Medium Matters: A Decade of Media Consumption Predicts Positive and Negative Dimensions of Self-Perceptions of Aging

Abstract

Objectives

Media consumption over time is suggested to be a significant contributor to how people develop their self-perceptions of aging (SPA); however, this association has only been investigated with cross-sectional methodologies. The current study used growth curve modeling to examine the influence of 10 years of television, newspaper, radio, and book consumption on positive and negative dimensions of SPA.

Methods

Growth curve modeling on 4 waves of data from the German Aging Survey (N = 2,969), a population-based representative survey of adults aged 40–95, was used to examine the longitudinal associations between media consumption and SPA trajectories.

Results

Across 10 years, more television intake (B = −0.58, 95% CI [−0.94, −0.21]) was associated with lower perceptions of continuous growth. Inversely, greater book (B = 0.10, 95% CI [0.06, 0.13]) and radio (B = 0.52, 95% CI [0.29, 0.74]) consumption was significantly linked to higher perceptions of continuous growth. In parallel, more television (B = 0.88, 95% CI [0.52, 1.25]) and newspaper consumption (B = 0.46, 95% CI [0.04, 0.88]) was associated with higher perceptions of physical decline, while greater radio (B = −0.40, 95% CI [−0.64, −0.16]) and book (B = −0.05, 95% CI [−0.09, −0.00]) consumption was associated with lower perceptions of physical decline.

Discussion

This study provides longitudinal evidence for the relationship between media consumption and SPA. However, not all types of media intake are negative as radio and book consumption were associated with better SPA across time. Age-group differences were investigated and are discussed in the Supplementary Materials.

Ageism, Attitudes toward aging, German Aging Survey, Longitudinal, Views on aging

Media consumption in the form of viewing television, reading books and newspapers, and listening to the radio takes up the majority leisure time in the United States and Europe (Aliaga & Winquist, 2003; U.S. Bureau of Labor Statistics, 2018). Specifically, adults watch an average of 4.5 hr of television a day and viewership increases across the life span (Grajczyk & Zöllner, 1998)—becoming the most frequently cited activity for older adults (Robinson et al., 2004). Increases in media consumption with age may pose a risk to the health of older adults that reside in primarily ageist societies as media plays a role in reflecting, shaping, and reinforcing social views and norms regarding the aging process and life during old age (Harwood et al., 1994).

The potential risk of consuming ageist media is highlighted by stereotype embodiment theory (SET) which posits that the general aging stereotypes we hold true become self-stereotypes and eventually affect our thoughts and beliefs regarding our own aging process. Such thoughts and beliefs are known as self-perceptions of aging (SPA; Levy, 2009). Internalized positive and negative stereotypes become positive and negative SPA, respectively. Moreover, positive SPA includes perceptions of continuous growth and increased self-knowledge and control, and negative SPA includes perceptions of physical decline and social loss (Steverink et al., 2001). Such perceptions have long-term effects on physiological and psychological functioning (Levy, 2009), with those having more positive SPA living an average of 7.5 years longer than those with poorer SPA (Levy et al., 2002). While there is a paucity of empirical research investigating how SPA develops, SET suggests that cultural immersion via media consumption is the primary factor leading to differences in positive and negative SPA among individuals in the same society (Levy, 2009). This position is based on other tenets of SET which hold that most societies have more negative than positive aging stereotypes (Boduroglu et al., 2006), and in turn, more cultural exposure leads to an increasingly disproportionate internalization of negative stereotypes that become negative SPA.

Cultivation theory provides theoretical support for the connection between media consumption and SPA as media is a socializing agent that shapes individuals’ beliefs and attitudes toward aging (Gerbner et al., 2002). Other social psychologists suggest media reflects a culture’s values and standards, which communicate group status and expectations (Haboush et al., 2012)—thus, ageist cultures propagate negative expectations that act as self-fulfilling prophecies. Conversely, overly positive images can create ideals that are not possible to reach, ultimately hurting one’s self-perceptions (Slevec & Tiggemann, 2011). An experimental study found that adults who identify as old performed worse on a memory performance tasks after viewing ageist commercials, supporting the possibility that media can prime individuals with a negative self-schema (Westerhof et al., 2010).

Beyond media’s ability to transmit age-related stereotypes, two studies have explored the direct relationship between television consumption and views on aging, finding more television intake is related to worse attitudes toward older adults (Donlon et al., 2005; Gerbner et al., 1980). While such findings provide some evidence for SET’s media hypothesis, prior studies have been cross-sectional, conducted with small samples, limited exclusively to television consumption, and focused on attitudes toward older adults instead of SPA. Thus, longitudinal research is needed to test SET’s assumption that more media consumption is related to more negative and less positive SPA.

As media rapidly changes to reflect cultural attitudes, there is conflicting evidence as to whether media is still ageist (Loos & Ivan, 2018; Ylänne, 2015). Although early work utilizing content analyses revealed a high degree of ageism in television programming, longitudinal research has found older adults are increasingly portrayed in a positive light, often being shown as active, healthy, and independent (Loos & Ivan, 2018). However, older adults in the media who do possess positive traits most often resemble the third age of life, creating a mostly unachievable representation of successful aging in late life (Kessler et al., 2004). Moreover, older adults are still vastly underrepresented in proportion to their demographic standing in society (Ylänne, 2015), tend to be in minor or peripherical roles, and possess fewer positive traits compared to younger groups (Loos & Ivan, 2018). With television no longer viewed as being inherently ageist, there is a possibility that more consumption may no longer be associated with worse SPA over time.

Significantly less is known about the effects of print and audio media consumption on SPA development. Although not as widely consumed as television, listening to the radio and reading the newspaper and books take up roughly an hour of older adults’ daily activity (Aliaga & Winquist, 2003; U.S. Bureau of Labor Statistics, 2018). Similar to television programming, print media is becoming less overtly ageist; however, older adults are still vastly underrepresented (Loos & Ivan, 2018; Ylänne, 2015). Moreover, like television, ageism is perpetuated by newspapers and radio through advertisements that sell products to overcome “negative” conditions of old age (Raman et al., 2008). Although research on the representation of older adults in books is scarce, studies have shown the rampant ageism in children’s literature as older adults tend to have negative attributes and often subsume the role of the villain (McGuire, 2003).

While it is difficult to determine the extent to which various forms of media are ageist due to constant changes in media content, all forms of media have the potential to support positive SPA. For instance, books can offer factual insights about the world and allow for more complex character development than shorter forms of media. Therefore, it is possible that consuming more books may reduce negative SPA and increase positive SPA. Additionally, gaining perspectives on the world through consuming the news and staying in touch with cultural trends may allow all forms of media consumption to enhance dimensions of positive SPA. More research is needed to investigate whether various forms of media consumption differentially affect dimensions of positive and negative SPA across time.

The Current Study

This study uses longitudinal methods to examine the association between various forms of media (i.e., television viewing, book reading, newspaper reading, and radio listening) and changes in positive and negative SPA in adulthood. Specifically, the study focuses on perceptions of continuous growth and physical decline, as they are most commonly used to represent positive and negative SPA, respectively (for review, see Wurm et al., 2017). Due to conflicting evidence for the portrayal of older adults across media platforms, the present study examines the general research question of whether cumulative media consumption is related to changes in perceptions of continuous growth and physical decline across a decade.

Method

Data and Sample

Data for this study come from the German Aging Survey (DEAS), an ongoing nationally representative, cohort-sequential survey of the German population aged 40 and older (Klaus et al., 2017). The first DEAS survey wave took place in 1996, with further waves following in 2002, 2008, 2011, 2014, and 2017. New panels of individuals 40 years and older were added in 2002, 2008, and 2014. The current study includes data from individuals who completed the 2008 survey and at least two additional surveys in 2011, 2014, and 2017 (N = 2,969). In general, ongoing participants tend to be younger, healthier, more educated, and have higher incomes and larger informal networks than respondents who drop out (Klaus et al., 2017). Participant characteristics are in Table 1.

Table 1.

Participant Characteristics at Baseline (N = 2,969)

Variables	Mean/percentage	SD
Age (years)	60.18	10.43
Gender
Female	51.4%
Male	48.6%
Education
Low (incomplete vocational training/high school)	6.0%
Medium (completed vocational training/high school)	49.7%
High (any higher education)	44.3%
Marital status
Single	16.5%
Partnered	83.5%
Net monthly income (Euros)	2,882.74	3,054.16
Number of physical illnesses	2.16	1.67
Depression	6.08	5.76
Television consumption (hours/weekday)	2.46	1.58
Newspaper consumption	5.40	1.28
Radio consumption (hours/weekday)	2.64	2.91
Books consumption (count per year)	10.90	22.14
Perceptions of continuous growth	64.61	16.43
Perceptions of physical decline	58.01	17.95

Notes: SD = standard deviation. These demographics represent each participant’s response during the 2008 wave. Newspaper consumption was rated on a Likert scale ranging from 1 (daily) to 6 (never). Scores were reverse-coded, so higher ratings reflect more consumption.

Measures

Media consumption

While prior work has measured television viewing exclusively as media use (Donlon et al., 2005), the present study better aligns with recent time use surveys by measuring television, radio, newspaper, and book consumption (U.S. Bureau of Labor Statistics, 2018). Television consumption was measured with the item, “How many hours in total, on an average weekday, do you spend watching television?” Radio usage was assessed with a single item, “How many hours per day do you have the radio on, either as background (music) while doing other things or actually listening to the content of radio programs?” This question was slightly changed in 2014 to focus on radio listening during a typical weekday. Newspaper consumption was measured through a 6-point Likert question asking respondents “How often do you read the newspaper?” with options ranging from 1 (daily) to 6 (never). Scores were reverse-coded, so higher scores equate to more newspaper consumption. To assess book reading, participants were asked “How many books have you read in your spare time over the past 12 months?”

Perceptions of continuous growth

Perceptions of continuous growth were assessed with the 4-item Ongoing Development scale, a component of the Aging-related Cognitions scales (Steverink et al., 2001). Participants rated their perceptions on a 4-point Likert scale, ranging from 1 (definitely false) to 4 (definitely true). A sample item is “Aging means to me that I continue to make plans.” Previous research has found the scale to be reliable (Steverink et al., 2001; α = 0.78). The intraclass correlation (ICC) indicates that 54.11% of variance was between-person.

Perceptions of physical decline

Perceptions of physical decline were measured with the 4-item Aging-related Cognitions scale of Physical Loss (Steverink et al., 2001). Participants were asked to rate their perceptions on aging with a 4-point Likert scale, ranging from 1 (definitely false) to 4 (definitely true) on items such as, “Aging means to me being less energetic and fit.” The scale has been found to be valid and reliable (Steverink et al., 2001; α = 0.79). The ICC for the measure indicates that 56.92% of variance was between-person.

Covariates

In addition to media consumption, longitudinal studies have emphasized the association between personal characteristics and SPA. For instance, being older, and having more functional limitations and depression are negatively related to SPA (Sargent-Cox et al., 2012) as deteriorating health influences one’s ability to view aging positively. Furthermore, one’s gender, marital status, and income are all associated with SPA (Kleinspehn-Ammerlahn et al., 2008; Kotter-Grühn et al., 2009), and can influence the impact of media intake. To account for these effects, age, sex, education, marital status, socioeconomic status, comorbidities, and depression were used as covariates. Educational attainment was assessed with the International Standard Classification of Education which distinguishes between low (incomplete vocational training), medium (complete vocational training and/or high school degree), and higher education (completion of any higher education). Socioeconomic status was measured by net monthly household income. Lastly, comorbidity was represented by total number of physical illnesses from a list of 11 chronic health conditions (e.g., cardiovascular diseases, arthritis; ranging from 0 to 11). The German version of the Center for Epidemiological Studies-Depression scale was used to assess depressive symptoms (Hautzinger, 1988). Scores were summed and transformed so that they range from 0 to 45.

Data Analysis

To determine the association between various types of media usage and perceptions of continuous growth and physical decline across a decade of adulthood, two growth curve models were created using SPSS version 26. A maximum likelihood estimator was used to account for missingness. As suggested for longitudinal studies to have more interpretable fixed-effect coefficients, the dependent variables were transformed using the percentage of maximum possible (POMP) method, such that the lowest and highest scores are transformed to 0 and 100, respectively (Moeller, 2015). In the conditional models, age, gender, education, and marital status were grand-mean-centered and modeled as time invariant covariates. The media consumption variables, income, depression, and number of physical illnesses were modeled as time-varying predictors. Random effects were modeled for all media consumption variables. All models utilized a variance components covariance matrix. Supplemental analyses were carried out to investigate whether the effect of media on SPA was the same across middle-aged (40–59 years of age) and older adults (60 years of age and older; see Supplementary Appendices A and B)

Results

Unconditional growth curve models suggested time for both perceptions of aging should be represented linearly. Specifically, perceptions of continuous growth were found to start above the midpoint of the transformed scale (B = 64.61, 95% CI [64.03, 65.19]) and decline at each wave (B = −0.42, 95% CI [−0.66, −0.17]). In contrast, perceptions of physical decline started above the scale’s midpoint (B = 57.84, 95% CI [57.21, 58.47]), but increased linearly at each wave (B = 0.20, 95% CI [−0.02, 0.42]). Although the linear decline in perceptions of physical decline was small, the significant random effect of the slope suggests that there are notable between-person differences in intraindividual change over time (see Supplementary Appendix C).

The results from each conditional analysis focus on the effect of the media consumption variables, while the estimates for all variables can be found in Table 2. More television consumption was associated with decreases in perceptions of continuous growth (B = −0.58, 95% CI [−0.94, −0.21]), while more radio (B = 0.52, 95% CI [0.29, 0.74]) and book (B = 0.10, 95% CI [0.06, 0.13]) consumption was related to increases in these perceptions at each wave. In contrast, newspaper consumption (B = −0.13, 95% CI [−0.54, 0.29]) was not significantly related to perceptions of continuous growth across waves. These findings held for middle-aged adults. For older adults, the influence of newspaper consumption on perceptions of continuous growth was not significant; however, the effect was in the positive direction. Additionally, the influence of book consumption did not reach statistical significance for older adults (see Supplementary Appendices D–F).

Table 2.

Fixed and Random Effects of Predictors on Self-Perceptions of Aging (N = 2,969)

	Perceptions of continuous growth	Perceptions of physical decline
Variable	B [95% CI]	B [95% CI]
Fixed effects
Intercept	65.98*** [63.12, 68.85]	49.74*** [46.84, 52.64]
Time	0.09 [−0.39, 0.56]	−0.42 [−0.90, 0.05]
Television consumption	−0.58** [−0.94, −0.21]	0.88*** [0.52, 1.25]
Newspaper consumption	−0.13 [−0.54, 0.29]	0.46* [0.04, 0.88]
Radio consumption	0.52*** [0.29, 0.74]	−0.40*** [−0.64, −0.16]
Books consumption	0.10*** [0.06, 0.13]	−0.05* [−0.09, −0.00]
Age	0.02 [−0.07, 0.11]	−0.12* [−0.33, −0.03]
Gender	0.12 [−1.75, 2.00]	−2.79** [−4.76, −0.82]
Education	1.22 [−0.37, 2.81]	0.74 [−0.93, 2.42]
Marital status	0.07 [−2.50, 2.63]	1.56 [−1.13, 4.24]
Net monthly income	0.58** [0.24, 0.93]	−0.50** [−0.85, −0.14]
Number of physical illnesses	−1.47*** [−1.83, −1.11]	2.80*** [2.43, 3.16]
Depression	−0.07 [−0.17, 0.03]	0.06 [−0.04, 0.16]
Linear effects on time
Age	−0.13*** [−0.18, 0.08]	0.08*** [0.03, 0.12]
Gender	0.41 [−0.53, 1.36]	0.64 [−0.30, 1.59]
Education	0.22 [−0.58, 1.02]	0.04 [−0.76, 0.83]
Marital status	0.69 [−0.62, 2.00]	−1.65* [−2.96, −0.35]
Random effects
Time	9.40*** [6.50, 13.58]	3.19* [1.26, 8.05]
Television consumption	2.53*** [1.56, 4.09]	1.72** [0.88, 3.34]
Newspaper consumption	2.86*** [2.35, 3.48]	2.96*** [2.43, 3.60]
Radio consumption	0.43 [0.12, 1.49]	1.15*** [0.63, 2.09]
Book consumption	0.01 [0.00, 0.02]	0.01 [1.14, 1.78]

Notes: CI = confidence interval. Net monthly income was divided by 1,000 to create more equal variances among predictors.

*p < .05. **p < .01. ***p < .001.

When modeling perceptions of physical decline as the dependent variable, increases in television (B = 0.88, 95% CI [0.52, 1.25]) and newspaper (B = 0.46, 95% CI [0.04, 0.88]) consumption were associated with more negative perceptions at each time point. In contrast, radio (B = −0.40, 95% CI [−0.64, −0.16]) and book consumption were inversely related to perceptions of physical decline (B = −0.05, 95% CI [−0.09, −0.00]; see Table 2). The direction of effects was the same across age groups; however, the effects of book and newspaper consumption on perceptions of physical decline did not reach statistical significance for middle-aged and older adults, respectively (see Supplementary Appendices D–F).

Discussion

The current study investigated SET’s assumption that more media consumption is responsible for worse SPA among individuals in the same culture (Levy, 2009). In line with previous work investigating attitudes toward older adults, television consumption was related to worse SPA (i.e., lower perceptions of continuous growth and higher perceptions of physical decline) across a 10-year period (Donlon et al., 2005; Gerbner et al., 1980). However, media consumption is not inherently harmful, as radio and book consumption were associated with more positive SPA over time. Additionally, results suggest that media’s longitudinal association with SPA is more complex than previously assumed since more newspaper consumption was related to increases in perceptions of physical decline, but unrelated to perceptions of continuous growth. Although it appears that cultural immersion is associated with differences in SPA across time, media consumption is not inherently harmful, just as it is not inherently ageist—especially as societies progress (Loos & Ivan, 2018).

With worse SPA being related to an increased risk of mortality (Levy et al., 2002), the negative impact of television consumption on SPA poses a health risk to the German people, as well as other nations where television is the most pervasive form of media intake (Aliaga & Winquist, 2003). As television intake increases with age (Robinson et al., 2004), there may be a cumulative negative effect on health and mortality risks. Our results provide some evidence that the content chosen by individuals promotes ageist stereotypes. However, research finds that television is becoming less ageist in Europe and United States (Loos & Ivan, 2018; Ylänne, 2015), and that older adults prefer more informative television like watching news channels (Robinson et al., 2004). Therefore, the harmful relationship between television consumption and SPA may be driven by ageist advertisements which are found across nations (e.g., China, Europe, India, and United States; Zhang et al., 2006).

Advertisements prompting comparisons to unrealistic aging expectations or priming negative age schemas (Westerhof et al., 2010) may also engender the observed negative influence of newspaper consumption on perceptions of physical decline. While the content of newspapers has not been found to be ageist beyond underrepresentation, the same cannot be said for the advertisements in this medium (Carrigan & Szmigin, 2000). The finding that newspapers were only related to perceptions of physical decline and not ongoing development may be explained by the extent to which newspaper advertisements promote stereotypes related to physical deterioration, the most prevalent domain of negative stereotypes associated with aging (Hummert, 2011). Beyond solutions offered to allay ageist media content (Loos & Ivan, 2018), policy makers in Germany and nations that have similar forms of ageist media content (e.g., China, India, and the United States) should limit ageist advertisements as a means of promoting public health.

Notably, not all forms of media consumption were related to worse SPA. Listening to the radio and reading books were beneficial forms of cultural immersion that allowed for more positive SPA over time. It is possible that books provide the pleasure of television and the information of newspapers without the advertising. Interestingly, while the radio does have advertisements, one study found that older adults are far less likely to recall age-related advertising on the radio (9%) compared to television (53%; Zhang et al., 2006). Again, while the radio is used for similar reasons as television and newspapers, the lack of influence by advertisements may allow the beneficial ties to culture provided by media consumption take hold without the promotion of ageist messages.

Limitations and Future Directions

Although the German Aging Survey allowed for a thorough investigation of the relationship between the quantity of media consumption and changes in SPA across 10 years, it was not possible to investigate the content of the media participants consumed. Future research should focus specifically on the interaction between media type and content in relation to SPA development. Relatedly, this study observed slight age-group differences in the effects of media on SPA. The direction of the effects was the same across age groups with the exception of the nonsignificant positive association between newspaper consumption and perceptions of continuous growth for older adults. Replication is needed to determine whether differences in significance levels were due to reductions in sample size or actual group differences. Overall, more research is needed to begin to disentangle these findings as not much is currently known about the factors that differentially affect younger and older adults’ internalization of age-related content, especially as it relates to various forms of media.

Even though this study revealed longitudinal associations between media intake and SPA, causality cannot be inferred from the findings due to the lack of experimental control needed to eliminate alternative causal hypotheses. Therefore, more longitudinal studies are needed to disentangle the direction of effects, and test whether worsening SPA may prompt more negative forms of media consumption. Moreover, with media content and interfaces constantly evolving (i.e., streaming services), more research is needed to see if the current findings hold for more recent media programming. With some paid streaming services not having advertisements it may be possible to test whether the harmful effects of television consumption on SPA are driven by advertisements. The current study is the first to explore the long-term link between media consumption and dimensions of positive and negative SPA. Based on the findings, it is clear that actions need to be taken to reduce ageism on television to enhance positive aging at the population level.

Funding

T. Chan is supported by NIH NIGMS BUILD (UL1GM118976 and RL5GM118975). The German Ageing Survey was funded by the German Federal Ministry for Family, Senior Citizens, Women and Youth.

Acknowledgments

All data come from the German Aging Study. Due to contractual obligations, the data cannot be released by the authors. However, all syntax for analyses can be made available and will be sent upon request. The data from this study were not preregistered due to the data set being owned by the German Government. This study was not preregistered. However, all analytic materials will be sent upon request.

Conflict of Interest

None declared.