Studying Stress Research Designs and Epistemologies

Everyone has experienced stress. This can range from minor problems, such as getting stuck in traffic, to more significant events, such as the death of a loved one. Understanding the impacts of stress is not as straightforward as it might appear. The topic of stress and how it impacts daily living is complex. Psychologists have attempted to understand stress through a variety of theories and different perspectives. This paper serves to highlight a few of the frameworks and important pieces in stress research. First, will be an examinaiton of principal problems associated with measuring stressors and outcomes. Second, will be an analysis of the relationships between stressors and mental health in children and adolescents. Third, will be a focus on how early life experiences may influence our relationships with stress. Last, is an examination of the attention to personality qualities and their role in stress reactivity.

What should we know about measuring stress?

Despite the fact that stress is one of the most researched variables in psychology, defining, measuring, and understanding stress is not as straightforward as it may appear. Grant et al. (2004), experts in stress research, conducted a review of the scientific literature in this area. Reviews in psychology are helpful because they systematically examine how studies in a particular area have been conducted, what we know from these studies, but also what remains unknown about the given topic.

Psychologists have conceptualized stressors/stress in a few different ways, and these ways of thinking have important implications for how researchers conduct their studies and what results they obtain. Grant et al. (2004) rely on a specific definition of stress: stress as an event or the possibility of an event outside the individual that has the potential to threaten an individual’s psychological or physical well-being and use this to derive their definition of stressors. This is sometimes referred to as the “stimulus-response” model. In the stimulus-response model of stressors, researchers often try to understand how stressors typically affect people. The stimulus-response model is different from another widely cited model of stress, the transactional model (Lazarus, 1990). In general, Lazarus (1990) theorizes that stress is not merely a byproduct of some type of adversity but is instead intrinsically related to a) how the person thinks about the stressor and b) what resources the individual believes they have at their disposal to cope with it. In other words, Lazarus (1990) argues that stressors do not cause stress responses; rather, it is how a person appraises the stressor and their ability to deal with the stressor that explains how it impacts them. Advocates of the stimulus-response model criticize the assumptions in the transactional model for a few reasons.

First, researchers using the stimulus-response approach argue that secondary processes, such as how one thinks about stressors, should be studied separately from the stressor-response relationship. These ways of thinking are reflected in the measures in Figure 1. The first measure represents one of the first attempts by researchers to study stress: The Social Readjustment Rating Scale (SSRS; Holmes and Rahe, 1967). The SSRS is a “checklist,” and participants are asked to rate how many times they experienced each event over the past twelve months. Figure 2 depicts the Perceived Stress Scale (PSS; Cohen, Kamarck, & Mermelstein, 1983). The PSS is associated with the transactional model of stress. Items 1 and 3 of the PSS measure two psychological processes: 1) how “upset” or “nervous” the respondent has felt over the past month and 2) if these reactions were due to an event. Researchers using a measure like the PSS cannot determine how much of the impact of the stressor is due to the event itself and how much is due to a person’s perceptions of a stressor. Researchers often use the term confounding variable when one or more variables are not controlled for and are measured in a study. Thus, measuring two variables at once (stressors/stress responses) represents a confound.

Researchers might choose the stimulus-response model as a theoretical basis for their study for a few reasons. First, the stimulus-response helps an understanding how stressors, in general, affect people. For instance, a person might ask, “How does the average person respond to this stressor?” Additionally, some stressors, such as natural disasters, are expected to be stressful regardless of how they are perceived.

Problems with stress management

Before looking at how stressors affect mental health in children and adolescents, Grant et al. (2004) spend a considerable amount of time in their review focused on how stressors are measured by psychologists. Grant et al. (2004) explain that one of the common forms of measurement of stressors is what is called a “checklist” (remember, the SSRS, described earlier, is a checklist). Checklists give participants events that are both negative and positive since many positive stressors may also impact people (Grant et al., 2004). Sometimes, researchers use very broad checklists, like the SSRS, to assess the most common types of stressors that respondents may have experienced (Grant et al., 2004). Other researchers use specialized checklists, including checklists based on very specific types of stressors (i.e., school-based stressors) or based on the needs of a specific population (i.e., urban youth). Although stressor checklists are widely used, they are far from perfect. First, the items listed on the checklists are limited because they are generated by researchers or small groups of respondents (Grant et al., 2004). Second, the way participants interpret and respond to the item varies quite a lot (Grant et al., 2004). For instance, a checklist item might be “death of a close family member.” Some participants might endorse this item if they lost someone very close and important to them, whereas others may endorse it if a distant relative with whom they had little contact died. Because of the nature of the checklists, it is difficult to know how great of an impact the event had on the respondent. Checklists also fail to assess the degree to which a respondent caused or brought about a particular event (Grant et al., 2004). For instance, a respondent may endorse an item indicating that they had problems at school, but the problems at school may have been because the individual failed to turn in assignments or argued with a teacher (Grant et al., 2004). On the other hand, a respondent might report experiencing the same stressor but because of a learning disability or a bad teacher. With a checklist, it is nearly impossible to know how of a role the person played in experiencing the stressor.

To combat this problem, some researchers advocate for the use of stressor interviews (Grant et al., 2004). Sometimes, these interviews will start with a list of stressors and allow the interviewers to probe (e.g., follow-up with more in-depth questions) about the specific stressors, such as how long ago the stressor occurred and what the implications were of experiencing the stressor. Grant et al. (2004) report that stressor interviews tend to pick up on events that are not captured by checklists. Additionally, interviews tend to be better at predicting which stressors will be associated with psychological symptoms, such as depression (Grant et al., 2004). However, stressor interviews are a) infrequently used, b) difficult and expensive to implement, c) limit reporting of certain stressors, particularly those that are embarrassing or difficult to talk about (Grant et al., 2004).

Overall, Grant and colleagues (2004) found that the measurement tools and methodologies used in studies on the relationship between stressors and psychological disorders in children and adolescents have improved, but stressor measurement for racial/ethnic minority youth is more limited. Taken as a whole, measuring stressors is more challenging than it may appear at first. A consideration is how stress is defined and measured in each study.

Does experiencing stressors impact the mental health of children and adolescents?

Following their evaluation of the measures and methods used in stressor studies, Grant et al. (2004) tested whether stressors increased children and adolescents’ symptoms of “psychopathology.” The term “psychopathology” is often used in place of other words you may be more familiar with, like “mental illness” or “psychological disorders.” Grant et al. (2004) found that stressors increased children/adolescents’ psychological symptoms. More specifically, the authors found that in 53 studies, higher levels of stressors were associated with heightened psychological disorder symptoms. Experiencing stressors predict both “internalizing” (i.e., keeping feelings contained within, such as feeling worried or sad) and “externalizing” (i.e., acting out symptoms, like breaking rules or starting fights) symptoms in youth (Grant et al., 2004). In many ways, these findings are consistent with popular definitions of stress. These state that experiencing more demands from the environment makes people feel worse internally and also impacts their behavior.

Less obvious is the relationship between stress and mental health and how mental health symptoms impact the amount and type of stressors we experience. This question of whether the children and adolescents’ psychopathology symptoms predicted stressors was another important one addressed by Grant et al. (2004). In other words, did youths’ psychological symptoms increase stressful experiences? Grant et al. (2004) found that psychological symptoms did predict higher levels of stressors over time. Grant et al. (2004) refer to this as a “reciprocal relationship.” This means that each of the variables (in this meta-analysis, the variables are stressors and psychological symptoms) are influencing each other: stressors are associated with more symptoms of psychopathology over time, but psychopathology is associated with higher levels of stressors over time. The researchers suggest that the psychopathology symptoms may influence stressors like strained personal relationships, poor school performance, and so forth.

These results suggest that stress is a meaningful issue for understanding the mental health of children and adolescents. However, a person should not just consider the relationship between stressors and outcomes as purely linear; stressors impact us, and, in turn, we create stressors. There is far more to understand about stressors.

Early life experiences

Animal studies provide another way to measure, study, and understand stress. Animals serve as useful subjects for a few reasons. First, animals can be used for experimental manipulations that are impossible, impractical or unethical in humans. Second, because of the shorter lifespans of animals, it is easier to scientifically study short-term and long-term implications.

One such animal study with important results for our purposes is a review study by Meaney (2001). More specifically, Meaney (2001) explained what we know about how early life experiences in rat pups influence their adult stress reactivity. Generally speaking, stress reactivity refers to the extent that an individual responds to a stressor, whether it be cognitively, physically, behaviorally, or affectively. In other words, to what degree does an individual respond to a stressor?

First, Meaney (2001) described studies that have examined how handling (e.g., picking up the rats, holding them in their hands, and so forth) by researchers impacts rat pups. In this case, it is inferred that being removed from one’s mother and being handled by researchers is a form of a stressor. Interestingly, numerous studies show that in adulthood, rat pups that were handled daily but briefly (researchers remove rat pups from their cages and mothers for 3–12 minutes) show less stress reactivity, fearfulness, and physiological responses to stress (Meaney, 2001) as adults. Based on this finding, one might presume that exposure to some levels of stress “inoculates” rat pups to future stress. These results should be interpreted with caution, however. In natural environments, mothers leave rat pups for periods of 20–30 minutes (Meaney, 2001). To address whether the handling is a stressor or whether it mimics a naturally occurring process, researchers exposed rat pups to more extended periods of separation (i.e., 180 minutes) from their mothers. Results showed that pups exposed to more extended maternal separation demonstrated greater physiological reactivity to stress and were also more fearful in new situations as adults (Meaney, 2001). In short, early life exposure to handling by researchers, at least in brief but regular intervals, is associated with lower stress responses in rats. However, extended separation from mothers produced more stress reactivity.

Researchers have also found important effects for how “warm” versus “cold” mothers influence their pups’ stress reactivity (Meaney, 2001). Researchers observed rat mothers and identified “warm” and “cold” mothers based on licking behaviors and how the mothers nursed their pups. Results indicated that the pups of “warm” mothers were less physiologically reactive to stress, more open to exploring new environments, and less easily startled than pups raised by “cold” mothers (Meaney, 2001). Researchers wondered if “warm” mothers passed on genes that were responsible for these results. To address this, the researchers switched offspring; the biological rat pups from “cold” mothers were raised by “warm” mothers and vice versa (Meaney, 2001). Interestingly, the results showed that the pups’ adult behavior most closely resembled the mother that raised them. In other words, the pups raised by “warm” mothers, even those genetically related to the “cold” mothers, were less stress-reactive and more open to exploration.

This article provides an entirely different lens to understand the relationship between stressors and responses. Not only were the participants not humans, but the methodologies were experimental (i.e., the researcher manipulated something in the study, rather than measuring naturally occurring variables and looking for associations). The results, overall, suggest that early life experiences may influence how we experience stress later in life.

How can personality impact our relationship with stress?

A question for inquiry is whether people have personalities that affect how they react to stressors. To understand this and other related questions, Chida and Hamer (2008) performed a meta-analysis on 729 studies. A meta-analysis is a statistical technique used by researchers to synthesize a large number of studies on the same topic. The benefit of a meta-analysis is that it provides much greater statistical information about a topic than any one study alone.

Chida and Hamer (2008) included very specific types of studies in their meta-analysis. In order for an article to be included, the study had to meet these criteria: a) measured stressors people were experiencing in their everyday lives, b) involved an experimental manipulation in which the researchers created stressors in controlled ways, like asking participants to do difficult math problems out loud, and c) measured how this impacted participants’ cardiovascular reactivity (defined as the difference between a person’s resting heart rate and his/her heart rate after experiencing a stressor). Chida and Hamer (2008) were interested in cardiovascular reactivity because it has been shown to be associated with greater risk for certain types of heart issues, such as high blood pressure and cardiovascular diseases (Chida & Steptoe, 2010). Chida and Hamer's (2008) studies also examined the length of time it takes a participant’s heart rate to return to the baseline heart rate from the post-stressor heart rate, referred to as cardiovascular recovery. To help you better understand the types of studies Chida and Hamer (2008) included, see Figure 3.

Across the studies, Chida and Hamer’s (2008) results showed that individuals higher in certain characteristics showed more cardiovascular reactivity. One of these personality traits was hostility (Chida & Hamer, 2008; see Figure 4). Hostility is defined as a person’s tendency to have thoughts or attitudes that are angry, aggressive, suspicious, and generally negative (Chida & Steptoe, 2009). Anger is defined as an emotional state ranging from annoyance to fury, often with behaviors such as verbal outbursts or physical aggression (Chida & Steptoe, 2009). Type A behavior is defined as a person’s tendency to be angry and hostile, but also driven, impatient, and ambitious (Chida & Steptoe, 2009). An important finding in Chida and Hamer’s (2008) review is that persons higher in hostility, anger, and Type A behavior had significantly higher cardiovascular reactivity from experiencing an acute stressor. In other words, participants who displayed more of these qualities had heart rates that changed a great deal from their heart rates before the stressor (i.e., cardiovascular reactivity). From these findings, we can conclude that some of our characteristics, like hostility, anger, and Type A tendencies, are associated with greater stress reactivity.

Interestingly, Chida and Hamer (2008) found that some surprising factors were associated with lower cardiovascular reactivity, including anxiety, negative affect (low mood), and neuroticism (Chida & Hamer, 2008; see Figure 5). Neuroticism is a trait in the “Big Five” theory of personality and is often defined as one’s tendency to become anxious, worried, depressed, or self-conscious (McCrae & Oliver, 1992). However, participants higher in neuroticism, anxiety, and negative affect had longer cardiovascular recovery rates (amount of time it took for participant’s heart rate to return to baseline following the post-stressor increase. In other words, people higher in neuroticism, anxiety, and negative affect had heart rates that changed less pre-stressor (baseline) to post-stressor than other peoples’ hearts. Even though their heart rates increased less than other participants, it took these same participants more time to return to their baseline heart rates (cardiovascular recovery rates).

Chida and Hamer (2008) also examined whether experiencing different types of stressors affected stress reactivity in laboratory settings. Results showed that participants experiencing more general life stressors also had significantly greater cardiovascular recovery times (Chida & Hamer, 2008).

Conclusions

Human lives are filled with stress, and people often discuss stress and its implications in their everyday conversations. However, using a scientific lens to study and understand stress and its impacts is complex for many reasons. Psychologists are not in consensus about what stress is (events versus appraisal) and how to measure it. Also, youth experiencing stressors are likely to have increased psychopathology symptoms; in turn, these psychopathology stressors may increase stressors. For example, one such stressor faced by many young people in the US is poverty.

Early life experiences may put people at increased risk of responding more physiologically to stressors, although these relationships are more tenuous, given that animal models were used as participants. A meta-analysis by Segerstrom and Miller (2004) synthesizes studies on stress and our immune systems. Last, personality qualities, like anger and hostility, influence how reactive people are to stressors. Other qualities, such as neuroticism and having general life stressors, may make it more difficult to recover mentally and physically after experiencing an acute stressor.

About the Author

Andrea M. Flynn, PhD, Associate Professor of Psychology at Concordia University Chicago; Dr. Flynn studies the relationship between stress and mental health in students. She is particularly interested in students from underserved backgrounds.

Bibliography

Chida, Y. et al. 2008). Chronic psychosocial factors and acute physiological responses to laboratory-induced stress in healthy populations: A quantitative review of 30 years of investigations. Psychological Bulletin, 134, 829–855.

Chida, Y. et al. (2009). The association of anger and hostility with future Coronary Heart Disease. Journal of the American College of Cardiology, 53, 936–946.

Chida, Y. et al. (2010). Greater cardiovascular responses to laboratory mental stress are associated with poor subsequent cardiovascular risk status: A meta-analysis of prospective evidence. Hypertension, 55, 1026–1032. doi.org/10.1161/HYPERTENSIONAHA.109.146621

Cohen, S. et al (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24, 386–396.

Dohrenwend, B. et al (1985). “Hassles” in the conceptualization and measurements of life stress variables. American Psychologist, 40, 780–785.

Grant, K et al. (2004). Stressors and child and adolescent psychopathology: Measurement issues and prospective effects. Journal of Clinical Child and Adolescent Psychology, 33, 412–425.

Holmes, T. H. & Rahe, R. H. (1967). The Social Readjustment Rating Scale. Journal of Psychosomatic Research, 11, 213–218.

Lazarus, R. S. (1990). Theory-based stress measurement. Psychological Inquiry, 1, 3–13.

McCrae, R. et al. (1992). An introduction to the five-factor model and its applications. Journal of Personality, 60, 175–215. doi.org/10.1111/j.1467-6494.1992.tb00970.x

Meaney, M. (2001). Maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations. Annual Review of Neuroscience, 24, 1161–1192.

Segerstrom, S. et al. (2004). Psychological stress and the human immune system: A meta-analytic study of 30 years of inquiry. Psychological Bulletin, 130, 601–630.

Simmons, J. (2023, December 20). NIA’s evolving toolbox to study the science of stress. National Institute on Aging. Retrieved May 31, 2024, from www.nia.nih.gov/research/blog/2023/12/nias-evolving-toolbox-study-science-stress.

(2024). The American Institute of Stress. Stress Research. Retrieved May 31, 2024 from www.stress.org/stress-research.

Yoshikawa, H. et al. (2012). The effects of poverty on the mental, emotional, and behavioral health of children and youth. American Psychologist, 67, 272–284.