Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Nov 16.
Published in final edited form as: Appl Psychol Health Well Being. 2024 May 24;16(4):1778–1800. doi: 10.1111/aphw.12553

Daily Social Comparisons among Women in Midlife with Elevated Risk for Cardiovascular Disease: A Within-Person Test of the Identification/Contrast Model

Kiri Baga 1, Gabrielle M Salvatore 2, Iris Bercovitz 3, Jacqueline A Mogle 4, Danielle Arigo 5
PMCID: PMC11567783  NIHMSID: NIHMS2018469  PMID: 38785210

Abstract

Women ages 40–60 are disproportionately affected by health problems that increase risk for cardiovascular disease (CVD; e.g., hypertension). Social comparisons (i.e., self-evaluations relative to others) are known to influence health in this and other groups, but their nature and consequences in daily life are poorly understood. We conducted ecological momentary assessment over 10 days (5x/day) with 75 women ages 40–60 who had ≥1 CVD risk condition (MAge=51.6 years, MBMI=34.0 kg/m2). Using a mix of frequentist and Bayesian analytic approaches, we examined characteristics of women’s naturally occurring comparisons and tested predictions from the Identification/Contrast Model within-person (e.g., identifying with an upward target results in positive affect, whereas contrasting results in negative affect). Comparisons occurred at 21% of moments, with considerable within-person variability in response. In line with predictions from the Identification/Contrast Model, women were more likely to experience positive affect after upward identification or downward contrast and more likely to experience negative affect after upward contrast or downward identification, though observed nuances warrant additional consideration. Overall, findings support the Identification/Contrast Model to describe women’s comparison experiences as they occur in daily life. Future work should determine pathways between immediate consequences of comparisons and longer-term health outcomes.

Keywords: social comparison, ecological momentary assessment, midlife, women’s health, social influence, theory testing

Introduction

Both theory and evidence indicate that people are naturally driven to evaluate themselves relative to others by making social comparisons (i.e., appraisal of oneself compared to others; Festinger, 1954; Gerber et al., 2018), and that comparisons influence physical and mental health outcomes across many populations (Arigo, Smyth, & Suls, 2014; Samra et al., 2022). Yet, social comparison processes warrant additional attention among women in midlife (ages 40–60; Brim et al., 2019). This group is particularly vulnerable to health concerns related to aging, the menopause transition, and gendered socialization that contribute to low quality of life and early mortality (Matthews et al., 2009; Mikkola et al., 2013). For many women, the midlife period also involves stress from shifting professional and social roles, increased caregiving responsibilities, financial concerns, depressive symptoms, and sleep problems (Freeman, 2015; Infurna et al., 2020; Polo-Kantola, 2011; Pope et al., 2012; Thomas et al. 2018), and women in midlife report higher levels of anxiety and depression and score lower on measures of mental well-being compared to men and women of other ages (Min et al., 2021; Reeves et al., 2011; Sassarini et al., 2016).

Further, the effects of aging on the cardiovascular system and hormonal changes during menopause increase women’s risk for cardiovascular disease (CVD), which remains the leading cause of death in the U.S. (Mozaffarian, 2015). Gendered socialization and associated learning experiences encourage women to prioritize others ahead of themselves, which can discourage protecting time for cardioprotective behaviors such as physical activity that are strongly recommended to women with CVD risk conditions such as hypertension (Hendry et al., 2010; Pope et al, 2012). We do not yet understand the role of social comparison processes in these tendencies, however; comparisons may either reinforce gendered notions of healthy self-care or buffer against them if healthy role models are available (Arigo et al., 2015; Yun & Silk, 2011). Comparisons may also influence daily affect, which may contribute to longer-term health outcomes. A deeper foundational understanding of how women experience comparisons in their daily lives is an important step toward elucidating influences on their broader social, emotional, and physical health.

Specifically, social comparisons provide feedback on one’s standing in a given domain and opportunities to learn about one’s life circumstances, especially for those experiencing the stress or uncertainty that accompanies chronic illness conditions (Arigo, Suls, & Smyth, 2014; Van Der Zee et al., 1999). Stronger affiliative needs under these circumstances lead individuals to seek more connections and often result in more frequent social comparisons (Buunk & Ybema, 1997). Women in midlife account for the largest proportion of U.S. healthcare costs associated with CVD risk, due to the high prevalence of conditions such as hypertension and type 2 diabetes as well as behavioral risk factors such as smoking and physical inactivity in this group (Centers for Medicare and Medicaid Services, 2020a, 2020b; Kim et al., 2021). The additional stress and uncertainty of such conditions, on top of the stressors of midlife and menopause, may lead women in midlife with conditions that increase their CVD risk to look to others’ standing in their daily lives – for comfort, useful information, or both.

Importantly, multiple aspects of social comparison are known to prompt immediate cognitive or emotional reactions, which can influence physical and mental health (Bandura, 1998; Muller & Fayant, 2010). Experiences such as positive and negative affect in daily life can contribute to longer-term outcomes such as stress, depression, anxiety, illness self-care, and overall well-being (Leger, Charles, & Almeida, 2018; Pressman, Jenkins, & Moskowitz, 2019; Verhees et al., 2021), particularly for women (Shenk et al., 2018; Stewart et al., 2018). At present, it is not clear whether (or which types of) comparisons are beneficial versus not, with respect to their immediate consequences for women in midlife with elevated risk for CVD. For example, making social comparisons has been positively associated with depressive symptoms (Yoon et al., 2019) and negatively associated with body satisfaction (Thøgersen-Ntoumani et al., 2017). Yet, people compare different aspects of themselves (which represent distinct dimensions of comparison), such as their appearance, status, abilities, and health behaviors (Lakerveld et al., 2018; Lee et al., 2018). Although appearance comparisons are associated with body dissatisfaction and disordered eating behavior (Jin et al., 2018), health- and coping-related comparisons can improve mood and motivation for self-care (Arigo, Smyth, & Suls, 2015).

People also have opportunities to make comparisons in different social contexts, or modes; we encounter potential comparison targets in person, over the phone, or online (e.g., on social media), and may only interact with a subset of targets (vs. hearing about or reading about them; Verduyn et al., 2020). Comparisons also differ in their direction: upward (to those perceived as better off), lateral (to those perceived as at the same level), or downward (to those perceived as worse off; Jin et al., 2018; Wood et al., 1989). Comparisons via social media are typically made upward and are associated with greater envy and negative well-being outcomes for women, such as anxiety, depression, and body dissatisfaction (Fitzsimmons-Craft et al., 2012; Watt & Konnert, 2020).

Critically, however, knowledge of social comparison experiences among women in midlife with elevated risk for CVD is primarily informed by studies that focus on a specific comparison mode or dimension. For example, most recent studies of women’s experiences focus on the effects of upward comparisons, made using social media, on negative mental health outcomes (Appel et al., 2016; Fitzsimmons-Craft et al., 2012; Lewallen & Behm-Morawitz, 2016; Pan, 2023). Yet, upward comparisons are also associated with positive outcomes, including improved chronic illness self-care and engagement in physical activity (Arigo & Butryn, 2019; Brakel et al., 2012). Similarly, downward comparisons are associated with both increased motivation to exercise (Diel & Hofmann, 2019) and lower levels of well-being for women with fibromyalgia (Terol Cantero et al., 2021). Thus, the potential affective consequences of social comparisons vary; outcomes may depend on the direction, dimension, or mode of comparison, though this information is typically aggregated over long periods (e.g., in self-report measures; cf. Gibbons & Buunk, 1999). The determinants of these experiences as they unfold are not clear, and social comparison represents a prime candidate for further examination. Specifically, the immediate affective consequences of comparisons in daily life may be most influenced by how women interpret a given comparison in the moment, though little is known about this process.

Immediate Consequences of Social Comparison: The Identification/Contrast Model

The Identification/Contrast Model of social comparison proposes that the extent to which a person focuses on similarities with (vs. differences from) the comparison target is key for determining their immediate response (Buunk & Ybema, 1997). Identifying, or focusing on similarities, with an upward target should lead to positive affect, because it highlights the comparer’s similarity with someone doing well and suggests that this target’s status is attainable. Similarly, contrasting oneself against a downward target leads to positive affect, because the comparer distances themselves from someone doing poorly; they see themselves as doing well relative to that person and perceive the target’s status as an unlikely outcome for themselves (Vogel et al., 2014). Alternatively, contrasting oneself against an upward target and identifying with a downward target should lead to negative affect, because they create distance between oneself and a more desirable target (Lewallen & Behm-Morawitz, 2016) or closeness with a worse-off target whose status may represent the comparer’s future state, respectively.

Initial evidence suggests this model may have utility for women in midlife with elevated risk for CVD (e.g., hypertension, type 2 diabetes). For example, stronger (vs. weaker) identification with upward targets has been linked to greater motivation for diabetes self-care behaviors (Arigo et al., 2015), and both upward identification and downward contrast are associated with positive affect and effective coping for women with cancer (Brakel et al., 2012; Wood et al., 1985). Conversely, downward identification and upward contrast have been associated with psychological distress for women with fibromyalgia and as well as with lower motivation for self-care behaviors among adults with type 2 diabetes (Arigo et al., 2015; Terol Cantero et al., 2020). Thus, identification and contrast may have important implications for mental health and illness self-care in this at-risk group. To date, however, no study has tested the Identification/Contrast Model among women in midlife – broadly or among those with elevated CVD risk – or examined the roles of dimension, mode, or direction on women’s immediate reactions to comparisons. This gap has left a fragmented understanding of the roles of different aspects of social comparison (and no unifying model for how they operate), particularly among women in midlife with elevated risk for CVD. Additional information in this area could identify opportunities to mitigate negative health outcomes or promote positive health outcomes for this and other at-risk populations.

The Benefits of Intensive Assessment Methods for Examining Social Comparison Processes

Use of intensive assessment methods to address the experience of social comparisons carries unique advantages. Many existing studies of comparison have used cross-sectional designs and/or examined comparison experiences by asking participants to recall over long periods using global, retrospective self-reports (e.g., Gibbons & Buunk, 1999; Verduyn et al., 2020). Yet, comparisons occur frequently in the context of daily life, often without warning or intention, and with limited awareness (Bocage-Barthélémy et al., 2018). Asking people to report on their comparison experiences over long periods, or based on what they “typically” experience, can lead to forgetting and biased recall (e.g., of comparisons that led to particularly intense reactions; Arigo et al., 2020). As a result, these methods have limited ecological validity, and the details of the internal comparison experience are often lost, even if the comparer experiences noticeable affective consequences.

Further, optimal methods to assess comparison experiences would account for how comparison experiences differ between persons and how they vary for the same person at different times (Arigo et al., 2020; Baldwin & Mussweiler, 2018; Perski et al., 2022). Global self-report measures capture only the former, whereas the latter would reveal how comparisons may be experienced differently at different times (e.g., with respect to dimension, mode, or direction). Ecological momentary assessment (EMA) captures people’s experiences in their natural environments by conducting brief, ambulatory assessments multiple times per day (Smyth et al., 2017), and is increasingly popular in psychological science. As a result, EMA is well-suited to deepening our understanding of social comparison occurrences, identification and contrast processes, and their immediate consequences in daily life. Use of EMA could also help to explain how the day-to-day experience of comparisons may be linked to longer-term emotional and physical health outcomes such as body image, depressive symptoms, and CVD risk.

Aims of the Present Study

The present study was designed as an initial step toward these goals. As noted, much existing research among women emphasizes comparisons of appearance and neglects comparison of other attributes that may be important to women in this age group who have risk markers for CVD (e.g., abilities, health behaviors). Few studies have examined the influence of comparison mode, and no existing study has used an intensive assessment approach to examine identification and contrast processes within-person (among women in midlife with elevated CVD risk or more broadly). The present report uses data from a larger EMA protocol that examined women’s experiences of social comparison in daily life, with attention to dimension, mode, direction, and identification/contrast (Arigo et al., 2020). This protocol was designed to capture multiple potential psychosocial determinants of health behaviors in women’s daily lives (e.g., social interactions, body satisfaction, social comparisons). Previous reports have described associations between these momentary experiences and subsequent physical activity behavior (Arigo, Mogle, & Smyth, 2021). Mode of comparison and identification and contrast processes were not examined in prior reports, as these were of particular interest and reserved for the present, more detailed investigation.

Our first aim was to describe the overall frequencies and variability in women’s comparison experiences within-person. Our second and primary aim was to examine differences in women’s immediate responses to comparisons (i.e., positive and negative affect) based on comparison context, including dimension, mode, direction, and identification/contrast. We tested hypotheses proposed by the Identification/Contrast Model, to determine whether this model accurately captures women’s experiences in the moment. Specifically, we predicted that when women made upward comparisons, they would be more likely to report experiencing positive affect at times when they identified with these targets (vs. not) and more likely to report experiencing negative affect at times when they contrasted against these targets (vs. not). Conversely, we predicted that when women made downward comparisons, they would be more likely to experience negative affect at times when they contrasted against these targets (vs. not) and would be more likely to experience negative affect at times when they identified with downward targets (vs. not).

Methods

Recruitment and Participants

Women were recruited through online and print advertisements, as well as through physician referrals and warm handoffs to research staff in university-affiliated primary care offices, from February 2019 to March 2020. Women between the ages of 40–60 (inclusive) were eligible if they endorsed smoking tobacco (current or quit in the last 3 months) or reported receiving a physician’s diagnosis of one or more of the following CVD risk factors: hypertension, prehypertension, prediabetes, type 2 diabetes, hyperlipidemia, hypercholesterolemia, or metabolic syndrome. Additional inclusion criteria were fluency in English, not currently pregnant, no medical or psychiatric symptoms that would interfere with participation (e.g., active psychosis), and access to a personal mobile device for completing EMA surveys. We selected a target sample size of 100 participants with 5 momentary assessments per day for 10 days (total = 50 assessments) to support within-person analyses such as those described below at power ≥ 0.85, as well as exploratory between-person moderation of within-person associations (Maas & Hox, 2005). Although the onset of the COVID-19 pandemic ended data collection earlier than anticipated, the resulting sample of 75 women with high protocol compliance still afforded overall power > 0.80 for within- person tests (see below for additional details). Participants were 75 women (MAge = 51.6 years, MBMI = 34.0 kg/m2); the majority (73%) identified as white, 22% as Black, 2% as Hispanic/Latina, 1% as Asian-American, and 3% mixed or other. The largest subset of participants was married (56%) and postmenopausal (39%). Additionally, 52% reported having a diagnosis of hyperlipidemia or hypercholesterolemia and 59% of participants had BMI in the obese range (M = 34.02; SD = 7.13). Please see Table 1 for additional demographic information.

Table 1.

Demographic information for participants (N = 75).

Demographics M (SD) n (%)
 Age 51.61 (5.43) Hispanic/Latina Ethnicity 2 (3)
 BMI 34.02 (7.13)
Education n (%) Household Income n (%)
 High School/GED 7 (9)  Under $25,000 5 (7)
 Associate’s degree/technical degree 16 (21)  $25,001–$50,000 12 (16)
 Bachelor’s degree 24 (32)  $50,000–$75,000 12 (16)
 Graduate degree 28 (37)  Over $75,000 45 (61)
CVD Risk Conditions n (%) Marital Status n (%)
 Hypercholesterolemia or Hyperlipidemia 39 (52)  Married 44 (59)
 Hypertension or Prehypertension 35 (47)  Widowed 4 (5)
 Type 2 diabetes 30 (40)  Divorced 11 (15)
 Metabolic syndrome 8 (11)  Separated 4 (5)
 Smoker (or quit in previous 3 months) 10 (14)  Never Married 12 (16)
Racial Identification n (%) Menopause Status n (%)
 White 55 (73)  Pre-menopause 14 (20)
 Black/African American 16 (22)  Perimenopause 16 (23)
 Asian American 1 (1)  Post-menopause 29 (39)
 Mixed/Other 1 (1)  Other 12 (17)
Open in a new tab

Measures

An initial electronic survey included demographic questions such as age, CVD risk conditions, education level, ethnicity, race, and menopause status. Height and weight were measured by research staff in the clinic and were used to calculate BMI. Social comparison experiences were captured via electronic survey 5 times per day for 10 consecutive days; items were developed with guidance from similar published work (Leahey, Crowther, & Mickelson, 2011; Wheeler & Miyake, 1992). We then undertook multiple rounds of pre-testing with this population prior to use (Arigo et al., 2021). We conducted a 7-day pilot EMA study with an initial set of items (N = 13; 5 assessments per day). Social comparison and other psychosocial experiences were reported less frequently than expected, suggesting potential challenges with recognizing comparisons in daily life. We conducted cognitive interviewing with a different sample of women (N = 10), which highlighted opportunities to improve the clarity of items and instructions. We then revised the items with their input and conducted a second round of EMA pilot testing with a different sample of women, again over 7 days (N = 13; 5 assessments per day). As item responses and narrative feedback indicated improved understanding and recognition of comparisons in daily life with low reporting burden, the revised set of items was used in the present study (Arigo et al., 2021).

Comparison items assessed the occurrence (yes/no), number, direction(s), and dimension(s) of any comparisons made in the relevant reporting window (see below). If a participant indicated “yes” to the occurrence of comparison, they were asked to report the details of their most recent comparison: direction, dimension, mode, whether they focused on similarities and differences from the target (separate items), and whether they experienced positive and negative affect after the comparison (separate items). Focusing on similarities and differences, as well as positive and negative affect, were originally assessed on a 3-point scale (0 = not at all, 1 = somewhat, 2 = very much). Based on the distribution of responses and for ease of interpretation, responses were dichotomized to represent not at all (0; all original responses of 0) and any (1; all original responses of 1 or 2). The full set of social comparison EMA items are presented in Table 2.

Table 2.

Ecological momentary assessment (EMA) items capturing social comparison in daily life.

Survey Item Response Options
In the past 3 hours/since you woke up this morning, did you think about or evaluate yourself or your behavior in comparison to someone else (or someone else in comparison to yourself)?  • Yes
 • No
How many times did you compare yourself in the last three hours/since you woke up this morning?  Numeric entry
Did you communicate with the person or people you compared to?  • Yes. I talked to them in person, or on the telephone or online.
 • No. I saw, heard about, read about, or thought about them but did not communicate.
 • Both
What aspect(s) of yourself did you compare? (Choose all that apply)  • Appearance (height, weight, shape, attractiveness, body part)
 • Health habits (physical activity, eating behavior, sleep)
 • Status (wealth, work position, possessions)
 • Emotions (how you responded to a positive or negative situation)
 • Personality (outgoing, perfectionistic)
 • Abilities (skills, talents)
 • Other (please specify)
Below are some types of comparisons you may have made. Please indicate how many of each type you made in the past three hours/since you woke up this morning. Comparisons to people who
 • Seem to be doing better than I am ___
 • Seem to be doing the same as I am ___
 • Seem to be doing worse than I am ___
 Numeric entry
Now consider only your most recent comparison. Did you communicate with the person you compared to?  • Yes. I talked to them in person, or on the telephone or online.
 • No. I saw, heard about, read about, or thought about them but did not communicate.
 • Both
What aspect of yourself did you compare?  • Appearance (height, weight, shape, attractiveness, body part)
 • Health habits (physical activity, eating behavior, sleep)
 • Status (wealth, work position, possessions)
 • Emotions (how you responded to a positive or negative situation)
 • Personality (outgoing, perfectionistic)
 • Abilities (skills, talents)
 • Other (please specify)
As you compared yourself, how much did you focus on each of the following: how similar I am to the person I compared to?; how different I am to the person I compared to?  • Not at all
 • Somewhat
 • Very much
 *Recoded to 0 (not at all) vs. any (1)
After the comparison, how much did you feel each of the following: inspired, encouraged, or hopeful about my own situation; anxious, frustrated, or discouraged about my own situation?  • Not at all
 • Somewhat
 • Very much
 *Recoded to 0 (not at all) vs. any (1)
Open in a new tab

Procedures

As part of a larger EMA study, all procedures were approved by the Institutional Review Boards at the supporting institutions and followed a registered protocol (Arigo et al., 2020). Women who expressed interest completed a 10-minute phone screening with a member of the research team and an initial electronic survey. A member of the research team then met with each participant individually for 60 minutes at the research center; these initial sessions included written documentation of informed consent and measurement of height and weight. Participants received a detailed overview of study procedures, including detailed information about EMA items, and were asked to complete surveys within 1 hour of receipt. To mitigate known hesitation to reporting the occurrence of comparisons (as some people frame it as negatively judging other people; Arigo, Mogle, Brown, & Gupta, 2021; Helgeson & Mickelson, 1995), comparisons were introduced as neutral: “there is nothing inherently good or bad about making comparisons, though they may have positive or negative consequences for us as we go about our day” (cf. Gibbons & Bunk, 1999). Participants were guided to define social comparison in the context of this study: “comparison” was defined as any instance of considering oneself in relation to another person (rather than only comparisons that elicited a strong psychological response; see Patrick, Neighbors, & Knee, 2004). Research staff also gave examples of comparisons, including noticing that a coworker is more or less efficient than you are or that someone has something that you do not have. Participants were encouraged to develop their own examples, which were discussed with research staff, and were told what to expect in the EMA survey (e.g., reporting on the number, direction, dimension, etc. in lay language).

The EMA protocol used a signal-contingent approach; signals (prompts) came to participants’ smartphones with an embedded link to the relevant survey on the Qualtrics platform. Surveys were sent on a semi-random schedule at 5 times between standard sleep/wake times for a 9:00 AM to 5:00 PM work schedule (8:30 AM, 12:15 PM, 3:45 PM, 6:30 PM, 9:45 PM), an early-rising schedule (6:30 AM, 9:15 AM, 12:00 PM, 3:30 PM, 6:45 PM), and a late-rising schedule (11:00 AM, 2:35 PM, 5:45 PM, 8:45 PM, 11:15 PM). Three versions of these schedules were created by the research team, which had different distribution times anchored to the times above, and a random number generator was used to assign each participant to a specific schedule that matched their sleep/wake routine. On all schedules, surveys were distributed at different times each day for 10 consecutive days (including weekends). Following this EMA period, participants completed an in-person exit interview at the research center, where they provided feedback on their experience. Participants received $15 for the orientation meeting, $30 for the follow-up interview, and a $10 bonus for completing ≥80% of EMA surveys.

Data Analysis

Initial analyses were conducted in SAS Version 9.4 (Cary, NC). Compliance was high: participants completed 90% of surveys overall and completed 80% within the designated 1-hour time window. Only those surveys completed within 1 hour of EMA signals were included in analyses; of these, 99% included responses to social comparison items. The resulting 2951 observations afforded sufficient power for the tests reported below (Murayama et al., 2022), with exceptions as described for lower-frequency combinations of direction, identification/contrast, and affective response (which were handled using a Bayesian approach). Participants indicated any social comparisons in the reporting window (~3 hours prior to receiving the prompt) versus not as yes/no, which was treated as a binary outcome. Similarly, reports of experiencing comparisons on any of the dimensions provided (e.g., appearance, health, status, other) were dichotomized as yes/no; as noted, focusing on similarities and differences from the most recent comparison target (i.e., identification and contrast) and experiencing positive and negative emotions in response to that target (total of 4 items) were also dichotomized as yes/no, and all of these variables were treated as binary outcomes.

The total number of comparisons reported and the numbers of upward, lateral, and downward comparisons were treated as continuous. These totals were zero-inflated when all valid responses were included and were positively skewed when only moments with any comparisons were included. Consequently, all tests of comparison type and response to were conducted using PROC GLIMMIX, with maximum likelihood estimation to flexibly address missingness (which was minimal). The likelihood of reporting the occurrence of comparison, the number of comparisons reported, and response to most recent comparisons all did not differ by day of observation, and day of observation did not moderate any of the temporal patterns described below (ps > 0.40), suggesting little (if any) reactivity to reporting on social comparisons. As controlling for observed differences between surveys and weekdays versus weekends (described below) did not meaningfully change conclusions, we report on models without these covariates, to maintain parsimony.

We initially describe the frequencies of all observed aspects of the comparison process: number, directions, and dimensions across all comparisons in the reporting window, and directions, dimensions, modes, identification/contrast, and affective responses to the most recent comparison. Empty models were used to determine overall frequencies and to calculate intraclass correlation coefficients (ICCs); the latter were used to determine the extent of stability (vs. variability) in participants’ experiences of social comparisons. To test hypotheses about identification and contrast processes in daily life, we then examined the most recent comparison in detail to determine the likelihood of specific responses (e.g., positive and negative affect) to certain types of comparisons (e.g., upward vs. downward). We used a frequentist approach for these analyses that employed multilevel logistic regression. These models controlled for the relevant person mean to distinguish between- and within-person associations (cf. Curran & Bauer, 2011; Hoffman, 2015). Statistical significance was set to p < 0.05 and effect sizes are expressed as odds ratios.

We observed that certain combinations of comparison direction, identification/contrast, and affect were less common in the present dataset than others (e.g., negative affect in response to contrasting against a downward target on 1% of occasions). To determine whether these lower-frequency experiences systematically followed the pattern predicted by the Identification/Contrast Model, we used generalized linear mixed modeling (GLMM) with Bayesian estimation. This approach complemented our frequentist models to examine the strength of the evidence for the Identification/Contrast Model, rather than relying on a binary significant/non-significant interpretation. GLMM with a binomial distribution and logit link were fit using uninformative priors in JASP (Love et al., 2019). All models used 3 chains and 4,000 iterations after 2,000 burn-in samples. Markov Chain Monte Carlo (MCMC) diagnostic plots were examined for predictors to ensure model stability prior to model interpretation. We report the average estimated relation between our target predictor (e.g., identification or contrast) and target outcome (i.e., positive or negative affect) as well as the 95% highest posterior density interval (HPDI). We used the HPDI as evidence of the extent to which the effect of interest was in the direction predicted by the Identification/Contrast Model.

Results

Across 10 days of observation, participants reported making comparisons at 21% of moments (621/2951). Comparisons were more likely to occur at the first or second survey of the day (relative to surveys 3–5; t[2870] = 2.63, p = 0.01, OR = 1.31, CI 1.07–1.61) and on weekdays (vs. weekends; t[2875] = 2.27, p = 0.02, OR = 1.30, CI 1.04–1.63). Each participant made an average of 14.74 comparisons (SD = 16.71). The number of comparisons per person across days ranged from 0 (3 participants) to 84 (1 participant). Participants reported an average of 0.36 comparisons per moment (SE = 0.05; range 0–10), representing the number of comparisons made since they woke up (first survey of each day) or in the past 3 hours (all other surveys). Approximately 20% of the variability in the number of comparisons reported was attributable to between-person stability (ICC = 0.20), with the majority of variability attributable to within-person variability across moments (and error).

As shown in Table 3, upward comparisons were most common with respect to direction (11% of all moments), followed by downward (8.8%) and lateral (8.5%). With respect to dimension, comparisons of abilities were most common (9.4% of all moments), followed by comparisons of appearance (6.6%). Participants’ most recent comparisons also followed these patterns (see Table 3): upward comparisons (43% of most recent) and comparisons of abilities (23% of most recent) were most common. Of note, comparisons on “other” dimensions included work ethic, responsibility, common sense, kindness, leisure time, decisions, age, hygiene, boundaries, life situation, parenting, menstrual symptoms, manners/ etiquette, spending habits, housework, motivation, and religious commitment. These represented 9.2% of the most recent comparisons (reported at 1.8% of all moments). Of note, participants also listed comparisons of dimensions such as social status and driving ability; these could have been characterized as comparisons of status and ability, respectively, though participants elected to list them separately.

Table 3.

Descriptive statistics for social comparison EMA responses (n = 2951 valid observations).

All Moments B (SE) ICC Most Recent Comparison (Y/N) Odds of Yes B (SE) ICC
 Total number 0.36 (0.05) 0.20  Focus on similarities 0.59 (0.05) 0.12
 Number of upward 0.16 (0.02) 0.11  Focus on differences 0.91 (0.02) 0.37
 Number of lateral 0.14 (0.02) 0.11  Positive affect 0.65 (0.04) 0.14
 Number of downward 0.09 (0.01) 0.07  Negative affect 0.62 (0.04) 0.14
 Interacted with target 0.46 (0.03) 0.05
Comparison Dimension: n {%) of all valid observations / n {%) of most recent comparisons
Appearance Health habits Status Emotion regulation Personality Abilities
197 (7%) / 82 (17%) 155 (5%) / 51 (11%) 166 (6%) / 51 (11%) 181 (6%) / 79 (17%) 140 (5%) / 35 (7%) 279 (9%) / 135 (28%)
Other 54 (2%) / 44 (9%)
Comparison Direction: n {%) of valid observations / n {%) of most recent Comparison Direction & Identification/Contrast: n {%) of most recent
Upward Lateral Downward Y/N Upward Downward
 322 (11%) / 234 (43%) 252 (8.5%) / 164 (30%) 261 (8.8%) / 145 (27%)  Focus on similarities 123 (24%) / 96 (18%) 73 (14%) / 68 (13%)
 Focus on differences 212 (39%) / 21 (4%) 138 (26%) / 6 (1%)
Upward Comparisons (Most Recent) Positive Affect (frequencies Y/N) Negative Affect (frequencies Y/N) Downward Comparisons (Most Recent) Positive Affect (frequencies Y/N) Negative Affect (frequencies Y/N)
 Focus on similarities 75 (35%) / 46 (21%) 91 (42%) / 29 (13%)  Focus on similarities 56 (40% / 16 (12%) 29 (21%) / 44 (32%)
 Focus on differences 104 (47%) / 98 (44%) 165 (73%) / 42 (19%)  Focus on differences 99 (69%) / 38 (27%) 53 (38%) / 82 (58%)
Open in a new tab

Note: Y/N = yes/no.

Most Recent Social Comparison: Mode, Similarities, Differences, and Affective Responses

Participants’ most recent comparisons occurred more often when hearing, reading, or thinking about the comparison target, rather than when communicating or interacting with them (interaction at 45% of most recent comparisons), though this varied considerably within-person (ICC = 0.05). Similarly, whether participants focused on similarities with their most recent comparison target (vs. not), as well as whether they experienced positive and negative affect after this comparison (vs. not), showed considerable variability within-person (ICCs = 0.12, 0.14, 0.14, respectively). Whether participants focused on differences from their most recent comparison target (vs. not) showed noticeably more stability (ICC = 0.37), though the majority of variability was within-person (and error). The odds of focusing on similarities (vs. not) was 4 times higher with a lateral target than with upward and downward targets (contrast t[456] = 5.12, p < 0.0001, OR = 4.00, 95% CI = 2.35–6.81), and was nearly 2 times higher with an upward than a downward target (contrast t[456] = 2.21, p = 0.03, OR = 1.92, 95% CI = 1.07–3.42; see Table 3 for frequencies and Table 4 for multilevel model estimates). The opposite patterns occurred for focusing on differences from the target, though the direct contrast between moments with upward and downward targets as most recent was not significant (t[472] = −1.83, p = 0.07). Focusing on similarities, focusing on differences, experiencing positive affect, and experiencing negative affect all did not significantly differ based on the dimension of comparison (ps > 0.06).

Table 4.

Multilevel model estimates and 95% highest posterior density interval (HPDI) summaries for tests of identification, contrast, and positive and negative affect in response to social comparisons.

All Comparison Directions Identification Y/N Contrast Y/N Positive Affect Y/N Negative Affect Y/N
B (SE) B (SE) B (SE) B (SE)
 Intercept 0.34 (0.24) 2.54 (0.29)** 0.16 (0.19) 1.38 (0.21)**
 Downward −0.65 (0.29)* 0.92 (0.51) 0.89 (0.26) −1.77 (0.27)**
 Lateral 1.06 (0.28)** −1.09 (0.32)** 0.89 (0.25) −1.41 (0.26)
 Upward -- -- -- --
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)
 Direct Contrast: Upward vs. Lateral/Downward 0.81 (0.51–1.30) 1.08 (0.55–2.14) 0.41 (0.27–0.62) 4.90 (3.13–7.66)
Positive Affect (Y/N) Negative Affect Y/N
Upward Comparisons Only Identification Y/N Contrast Y/N Identification Y/N Contrast Y/N
B (SE) B (SE) B (SE) B (SE)
 Intercept 0.58 (0.26)* 0.12 (0.23) 1.15 (0.50) 1.44 (0.25)
 Predictor (Y/N) 1.20 (0.34)** −0.38 (0.42) −0.68 (0.34) 0.43 (0.54)
OR (95% CI) OR (95% HPDI) OR (95% HPDI) OR (95% HPDI)
3.31 (1.68–6.52) 0.68 (0.29–1.53) 0.51 (0.25–0.96) 1.54 (0.58–4.72)
Positive Affect (Y/N) Negative Affect Y/N
Downward Comparisons Only Identification Y/N Contrast Y/N Identification Y/N Contrast Y/N
B (SE) B (SE) B (SE) B (SE)
 Intercept 1.82 (0.54) 1.16 (0.34)** −0.44 (0.31) −0.48 (0.24)*
 Predictor (Y/N) 1.10 (0.57) 0.13 (1.07) 0.07 (0.41) −0.41 (0.92)
OR (95% HPDI) OR (95% HPDI) OR (95% HPDI) OR (95% HPDI)
2.34 (1.08–5.78) 1.29 (0.23–6.41) 1.07 (0.61–1.95) 0.78 (0.18–3.34)
Open in a new tab

Note: -- = reference group; model estimates for upward and downward comparisons reflect the likelihood of yes (1) and control for relevant person means; HPDI = Highest Posterior Density Interval

t

= 0.05

*

p ≤ 0.05

**

p ≤ 0.01.

With respect to affective responses, participants were 2 times more likely to experience positive affect if they interacted with the comparison target (vs. not; t[526] = 3.41, p = 0.0007, OR = 2.00, 95% CI = 1.34–2.98), and 44% less likely to experience negative affect with interaction (vs. without; t[531] = −2.95, p = 0.003, OR = 0.56, 95% CI = 0.38–0.82). Participants were also 59% less likely to experience positive affect (contrast t[459] = −4.21, p < 0.0001, OR = 0.41, 95% CI = 0.27–0.62) and almost 5 times more likely to experience negative affect (contrast t[462] = 7.13, p < 0.0001, OR = 4.98, 95% CI = 3.20–7.75) in response to upward comparisons, relative to lateral and downward comparisons. Importantly, however, and in line with our hypotheses, participants were more than 3 times more likely to experience positive affect in response to upward comparisons if they focused on similarities (vs. not; t[160] = 3.09, p = 0.002, OR = 3.27, 95% CI = 1.53–6.97).

Given the small percentage of observations in remaining identification/contrast categories (i.e., affective responses to focusing on differences with upward targets and focusing on either similarities or differences with downward targets), we examined the HPDI for each combination. At times when participants focused on differences from upward targets, positive affect was less likely on average (b = −0.40, 95% HPDI = −1.26, 0.41) and negative affect was more likely (b = 0.43, 95% HPDI = −0.55, 1.55). At times when participants focused on similarities with upward targets, negative affect was less likely on average (b = −0.68, 95% HPDI = −1.40, −0.05). Conversely, at times when participants focused on differences from downward targets, positive affect was more likely on average (b = 0.255, 95% HPDI = −1.453, 1.858) and negative affect was less likely on average (b = −0.254, 95% HPDI = −1.708, 1.206). At times when participants focused on similarities with downward targets, negative affect was more likely on average (b = 0.069, 95% HPDI = −0.501, 0.669). These patterns all align with our predictions informed by the Identification/Contrast Model. There was one exception, however: at times when participants focused on similarities with downward targets, positive affect was more likely on average (b = .849, 95% HPDI = 0.076, 1.755), rather than less (as expected in alignment with the Identification/Contrast Model).

Discussion

Social comparisons naturally occur during daily life and show associations with physical and mental health outcomes (Arigo, Gray et al., 2023; Thøgersen-Ntoumani et al., 2017; Yoon et al., 2019). Yet, the potential mechanisms linking social comparison experiences to these health outcomes remains poorly understood; in particular, it is not clear what determines the immediate positive or negative consequences of comparisons in daily life, which may represent the first step in a longer chain of effects. For example, among women in midlife with elevated CVD risk, identification with upward targets may result in positive affect that buffers against the unique stress they experience (Stewart et al., 2018), or that motivates engagement in behaviors (e.g., illness self-care) to achieve the target’s better-off status (Arigo et al., 2015). Examining social comparison experiences as they unfold in daily life can help to identify the nuances of specific contexts associated with immediate positive and negative consequences, which has implications for both theoretical and applied work moving forward.

The present findings align with prior work that documents social comparison as a common experience (Wheeler & Miyake, 1992). Women in this study reported making 0–84 comparisons per person over 10 days (~15 comparisons per person), though their comparisons varied considerably within- person. These reports may even underestimate the frequency of social comparisons, which are not always consciously processed and awareness of them may fluctuate (Bocage-Barthélémy et al., 2018). Women in the present study also compared on a variety of dimensions. Although comparisons of abilities and appearance were most common, comparing status, health behaviors, and emotions were also popular, and many participants indicated comparing other aspects of themselves (e.g., common sense). Some of these entries were for dimensions that fit into the categories provided (e.g., driving ability not designated as ability, kindness not designated as a personality trait), which may reflect a misunderstanding of comparison dimensions or errors resulting from efforts to balance study tasks with many other competing priorities (Hendry et al., 2010). The range of dimensions reported may reflect women’s social-mindedness and particular interest in evaluating their abilities or status (Guimond & Chatard, 2014). Dimension may influence reactions, if women particularly value a given domain or if the context encourages comparisons, and additional work is needed to examine the role of dimension in comparison reactions.

Similarly, the direction of women’s comparisons, and whether they focused on similarities and/or differences with these targets (reflecting components of identification and contrast processes), also fluctuated within-person in daily life. In line with potential efforts toward self-improvement and affiliation (Helgeson & Mickelson, 1995), as well as efforts to support a positive self-concept (Arigo & Butryn, 2019; Brakel et al., 2012; Buunk & Ybema, 2003), women were twice as likely to focus on similarities with upward targets (vs. not) and four times as likely to focus on similarities with lateral targets (vs. not). Women’s inclination for identification with lateral and upward targets is one reason that the presence of similar-level peers and role models who recently achieved a desired outcome may have a particularly strong influence for women in midlife (cf. Rowland et al., 2018).

To our knowledge, this is the first evidence documenting identification and contrast as within-person, variable processes (rather than as stable, person-level traits). In addition to advancing our basic understanding of social comparison, this information is critical to avoiding common pitfalls such as assuming that people will be inspired by (or feel dejected by) exposure to upward targets, as might be inferred from global self-reports (e.g., Van der Zee et al., 2000). The present findings show that identification and contrast are contextually bound, and an essential next step will be to determine the contexts that lead to identification, contrast, or a combination – particularly for populations with existing health risks that may be affected by social comparisons in daily life, such as women in midlife with health conditions that increase their risk for CVD. Equally important will be to identify the downstream consequences of identification and contrast processes as they relate to health outcomes, for this population and more broadly.

Does the Identification/Contrast Model Accurately Describe Women’s Comparisons in Daily Life?

Overall, the present within-person findings were consistent with the Identification/Contrast Model of social comparison (Buunk & Ybema, 1997). Women were more likely to report feeling negative affect (i.e., anxious, frustrated, or discouraged vs. not) when identifying with downward targets or contrasting against upward targets. Conversely, women were more likely to report feeling positive affect (i.e., inspired, confident, hopeful vs. not) when identifying with upward targets and contrasting against downward targets. Only one finding was inconsistent with the Identification/Contrast Model: women were more likely to experience positive affect (vs. not) when identifying with downward targets, though the model would suggest that this outcome should be less likely. It is possible that in these situations, although women recognized the target’s worse-off status, their acknowledgement of similarities promoted affiliation. Such circumstances may reflect Schacter’s (1959) early observation that “misery likes miserable company,” as recognition that we are not alone in our “misery” can be comforting. Women were also more likely to report feeling positive affect (vs. not) after comparing with others who they actively engaged with, relative to those they learned about from a distance. Having contextual cues such as tone of voice or visuals may also promote affiliation and its expected benefits for one’s internal experience and might buffer against any potential negative consequences of downward identification. This will be an important hypothesis to test in future work, as it requires repeated observation of a specific combination of circumstances that is beyond the scope of the present study.

Together, the current findings demonstrate that the Identification/Contrast model accurately describes experiences of naturally occurring social comparisons among women in midlife with elevated risk for CVD. These findings have important implications for health in the specific population of interest and possibly, more broadly. As noted, this population experiences greater psychological distress than men and women of other age groups, as well as high rates of depression, anxiety, and physical inactivity that further exacerbate CVD risk (Min et al., 2021; Reeves et al., 2011; Sassarini et al., 2016; Stewart et al., 2018). The present findings show that social comparisons are common in their daily lives and that interpretations of these experiences in their natural environments have predictable consequences for their immediate affect. These consequences may have short-and long-term implications for health: indirectly, via behaviors such as chronic illness self-care and physical activity (Arigo & Butryn, 2019; Arigo, Mogle, & Smyth, 2021; Brakel et al., 2012), or directly, via stress responses and related physiological processes (Shenk et al., 2018). As women in midlife represent a large, diverse subset of the population with substantial healthcare costs as well as workforce and caregiving responsibilities (Hardy et al., 2018; Infurna et al., 2020; Thomas et al. 2018), advancing our understanding of social comparisons and their effects in women’s daily lives will enable improvements in health promotion interventions for this group.

To that end, the present findings also raise critical questions about the use of social comparison opportunities in health promotion interventions. For example, women in midlife with CVD risk (and many other at-risk populations) are often encouraged to interact with one another in group interventions for health behavior change, with social comparison as a key motivator that is activated in these settings (Arigo et al., 2022; Olander et al., 2012). Use of leaderboards and competitive challenges are also popular in digital interventions and are designed to induce relevant comparisons between users (Arigo, Brown, Pasko, & Suls, 2020), with the assumption that such comparisons motivate a behavior of interest. The Identification/Contrast model – and the present findings in particular – indicate that social comparison might facilitate or undermine such efforts, as they can activate negative affective experiences that are associated with decreased engagement in healthy behavior. The model also does not offer explicit predictions about identification and/or contrast with lateral targets; these are often ignored, in part because they tend not to prompt noteworthy cognitive or emotional responses. Thus, the current study highlights the need for future work to determine the generalizability of the Identification/Contrast Model in daily life and its implications for health promotion, especially in at-risk groups.

Strengths, Limitations, and Additional Future Directions

This study used a registered and pre-tested EMA design to examine social comparison experiences within-person, in an at-risk clinical population (i.e., women in midlife with one or more CVD risk factors). Compliance with the EMA protocol was high and responses captured a range of comparison experiences. This study was among the first to use an intensive assessment approach to examine identification and contrast processes as they occur in daily life, and the first to do so in the at-risk population of interest. EMA methodology is particularly advantageous for this line of inquiry, as the details of social comparison experiences can be difficult to capture retrospectively. We also combined the unique strengths of frequentist and Bayesian analytic approaches to maximize knowledge generated.

As noted, however, a fine-grained understanding of social comparisons as they occur – including immediate responses as predicted by the Identification/Contrast Model – is only one step in mapping the pathway(s) linking social comparison to health outcomes. Specifically, there may be more than one pathway from social comparison to health: upward identification may inspire behavior change via positive affect (e.g., hope, increased self-efficacy), though downward identification may inspire similar change via negative affect (e.g., anxiety, increased motivation to avoid a similar situation; Wills, 1981). In our prior work with the present dataset, at times when women in midlife with elevated CVD risk made more (vs. fewer) comparisons than they usually did, their subsequent physical activity increased 42% of the time and decreased 58% of the time (Arigo et al., 2021). The present findings suggest that identification and contrast processes may explain these contextual distinctions in associations with subsequent physical activity (i.e., a key cardioprotective and self-care behavior for this population), though additional work is needed to capture the full mechanistic pathway(s) in daily life.

Specifically, it will be critical for future research to examine whether and how identification, contrast, and their affective consequences predict health indicators (e.g., engagement in health behaviors, direct effects such as changes in cortisol), for whom and under what circumstances, as well as to what extent these associations are causal (vs. co-occurring). This work will be most informative with a considerable sample size (to detect between-person differences) and a considerable number of observations (to detect within-person, contextual distinctions and combinations of identification/contrast, affect, health behavior, physiological response), as well as assessment of clinical outcomes (e.g., blood pressure, CVD risk score). The detailed, foundational work offered in the present report and in our prior work offers strong justification for directing the necessary resources to these next steps.

In this vein, future studies may also explore individual differences in within-person patterns of comparison and responses; for example, those who have already developed depression or anxiety compare upward more often than those who do not (McCarthy & Morina, 2020) and may respond more negatively to these comparisons in daily life, which may contribute to symptom maintenance. As these conditions are highly prevalent among women in midlife and exacerbate risk for CVD (Kravitz et al., 2014; Sassarini, 2016), understanding how their symptoms unfold in daily life and individual differences in these patterns could identify optimal intervention targets and contexts in the natural environment. As noted, this will require more careful consideration of opportunities for comparison in interventions to support health, including the potential for improving immediate responses to comparisons (e.g., increased use of upward identification, if this leads to both positive affect and goal-aligned behavior change).

Notable limitations of the present study also point to important next steps in this line of work. The majority of participants identified as white, married, and highly educated; as such, we had limited ability to understand the experiences of women from underrepresented backgrounds, which may be unique. For instance, women who identify with minoritized racial or ethnic groups may choose different comparison targets and respond differently to these targets than white women (cf. Davis & Wu, 2014), and these potential distinctions warrant further attention. In addition, although participants were asked to describe several features of their most recent comparisons (e.g., direction, dimension, mode), assessment of additional characteristics would be informative. For example, the majority of recent comparisons were with people who were not actively engaged with the respondent, and these comparisons may be less likely to trigger positive affect than those that occur with active engagement. Thus, it would be useful to better understand comparisons via digital means (e.g., social media, text message, phone call). Additionally, we did not assess the relationship between the participant and the comparison target. The Identification/ Contrast Model does not propose that this is a determinant of affective response to comparisons (Buunk & Ybema, 1997); the model indicates that perceived similarity in one’s situation is critical, regardless of familiarity with the comparison target, and we sought to minimize reporting burden. However, as evidence suggests differences in who is selected as a relevant comparison target (e.g., young adults compare to friends more often than family; Wheeler & Miyake, 1992) and it is plausible that identification and contrast could differ in closer than more distal relationships, this represents an important avenue for future work.

We also operationalized focusing on similarities and differences, as well as positive and negative affect, as binary outcomes. This was intentional and still allowed us to test hypotheses guided by the Identification/Contrast Model. As this model describes identification and contrast as continuous spectra, however, an optimal approach would be to capture the full extent of these experiences and their variability within-person (vs. all/none). Further, the model and its prior assessment tools have conceptualized identification and contrast as the combined experiences of perceived similarity (cognitive interpretation of similarities/differences) and affect (positive/negative). Our separation of these into separate steps was intentional and guided by prior work (cf. Arigo, Smyth, & Suls, 2015). Although we believe that this is essential for fully mapping the process, we acknowledge that it may not reflect the original intentions of the model.

Finally, we selected women in midlife with elevated CVD risk for the present, initial within-person investigation of the Identification/Contrast Model for the relevance of comparisons to their daily and longer-term health experiences. As this represents only a subset of the larger population for whom comparisons are common and may have important health consequences, however, it is not clear whether the present findings generalize to daily life in other groups. Expanding the use of EMA and similar methods to study social comparison processes in other groups (e.g., women without CVD risk conditions, men, those who identify with sexual and gender minority groups, younger and older adults) will more fully address the potential of the Identification/Contrast Model, for explaining comparison experiences and relevant health outcomes in daily life.

Conclusions

The present findings improve our understanding of naturally occurring social comparisons in daily life, by illustrating that immediate comparison experiences fluctuate within-person – in particular, that identification and contrast processes tend to unfold as proposed by the Identification/Contrast Model of comparison and vary within-person, among women with health risks. These findings have important implications for chronic disease prevention and overall well-being, including the need for more attention to comparison processes in interventions to promote these outcomes. Future work is needed to determine how immediate affective responses influence subsequent health behaviors and longer-term health outcomes, among women in midlife and more broadly.

Funding:

This work was supported by the U.S. National Heart, Lung, and Blood Institute (National Institutes of Health) under award numbers K23 HL136657 and DP2 HL173857 (PI: Arigo).

Footnotes

Conflict of interest: Nothing to disclose.

Ethics approval: This study was approved by the Institutional Review Board at Rowan University under protocol number Pro2018002377.

Contributor Information

Kiri Baga, Department of Psychology, Rowan University; 201 Mullica Hill Road, Robinson H109, Glassboro, NJ 08028.

Gabrielle M. Salvatore, Department of Psychology, Rowan University; 201 Mullica Hill Road, Robinson 118F, Glassboro, NJ 08028.

Iris Bercovitz, Department of Psychology, Rowan University; 201 Mullica Hill Road, Robinson H109, Glassboro, NJ 08028.

Jacqueline A. Mogle, Department of Psychology, Clemson University; 312E Brackett Hall, Clemson, SC 29634.

Danielle Arigo, Department of Psychology, Rowan University; Department of Family Medicine, Rowan-Virtua School of Osteopathic Medicine; 201 Mullica Hill Road, Robinson 116G, Glassboro, NJ 08028.

Data availability:

Data are available upon reasonable request to the corresponding author.

References

  1. Appel H, Gerlach AL, & Crusius J (2016). The interplay between Facebook use, social comparison, envy, and depression. Current Opinion in Psychology, 9, 44–49. [Google Scholar]
  2. Arigo D, Brown MM, Pasko K, Ainsworth MC, Travers L, Gupta A, & Smyth JM (2020). Rationale and design of the women’s health and daily experiences project: Protocol for an ecological momentary assessment study to identify real-time predictors of midlife women’s physical activity. JMIR Research Protocols, 9(10), e19044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arigo D, & Butryn ML (2019). Prospective relations between social comparison orientation and weight loss outcomes. Behavioral Medicine, 45(3), 249–254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Arigo D, Gray RC, Dallal DH, Villareale J, & Zhu J (2023). Selection of and response to physical activity–based social comparisons in a digital environment: Series of daily assessment studies. JMIR Human Factors, 10, e41239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Arigo D, Mogle JA, Brown MM, & Gupta A (2021). A multi-study approach to refining ecological momentary assessment measures for use among midlife women with elevated risk for cardiovascular disease. Mhealth, 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Arigo D, Mogle JA, Brown MM, Pasko K, Travers L, Sweeder L, & Smyth JM (2020). Methods to assess social comparison processes within persons in daily life: A scoping review. Frontiers in Psychology, 10, 2909. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Arigo D, Mogle JA, & Smyth JM (2021). Relations between social comparisons and physical activity among women in midlife with elevated risk for cardiovascular disease: An ecological momentary assessment study. Journal of Behavioral Medicine, 44(5), 579–590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Arigo D, Suls JM, & Smyth JM (2014). Social comparisons and chronic illness: Research synthesis and clinical implications. Health Psychology Review, 8(2), 154–214. [DOI] [PubMed] [Google Scholar]
  9. Arigo D, Smyth JM, & Suls JM (2015). Perceptions of similarity and response to selected comparison targets in type 2 diabetes. Psychology & Health, 30(10), 1206–1220. [DOI] [PubMed] [Google Scholar]
  10. Baldwin M, & Mussweiler T (2018). The culture of social comparison. Proceedings of the National Academy of Sciences, 115(39), E9067–E9074. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bandura A (1998). Health promotion from the perspective of social cognitive theory. Psychology and Health, 13(4), 623–649. [Google Scholar]
  12. Barone Gibbs B, Hivert MF, Jerome GJ, Kraus WE, Rosenkranz SK, Schorr EN, & American Heart Association Council on Lifestyle and Cardiometabolic Health, Council on Cardiovascular and Stroke Nursing, & Council on Clinical Cardiology. (2021). Physical activity as a critical component of first-line treatment for elevated blood pressure or cholesterol: Who, what, and how?: A scientific statement from the American Heart Association. Hypertension, 78(2), e26–e37. [DOI] [PubMed] [Google Scholar]
  13. Bocage-Barthélémy Y, Chatard A, Jaafari N, Tello N, Billieux J, Daveau E, & Selimbegović L (2018). Automatic social comparison: Cognitive load facilitates an increase in negative thought accessibility after thin ideal exposure among women. Plos One, 13(3), e0193200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Bom J, Bakx P, Schut F, & Van Doorslaer E (2019). The impact of informal caregiving for older adults on the health of various types of caregivers: A systematic review. The Gerontologist, 59(5), e629–e642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Brakel TM, Dijkstra A, Buunk AP, & Siero FW (2012). Impact of social comparison on cancer survivors’ quality of life: An experimental field study. Health Psychology, 31(5), 660. [DOI] [PubMed] [Google Scholar]
  16. Brim OG, Ryff CD, & Kessler RC (Eds.). (2019). How healthy are we?: A national study of well-being at midlife. University of Chicago Press. [Google Scholar]
  17. Buunk BP, & Ybema JF (2003). Feeling bad, but satisfied: The effects of upward and downward comparison upon mood and marital satisfaction. British Journal of Social Psychology, 42(4), 613–628. [DOI] [PubMed] [Google Scholar]
  18. Buunk BP, & Ybema JF (1997). Health, coping, and well-being: Perspectives from social comparison theory / Social comparisons and occupational stress: The identification- contrast model. 359–388. [Google Scholar]
  19. Centers for Medicare & Medicaid Services (2020a). Health Expenditures by Age and Sex. Dept. of Health and Human Services. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData/Age-and-Gender. [accessed 2024-03-27] [Google Scholar]
  20. Centers for Medicare & Medicaid Services (2020b). National Health Expenditure Data. Dept. of Health and Human Services. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData. [accessed 2024-03-27] [Google Scholar]
  21. Chae CU, & Derby CA (2011). The menopausal transition and cardiovascular risk. Obstetrics and Gynecology Clinics, 38(3), 477–488. [DOI] [PubMed] [Google Scholar]
  22. Cohen BE, Edmondson D, & Kronish IM (2015). State of the art review: Depression, stress, anxiety, and cardiovascular disease. American Journal of Hypertension, 28(11), 1295–1302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Curran PJ, & Bauer DJ (2011). The disaggregation of within-person and between-person effects in longitudinal models of change. Annual Review of Psychology, 62, 583–619. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Davis L, & Wu S (2014). Social comparisons and life satisfaction across racial and ethnic groups: The effects of status, information and solidarity. Social Indicators Research, 117(3), 849–869. [Google Scholar]
  25. Dhingra R, & Vasan RS (2012). Age as a risk factor. Medical Clinics of North America, 96(1), 87–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Diel K, & Hofmann W (2019). Inspired to perspire: The interplay of social comparison direction and standard extremity in the context of challenging exercising goals. Social Cognition, 37(3), 247–265. [Google Scholar]
  27. Festinger L (1954). A theory of social comparison processes. Human Relations. [Google Scholar]
  28. Freeman EW (2015). Depression in the menopause transition: Risks in the changing hormone milieu as observed in the general population. Women’s Midlife Health, 1(1), 1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Gerber JP, Wheeler L, & Suls J (2018). A social comparison theory meta-analysis 60+ years on. Psychological Bulletin, 144(2), 177–197. [DOI] [PubMed] [Google Scholar]
  30. Gibbons FX, & Buunk BP (1999). Individual differences in social comparison: Development of a scale of social comparison orientation. Journal of Personality and Social Psychology, 76(1), 129. [DOI] [PubMed] [Google Scholar]
  31. Guimond S, & Chatard A (2014). Basic principles of social comparison: Does gender matter. Communal functions of social comparison, 205, 205–229. [Google Scholar]
  32. Hardy C, Thorne E, Griffiths A, & Hunter MS (2018). Work outcomes in midlife women: the impact of menopause, work stress and working environment. Women’s Midlife Health, 4, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Helgeson VS, & Mickelson KD (1995). Motives for social comparison. Personality and Social Psychology Bulletin, 21(11), 1200–1209. [Google Scholar]
  34. Hendry P, Solmon M, Choate LH, Autrey P, & Landry JB (2010). Midlife women’s negotiations of barriers to and facilitators of physical activity: Implications for counselors. Adultspan Journal, 9(1), 50–64. [Google Scholar]
  35. Hickey M, Bryant C, & Judd F (2012). Evaluation and management of depressive and anxiety symptoms in midlife. Climacteric, 15(1), 3–9. [DOI] [PubMed] [Google Scholar]
  36. Hoffman L (2015). Longitudinal analysis: Modeling within-person fluctuation and change. Routledge. [Google Scholar]
  37. Infurna FJ, Gerstorf D, & Lachman ME (2020). Midlife in the 2020s: Opportunities and challenges. The American Psychologist, 75(4), 470–485. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Jin SV, Ryu E, & Muqaddam A (2018). Dieting 2.0!: Moderating effects of Instagrammers’ body image and Instafame on other Instagrammers’ dieting intention. Computers in Human Behavior, 87, 224–237. [Google Scholar]
  39. Kim H, Yoo J, Han K, Fava M, Mischoulon D, Park MJ, & Jeon HJ (2021). Associations between smoking, alcohol consumption, physical activity and depression in middle-aged premenopausal and postmenopausal women. Frontiers in Psychiatry, 12, 761761. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Kravitz HM, Schott LL, Joffe H, Cyranowski JM, & Bromberger JT (2014). Do anxiety symptoms predict major depressive disorder in midlife women? The Study of Women’s Health Across the Nation (SWAN) Mental Health Study (MHS). Psychological Medicine, 44(12), 2593–2602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Lakerveld J, Mackenbach JD, de Boer F et al. (2018) Improving cardiometabolic health through nudging dietary behaviours and physical activity in low SES adults: Design of the supreme nudge project. BMC Public Health, 18, 899. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Lee M, Lee H, Kim Y, Kim J, Cho M, Jang J, & Jang H (2018). Mobile app-based health promotion programs: A systematic review of the literature. International Journal of Environmental Research and Public Health, 15(12), 2838. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Leger KA, Charles ST, & Almeida DM (2018). Let it go: Lingering negative affect in response to daily stressors is associated with physical health years later. Psychological Science, 29(8), 1283–1290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Love J, Selker R, Marsman M, Jamil T, Dropmann D, Verhagen J, & Wagenmakers EJ (2019). JASP: Graphical statistical software for common statistical designs. Journal of Statistical Software, 88, 1–17. [Google Scholar]
  45. Maas CJ, & Hox JJ (2005). Sufficient sample sizes for multilevel modeling. Methodology, 1(3), 86–92. [Google Scholar]
  46. Matthews KA, Crawford SL, Chae CU, Everson-Rose SA, Sowers MF, Sternfeld B, & Sutton-Tyrrell K (2009). Are changes in cardiovascular disease risk factors in midlife women due to chronological aging or to the menopausal transition?. Journal of the American College of Cardiology. 54(25), 2366–2373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. McCarthy PA, & Morina N (2020). Exploring the association of social comparison with depression and anxiety: A systematic review and meta-analysis. Clinical Psychology & Psychotherapy, 27(5), 640–671. [DOI] [PubMed] [Google Scholar]
  48. Mikkola M, Mika G, Marko M, Pauliina T, & Olavi Y (2013). Sex differences in age-related cardiovascular mortality. Plos One, 8(5), e63347. 10.1371/journal.pone.0063347 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Min SH, Docherty SL, Im EO, & Yang Q (2021). Identification of symptom clusters among midlife menopausal women with metabolic syndrome. Western Journal of Nursing Research, 1–16. [DOI] [PubMed] [Google Scholar]
  50. Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, & Turner MB (2015). Heart disease and stroke statistics—2015 update: A report from the American Heart Association. Circulation, 131(4), e29–e322. [DOI] [PubMed] [Google Scholar]
  51. Muller D, & Fayant MP (2010). On being exposed to superior others: Consequences of self-threatening upward social comparisons. Social and Personality Psychology Compass, 4(8), 621–634. [Google Scholar]
  52. Murayama K, Usami S, & Sakaki M (2022). Summary-statistics-based power analysis: A new and practical method to determine sample size for mixed-effects modeling. Psychological Methods, 27(6), 1014–1038. [DOI] [PubMed] [Google Scholar]
  53. Olander EK, Fletcher H, Williams S, Atkinson L, Turner A, & French DP (2013). What are the most effective techniques in changing obese individuals’ physical activity self-efficacy and behaviour: a systematic review and meta-analysis. International Journal of Behavioral Nutrition and Physical Activity, 10, 1–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Pan W, Mu Z, Zhao Z, & Tang Z (2023). Female users’ TikTok use and body image: Active versus passive use and social comparison processes. Cyberpsychology, Behavior, and Social Networking, 26(1), 3–10. [DOI] [PubMed] [Google Scholar]
  55. Patrick H, Neighbors C, & Knee CR (2004). Appearance-related social comparisons: The role of contingent self-esteem and self-perceptions of attractiveness. Personality and Social Psychology Bulletin, 30(4), 501–514. [DOI] [PubMed] [Google Scholar]
  56. Perski O, Keller J, Kale D, Asare BYA, Schneider V, Powell D, & Kwasnicka D (2022). Understanding health behaviours in context: A systematic review and meta-analysis of ecological momentary assessment studies of five key health behaviours. Health Psychology Review, 16(4), 576–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Polo-Kantola P (2011). Sleep problems in midlife and beyond. Maturitas, 68(3), 224–232. [DOI] [PubMed] [Google Scholar]
  58. Pope ND, Kolomer S, & Glass AP (2012). How women in late midlife become caregivers for their aging parents. Journal of Women & Aging, 24(3), 242–261. [DOI] [PubMed] [Google Scholar]
  59. Pressman SD, Jenkins BN, & Moskowitz JT (2019). Positive affect and health: What do we know and where next should we go? Annual Review of Psychology, 70, 627–650. [DOI] [PubMed] [Google Scholar]
  60. Rancourt D, Leahey TM, LaRose JG, & Crowther JH (2015). Effects of weight-focused social comparisons on diet and activity outcomes in overweight and obese young women. Obesity, 23(1), 85–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Reeves WC, Pratt LA, Thompson W, Ahluwalia IB, Dhingra SS, McKnight-Eily LR, Harrison L, D’Angelo DV, Williams L, Morrow B, Gould D, & Safran MA (2011). Mental illness surveillance among adults in the United States. Centers for Disease Control and Prevention, 60, 1–32. [PubMed] [Google Scholar]
  62. Rowland SA, Berg KE, Kupzyk KA, Pullen CH, Cohen MZ, Schultz PS, & Yates BC (2018). Feasibility and effect of a peer modeling workplace physical activity intervention for women. Workplace Health and Safety, 66(9), 428–436. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Samra A, Warburton WA, & Collins AM (2022). Social comparisons: A potential mechanism linking problematic social media use with depression. Journal of Behavioral Addictions, 11(2), 607–614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Sassarini J (2016). Depression in midlife women. Maturitas, 94, 149–154. [DOI] [PubMed] [Google Scholar]
  65. Schenk HM, Jeronimus BF, van der Krieke L, Bos EH, de Jonge P, & Rosmalen JG (2018). Associations of positive affect and negative affect with allostatic load: a lifelines cohort study. Psychosomatic Medicine, 80(2), 160–166. [DOI] [PubMed] [Google Scholar]
  66. Smyth JM, Juth V, Ma J, & Sliwinski M (2017). A slice of life: Ecologically valid methods for research on social relationships and health across the life span. Social and Personality Psychology Compass, 11(10), e12356. [Google Scholar]
  67. Stewart AL, Kathawalla UK, Wolfe AG, & Everson-Rose SA (2018). Women’s heart health at mid-life: what is the role of psychosocial stress?. Women’s Midlife Health, 4, 1–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Taylor SE, & Lobel M (1989). Social comparison activity under threat: Downward evaluation and upward contacts. Psychological Review, 96(4), 569. [DOI] [PubMed] [Google Scholar]
  69. Terol Cantero MC, Bernabé M, Martín-Aragón M, Vázquez C, & Buunk AP (2021). Social comparison and stress appraisal in women with chronic illness. International Journal of Environmental Research and Public Health, 18(10), 5483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Thøgersen-Ntoumani C, Dodos L, Chatzisarantis N, & Ntoumanis N (2017). A diary study of self-compassion, upward social comparisons, and body image-related outcomes. Applied Psychology: Health and Well-Being, 9(2), 242–258. [DOI] [PubMed] [Google Scholar]
  71. Thomas AJ, Mitchell ES, & Woods NF (2018). The challenges of midlife women: themes from the Seattle midlife women’s health study. Women’s Midlife Health, 4, 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Van Der Zee KI, Buunk BP, Sanderman R, Botke G, Van Den Bergh F (1999). The big five and identification–contrast processes in social comparison in adjustment to cancer treatment. European Journal of Personality, 13(4), 307–326. [Google Scholar]
  73. Verduyn P, Gugushvili N, Massar K, Täht K, & Kross E (2020). Social comparison on social networking sites. Current Opinion in Psychology, 36, 32–37. [DOI] [PubMed] [Google Scholar]
  74. Verhees MW, Finet C, Vandesande S, Bastin M, Bijttebier P, Bodner N, ... & Bosmans G. (2021). Attachment and the development of depressive symptoms in adolescence: The role of regulating positive and negative affect. Journal of Youth and Adolescence, 50(8), 1649–1662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Vogel EA, Rose JP, Roberts LR, & Eckles K (2014). Social comparison, social media, and self-esteem. Psychology of Popular Media Culture, 3(4), 206. [Google Scholar]
  76. Watt AD & Konnert CA (2020). Body satisfaction and self-esteem among middle-aged and older women: The mediating roles of social and temporal comparisons and self-objectification. Aging & Mental Health, 24(5), 797–804. [DOI] [PubMed] [Google Scholar]
  77. Wemken G, Janurek J, Junker NM, & Häusseret JA (2021). The impact of social comparisons of job demands and job control on well-being. Applied Psychology: Health and Well-Being, 13(2), 419–436. [DOI] [PubMed] [Google Scholar]
  78. Wheeler L, & Miyake K (1992). Social comparison in everyday life. Journal of Personality and Social Psychology, 62(5), 760. [Google Scholar]
  79. Wills TA (1981). Downward comparison principles in social psychology. Psychological Bulletin, 90(2), 245. [Google Scholar]
  80. Wood JV, Taylor SE, & Lichtman RR (1985). Social comparison in adjustment to breast cancer. Journal of Personality and Social Psychology, 49(5), 1169. [DOI] [PubMed] [Google Scholar]
  81. Yoon S, Kleinman M, Mertz J, & Brannick M (2019). Is social network site usage related to depression? A meta-analysis of Facebook–depression relations. Journal of Affective Disorders, 248, 65–72. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data are available upon reasonable request to the corresponding author.

RESOURCES