NICHD Early Child Care Research Network

In this paper, we use data from the NICHD Study of Early Child Care, a prospective longitudinal study of over 1300 children, to examine this question. This paper is based on data from the 321 families in which the mother returned to work by six-months post-partum, and remained employed through 36-months post-partum. Participants in the NICHD Study of Early Child Care were recruited from hospitals located in or near Little Rock, AR; Orange County, CA; Lawrence, KS; Boston, MA; Philadelphia, PA; Pittsburgh, PA; Charlottesville, VA; Morganton, NC; Seattle, WA; and Madison, WI. The recruited families included 24% ethnic minority children, 10% mothers without a high school education, and 14% single mothers. Of the enrolled mothers, 53% were planning to work full time, 23% part time, and 24% were not planning to be employed.

In this paper, we [1] examine the change over time in individual well-being following childbirth, among employed women, and [2] test a model of the relationship between job stress and individual well-being during the first three years post-partum. We posit that individual well-being will improve over time, and that this improvement is partially a function of the psychosocial risk and resilience factors. Specifically, the model posits that job stress is most strongly related to concurrent individual well-being, but also has a lagged effect on well-being. The model posits the following variables will serve as risk factors, and will exacerbate the relationship between job stress and individual well-being: working more hours; working evenings, nights or other non-day shifts; not working at home; having more children in the household; the birth of an additional child after the study child; returning to work sooner post-partum; being younger at the time of the first birth. The model further posits that the following variables will serve as resilience factors, ameliorating the relationship between job stress and individual well-being: higher maternal education; greater parental coping skills; having been employed during pregnancy; living with a husband or partner; having other adults in the household; reporting greater emotional support; having flexible child care arrangements, and having better quality child care.

CORRESPONDING AUTHOR: Nancy L. Marshall, Center for Research on Women, Wellesley College, Wellesley, MA 02481, USA

James T. Bond and Ellen Galinsky, Families and Work Institute and Jennifer Swanberg, National Center for Addiction and Substance Abuse at Columbia University

Findings from the Families and Work Institute's 1997 National Study of the Changing Workforce suggest that job-related stress mediates the effects of job and workplace characteristics on employee attitudes, productivity, and retention. Specifically, jobs and workplaces that are less stressful for employees appear to be more conducive to greater job satisfaction, higher loyalty and commitment, more productivity on the job, and lower turnover. A win-win situation for employers and employees.

The 1997 study is part of an ongoing research program that surveys representative cross-sectional samples of the U.S. labor force every five years. The 1997 sample includes 2,877 wage and salaried workers who are the focus of this paper.

Previously published analyses of 1997 data have examined the relationships between job/workplace characteristics and various on-the-job outcomes, but have not included job-related stress in the analytic model. Here job-related stress is evaluated as a possible mediating variable that could help to explain strong associations between various job and workplace factors and on-the-job outcomes of great interest to employers.

Multiple indices of job demands, job quality, and workplace support are considered. The findings confirm expectations that as job demands increased so would reported levels of job-related stress. The findings also suggest, however, that the negative impact of demanding jobs is mitigated by having jobs of higher quality and workplaces that are more supportive of personal and family needs. Analyses also identify characteristics of life off the job that may exacerbate job-related stress.

CORRESPONDING AUTHOR: James T. Bond, Families and Work Institute, 330 Seventh Avenue, New York, NY 10001

Jennie Ward Robinson, Ph.D., National Institute for Occupational Safety and Health, and Sharon K. Hall, Ph.D., University of Houston-Clear Lake

The surge in working women, particularly those parenting young and school-aged children, underpins empirical research targeting the work-family phenomenon (Auerbach, 1994). Today, a diversity of working women comprise a significant component of the workforce, leading to concerns of role (s) execution, benefits/costs of employment, and the familial/organizational implications. Preliminary findings note a complex interplay of factors underpinning the work/family phenomenon (Zedeck, 1992). However, challenges of validity and generalizability of findings are linked to inappropriateness of research designs, and homogeneity of the participant population.

This absence remains a concern (Nkomo & Cox,1990), and investigations uncover assertions of "professional risk" (Alderfer,1990), and the unavailability of population (Nkomo & Cox, 1990). This has resulted in perceptions of monolithic experiences, shared outcomes, and similar coping/responsive mechanisms for all working women. Yet we know that workplace stressors are experienced differentially among women (e.g. harassment-gender/racial), and that perceptions and responses to organizational stressors differ from those of their white counterparts (Bell, 1990). However, a dearth of literature exists with respect to the work-family phenomenon cross-culturally.

Guided by a qualitative design, this study contributes to the minority perspective within the work-family dialogue. Additionally, it challenges the monolithic perspective of workplace experiences and responsive mechanisms across racial groups. Next, the research presents an integrated picture of the work-family dialogue by examining and capturing the subjective experiences, perceptions, and outcomes of working women across cultural groups.

With respect to the minority participants, the complexity of the work-family phenomenon is revealed as issues of race, ethnicity and culture are juxtaposed with the demands and resolutions addressing their work and family lives. This group highlighted the complexity of these issues when experiences of ethnic and racial stereotypes are confronted relative to work ethics, role performance and job qualifications. The findings also revealed negative outcomes associated with factors of organizational culture/climate with respect to career and professional mobility coupled with the absence of mentors or role-models. Finally, this research provides a framework for exploring work-family issues cross-cultura, and discusses the implications for integrative cross-culturally research.

CORRESPONDING AUTHOR: Jennie Ward Robinson, Ph.D., National Institute for Occupational Safety and Health, 4676 Columbia Parkway, Cincinnati, Ohio 45226

Robert T. Brennan, Ed.D., Graduate School of Education, Harvard University, and (*) Rosalind C. Barnett, Ph.D., Murray Research Center, Radcliffe College

Dual-earner couples constitute roughly 66% of the workforce and women earn as much as or more than their husbands in about 26% of these couples. Little systematic attention has been paid to the effects of relative earnings on marital quality. Most theories suggest that there will be negative effects in the marriage for both spouses when she earns more, although these relationships might depend on gender, gender-role ideology (GRI) and identification with the provider role (i.e., salary affect).

We tested two hypotheses: (1) when wives earn more than their husbands, both partners’ evaluations of their marriage decrease; and (2) the relationship between salary differences within couples and marital-role quality (MRQ) varies with GRI and/or with salary affect. These relationships have not been tested before because no extant studies include measures of both partners’ salaries, GRI, and salary affect over time.

We interviewed a random sample of 287 full-time employed dual-earner couples at three time points. We conceptualized salary as a couple predictor with four components: (1) stable mean (average) couple salary; (2) change over time in mean couple salary; (3) stable between-partner salary gap; and (4) change over time in salary gap.

For men, the strongest relationships between salary and MRQ emerged after other variables were introduced into the model. The larger the stable salary gap favoring the wife, the lower the husband’s MRQ in couples whose stable GRI was traditional. Conversely, in these same couples, the larger the salary gap favoring the husband, the higher his MRQ. In contrast, in couples whose stable GRI was more egalitarian, the larger the salary gap favoring the husband, the lower his MRQ. Interestingly, for women, earning more or less than their husbands had no significant impact on their MRQ, even when considered in the context of the couple’s GRI.

With respect to the relationship between change components of salary and change over time in MRQ, for men who earn more over time relative to their wives, MRQ increases, especially if the men derive increasing rewards from their own salaries. In contrast, among married men for whom earnings are not so prized, a widening salary gap is associated with decreasing MRQ. The second time-varying component, change in mean couple salary, was not significantly related to change in MRQ for men or women.

As women increasingly work full-time/full-year, and men’s earnings continue to stagnate or decline, women’s earnings are beginning to catch up to men’s, albeit slowly. In this climate, adherence to the good-provider mandate of the traditional male role seems to increase the vulnerability of men in full-time employed dual-earner couples.

CORRESPONDING AUTHOR: Robert Brennan, Harvard University Graduate School of Education, Gutman Library, Appian Way, Cambridge, MA 02138-3704, USA

Chair: Satoru Shima, MD, PhD, Tokyo Keizai University; Presenters: Rumii Kurabayashi, National Institute of Industrial Health, Norio Kuroki, Toho University, MD, PhD, Koji Takahashi, MBA, Mie University and Minoru Arai, MD, PhD, Juntendo University; Discussant: Takashi Haratani, National Institute of Industrial Health

As reported by a number of studies in Japan, more and more workers have been suffering from stress at work. The causes of the increasing stress at work are diverse, such as restructuring, leaner organization, globalization and new technologies. This high level of stress among employees may considerably affect the level of employees’ health, leading to increasing numbers of employees with stress related diseases, especially psychosomatic disorders. In response to these findings, the Ministry of Labor has established a research group, focusing on stress management programs. In this symposium, some of the results will be presented.

The first speaker will address the current stress management system along with related laws in Japan. This will reflect the basic stress management framework with its various advantages as well as limitations. The second speaker will report the results of a survey on employees with mental disorders focusing on mental health care facilities. What system would be desirable to deal with mental disorders in the workplace will be clarified. The third speaker will focus on commitment and stress management from the point of view of organizational psychology. This strategy may create healthier organization with more productivity and healthier employees. The last speaker will talk about the utilization of intranet/internet in stress management.

CORRESPONDING AUTHOR: Satoru Shima, Department of Management, Tokyo Keizai University, 1-7-34 Minami-cho, Kokubunji, Tokyo, 185-8502 Japan

Lumie Kurabayashi, M.D., Takashi Haratani, Ph.D., and Akinori Nakata, Ph.D., National Institute of Industrial Health, Ministry of Labor, Japan

According to the Industrial Safety and Health Law in Japan, in a company with more than fifty workers, the employer must nominate a doctor as an occupational physician (OP). They are supposed to plan and assess various aspects of health care. They put medical health examination into practice. They do health counseling based on the examination and referral to hospitals if necessary. Also, they are in charge of risk assessments in working environment.

One of the new health care guidelines, Total Health Promotion Plan (THP), was published in 1988 by the Ministry of Labor (MOL). In the THP activities, OPs give suitable physical training programs and advice for each worker to promote his/her health based on his/her health examination. One of the other guidelines to make "comfortable working place" to reduce job stressors is shown by MOL in 1992. To promote the workers' health in these companies, MOL has established the Occupational Health Promotion Center and the Regional Occupational Health Center. To follow these guidelines and to make use of these services, it is expected to promote workers' health against job stress.

CORRESPONDING AUTHOR: Lumie Kurabayashi, M.D., Division of Work Stress Control, National Institute of Industrial Health, 6-21-1, Nagao, Tama-Ku, Kawasaki 214-8585, JAPAN

Norio Kuroki, MD, PhD, Sakura Hospital, Toho University, Minoru Arai, MD, PhD, Juntendo University, Satoru Shima, Tokyo Keizai University

Although the quality of mental health care services in public have got higher, mental health services are still quite insufficient in the workplace. Mental health care in the workplace is very important when we consider that about the Japanese population is employed. Although physical health checks for worker are regularly carried out, mental health checks are generally insufficient.

The subjects of this study consist of 504 patients who are being treated at 5 company owned medical facilities and at 8 public or private facilities. The subjects were classifications by the facilities as follows: 1) patients being treated at company facilities and those being treated at other facilities, 2) patients who classified by mental diseases, 3) patients were classified by the size of their company.

The results showed that it was much easier for patients to use company facilities to deal with their mental health problems than for them to use other facilities. It was less time consuming and more convenient to use company facilities. Also, company doctors could consult with an employee’s boss and decide which job would be appropriate for the patient. In the case of schizophrenic patients, however, it was found that being treated at a hospital or a clinic near their home is better for them. For patients with mood disorders, having company doctors and nurses would be better because they can get support from professionals easily.

CORRESPONDING AUTHOR: Norio Kuroki, MD, PhD, Department of Psychiatry, Sakura Hospital, Toho University, 564-1 Simosizu, Sakura-shi, Tiba 285 Japan

Koji Takahashi, M.B.A., Mie University and Naotaka Watanabe, Ph.D., Keio University

Past research suggested that individual internal resources and traits(e.g., cognitive style, coping strategy, personality) work well as their stress tolerance factors. However, organizational culture/climate as a form of psychological characteristic of organization(PCO) and its effect on stress prevention have been seldom mentioned.

In this study, we attempted (1) to describe PCOs seemingly similar but different two Japanese companies, and (2) to estimate employees' stress tolerance by focusing on organizational commitment (affective, continuance, and normative).

We conducted a questionnaire survey for 849 middle-management of two Japanese electronics companies (Company A and B). Results revealed that Company A had a PCO in which employees were well socialized to their company, well continuously committed (very sensitive to side-bet and sunk-cost recognition thus committed), more experienced as protege in career function in the past, well working as mentor both in career and psycho-social functions currently, and more depressed. In Company B, on the other hand, employees were well socialized to their company, well continuously and normatively committed, less experienced as protege in career function previously, less effective mentors either in career and psycho-social functions presently, and less depressed.

CORRESPONDING AUTHOR: Koji Takahashi, M.B.A., Assistant Professor of Human Resource Management, Faculty of Humanities and Social Sciences, Mie University, 1515 Kamihama-cho, Tsu, Mie 514-8507, JAPAN

Minoru Arai, MD, Daisuke Mori, MD, Tetsu Kawamura, MD, Hideo Fumimoto, MD, Masagi Shimazaki, MD, Usuke Nozaki, MD, Reiichi Inoue, MD, Department of Psychiatry, Juntendo University School of Medicine

Objectives and method: Intranet Settings are available in many companies in recent years. We planned to screen employees with mentally ill health on the net instead of using paper-based screening test. The items of the screening test are as follows; easy fatigability, hypochondriasis, irritability, anxiety, isolation, thought inhibition, suicide ideation, sleep disorder, loss of appetite, and so forth. The answers were evaluated according to the severity of stress state and mental health state, we advised on the net how to cope with stresses and to whom they could consult.

Results: Of 80,000 employees, about 20,000 individuals completed the screening test during six months. Sex ratio of the subject was 77:23(male: female). The rate of the highest stressed subjects was 4.7% and that of the second highest subjects was 34.5%. Numbers of clients of the clinics in the company increased considerably after the introduction of intranet screening.

Conclusion: Intranet screening test may be beneficial to employees in terms of self-checking of mental health state, seeking consultation sites, and learning how to cope with stresses. Further elaboration of intranet screening test of mental health must be required.

CORRESPONDING AUTHOR: Minoru Arai, MD, PhD, Department of Psychiatry, Juntendo University School of Medicine, 2-1-1 Hongo Bunkyo-Ku Tokyo 113-8421 Japan

Chair: E. Scott Geller, Ph.D., Center for Applied Behavior Systems and Department of Psychology, Virginia Tech; Presenters: Charles Pettinger, MS, Joshua Williams, MS, Jason DePasquale, MS, Steven Clarke, MS, & Thomas Boyce, MS, Center for Applied Behavior Systems, Department of Psychology, Virginia Tech.

This symposium presents five intervention-relevant studies conducted as portions of two NIOSH-funded research grants. Each paper is firmly grounded in psychological theory and relevant for safety practitioners with regard to the application of behavior-based safety (BBS) to prevent workplace injuries.

The first paper discusses the differential impact of involving employees in BBS training and process implementation. Effects of various involvement manipulations on satisfaction with training, retention of key BBS concepts taught in training, and participation in an observation and feedback process are presented. The second paper reports data on the differential effects of specific vs. global feedback on the occurrence of safety-related behaviors. Specifically, behavioral feedback to workers in a beverage bottling company was given on specific behaviors or as a global "percent safe" score. Results are presented with regard to effects on process measures (i.e., changes in target behaviors) and outcome measures (i.e., impact on injuries). The third paper discusses the results of a two-year quantitative and qualitative assessment of industries who have used or are using BBS in their organization. Industries following a mandatory approach were compared with those using a voluntary approach. Trained research personnel conducted focus group meetings at 20 different companies and gave participants a culture survey to complete. Employee perceptions and organizational culture variables predicting success versus lack of success with BBS are presented. The fourth paper discusses an innovative self-management intervention that increased driving safety among short-haul truck drivers. Theory and techniques of safety self-management are presented. It is argued that safety self-management grounded in the principles of BBS may be the best approach to improve safety-related behavior among employees, like delivery drivers, who work in solitary jobs. The final paper brings technology to the field of BBS to improve driving safety. Specifically, data are presented on relationships between individual differences and various driving behaviors as measured by an intelligent transportation systems instrumented vehicle. The development of a reliable methodology to collect unobtrusive driving data are also presented, as well as applications of this technology and findings to transportation-oriented industries.

The chair of this symposium is E. Scott Geller, Ph.D., professor of psychology at Virginia Tech and director of the Center for Applied Behavior Systems. He is a recognized leader in the field of BBS and the principal investigator for NIOSH funded research.

CORRESPONDING AUTHOR: T. Boyce, MS, Center for Applied Behavior Systems, Dept. of Psychology, Virginia Tech, 5100 Derring Hall, Blacksburg, VA 24061-0436

Charles Pettinger, MS*, Jason DePasquale, MS, Thomas Boyce, MS, & Joshua Williams, MS., Center for Applied Behavior Systems, Dept. of Psychology, Virginia Tech

This research began by training volunteer "safety facilitators" from representative areas of a large manufacturing plant and then conducting facility-wide education/training sessions for the remaining employees (n=476). The format of the training sessions was manipulated to investigate the impact of employee participation during BBS education/training, but the materials for all sessions were held constant. The safety trainers in the Choice condition (n=230 on Shift 1) were instructed to ask questions to facilitate group discussion and involvement with workers. This included group exercises where workers presented their suggestions for their site’s BBS process. The same trainers in the Assigned conditions (n=246 on Shift 2) presented the safety material in a lecture format without asking questions or facilitating workers’ input.

Four-hour training sessions were held for 12 Choice groups and 14 Assigned groups, ranging in size from 7 to 30 individuals. To assess the impact of the two training approaches, three variables were measured – the amount of verbal participation directed at the trainers, participants’ reported satisfaction with the training, and the participants’ retention of safety training information. There were no significant differences between conditions regarding information retention, satisfaction, and involvement.

Many involvement manipulations were made to give the Shift 1 facilitators (n=8) as many opportunities to make key decisions in the BBS process as possible. For example, the Shift 1 safety facilitators selected: a) the initial safety-related behavior to observe plant-wide (hearing protection was chosen), b) the design and location of group feedback prompts, c) the initial plant-wide safety intervention, d) the design of a safety slogan contest, and e) the safety-related behaviors to observe in individual work areas. The choices made by the Shift 1 safety facilitators were assigned to the Shift 2 safety facilitators (n=6), thus limiting Shift 2 input into the BBS process. This extended the Choice vs. Assigned manipulations implemented during training.

For nine weeks, the safety facilitators of Shift 1 and Shift 2 made behavioral observations on hearing protection. The data were graphed and posted on a safety bulletin board located at the entrance to the production areas of the plant. Over the nine-week period, the Shift 1 safety facilitators made significantly more observations per facilitator per week than the Shift 2 facilitators. Also, Shift 1 safety facilitators conducted significantly more observations per person each week than the Shift 2 facilitators.

A time series tabulation of this plant’s injury statistics shows lost days due to injuries decreased markedly after the BBS training and observation/feedback programs were put in place (10.9 lost days per month prior to BBS versus 1.5 lost days per month after the initiation of BBS).

CORRESPONDING AUTHOR: C. Pettinger, Center for Applied Behavior Systems, Dept. of Psychology, 5100 Derring Hall, Virginia Tech, Blacksburg, VA 24061-0436

Joshua Williams, M.S.*, Steven Clarke, M.S., & E. Scott Geller, Ph.D.
Center for Applied Behavior Systems, Dept. of Psychology, Virginia Tech

According to the literature, behavior-based safety feedback (BBS) works whether given on specific behaviors or as an overall (global) score. However, the relative impact of global vs. specific feedback has not been addressed. The effectiveness of specific vs. global feedback was investigated at a soft-drink bottling plant in southwestern Virginia.

Participants were 40 front-line workers. Employees were observed twice a day (once per shift) by trained observers using a critical behavior checklist. For each behavioral category, the observers marked either "safe" or "at risk" for the target behaviors. Behavioral categories included: personal protective gear, lifting, fork truck driving, and general safety. Inter-rater agreement was calculated each week and was above 85% for all behaviors observed.

The research design included five phases (Baseline, 15 weeks; Awareness, 9 weeks; Feedback 1, six weeks; Feedback 2, five weeks; Withdrawal, 20 weeks). For the awareness intervention, employees were informed by their supervisors what behaviors were being observed. During the first feedback intervention, Shift 1 received global feedback (a graph depicting weekly changes in percent safe scores aggregated across behaviors). Shift 2 received specific graphical feedback for each behavioral category, with no global feedback. Then during the second feedback intervention these conditions were reversed. Feedback was provided in weekly meetings.

Results of a 2 Feedback (Global vs. Specific) x 5 Phase (Baseline vs. Awareness vs. Feedback 1 vs. Feedback 2 vs. Withdrawal) ANOVA indicated significantly lower safety performance for Shift 2 (60%) than Shift 1 (77%) during baseline. During the awareness phase safety performance for Shift 1 (79%) increased to the same level as Shift 2 (81%). During Feedback 1, global feedback (Shift 1, 89%) resulted in a 13% increase in safety performance over the awareness phase. There was no increase for specific feedback (82%). During Feedback 2, safety performance returned to awareness phase levels in the both the Global-Specific condition (Shift 1, 77%) and Specific-Global condition (Shift 2, 76%). During withdrawal there was no difference between the safety performance of Shift 1 and 2. The overall frequency of recordable injuries fell by more than 50% following the implementation of BBS observation and feedback.

CORRESPONDING AUTHOR: J. Williams, MS, Center for Applied Behavior Systems, Dept. of Psychology, Virginia Tech, 5100 Derring Hall, Blacksburg, VA 24061-0436