Health Education Research, Vol. 16, No. 5, 579-592,
October 2001
© 2001 Oxford University Press
The Revised Health Hardiness Inventory (RHHI-24): psychometric properties and relationship with self-reported health and health behavior in two Dutch samples
Department of Clinical and Health Psychology, Leiden University, PO Box 9555, 2300 RB Leiden, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Our objective was to examine the psychometric properties of a Dutch translation of the Health Hardiness Inventory (HHI), and to determine relationships between health hardiness scales and self-reported health and preventive health behavior. Data from a cross-sectional study with two samples [a general population sample (n = 205) and a student sample (n = 286)] were analyzed. The Revised Health Hardiness Inventory (RHHI-24) was found to consist of four stable and reliable scales: (1) Health Value, (2) Internal Health Locus of Control, (3) External Health Locus of Control and (4) Perceived Health Competence. Women valued their health more than men, older individuals (>45 years) valued their health more than younger individuals (
45 years) and elderly people (>65 years) were more externally orientated with respect to health locus of control. Preventive health behavior was related to a higher value placed on health, a lower external health locus of control and a higher perceived health competence. Better self-reported health was related to a higher perceived health competence and a more internally orientated health locus of control. We conclude that the RHHI-24 is a theoretically sound instrument for the measurement of health cognitions. |
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
In 1979 Kobasa introduced the concept of hardiness. This personality characteristic, derived from existential psychology, expresses a general quality of an individual to regard stressful life events as amenable, and to consider changes as a normal and interesting part of life. It is assumed that this positive orientation towards life helps a person to stay healthy under stressful circumstances (Kobasa et al., 1982a
). According to Kobasa, the concept consists of three inter-related dimensions: (1) commitment, a sense of involvement in life-activities, (2) challenge, a tendency to regard potentially stressful events as interesting opportunities for growth, and (3) control, a perception of influence over one's life.
In the years following the introduction of this concept, the relationship between hardiness and health has been investigated in many different studies [e.g. (Kobasa et al., 1982a,b, 1983; Funk and Houston, 1987; Roth et al., 1989; Kobasa, 1993)]. Results from these studies consistently indicate that hardiness is associated with better self-reported health and with fewer physical complaints. However, the assumed stress-buffering effect of hardiness has not been supported (Hull et al., 1987; Funk, 1992).
One of the pathways through which hardiness is expected to improve health is the practice of healthy life styles (Funk, 1992). Individuals with a hardy personality are assumed to practice more health behaviors than those who do not possess this characteristic. Empirical results seem to, at least partially, support this notion.
For example, Pollock studied 244 adults, including patients who suffered from a chronic illness, and found a significant relationship (r = 0.23) between the presence of hardiness and participation in health promotion activities (Pollock, 1989). Similarly, Nicholas investigated the relationship between hardiness, self-care practices and perceived health status in 72 older adults (Nicholas, 1993). Again, the results indicated that hardiness is significantly correlated (r = 0.68) with the practice of health-care activities, such as exercise, healthy nutrition, relaxation and general health promotion. In addition, this study showed that persons with a high score on both hardiness and health-care practices reported a better health status than individuals with low scores on these two variables. Finally, results of a study by Nagy and Nix with 211 college students yielded that hardiness is significantly, although not highly (r = 0.16), correlated to (un)healthy lifestyles, such as physical activity, drug use, smoking status, stress and eating habits (Nagy and Nix, 1989). The only prospective study on this topic was conducted by Wiebe and McCallum and included 86 undergraduate students (Wiebe and McCallum, 1986). Again, the authors found that hardy individuals, as compared to non-hardy participants, were far more likely to be involved in health behaviors, such as dietary and hygiene practices, non-use of drugs, and exercise, even under stressful circumstances.
Therefore, a generally positive association between hardiness and general health practices has been reported. However, studies, which have focussed on the relationship between hardiness and exercise behavior as an isolated health behavior have produced less positive results. In his review, Funk identified several studies (Kobasa et al., 1982b; Nagy and Nix, 1989; Roth et al., 1989) which suggested that exercise and hardiness have independent effects on health, and exercise has no mediating effect in the hardiness–health relationship (Funk, 1992).
Since the introduction of the hardiness construct numerous measurement scales have been designed and applied. Initially, different items of existing scales, such as the Alienation Test (Maddi et al., 1979), the Internal–External Locus of Control Scale (Rotter et al., 1962), the Personality Research Form (Jackson, 1974) and the California Life Goals Evaluation Schedules (Hahn, 1966), were combined in the so-called Unabridged Hardiness Scale (Ouellette, 1993) to measure the three dimensions of hardiness. On the basis of research with this instrument a refined 36-item questionnaire was constructed (Kobasa et al., unpublished; Jennings and Staggers, 1994). The items were negatively formulated and as such the instrument measured the absence of hardiness rather than the presence of it. This has led to substantial conceptual problems (Funk and Houston, 1987). For example, hardiness measured in such a manner may be confounded with maladjustment or neuroticism, while an acquiescent response style may distort scores in a way that respondents appear less hardy (Ouellette, 1993). In later instruments, such as the Personal Views Survey (Hardiness Institute, 1985) and the Dispositional Resilience Scale (Bartone et al., 1989), the number of positively and negatively formulated items are more equal.
According to Pollock, the predictive validity of the concept could be greatly improved by measuring hardiness at a domain-specific level (Pollock, 1986, 1989). This implies that instruments that assess hardiness should consist of items that are particularly relevant to the specific field of research. Thus, when studying health-related issues, control could be defined as the sense of mastery needed to appraise health stressors appropriately, challenge as the ability to reappraise a health stressor as potentially beneficial and commitment as the motivation to effectively cope with the threat of a health stressor. An example of such a questionnaire, developed by Pollock and Duffy (Pollock and Duffy, 1990), is the Health-Related Hardiness Scale (HRHS). However, the authors found in a study with 389 patients, who suffered from diabetes, multiple sclerosis, hypertensive or rheumatoid arthritis, that the items for commitment and challenge loaded together on one factor. They propose that the commitment and challenge scales may not be conceptually distinct.
The most recent instrument to measure health hardiness is the Health Hardiness Inventory (HHI), constructed by Wallston and Abraham (Wallston and Abraham, unpublished communication) as an alternative to Pollock's HRHS. The HHI was developed to improve the measure of health control by adding items which reflect the concept of perceived health competence. While locus of control is concerned with the source of control (internal or external) only, perceived health competence refers to an individual's generalized expectancy regarding the ability to interact effectively with the environment in order to obtain the desired outcome, i.e. being healthy (Wallston, 1992; Smith et al., 1995). As such, perceived health competence is a composite of a behavioral expectancy (being capable of performing the behavior) and an outcome expectancy (the behavior will lead to good health). In addition, Wallston and Abraham formulated new items to improve the face-validity of the health commitment and health challenge scales. The HHI has been previously investigated in a sample of 116 university staff members (Abraham, 1992). Factor analyses yielded results which were fairly consistent with the proposed underlying structure of the questionnaire. The reliability coefficients were adequate for all three subscales (ranging from 
= 0.61 for Challenge to = 0.82 for Control). It was concluded from the study that the three components are relatively distinct constructs that differ in their relationship with various health indicators (e.g. health behavior) and personality measures (e.g. general hardiness as measured with the Personal Views Survey). Furthermore, one of the subscales of the HHI, perceived health competence (as part of the control dimension), has been examined in more detail by Smith et al. with college students, adults and patients with rheumatoid arthritis (Smith et al., 1995). It appeared that chronic patients reported significantly lower levels of health competence than healthy participants, and that perceived health competence was correlated with health status, positive state of well-being and mental adjustment in the patient group, as well as in the healthy population (r = 0.44–0.55).
In the present study, the HHI was translated into Dutch and administered to a sample from the general population and a student sample. The factor structure and reliability of the Dutch translation were examined. Age and gender differences on health hardiness were investigated. Furthermore, construct validity of the questionnaire was determined by investigating the relationships between health hardiness, preventive health behavior and health.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Participants
Data were collected in two samples. The first sample consisted of 205 members of the general Dutch population and the second sample included 286 university students.
General population sample
The general population sample involved 205 individuals of the general Dutch population. The procedure was designed to acquire equal numbers of female and male respondents across age groups ranging from young adults (<26 years) to elderly (>65 years). The sample included 99 males and 106 females. The average age was 42.3 years (SD = 19.8), ranging from 13 to 87 years.
Student sample
The second sample consisted of 286 Dutch university students. The sample included 107 males and 179 females. The age of the participants ranged from 18 to 57 years with an average age of 25.2 years (SD = 7.5). Seventy-nine percent of the sample was under 30 years of age.
Questionnaire
Health hardiness
A Dutch translation of Wallston's HHI was used to measure the health hardiness characteristic. A back-translation procedure was followed, and did not reveal any main differences between the English and the Dutch version of the questionnaire. The HHI is a self-report questionnaire consisting of 35 items (see Appendix). Each item of the questionnaire is a belief statement, referring to one's position regarding one's own health. Respondents can indicate on a five-point scale the degree to which they agree with the statement, ranging from ‘strongly disagree’ to ‘strongly agree’. The questionnaire encompasses three scales: health commitment (10 items), health challenge (seven items) and health control (18 items). A fourth scale, perceived health competence, is formed by an eight-item subset of the scale for health control.
|
|
Self-reported preventive health behavior
To assess health habits, respondents were presented with questions based on the ‘Alameda seven’ (Belloc and Breslow, 1972
; Breslow and Enstrom, 1980). In the Alameda Study, seven health behaviors were related to better physical health status and lower mortality. These behaviors are: (1) eating three regular meals each day without snacking between meals (‘Do you snack between meals?’), (2) having breakfast every day (‘Do you have breakfast every day of the week?’), (3) exercising moderately, twice or more a week (‘Do you exercise at least twice a week during 30 or more minutes?’), (4) sleeping 7–8 h each night (‘Do you sleep 7–8 h a night?’), (5) no smoking (‘Do you smoke?’), (6) drinking alcohol in moderation, if at all (‘Do you drink less than two drinks of alcohol per day?’), and (7) maintaining normal weight (‘What is your weight?’ and ‘What is your height?’). The first six questions could be answered by yes or no. Normal weight was assessed by means of the body mass index (BMI). Respondents were asked to give their weight and height; a BMI [weight (kg)/height (cm2)] within the range 20–25 was considered to reflect a healthy weight. The total score for health behavior was assessed by a summation of these seven health behaviors.
Self-reported health
Self-reported health was measured by one (five-point scaled) item, referring to how healthy the respondent generally felt.
Procedure
General population sample
The aim of the procedure was to acquire equal numbers of female and male respondents across various age groups ranging from young adults (<26 years) to (retired) elderly (>65 years). In several large Dutch cities, a number of streets were selected at random. In order to involve equal numbers of respondents in each gender and age group, households were approached with the question whether it included a person in a specific gender and age group (e.g. female older than 64 years). This individual was then asked to participate in the study. Participants received a questionnaire and accompanying instructions. Questionnaires could be completed anonymously and were collected from their homes within 3 days. All of the questionnaires were returned.
Student sample
Students were recruited during lectures and by poster adverts, and were asked to complete the questionnaire on a voluntary basis. The completed questionnaires were gathered in a collection box at the university. All of the questionnaires which were handed out, were returned.
Statistical analysis
The factor structure of the HHI was determined on the data from the general population sample, by means of principal component analyses with varimax rotation. On the basis of internal consistency (Chronbach's ) and inter-correlations between scales decisions were made regarding the composition of the scales. For both the general population sample and the student sample, scores on the scales were computed by adding the item scores and dividing the sum by the number of items in the scale. As a result, scores on all scales ranged from 1 to 5. To determine age and gender differences, scores were compared using Student's t-tests and analyses of variance (post hoc Scheffé test). The relationships between the health hardiness scales, preventive health behavior and self-reported health were assessed by means of Pearson correlations and multiple hierarchical regression analyses. In the regression analyses variables were entered in the following order: (1) age and gender, (2) the four subscales separately, (3) the products of all possible combinations of two subscales (two-way interactions), (4) the products of all possible combinations of three subscales (three-way interactions), and (5) the products of the four subscales (four-way interaction).
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Reliability and inter-correlations of the original HHI scales
Table I
gives the reliability coefficients of the original HHI scales (Wallston and Abraham, unpublished communication). They ranged between 0.64 and 0.77 for the general population sample, and 0.66 and 0.79 for the student sample. Worrysome, however, were the high inter-correlations between the scales for Health Commitment and Health Challenge, and the scale Health Control and its subscale Perceived Health Competence. These inter-correlations suggested that these factors are not conceptually distinct. Therefore, the factor structure of the HHI was re-examined.
|
Factor structure and psychometric properties of the RHHI-24
An explanatory free factor analysis on the data from the general population sample was carried out including the 35 items of the HHI. Inspection of the scree plot, applying the elbow criterion, indicated a four-factor solution. In a forced four-factor analysis, these four factors together explained 38.5% of the variance (initial eigenvalues of, respectively, 6.0, 3.3, 2.4 and 1.7). The factors could be labeled as ‘Health Value’ (HV), ‘Internal Health Locus of Control’ (IHLOC), ‘External Health Locus of Control’ (EHLOC) and ‘Perceived Health Competence’ (PHC). Scales were constructed on the basis of: (1) factor loadings (factor loadings >0.30), (2) internal reliability (i.e. deleting items until an optimal Cronbach's
occurred) and (3) content of the items (see Appendix for the final scales). As a result, it was decided to exclude 11 items from further analyses. Scores on the scales were computed by adding the item scores and dividing the sum by the number of items in the scale.
Table II shows the reliability and inter-correlations of the four scales of the Revised Health Hardiness Inventory (RHHI-24) in both samples. In the general population sample, the reliability of the scales was adequate, with 's ranging from.66 to.79. The highest inter-correlations were found for PHC and EHLOC (r = –0.39), and PHC and IHLOC (r = 0.40).
|
The reliability coefficients of the RHHI-24 scales in the student sample were comparable to those in the general population sample with one exception; the scale EHLOC had a lower internal consistency. Although the patterns of inter-correlations were comparable to the general population sample, the scales of the RHHI-24 were more strongly inter-correlated in the more homogeneous student sample. Particularly, the scale for HV was more strongly related to the scales for EHLOC and PHC.
The RHHI-24 scales: age and gender differences
A comparison of the female and male respondents within the general population sample indicated that the two groups differed significantly on the scale for HV (see Table III). The female respondents had a higher mean score on this scale than the males, indicating that they were more involved with issues related to their health.
|
The same difference was found between the female and male respondents in the student sample. Furthermore, male students had a higher EHLOC than the female students.
Due to the restricted age range in the student sample, age differences were only examined in the general population sample. A comparison of four age groups with respect to scores on the RHHI-24 scales indicated two significant age differences in this sample (see Table IV). Firstly, the older respondents (>45 years) valued their health more than the younger respondents (45 years). Secondly, the elderly respondents (>65 years) had a higher EHLOC than all other participants. No significant age differences were observed on the IHLOC and PHC scale.
|
The RHHI-24 scales and their relation to self-reported health behavior and health
To assess the construct validity of the RHHI-24 scales, the relationships between the RHHI-24 score, on the one hand, and self-reported health and preventive health behavior, on the other hand, were examined for both samples. As Table V
shows, the number of preventive health behaviors was significantly related to the extent to which the respondents: (1) valued their health, (2) felt competent in dealing with issues concerning their health and (3) were less externally oriented. The score on IHLOC was unrelated to preventive health behavior.
|
The RHHI-24 scales appeared to be differently related to self-reported health. The respondents with better self-reported health felt more competent in dealing with their health and had a stronger IHLOC. In both samples self-reported health was, however, unrelated to EHLOC. Moreover, in the general population, but not in the student population, self-reported health was unrelated to HV.
Multiple hierarchical regression analyses confirmed the relation between the RHHI-24 scales and preventive health behavior (see Table VI). In the general population sample HV and PHC were significant predictors of preventive health behavior. In the student sample HV significantly predicted health behavior, whereas for PHC a tendency (P < 0.10) was found. Controlling for age and gender, in the two samples the RHHI-24 scores explained, respectively, an additional 22.3 and 19.8% of the variance in preventive health behavior. Higher scores on HV and PHC were related to the participation in more preventive health behaviors.
|
In the general population sample an additional 23.2% of the variance in self-reported health was explained by IHLOC and PHC, i.e. higher scores on IHLOC and PHC were related to a better self-reported health status. In the student sample the results were somewhat different, as HV and PHC were significant predictors of self-reported health status, and for IHLOC only a tendency (P < 0.10) was found (
R2 = 28.5%). Explorative analyses on all two-way, three-way and four-way interactive effects between the four concepts yielded no significant results for either preventive health behavior or self-reported health. |
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Factor structure and psychometric properties of the RHHI-24
The original scales for Health Commitment and Health Challenge were strongly correlated, in line with findings of Pollock and Duffy (Pollock and Duffy, 1990
). Also, the scale Health Control and its subscale Perceived Health Competence appeared to be highly inter-correlated. This indicates that these scales do not concern discrete dimensions. Therefore, analyses were performed to re-examine the factor structure of the HHI with the data of both research samples. Factor analyses on data gathered with the HHI in the general population yielded a stable underlying structure of four factors. The factors could be interpreted as follows: (1) Health Value (six items), (2) Internal Health Locus of Control (five items), (3) External Health Locus of Control (seven items) and (4) Perceived Health Competence (six items) (see Appendix). The scales designed on the basis of these results were found to have an acceptable internal consistency. Contrary to the original scales, the scales were not strongly correlated with each other, although the correlations between the scale for perceived health competence, on the one hand, and the scales for internal and external health locus of control, on the other hand, were somewhat higher. These results support the notion that perceived health competence is an extension of the theoretical concept of locus of control, while at the same time remaining substantially distinct. However, prudence is called for when interpreting results with regard to the unique contributions of these scales.
The health value scale was composed of two of the original seven items measuring ‘health challenge’, three of the original 10 items measuring ‘health commitment’ and one item of the ‘perceived health competence’ subscale of the original scale measuring ‘health control’. All of these items reflect a concern with health issues and a determination to do everything possible to maintain or improve one's health. The internal health locus of control scale consisted of five items of the original 18 ‘health control’ items. These items refer to a general notion that one is capable of influencing one's own health. External health locus of control was assessed by four items of the original ‘health commitment’ scale and three items of the original ‘health control’ scale. The four items originally designed to measure ‘health commitment’ all refer to beliefs that personal efforts to improve health are useless. Similarly, the three ‘health control’ items involve the belief that health outcomes are uncontrollable. Finally, six of the original eight items for ‘perceived health competence’ were incorporated in the new scale for perceived health competence. It should be noted that the 24 items included in these scales consisted of an almost equal number of positively and negatively formulated beliefs (13 versus 11).
Thus, only two original ‘health challenge’ items remained in the scales. Either the factor loadings of the removed ‘health challenge’ items were low or the items did not contribute to the reliability of a scale. This is in agreement with results from other studies, which consistently found lower coefficients for ‘health challenge’ scales [e.g. (Hull et al., 1987; Funk, 1992)]. Three ‘health commitment’ items were concerned with efforts to improve one's health and, therefore, loaded on the health value factor. Four other original ‘health commitment’ items involved a lack of personal responsibility for one's own health and loaded on the factor for external health control. The items originally designed to measure ‘health control’ could be divided into items assessing external health locus of control or internal health locus of control. This is consistent with the literature on general locus of control, in which the uni-polar nature of the concept has been rejected in recent years. It has been argued that internal and external locus of control should be considered as two independent, albeit related, dimensions, and that even within these two dimensions other subdimensions can be distinguished. For example, the widely used Multi-dimensional Locus of Control Scale, developed by Levenson (Levenson, 1972, 1974), differentiates between an internal locus of control, a powerful others locus of control (i.e. the belief that others in power determine outcomes) and a chance locus of control (i.e. the belief that outcomes cannot be controlled by humans, but are determined by chance alone). The HHI does not include items referring to powerful others and the external health locus of control scale focuses more on a chance related orientation towards health.
The RHHI-24 scales: age and gender differences
In the student population the males had higher scores on the External Health Locus of Control than the females. This has been observed previously [e.g. (Galanos et al., 1994)]. Furthermore, the females from both populations appeared to value their health more than the males, which is in agreement with findings from previous research of, for example, Lau et al. (Lau et al., 1986). They suggest that because females are typically caretakers of the family, or preparing to perform this role later in life, they are naturally more inclined to be concerned with health matters. Other researchers [e.g. (Verbrugge, 1985)] have proposed that women are also more sensitive to body discomforts that men, due to socialization processes in childhood. By being more aware of potential or realistic health problems (e.g. menstrual-related problems), women are conceivably more interested in health matters and value health more than men.
In the present study adults aged 46 and older appeared to value their health more than younger individuals. Furthermore, in agreement with Lachman's study, the locus of control of adults aged 66 and older is more externally orientated than that of all other participants in our study (Lachman, 1986). An explanation for these results may be that as people age they take their health less for granted. Older people may place a higher value on health than younger people, as they start to experience the consequences of aging, either personally or vicariously through people in the near social environment. In addition, they may be confronted with more situations in which they have virtually no control over their healt and consequently their health locus of control may have shifted to a more external orientation. Several authors have suggested that an external (health) locus of control may be highly adaptive, particularly when the outcomes of a situation are uncontrollable and/or unpredictable. For example, it has been found that for elderly people who moved to a residential home for the elderly, a higher external locus of control is related to better adjustment to the new situation (Felton and Kahana, 1974; Cicirelli, 1987). Rothbaum et al. even argue that an external locus of control often represents a form of secondary control, by bringing oneself in line with environmental forces (Rothbaum et al., 1982). For example, by identifying with ‘powerful others’ (e.g. physicians), one could regain a sense of control over an otherwise uncontrollable situation.
The RHHI-24 scales and their relation to self-reported health behavior and health
As mentioned in the introduction, earlier research shows a significant relationship between hardiness scales and health behavior (Wiebe and McCallum, 1986; Nagy and Nix, 1989; Nicolas, 1993; Pollock, 1989). In the present study three of the four RHHI-24 scales, i.e. Health Value, Perceived Health Competence and External Health Locus of Control, were also related to health behavior. In the multiple hierarchical regression analyses health value and perceived health competence emerged as independent predictors of health behavior, when controlling for age and gender. It should be noted that the two samples were comparable with respect to these results. The findings are consistent with Wallston's theory, who states that in order to be motivated to perform a particular behavior individuals need to value the expected outcomes as well as feel that they have control over them (Wallston, 1992). Likewise, Norman (Norman, 1995) and Smith et al. (Smith et al., 1995) argue that behavior-specific efficacy beliefs play an important role in the performance of health behaviors.
The previously reported direct effect of hardiness on self-reported health (Funk and Houston, 1987; Hull et al., 1987; Roth et al., 1989) was replicated in this study. Specifically, perceived health competence and internal health locus of control were significantly related to self-reported health.
Naturally, health behaviors require personal effort and investment of time. A personal commitment to performing these behaviors over time is therefore only likely to occur when the individual values his or her health highly and feels competent to perform the behavior. On the other hand, the personal experience of being healthy or not is likely to be related to beliefs concerning control and competence. Healthy individuals, who feel that their health is largely accounted for by their own actions, may change this internal orientation to some extent when confronted with illness. The experience of having an illness may indeed lead to the perception that health outcomes are not under such complete control as thought when being in a healthy state.
Methodological issues
Two samples were included in this study. The general population sample consisted of an almost equal amount of male and female participants. The sample was composed of individuals from various age groups, although the number of participants older than 65 was somewhat smaller than those under 65 years of age. Furthermore, all participants lived in Dutch cities. Generalization of the results to other specific groups may be unwarranted. Most importantly, the samples did not include specific patient groups, while it is likely that those who are confronted with serious health problems differ from healthy individuals with respect to the health beliefs they hold, and to the relationships between these beliefs and health behaviors and health status.
Health behavior was measured by assessing the performance of seven different health behaviors, which have been found to be predictive of health status and mortality in the Alameda study (Belloc and Breslow, 1972; Breslow and Enstrom, 1980). Comparison of the results with respect to health behavior with findings from previous studies in this field is hampered, as all studies vary considerably with respect to the operationalization of health behavior.
In our study the outcome variables were assessed by self-report measures. Furthermore, self-reported health was determined by one item only. Previous research with a single item measure of self-rated health has shown that this is a valid measure to assess health status, even within different ethnic groups [e.g. (Chandola and Jenkinson, 1997)]. However, for future research it is recommended that multiple measures are used, including more objective measures, such as observation of behavior and medical information, in order to enhance the validity of the study. Also, by applying more objective measures the possible effect of a response bias, e.g. due to negative affectivity, could be largely reduced. Finally, in order to examine causal relationships between the health hardiness scales and health outcomes, such as health behaviors and health status, prospective studies with the RHHI-24 should be carried out.
In conclusion, this study with the RHHI-24 indicates that the questionnaire consists of four underlying scales: Health Value, Internal Health Locus of Control, External Health Locus of Control and Perceived Health Competence, all of which have adequate psychometric properties. Several age and gender differences emerged on these scales, which were consistent with earlier research outcomes. Relationships between the four scales and two outcome variables, Self-reported Health and Preventive Health Behavior, were also in line with previous findings. Therefore, the RHHI-24 is a reliable and useful instrument to assess important dimensions of individual variation in one's perspective towards one's own health.
|
Acknowledgments |
|---|
The authors are indebted to Professor K. A. Wallston for his valuable remarks and to K. Joekes for reviewing the English.
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Abraham, A. (1992) Construct validation of the Health Hardiness Inventory. Dissertation. Vanderbilt University, Nashville, TN.
Bartone, P. T., Ursano, R., Wright, K. and Ingraham, L. (1989) The impact of military air disaster on the health of assistance workers. Journal of Nervous and Mental Disease, 177, 317–328.[ISI][Medline]
Belloc, N. B. and Breslow, L. (1972) Relationship of physical health status and health practices. Preventive Medicine, 1, 409–421.[Medline]
Breslow, L. and Enstrom, J. E. (1980) Persistence of health habits and their relationship to mortality. Preventive Medicine, 9, 469–483.[ISI][Medline]
Chandola, T. and Jenkinson, C. (2000) Validating self-rated health in different ethnic groups. Ethnicity and Health, 5, 151–159.[ISI][Medline]
Cicirelli, V. G. (1987) Locus of control and patient role adjustment of the elderly in acute-care hospitals. Psychology and Ageing, 2, 138–143.
Felton, B. and Kahana, E. (1974) Adjustment and situationally-bound locus of control among institutionalised aged. Journal of Gerontology, 29, 295–301.[ISI][Medline]
Funk, S. C. (1992) Hardiness: a review of theory and research. Health Psychology, 11, 335–345.[ISI][Medline]
Funk, S. C. and Houston, B. K. (1987) A critical analysis of the hardiness scale's validity and utility. Journal of Personality and Social Psychology, 53, 572–578.
Galanos, A. N., Strauss, R. P. and Pieper, C. F. (1994) Socio-demographic correlates of health beliefs among black and white community dwelling elderly individuals. International Journal of Ageing and Human Development, 38, 339–350.
Hahn, M. E. (1966) California Life Goals Evaluation Schedule. Western Psychological Services, Palo Alto, CA.
Hardiness Institute (1985) Personal Views Survey. Hardiness Institute, Arlington Heights, IL.
Hull, J. G., Van Treuren, R. R. and Virnelli, S. (1987) Hardiness and health: a critique and alternative approach. Journal of Personality and Social Psychology, 53, 518–530.[ISI][Medline]
Jackson, D. N. (1974) Personality Research Form Manual. Research Psychologist Press, Goshen, NY.
Jennings, B. M. and Staggers, N. (1994) A critical analysis of hardiness. Nursing Research, 43, 274–281.[ISI][Medline]
Kobasa, S. C. (1979) Stressful life events, personality and health: an inquiry into hardiness. Journal of Personality and Social Psychology, 37, 1–11.[ISI][Medline]
Kobasa, S. C. (1993) Hardiness and health: a prospective study. Journal of Personality and Social Psychology, 65, 207.
Kobasa, S. C., Maddi, S. R. and Kahn, S. (1982a) Hardiness and health: a prospective study. Journal of Personality and Social Psychology, 42, 168–177.[ISI][Medline]
Kobasa, S. C., Maddi, S. R. and Pucetti, M. C. (1982b) Personality and exercise as buffers in the stress–illness relationship. Journal of Behavioral Medicine, 5, 391–404.[ISI][Medline]
Kobasa, S. C., Maddi, S. R. and Zola, M. A. (1983) Type A and hardiness. Journal of Behavioral Medicine, 6, 41–51.[ISI][Medline]
Lachman, M. E. (1986) Locus of control in ageing research: a case for multi-dimensional and domain-specific assessment. Psychology and Ageing, 1, 34–40.
Lau, R. R., Hartman, K. A. and Ware, J. E. (1986) Health as a value: methodological and theoretical considerations. Health Psychology, 5, 25–43.[ISI][Medline]
Levenson, H. (1972) Distinctions within the concept of internal-external control: development of a new scale. In Proceedings of the 80th Annual Convention of the American Psychological Association, pp. 261–262.
Levenson, H. (1974) Activism and powerful others: distinctions within the concept of internal-external control. Journal of Personality Assessment, 38, 377–383.[ISI]
Maddi, S. R., Kobasa, S. C. and Hoover, M. (1979) An alienation test. Journal of Humanistic Psychology, 19, 73–76.[ISI]
Nagy, S. and Nix, C. L. (1989) Relations between preventive health behavior and hardiness. Psychological Reports, 65, 339–345.[ISI][Medline]
Nicholas, P. K. (1993) Hardiness, self-care practices and perceived health status in older adults. Journal of Advanced Nursing, 18, 1085–1094.[ISI][Medline]
Norman, P. (1995) Health locus of control and health behavior: an investigation into the role of health value and behavior-specific efficacy beliefs. Personality and Individual Differences, 18, 213–218.
Ouellette, S. C. (1993) Inquiries into hardiness. In Goldberger, L. and Breznitz, S. (eds), Handbook of Stress: Theoretical and Clinical Aspects, 2nd edn. Free Press, New York, pp. 77–100.
Pollock, S. E. (1986) Human responses to chronic illness: physiologic and psychosocial adaptation. Nursing Research, 35, 90–95.[ISI][Medline]
Pollock, S. E. (1989) The hardiness characteristic: a motivating factor in adaptation. Advances in Nursing Science, 11(2), 53–62.
Pollock, S. E. and Duffy, M. E. (1990) The Health-Related Hardiness Scale: development and psychometric analysis. Nursing Research, 39, 218–222.[ISI][Medline]
Roth, D. L., Wiebe, D. J., Fillingim, R. B. and Shay, K. A. (1989) Life events, fitness, hardiness and health: a simultaneous analysis of proposed stress-resistance effects. Journal of Personality and Social Psychology, 57, 747–768.
Rothbaum, R., Weisz, J. R. and Snyder, S. S. (1982) Changing the world and changing the self: a two process model of perceived control. Journal of Personality and Social Psychology, 42, 5–37.
Rotter, J. B., Seeman, M. and Liverant, S. (1962) Internal vs. external locus of control of reinforcement: aA major variable in behavior theory. In Washburne, N. F. (ed.), Decisions, Values, and Groups. Pergamon, London, pp. 473–516.
Smith, M. S., Wallston, K. A. and Smith C. A. (1995) The development and validation of the Perceived Health Competence Scale. Health Education Research, 10, 51–64.
Verbrugge, L. M. (1985) Gender and health: an update on hypotheses and evidence. Journal of Health and Social Behavior, 26, 156–182.[ISI][Medline]
Wallston, K. A. (1992) Hocus-pocus, the focus isn't strictly on locus: Rotter's social learning theory modified for health. Cognitive Therapy and Research, 16, 183–199.
Wiebe, D. J. and McCallum, D. M. (1986) Health practices and hardiness as mediators in the stress–illness relationship. Health Psychology, 5, 425–438.[ISI][Medline]
Received on June 9, 2000; accepted on February 2, 2001
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. M. Tromp, X. D. R. Brouha, G. J. Hordijk, J. A. M. Winnubst, W. A. Gebhardt, M. P. van der Doef, and J. R. J. De Leeuw Medical care-seeking and health-risk behavior in patients with head and neck cancer: the role of health value, control beliefs and psychological distress Health Educ. Res., December 1, 2005; 20(6): 665 - 675. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M Tromp, X. D. Brouha, G.-J. Hordijk, J. A. Winnubst, and J R. J de Leeuw Patient factors associated with delay in primary care among patients with head and neck carcinoma: a case-series analysis Fam. Pract., October 1, 2005; 22(5): 554 - 559. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||