![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Division of Nutritional Sciences, Cornell University, Ithaca NY 14853
3To whom correspondence should be addressed. E-mail: kld12{at}cornell.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
KEY WORDS: • program effectiveness • low-income population • nutrition education • program management • community health educators
Effective promotion of healthy eating practices is critical to the health and development of children and the prevention of chronic diseases. Making improvements in established food habits is always challenging, and low-income populations face additional economic and psychological barriers that constrain their ability to change habitual behaviors (1–4). Effective strategies for promoting behavior change in high-risk, low-income populations are urgently needed (3). Efforts to enhance the effectiveness of nutrition education have focused primarily on curricula and the training of personnel (5), and now, after decades of experience and refinement, neither educational content nor lack of training is likely to be a major limiting factor. In contrast, little is known about the job characteristics, management, and motivational strategies that constitute the work context of front-line health and nutrition workers or how these factors influence program effectiveness (6). Research in business settings may not apply directly to the nutrition program context because human services organizations differ appreciably from profit-oriented organizations (7–10). Our research was designed to open the "black box" of nutrition program management and explore work context factors expected to influence program effectiveness.
This research was conducted within the Expanded Food and Nutrition Education Program (EFNEP),4 which is funded by the USDA and implemented by Cooperative Extension. EFNEP provides nutrition education services to low-income families with children (11), through individual or small group sessions designed to promote improved dietary habits, food resource management, and food preparation. EFNEP hires front-line paraprofessional Community Nutrition Educators (CNEs) from among the population served, on the premise that the life experiences of indigenous CNEs will enhance their rapport and credibility with the program audience (12,13). Cooperative Extension Educators, most of whom are professional nutritionists, manage EFNEP and provide supervision, training, and technical guidance to CNEs.
This research is part of a larger study exploring how work context relates to CNE work attitudes (e.g., job satisfaction) and program outcomes. Based on interviews conducted with EFNEP CNEs and supervisors in the first phase of the project, we developed a multidimensional construct we call the perceived work context of these workers. This includes CNEs’ perceptions of the value of the program, management and supervision, work relationships, workload, and pressure to graduate large numbers of participants (14). The objective of the research reported here was to examine CNEs’ work context in relation to program effectiveness, measured as participant behavior change.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
A survey of EFNEP CNEs and supervisors was conducted in all counties in New York State (NY) where EFNEP was actively implemented at the time of the survey (April–June 2001). The 4 sites in New York City (NYC) were subsequently excluded because of extreme outlying values on population and program size-related variables, resulting in a sample of 30 sites. All CNEs who had completed initial training and had their own caseloads (i.e., had been on the job 
2 mo or more) were eligible to participate, and 97% of those eligible agreed to participate (100 CNEs). All 30 supervisors completed the questionnaire.
Data on program characteristics (30 sites) and on nutrition behaviors reported pre- and postparticipation (by 6321 EFNEP participants) were abstracted from the fiscal year (FY) 2001 Evaluation and Reporting System (ERS) dataset used by EFNEP for program monitoring (15).
Survey data collection.
Separate questionnaires for CNEs and for supervisors were developed and pretested (14) to collect data on characteristics of CNEs, supervisors, and programs and on CNEs’ perceived work context (variables described below). CNEs (71%) completed the survey at in-service training sessions, with the rest submitting questionnaires by mail. All respondents gave written informed consent. The research was approved by the Cornell University Committee on Human Subjects and the state EFNEP Leader (J.S.D.). To preserve confidentiality, participating counties are not identified by name.
Perceived work context variables.
A multidimensional construct of perceived work context was developed through prior in-depth interviews with EFNEP personnel on perceptions of EFNEP, the CNE work role, and factors influencing effectiveness and work attitudes (14). Based on these interviews, we adapted an existing instrument to measure managerial practices (16) and developed new scales to measure the other components of CNEs’ perceived work context. These variables are described below, with the names used throughout this paper for each variable given in italics; the multi-item scales assessing work context are summarized in Table 1.
|
All items in these scales were scored on 5-point Likert-type scales with response options from strongly agree to strongly disagree; some items were worded negatively to avoid response set bias and were reverse-coded. Factor analysis followed by orthogonal rotation (17) was used to investigate the covariance of scale items and refine the scales based on factor loadings, intercorrelations, and analysis of internal reliability (14). For 3 scales, 1–2 items with low factor loadings were dropped (Table 1), resulting in simpler scales with no loss of internal reliability. Because there was no basis for differential weighting of items, simple summative scales were used to create all scaled variables (18,19). Most scales had Cronbach’s 
coefficients ranging from 0.71 to 0.92 (Table 1), indicating good internal reliability (18,19). Site-level perceived work context variables were created from the means of all CNEs in a site.
Program and county variables.
Supervisory structure and the presence of Food Stamp Nutrition Education (FSNE)6 were assessed in the survey. Data on numbers of participants/y at each site; mean number of lessons/participant; and program delivery methods (individual or group instruction) were abstracted from the ERS dataset (15). A continuous variable created to represent the proportion of participants in a site who received individual instruction included a small number of participants (<5%) who received both individual and small group instruction. Demographic and socioeconomic characteristics of counties included county population, population density, Supplementary Feeding Program for Women Infants and Children (WIC)-eligible population, median household income, and percentage of children living under poverty (20). The WIC-eligible population size is a proxy for the EFNEP-eligible population size because both programs use the same income criteria.
CNE and supervisor variables.
Background characteristics of CNEs included age, gender, ethnicity, having ever participated in programs such as EFNEP, WIC, or Food Stamps, and years of postsecondary education. Variables describing CNEs’ work roles included tenure in current position and categories of hourly wage and work hours/wk. Supervisor characteristics included age, gender, ethnicity, years of postsecondary education, tenure in current position and with Extension, and categories of work hours/wk and percentage of work time spent on EFNEP.
Behavior change variables.
The dependent variable was behavior change, i.e., the difference between scores on behavioral items administered to all EFNEP participants at the beginning and end of their participation in the program. Our behavior change scale was based on the difference score for 6 self-reported items from a federally mandated 10-item EFNEP Behavior Checklist (15). This checklist includes items such as "When deciding what to feed your family, how often do you think about healthy food choices?" Participant responses represent increasing frequency of the behavior, from 0 (do not do) to 5 (almost always do). Previous factor analysis (21), confirmed in this study, revealed that 6 of 10 items in the Checklist clustered as a single factor representing food choice behaviors related to diet quality and management of food resources.7 These 6 items were the frequencies of planning meals ahead of time, comparing prices, shopping with a grocery list, thinking about healthy food choices when deciding what to feed the family, using "Nutrition Facts" on food labels to make food choices, and having children eat something within 2 h of waking up in the morning. These items all had similar factor loadings and were used to create an unweighted 30-point summative scale (18,19), converted to a 100-point scale for ease of interpretation. The dependent variable modeled in the regression analysis was the mean change in this scale between entry and graduation.
Many participants scored highly on 1 or more questions at program entry and therefore could not be expected to improve substantially on these behaviors. To account for this ceiling effect, we controlled for entry score in our analyses by using a potential for change variable. Potential for change was calculated as the maximum possible behavior change score (100 points) minus the entry score, i.e., participants with a low entry score had high potential for change.
Statistical analysis.
We conducted site-level analyses of behavior change in relation to CNE perceived work context and other program, site, and staff characteristics, using Stata 7.0 (17). Recognizing that the dependent variable, behavior change (B), was conditional on the potential for change (Po), we included Po as an independent control variable in the statistical analyses. Initially, we estimated the associations of all independent variables with B using partial correlation coefficients controlling for Po. Log transformation was used as needed to correct for skewness in variables such as population. Correlations with a P-value < 0.05 were considered significant. Significant independent variables were then included in multiple regression analyses using the following multiplicative model:
 | (1) |
where Xi = independent variables. The multiplicative model allowed testing of B in proportion to Po (22). Use of logarithmic transforms of variables on both sides of the equation creates a model (Eq. 2) that is additive and therefore amenable to usual ordinary least squares (OLS) analyses. For this reason, the other independent variables were also log transformed, except for individual instruction because transformation reduced its association with behavior change.
 | (2) |
To interpret the regression results, the additive equation was back transformed to correspond to Eq. 1 and each independent variable’s association with behavior change was plotted, holding all other variables constant at their mean values.
After the main effects of perceived work context, supervisory, and program characteristics on behavior change were analyzed in the regression, we examined whether any of the county or CNE characteristics confounded the results or were effect modifiers. Significance for main effects and effect modification were set at P < 0.05 and P < 0.10, respectively. All models were examined for influential outliers or violations of assumptions underlying linear OLS analysis.
Collinearity was addressed in 2 ways. Collinear variables thought to represent a similar underlying construct were combined to form a single composite variable. This was the case for 5 MPS categories that were significantly correlated with behavior change and highly collinear. A summative composite variable was created from these 5 variables. Networking behavior, the 6th MPS category correlated with behavior change, was not included in the composite variable because networking was neither conceptually nor statistically as closely related to the other 5 categories, and its inclusion reduced the internal reliability of the scale. Networking was tested separately in the regression models. When variables were collinear but represented distinct constructs, it was not appropriate to create composites. This was the case for the mean number of lessons/participant and the proportion of participants reached by individual instruction, for which we tested each variable in separate models.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
There was considerable variation across sites in program and county characteristics (Table 2). On average, about half of the program participants received individual instruction, but program delivery method varied from almost 100% individual instruction in some (mostly rural) counties to 100% group instruction in others. Mean behavior change score among all graduates in a site was an increase of > 13 points, ranging from 3 to 41 points, and potential for change ranged from 27 to 57 points/site.
|
Most CNEs were female, white, and had at least 1 y of postsecondary education (Table 3). They also had substantial life experience and tenure in the position, and over one third had participated in food or welfare programs themselves, reflecting EFNEP efforts to hire CNEs from among the population served (Table 3).
|
4 out of a possible 5. Relationships within Extension offices (outside the nutrition unit) were viewed somewhat less positively. The 2 factors that assessed undesirable factors, pressure for numbers and perceived workload, received the lowest scores (65% of maximum possible) and had the greatest variability across sites. Results of partial correlation analysis.
Taking potential for change into account, behavior change was significantly positively correlated with 2 aspects of perceived work context, i.e., value and managerial practices (Table 4). Six of 12 MPS categories were significantly correlated with behavior change, as shown in Table 4. None of the other work context or background characteristics of CNEs or supervisors was significantly associated with behavior change. Among program and site characteristics, behavior change was significantly positively correlated only with the proportion of participants receiving individual instruction and the mean number of lessons/participants at a site (Table 4). Behavior change was negatively related to the number of graduates/worker, an indicator of program output, but not significantly related to indicators of program size such as total number of participants/y in a site.
|
In the final multiple regression model controlling for potential for change (Table 5), greater behavior change was reported for sites in which a higher percentage of participants received individual instruction and in which CNEs reported more positive perceptions of the value of the program and of a composite of 5 managerial practices (planning, monitoring, problem-solving, motivating, and clarifying).
|
Figure 1allows visual comparison of the slopes of each of the independent variables in Table 5 regressed on behavior change, holding constant all other variables in the model. Across the 20th–80th percentile range, individual instruction accounted for an increase of 8 points in the behavior change score. The plots of the 2 work context variables, perceived value of the program and managerial practices, were similar to each other and accounted for increases of 3–4 points of behavior change in this range. The slope for potential for change was slightly greater than these, but less than the slope of individual instruction.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The importance of CNEs’ perceived value of the program is consistent with our earlier analyses showing that a strong belief in the value of EFNEP was critical to CNEs’ motivation and job attitudes (14). Our value scale is specific to EFNEP but appears to capture a construct closely related to the more general construct of job "meaningfulness," central to models of both intrinsic work motivation and employee empowerment (23–26). CNEs’ perceptions of value reflect a good fit between the goals and approach of the program, the responsibilities and rewards of the job, and the motivations of paraprofessional CNEs.
The positive association of behavior change score with perceived managerial practices suggests that, as expected, aspects of leadership that differ across EFNEP sites can influence program success. Planning, monitoring, problem-solving, motivating, and clarifying of roles and objectives were the most important managerial practices in our regression equation, and are consistent with key aspects of supervision described and appreciated by CNEs (14). Our use of CNEs’ perceptions of managerial practices was based on previous evidence that, compared with managers’ self-reports, subordinates’ ratings of managers are stronger predictors of independent measures of managerial effectiveness (27). Our findings on managerial practices are notable because we are not aware of previous research linking the MPS categories to any measures of program effectiveness (e.g., behavior change). These MPS categories were found to be related to independent criteria of managerial practices (28) and to managerial effectiveness (27). However, program effectiveness is a more distal outcome than managerial effectiveness, and one more likely to be of interest to organizations that fund and implement nutrition programs.
Other aspects of work context expected to be relevant to program effectiveness did not explain additional variation in behavior change. Some, such as work relationships and sense of having a voice, were significantly related to job satisfaction in other analyses of our data (14). These differential findings suggest that the influence of work context factors varies depending on whether the outcome of interest is staff attitudes or program outcomes. Previous research showed that the connection between job satisfaction and performance is complex (29,30) and that interventions that improve employee attitudes have varying effects on productivity (31).
The positive association of individual instruction with greater behavior change among EFNEP participants is in line with the results of previous studies (21,32) and with CNEs’ views that group lessons provide fewer opportunities to develop supportive relationships and tailor lessons to particular needs (14,32). A trend toward group methods occurred within EFNEP in response to conditions such as decreased availability of participants at home since the advent of welfare reform and the need to increase efficiency of service provision within funding constraints (21,32). Use of group methods often coincides with reductions in the number of lessons participants receive, but our study could not examine their relative contributions to behavior change because of collinearity. Group methods are used more extensively in urban vs. rural sites in EFNEP NY, but regression analyses indicated that population variables did not account for the relation of individual instruction with behavior change. Although our results suggest that group nutrition education may have less effect than individual instruction on some reported behaviors, definitive conclusions require studies that control the number of lessons. Experiences in EFNEP sites in NYC suggest that group methods in urban settings can achieve good outcomes if CNEs are given adequate training and support to become effective group educators and if participants receive sufficient numbers of lessons (14). Other variables likely to influence behavioral outcomes of nutrition education that could not be measured in this study include the quality of educational sessions and the life situations of program participants.
A statistical strength of this study was the use of a multiplicative regression model of behavior change in proportion to the potential for change, taking into account the ceiling effect limiting behavior change among participants reporting positive behaviors at program entry. The importance of controlling for potential for change was confirmed by our finding of a strong relation with behavior change, e.g., less behavior change occurred when highly positive behaviors were reported at program entry.
We note several limitations to this study related to the validity of the outcome measure, the generalizability of results, and causal inferences. The Behavior Checklist was completed by participants and forms were collected by CNEs delivering the program, creating 2 possible sources of bias. If CNEs who valued the program highly were to overreport behavior change, this bias could inflate the association between value and behavior change. This is unlikely because in other analyses, we found that the CNE characteristics that were determinants of value (CNEs’ educational level and personality characteristics) were not significantly related to behavior change. It is also possible that participants overreported behavior change. However, CNEs contend that comparing pre- and post-test reports of behavior change tends to underestimate program benefit because participants are more honest and accurate in their self-reports at program completion when they have developed rapport with CNEs and understand what the practices described in the Behavior Checklist actually entail. Previous research demonstrated this problem of underestimation when behavior change is assessed using the pre- and post-test self-report method (33). More importantly, the Behavior Checklist does not tap the full range of benefits provided by a program that is locally tailored to participant interests and need. Many participants report benefits not captured by this program monitoring tool (14,34–36).
Overall, the limitations of this tool suggest that the behavior change scale may underestimate program effectiveness. Lack of reliability would attenuate the associations found; thus, the actual associations with work context may be stronger or more numerous than those detected in this study. However, our results would be strengthened if the behavior change data were validated against indicators of dietary intake or other objective criteria, and future research on program quality would benefit from the development of better outcome measures. This is important in light of the growing emphasis on "outcome-based evaluation" to guide decision-making and support of programs in health and human services (37).
Because our goal was to understand the influence of work context, not of program type, we included multiple sites of a single program in a single state. Although appropriate for our purposes, this sampling approach does not ensure generalizability of the results, and research in additional settings is warranted. However, our findings have important implications for other nutrition programs, despite variation in program type and implementation. Many nutrition programs worldwide employ community-based workers to promote behavior change among low-income populations, and although program value might be defined somewhat differently in other contexts, it is plausible that these workers’ views of value affect their work attitudes and performance. We recommend greater attention to front-line workers’ perceptions of program value as an indicator of staff attitudes relevant to program success. Greater attention to the development of relevant managerial skills among supervisors of nutrition programs is also warranted because we expect that supportive supervision combined with monitoring and problem-solving techniques will be beneficial in all nutrition program settings. Although this study identified 5 aspects of managerial behavior as most important, it is advisable to begin broadly when assessing management practices in different program contexts.
Also, to guide future efforts to improve nutrition programs, it is essential to implement intervention research that tests the effect of changes in work context on nutrition program outcomes. Cross-sectional studies such as this one cannot establish causality.
In conclusion, work context as experienced by front-line nutrition educators can have important implications for the effectiveness of nutrition programs. Nutrition workers’ perceived value of the program is an influential aspect of the work context that has not received explicit attention from researchers or program planners. Managerial practices, as perceived by front-line educators, are also related to program outcomes, suggesting that improvements in this area could enhance nutrition program success. We hope that our results and the methods developed for this study will facilitate an expansion of assessments of nutrition program success to include the influences of management, job characteristics, and program characteristics on the performance of nutrition educators. Work context factors should be considered in the design, implementation, and evaluation of nutrition programs.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported in part by a grant from the U.S. Department of Agriculture, Cooperative State Research, Education, and Extensive Service through Cornell University Agricultural Experiment Station federal formula funds, Project No. NYC-199411. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture. Additional support was received from the Division of Nutritional Sciences, Cornell University.
4 Abbreviations used: CNE, Community Nutrition Educator; EFNEP, Expanded Food and Nutrition Education Program; ERS, Evaluation and Reporting System; FSNE, Food Stamp Nutrition Education; FY, fiscal year; MPS, Managerial Practices Survey; NYC, New York City; NY, New York State; OLS, ordinary least squares; value, perceived value of the program; WIC, Supplementary Feeding Program for Women, Infants and Children.
5 The value scale items were: 1) I really think EFNEP matters. I see people improving their lives with this program; 2) I believe that, in the long run, EFNEP helps people improve their health and well being; and 3) I see EFNEP participants take real pride in learning new skills and it helps them develop the self-confidence to go on to do other things.
6 FSNE is similar to and often implemented concurrently with EFNEP, but targets people receiving Food Stamps. Its presence could affect program resources, staffing, and implementation strategies.
7 The other 4 items in the Behavior Checklist were not included in the scale because 1) they were conceptually different (1 item on food security, 2 on food safety) or poorly worded (1 item on adding salt to food) and 2) in factor analysis, they did not load strongly with the 6 items selected.
Manuscript received 29 January 2005. Initial review completed 3 March 2005. Revision accepted 7 July 2005.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Lindbladh, E. & Lyttkens, C. H. (2002) Habit versus choice: the process of decision-making in health-related behaviour. Soc. Sci. Med. 55:451-465.
2. Shannon, J., Kirkley, B., Ammerman, A., Keyserling, T., Kelsey, K., DeVellis, R. & Simpson, R. J. (1997) Self-efficacy as a predictor of dietary change in a low-socioeconomic-status southern adult population. Health Educ. Behav. 24:357-368.
3. Singleton, J. C. (1994) Nutrition and health education for limited income, high-risk groups: implications for nutrition educators. J. Nutr. Educ. 26:153-155.
4. Drewnowski, A. & Spector, S. (2004) Poverty and obesity: The role of energy density and energy costs. Am. J. Clin. Nutr. 79:6-16.
5. Contento, I., Balch, G., Bronner, Y., Lytle, L., Maloney, S., Olson, C. & Swadener, S. S. (1995) The effectiveness of nutrition education and implications for nutrition education policy, programs, and research: a review of research. J. Nutr. Educ. 27:279-422.
6. Hongoro, C. & McPake, B. (2004) How to bridge the gap in human resources for health. Lancet 364:1451-1456.[Medline]
7. Drucker, P. (1990) How to bridge the gap in human resources for health. Managing the Non-Profit Organization: Principles and Practices Harper Collins New York, NY.
8. Scott, W. R. (1992) How to bridge the gap in human resources for health. Organizations: Rational, Natural, and Open Systems Prentice Hall Englewood Cliffs, NJ.
9. Hasenfeld, Y. (1992) The nature of human service organizations. Hasenfeld, Y. eds. Human Services as Complex Organizations :3-23 Sage Newbury Park, CA.
10. D’Aunno, T. (1992) The effectiveness of human service organizations. Hasenfeld, Y. eds. Human Services as Complex Organizations :341-361 Sage Newbury Park, CA.
11. New York State Cornell Cooperative Extension (1998) The effectiveness of human service organizations. EFNEP Guidelines Cornell Cooperative Extension Ithaca, NY.
12. Bremner, B., Campbell, C. C. & Sobal, J. (1994) Comparison of the beliefs and practices of EFNEP clients with staff perceptions of clients. J. Nutr. Educ. 26:123-130.
13. Randall, M. J., Brink, M. S. & Joy, A. B. (1989) EFNEP: an investment in America’s future. J. Nutr. Educ. 21:276-279.
14. Dickin, K. L. (2003) EFNEP: an investment in America’s future. The Work Context of Community Nutrition Educators: Relevance to Work Attitudes and Program Outcomes. Doctoral thesis Cornell University Ithaca, NY.
15. ERS4 Evaluation and Reporting System [Online] () EFNEP: an investment in America’s future. USDA. http://www.csrees.usda.gov/nea/food/efnep/ers/ers4.html [accessed May 19, 2005].
16. Yukl, G. (1988) EFNEP: an investment in America’s future. Development and Validation of the Managerial Practices Questionnaire State University of New York Albany, NY.
17. StataCorp. (2001) EFNEP: an investment in America’s future. Stata Statistical Software: Release 7.0 Stata Corporation College Station, TX.
18. DeVellis, R. F. (1991) EFNEP: an investment in America’s future. Scale Development: Theory and Applications Sage Newbury Park, CA.
19. Nunnally, J. C. (1978) EFNEP: an investment in America’s future. Psychometric Theory McGraw-Hill New York, NY.
20. State & County QuickFacts [Online] () EFNEP: an investment in America’s future. US Census Bureau. http://quickfacts.census.gov/qfd/ [accessed October 14, 2002].
21. Dollahite, J. & Scott-Pierce, M. (2003) EFNEP: an investment in America’s future. Outcomes of individual vs. group instruction in EFNEP J. Extension 41: http://www.joe.org/joe/2003april/a4.shtml [accessed December 19, 2003].
22. Snedecor, G. W. & Cochran, W. G. (1980) EFNEP: an investment in America’s future. Statistical Methods Iowa State University Press Ames, IA.
23. Hackman, J. R. & Oldham, G. R. (1980) EFNEP: an investment in America’s future. Work Redesign Addison-Wesley Reading, MA.
24. Spreitzer, G. M. (1995) Psychological empowerment in the workplace: Dimensions, measurement, and validation. Acad. Manag. J. 38:1442-1465.
25. Menon, S. T. (1999) Psychological empowerment: Definition, measurement and validation. Can. J. Behav. Sci. 31:161-164.
26. Thomas, K. W. & Velthouse, B. A. (1990) Cognitive elements of empowerment: An "interpretive" model of intrinsic task motivation. Acad. Manag. Rev. 15:666-681.
27. Kim, H. & Yukl, G. (1995) Relationships of managerial effectiveness and advancement to self-reported and subordinate-reported leadership behaviors from the Multiple-Linkage Model. Leadership Q. 6:361-377.
28. Yukl, G., Wall, S. & Lepsinger, R. (1990) Preliminary report on validation of the Managerial Practices Survey. Clark, K. Clark, M. eds. Measures of Leadership :223-238 Leadership Library of America West Orange, NJ.
29. Judge, T. A., Thoresen, C. J., Bono, J. E. & Patton, G. K. (2001) The job satisfaction—job performance relationship: a qualitative and quantitative review. Psychol. Bull. 127:376-407.[Medline]
30. Iaffaldano, M. T. & Muchinsky, P. M. (1985) Job satisfaction and job performance: a meta-analysis. Psychol. Bull. 97:366-373.
31. Cotton, J. L. (1993) Job satisfaction and job performance: a meta-analysis. Employee Involvement: Methods for Improving Performance and Work Attitudes Sage Publications Newbury Park, CA.
32. Cason, K. L., Scholl, J. F. & Kassab, C. (2002) A comparison of program delivery methods for low income nutrition audiences. Top. Clin. Nutr. 17:63-73.
33. Rohs, F. R., Langone, C. A. & Coleman, R. K. (2001) Response shift bias: a problem in evaluating nutrition training using self-report measures. J. Nutr. Educ. 33:165-170.[Medline]
34. Amstutz, M. K. & Dixon, D. L. (1986) Dietary changes resulting from the Expanded Food and Nutrition Education Program. J. Nutr. Educ. 18:55-60.
35. Del Tredici, A. M., Joy, A. B., Omelich, C. L. & Laughlin, S. G. (1988) Evaluation study of the California Expanded Food and Nutrition Education Program: 24-hour recall data. J. Am. Diet. Assoc. 88:185-190.[Medline]
36. Arnold, C. G. & Sobal, J. (2000) Food practices and nutrition knowledge after graduation from the Expanded Food and Nutrition Education Program (EFNEP). J. Nutr. Educ. 32:130-138.
37. Schalock, R. L. (2001) Food practices and nutrition knowledge after graduation from the Expanded Food and Nutrition Education Program (EFNEP). Outcome-Based Evaluation Kluwer Academic/Plenum Publishers New York, NY.
This article has been cited by other articles:
|
|
 |
|
  P. Menon, M. Mbuya, J.-P. Habicht, G. Pelto, C. U. Loechl, and M. T. Ruel Assessing Supervisory and Motivational Factors in the Context of a Program Evaluation in Rural Haiti J. Nutr., March 1, 2008; 138(3): 634 - 637. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. S. Wellman, B. Kamp, N. J. Kirk-Sanchez, and P. M. Johnson Eat Better & Move More: A Community-Based Program Designed to Improve Diets and Increase Physical Activity Among Older Americans Am J Public Health, April 1, 2007; 97(4): 710 - 717. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
