https://orcid.org/0000-0002-0715-8121
![Sample our Engineering & Technology journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5933/TOC-Subject-Sample-2021-Engineering-Tech.jpg"})
Abstract
Walking is a mode of transport that offers many benefits. This study aims to provide insights on the emotions associated with different types of walking episodes – namely, utilitarian walking episodes that are undertaken with the purpose of fulfilling an activity at a destination and recreational walking episodes that are undertaken with no specific purpose/destination. A knowledge of the emotions associated with different types of walking episodes can help steer policies and investments in a way that would engender greater levels of walking. The paper utilizes the well-being module of the American Time Use Survey (ATUS) data set to model subjective ratings on five different emotions associated with walking, employing a total of 1583 walk episodes. The five emotions include happiness, meaningfulness, tiredness, stress, and painfulness. A multivariate ordered probit model is estimated to account for unobserved attributes that may simultaneously affect multiple emotions. Model estimation results show that a number of socio-demographic variables and walking episode attributes affect how people feel about recreational and utilitarian walking episodes. Overall, it is found that utilitarian walking episodes offer a lower level of positive emotions than recreational episodes – presumably because recreational episodes are undertaken in more pleasant environments under more relaxing conditions. The results suggest that investments in recreational walking infrastructure and green spaces may yield richer dividends in terms of engendering higher levels of walking. Targeting specific socio-demographic groups for awareness campaigns and having strategic parking policies based on the day of the week are some of the other important implications.
Disclosure statement
No potential conflict of interest was reported by the authors.
Acknowledgement
The authors are grateful to Lisa Macias for her assistance in formatting the manuscript, and appreciate the comments of three anonymous reviewers on an earlier version of the paper.
Notes
1 In fact, it was a personal experience along these lines that led to the research in this paper. When a visiting professor in Hong Kong, the corresponding author found himself preferring to take a hotel shuttle that serviced an area near to his workplace and then walking about 25 minutes to work, rather than taking a more direct subway system that entailed little walking and a lower total travel time by about 15 minutes. But the two 25 minute walk episodes (to and fro from work) served, in the perception of the corresponding author, as the exercise episodes of the day and relieved the need to allocate separate time to pursue aerobic exercises at the beginning or end of the day.
2 The “sadness” variate was included in our initial modeling efforts; however, only three exogenous variables (“Age over 65 years,” “Individual self-reports health as good, very good or excellent,” and “Social interaction while walking”) turned out to be significant for this variate at 95% confidence level. Furthermore, all the three variables had the opposite signs compared to the “happiness” variate, which is quite intuitive. The overall unconditional Pearson correlation between the “sadness” and “happiness” ratings was in the order of −0.6, while the polychoric correlation between the two variates (that is, the correlation in the underlying latent variables for the two variables) was much higher in the order of −0.7. As indicated by Schmukle and Egloff (Citation2009), page 277, emotive state ratings tend to be positively skewed rather than being normal distributions, and multivariate distributions of emotive states are far from being multivariate normally distributed; however, these are the assumptions made in computing the Pearson correlations for the linear association between emotive states. On the other hand, Schmukle and Egloff indicate that the underlying latent variables (and the resulting polychoric correlation) provide a much better representation to assess linearity association. That is also the reason why the modeling framework used in the current paper is based on an underlying latent variable approach. The overall suggestive implication from the high negative polychoric correlation is that the “sadness” and “happiness” emotions are, relatively speaking, polar opposites and do not co-occur much at least in the context of the underlying ExWB propensities associated with walk episodes. The psychological literature does typically differentiate happiness and sadness, because the state of happiness today is relative to the state of sadness in the past. That is, deeper states of sadness in the past can engender a deeper state of happiness in the present or vice-versa, because happiness and sadness are each relative emotions and opponent process theory from physiology suggests that each of us has a baseline state of homeostasis and ups and down are followed by reverses to re-stabilize to that baseline state (see Solomon & Corbit, Citation1974; Solomon, Citation1980). But this positive association between happiness and sadness is likely to be more pertinent when investigating relationships among well-being emotions at different points in time in a longer-term longitudinal fashion than for momentary ExWB measures. Of course, in the context of momentary ExWB measures too, a high association and high co-occurrence of both happiness and sadness is not impossible, though is known to be relatively rare and is typically confined to emotionally complex and intense situations (such as the bitter-sweet moments associated with a graduation when a person may feel sad to leave the past but also be excited/happy about what a person has achieved and is about the explore in the future; or the death of a loved one who has had a good long life, when an individual feels sad to lose a beloved one but may also be happy that the loved one is no more in pain/distress; see Larsen, Citation2017). In most common (less emotionally complex) activities/events (such as walking episodes), the near mirror inverseness and low co-occurrence of the ExWB states of sadness and happiness is likely to hold (see, for example, Larsen et al., Citation2017; Russell & Carroll, Citation1999).
3 Note that the 54% of episodes (of the 1583 episodes) contributed by employed individuals is commensurate with the 54% of individuals (of the 1326 individuals) who are employed in the sample. That is, the frequency distribution of the number of episodes contributed by employed individuals and unemployed individuals are about the same. Thus, it is not the case in our sample that employed and unemployed individuals contribute very different numbers of episodes on a per individual basis (the ratio of the number of episodes to individuals is 1.208 among employed individuals and 1.177 among unemployed individuals). Also, important to note is that the percentage of individuals employed in our sample reasonably closely reflects the percentage of individuals in the US population who were employed in 2012. According to the Bureau of Labor Statistics (see https://www.bls.gov/opub/ted/2012/ted_20120911_data.htm), the labor force participation rate was 63-64% (say 63.5%) in 2012, with an unemployment rate of 8.2%. Thus, the employment rate may be derived as 63.5*(0.918) = 58%. Thus, our sample has a slightly lower percentage of employed individuals, but not dramatically less so than the US population percentage of employed individuals in 2012.