Home Random Page


CATEGORIES:

BiologyChemistryConstructionCultureEcologyEconomyElectronicsFinanceGeographyHistoryInformaticsLawMathematicsMechanicsMedicineOtherPedagogyPhilosophyPhysicsPolicyPsychologySociologySportTourism






Evidence From Studies With Objective Measures of Stressors

To examine whether the relationship between stressors and strains can be primarily explained by the use of self-report measures and the associated methodological problems, stud­ies are needed in which stressors are assessed by non-self-report measures. There is an increasing number of such studies. In some of these studies, researchers inferred objec­tive stressors from occupational titles and similar information. Analyses revealed significant relationships between stressful jobs and poor health and well-being. For example, Tsutsumi, Theorell, Hallqvist, Reuterwall, and de Faire (1999) reported increased odd ratios of plasma fibrinogen concentrations—a physiological indicator assumed to be associated with coro­nary heart disease—in study participants working in highly demanding jobs.

Other researchers assessed objective stressors by means of observations. These studies also showed association between objective stressors and impaired health and well-being. Forex-ample, Frese (1985) found correlations of r = .18 and r = .19 between observer ratings of psychological stressors and psy­chosomatic complaints. Melamed et al. (1995) measured mo­notony with observational ratings and found that short-cycle and medium-cycle repetitive work was significantly associ­ated with psychological distress, particularly in women. Greiner et al. (1997) reported increased odd ratios of psycho­somatic complaints in observed high-stress jobs.

In summary, these findings show that stressors at work are related to poor health and well-being—even when objective measures of stressors are used. Often the correlations be­tween objective stressor measures and strains are smaller in size than are the correlations between self-report measures of stressors and strains (cf. Frese, 1985), but they do not break down completely; this suggests that common method variance inflates the relationships between self-reported stressors and self-reported strains, but it does not fully ex­plain the empirical relationship between organizational stres­sors and strains. For methodological reasons, the correlations found between objective stressors and self-reported strains


present the lower boundary of the stressor-illness relation­ships (Frese, 1993).

Evidence From Longitudinal Studies

To arrive at a clearer picture about the causal processes be­tween stressors and strains, longitudinal studies are needed. Although they do not solve all the methodological prob­lems (Zapf, Dormann, et al., 1996), they at least allow re­searchers to rule out some of the alternative interpretations. Table 18.3 gives an overview over longitudinal studies published between 1981 and 2000 that meet the following criteria: (a) data collection on work-related stressors and strains and (b) control for initial level of strains in the analyses.

Table 18.3 shows the number of time lags, the time interval between the various measurement points, sample size, type of Stressors assessed, type of strains assessed, results with re­spect to lagged effects, concurrent effects, reverse effects (i.e., effects of strains on stressors), and nonsignificant findings. Most of the studies assessed data at two measurement points. Time lags ranged between 1 month and 180 months, with most studies using time lags of 12 months or less. A wide range of stressors were assessed, including workload, social stressors, and job insecurity. Also strain measured covered a large variety of indicators, including physiological measures, distress symptoms, depression, psychosomatic complaints, and physical illnesses. Most researchers analyzed their data with variants of cross-lagged panel correlations (CLPC), mul­tiple regression analyses, or structural equation approaches (e.g.,LISREL).



We discuss the study findings separately for concurrent, ^gged, and reverse effects. Concurrent effects refer to syn-( chronous effects of stressors (Time 2) on strain (Time 2) with controlling for strain (Time 1). Lagged effects imply effects of stressors (Time 1) on strain (Time 2) when controlling for straui (Time 1). Reverse effects refer to effects of strains (Time 1) on stressors (Time 2) with controlling for stressors Himel; drift hypothesis).

Most studies that examined concurrent effects focused on

Psychological strains (exceptions: Howard, Cunningham, &

Kechniter, 1986; Spector, Chen, & O'Connell, 2000, which

ooked at physiological strain). About half of the studies

nd concurrent effects of all measured stressors on strains.

other half of the studies found support for relationships

een some combinations of stressors and strains. Stressors

concurrent effects on strains included workload, role

lcts, and role ambiguity. Strains affected were depressive

P °ms, burnout, and fatigue spillover into leisure time.


Empirical Evidence 461

There was no systematic pattern of stressor-strain relation­ships for which concurrent effects were found.

Studies that addressed lagged effects of stressful work situ­ations examined both psychological and physical strain symp­toms. Psychological symptoms included strains such as distress, anxiety, depressive symptoms, and exhaustion. Physi­cal symptoms included mainly (psycho)somatic health com­plaints, cardiovascular disease, and other illnesses. Lagged effects of stressors on psychological strain symptoms appeared in more than half of the studies, at least for some of the stressors or strains tested. Significant effects were more often found when stressors such as high demands and high workload were examined (as opposed to social stressors), when the time lag was relatively short (not longer than 12 months), and when no concurrent effects were tested simultaneously.

There is rather strong evidence that stressors at work have a lagged effect on physical strain symptoms, particularly (psy­chosomatic health complaints (Carayon, 1993; Frese, 1985; Leitner, 1993; Parkes et al., 1994; for an exception, cf. Mauno & Kinnunen, 1999). Stressors have lagged effects on cardio­vascular disease, particularly in men (Hibbard & Pope. 1993; Karasek, Baker, Marxner, Ahlbom, & Theoreli, 1981). How­ever, stressors seems to have none or only a minor lagged effect on other illnesses such as cancer (Hibbard & Pope, 1993). Taken together, these longitudinal studies suggest that there are lagged effects of stressors on strains, particularly if the time lag between two measurement points does not exceed 12 months.

Most of the studies tested either concurrent or lagged ef­fects. The majority of these studies found evidence for an effect of stressors on strains, at least for some of the stressor or strain indicators. There are only a few studies that ana­lyzed both lagged and concurrent effects within the same data set (Glickman, Tanaka, & Chan, 1991; Kohn & Schooler, 1982; Moyle, 1998; Roy & Steptoe, 1994; Schonfeld, 1992; Wolpin, Burke, & Greenglass, 1991). All these studies found concurrent effects (at least for some of the indicators). How­ever, half of the studies failed to find lagged effects when concurrent effects were present. Only Wolpin et al. (1991), Schonfeld (1992), and Moyle (1998) reported lagged effects in the presence of concurrent effects. These findings indicate that individuals develop distress reactions to stressful situa­tions rather quickly; this implies that having experienced stressful work situations in the past may have little effect on one's psychological well-being unless the stressful situation continues into the present. We assume, however, that the situation is different for physical symptoms. More studies on physical indicators are needed that examine concurrent and lagged effects simultaneously.


TABLE 18.3 Longitudinal Studies on Stressors,' Weil-Being, and Health            
  Number of Time Sample Statistical     Third Lagged Main Concurrent   Effects Tested
Study Waves Lags Size Procedure Stressors Strains Variables Effects Main Effects Reverse Effects But Not Found
Bakker. 2 60 LISREL Patient Burnout Not tested Patient demands Depersonalization
Schaufeli,       demands       71 emotional 71 patient demands  
Sixma.               exhaustion;    
Bos veld,               effects on other    
& van               burnout    
Dierendonck (2000)               components are mediated by emotional exhaustion    
Beg ley & Czajka 2 3 Multiple regression Experienced stressfulness Jobdispleasure Age, gender, marital status, Not tested Experienced stressfulness 71 Not tested
(1993)       of organizational change   education, organizational tenure, NA, organizational commitment   job displeasure    
Bromet, Dew, 2 12 322-325 Multiple Job demands Affective Age, coworker Job demands 7t Not tested Not tested No effect of job
Parkinson, &     regression.   disorders, support, affective     demands on
Schulberg (1988) i     Logistic regression   alcohol-related problems, distress symptoms friendship support disorders; Job demands 71 distress symptoms (P< .10)     alcohol-related problems
Carayon 2 12 CRPC Workload Daily life Not tested in Workload 71 Not tested No reverse No lagged effect
(1993)         stress, physical health complaints CLPC physical health complaints   effects of workload on daily life stress
Chapman, Mandryk, Fronimer, Edye, & 3 36 + 24 2,634 Multiple regression Quantitative demands, qualitative demands, Systolic blood pressure, diastolic Age, education, weight, fitness, alcohol Young women: Quantitative demands 71 diastolic blood ** chronicity scores: mixture of lagged and concurrent Not tested No effects on systolic blood pressure; no effects for
Ferguson (1990)       outside stress blood pressure consumption, family history, etc. pressure effects   men; no effects for all women together
Daniels & 2 1 Multiple Various Well-being No Not tested Stressors il Not tested  
Guppy(1994)     regression stressors       well-being    
Dormann & 3 4 + 8 LISREL Social Depressive No effect of Not tested Not tested No effect of
Zapf(1999)       stressors symptoms   social stressors     social stressors
Frese(1985) 2 16 53-79 CLPC Psychological stressors Psycho somatic complaints Not tested in CLPC Psychological stressors 71 psychosomatic complaints Not tested No reverse effects No effect when observational measure of stressors was used
                ;.- ■ _    

 

                Uncertainty 71      
Garst, 6 4, 10, Growth Job Depression, Not in growth Stressors 7t Strains ^1 No lagged
Frese, &   12,   curve insecurity. psychos omatic curve model depression. strains Stressors effects of job
MoJenaar   12,24   model time complaints.   psychosomatic     insecurity
(2000)         pressure, organizational problems, social stressors, uncertainty irritation, worrying   complaints, irritation, worrying; social stressors 7\ psychosomatic complaints, worrying      
Glickman. 2,506 LISREL Work load Distress Age, life No lagged effect Workload and Distress 7\ No lagged effect
Tanaka,&         and economic   events of workload and economic strain workload and of workload and
Chan (1991)         strain     economic strain on distress 7\ distress economic strain (lagged) economic strain on distress
Hibbard & 2,157 Prospective Work stress Ischiamic Age, Men: work stress Not tested Not applicable No effects for
Pope(1993)       design   heart disease (IHD). malignancy, stroke, death education, self-reported health, marital and parental roles *IHD     women: no effects on malignancy, stroke, or death
Howard, Multiple Role Systolic blood Hardiness   Type A   No effects for
Cunningham,       regression ambiguity pressure.     individuals:   TypeB
& Rechniter           diastolic blood     change in role   individuals;
(1986)           pressure, cholesterol, triglycerides, uric acids     ambiguity 71 systolic blood pressure, diastolic blood pressure, triglycerides   no effects on cholesterol and uric acids
Karasek, 1,461 Logistic Jobdemands Cardiovascular Age. Job demands 71 Not tested Not applicable
Baker,       regression   disease intelligence cardiovascular      
Marxner,             discretion. disease      
Ahlbom, &             personal        
Theorell             schedule        
(1981)             freedom, education, smoking, overweight        
Kohn& LISREL Time Distress No lagged Dirtiness 7% Distress 7t time No effects of
Schooler         pressure,     effects on distress; pressure time pressure
(1982)         heaviness, dirtiness, hours of work     distress hours of work )l distress (concurrent); distress ^1 heaviness (lagged) and heaviness on distress; no lagged effects of dirtiness and hours of work on distress
                      (Continued)

TABLE 18.3 t Continued)                    
  Number of Time Sample Statistical     Third Lagged Main Concurrent   Effects Tested
Study Waves Lags Size Procedure Stressors Strains Variables Effects Main Effects Reverse Effects But Not Found
Ue& LISREL Role stress Emotional No Lagged effects Role stress 79 Not tested  
Ashforth           exhaustion   not testable emotional    
(1993)           (EE); depersonal ization personal accomplishment i. effect of role exhaustion; stress on depersonali zation and personal accomplishment mediated by EE    
Leitner (1993) 12 + 12 CLPC Barriers in work process Psychosomatic complaints, irritation, strain, depression, anxiety, somatic symptoms   Barriers 7t psychosomatic complaints, irritation, strain, depression, somatic symp­toms, illness Not tested No reverse effect No effect on anxiety or illness
Mauno & LISREL Job Exhaustion, No Women: Not tested No reverse No effects for
Kinnunen         insecurity somatic   job insecurity 71   effects men; no effects
fc (1999)           symptoms.   exhaustion;     on somatic
            spillover into parenthood   job insecurity 7\ spillover     symptoms
Mohr (2000) 62-65 Partial correlations Job insecurity Irascibility, anxiety, psychosomatic complaints, depression   Not tested Job insecurity 7\ anxiety; job insecurity 71 psychosomatic complaints Not tested No effects on irascibility and depression
Moyle(1998) 7 + 5 LISREL Demands Distress (GHQ) Neuroticism Demands 7\ distress Demands 31 distress Distress 3* demands  
Moyle & 2.5 +6 Multiple Demands Distress Not entered Not tested Demands 7\ Not tested Relocation per se
Parkes(l999)       regression relocation (GHQ) into regression equation before other variables   distress   does not affect distress
Muntaner, 11,789 Prospective Psychological Psychotic __ Physical Not tested Not applicable/ No effects on
Tien, Eaton, &         demands. disorders   demands 7t   not tested psychotic-affective
Garrison         physical (delusions,   delusions,     disorders, no effect
(1991)         demands schizophrenia, psychotic affective)   psychological demands ^1 schizophrenia     of psychological demands on delusions, effect of physical demands on schizophrenia marginally sigmficjiil {7t)

 

                V      
Nelson & 6 + 3 Multiple Chronic EH stress Coping Chronic stressors Not tested Not tested
suttan (i yy<»       regression stressors symptoms   7t distress symptoms      
                (3-month time lag)    
Newton & Multiple Role conflict. Job dissatis- Not tested A role conflict 7t Not tested Quantitative
Keenan(1990)       regression role ambiguity, quantitative high load, qualitative low load faction, anxiety, anger, frustration, hostility     job dissatisfac­tion, anxiety, anger, frustra­tion, hostility; A role ambiguity 71 job dissatis­faction, anxiety, anger, frustration, hostility; A quantitative high load 71 anxiety; A quantitative high load 51 job dissatisfaction, anger: A qualitative low load 7\ job dissatisfaction, anger, frustra­tion, hostility   high load has no effect on frustration or hostility: qualitative low load has no effect on anger
Noor(1995) Multiple regression Role overload GHQ scores. happiness Age, NA Not tested Role overload 71 GHQ scores Not tested No effect on happiness
Parkes(1991) Multiple regression Demands (i.e., time pressure) Anxiety Age, gender, discretion, locus of control. social dysfunctioning Demands 7\ anxiety Demands 7\ Not tested Not tested Not tested Not tested  
Parkes, Multiple Demands Somatic Gender, age. -_
Menham, &       regression   symptoms neuroticism somatic      
von Rabenau               symptoms      
(1994)                      
Revicki, 12 + 12 369 (1st Multiple (low) Role Depressive Age. gender. No effects Not tested Not tested No effects
Whitley,     time lag); regression clarity symptoms marital status,        
Gallary, &     192 (2nd       other strain        
Allison (1993)     rime lag)       symptoms        
Roy & Steptoe 3 + 3 + Multiple Daily Depression NA, social No lagged Daily stressors No reverse No lagged
(1994)     regression stressors   support effects 7t depression for all three time lags effects effects
                      {Continued)

 

 

 

TABLE 18.3 {Continued)                    
  Number of Time Sample Statistical     Third Lagged Main Concurrent   Effects Tested
Study Waves Lags Size Procedure Stressors Strains Variables Effects Main Effects Reverse Effects But Not Found
Rydstedt, Multiple Workload Perceived Gender Not tested Delta-workload Not tested  
Johansson. &       regression   effort, fatigue     31 perceived    
Evans(1998)           spillover, intake of stress-related drugs     efforts, fatigue spillover    
Sehonteld Approx. LISREL Episodic and Depressive No control Stressors 31 Stressors 31 No reverse effects  
(1992)       chronic stressors symptoms variables in LISREL models depressive symptoms depressive symptoms; concurrent models fits the data better than lagged model    
Shirom, 2 24-36 Multiple Overload Cholesterol Age, body Women; Not tested   No effects in
Westman,       regression   triglycerides mass index, Overload 31     men; no
Shantai, &             emotional cholesterol     effect on
Carel(1997)             reactivity, burnout, fatigue       triglycerides
Siegrist, Peter, 2 Logistic Status Ischiamic Age, body Status Not tested Not applicable
Junge, Cremer,       regression inconsistency, heart disease mass index. inconsistency 31      
& Seidel         job insecurity. (IHD) blood pressure, IHD, job      
(1990)         work   cholesterol. insecurity 31 IHD,    
          pressure   coping work pressure      
                31 IHD      
Spector, Chen. 2 Approx. Partial Interpersonal Anxiety, NA Not tested AH stressors 31 Not tested No effect on
& O'Connell     Korr conflict frustration.     anxiety and   physical
(2000)         constraints, role ambiguity, role conflict, workload job satisfaction, physical symptoms     frustration, role ambiguity and role conflict X satisfaction   symptoms
Tang& 2 Multiple Stressors in Illness Hardiness Stressors 31 Not tested Not tested
Hammontree       regression police jobs     illness      
(1992) 2                
Wolpin, Multiple Various Burnout, Stressors Si job Stressors ^1 job Not tested No lagged
Burke. &       regression stressors job   satisfaction satisfaction.   effects on
Greenglass           satisfaction     stressors 31   burnout
(1991) 2 CLPC         burnout    
Zapf & Frese, Social Psychosomatic Not in CLPC No lagged Not tested No reverse effects No lagged
(1991)         stressors complaints, irritation, strain, anxiety, depression   effects     effects
Nok'. Time lags refer to months. CLPC = cross-lagged panel correlation. NA = negative affectivity.          

      '-■"■"■ '"*&y            
Nt*|s«n <fc 6 + 3 9f Multiple Chronic Distress Coping Chronic stressors Not tested Not tested
Sutton (1990)       regression stressors symptoms   7f distress symptoms      
                (3-month time lag)    
Newton & Multiple Role conflict. job dissatis- _ Not tested A role conflict 7* Not tested Quantitative
Keenan(1990)       regression role ambiguity, quantitative high load, qualitative low load faction, anxiety, anger, frustration, hostility     job dissatisfac­tion, anxiety, anger, frustra­tion, hostility: A role ambiguity 7\ job dissatis­faction, anxiety, anger, frustration, hostility; A quantitative high load 7\ anxiety: A quantitative high load 51 job dissatisfaction, anger: A qualitative low load 7\ job dissatisfaction. anger, frustra­tion, hostility   high load has no effect on frustration or hostility: qualitative low load has no effect on anger
Noor(1995) Multiple regression Role overload GHQ scores, happiness Age, NA Not tested Role overload 7\ GHQ scores Not tested No effect on happiness
Parkes(199l) Multiple regression Demands (i.e., time pressure) Anxiety Age, gender, discretion, locus of control, social dysfunctioning Demands 7\ anxiety Not tested Not tested  
Parkes, Multiple Demands Somatic Gender, age. Demands 7t Not tested Not tested
Menham, &       regression   symptoms neuroticism somatic      
von Rabenau               symptoms      
(1994)                      
Revicki, 12 + 12 369 (1st Multiple (low) Role Depressive Age, gender. No effects Not tested Not tested No effects
Whitley,     time lag); regression clarity symptoms marital status.        
Gallary, &     192 (2nd       other strain        
Allison (1993)     time lag)       symptoms        
Roy & Steptoe 3 + 3 + Multiple Daily Depression NA, social No lagged Daily stressors No reverse No lagged
(1994)     regression stressors   support effects 71 depression for all three time lags effects effects
                      (Continued)

Agrowing number of studies have tested reverse effect. These studies addressed the question of whether strains lead to an increase in stressors as suggested in the drift hypothesis (cf. Zapf, Dormannn, et al., 1996). In 9 out of 12 studies, no such reverse effects were found (Carayon, 1993; Frese, 1985; Garst et al., 2000; Leitner, 1993; Mauno & Kinnunen, 1999; Moyle, 1998; Roy & Steptoe, 1994; Schonfeld, 1992; Zapf & Frese, 1991). Three studies reported reverse effects for (some of the) strain symptoms on (some of the) stressors (Bakker etal.,2000;Glickmanetal., 1991; Kohn & Schooler, 1982). It is interesting to note that in most of the studies that found such reverse effects, both types of effects were present—effects of stressors on strains and effects of strains on stressors. This suggests that—at least for some individuals—experiencing organizational stress may be linked to a negative spiral: Stres­sors increase strain, which in turn increases stressors. Moyle (1998) and Garst et al. (2000), however, found an effect oppo-9 site to the drift hypotheses (a sort of refuge model). People with high strain eventually received workplaces that had fewer demands and stressors.

In summary, there is good and increasing evidence that stressors at work have a causal effect on health and well-being. The support for concurrent effects is stronger than for lagged effects, at least for psychological strains. Consistent lagged effects were mainly found for physical strain symp­toms. This implies that an individual's present work situation seems to be more relevant for developing psychological dis­turbances, whereas an individual's past work situation may also have long-term effects on his or her physical health and well-being. Clearly more research is needed that examines concurrent versus lagged effects more systematically. More­over, more attention should be paid to the time intervals at which data are gathered (cf. Dormann & Zapf, 1999). Differ­ential effects of different stressors and different models of stressor-strain relationships should be examined (Frese & Zapf, 1988; Garst etal., 2000).


Date: 2016-03-03; view: 778


<== previous page | next page ==>
Theoretical Models on the Relationship Between Stressful Situations and Strains | The Role of Resources
doclecture.net - lectures - 2014-2024 year. Copyright infringement or personal data (0.015 sec.)