Sedentary Behavior Affects the Ratings of Workplace Physical Activity Sessions

Abstract

It is not known how individuals with various levels of sedentary behavior rate their experiences with physical activity. The absence of these data diminishes the ability to tailor programs and interventions for those with different sedentary behavior patterns. We hypothesize that those who engage in greater levels of sedentary behavior will rate their experiences of workplace physical activity sessions less favorably than those who engage in lower levels of sedentary behavior.

The study was a within subject, repeated-measures design. Sixty participants completed six distinct workplace physical activity sessions (i.e., 360 data points) and rated each session immediately after with self-report measures of social cognitive constructs (e.g., benefit, confidence, enjoyment, and likability). Generalized estimating equations were used to analyze the data. Those who sat more than 6 hours/day on non-workdays rated sessions more favorably on enjoyment and likability constructs, but less favorably on confidence. Those who sat 10 hours/day on workdays rated physical activity sessions less favorably on the constructs of enjoyment and likability. Furthermore, sedentary behaviors had no effect on ratings of perceived benefits. Interventions designed to promote physical activity should assess sedentary behavior patterns (both workday and non-workday) and evaluate the effects on social cognitive perceptions of physical activity.

Introduction

Social cognitive processes motivate, energize, and predict human behavior (Bandura, 1986, 2001). During training and competition, professional athletes employ self-talk such as “I got this,” “Be relentless,” “Why not today,” “Stay after it,” and “This is my moment.” In contrast, we know little about what motivates nonprofessionals and recreational physical activity enthusiasts; different levels of sedentary behavior may be related to individuals’ willingness to be physically active or their perceived confidence levels. From a public health perspective, this lack of information undermines our ability to develop effective interventions to reduce sedentary behavior and promote physical activity.

There are many benefits associated with reducing sedentary behavior and promoting physical activity. Based on evidence from the 2018 Physical Activity Guidelines Advisory Committee Scientific Report, “for individuals who perform no or little moderate-to-vigorous physical activity, replacing sedentary behavior with light-intensity physical activity reduces the risk of all-cause mortality, cardiovascular disease incidence and mortality, and the incidence of type 2 diabetes. Before this report, evidence that light-intensity physical activity could provide health benefits had not been clearly stated” (2018 Physical Activity Guidelines Advisory Committee, 2018). In addition, interruptions or breaks in sedentary time reduce health risks irrespective of total time spent in sedentary behavior (i.e., volume of sedentary time). Thus, reducing sedentary behaviors with breaks or interruptions and increasing light-intensity physical activities have documented health benefits (Bellettiere et al., 2017; Dunstan et al., 2011; Healy et al., 2011; Reid et al., 2018).

To increase physical activity and decrease sedentary behavior, it is important to know what factors (i.e., correlates) are associated with physical activity and sedentary behavior. In a comprehensive review of physical activity correlates, social cognitive constructs were consistently associated with physical activity (Bauman et al., 2012). Conversely, fewer studies have reported on the correlates of sedentary behavior (Paxton et al., 2016). One study reported that implicit attitudes toward sedentary behavior were not associated with physical activity (Chevance et al., 2018) which confirms other studies that indicated physical and sedentary behaviors have different correlates (Owen et al., 2011). What is less understood is how the time spent in sedentary behaviors influences perceptions of physical activity.

There are two distinct hypotheses related to the amount of sedentary behavior and willingness to be physically active. One hypothesis is that given the widespread dissemination of information that sedentary behavior has adverse health effects, both in the popular press (Fiorenzi, 2021) and scientific literature (Hamilton et al., 2008; Owen, 2012), perhaps, people who have high amounts of sedentary behavior will be more receptive to participate in an easy, time-limited session to reduce sedentary behavior and increase physical activity. An alternative hypothesis is that if a person is sedentary, there may be less receptivity and more resistance to participate in any type of program to reduce sedentary behavior and increase physical activity, irrespective of the incentives and ease of participation. These two hypotheses have not been empirically tested. We hypothesize that those who engage in greater levels of sedentary behavior will rate physical activity less favorably than those who engage in lower levels of sedentary behavior.

Methods

Participants and Recruitment

This study was approved by the Committee for the Protection of Human Subjects, the institutional review board (IRB), at The University of Texas Health Science Center at Houston

The university’s institutional review board (IRB) approved all the study protocol and procedures. (protocol number HSC-SPH-14-0171). Before any study procedures, participants read and signed a written informed consent. Eligibility criteria included participants who were: (a) 18 to 69 years of age; (b) employed full time; and (c) willing to participate in mild- to- moderate-intensity, physical activity sessions. Participants were recruited from a large church and three medical and health science center institutions in an urban city in the United States. Before engaging in any of the workplace physical activity sessions, participants completed the Physical Activity Readiness Questionnaire (PAR-Q) (Shephard, 1988; Thomas et al., 1992). This questionnaire screened participants for medical issues that contraindicated physical activity.

Study Design and Setting

This study was a within subject, repeated-measures design wherein participants completed six different physical activity sessions and rated each session immediately thereafter. Ratings were based on twelve different dimensions. Participants were asked to arrive in their usual work attire and attend two evening sessions to complete all six sessions. The physical activity sessions took place at two venues that had large open spaces, were temperature controlled, and featured a large stage where the certified personal trainees delivered the physical activity session instructions and demonstrations. Three sessions were completed each evening. Immediately following each routine, there was a 15-minute break, whereby participants sat down and completed a questionnaire to rate the session (no consultation with others was permitted).

Description of Workplace Physical Activity Sessions

The workplace physical activity sessions were designed for the 15-minute work break by certified personal trainers and research staff (Taylor et al., 2021). The sessions consisted of a 1- to 2-minute warm-up period, a 10- to 12- minute physical activity period, and a 1- to 2-minute cool down period. To maintain consistency, the warm-up and cool-down movements were the same for all sessions. The sessions targeted major muscle groups in the upper and lower body, ranged from light-to moderate-intensity, and included different types of exercises (aerobic, muscle strengthening, stretching, or a combination). The descriptions of the six physical activity sessions are in Table 1.

Table 1.

Description of the Six Different Physical Activity Sessions.

Exercise Routine	Type(s) of Exercises Included	Target Intensity	Number of Different Exercises (or Movements)	Exercise Description
1	Muscle Strengthening	Light-to-Moderate	9	Exercises were performed with body weight. Each exercise included 2 sets of 15 reps, with a 30-second break between sets. Examples included wall push-ups, bent row, and squats.
2	Stretching and Balance (Yoga inspired)	Light	11	Exercises were inspired by beginner yoga poses. Each stretch or pose was held for 15 seconds and repeated at least once with a 10 second breaks in between. Examples included open chest stretch and balance tree pose.
3	Aerobic and Muscle Strengthening (Circuit training inspired)	Moderate	12	Exercises were performed in 30-second bouts with 30 seconds of rest between exercises. Examples included modified jumping jack, simulated jump rope, and alternating triceps kickbacks.
4	Stretching, Aerobic, and Muscle Strengthening (Ballet inspired)	Light-to-Moderate	10	Exercises were inspired by ballet movements and featured a choreographed sequence of movements with some breaks. Examples included simulated rope climb and plies with calf raises (e.g., plie squat with calf raise).
5	Muscle Strengthening and Stretching (Booster Break inspired)	Light-to-Moderate	10	Exercises were inspired by an existing workplace routine (Booster Break). Some exercises included 1 set of 10 reps while other exercises included up to 2–5 sets performed for 5–20 seconds. Examples included modified lunges, modified neck roll, wrist flexions, and extensions.
6	Aerobic Dance (Aerobic dance inspired)	Light-to-Moderate	10	Exercises were inspired by aerobic dance wherein a series of movements were sequenced with breaks of marching in place to maintain a steady heart rate.

To characterize differences, each routine was named based on source of inspiration: aerobic dance, ballet, Booster Break, circuit training, muscle strengthening, and yoga. Certified personnel trainers and research staff reviewed and practiced each session to confirm that the main movements of each routine were different and that participants could complete the session in office work attire. The six sessions were selected to represent a range of movements, requiring no equipment, and appropriate for the 15-minute work break. The sessions incorporated a variety of light- to moderate-intensity movements (Taylor et al., 2021).

Primary Outcome Measures

The research team developed a series of questions to evaluate dimensions of the physical activity sessions. A 5-point Likert-type scale was used, where “1” represented the lowest level, “3” was moderate, and “5” represented the highest level of the dimension. For this study, four dimensions were selected with a particular emphasis on social cognitive correlates cited in previous reviews (Bauman et al., 2012). The Confidence dimension is self-efficacy [i.e., the extent to which you are certain that you can perform the routine (belief in one’s own ability)] (Bandura, 1986, 2001). Other constructs included Benefit (i.e., the extent to which the routine enhances one’s health and well-being), Enjoyment (i.e., the extent to which a feeling of pleasure, satisfaction, or fun is experiences by doing the routine) and Likability (i.e., the extent to which the routine is engaging or appealing). The measures of each construct are presented in Table 2.

Table 2.

Social Cognitive Dimensions and Definitions of Physical Activity Ratings.

Social Cognitive Dimensions	Definitions of Physical Activity Ratings
Confidence level (Self-efficacy)	The extent to which you are certain that you can perform the routine (belief in one’s own ability).
Benefit	The extent to which the routine enhances one’s health and well-being
Enjoyment level	The extent to which a feeling of pleasure, satisfaction, or fun is experienced by doing the routine.
Likability level	The extent to which the routine is engaging or appealing.

Note: The rating scale was from one (low) to five (high)

Assessments and Measures

All surveys were in English. The first part of the survey collected demographic data to describe the sample and enable subgroup analyses. Self-report surveys were administered to assess participants’ demographic characteristics, physical activity, and sedentary behavior. Demographic questions included sex, age, race/ethnicity, occupation, and place of work. In addition, participants self-reported their height and weight, which were used to calculate body mass index (BMI). Self-reported physical activity was assessed using the International Physical Activity Questionnaire (IPAQ). The IPAQ was developed to measure physical activity in adult populations (ages 18–65 years old) and has strong psychometric properties (Hallal et al., 2004). Participants’ sitting patterns were assessed using the Multicontext Sitting Questionnaire (MSTQ); the questionnaire has acceptable psychometric properties (Whitfield et al., 2013). The MSTQ assesses typical workday and non-workday sitting behavior across several sitting contexts including sitting while at work, watching TV or movies, using a computer or videogame, transportation, and while talking or socializing. An important distinction is that workday and non-workday sitting was reported separately.

Statistical Model

The study analyzed ratings on 4 dimensions: Benefit, Confidence (Self-efficacy), Enjoyment, and Likability. The original rating scale was a 5-point Likert-type response scale with 1 being the lowest and 5 highest. Upon examining the data, we noticed that ratings were skewed (i.e., predominantly 4 or 5, see Table 4). Thus, we made the decision to collapse ratings into “favorable” (rating ≥ 4) vs. “unfavorable” (rating < 4). There may be some lose information due to dichotomization. However, the skewedness of the data should minimize any loss of information. The resulting outcomes were binary (i.e., favorable vs. unfavorable), and were analyzed with a logit link. For each outcome, the marginal model (i.e., GEE-type) was given by:

Table 4.

Average Ratings of the Social Cognitive Dimensions

Social Cognitive Dimensions	Enjoyment		Likability		Benefit		Confidence (Self-efficacy)
Type of Exercise	Mean (SD)	%	Mean (SD)	%	Mean (SD)	%	Mean (SD)	%
Ballet	4.42 (.70)	.88	4.38 (.78)	.82	4.22 (.84)	.78	4.18 (.94)	.76
Booster Break	4.54 (.65)	.92	4.44 (.80)	.85	4.50 (.80)	.90	4.35 (.86)	.83
Circuit training	4.36 (.75)	.88	4.18 (.85)	.80	4.62 (.64)	.92	3.32 (1.38)	.54
Aerobic dance	4.25 (.94)	.80	4.05 (1.29)	.80	4.11 (.97)	.73	3.54 (1.28)	.59
Muscle Strengthening	4.59 (.64)	.96	4.35 (.82)	.86	4.15 (.92)	.78	4.51 (.70)	.88
Yoga	4.06 (1.08)	.74	3.96 (1.07)	.70	4.21 (.95)	.77	3.96 (1.18)	.72

Notes: % = the proportion of ratings that are ≥ 4; SD = standard deviation

$$\text{log}(\text{P}({\text{Y}}_{\text{ji}}=1)/(1-\text{P}({\text{Y}}_{\text{ji}}=1)))=\mathrm{\alpha }+{\mathrm{\beta }}_1\hspace{0.17em}{\text{X}}_{\text{ji}}+{\mathrm{\beta }}_2{\text{Z}}_{\text{i}}$$

where Y_ji the rating that participant i rates routine j on a particular dimension, P(Y_ji=1) is the probability of Y_ji=1 (i.e., probability of receiving favorable rating). X_ji represents covariates that vary between sessions. In the study, X_ji include a set of dummy-coding variables to indicate the 6 routines. Z_i stands for covariates that remain unchanged across session. In the study, of interest were sitting times on workdays and non-workdays. Other (routine-invariant) covariates included as controlled variables were gender, age, race, body mass index, and total METS. The conventional way to interpret β is the odds ratio, which is the exponent of β. A positive (negative) β implies an increased (decreased) odd of receiving favorable rating, as the value of the corresponding covariate increases by one unit. The empirical (“sandwich”) covariance matrix was used for statistical inference. The empirical (“sandwich”) covariance matrix was used for statistical inference. For a binary outcome, a minimum of 30-40 clusters (subjects, in longitudinal/repeated measures data) are needed for GEE to produce unbiased standard error estimates (McNeish & Harring, 2017).

Results

Sample Description

Workday and non-workday sitting times (per day) were derived from the MSTQ by summing self-reported sitting hours for different categories. For a participant who missed reporting sitting time for a category, the (total) sitting time was set to missing. This process ensures that sitting time is comparable among participants. For instance, a participant whose sitting time is 8 hours/day can be said to sit less than another whose sitting time is 10 hours/day because the individual indeed sits less, not because the individual misses reporting on some category. The initial sample included 94 participants, out of which 80 participants had non-missing data on both workday and non-workday sitting time. The analysis was performed on the sample of 60 participants whose total hours of sitting plus sleeping were less than 23 hours/day. It is reasonable to believe that a typical person would spend some time on activities that are not captured in the MTSQ.

The average sitting time on workdays and non-workdays are 11.30 hours/day (± 3.08) and 9.22 hours/day (± 3.28), respectively. The correlation between workdays and non-workdays sitting time was significant 0.38 (p=.003). On average participants were mostly female (68%), 47 (±12.77) years old, overweight or obese (average BMI= 31.38 ±7.09), and inactive (METs 251.94 ± 538.84). In addition, the sample was racially and ethnically diverse with 41.7% non-Hispanic White, 38.3% African American/Black, and 20.0% classified as other races (Table 3). The Ratings on Benefit, Confidence (Self-efficacy), Enjoyment, and Likability for each of the sessions are summarized in Table 4.

Table 3.

Demographics Table of the Study Participants.

Variables	Percentage	Mean (SD)
Sex:	68.3
Female	31.7
Male
Race:	41.7
Non-Hispanic White	38.3
African American	20.0
Others
Age (yrs)		46.68 (12.77)
BMI		31.38 (7.10)
Inactive METs		251.80 (538.85)
Workday sedentary (hrs.)		11.30 (3.08)
Non-workday sedentary (hrs.)		9.22 (3.28)

Notes: SD=standard deviation; METs=metabolic equivalents of tasks; yrs=years; hrs. =hours

K-means clustering suggested 5 clusters based on sitting time (Figure 1). In cluster 1, participants appeared to have highest sitting time, while participants in cluster 4 had lowest sitting time on both workdays and non-workdays. Clusters 1, 3, and 5 had workday sitting times >10 hours/day; their non-workday sitting time straddled 3-16 hours/day. Clusters 2 and 4, sat <10 hours on workdays, with the former having greater sitting times on non-workdays. Proc HPCLUS was used to run k-means clustering. The aligned box criterion (ABC) was used to determine the number of clusters.

Figure 1.

Clustering of Sedentary Behaviors.

Note: K-means clustering suggested 5 clusters based on sitting time.

GEE Model Results

Based on the exploratory analysis, two categorical variables wd-sit and wk-sit were defined and used as explanatory variables in the primary analyses using the GEE models. Variable wd-sit dichotomizes workday sitting time such that wd-sit = 1 if a participant sits (strictly) >10 hours/day and = 0 otherwise. Variable wk-sit categorized non-workday sitting time, defined as wk-sit = 1 if a participant sat (strictly) >6 hours/day and = 0 otherwise. The hypothesis was that those who engage in greater levels of sedentary behavior would rate workplace physical activity sessions less favorably than those who engage in lower levels of sedentary behavior. This hypothesis, however, did not account for differences between workday and non-workday sitting. Nevertheless, sedentary behaviors did affect ratings of physical activity sessions. These results are presented in Table 5.

Table 5.

Effects of Sedentary Behaviors on Ratings of Social Cognitive Dimensions

Social Cognitive Dimensions	Enjoyment		Likability		Benefit		Confidence (Self-efficacy)
Ratings	β (se)	p	β (se)	p	β (se)	p	β (se)	p
Non-workday sitting >6 vs. ≤6 hours/day	1.363 (.542)	.012	1.183 (.543)	.029	0.766 (.491)	.119	−1.436 (.591)	.015
Workday sitting >10 vs. ≤10 hours/day	−1.460 (.692)	.035	−1.076 (.544)	.048	−0.829 (.525)	.115	0.355 (.470)	.450

Notes: se = standard error; p = p-value

Enjoyment and Likability: Participants who reported sitting >6 hours/day on non-workdays were more likely to rate a routine favorably on Enjoyment (β^{^}= 1.363, odds ratio (OR) = 3.91, p = .012) and Likability (β^{^} = 1.183, OR = 3.26, p = .029). Alternatively, those who sat >10 hours/day on workdays were less likely to give a routine a favorable rating on Enjoyment (β^{^} = −1.460, OR = 0.23, p = .035) and Likability (β^{^} = −1.076, OR = 0.34, p = .048).

Benefit: Whether on workdays or non-workdays, sedentary behaviors had no significant effect on Benefit ratings of the physical activity sessions.

Confidence: Participants who sat >6 hours/day on non-workdays were less likely to rate a routine favorably on Confidence (β^{^} = −1.436, OR = 0.24, p = .015) (i.e., less likely to report being confident in performing the routine at work). Whereas sedentary behaviors on workdays did not have a significant effect on Confidence ratings.

Discussion

Summary and Interpretation of Findings

The hypothesis, that sedentary behavior affects ratings of physical activity sessions, was confirmed. We found significant differences among various levels of sedentary behavior. Those who sat greater hours on non-workdays rated physical activity sessions more favorably on Enjoyment and Likability. In contrast, those who sat more hours on workdays rated sessions less favorably on Enjoyment and Likability. For Benefit, there was no difference between workday and non-workday sitting. Non-workday rather than workday sedentary behavior was related to perceived confidence in performing the sessions. These results are novel and provide insight into future studies that may elucidate the underlying mechanisms of these associations.

Involuntary and voluntary sedentary behaviors are both characterized by prolonged periods of sitting or reclining. Both sedentary behaviors pose significant health risks. However, these behaviors differ in their origins, contexts, and mechanisms. Recognizing the distinctions between these types of sedentary behavior is essential to develop targeted interventions to mitigate the adverse health effects associated with sedentary behaviors (Taylor, 2022). Voluntary sedentary behavior is described as sitting or lying down by choice (i.e., self-chosen). In contrast, involuntary sedentary behavior is being sedentary without choice (i.e., due to external constraints or circumstances) (Eklund et al. 2021).

The literature on voluntary and involuntary sedentary behavior includes one taxonomy and one empirical study. The Taylor Taxonomy of Sedentary Behaviors includes four domains: (a) social interaction, (b) novelty, (c) choice, and (d) cognition (SNCC) (Taylor, 2022). High levels of the four domains in any sedentary behavior leads to more favorable cognitive and psychological outcomes than low levels of the four domains. The domain of choice is directly related to a voluntary versus involuntary decision (Taylor 2022). In an empirical study, participants reported that voluntary sedentary behavior was positively related to general health and well-being. Conversely, participants described involuntary sedentary time was negatively related to health (Eklund et al. 2021). Distinguishing between voluntary and involuntary sedentary behaviors is important to inform targeted health interventions. In our study, this distinction affects perception of physical activity.

One interpretation is that sitting on non-workdays may be more voluntary than sitting at work and thus an explanatory factor for the discrepant findings. In general, the weekend is consumed primarily of leisure activities and our choice to sit may impact our perceptions. In contrast, prolonged work-related sitting may result in subconscious resentment and resistance to change and may be the culprit behind the lower ratings on Enjoyment and Likeability. Perhaps, voluntary sitting in our leisure time (i.e., non-workdays) may result in a greater motivation to be active; thus, the physical activity sessions received higher ratings on Enjoyment and Likability. Therefore, when one is sedentary on non-workdays there is an urge or readiness to move on the next days (which are workdays). But forced sitting on workdays results in greater challenges to be physically active yielding less favorable ratings of physical activity sessions. Additional studies are needed to examine the relationship between weekend sitting and next day perceptions of physical activity across different social cognitive constructs.

A consequence of greater sitting patterns on non-workdays resulted in lower ratings on Confidence (i.e., self-efficacy). Perhaps, the voluntary nature of sedentary behavior on non-workdays undermines Confidence in participating in physical activity sessions at work even though the sessions are rated as Enjoyable and Likeable. This study indicates a need to broaden the scope of research that examines determinants of confidence and self-efficacy for physical activity. Clearly, more research is needed to disentangle and clarify why workday and non-workday sedentary behaviors yield different findings.

From a public health perspective, neither non-workday nor workday sitting patterns affected the dimension of Benefit. Perhaps, the benefits of workplace, physical activity sessions are not diminished or enhanced by sedentary behavior patterns which is promising in terms of promoting physical activity and advancing public health objectives. The lack of associations indicated that both groups are aware of the benefits of physical activity. Thus, knowing the benefits of physical activity alone is not sufficient to impact social cognitive constructs. We found that different levels of sedentary behavior affected ratings of physical activity sessions. However, Chevance et al. (2018) reported that implicit attitudes toward sedentary behavior were not associated with physical activity. These two studies differed in several ways. The purpose of the current study was to assess participants’ contemporary sedentary behaviors and concomitant ratings of physical activity sessions. Participants had a wide range of body mass indices and rated the physical activity sessions immediately after completing them. The purpose of the Chevance et al. (2018) study was to examine the specific contribution of implicit attitudes, toward both physical activity and sedentary behavior, to predict of physical activity in obese adults, and used self-report measures of implicit attitudes toward sedentary behavior to predict physical activity four months later. Another distinction between the two studies is that implicit attitudes are methodologically and conceptually different (Chevance et al., 2018) from explicit social cognitive factors as motivational constructs.

It is difficult to know definitively why individuals with greater levels of sedentary behavior perceive physical activity differently than those with lower levels of sedentary behavior. An actionable recommendation for workplace program designers is to evaluate sedentary behaviors of employees both in the workplace and during the weekends. This recommendation emphasizes the importance and awareness that different settings may be associated with different psychosocial perceptions and health consequences. Then workplace program designers can identify barriers and facilitators for participation in workplace physical activity programs by type and setting of sedentary behavior. This knowledge will enable practitioners to reinforce facilitators and minimize barriers to maximize participation in physical activity sessions.

A strategy to identify explanatory mechanisms is to conduct a series of systematic qualitative studies (e.g., focus groups, Nominal Group Technique, etc.) and then quantitative studies (or Mixed-Method Studies) to confirm the hypotheses derived from qualitative research. One can only speculate that individuals enjoy different types of physical activity sessions, contextual settings, and stimuli (e.g., music or not) to elicit favorable responses and participation. In this study, group based physical activity programs accommodate the needs of the group (i.e., the collective) and are not designed to serve a particular individual’s preferences and needs, per se. Therefore, the incentives, activities, context, and stimuli are derived from group consensus to maximize participation and enjoyment.

A comprehensive approach includes both individual and group interventions. Interventions designed to change sedentary behavior should consider encouraging self-regulatory skills such as repeated breaks in sedentary time as well as cues (e.g., reminders) to influence these breaks. Additional strategies such as self-monitoring, forming action plans, and recognizing and overcoming barriers are needed to develop new behaviors (Fleig et al., 2011).

Limitations and Strengths of the Study

The findings should be interpreted in light of study limitations. The most relevant limitations are the dimensions and self-report measures. The psychosocial measures were single items used to measure various constructs, which limits the ability to determine construct validity (i.e., internal consistency reliability). We however, note that the items selected were chosen based on comprehensive reviews that consistently indicated that they were key determinants of physical activity and sedentary behavior. In addition, our measures of physical activity and sedentary behavior were self-report, which were subject to recall and reporting biases. Furthermore, research has documented that, apart from total sitting time, the duration of sitting bouts and interruptions (i.e., breaks) in sitting time are important (Bellettiere et al., 2017; Dunstan et al., 2011; Healy et al., 2011; Reid et al., 2018). Breaks in sedentary time were not assessed in the MSTQ questionnaire. Future studies, whenever technologies permit, should utilize objective measures of sedentary time. Other limitations include having a volunteer sample (i.e., selection bias), and having participants rate workplace physical activities all in one session instead of multiple sessions. The participants were office- and desk-based workers; therefore, it is inappropriate to generalize the findings to other types of jobs (e.g., manufacturing, construction, manual labor, etc.).

The strengths include a study sample representing various age groups and different racial and ethnic identities. The order of physical activity sessions was varied to eliminate sequence effects; for consistency, standard warm-up and cool-down sessions were implemented. There are no known studies that developed and described six, distinct physical activity sessions for 15-minute work breaks and evaluated the sessions on pilot tested and documented dimensions. For this study, the most relevant and unique strengths was the inclusion of both work and non-workday sitting. This distinction is essential because office workers sitting at work may be expected and required to do so, whereas sitting on non-workdays is voluntary or more of a choice. Furthermore, the MSTQ covered a broad range of sitting behaviors from sitting while working, reading, or studying to sitting while talking, texting, or other socializing; the physical activity ratings were a composite of six different routines ranging from yoga to circuit training (Table 1). Therefore, the unique features of one routine were not the basis for the overall ratings. Importantly, there are no known studies that analyzed sedentary behaviors and corresponding effects on ratings of different physical activity routines.

Conclusion

The hypothesis that sedentary behavior affects the ratings of physical activity sessions was confirmed. Interventions designed to promote physical activity and/or reduce sedentary behavior should assess participants’ sedentary behavior patterns (both workday and non-workday) and recognize the influences on social cognitive constructs and readiness to be physically active. Given the limited research on how people with differing patterns of sedentary behavior perceive physical activity, our study contributes to the literature on how sedentary behaviors can influence psychosocial outcomes. Finally, our study highlights and underscores the importance of context-specific sedentary behaviors and implications for perceptions and health outcomes.