Regulating employees' health behaviors: The effects of personal health-related orientations on legitimacy perceptions of organizational programs and policies

Klautke, Hannah; Park, Hee Sun Article Regulating employees' health behaviors: The effects of personal health-related orientations on legitimacy perceptions of organizational programs and policies International Journal of Business Science & Applied Management (IJBSAM) Provided in Cooperation with: International Journal of Business Science & Applied Management (IJBSAM) Suggested Citation: Klautke, Hannah; Park, Hee Sun (2011) : Regulating employees' health behaviors: The effects of personal health-related orientations on legitimacy perceptions of organizational programs and policies, International Journal of Business Science & Applied Management (IJBSAM), ISSN 1753-0296, International Journal of Business Science & Applied Management, s.l., Vol. 6, Iss. 1, pp. 1-15 This Version is available at: https://hdl.handle.net/10419/190619 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/2.0/uk/ Int. Journal of Business Science and Applied Management, Volume 6, Issue 1, 2011 Regulating employees' health behaviors: The effects of personal health-related orientations on legitimacy perceptions of organizational programs and policies Hannah Klautke Department of Communication, Michigan State University East Lansing, MI 48824, USA Tel: +01 (314) 814-4262 Email: [email protected] Hee Sun Park Department of Communication, Michigan State University East Lansing, MI 48824, USA Tel: +01 (517) 355-3480 Email: [email protected] Abstract The current study investigated individuals' responses and evaluations of worksite health programs and policies. Upper level undergraduates rated the legitimacy of policies and programs exerting low, moderate, or high levels of control over employees’ health behaviors. The findings showed that individuals’ nutrition orientations accelerated the decrease in legitimacy ratings that accompanied increased control over employee fitness and health risk appraisal. On the other hand, individuals’ antismoking orientations slowed the decrease in legitimacy ratings that accompanied increased control over employee smoking behavior. Implications of the findings for job counselors and human resource personnel are discussed. Keywords: worksite, health, smoking, fitness, health risk appraisal Int. Journal of Business Science and Applied Management / Business-and-Management.org 2 1 INTRODUCTION The growth in corporate health programs and policies constitutes one of the most pronounced trends in business over the past 20 years, and one that is likely to continue, and even accelerate, in the future (Goetzel & Ozminkowski, 2000; Harris, 1994). Programs range from health awareness - and supportive environment - programs to programs aiming at behavioral change, and target areas such as smoking cessation, fitness coaching, health risk awareness, hypertension control, stress management, and even spiritual employee health (e.g., Kirby, 2006; O'Donnell, 1991). Motivations for organizations to implement such programs can include demonstrating concern for employees, improving management-labor relations, and raising employee morale, although it may be impossible to disentangle the intermediate goal of healthier employees from the ultimate goal of cost savings (Mayer, 1991). In cases where programs are being offered on a truly voluntary basis and, more importantly, perceptions of organizational support outweigh the restrictions that are being imposed (Dalsey & Park, 2009), individuals may well enjoy personal benefits provided by these programs rather than feel victimized by them. Types of health programs and policies at worksite can vary in the extent to which they regulate individual employees' health-related behaviors. When designing and implementing health programs and policies, companies may need to consider factors that may affect individuals' responses to the health programs and policies. It is important to see whether individuals who will be affected by the health programs and policies will regard the programs and policies as the legitimate and appropriate level of organizational control. Considering that undergraduates are about to enter the workforce and apply for jobs in near future, how they will evaluate various health programs and policies can be useful information for companies that consider recruiting college graduates and implementing various health programs and policies. Furthermore, undergraduates' orientations and behaviors regarding maintaining or improving their health, nutrition, and fitness can be a factor that affects how openly they welcome voluntary or mandatory participation in worksite health programs and policies. The current paper focuses on health programs and policies in smoking cessation, fitness, and health risk appraisal and investigates individual health orientation factors that are likely to affect legitimacy perceptions of low, moderate, and high control levels of health programs and policies. 2 LITERATURE REVIEW 2.1 Health promotion in the workplace The worksite can be an efficient place for providing public health education such as encouraging smokers to engage in cessation techniques (Osinubi, Barbeau, Williams, & Sorensen, 2005). First, there is the advantage of broad reach. With half of the adult population working outside of the home, worksite programs have the potential to reach even those who traditionally lack good connections to healthcare and health education networks (Osinubi et al., 2005). Second, there is the advantage of exposure. The sheer amount of time individuals spend at work makes the worksite a valuable platform for health campaigns. A third and equally important advantage lies in the opportunities for social support, both in the form of colleagues working toward similar health goals together . Employer-based initiatives may provide the little "nudge" that may be needed to move from good intentions to actual healthy behaviors in order to maintain long-term changes (Mayer, 1991). 2.2 Effects on employees As human resources strategists note, improving the company's bottom line and demonstrating true care for employees are not mutually exclusive goals (Pfeffer, 1998). In the case of corporate health interventions, a concern for employees’ well-being may well be an additional motive for program implementation and employees may well appreciate ―being treated like family‖ (Hunnicutt, 2001). By the same token, some policies can backfire and cause reactance, reduce organizational attractiveness, and worsen the individuals' and company's well-being at the same time. This is especially true if programs get more involved with health screenings and are perceived as intruding into employees’ privacy (Greer & Labig, 1987; Konovski & Cropanzano, 1991; McGregor, 2007; Truxillo, Baier, & Paronto, 2002). The question is: at what point does a caring, big-brother-like company start turning into an Orwellian version of a big brother ―watching‖ and penalizing employees for unhealthy behaviors previously thought private? Regarding the implementation of employee health-related interventions, more attention needs to be paid to the perceptions of those on the receiving end of various policies formulated today (Dalsey & Park, 2009; Konovski & Cropanzano, 1991; O'Donnell, 2000). At what point are companies' health Hannah Klautke and Hee Sun Park 3 regimens perceived as "going too far," and what types of interventions are being perceived as legitimate and/or desirable, if any? Critical voices have pointed to the paternalistic nature of these new forms of corporate control over one's life (Kirby, 2006; Zoller, 2003), and anecdotal evidence for employee concern over a ―slippery slope‖ exists (Jones, 2007; Park, Dalsey, Yun, Guan, & Cherry, 2008). Individuals reported the feeling of having their privacy being whittled away by attempts to change employee behavior after five o’clock, while employers argue that it is up to the individual whether or not they want to work under certain mandates (Jones, 2007). Although certain health issues, such as obesity, are somewhat less amenable to employer control for legal reasons, halting behaviors such as smoking seems to be more widely accepted; there is a move from incentive based, honorsystem approaches toward penalty-based approaches to noncompliance and more intrusive testing (McGregor, 2007). For example, one employer started charging employees a biweekly penalty of $30 unless they meet weight, cholesterol and blood-pressure guidelines set forth by the company (Rose, 2008). For some, such intrusive health policies are overstepping sacred boundaries, and some firms endorsing aggressive anti-smoking laws have attracted negative attention with the public both within and outside the organization (Jones, 2007), while others are open to and even supportive of organizational control in these areas; some may even find it desirable to work at a place where healthy behavior is forced upon individuals. 2.3 Legitimacy perceptions of health programs and policies Legitimacy pertains to the belief that social arrangements, institutions, authorities and their decisions and rules are appropriate, proper and just (Tyler, 2006). Individuals’ perception of legitimacy regarding organizational policies and programs can be important for organizations to positively influence their members to participate in the programs and adhere to the policies. Assuming that organizations do not prefer to simply impose a certain type of health behaviors to their members and garner undesirable consequences, organizations may want to know how individuals will think about legitimacy and acceptability level of organizational control when enforcing health-related regulations. For example, when organizations change their smoking regulations from smoking allowed only in designated break rooms to administering a mandatory pre-employment nicotine test, some individuals may consider the new change to be much less legitimate than others may. 2.4 Types of interventions investigated in the current study This paper focuses on three of the most common interventions: Smoke-free programs, employee fitness programs, and health-risk appraisals (HRAs). All three approaches are well established in organizational practice (Harris, 1994), and remain at the core of current health initiatives. For example, the ―Wellness Management‖ program advocated by corporate healthcare provider Meritain is based on three key initiatives labeled ―Nicotine Free,‖ ―Physically Fit,‖ and ―Managed Metrics,‖ (Meritain Health, 2007). These programs indicate cost-saving potentials and are likely to continue to play a key role in corporate health promotion. Smoke-free programs: Whether in the form of incentives for quitting smoking or disincentives for being a smoker, programs designed to eliminate or regulate smoking at the worksite do not only have a long standing in business practice, but continue to gain in prevalence (Rose, 2008; Sofian, McAfee, Doctor, & Carson, 1994). After all, cigarette smoking has been identified as the leading preventable cause of illness and premature death in the U.S., increasing the risk for heart disease, stroke, emphysema, and many cancers (CDC, 2005). It is associated with direct costs to businesses reaching $75 billion a year in direct medical costs, $92 billion in lost productivity and $10 billion in exposure to second-hand smoke (CDC, 2005; Meritain, 2008; Osinubi et al., 2005). Fitness programs: After cigarette smoking, obesity is the second-leading cause of preventable death in the U.S. A study commissioned by the AOA found that the direct healthcare costs related to obesity reached over $102 billion in 1999 (American Obesity Association, 2002). Efforts to reduce obesity and improve fitness are well-reflected in exercise and fitness programs pervasive in U.S. worksites (Collingwood, 1994). Health-risk appraisals (HRAs): Even if no immediate medical attention is needed, periodic health reports have been found to effectively increase employees' awareness of health issues, a prerequisite for long-term behavior modification (Dunton, 1991). While awareness and early detection are the basis for prevention and effective treatment of most illnesses, health-risk-appraisals are one of the most controversial elements of corporate health policies, especially if they go beyond self-report data, as in Int. Journal of Business Science and Applied Management / Business-and-Management.org 4 blood chemistry analyses (Konovski & Cropanzano, 1991; McGregor, 2007; Wallston & Armstrong, 1994). 2.5 Levels of control exerted via health interventions Canby (2007) describes the range of incentives and disincentives that may be used to encourage compliance with corporate health programs. Rewards may range from public recognition, gift certificates and gym memberships to days off, cash payments and reduced medical premiums. Penalties for noncompliance, may include increased health insurance premiums, paycheck reductions, and even termination of the employee. Some employers have elaborate calculation procedures for insurance copays and deductibles in place, based on employees’ BMI, blood pressure, and even cholesterol levels (McGregor, 2007). The current study grouped various health initiatives into three levels of "control" based on their intrusiveness into employees' life. Interventions labeled as low in control comprised those programs that are implemented on a voluntary basis, are designed to raise awareness and encourage practices that do not infringe on behaviors off the job. Examples are the limitation of smoking to designated areas, encouragement to join the corporate fitness club membership program, and voluntary sign-ups for health screenings. Interventions labeled as moderate in control are those that regulate employees' behaviors on and beyond the job, such as hiring only nonsmokers (while encouraging smoking cessation and offering support to current employees who smoke), ordering mandatory fitness regimens, or creating annual health files for employees. Lastly, interventions labeled as high in control refer to drastic measures such as terminating employment of smokers for their failure to quit smoking, or denying medical benefits to high-risk individuals who fail to improve their personal fitness or reduce health risk indicators. 2.6 Research question Among a host of factors that can affect individuals' legitimacy ratings of various types of health programs and policies, the current study focuses on individuals’ orientations related to health and fitness. Individuals differ in their health and fitness related attitudes, habits, and perceived healthrelated self-efficacy. These individual differences are likely to explain some variation in perceived legitimacy and desirability of corporate health policies and programs. A match between a job applicant’s personal values and a company’s values has been shown to positively affect the job applicant’s evaluation of the organization (Bretz, Ash, & Dreher, 1989; Chatman, 1991; Dalsey & Park, 2009). Similarly, a match between individuals’ health-related orientations and specific organizational health policies that relate to these orientations is likely to increase perceptions of legitimacy of such policies. Specifically, compared to smokers and individuals with favorable attitudes about smoking, non-smokers and individuals with anti-smoking attitudes may be more likely to welcome severe anti-smoking policies. Dalsey and Park (2009) showed that, compared to smokers, non-smokers indicated higher attraction for an organization that encouraged employees to quit smoking. Similarly, individuals with greater concern of and care for their fitness, nutrition, and wellness can be more likely to have stronger legitimacy perceptions about higher levels of employeefitness and health risk appraisal interventions. The research question examined in this study is: How do personal fitness, nutrition, anti-smoking, and wellness orientations affect the perceptions of legitimacy of employer control of various health behaviors? 3 METHOD 3.1 Participants Participants (n = 115, age M = 21.90, SD = 2.79, 65.3 % women) were recruited from upper division undergraduate classes at a large Midwestern university in the U.S. The sample consisted of 84.3% Whites/Caucasians, 7.0% African Americans, 3.5% Asians, and 5.2% who indicated other ethnicities. The majority (76.7%) was in their senior year, and 67.0 % indicated their plans to enter the workforce within the next year. This student sample thus possessed two desirable properties for the purpose of the current investigation: While these individuals are seriously thinking about potential workplaces and their characteristics, they are still less concerned with an immediate need for employment. It is likely that their personal orientations and opinions towards corporate health policies are relatively free from practical necessity considerations that may enter the picture for long-term unemployed job seekers, and from influences through previous corporate health promotions. Based on considerations outlined above, but also as an induction for respondents to seriously think about their working future, we assessed what industries, job types, and geographical regions were sought. Answers reflected a high popularity for marketing-related jobs (26.0%), the public relations Hannah Klautke and Hee Sun Park 5 industry (23.4%), health communication (11.7%), the media and entertainment industry (8.1% and 9.0%), and legal/financial/other services (7.4% each). Of the participants, 44% wished to stay in the Midwest for their work, while 37.2% felt drawn to the West coast (19.6.2%), the East coast (17.6%), or the South (8.8%). 3.2 Measures All measurement items were constructed for this study, using a 7-point Likert style response format (1 = strongly disagree, 7 = strongly agree). Appendix shows all the measurement items. Table 1 shows the reliabilities, correlations, means, and standard deviations. Table 1: Reliabilities, Correlations, Means, and Standard Deviations Personal health-related orientations Legitimacy ratings of interventions Fitness Nutrition Antismoking Wellness Psc low Psc mod Psc high Pfc low Pfc mod Pfc high Phc low Phc mod Phc high Fitness (.87) Nutrition .53** (.72) Antismoking .13 .15 (.83) Wellness .47** .52** .18 (.82) Psclow .22* .14 .30** .18 (.81) Pscmod .22* .14 .25** .16 .41** (.82) Pschigh .13 .04 .30** .17 .32** .71** (.91) Pfclow .27** .26** .24** .06 .42** .31** .38** (.88) Pfcmod .19* .05 .23* .07 .18 .55** .69** .57** (.83) Pfchigh .17 -.02 .11 .04 .19 .55** .72** .38** .78** (.91) Phclow .19 .25 .20* .03 .32** .30** .26** .65** .50** .25** (.91) Phcmod .17 .09 .10 .15 .30** .50** .60** .47** .72** .71** .47** (.92) Phchigh .06 -.06 .12 10 .16 .45** .66** .31** .67** .79** .26** .80** (.95) M 5.00 4.84 6.15 5.59 6.19 4.14 3.34 5.40 3.85 2.79 5.53 3.46 2.58 SD 1.28 1.19 1.16 0.86 0.88 2.28 1.73 1.33 1.40 1.32 1.29 1.41 1.43 *p < .05, **p < .01. Reliabilities are reported on the diagonal. 1 = strongly disagree; 7 = strongly agree Fitness: personal orientation placed on fitness, assessed with 3 items Nutrition: personal orientation placed on healthy nutrition, assessed with 3 items Anti-Smoking: anti-smoking-related orientation, assessed with 5 items Wellness: wellness awareness orientation, assessed with 7 items Psclow: legitimacy rating of low employer control of smoking behaviors, assessed with 7 items Pscmod: legitimacy rating of moderate employer control of smoking behaviors, assessed with 5 items Pschigh: legitimacy rating of high employer control of smoking behaviors, assessed with 5 items Pfclow: legitimacy rating of low employer control of employee fitness, assessed with 6 items Pfcmod: legitimacy rating of moderate employer control of employee fitness, assessed with 5 items Pfchigh: legitimacy rating of high employer control of employee fitness, assessed with 8 items Phclow: legitimacy rating of low employer control of health risk appraisals, assessed with 5 items Phcmod: legitimacy rating of moderate employer control of health risk appraisals, assessed with 9 items Phchigh: legitimacy rating of high employer control of health risk appraisals, assessed with 7 items 3.3 Personal health-related orientations For this study, four categories of personal health related orientations were expected to potentially influence perceptions of policies that related to them. These orientations were assessed through topicrelated attitudes, behaviors, and values. Fitness orientation describes personal value placed on fitness, getting or staying in shape, and exercise habits and was assessed using three items (α = .87) such as ―I exercise on a regular basis.‖ Nutrition orientation was measured with three items (α = .72) such as "I Int. Journal of Business Science and Applied Management / Business-and-Management.org 6 maintain a well-balanced diet" and assessed respondents’ views of the importance of healthy nutrition. Anti-smoking orientation refers to an overall tendency to be unfavorable about smoking, rather than a simple smoker-versus-nonsmoker dichotomy. It was assessed with a continuous measure that integrated four items (α = .83) on smoking behaviors and attitudes about being around smokers, such as ―I prefer my environment to be smoke-free.‖ Finally, seven items (α = .82) assessed wellness orientation, or respondents' general health awareness and behaviors regarding regular medical check-ups and taking preventive measures (e.g., "I am interested in ways of preventing illness."). Confirmatory Factor Analysis (CFA) showed that four-factor model (four types of orientations) fit the data well (CFI [Comparative Fit Index] = .90, IFI [Incremental Fit Index] = .90) and was better than one-factor model (CFI = .73, IFI = .73), Δχ2(6) = 269.78, p < .001. 3.4 Legitimacy of interventions. Health programs and policies were grouped into three levels varying in severity for the three general areas of interest, smoke-free programs, employee fitness programs, and health-risk-appraisals. Legitimacy ratings were assessed by individuals indicating the extent to which they agreed or disagreed with each program and/or policy. For smoke-free interventions, CFA showed that three-factor models (low, moderate, and high control) fit the data well (CFI = .95, IFI = .95) and was better than one-factor model (CFI = .87, IFI = .87), Δχ2(3) = 305.87, p < .001. Individuals' legitimacy rating of low control of smoking behaviors (Psclow) was assessed with seven items (α = .81) such as ―An employer has the right to limit smoking to designated areas." Legitimacy ratings of moderate control of smoking behaviors (Pscmod) were assessed with five items (α = .82) such as ―A company may use mandatory pre-employment nicotine testing.‖ Legitimacy ratings of high control of smoking behaviors (Pschigh) were assessed with five items (α = .91) such as ―One year after making a company smoke-free, an employer has the right to fire smokers that fail to quit smoking.‖ For employee fitness interventions, CFA showed that the three-factor model fit the data well (CFI = .91, IFI = .91) and was better than one-factor model (CFI = .83, IFI = .83), Δχ2(3) = 455.04, p < .001. Individuals' legitimacy ratings of low control of employee fitness were assessed with six items (α = .88) such as ―It is okay for the company to encourage regular exercise by inviting employees to a companywide fitness-challenge event‖ (Pfclow). Moderate control was assessed with five items (α = .83) such as ―Depending on the employee’s fitness level, a mandatory meeting with an assigned fitness coach may be imposed‖ (Pfcmod). High control was assessed with eight items (α = .91) such as "A company has the right to charge higher out-of-pocket health insurance contributions of employees who fail to improve their fitness scores substantially after one year" (Pfchigh). Finally, for health risk appraisal interventions, CFA showed a three-factor model to be a better fit (CFI = .95, IFI = .95) than one-factor model (CFI = .87, IFI = .87), Δχ2(3) = 897.53, p < .001. One example of the five (α = .91) items measuring individuals' legitimacy rating of low control over employee-health-risk appraisals (Phclow) was "A voluntary sign-up opportunity for complementary physical check-ups is a valuable service to employees.‖ Nine items (α = .92) measured moderate control over employee-health-risk-appraisals (Phcmod) (e.g., "An employer has the right to create annual health reports on all employees"). Seven items (α = .95) assessed high control over employeehealth-risk appraisals (Phchigh) (e.g., "It is okay for an employer to prescribe annual blood tests to check for cholesterol levels."). 4 RESULTS 4.1 Overview The data were examined with Hierarchical Linear Modeling (HLM) (Raudenbush & Bryk, 2002) because the research design involved repeated measures, and HLM allows separating the variance in the dependent variables (i.e., legitimacy ratings) into within-individual variance and betweenindividual variance. Control types (low, moderate, and high) was used to explain within-individual level variance in the dependent variables. For between-individual level variance in the dependent variable, individual health orientation variables (fitness, nutrition, anti-smoking, and wellness orientations) was used to explain the variance in the extent to which individuals differ in how they respond to each of the three control types for each health program. That is, HLM allowed for partitioning of variance in legitimacy ratings into segments accounted for by the three types of control (level-1 predictor) and stable individual differences (level-2 predictors). Table 2 shows HLM results. Hannah Klautke and Hee Sun Park 7 Table 2: Multilevel analyses results Coefficient SE t df p-value Smoke-Free Programs1 For intercept 1, β0j Intercept 2, γ00 5.98 0.11 52.77 339 < .001 For control slope, β1j Intercept 2, γ10 –1.42 0.12 –12.19 110 < .001 Fitness, γ11 0.13 0.09 1.44 110 .152 Nutrition, γ12 –0.06 0.10 –0.61 110 .543 Anti-smoking, γ13 0.26 0.08 3.06 110 .003 Wellness, γ14 0.06 0.14 0.44 110 .661 Fitness Programs2 For intercept 1, β0j Intercept 2, γ00 5.32 0.12 45.91 110 < .001 Fitness, γ01 0.21 0.11 1.83 110 .070 Nutrition, γ02 0.18 0.13 1.41 110 .162 Anti-smoking, γ03 0.27 0.10 2.66 110 .009 Wellness, γ04 –0.17 0.17 –0.95 110 .343 For control slope, β1j Intercept 2, γ10 –1.30 0.07 –19.43 110 < .001 Fitness, γ11 0.02 0.06 0.36 110 .723 Nutrition, γ12 –0.19 0.07 –2.65 110 .010 Anti-smoking, γ13 –0.07 0.06 –1.09 110 .279 Wellness, γ14 0.08 0.10 0.84 110 .404 Health Risk Appraisals3 For intercept 1, β0j Intercept 2, γ00 5.33 0.11 47.06 110 < .001 Fitness, γ01 0.12 0.11 1.11 110 .269 Nutrition, γ02 0.26 0.12 2.08 110 .040 Anti-smoking, γ03 0.18 0.10 1.78 110 .077 Wellness, γ04 –0.19 0.17 –1.09 110 .278 For control slope, β1j Intercept 2, γ10 –1.47 0.07 –19.76 110 < .001 Fitness, γ11 0.01 0.07 0.08 110 .938 Nutrition, γ12 –0.25 0.08 –3.07 110 .003 Anti-smoking, γ13 –0.04 0.07 –0.60 110 .553 Wellness, γ14 0.19 0.11 1.74 110 .085 Note. Equations illustrating the model with grand mean centered level-2 predictors. 1 legitimacy of smoke-free programsij = β0j + β1j(Control Type) + rij. β0j = γ00§, β1j = γ10 + γ11(Fitnessj) + γ12(Nutritionj) + γ13(Anti-Smokingj) + γ14(Wellnessj) + u1j, § This intercept was treated as fixed because the variance in the intercept 1 was not significant. 2 legitimacy of fitness programsij = β0j + β1j(Control Type) + rij. β0j = γ00 + γ01(Fitnessj) + γ02(Nutritionj) + γ03(Anti-Smokingj) + γ04(Wellnessj) + u0j , β1j = γ10 + γ11(Fitnessj) + γ12(Nutritionj) + γ13(Anti-Smokingj) + γ14(Wellnessj) + u1j, 3 legitimacy of health risk appraisal programsij = β0j + β1j(Control Type) + rij. β0j = γ00 + γ01(Fitnessj) + γ02(Nutritionj) + γ03(Anti-Smokingj) + γ04(Wellnessj) + u0j , β1j = γ10 + γ11(Fitnessj) + γ12(Nutritionj) + γ13(Anti-Smokingj) + γ14(Wellnessj) + u1j, Int. Journal of Business Science and Applied Management / Business-and-Management.org 8 4.2 Smoke-Free Programs For individuals’ legitimacy ratings of smoke-free programs, control type was a significant predictor of the within-individual variance, coefficient = –1.42, SE = 0.09, t = –15.34, p < .001, indicating that legitimacy ratings decreased as the control type moved from low to high. The control type explained 61.50% of the within-individual variance. To clarify the effect of control type, paired ttests were conducted. Low control received higher legitimacy rating than moderate control, t (114) = 10.56, p < .001, η2 = .49. Moderate control received higher legitimacy ratings than high control, t (114) = 5.33, p < .001, η2 = .20. Means are reported in Table 1. The analysis showed that the level-1 intercept (i.e., individuals’ legitimacy ratings averaged across the three control types) did not vary significantly across individuals, variance = 0.48, χ2(114) = 114.91, p = .46, indicating no need for level-2 predictors. On the other hand, the level-1 slope (i.e., individual changes from low control to high) had a significant amount of variance across individuals, variance = 0.78, χ2(114) = 364.71, p < .001, indicating that the extent to which legitimacy rating decreased from low control to high was greater for some individuals than for others. As shown in Table 2, individual health orientation variables (level-2 predictors) were included in the analysis to explain between-individual variance in the extent to which legitimacy ratings changed from low control to high (i.e., the variance in the level-1 slope). Including these level-2 predictors explained 12.29% of the variance in the level-1 slope. Among the predictors, only anti-smoking orientation was a significant and positive predictor, whereas fitness, nutrition, and wellness orientations were not significant. This finding indicated that as individuals had higher anti-smoking orientations, the slope became less negative; individuals with higher anti-smoking orientations showed less decrease in their legitimacy ratings from low control to high. To put it differently, legitimacy rating decrease from low control to high was more pronounced among those with lower anti-smoking orientations. 4.3 Fitness programs For individuals’ legitimacy ratings of fitness program, the control type was a significant predictor of the within-individual variance, coefficient = –1.30, SE = 0.07, t = –18.97, p < .001, indicating that legitimacy ratings decreased as the control type moved from low to high. The control type explained 80.44% of the within-individual variance. To clarify the effect of control type, paired t-tests were conducted. Low control received higher legitimacy rating than moderate control, t (114) = 13.19, p < .001, η2 = .60. Moderate control received higher legitimacy rating than high control, t (114) = 12.43, p < .001, η2 = .58. The analysis showed that the level-1 intercept (i.e., individuals’ legitimacy ratings averaged across the three control types) varied significantly across individuals, variance = 1.33, χ2(114) = 448.32, p < .001, indicating the need for level-2 predictors to explain the variance in the level-1 intercept. The level-1 slope (i.e., individual changes from low control to high) had a significant amount of variance across individuals, variance = 0.30, χ2(114) = 255.15, p < .001, indicating that the extent to which legitimacy rating decreased from low control to high was greater for some individuals than for others. As shown in Table 2, individual health orientation variables (level-2 predictors) were included in the analysis to explain between-individual variance in the individual average legitimacy ratings (i.e., the variance in the level-1 intercept) and also in the extent to which legitimacy ratings changed from low control to high (i.e., the variance in the level-1 slope). Including these level-2 predictors explained 14.29% of the variance in the level-1 intercept and 8.31% of the variance in the level-1 slope. Among the predictors, only anti-smoking orientation was a significant and positive predictor of the level-1 intercept, whereas fitness, nutrition, and wellness orientations were not significant. This finding indicated that the higher an individual's anti-smoking orientation, the higher his or her average legitimacy ratings across the three types of control. Among the predictors, only nutrition was a significant and negative predictor of the level-1 slope. This finding indicated that the higher an individual's nutrition orientation, the more negative the slope became; individuals with higher nutrition orientations showed greater decrease in their legitimacy ratings from low control to high. Put differently, the decrease in perceived legitimacy from low control to high was less pronounced among those with lower nutrition orientations. 4.4 Health risk appraisal programs Finally, for individuals’ perceived legitimacy scores of health risk appraisal program, the control type was a significant predictor of the within-individual variance, coefficient = –1.47, SE = 0.08, t = – 19.17, p < .001, again indicating that legitimacy ratings decreased as the control type moved from low control to high. The control type explained 78.27% of the within-individual variance. To clarify the effect of control type, paired t-tests were conducted. Low control received higher legitimacy rating than moderate control, t (114) = 15.98, p < .001, η2 = .69. Moderate control received higher legitimacy Hannah Klautke and Hee Sun Park 15 Kouvonen, A., Vahtera, J., Elovainio, M., Cox, S., Cox, T., Linna, A., et al. (2007). Organizational justice and smoking: The Finnish public sector study. Journal of Epidemiology and Community Health, 61, 427-433. Mayer, J. P. (1991). Worksite health promotion: Needs, approaches, and effectiveness. In J. P. Mayer & J. K. David (Eds.), Worksite health promotion: Needs, approaches, and effectiveness (pp. 112151). Michigan Department of Public Health: Health Monograph Series No. 4. McGregor, J. (2007, November 5). Sweat more, pay less. Business Week, 4057, 76. Meritain Health. (2007). Healthy Merits. [Brochure]. Retrieved January 10, 2008, from http://www.meritain.com/Home/BenefitSolutions/HealthyMerits O’Donnell, M. (1991). The third and fourth steps in the evolution of worksite health promotion. In: In J. P. Mayer & J. David (Eds.), Worksite health promotion: Needs, approaches, and effectiveness (pp. i-ix). Michigan Department of Public Health: Health Monograph Series No. 4. Osinubi, O. Y. O., Barbeau, E. Williams, J. M, & Sorensen, G. (2005). Curbing the tobacco epidemic: The role of workplace tobacco control policies and programs. In J. H. Owing (Ed.), Focus on smoking and health research (pp. 1-75). New York: Nova. Park, H. S., Dalsey, E., Yun, D., Guan, X., & Cherry, J. (2008, May). Employee's responses to implementation of smoke-free workforce policy. Paper presented at the annual convention of International Communication Association, Montreal, Canada. Pfeffer, J. (1998). The human equation: Building profits by putting people first. Boston, MA: Harvard Business School Press. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd Edition). Thousand Oaks, CA: Sage. Rose, B. (2008, April 23). Lies about smoking burn workers. Tribune Business News. Retrieved July 10 from http://p2047ezproxy.msu.edu.proxy1.cl.msu.edu/login?url=http://proquest.umi.com.proxy1.cl.msu.edu/pqdwe b?did=1467160631&sid=1&Fmt=3&clientId=3552&RQT=309&VName=PQD Sofian, N. S., McAfee, T., Doctor, J., & Carson, D. (1994). Tobacco control and cessation. In M. P. O'Donnell & J. S. Harris (Eds.), Health promotion in the workplace (pp. 343-366). New York: Delmar. Truxillo, F. M., Baier, T. N., & Paronto, M. E. (2002). Reactions to organization alcohol testing and treatment programs. Journal of Business and Psychology, 17, 31-45. Tyler, T. R. (2006). Psychological perspectives on legitimacy and legitimation. Annual Review of Psychology, 57, 375-400. Wallston, K., & Armstrong, C. (1994). Theoretically-based strategies for health behavior change. In M. P. O'Donnell & J. S. Harris (Eds.), Health promotion in the workplace (pp. 182-203). New York: Delmar. Zoller, H. M. (2003). Working out: Managerialism in Workplace Health promotion. Management Communication Quarterly, 17, 171-205.