Patricia Manganelli 

Hardiness and Health


A thesis  submitted in partial fulfillment of the
requirements for the degree of Bachelor of Arts
Department of Psychology
Saint Anselm College
 Manchester, NH 03102

November 24, 1998

Why do certain types of people fall ill and have a difficult time overcoming sickness while others seem to be unaffected by illness and can buffer a stressful situation?


The purpose of this study is to investigate the relationship between the hardy personality and health.   Hardiness  was first described by Kobasa (1979) as a cognitive appraisal processes composed of three characteristics: challenge, commitment and control. Participants were 34 college students (28 females, 6 males) whose levels of hardiness were related to their general health.  Hardiness was measured by the PVSII and health by the SF-36.  Results indicate control was the only significant hardiness measure that was positively correlated with five health measures.  Intercorrelation analysis of the three hardiness components were significant for control and challenge and control and commitment.  This thesis illustrates that personality and health are indeed related and provides evidence that the hardy personality exists.  Research should be encouraged to study hardiness and health so that people may become aware of the impact that hardiness has on health.

   Questions or Comments please email me at

And now I present to you my thesis.........



This thesis would not have been a success without the help and understanding of many people.  I would like to thank all the Psychology professors for their guidance and attempt to lessen the strain that this thesis caused! I would especially like to thank professors Flannery and Finn for their guidance and patience throughout the whole Experimental Psych process- from choosing a topic to its total completion.  And a thanks to Barbara who had to deal with the Hardiness Institute yet again this year.
A special thanks to my friends who listened to my complaints and eased my anxiety from the very beginning.  I would like to thank my family who has been the strongest influence in my life.  They have been very understanding and supportive while I was constantly rambling about this thesis, even though they had no idea what I was talking about most of the time! Thanks again to everyone who listened, advised and helped out in ways that they are unaware of.  It's finally over!


Hardiness and Health

Previous research has led many psychologists to theorize why certain types of people fall ill and have a difficult time overcoming sickness while others seem to be unaffected by illness and can buffer a stressful situation (Florian, Mikulincer, & Taubman,1995; Hull, Van Treuren, Virnelli, 1987; Kobasa,1979a; Kobasa, 1979b;Kobasa, Maddi, & Kahn, 1982; Li-Ping Tang & Hammontree,1988).  Selye (1979) defined the concept of stress as the nonspecific response of the body to any demand made upon it.  Through experimental data, a direct link has been made between abnormal immune responses and stress.  In addition to the stress- disease connection, it seems personality and other mediators are the connection between stressful life events and illness or health.

Kobasa (1979) hypothesized that people who experience high degrees of stress without illness have a personality structure characterizing them differently from people who become sick under stress.  Kobasa coined the term "hardiness" to describe these people. Kobasa, Maddi and Kahn (1982) defined the concept of hardiness as "a constellation of personality characteristics that function as a resistance resource in the encounter with stressful life events" (p.169). Kobasa (1979a) defines hardiness as a cognitive appraisal processes composed of three characteristics: 1) control was described as a belief that people could control or influence the events of their experience; 2) commitment was described as an ability to feel deeply involved in the activities in their lives; and 3) challenge was described as anticipating change as an exciting challenge to further development. The implications of a hardy personality could profoundly affect the way health promotion and preventative measures are used in the health field.

Hardiness and Health
 According to Kobasa (1979a), hardy people are buffered against stressful life situations because they engage in certain affective, cognitive, and behavioral responses. In turn, buffering the stressors leads to better overall health.
  Kobasa (1979a) worked with executives working under conditions of stress. Control was measured through four instruments, commitment with the Alienation Test, and challenge with 6 instruments. Kobasa suggests that a hardy personality plays the role of a buffer in the stress-illness relationship. This study was one of the first to significantly correlate the role of personality and other mediators in the connection between stressful life events and illness or health.  In this study, Kobasa's results indicated that high stress/low illness executives can be distinguished from high stress/ high illness subjects.
 Since Kobasa's 1979a study, many other researchers have investigated the health and hardiness connection (Florian, 1995; Hull, 1987; Kobasa, 1979b; Kobasa et al. 1981; Kobasa et al. 1982; Greene & Nowack, 1988).  Greene & Nowack (1988) studied coping styles in relation to hardiness and health and supported the notion that a positive association exists between stress, coping and health.  Specifically, hardiness was correlated with four coping styles.  In accordance with this finding, Li-Ping Tang & Hammontree (1988) determined hardiness was significantly associated with future strain and stress. Results ultimately showed that "Hardiness will operate as a resistance resource in the stress and strain relationship and also the stress and illness relationship." (21).

Stress and Illness
Hans Selye (1956, 1976) first described the connection between stress and disease.  The sequence of physical responses to stress that occurs in a consistent pattern and is triggered by the effort to adapt to any stressor is called general adaptation syndrome (GAS).  Stress and stressors are the initiators of physiological responses that begin in the brain and spread to the many organs throughout the body.  For example, the pituitary gland triggers the release of corticosteroids which help resist stress but also tends to suppress the immune system making the body more susceptible to infection.  Thus, Selye notes stress plays some role in the development of every disease.  Yet when coping is effective, it can either ameliorate or nullify the stressor so that the risk of illness is decreased (Pollock, 1984).
According to Selye (1979), biochemical homeostasis in the body depends on two types of reactions, synotoxic and catatoxic, that can take place in order to resist different toxic stressors.  Syntoxis stimuli would pacify and allow the aggressors to live in a peaceful coexistence with the body, while catoxic agents destroy pathogens.  An example of how our own actions can induce disease is the following.  A homeless drunk insults you as you walk by him.  A syntoxic response would be to just walk away, while a catatoxic response would increase your blood pressure and pulse, and alarm your nervous system in preparation for a fight.  The latter action could result in a fatal brain hemorrhage or a heart attack.  Selye calls this action "biological suicide" and states that "death is caused by choosing the wrong reaction."  We do not always recognize what is and what is not fighting and this can ultimately lead to illness. Selye referred to diseases that are caused by or promoted by stressors as diseases of adaption.
Research by Pollock (1984) has shown evidence that personality, as well as stress and coping, is related to various psychosomatic complaints and diseases.  "Current research suggests a positive relationship between life stresses and illness." (4).

With past research investigating stress and illness, some research has also been done on the stress and hardiness model (Collins,1992; Florian, 1995; Li-ping Tang & Hammontree, 1988). These researchers are basically in agreement that hardiness plays an integral part in buffering stress which may ultimately result in better health.  The participants in the Collins paper (1992) were adolescents.  She found that hardiness demonstrated statistically significant negative correlations with stressful life events, aberrant behavior, strain, school absence and drug use which supports the hypothesis that hardiness is a buffer against these stressors.  Collins concludes, "Conceptualization of hardiness as a stress resistant resource in adolescents was given empirical support." (7).
 Real life stress events and their relationship to hardiness was also investigated by Florian et al. (1995).  Military recruits were evaluated during their four month combat training period.  Evidence was found to support hardiness as positively contributing to health by means of coping and appraisal techniques.
 Li-Ping Tang and Hammontree (1988) utilized a professional police officer population, who are immersed in stressful situations daily.  Results showed that hardy police officers have less future and concurrent strain, in addition to less future and concurrent illness.  The three characteristics of hardiness, even under highly stressful situations, may promote health by decreasing the incidence of disease (Pollock, 1984).

Measuring Hardiness
 A controversy exists in the literature about the best way to test hardiness and whether it does indeed affect health or illness by it's three factors of control, commitment and challenge (Benishek, 1996;  Greene & Nowack, 1995; Hull et al., 1987).
 Hull et al. (1987) concluded that the variable of commitment is being more precisely measured than either control or challenge and that the imprecision of the control measure used is due to the mixing of items from different scales.  A more recent study done by Benishek (1996) indicated that criticisms of hardiness have been voiced because of the variety of measures used to asses hardiness and the factor structure underlying hardiness. Benishek (1996) addressed the issue of how hardiness is to be measured and operationally defined. Benisheck proposed that the large variety of hardiness measures makes it extremely difficult to interpret research findings. Upon questioning the psychometric properties of two hardiness measures, the Revised Hardiness Scale and the Personal Views Survey (PVS), it was found that the correlation was slightly lower than would be expected. The intercorrelation between two challenge factors was extremely low and was independent of the commitment and control measures. Benishek found that PVS was the more reliable of the two measures. Studies originated from college-age populations tend to find more than one factor underlying hardiness but this area of research has not yet been explored.
 Hull et al. (1987) found that commitment was positively correlated with control whereas challenge was negatively correlated with commitment and control. Due to this invalidity, Hull et al. recommend that research proceed to investigate the independent contributions of commitment and control to health outcomes and eliminate the sub scale of challenge altogether. Greene and Nowack (1995) found that there was no significant correlations between the original Kobasa hardiness scales and predictions of self-reported hospitalization over three years.
 Kobasa, Maddi and Courington (1981) and Kobasa et al. (1982) both used a prospective design to validate that hardiness predicts both simultaneous and future health. These studies reported analyses using the same set of data.  In their study, they used the same subject pool of middle and upper level management that they used in the 1979 study.  They measured stressful life events with the Schedule of Life Events and symptomatology was measured with the Seriousness of Illness Survey.  Hardiness was measured by the Alienation Test and California Life Goals Evaluation Schedule. Results indicated that stressful life events and illness were positively correlated as they both increased over years.  A main effect of hardiness suggests that it functions prospectively as a resistance resource.
 Hull et al. (1987) found the practice of using the same data set to draw conclusions to be flawed and states that the samples cannot be treated as independent.  All correlations were found to be highly significant, suggesting that the presence of an overall style of hardiness exists. Commitment and control had high mean intercorrelations whereas the results for challenge were more complicated. Hull et al. (1987) state that his correlation of challenge was unacceptably low and that Kobasa's versions of challenge lacks validity. Control and commitment are more accepted to be adequate measures than is challenge. Hull suggested that commitment should be measured by the short form developed by Kobasa but that control should not be measured by her scales.
Instead, Hull et al. claim The Rotter Locus of Control Scale is a reasonable measure. Correlational analysis revealed that perceptions of personal control and perceptions of personal responsibility were both significantly correlated with health. Specifically, that a stronger sense of control and stronger sense of responsibility were related to a higher self-rating of health (Ziff 1995).
Control is one of the three characteristics of hardiness and seems to be the one that is most valid, in comparison to commitment and especially to challenge.  Cohen and Edwards (1989), in the Flowers article (1994), concluded that perceived control is the best documented mechanism for buffering the stress-disorder relationship in general. In their study which examined perceived control, illness status, stress and adjustment to cardiac illness, it is stated that the interactive effect between level of life stress and perception of control was significantly correlated with psychological distress.
There is much evidence to suggest that hardiness is not a unitary trait.  Hull et al. (1987) reports that researchers must understand the link between personality and health and examine such relationships systematically. Benishek (1996) agrees that hardiness continues to appeal to the health field and that evidence found could improve the validity and utility of future research on hardiness. Greene and Nowack (1995) conclude that hardiness as a mediator of health is provocative, yet inconclusive. Most current research suggests that a hardy cognitive outlook and explanatory style (optimism, internal locus of control and self appraisal) may be directly related to health functioning.


 The participants of this experiment were college students.  A total of 34 participants (28 female, 6 male) from Saint Anselm College engaged in this study. Sixteen of the participants were freshmen enrolled in the general psychology classes and the rest were of various majors and grades who volunteered to participate.  All participants signed an informed consent form (see Appendix A) before begining.

Materials and Procedure
 Testing the participants began with verbal directions.  Two questionnaires were administered in this study, the PVSII and the SF- 36.
  SF- 36: The first questionnaire that was administered was the SF-36 which is a general health survey.  It assesses the overall health of an individual.  The SF-36 includes a multi-item scale that evaluates eight health concepts: Physical Funtioning, Role Physical, Bodily Pain, General Health, Vitality, Social Functioning, Role Emotion and Mental Health.     Content validity of the SF-36 compared to other commonly used generic health surveys has been found to include eight of the most widely represented health concepts (Ware, Keller, Gandek, Brazier & Sullivan, 1995).  This general health inventory also has been investigated in predictive validity studies which have linked SF-36 scales and summary measures to utilization of health care services (Ware & Gandek, 1994).
 The SF-36 is composed of 11 sections with a total of 36 questions.  The breakdown of questions to health concepts is as follows: numbers 1 and 11a-11d  measure general health; 3a-3j score physical functioning; 4a-4d determine role physical; 5a-5c score role emotion; numbers 6 and 10 measure social functioning; 7 and 8 are for bodily pain; 9a, 9c, 9g, 9i measure vitality; and lastly 9b-9d, 9f, and 9h gauge mental health.  Question 2 is not used to score any of the eight health scales.  Physical Functioning, Role-Physical and Bodily Pain correlate most highly with the Physical Component Summary measure (PCS).  The Mental Component Summary measure correlates most with the scales of Mental Health, Role Emotion, an Social Functioning.  This test took approximately 10 minutes to complete.
  PVSII:  The second questionnaire that was given was the Personal Views Survey II (PVS II). The PVSII is a 50-item scale created by Maddi (1987) that measures the degree to which a person feels they are hardy, in respect to the interaction between themselves and the world.  This instrument scores for 3 factors: commitment, challenge and control.  Each of these factors has a sub scale that is scaled from 0 (completely disagree) to 3 (completely agree).  Sample statements from the PVSII include "I often wake up eager to take up my life where it left off the day before" and "most of my life gets spent doing things that are worthwhile".
 Maddi (1994) has given the PVSII sufficient internal reliability (.70 to .75 for commitment, .61 to .84 for control, .60 to .71 for challenge and .80 to .88 for total hardiness) and validity (.68 for commitment, .73 for control, .71 for challenge and .77 for total hardiness). This test also took about 10 minutes to complete.
 Upon completion of both questionnaires, participants were given a debriefing form (see Appendix A) and course credit slips to those in the general psychology classes.


 The research question examined the relationship between hardiness and health.  A correlation matrix was developed to investigate the intercorrelation of the three measures of hardiness (see Table 1).  Findings suggested there is a strong intercorrelation between control and challenge (r=.329, p<.05) and control and commitment (r= .477, p<.01).  In addition, total hardiness was significantly correlated with all three measures at an alpha level of .01, challenge (r= .739), commitment (r=7.61) and control (r= .701).  No significant correlation was found for neither challenge and control nor challenge and commitment.

Table 1
Intercorrelation Between Three Hardiness Measures
                          Challenge     Commitment        Control        Hardiness Total
Commitment         .241
Control                 .329*            .477**
Hardiness              .739**          .761**              .701**

* Correlation is significant at the 0.05 level (2-tailed)
** Correlation is significant at the 0.01 level (2-tailed)

 Another correlation matrix was calculated in order to identify the intercorrelation between the eight health measures that were used (see Table 2).  Bodily pain was positively correlated at an alpha level of .01 with physical functioning (r= .339) and role physical (r= .412).  General health was correlated with mental health (r= .689), social functioning (r= .566) and vitality (r=.502) at an alpha level of .01 and also with role emotion (r= .392, p<.05).  Mental health was correlated with role emotion (r= .620), social functioning (r= .707) and vitality (r= .760) at p<.01 and also with role physical (r= .389, p<.05).  Physical functioning was not found significant with any of the eight measures.  Role emotion was correlated with social functioning (r= .706) and vitality (r= .618) at an alpha level of .01.  Role physical (r= .484) and social functioning (r= .619) were both found significant with vitality at an alpha level of .01.

Table 2
Intercorrelation Between Eight Health Measures
                         BP             GH             MH          PF              RE            RP        SF      VT
GH .155
MH .096         .689**
PF .339*         .328          .205
RE .243          .392*         .620**          .215
RP .412*        .277          .389*             -.071     .261
SF .306          .566**       .707**           .347*     .706**        187
VT .293         .502**       .760**          .351*     .618**           .484**      .619**

* Correlation is significant at the 0.05 level (2-tailed)
** Correlation is significant at the 0.01 level (2-tailed)

 Next, a Pearson correlation between general health and hardiness was used.  The analysis indicated that of the three hardiness measures, control was the only statistically significant one.  In accord with most norms researched, control has been singled out as a hardiness measure that may most accurately determine health (Hull et al., 1987; Rosolack & Hampton, 1991; Ziff, 1995). Control was significantly correlated at an alpha level of .05 with general health (r= .384), role emotion (r= .418), social functioning (r= .401) and vitality (r= .342).  Control was also found significant with mental health (r= .524, p<.001).
 In order to better understand these findings, a stepwise linear regression analysis was calculated for control being the dependent variable with the five independent health measures which are general health, role emotion, social functioning, vitality and mental health. The regression showed an R Square of .0.524 and an R Square variance of 0.274 for the variable of mental health.  No other health measures made an impact on the effect of control.  This would suggest that the other four health measures were intercorrelated enough not to have their own individual influence over the dependent variable of control.
 The health measure designated mental component summary was significantly and positively related to the commitment measure of hardiness (r= .352, p<.05), control measure of hardiness (r= .483, p<.01) and total hardiness (r= .417, p<.05).  No other statistical significance was found between measures of hardiness and measures of health.
  No significance was found in respect to hardiness and health when analyzing the data set by year.  That is, there was no significant difference in hardiness and health scores between the two larger groups, freshmen and seniors.


 The present study has investigated the relationship between hardiness and health.  Students were asked to complete the PVSII and the SF-36 to assess hardiness and general health.  The research hypothesis proposed higher hardiness scores would positively correlate with higher general health scores.  The current study has demonstrated that a relationship exists between hardiness and health.
 Although not all three components of hardiness were significant, control did positively correlate with five of the eight health measures.  This analysis confirms the literature of Hull et al. (1987), Ziff (1995) and Flowers (1994) who suggest control is the dominant component of hardiness which may buffer stress.
 Findings also showed the intercorrelations of the measures of both hardiness and general health.  The intercorrelation of hardiness suggests that control and commitment are correlated but that challenge is not.  In accordance with critical research by Hull et al. (1987), it could be determined that hardiness is not a unidimensional measure but that control and commitment are being more precisely measured than challenge.  Total hardiness was correlated with all three dimensions which would suggest that total hardiness does indeed measure all three components.
 The intercorrelation of the eight health measures seem to suggest that this generic measure of health may not be testing for each individual concept.  Although it has been studied for its validity (Ware et al. 1995), the results implied that the SF-36 was not an accurate predictor of individual health measures in this study.
 The relationship between hardiness and health in this study has not been as conclusive as previous research.  This may be for a variety of reasons.  All participants were students which is a homogeneous population.  Also, a small sample size was used whereas a larger selection may have shown greater results.
 The PVSII is still considered a controversial tool in investigating hardiness. Debate still exists about whether hardiness is a unitary trait or not.  The health survey that was utilized in this study was found to have a high intercorrelation which suggests that each health measure may not be uniquely identified.
 Future research into the area of hardiness and health may take into consideration the impact stress may have on a participant.  This cross-sectional study may be further developed by doing a test-re test of both measures before and after a stressful event (e.g. midterms).  This would present the opportunity to collect more data and analyze it to show specific differences in health.  Other recommendations would be to use a consistent sample size for each grade level.
 In summary, most literature is optimistic about the ability of hardiness to act as a buffer against stress and therefore lead to better health (Collins, 1993; Florian et al., 1995; Hull et al., 1987; Kobasa, 1979a; Kobasa, 1979b; Kobasa et al., 1981; Kobasa et al., 1982; Kobasa et al., 1985; Li-Ping Tang & Hammontree, 1988).  The results presented here continue to provide evidence that hardiness exists and that it is a substantial factor in health.  Along with stress and other factors, personality provides an important clue for how to live a health life.  With further research and more accurate instruments, the relationship between personality and health may, in the words of Suzanne Kobasa, "...illuminate ways of developing the personality characteristics that can aid in a productive and healthy life led in full complexity of modern, urbanized, industrialized societies." (10).


 Benisheck, L. (1996). Evaluation of the factor   structure underlying two measures of hardiness.    Assessment, 3 (4), 423-435.

 Collins, C. (1993, Aug).  Hardiness as a stress   resistance resource.  Paper presented at the  annual meeting of the American Psychological Association, Toronto, Ontario, Canada.

 Florian, V., Mikulincer, M. & Taubman, O. (1995).  Does  hardiness contribute to mental health during a   stressful real-life situation? The roles of appraisal  and coping.  Journal of Personality and  Social Psychology, 68 (4), 687-695.

 Flowers, B. (1994). Perceived control, illness status, stress, and adjustment to cardiac illness.   Journal of Psychology Interdisciplinary & Applied,  128 (5), 567.

 Gala, C., Musicco, F., Durbano, F. & Cesara, B. (1995).  Italian validation of the multidimensional scale  of "Health Locus of Control". New Trends in    Experimental and Clinical Psychiatry, 11 (2), 79-86.

 Greene, R., Nowack, K. (1995). Hassles, hardiness   and absenteeism: results of a 3-year longitudinal study. Work & Stress, 9 (4), 448-462.

 Hull, J., Van Treuren, R., Virnelli, S. (1987).   Hardiness and Health: A critique and alternative    approach. Journal of Personality and Social Psychology, 53 (3), 518-530.

  Kobasa, S. (1979a). Stressful life events, personality, and health: An inquiry into hardiness.   Journal of Personality and Social Psychology, 37 (1), 1-11.

 Kobasa, S. (1979b). Personality and resistance to   illness. American Journal of Community Psychology, 7 (4), 413-423.

 Kobasa, S., Maddi, S. & Courington, S. (1981).   Personality and constitution as mediators in the  stress-illness relationship. Journal of Health and   Social Behavior, 22, 368-378.

 Kobasa, S., Maddi, S. & Kahn, S. (1982). Hardiness and   health: A prospective study. Journal of Personality and Social Psychology, 42 (1), 168-177.

 Kobasa, S., Maddi, S. & Puccetti, M. (1982).    Personality and exercise as buffers in the stress-illness relationship. Journal of Behavioral  Medicine, 5 (4), 391-404.

 Kobasa, S., Maddi, S., Puccetti, M. & Zola, M. (1985).  Effectiveness of hardiness, exercise, and social support as a resource against illness. Journal of Psychometric Research, 29 (5), 525-533.

 Li-Ping Tang, T. & Hammontree, M. (1988, April).  The effects of hardiness, job related stress and life   stress on health and absence from work.  Paper    presented at the annual meetin of the Southwestern   Psychological Association, Tulsa, OK.

 Rosolack, T. & Hampson, S. (1991). A new typology of health behaviors of personality-health predictions: The case of locus of control.  European Journal of Personality, 5, 151-168.

  Selye, H. Cancer, Stress and Death. New York: Plenum Publishing Co., 1979.

 Ware, J. and Gandek, B. (1994).  The SF-36 Health   Survey: development and use in mental health  research and the IQOLA Project.  International Journal of Mental Health, 23 (2) , 49-73.

  Ware J., Keller S., Gandek B., Brazier J. and Sullivan M., (1995). Evaluating translations of health   status questionnaires: methods from the IQOLA Project.  International Journal of Technology Assessing  Health Care, 11(3), 525-51.

  Ziff, M. (1995). The relative effects of perceived personal control and responsibility on health and   health-related behaviors in young and middle-aged
adults. Health Education Quarterly, 22 (1), 127.



All psychological research at Saint Anselm College is conducted according to strict ethical principles outlined by the American Psychological Association and is in full compliance with Federal law. The Department of Health and Human Services, for example, specifies that informed consent must be given prior to research studies, that is, "the knowing consent of an individual or his legally authorized representative so situated as to be able to exercise free power of choice without undue inducement or any element of force, fraud, deceit, duress, or other form of constraint or coercion".

Simply put, this means when you participate in any research study, you will be given a clear explanation of the procedures involved. You may ask for clarification either before or during the procedure, and you may terminate the procedures at any time.

After having carefully read and considered the foregoing, I consent to participate in research activities according to the terms heretofore enumerated.

Date__________________________ Signature_______________________

Class/Student I.D.#____________________  Other_____________________



 Thank you for taking the time to participate in my study.  The purpose of this experiment is to examine the relationship between personality and health.  A question that intrigued me is why do some people fall ill when others seem to buffer sickness.  The SF-36 is a general health survey while the Personal Views Survey II is a hardiness survey.  Hardiness was coined by Suzanne Kobassa in 1987.  Hardiness is a personality trait characterized by having a sense of commitment, challenge and control over life situations.  The hypothesis of this study is that people who demonstrate a higher level of hardiness will have lower instances of illness and better overall health.  I am testing all four grade levels (freshmen, sophomores, juniors and seniors) to see if there is any relationship there as well.
 Again, thank you for taking part in this study and if you have any questions, please feel free to contact me (ex. 6250, box #1253).

    Patricia Manganelli

Thank you for visiting my homepage. I hope it has been informative and helpful to those embarking on thier own thesis!

Good Luck!