Skip to main content
NCBI home page
Health Psychology and Behavioral Medicine logoLink to Health Psychology and Behavioral Medicine
. 2026 Aug 22;14(1):2721026. doi: 10.1080/21642850.2026.2721026

Social comparison target selections as health information preferences: exploring contextual influences among women in midlife with elevated risk for cardiovascular disease

Emmanuel Lapitan a, Andrea F Lobo b, Danielle Arigo a,c,d,*
PMCID: PMC13501744  PMID: 42639104

Abstract

Background

Insufficient physical activity heightens risk for cardiovascular disease (CVD), particularly among women ages 40−65 whose risk is already elevated by conditions like hypertension. Physical activity interventions exist for this group, but whether health information preferences (which may affect intervention outcomes) vary for the same individual across contexts is unclear. Little is known about contextual predictors of selections of health-relevant social comparisons (i.e. evaluating oneself relative to others), which represent sources of health information.

Methods

To explore social comparison selections as indicators of health information preferences, we conducted secondary analyses using data from tests of a web application designed to increase physical activity for women in midlife with elevated risk for CVD (N = 88). Participants completed an initial survey, used the web application each morning for 7 days (where they selected and viewed a comparison target), and completed daily surveys and exit interviews. We used multilevel models to assess stability versus fluctuation and associations between daily context and selections, and highlighted statements from interviews for additional context.

Results

Women showed considerable daily variability in physical activity satisfaction and social comparison target selections (ICCs < 0.63; Design Effects > 1.3); they attributed the latter in part to the desire to explore web app content. On days women reported greater-than-usual satisfaction with their steps, they were less likely to select a not very active peer (ORs < 0.78, ps < 0.03) and more likely to avoid making a specific target selection (ORs > 1.25, ps < 0.02).

Conclusion

Women showed meaningful daily fluctuations in social comparison target selection, and contextual factors such as satisfaction with the previous day's physical activity predicted these variations. Findings indicate that health information preferences may be dynamic, underscoring the potential benefits of tailoring content to daily or momentary experiences to enhance the effectiveness of digital health interventions.

Introduction

Health information processing in the context of physical activity promotion

Inadequate physical activity (PA) increases risk for chronic health conditions such as cardiovascular disease (CVD), especially for populations with CVD risk factors such as hypertension (a condition in which blood pressure is higher than normal; Elagizi et al., 2020; Jakubisin Konicki, 2019; Kraus et al., 2019). One such group is women in midlife (ages 40-65), in part due to the combined biological influences of aging and the menopause transition (El Khoudary et al., 2020; Shufelt et al., 2019). PA promotion to minimise CVD risk often occurs via behavioural interventions that provide psychoeducation and teach psychological skills such as effective goal setting (Arigo et al., 2022). However, it is not clear how women interact with the health information provided by these interventions and whether these interactions meaningfully impact daily PA-related experiences.

Interest in how individuals interact with health information and the effects of these interactions informed various theories of health information preferences. For example, monitoring and blunting constructs represent two styles of coping with information about threats to one’s health—such as medical diagnoses—that reflect contrasting information-seeking tendencies (Miller, 1987). Monitoring describes the tendency to seek out health-threatening information whereas blunting refers to the tendency to distract oneself from such information. At the person level, monitors tend to be less satisfied with their health and healthcare than blunters, including lower satisfaction with the health information they receive, indicating that they would prefer more information (Miller, 1995; van de Wal et al., 2023). A monitoring style is more common among women than men (van de Wal et al., 2023) and among individuals with higher (vs. lower) trait anxiety and appraisals of threats (Muris et al., 1994). For example, dissatisfaction with health care received was independently predicted both by identifying as a woman and a monitoring (vs. blunting) coping style (van de Wal et al., 2023).

Preferences for how content is framed also vary and can be explained by regulatory focus theory. Individuals may prefer to receive information that is promotion-focused (i.e. framed to promote positive health outcomes and/or behaviours) or prevention-focused (i.e. framed to prevent negative health outcomes and/or behaviours; Higgins, 2012). Both a health promotion focus and a health prevention focus are positively associated with cognitive factors such as internal and external motivation. As these types of motivation are both positively associated with PA behaviour (Laroche et al., 2019), however, the nature of the association(s) and the potential power of regulatory focus style to predict response to PA intervention (and ultimately, PA behaviour) remain unclear.

To date, theories describing monitoring-blunting and regulatory focus have relied on assessments of global, person-level preferences for whether individuals tend to seek or avoid information and how they prefer information to be framed, across settings or contexts. As styles of behaviour, these tendencies are important to consider, as research demonstrates psychological and behavioural benefits for individuals who receive information in the way that matches their preference (versus not matching their preference; Latimer et al., 2008; Miller, 1995; Williams-Piehota et al., 2009).

However, little evidence speaks to whether information preferences are indeed stable for the same person across time and context, and whether they generalise beyond explicit, text or verbal communication to other types of information, such as social information shared in digital environments. Specifically, prior research has relied on global, retrospective self-report via questionnaires, interviews, and focus groups to assess person-level information preferences (Ramsey et al., 2017). While informative, these methods are prone to recall bias (i.e. reporting experiences that are more accessible or easier to remember) and summarising experiences across varying contexts into single or limited responses (Stone & Shiffman, 2002; Veenman, 2011). This precludes a clear understanding of information preferences; these may differ for the same individual at different times, as: 1) information needs are tied to an individual’s personal context (e.g. their health status and environment), which changes over time, and 2) information-seeking is posited as a dynamic process that continues to evolve, particularly as new technologies continuously change the way individuals interact with health information (Ramsey et al., 2017).

As information preferences show associations with health outcomes, understanding whether these preferences vary based on contextual factors is critical. Preliminary evidence highlights cognitive and affective experiences as potential contextual influences on preference (Christensen & Brooks, 2006; Moody et al., 2010; Xue et al., 2018), but this work too is limited by reliance on global, retrospective self-report. Thus, examining whether contextual factors predict health information preferences in daily life is a promising approach to advancing the current understanding of these preferences and their associations with health behaviours such as PA.

Preferences for health-relevant social comparison

Social comparison is the process of self-evaluation relative to others (Festinger, 1954), which can be used to obtain information about oneself in a range of domains, including health (Bandura, 2008). Evidence shows that social comparison can positively influence health behaviours, such as by increasing PA motivation (Diel & Hofmann, 2019), and greater trait-level interest in social comparison is associated with better PA maintenance after weight loss treatment (Arigo & Butryn, 2019). For these reasons, information that prompts comparison is embedded in many digital PA promotion tools (e.g. apps; Arigo et al., 2020). In particular, women in midlife often seek connections with other similar women around PA, which can facilitate beneficial social comparisons (Arigo et al., 2023; McArthur et al., 2014).

Individuals can compare themselves to others who they perceive as better off than them in a given domain (upward comparison target), comparable to them (lateral comparison target), or worse off in that domain (downward comparison target; Suls et al., 2002; Wood et al., 1985). Yet, existing evidence shows that PA-related comparisons in the same direction can lead to opposite outcomes. For example, downward comparisons are positively associated with motivation to achieve PA goals (Diel & Hofmann, 2019), but can also lead to feelings of hopelessness or convey an already satisfactory status with their PA behaviour, and thereby reduce motivation for high performance (Buunk et al., 1990). Conversely, upward comparisons may lead to feelings of hopelessness about reaching a better health status, and thereby lower PA motivation (Caltabiano & Ghafari, 2011). But upward comparisons are also associated with greater PA than engagement in lateral or downward comparisons (Rancourt et al., 2015), as they can provide inspiration and increased motivation to achieve a health goal or keep up with the upward target (Diel et al., 2024). Thus, electing to make an upward comparison when other options are available could reflect the use of comparison for self-improvement, whereas a preference to make a downward comparison may reflect the use of comparison for self-enhancement (Pavlova et al., 2018; Taylor & Lobel, 1989).

Considering the selection of social comparison targets could be helpful in predicting outcomes of comparison (e.g. PA). These selections could reflect stable preferences for social information, but minimal research has assessed these selections as potentially variable at different times. For example, in a series of studies where participants used a web platform where they selected a PA-focused peer profile to view, Arigo et al. (2023) showed that participants’ profile selections varied across days and were not explained by stable individual differences, suggesting that social comparison information preferences can differ for the same person on different days (see also Arigo et al., 2021).

Additionally, little evidence speaks to whether and under what circumstances exposure to digital social comparison information is associated with PA (Arigo et al., 2020). Research methods that assess social comparison processes in real time can help to determine the potentially different consequences of comparison in different contexts. As such, examining these social comparison experiences with more temporal sensitivity is paramount. While social comparisons have been discussed as a way to obtain self-relevant information (Buunk & Ybema, 1997), selections of comparison targets have rarely been conceptualised as representing health information preferences, which are known to be associated with health behaviours such as PA at the between-person level. Also, little is known about real-time associations between contextual factors and PA-based target selections—particularly among women in midlife with elevated risk for CVD and low PA engagement. Overall, better understanding whether and to what extent social comparison target selections (as reflections of health information preferences) vary and are associated with contextual factors in daily life can inform interventions that use health-relevant social information to increase PA for these women.

The present study

The purpose of the present study was to examine PA-relevant social comparison target selections as reflecting health information preferences—including potential within-person variability in these selections across consecutive days—and to test for associations between contextual factors and comparison target selections. We conducted secondary analyses using an existing dataset compiled from two studies that tested a proprietary PA web application (app) for women in midlife with elevated CVD risk. This app captured PA-relevant cognitive and social comparison experiences relevant to personalised and adapted peer profiles (i.e. comparison targets) in real time. In both studies, each participant used the app once per day for 7 consecutive days. Users’ information preferences were operationalized as their real-time selection of social comparison information on the web app. Cognitive factors measured in these studies included women’s satisfaction with their PA from the previous day, which was captured immediately prior to comparison target selection.

We tested the following hypotheses: 1) On days when participants’ satisfaction with their previous day’s PA is lower than usual, they are less likely to select an upward comparison target than other types of comparison targets. Participants who already experience low satisfaction with their achievements may be inclined to seek information that increases positive affect, which is associated with self-efficacy (Joshanloo, 2022). Thus, people may be motivated to avoid upward comparisons when they are particularly dissatisfied, to minimise risk of experiencing the frustration or disappointment that can occur in response to upward comparisons. 2) On days when participants’ satisfaction with their previous day’s PA is higher than usual, they are more likely to indicate “no preference” for a specific target than to select an upward, lateral, or downward comparison target. Recent PA success can be recalled as health-relevant information to prompt high PA self-efficacy and as such, people may be ambivalent about additional health information they would receive from social comparison.

Methods

Recruitment and participants

Across two studies, women ages 40-65 (inclusive) were eligible to participate if they were interested in increasing their PA, had at least 1 CVD risk factor (e.g. high cholesterol), had a device with internet capability, were fluent in English, and were not pregnant. Women were recruited (1) from a database of participants in earlier studies of the same overarching line of research at the supporting institution, (2) via email announcements to members of the supporting institution, and (3) via social media (Facebook, Twitter/X). The final sample across both studies consisted of 88 women (M Age  = 51.6; 66.7% white; see Table 1 for additional participant characteristics).

Table 1.

Participant characteristics (N = 88).

  M (SD)        
Age 51.6 (5.8)        
BMI 31.9 (7.3)        
Beck Anxiety Inventory (BAI)
 15.2 (14.0)
  
  
  
  
 
 n (%)
 
 n (%)
 
 n (%)
Racial Identification*   Menopause Status   Marital Status *  
White 58 (66.7%) Pre-menopause 23 (26.1%) Never married 14 (16.1%)
Black 22 (25.3%) Perimenopause 16 (18.2%) Married 52 (59.8%)
East Asian 1 (1.2%) Post-menopause 35 (39.8%) Divorced 14 (16.1%)
South Asian 2 (2.3%) Other (surgery) 14 (15.9%) Separated 4 (4.6%)
Other 4 (4.6%) CVD Conditions   Widowed 3 (3.5%)
Ethnic Identity   Hypertension 44 (50.0%) Children*  
Hispanic/Latina 6 (6.8%) High cholesterol 44 (50.0%) Yes 67 (77.0%)
Non-Hispanic/Latina 82 (93.2%) Type 2 diabetes 26 (29.5%) No 20 (23.0%)
Highest Educational Level   Metabolic syndrome 7 (8.0%) Provide Childcare  
High school graduate/GED 4 (4.6%) Current smoker 6 (6.8%) Yes 48 (54.6%)
Associate’s degree, partial college, technical degree 14 (15.9%) Household Income*   No 40 (45.4%)
Bachelor’s degree 33 (37.5%) <$25,000 6 (6.9%) Provide Other Care*  
Graduate degree 37 (42.0%) $25,000-$50,000 12 (13.8%) Yes 12 (13.8%)
    $50,000-$75,000 22 (25.3%) No 75 (86.2%)
    > $75,000 47 (54.0%)    
Open in a new tab
*

n = 87 (1 declined to answer).

Measures

Demographics and background characteristics

In an initial online survey, participants reported their demographics: age, CVD risk factors (i.e. high cholesterol, hypertension or prehypertension, type 2 diabetes or prediabetes, metabolic syndrome, or currently smoke or had quit smoking within the last 3 months), height (inches) and weight (pounds; used to calculate body mass index [BMI]), racial identity, ethnicity, highest level of completed education, employment status, total household income, marital status, number of children, caregiving responsibilities, and menstrual status. As a potential covariate, we also assessed baseline level of anxiety using the Beck Anxiety Inventory (BAI), a 21-item measure that assesses anxiety-related symptoms on a four-point scale from 0 (not at all) to 3 (severely; Beck et al., 1988). The BAI has demonstrated high internal consistency (α = 0.91) and is associated with other anxiety measures (e.g. Hamilton Anxiety Rating Scale and Generalised Anxiety Disorder-7; Beck et al., 1988; Spitzer et al., 2006). A total score was calculated by summing across individual item scores; possible scores range from 0-63, with scores from 0-7 indicating minimal-to-no anxiety symptoms, 8-15 indicating mild anxiety, 16-25 indicating moderate anxiety, and 26-63 indicating severe anxiety (Beck & Steer, 1990). Finally, the study that participants were enroled into were coded as ‘1’ and ‘2,’ corresponding to Study 1 and Study 2, respectively.

Physical activity-related cognition

PA-related cognitions were operationalised as participants’ satisfaction with their overall activity (i.e. step count) from the previous day, as self-reported in the web app in real time immediately prior to selecting a comparison target (peer profile). Response options were on a scale of 1 (Very dissatisfied) to 6 (Very satisfied).

Comparison target selection

Comparison target selection was recorded by the web app when participants indicated which peer profile they wanted to view: a very active (upward comparison target), somewhat active (lateral comparison target), or not very active (downward comparison target) peer, or no preference—choose for me (see below). The categories of selection and of the peer viewed each day were represented numerically in the user database.

Physical activity behaviour

PA was operationalised as total steps taken per day to capture overall movement; steps often seem more modifiable than structured exercise (Allmeta et al., 2026; Dąbrowska et al., 2015), particularly for individuals who are insufficiently active. Steps were measured using the study-issued Accusplit AX2720MV pedometer or a participant’s own PA monitor (e.g. Fitbit, Apple Watch), and participants reported the total number of steps at the end of each day. During their web app use in the morning, participants also reported their number of steps taken so far that day.

Procedure

The only procedural difference between Studies 1 and 2 was the inclusion of a 60-minute interview after 7 days of participation; in Study 2, this interview was shortened to 15 minutes. Study 1 focused on understanding participants’ web app use and user experience, and we conducted in-depth interviews and achieved a sufficient sample size to reach theme saturation (for qualitative thematic analysis; cf. Wutich et al., 2024). As a follow-up, Study 2 was focused on testing the feasibility and usability of the web app technology itself, and we collected feedback in an abbreviated manner only.

Following eligibility screening, women who did not have their own PA monitor were sent a study-owned pedometer and prepaid return mailing materials via U.S. mail, to arrive to the participant prior to their scheduled orientation session. During the orientation session (conducted via Zoom), women were informed of study procedures, were provided with opportunities to ask questions, received a list of study instructions, and scheduled their study exit interview for immediately after the last day of their study participation. Interested women provided written informed consent to participate before completing the orientation and then were sent an initial survey that included demographic questions. Women then participated in 7 consecutive days of data collection, including: 1) using a web app designed to increase PA among women in midlife with elevated CVD risk each morning prior to starting their regular daily activities; 2) wearing a study-owned pedometer or their own PA monitor each day to record daily PA, and; 3) completing an end-of-day survey each night approximately 1 hour before they went to sleep. Following the 7 days of data collection, women completed an exit interview and were compensated $50 for their participation.

Daily web application use

Upon initiating a use episode, the web app first assessed women’s PA goals for the day, steps so far today, and their satisfaction with their PA from the previous day. Participants then indicated which peer profile they wanted to view on the web app: available selections included a very active, somewhat active, or not very active peer, or no preference—choose for me, for which the web app displayed the type of peer profile that the participant viewed least recently. Profiles were descriptions of fictional women in midlife and included PA-relevant and demographic information such as average PA behaviour in the past week, caregiving responsibilities, and barriers to PA (see Figure 1 for an example peer profile).

Figure 1.

An example peer profile with a photograph of a woman in midlife who is of Asian racial background, preparing food. Underneath the photograph is information about this peer, such as recent physical activity behaviour, career/employment, caregiving responsibilities, and barriers to being active. Captions are added to highlight that peer profiles were personalised to match users’ age and racial/ethnic background, and adapted so the peer’s physical activity was similar to users’ own physical activity from the previous day.

Open in a new tab

Example of a Peer Profile From the Web App.

Each selection generated personalised profiles that were matched to users’ demographic characteristics (i.e. age and racial/ethnic background; Arigo et al., 2022). Images of peers were selected from open-access web-based databases to portray women ages 40-65 who represent a diverse range of racial and ethnic backgrounds and professions. Additionally, profiles were adapted to show peers whose PA was anchored to the participant’s own PA from the previous day. Selecting very active generated a peer profile whose steps and exercise minutes were 130% of the participant’s reported PA from the previous day (upward comparison target), somewhat active depicted PA levels that were 95-105% of the participant’s PA from the previous day (lateral target), and a not very active peer profile was shown to have PA levels that were 68% of the participant’s PA from the previous day (downward target; Arigo et al., 2022).

Data analysis

Analyses were conducted in RStudio using 2-level multilevel models with days nested in participants. Models included 80-88 of participants’ data and 455-532 unique observations. Reasons for missing data included participants not using the web app as intended (e.g. forgetting, prioritising other responsibilities, technical difficulties). This sample size afforded adequate power to detect small- to medium-sized effects with the statistical tests described below (i.e. power > 0.80; Maas & Hox, 2005; Murayama et al., 2022).

Intraclass correlation coefficients (ICCs) were calculated using empty models and were used to indicate the proportion of total variability attributable to between-person stability across days; the remaining variability is attributable to within-person fluctuation and error. We also included the Design Effect statistic, which indicates how much the estimates of error need to be increased to account for data that are not independent (Peugh, 2010). Statistical approaches such as multilevel modeling (MLM) are warranted when the Design Effect >1.1 (Lai & Kwok, 2015). Person-centred means and deviations were calculated by subtracting the individual's mean from each observation of the predictor variable to determine deviation from the participant’s average (i.e. person-mean centreing; Curran & Bauer, 2011; Hoffman, 2015). Within-person associations were tested between participants’ satisfaction with their previous day’s steps and their comparison target selection. PA was treated as continuous while target selections were re-coded and treated as binary for each selection option (i.e. selected [1] versus not [0]). Effect sizes are expressed as day-level semipartial correlation coefficients (sr) or as odds ratios (OR), as appropriate for the type of test. We considered effect sizes of sr ≥ 0.10 and OR < 0.90 or OR > 1.10 (and 95% confidence intervals of odds ratios entirely below or above 1.0; confidence intervals do not include 1.0) to indicate potential for small but meaningful differences at the day level (Cohen, 2013); p-values are also reported.

Participant characteristics such as age, study (i.e. Study 1 vs. Study 2), global anxiety score (BAI), and steps recorded in the morning prior to using the web app were examined for associations with the dependent variables (e.g. comparison target selection) and considered as covariates. For continuous between-person covariates (i.e. age and anxiety), global means and deviations were calculated; person-centred means and deviations were calculated for covariates measured repeatedly (i.e. steps so far today; Curran & Bauer, 2011; Hoffman, 2015). Sensitivity analyses compared models with and without these covariates and all results are reported below. Qualitative data from exit interviews were not formally incorporated as this subset of analyses were not intended as a multi/mixed methods study. However, we include quotations from these interviews, which were selected to illustrate and provide context for specific quantitative findings.

Ethics Statement

Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki and was approved by an Institutional Review Board/Ethics committee.

Results

Descriptives and variability in constructs of interest

Women reported low levels of global anxiety at baseline, though there was notable variability (M = 15.2, SD = 14.0; Beck et al., 1988). On average across days, women reported being somewhat satisfied with their step count from the previous day (B = 4.0, SD = 1.3). There was low between-person stability in women’s satisfaction with the previous day’s step count; 92.8% of the variability was due to within-person variability (and error; ICC = 0.072, Design Effect = 1.4; see Table 2).

Table 2.

Descriptive statistics for variables of interest.

  n        
Study 1 27        
Study 2
61
  
  
  
  
 
 M/B
 SD/SE
 Range
 ICC
 Design Effect
Beck Anxiety Inventory (BAI) * 15.2 14.0 0-60 - -
Daily steps taken before web app use ** 580.1 143.1 0-15,000 0.633 4.6
Satisfaction with previous day’s steps** 4.0 0.1 1-6 0.072 1.4
Steps per day**
5659.0
 314.6
 0-24,206
 0.575
 4.1
 
 n
 ICC (Chose Yes/No)
 Design Effect
  
  
Very active 124 0.113 1.6    
Somewhat active 138 0.086 1.4    
Not very active 96 0.087 1.4    
No preference—choose for me 174 0.145 1.7    
Open in a new tab
*

BAI mean represented as M and standard deviation (SD) is reported.

**

Bs represent estimates of averages from empty multilevel models (reflecting day-level averages across participants) and standard errors (SEs) are reported.

Across all days with recorded web app use, women selected to view a very active peer on 124 days (23.3%), a somewhat active peer on 138 days (25.9%), and a not very active peer on 96 days (18.0%; see Table 2). They selected no preference—choose for me most frequently, on 174 days (32.7%). 85.5%-91.4% of variability in peer profile selections were attributable to within-person fluctuation (and error; ICCs = 0.086-0.145; Design Effects > 1.3). Women recorded an average of approximately 5659.0 steps each day across all days of observation (full day, recorded before bedtime; SD = 2315.5, Range = 0-24,206). Between-person differences accounted for 57.5% of the variability in women’s steps per day (ICC = 0.575; Design Effect = 4.1). Study 1 participants recorded higher daily step counts (albeit with a low effect size; β = -2404.13, SE = 1117.91, p = 0.03, sr = 0.019) than Study 2 participants, thus justifying controlling for study in primary analyses. Across variables of interest, Design Effects also confirmed the need to use a statistical approach like MLM to account for nesting (i.e. effects > 1.1).

Primary results

Consistent with hypotheses, on days when women’s satisfaction with their step count from the previous day was higher than their average level of satisfaction, they were 28% more likely to choose no preference—choose for me than to select a directional comparison target (OR = 1.276, 95% CI [1.079, 1.508], p < 0.01). In contrast, there was no difference in the likelihood of selecting to view a very active peer (upward target; OR = 1.017, 95% CI [0.854, 1.209], p = 0.85; see Figure 2) or a somewhat active peer (lateral target; OR = 0.926, 95% CI [0.786, 1.092], p = 0.36) based on daily satisfaction with steps. Additionally, relative to their average level of satisfaction, on days when women were more satisfied with their step count from the previous day, they were 23% less likely to select a not very active peer (downward target; OR = 0.773; 95% CI [0.636, 0.941], p = 0.01; see Figure 2). We found similar results from models controlling for age, study, global anxiety, and steps so far (see Table 3 for additional results).

Figure 2.

Three Line Graphs with x Intercepts from -4 to +3. The first graph shows a single blue line that rises slightly from about y=0.19 at x=−4 to about y=0.21 at x=3. A wide grey shaded band around the blue line (uncertainty interval) is largest at the ends and narrowest near x=0 (about y=0.16–0.25), indicating greater variability at the extremes and more precision near the centre. The second graph shows a single blue line that falls from about y=0.33 at x=−4 to about y=0.08 at x=3. A wide grey shaded band around the blue line (uncertainty interval) is largest where x=-4 (y=0.18–0.52) and narrowest near x=0 (about y=0.12–0.19). The third graph shows a single blue line that rises from about y=0.14 at x=−4 to about y=0.48 at x=3. A wide grey shaded band around the blue line (uncertainty interval) is largest at the ends and narrowest near x=0 (about y=0.25–0.35), indicating slightly greater variability at the extremes and more precision near the centre.

Open in a new tab

Plots of results from primary models.

Table 3.

Results from primary analyses: associations with satisfaction with previous days’ steps.

  OR 95% CI p
Very active peer selection 1.017 0.854, 1.209 0.85
Very active peer selection (with covariates) 1.009 0.844, 1.206 0.92
Somewhat active peer selection 0.926 0.786, 1.092 0.36
Somewhat active peer selection (with covariates) 0.940 0.793, 1.115 0.48
Not very active peer selection 0.773 0.636, 0.941 0.01
Not very active peer selection (with covariates) 0.776 0.633, 0.952 0.02
No preference 1.276 1.079, 1.508  <0.01
No preference (with covariates) 1.252 1.053, 1.488 0.01
Open in a new tab

Participants’ responses during exit interviews suggest that interest and curiosity in seeing a variety of information as a potential explanation for the observed within-person fluctuation of social comparison target selections. For instance, this interest was especially noted as a reason for women selecting no preference on the web app, such as participants stating, “I selected mostly no preference and that reason was curiosity… You don't know what tidbits you're going to pick up” (age 62, White) and “I wanted to leave it up to the [web app] to choose for me and then I would just deal with whatever” (age 59, Black). This concept is described further in a published study of the same dataset (see Arigo et al., 2023). Interest in different information may have also been demonstrated by choosing other target selections, as described by one participant: “I changed it up, mainly just out of curiosity to see. Once I picked very active and… then not very active… more out of curiosity than anything, I mixed it up. I never went with just one thing all the time” (age 50, White).

Discussion

Health information coping styles and preferences have been associated with health behaviours and outcomes (Laroche et al., 2019) and are primarily assessed as consistent styles (versus experiences that vary for the same person at different times; Ramsey et al., 2017). For the present study, we used an existing dataset to examine variability in PA-based social comparison target selections, to represent preferences for health-related social information. Across all variables of interest, results indicated that women's PA-related experiences fluctuated day-to-day, with only a small proportion of the variability attributable to stable individual differences: 7.2% for their satisfaction with their steps from the previous day, and up to 14.5% of their daily peer profile selections on the web app, were stable across days. A larger proportion of the variability in actual PA behaviour (57.5%) was attributable to stable individual characteristics. This may be due to recruiting an insufficiently active population that struggles to increase PA, and our web app was not designed to be an effective standalone PA promotion intervention—particularly over only 7 days. Specifically, these women may normally engage in low levels of PA (e.g. steps) on any given day, and exposure to the web app for 7 days was not intended to induce substantial behaviour change.

These findings replicate recent work showing that social comparison selections vary for the same person over different time points (Arigo et al., 2023). Specifically, results from the present study indicated substantial within-person variability in selections of social comparison targets (i.e. upward, lateral, or downward targets, or no preference). This finding adds support to the notion that comparison selections are not stable in the context of health and PA and extends this observation to women in midlife with elevated CVD risk (Arigo et al., 2020). Variations in social comparison experiences may be linked to individuals’ dynamic and changing needs, goals, and/or preferences for social comparison information (Mollee & Klein, 2017; Pavlova et al., 2018; Taylor & Lobel, 1989).

Furthermore, fluctuation in social comparison target selections may indicate that health information preferences vary across days for the same person. This is particularly noteworthy, as much of our understanding of information preferences focus on stable differences between individuals (Ramsey et al., 2017) and individual-level predictors (Lewis et al., 2024). Previous work has highlighted the benefit of understanding differences in individuals’ health information needs and preferences (Li et al., 2016); unlike the present study, this work has seldom tested for fluctuating needs and preferences for the same person over time. One recent study found that across multiple moments over several days, participants indicated dynamic preferences for PA and diet-related information (e.g. daily health intention reminders; easy exercise options nearby) on the study-designed weight management mobile app (Nicol et al., 2020). Consequently, there is emerging evidence that health information processing is a fluctuating (vs. stable) construct. Existing measures that assess information seeking and preferences, such as regulatory focus and monitoring and blunting scales, have utility for person-level differences, as they were constructed to have high test-retest reliability (and thus, capture stability in differences between people; Higgins et al., 2001; Miller, 1987). However, they are unlikely to capture important context-specific preferences, which we detected in the present study.

Consequently, there is a need to develop and test measures that assess health information coping styles and preferences as they occur in daily life (vs. global styles). This process should include collecting feedback from the target and general populations (e.g. via qualitative interviews, focus groups, advisory boards) to refine draft items into survey items that are: 1) understandable to the broader population, and; 2) accurately measure individuals’ experiences of these constructs in daily life (Willis, 2004). As noted, measures developed for cross-sectional, between-person research do not capture moment-to-moment variations effectively (Bolger & Laurenceau, 2013; Hamaker, 2012), potentially missing important momentary and fluctuating experiences and data. Studies that use new within-person measures could elucidate whether and how information preferences vary for the same person at different times using intensive ambulatory assessment designs, such as ecological momentary assessments (EMA) that use repeated sampling of individuals’ real-time experiences in their natural environments. Assessing these preferences more frequently would clarify the degree of their time sensitivity (e.g. shifting day-to-day versus within days) and provide guidance for the optimal timing of assessment. Better insight into momentary fluctuations of information preferences may further elucidate whether such preferences are associated with daily health behaviours and outcomes (Lewis et al., 2024; Nicol et al., 2020).

Influences of contextual factors

Results from this study also showed that contextual factors—namely, satisfaction with PA from the previous day—predict selection of specific social comparison information among women in midlife with elevated CVD risk. Previous work has assessed predictors of upward and downward social comparisons (Arigo et al., 2018), including measuring these predictors across several months (Brown et al., 2007), but even studies that used longitudinal repeated measures across several months are unable to identify momentary antecedents. Importantly, the results from the present study expand our understanding of social comparison processes to include real-time antecedents to social comparison target selection, by considering cognitive contextual features.

Women’s satisfaction with their step count from the previous day was positively associated with increased odds of selecting no preference for a comparison target. Thus, on days when they were particularly dissatisfied with their previous day’s PA, and potentially more anxious about their health, they were more likely to select specific social comparison information (through selecting a specific comparison target to view rather than selecting no preference), aligning with previous research (Deniz et al., 2021; Herrera-Espiñeira et al., 2009; Howell & Sweeny, 2025). Although daily satisfaction did not predict selection of an upward comparison target, the odds of selecting a downward target were significantly lower on days when women were more satisfied than usual with their step count from the previous day, consistent with our expectations. Thus, the observed pattern aligns with the broader premise that women’s daily satisfaction with their PA from the previous day influences real-time comparison target selections. Furthermore, obtaining information relevant to one’s health could be an important contributor to engagement in health-promoting behaviours such as PA at the person level (Latimer et al., 2008; Miller, 1995; Williams-Piehota et al., 2009). An important next step in this work is to test the potential downstream associations between women’s satisfaction with their PA from the previous day, social comparison target preferences, and actual daily PA behaviour, to better understand whether contextual factors and health information preferences influence health behaviour.

The present findings are particularly timely, as technological developments have made personalisation and adaptation of digital interventions feasible to programme and implement. For instance, digital PA promotion tools that facilitate social comparison, such as the one designed and used in this study, can incorporate inputs of users’ satisfaction with their PA from the previous day to anticipate comparison target preferences each day. Anticipating these preferences may improve personalisation and adaptation of PA interventions that incorporate social comparison such that interventions can optimise delivery of the preferred social comparison information to the right participant at the right time (Hardeman et al., 2019). However, future work should also test whether beneficial outcomes (e.g. increased PA behaviour) are more likely on days when women receive information they prefer (vs. not prefer). Finally, given that CVD risk is increasing among younger women (Joynt Maddox et al., 2026), future research should examine these phenomena among younger women, to determine the extent to which they generalise among women with heightened risk for CVD.

Strengths and limitations

A primary strength of this study was that we leveraged a daily assessment design administered through a digital PA platform to examine dynamic PA-based social comparison target selections and their real-time predictors. Specifically, the study captured social comparison information preferences and PA engagement behaviourally and in daily life using an appropriate modelling approach. Limitations are also noteworthy. As an initial step toward understanding within-person variability in health information preferences, this project used a secondary analysis approach on an existing dataset, and the original study did not include measures specifically designed to assess health information preferences. This study focused on PA-based social comparison target selections only, and it is not clear whether these findings can be generalised to other types of social comparison information or to health information preferences more generally.

Further, whether findings can be generalised to a broader population of insufficiently active adults of a more diverse age range, as well as racial/ethnic, educational, and socioeconomic background, is unclear. In particular, women in this study were primarily white and well-educated. Future studies that include more women from underrepresented communities, as well as younger women, will be crucial in clarifying whether these findings generalise to other groups. Finally, the total number of observations collected in this study yielded adequate but modest statistical power (Maas & Hox, 2005; Murayama et al., 2022), which should be addressed with larger sample sizes in future work. As indicated above, additional work that is specifically aimed at measuring health information seeking processes and preferences in daily life would contribute value and is critical to understanding the role of dynamic information preferences in PA behaviour and related experiences, including among women in midlife with elevated CVD risk.

Conclusions

This study provides preliminary evidence that health information preferences—specifically social comparison target selections—exhibit substantial within-person variability among women in midlife with elevated CVD risk. While much existing work shows health information preferences as stable, our findings suggest that they may indeed fluctuate day-to-day and be influenced by contextual factors, such as women’s satisfaction with their PA from the previous day. The dynamic nature of these processes highlights the importance of assessing health information preferences as they occur, rather than relying solely on global, retrospective measures. Further research is needed to develop appropriate measures of dynamic health information preferences, explore additional contextual factors influencing these experiences, and determine to what extent these information preferences are linked to actual PA behaviour. Ultimately, integrating these dynamic assessments into adaptive health technologies may enhance intervention efficacy and support sustained behaviour change among women in midlife with elevated CVD risk and other at-risk groups.

Funding Statement

This work was supported by the National Institutes of Health under Grants K23 HL136657 and DP2 HL173857 (PI: Arigo).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Informed consent

Participants provided written informed consent before participating in the study. See details under Methods.

References

  1. Allmeta, A. , Arigo, D. , & König, L. M. (2026). Easy or difficult? Investigating perceived ease of changing eating and physical activity behaviors. Applied Psychology: Health and Well-Being , 18 (1), e70124. 10.1111/aphw.70124 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Arigo, D. , & Butryn, M. L. (2019). Prospective relations between social comparison orientation and weight loss outcomes. Behavioral Medicine , 45 (3), 249–254. 10.1080/08964289.2018.1481010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arigo, D. , Cornell, M. , & Smyth, J. M. (2018). Social comparisons in adults with type 2 diabetes: patients’ reasons for target selection. Psychology & Health , 33 (7), 906–924. 10.1080/08870446.2018.1432759 [DOI] [PubMed] [Google Scholar]
  4. Arigo, D. , Mogle, J. A. , & Smyth, J. M. (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. 10.1007/s10865-021-00229-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Arigo, D. , Brown, M. M. , Pasko, K. , & Suls, J. (2020). Social comparison features in physical activity promotion apps: scoping meta-review. Journal of Medical Internet Research , 22 (3), e15642. 10.2196/15642 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Arigo, D. , Lobo, A. F. , Ainsworth, M. C. , Baga, K. , & Pasko, K. (2022). Development and initial testing of a personalized, adaptive, and socially focused web tool to support physical activity among women in midlife: multidisciplinary and user-centered design approach. JMIR Formative Research , 6 (7), e36280. 10.2196/36280 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Arigo, D. , Gray, R. C. , Dallal, D. H. , 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 , 1–18, e41239. 10.2196/41239 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Arigo, D. , Lassiter, J. M. , Baga, K. , Jackson, D. A. , Lobo, A. F. , & Guetterman, T. C. (2023). You get what you need when you need it”: A mixed methods examination of the feasibility and acceptability of a tailored digital tool to promote physical activity among women in midlife. Digital Health , 9 , 20552076231210654. 10.1177/20552076231210654 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Arigo, D. , Mogle, J. A. , Brown, M. M. , Pasko, K. , Travers, L. , Sweeder, L. , & Smyth, J. M. (2020). Methods to assess social comparison processes within persons in daily life: A scoping review. Frontiers in Psychology , 10 , 2909. 10.3389/fpsyg.2019.02909 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Arigo, D. , Romano, K. A. , Pasko, K. , Travers, L. , Ainsworth, M. C. , Jackson, D. A. , & Brown, M. M. (2022). A scoping review of behavior change techniques used to promote physical activity among women in midlife. Frontiers in Psychology , 13 , 855749. 10.3389/fpsyg.2022.855749 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bandura, A. (2008). “An agentic perspective on positive psychology.” In Positive psychology: Exploring the best in people, Vol 1: Discovering human strengths (pp. 167–196). Praeger Publishers/Greenwood Publishing Group. [Google Scholar]
  12. Beck, A. T. , & Steer, R. A. (1990). Beck Anxiety Inventory: BAI; Manual . Psychological Corporation, Harcourt Brace Jovanovich. [Google Scholar]
  13. Beck, A. T. , Epstein, N. , Brown, G. , & Steer, R. A. (1988). An inventory for measuring clinical anxiety: psychometric properties. Journal of Consulting and Clinical Psychology , 56 (6), 893–897. 10.1037//0022-006x.56.6.893 [DOI] [PubMed] [Google Scholar]
  14. Bolger, N. , & Laurenceau, J.-P. (2013). Intensive longitudinal methods: An introduction to diary and experience sampling research . xv–256. Guilford Press. [Google Scholar]
  15. Brown, D. J. , Ferris, D. L. , Heller, D. , & Keeping, L. M. (2007). Antecedents and consequences of the frequency of upward and downward social comparisons at work. Organizational Behavior and Human Decision Processes, Special Issue on Social Comparison Processes , 102 (1), 59–75. 10.1016/j.obhdp.2006.10.003 [DOI] [Google Scholar]
  16. Buunk, B. P. , & Ybema, J. F. (1997). “Social comparisons and occupational stress: the identification-contrast model.” In Health, coping, and well-being: Perspectives from social comparison theory . 359–388. Lawrence Erlbaum Associates Publishers. [Google Scholar]
  17. Buunk, B. P. , Collins, R. L. , Taylor, S. E. , VanYperen, N. W. , & Dakof, G. A. (1990). The affective consequences of social comparison: either direction has its ups and downs. Journal of Personality and Social Psychology , 59 (6), 1238–1249. 10.1037/0022-3514.59.6.1238 [DOI] [PubMed] [Google Scholar]
  18. Caltabiano, M. L. , & Ghafari, N. (2011). The effect of social comparison processes, athletic prototypes, perceived behavioural control and perceived threat on physical activity. International Journal of Health Promotion and Education , 49 (2), 44–52. 10.1080/14635240.2011.10708208 [DOI] [Google Scholar]
  19. Christensen, L. , & Brooks, A. (2006). Changing food preference as a function of mood. The Journal of Psychology , 140 (4), 293–306. 10.3200/JRLP.140.4.293-306 [DOI] [PubMed] [Google Scholar]
  20. Cohen, J. (2013). Statistical power analysis for the behavioral sciences (2nd ed.). Routledge. 10.4324/9780203771587 [DOI] [Google Scholar]
  21. Curran, P. J. , & Bauer, D. J. (2011). The disaggregation of within-person and between-person effects in longitudinal models of change. Annual Review of Psychology , 62 (1), 583–619. 10.1146/annurev.psych.093008.100356 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Dąbrowska, J. , Dąbrowska-Galas, M. , Naworska, B. , Wodarska, M. , & Plinta, R. (2015). The role of physical activity in preventing obesity in midlife women. Przegla̜d Menopauzalny = Menopause Review , 14 (1), 13–19. 10.5114/pm.2015.49252 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Deniz, M. , Güler, K. , & Öcal, S. (2021). Analysis of the relationship between coronaphobia, health anxiety, and life satisfaction. Uluslararası Türk Kültür Coğrafyasında Sosyal Bilimler Dergisi , 6 (2), Article 2. https://izlik.org/JA96AU52JX [Google Scholar]
  24. 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. 10.1521/soco.2019.37.3.247 [DOI] [Google Scholar]
  25. Diel, K. , Hofmann, W. , Grelle, S. , Boecker, L. , & Friese, M. (2024). Prepare to compare: effects of an intervention involving upward and downward social comparisons on goal pursuit in daily life. Personality and Social Psychology Bulletin , 01461672231219378 , 1523–1537. 10.1177/01461672231219378 [DOI] [PubMed] [Google Scholar]
  26. El Khoudary, S. R. , Aggarwal, B. , Beckie, T. M. , Hodis, H. N. , Johnson, A. E. , Langer, R. D. , Limacher, M. C. , Manson, J. E. , Stefanick, M. L. , & Allison, M. A. (2020). Menopause transition and cardiovascular disease risk: implications for timing of early prevention: A scientific statement from the American heart association. Circulation, 142 (25), e506–e532. 10.1161/CIR.0000000000000912 [DOI] [PubMed] [Google Scholar]
  27. Elagizi, A. , Kachur, S. , Carbone, S. , Lavie, C. J. , & Blair, S. N. (2020). A review of obesity, physical activity, and cardiovascular disease. Current Obesity Reports , 9 (4), 571–581. 10.1007/s13679-020-00403-z [DOI] [PubMed] [Google Scholar]
  28. Festinger, L. (1954). A theory of social comparison processes. Human Relations , 7 (2), 117–140. 10.1177/001872675400700202 [DOI] [Google Scholar]
  29. Hamaker, E. L. (2012). Why researchers should think “within-person”: A paradigmatic rationale, Handbook of research methods for studying daily life . 43–61. The Guilford Press. [Google Scholar]
  30. Hardeman, W. , Houghton, J. , Lane, K. , Jones, A. , & Naughton, F. (2019). A systematic review of just-in-time adaptive interventions (JITAIs) to promote physical activity. International Journal of Behavioral Nutrition and Physical Activity , 16 (1), 31. 10.1186/s12966-019-0792-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Herrera-Espiñeira, C. , Rodríguez del Aguila, M. , del, M. , Rodríguez del Castillo, M. , Valdivia, A. F. , & Sánchez, I. R. (2009). Relationship between anxiety level of patients and their satisfaction with different aspects of healthcare. Health Policy (Amsterdam, Netherlands), 89 (1), 37–45. 10.1016/j.healthpol.2008.04.012 [DOI] [PubMed] [Google Scholar]
  32. Higgins, E. T. (2012). Regulatory focus theory, Handbook of theories of social psychology (Vol. 1 , pp. 483–504). Sage Publications Ltd. 10.4135/9781446249215.n24 [DOI] [Google Scholar]
  33. Higgins, E. T. , Friedman, R. S. , Harlow, R. E. , Idson, L. C. , Ayduk, O. N. , & Taylor, A. (2001). Regulatory focus questionnaire (RFQ). APA PsycTests . 10.1037/t67916-000 [DOI] [Google Scholar]
  34. Hoffman, L. (2015). Longitudinal analysis: Modeling within-person fluctuation and change . Routledge. [Google Scholar]
  35. Howell, J. L. , & Sweeny, K. (2025). Information-management behavior during stressful waiting periods. Emotion (Washington, D.C.), 25 (7), 1774–1783. 10.1037/emo0001529 [DOI] [PubMed] [Google Scholar]
  36. Jakubisin Konicki, A. (2019). Promoting cardiovascular health in midlife women. Current Obstetrics and Gynecology Reports , 8 (4), 145–151. 10.1007/s13669-019-00275-8 [DOI] [Google Scholar]
  37. Joshanloo, M. (2022). Self-esteem predicts positive affect directly and self-efficacy indirectly: A 10-year longitudinal study. Cognition and Emotion , 36 (6), 1211–1217. 10.1080/02699931.2022.2095984 [DOI] [PubMed] [Google Scholar]
  38. Joynt Maddox, K. E. , Reynolds, H. R. , Adedinsewo, D. , Bushnell, C. , DeVon, H. A. , Gooding, H. C. , Howard, V. J. , Mauricio, R. , Miller, E. C. , Sharma, G. , & Waken, R. J. on behalf of the American Heart Association Women’s Health Science Committee of the Council on Clinical Cardiology and Stroke Council, Council on Basic Cardiovascular Sciences; Council on Cardiovascular and Stroke Nursing, Council on Lifelong Congenital Heart Disease and Heart Health in the Young, & Council on Peripheral Vascular Disease (2026). Forecasting the burden of cardiovascular disease and stroke in women in the United States through 2050: A scientific statement from the American heart association. Circulation, 153 (13), e1061–e1077. 10.1161/CIR.0000000000001406 [DOI] [PubMed] [Google Scholar]
  39. Kraus, W. E. , Powell, K. E. , Haskell, W. L. , Janz, K. F. , Campbell, W. W. , Jakicic, J. M. , Troiano, R. P. , Sprow, K. , Torres, A. , & Piercy, K. L. (2019). Physical activity, all-cause and cardiovascular mortality, and cardiovascular disease. Medicine and Science in Sports and Exercise , 51 (6), 1270–1281. 10.1249/MSS.0000000000001939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lai, M. H. C. , & Kwok, O. (2015). Examining the rule of thumb of not using multilevel modeling: the “design effect smaller than two” rule. Journal of Experimental Education , 83 (3), 423–438. 10.1080/00220973.2014.907229 [DOI] [Google Scholar]
  41. Laroche, M. , Roussel, P. , Cury, F. , & Boiché, J. (2019). Understanding the dynamics of physical activity practice in the health context through regulatory focus and self-determination theories. PLoS One , 14 (8), e0216760. 10.1371/journal.pone.0216760 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Latimer, A. E. , Rivers, S. E. , Rench, T. A. , Katulak, N. A. , Hicks, A. , Hodorowski, J. K. , Higgins, E. T. , & Salovey, P. (2008). A field experiment testing the utility of regulatory fit messages for promoting physical activity. Journal of Experimental Social Psychology , 44 (3), 826–832. 10.1016/j.jesp.2007.07.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Lewis, N. , Andrews, E. A. , Keefe, D. , & Walter, N. (2024). A meta-analytical review of the relationship, antecedents, and consequences of information seeking and information scanning. Human Communication Research , 50 (4), 468–479. 10.1093/hcr/hqae012 [DOI] [Google Scholar]
  44. Li, J. , Theng, Y.-L. , & Foo, S. (2016). Predictors of online health information seeking behavior: changes between 2002 and 2012. Health Informatics Journal , 22 (4), 804–814. 10.1177/1460458215595851 [DOI] [PubMed] [Google Scholar]
  45. Maas, C. J. , & Hox, J. J. (2005). Sufficient sample sizes for multilevel modeling. Methodology , 1 (3), 86–92. 10.1027/1614-2241.1.3.86 [DOI] [Google Scholar]
  46. McArthur, D. , Dumas, A. , Woodend, K. , Beach, S. , & Stacey, D. (2014). Factors influencing adherence to regular exercise in middle-aged women: A qualitative study to inform clinical practice. BMC Women’s Health , 14 (1), 49. 10.1186/1472-6874-14-49 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Miller, S. M. (1987). Monitoring and blunting: validation of a questionnaire to assess styles of information seeking under threat. Journal of Personality and Social Psychology , 52 (2), 345–353. 10.1037/0022-3514.52.2.345 [DOI] [PubMed] [Google Scholar]
  48. Miller, S. M. (1995). Monitoring versus blunting styles of coping with cancer influence the information patients want and need about their disease. Implications for cancer screening and management. Cancer , 76 (2), 167–177. 10.1002/1097-0142(19950715)76:2<167::aid-cncr2820760203>3.0.co;2-k [DOI] [PubMed] [Google Scholar]
  49. Mollee, J. S. , & Klein, M. C. A. (2017). The effectiveness of upward and downward social comparison of physical activity in an online intervention: 15th International Conference on Ubiquitous Computing and Communications and 2016 8th International Symposium on Cyberspace and Security, IUCC-CSS 20162016 15th International Conference on Ubiquitous Computing and Communications and 2016 8th International Symposium on Cyberspace and Security (IUCC-CSS)109-115. 10.1109/IUCC-CSS.2016.023 [DOI]
  50. Moody, W. , Kinderman, P. , & Sinha, P. (2010). An exploratory study: relationships between trying on clothing, mood, emotion, personality and clothing preference. Journal of Fashion Marketing and Management: An International Journal , 14 (1), 161–179. 10.1108/13612021011025483 [DOI] [Google Scholar]
  51. 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. 10.1037/met0000330 [DOI] [PubMed] [Google Scholar]
  52. Muris, P. , van Zuuren, F. J. , De Jong, P. J. , De Beurs, E. , & Hanewald, G. (1994). Monitoring and blunting coping styles: the miller behavioural style scale and its correlates, and the development of an alternative questionnaire. Personality and Individual Differences , 17 (1), 9–19. 10.1016/0191-8869(94)90257-7 [DOI] [Google Scholar]
  53. Nicol, G. E. , Ricchio, A. R. , Metts, C. L. , Yingling, M. D. , Ramsey, A. T. , Schweiger, J. A. , Miller, J. P. , & Lenze, E. J. (2020). A smartphone-based technique to detect dynamic user preferences for tailoring behavioral interventions: observational utility study of ecological daily needs assessment. JMIR mHealth and uHealth , 8 (11), e18609. 10.2196/18609 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Pavlova, M. K. , Lechner, C. M. , & Silbereisen, R. K. (2018). Social comparison in coping with occupational uncertainty: self-improvement, self-enhancement, and the regional context. Journal of Personality , 86 (2), 320–333. 10.1111/jopy.12317 [DOI] [PubMed] [Google Scholar]
  55. Peugh, J. L. (2010). A practical guide to multilevel modeling. Journal of School Psychology , 48 (1), 85–112. 10.1016/j.jsp.2009.09.002 [DOI] [PubMed] [Google Scholar]
  56. Ramsey, I. , Corsini, N. , Peters, M. D. J. , & Eckert, M. (2017). A rapid review of consumer health information needs and preferences. Patient Education and Counseling , 100 (9), 1634–1642. 10.1016/j.pec.2017.04.005 [DOI] [PubMed] [Google Scholar]
  57. Rancourt, D. , Leahey, T. M. , LaRose, J. G. , & Crowther, J. H. (2015). Effects of weight-focused social comparisons on diet and activity outcomes in overweight and obese young women. Obesity, 23 (1), 85–89. 10.1002/oby.20953 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Shufelt, C. , Dutra, E. , Torbati, T. , & Ramineni, T. (2019). A clinical prescription for heart health in midlife women. Maturitas, 119 , 46–53. 10.1016/j.maturitas.2018.11.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Spitzer, R. L. , Kroenke, K. , Williams, J. B. W. , & Löwe, B. (2006). A brief measure for assessing generalized anxiety disorder: the GAD-7. Archives of Internal Medicine , 166 (10), 1092–1097. 10.1001/archinte.166.10.1092 [DOI] [PubMed] [Google Scholar]
  60. Stone, A. A. , & Shiffman, S. (2002). Capturing momentary, self-report data: A proposal for reporting guidelines. Annals of Behavioral Medicine: A Publication of the Society of Behavioral Medicine , 24 (3), 236–243. 10.1207/S15324796ABM2403_09 [DOI] [PubMed] [Google Scholar]
  61. Suls, J. , Martin, R. , & Wheeler, L. (2002). Social comparison: why, with whom, and with what effect? Current Directions in Psychological Science , 11 (5), 159–163. 10.1111/1467-8721.00191 [DOI] [Google Scholar]
  62. Taylor, S. E. , & Lobel, M. (1989). Social comparison activity under threat: downward evaluation and upward contacts. Psychological Review , 96 (4), 569–575. 10.1037/0033-295x.96.4.569 [DOI] [PubMed] [Google Scholar]
  63. van de Wal, D. , van Doorn, B. , den Hollander, D. , Desar, I. M. E. , Gelderblom, H. , Oosten, A. W. , Reyners, A. K. L. , Steeghs, N. , van der Graaf, W. T. A. , & Husson, O. (2023). The association of having a monitoring or blunting coping style with psychological distress, health-related quality of life and satisfaction with healthcare in gastrointestinal stromal tumour (GIST) patients. Acta Oncologica , 62 (12), 1616–1624. 10.1080/0284186X.2023.2269302 [DOI] [PubMed] [Google Scholar]
  64. Veenman, M. V. J. (2011). Alternative assessment of strategy use with self-report instruments: A discussion. Metacognition and Learning , 6 (2), 205–211. 10.1007/s11409-011-9080-x [DOI] [Google Scholar]
  65. Williams-Piehota, P. , Latimer, A. E. , Katulak, N. A. , Cox, A. , Silvera, S. A. N. , Mowad, L. , & Salovey, P. (2009). Tailoring messages to individual differences in monitoring-blunting styles to increase fruit and vegetable intake. Journal of Nutrition Education and Behavior , 41 (6), 398–405. 10.1016/j.jneb.2008.06.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Willis, G. B. (2004). Cognitive interviewing: A tool for improving questionnaire design . SAGE Publications. [Google Scholar]
  67. Wood, J. V. , Taylor, S. E. , & Lichtman, R. R. (1985). Social comparison in adjustment to breast cancer. Journal of Personality and Social Psychology , 49 (5), 1169–1183. 10.1037/0022-3514.49.5.1169 [DOI] [PubMed] [Google Scholar]
  68. Wutich, A. , Beresford, M. , & Bernard, H. R. (2024). Sample sizes for 10 types of qualitative data analysis: an integrative review, empirical guidance, and next steps. International Journal of Qualitative Methods , 23 . 16094069241296206. 10.1177/16094069241296206 [DOI] [Google Scholar]
  69. Xue, C. , Li, T. , Yin, S. , Zhu, X. , & Tan, Y. (2018). The influence of induced mood on music preference. Cognitive Processing , 19 (4), 517–525. 10.1007/s10339-018-0872-7 [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

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Health Psychology and Behavioral Medicine are provided here courtesy of European Health Psychology Society

RESOURCES