middle-aged people (the out-group for such songs), in comparison with young people.
Methods
Participants
Data were obtained from 214 participants (age range = 30–92, M = 56.14, SD = 16.39), of which 91 (42.5%) were male. Since the songs were Israeli songs, the inclusion criteria for the study were Israeli Jewish adults, non-musicians, (to control for the possible confounding effect of professional musical proficiency; see Shifriss et al., 2015), living in Israel for at least 20 years (long enough to familiarize with the culture, language, and songs). Most of the participants (80.37%) were born in Israel, and the rest were living in Israel for an average of 60.29 years (SD = 15.27).
Measures and Procedure
Information was collected using a convenience sample and throughout an online Qualtrics survey. After signing a digital informed consent form, subjects were provided with an anonymous link to a designated website, which contained the study questionnaires. Participants filled out the questionnaires voluntarily in their homes, places of work, or at any other place convenient for them, and upon completion, data were stored in the Qualtrics website. The study received an ethical approval by the IRB (E.MUS.2018–10) of the first author’s university.
Ageism was assessed by the Hebrew version (Bodner et al., 2012) of the Fraboni Scale of Ageism (Fraboni et al., 1990). Participants were required to rate the extent of their agreement with 18 statements (e.g., “Old people should find friends their own age”) on a scale ranging from 1 (strongly disagree) to 6 (strongly agree). The internal validity of this scale has been demonstrated in many cultures, including Israel (Bodner et al., 2012). A mean score of ageism was calculated. Cronbach’s alpha was .77.
Characteristics of Research Songs, Perceived Liking for Songs and Perceived Age of Songs.
Note. n = 214.
aThe Finjan is a small pot/kettle used for bonfires, which was popular before Israel was established.
bThis was a derogatory term for British soldiers during the British mandate of Israel.
Each song was played for 30 seconds, and excerpts included the opening parts and the refrain. Songs were played in random order. After hearing a given excerpt, participants were requested to indicate the age attributed to those who like each of the songs (“how old would be the person who liked this song?”). The age range given to the answer was between 0–100 years. Confirmatory factor analysis (CFA) was carried out to assess the reliability and validity of the proposed division. The results of the CFA for age assigned to preference for songs demonstrated a good model: Model Fit: χ2 (132) = 236.41; RMSEA = .061 (.048–.073); SRMR = .058; CFI = .924; TLI/NNFI = .912. These results matched the proposed focus group division of the songs to new/intermediate/old songs and were old songs (a person aged 63.31 years), intermediate songs (a person aged 44.02 years), and new songs (a person aged 23.91 years).
After listening to each song, the participants were asked how much they like the song on a scale ranging from 0–10. Mean scores were calculated for liking for old songs (Cronbach’s alpha was 0.92), liking for intermediate songs (Cronbach’s alpha was 0.73) and liking for new songs (Cronbach’s alpha was 0.80), and high scores reflect high levels of liking. In order to control for confounds of musical background (see Shifriss et al., 2015), participants were asked to rate their love for music on a scale ranging from 1 (very low) to 7 (very high), to report whether they play or had played musical instruments in the past for at least a year (yes/no), and to report familiarity with each of the 18 songs (yes/no), so that a mean of familiarity of each of the three groups of songs was calculated. Moreover, we collected socio-demographic data: gender, relationship status (not in a relationship/in a relationship), and self-rated economic status (less than average/like average/above average). Based on a previous study, these data were also controlled (Bodner & Bergman, 2017).
Data Analysis
Data were analyzed using the SPSS-25 software, and significant interactions were probed using Model 1 of the PROCESS 3.4 macro for SPSS (Hayes, 2018). Initial correlations between the study variables were calculated, and the hypotheses were examined by a hierarchical regression, with liking for songs (old songs, intermediate songs, and new songs) as the dependent variable. The first step included demographic (gender, relationship status, economic status) and music background variables (playing musical instruments, love for music, familiarity with the songs). The second step examined the main effects of participants’ age and ageist attitudes, and the third and final step included the age × ageist attitudes interaction. Using the PROCESS, we analyzed the moderated effects of age (X) on liking for old, intermediate, and new songs (the Y variables) at three age points (M): 5%–36 years (young-age), 50%–57 years (middle-age), 95%–75 years (old-age). A power analysis for detecting a medium effect size (0.15) with 11 predictors required a sample size of 107, indicating that the current sample was sufficient for the study model. Potential multicollinearity was rejected, as the tolerance and VIF ratios ranged between 0.44–0.92 and 1.09–2.25, respectively, which is in line with literature requirements (O'Brien, 2007).
Results
Means, SDs, and Pearson Correlations among Study Variables for the Cohort.
Note. n = 214; Correlation values represent Pearson coefficients except for gender, relationship status, economic status, playing musical instruments and love for music, which represent biserial correlations.
a0 = male; 1 = female.
b0 = not in a relationship; 1 = in a relationship.
c0 = less than average; 1 = like average; 2 = above average.
d0 = don’t play; 1 = play.
e0 = don’t like; 1 = neutral; 2 = like; 3 = very like.
f0 = no; 1 = yes.
gscale ranging from 0–10.
hscale ranging from 0–6.
* = p < .05; ** = p < .01; *** = p < .001.
Regression Coefficients for Predicting Liking for Songs.
Note.
a0 = male; 1 = female.
b0 = not in a relationship; 1 = in a relationship.
c0 = less than average; 1 = like average; 2 = above average.
d0 = don’t play; 1 = play.
eFor each category of songs, the variable included familiarity with the songs that belong to the category.
* = p < .05; ** = p < .01; *** = p < .001.
Old Songs
Ageist attitudes were only negatively significantly associated with liking for old songs among young (B = −1.31, SE = .41 p < .01) and middle-aged participants (B = −.53, SE = .26 p < .05), but not among older participants (B = .14, SE = .36, p > .05; see Figure 1). Therefore, the first hypothesis, that the association between ageist attitudes and the liking for old songs would be negative and more pronounced among young and middle-aged people (the age out-group for such songs) in comparison with older adults, was confirmed. The results of the interaction remained unchanged with and without covariates. The association between liking for old songs and ageism in interaction with age. The figure shows the moderated effects of ageism on liking for old songs at three age points: 5% (young age: 36 years), 50% (middle age: 57 years), and 95% (old age: 75 years).
Intermediate Songs
The second hypothesis was that the association between ageist attitudes and liking for intermediate songs would be negative among young adults, and positive among older adults. In line with this hypothesis, ageist attitudes were only negatively significantly associated with liking for intermediate songs among young participants, (B = −.68, SE = .31, p < .05), but not among middle-aged participants (B = −.23, SE = .20, p > .05) and older participants (B = .18, SE = .26, p > .05; see Figure 2). Therefore, the second hypothesis was only partially confirmed (for young but not for older participants). The results of the interaction remained unchanged with and without covariates. The association between liking for intermediate songs and ageism in interaction with age. The figure shows the moderated effects of ageism on liking for intermediate songs at three age points 5% (young age: 36 years), 50% (middle age: 57 years), and 95% (old age: 75 years).
New Songs
The third interaction of ageist attitudes × age for new songs was not significant (Table 2); and therefore, the third hypothesis was not confirmed.
Discussion
This study examined if younger, middle-aged and older adults demonstrate different associations between ageist attitudes and liking for old, intermediate and new songs. Such a perspective on ageist attitudes complies with studies which demonstrated that ageism can be prevalent in different life domains (e.g., Loos & Ivan, 2018; Spencer, 2013; Stypińska & Nikander, 2018; Wyman et al., 2018), and yet, to the best of our knowledge, this is the first study that examined whether ageism can also be linked with people’s attitude toward the songs that people like.
In line with the first hypothesis, while higher levels of ageism were negatively associated with increased liking for old songs among young and middle-aged individuals, this effect was nullified among older adults. Moreover, in line with the second hypothesis, higher levels of ageism were negatively associated with increased liking for intermediate songs among young individuals and this effect was nullified among middle-aged people. These findings are important as they provide an initial indication that ageism might be expressed in the attitudes toward music. Raising awareness of ageist attitudes toward music, which represents the preferences of older adults, can be the first step that will pave the way to their reduction.
The second part of the second hypothesis was not confirmed since older adults did not demonstrate a positive association between ageist attitudes and liking for intermediate songs. Moreover, the third hypothesis, that older adults and middle-aged people will demonstrate a positive association between ageism and liking for new songs was not confirmed. The rationale for these two hypotheses was that the connection between ageist attitudes and liking for new/intermediate songs would result from a preference of older and middle-aged adults for the more prestige out-group of younger adults. However, another examination of the literature has found various quotes of older adults referring to the music of the younger generation as shallow, depthless, instant and computer-controlled, linking it to their young age (Eliram, 2006; Katz, 2014). In other words, it seems that the ageist perceptions toward music also exist toward new songs. This attitude toward new songs can be defined as a kind of adultism. Adultism is the assumption that youngsters are not as competent as adults due to their young age and are less experienced (Bell, 2010). In this study, we did not control for adultism. Future studies need to examine negative attitudes toward the new and intermediate songs among participants of different ages alongside the examination of ageism and adultism.
Our findings show that the negative association between ageist attitudes and liking for old songs exists among the younger “out groups”; that is—among the young and the middle-aged people, while the out-group of the intermediate songs is the young age-group. This finding is consistent with Reyna et al. (2009), who reported that negative attitudes toward African Americans were associated with dislike of rap songs among non-African Americans. Interestingly, this study also found that while Caucasian Americans believe that rap music represents the African American group, African Americans do not share this view, as they do not define their group through this genre. In a similar manner, in the current study, it seems that older adults did not perceive old songs as representing “oldness,” and middle-aged people did not perceive intermediate songs as indicating “oldness,” because for them, old songs and intermediate songs, respectively, were part of their youth, and can be connected to memories from their youth and not to old age. In contrast, the out-group—the participants of the young age-group did link between oldness and intermediate songs. This explanation of the findings also corresponds with individuals’ tendency to perceive their out-group members as more similar to each other than their in-group members, known as “the out-group homogeneity effect” (Quattrone & Jones, 1980). Upon listening to old songs, some older adults may be responsive to the melody, others to the words, and some to their personal biography and react with nostalgia (Chou & Lien, 2014; Wildschut et al., 2018). However, as younger generations lack the emotional connection with this old music, it might be that ambivalence and/or dislike toward this old music corresponds with general negative attitudes toward the age-group it represents.
Another way to interpret the findings of this study relates to the perception of what is the age of becoming old among people of different ages. The age groups defined differently what old songs are. While for the old songs the association between ageist attitudes and liking for the songs exists both among the young age group and the middle-age group, for the intermediate songs, this association is maintained only among the young age group. This association with oldness of songs which are older than the participants’ age group corresponds well with the statement attributed to Francis Bacon: “I will never be an old man. To me, old age is always 15 years older than I am” (Voorhees, 2001). Indeed, studies show that older adults tend to perceive older adulthood as starting at a later age relative to younger adults (Barrett & Toothman, 2016). In this regard, a survey by Taylor et al. (2009) found that while young people (aged 18–29 years) defined the average age in which a person becomes “old” as 60 years, middle-aged people (aged 50–64 years) defined an old person as 72, and respondents who were 65 years old and over indicated that an old person is a person at the age of 74. This phenomenon was explained as a way for people to protect themselves and to keep away from ageist perceptions which are associated with old age (Weiss & Freund, 2012). Following this line of thought, it seems that while from the point of view of younger adults, the intermediate songs are seen as old-age songs, the middle-age group did not deem these songs as old. The same explanation may be relevant for the songs of old age, which in the eyes of the old
On a sidenote, another finding which may signify ageism was the fact that although the participants could indicate the age attributed to those who would like each of the songs on a range of 0–100 years, the average age they attributed to people who like old songs was 63.61 years, and the oldest age (in average) that the participants attributed to those who liked the oldest song (“Kalaniot”) was 65.14 years (see Table 1). It seems that there is an older age exclusion which is reflected in the perception that the love for these songs should end at the beginning of the age of 63.61 years. According to this evaluation, people at older ages are not relevant for liking music anymore.
The findings of this study should be interpreted in light of its limitations and strengths. This is a cross-sectional study, and lab studies are required so that causality can be inferred. Moreover, it is subjected to cultural influences. To demonstrate a cross-cultural effect, its findings need to be replicated in other countries and cultures. Another limitation is related to the fact that the songs include many characteristics that may define them as old, intermediate or new which were not examined (the content of the song, the singer, the accompanying musical instruments, the melody, etc.). The findings open the door for studies that will examine the effect of specific components defining the level of oldness in music on participants’ attitudes toward the age out-group. Moreover, future studies can manipulate these components with the aim to decrease the intergenerational gap. Additionally, this study is the first to demonstrate that ageist attitudes can be differently associated with liking for songs among people of different ages. Creating a map of the liking for songs in different age groups may serve as the first step in bridging intergenerational gaps. An intergenerational music intervention that will use songs, can create a dialogue between young and older generations, so that both generations will understand and even experience the meaning of these songs for each generation.
Footnotes
Acknowledgments
We would like to thank Dr. Yoav S. Bergman for his good advices on statistics and writing issues.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Ethical Approval
The study received ethical approval (E.MUS.2018–10) from a departmental ethical review committee in the authors’ university in accordance with the Declaration of Helsinki, on 19/11/2018.
