Longitudinal measurement models in evaluation research: Examining stability and change

This article provides an introduction to the use of Confirmatory Factor Analysis to test measurement invariance and stability in longitudinal research. The approach is illustrated through examples representing: (a) one construct, two measurement waves; (b) one construct, three waves; (c) two constructs, two waves; and (d) comparison of treatment and control groups in pre-post designs. Basic issues in establishing measurement invariance over time, across treatment groups, and within measurement waves are discussed. Estimates of the stability coefficients that are corrected for measurement error and method variance associated with each specific measured variable are provided. Establishing measurement invariance is a critical requirement for making inferences about treatment effects and changes in constructs over time.     

EVALUATION OF AN ACADEMIC FAMILY THERAPY TRAINING PROGRAM: CHANGES IN TRAINEES' RELATIONSHIP AND INTERVENTION SKILLS

Students in an academic training program were assessed for changes in their relationship and intervention skills following completion of the program. Positive improvements were found on some measures but not all. Additionally, trainee variables such as previous clinical experience. gender, age, marital stauts and scores on standardized admission tests were examined in relation to trainees' skills changes. Trainee characteristics were unrelated to changes in intervention skills. However, a number of relationships were found between trainee characteristics and changes in relationship skills.     

Lesbian and Gay Parents: Their Experiences of Children's Health Care in Australia

This study investigates the experiences of lesbian and gay parents seeking health care for their children within the Australian medical system. Previously unavailable demographic data was collected on lesbian and gay families, and qualitative information was sought from parents about positive or problematic health care experiences related to sexual orientation and family constellation. Ninety-two lesbian or gay parents responded to a national postal and Web-based survey. Eighty-nine percent of respondents reported a high level of satisfaction with the health care received by their children. Despite overall satisfaction, however, 49 percent of parents identified fear of disclosure of sexual orientation, and 27 percent reported negative or problematic experiences with their children's health care related to sexual orientation or family constellation. This study indicates that fear of discrimination and homophobic attitudes towards lesbian and gay families remain a factor within the Australian health care system.     

Effects of residential density on sonoran desert nocturnal rodents

We compared nocturnal rodent distributions among residential developments and in adjacent undeveloped native habitat remnants in Tucson, Arizona during October 1994. We trapped rodents in 2 low density (0.5 houses/ha) developments, 2 medium density (7.5 houses/ha) developments, and in 4 adjacent blocks of undeveloped native habitat using trap lines of 50 snap traps spaced at 2 traps every 10 meters. We compared species abundance, species richness, and total abundance among the 4 site types in a contingency table using a 2-tailed Fisher's exact test, and used post-hoc tests where the null hypothesis of no difference among sites was rejected. We encountered 144 individuals representing 5 species of nocturnal rodents in 1200 trap nights. Of 4 native rodent species, none were captured in medium housing density sites. House mice (Mus musculus) were captured only in medium housing density sites. Total rodent abundance was lower in both the medium density residential and the adjacent undeveloped sites than in either low density or adjacent undeveloped sites. Low density housing must be included in the developing matrix if the full complement of native nocturnal rodents is to be retained in Tucson.     

Neighborhood disadvantage, stress, and drug use among adults

This paper explores the relationships among neighborhood disadvantage, stress, and the likelihood of drug use in a sample of adults (N = 1,101). Using the 1995 Detroit Area Study in conjunction with tract-level data from the 1990 census, we find a positive relationship between neighborhood disadvantage and drug use, and this relationship remains statistically significant net of controls for individual-level socioeconomic status. Neighborhood disadvantage is moderately associated with drug related behaviors, indirectly through increased social stressors and higher levels of psychological distress among residents of disadvantaged neighborhoods. A residual effect of neighborhood disadvantage remains, net of a large number of socially relevant controls. Finally, results from interactive models suggest that the relationship between neighborhood disadvantage and drug use is most pronounced among individuals with lower incomes.     

Defining contracts between the state and charitable organisations in national accounts: a perspective from the UK

This paper explores the issues inherent in defining contracts for service delivery between the state and charitable organisations, and in differentiating them from grants. It begins by making the point that there is no one correct definition of a contract; rather, it is a function of its purpose. It then details our work in developing a definition of a contract within the national accounting tradition, as part of a study of the contribution of charitable organisations to the GDP of the United Kingdom. The paper ends by drawing some general conclusions from this experience.     

Reliability of the Demand-Control Questionnaire for sewing machine operators

The Demand-Control Questionnaire (DCQ), a 20-item scale that measures psychological work demands, job control and workplace social support, has frequently been used to assess occupational stress. The purpose of this study was to determine the test-retest reliability and internal consistency of the DCQ with sewing machine operators. Forty-six sewing machine operators completed the DCQ on two occasions with an 11-week time interval. A repeated measures analysis of variance model and subsequent application of generalizability theory were used to calculate the test-retest reliability of the subjects' ratings on the DCQ. Cronbach's alpha was used to determine the internal consistency of the scale. The test-retest reliability was 0.33 (95% confidence interval = 0.05-0.61), indicating fair reliability. Good internal consistency (Cronbach's alpha = 0.70) was found. The DCQ appears to be a reliable measure for assessing occupational stress in sewing machine operators. Workplaces need to place greater emphasis on the role of occupational stress in the prevention and treatment of musculoskeletal injuries among sewing machine operators.     

Ethnic Labels and Ethnic Identity as Predictors of Drug Use among Middle School Students in the Southwest

This article explores differences in the self‐reported drug use and exposure to drugs of an ethnically diverse group of 408 seventh‐grade students from a large city in the southwest. We contrast the explanatory power of ethnic labels (African American, non‐Hispanic White, Mexican American, and mixed ethnicity) and two dimensions of ethnic identity in predicting drug use. One dimension focuses on perceived ethnically consistent behavior, speech, and looks, while the other gauges a sense of ethnic pride. Ethnic labels were found to be somewhat useful in identifying differences in drug use, but the two ethnic identity measures, by themselves, did not generally help to explain differences in drug use. In conjunction, however, ethnic labels and ethnic identity measures explained far more of the differences in drug use than either did alone. The findings indicate that the two dimensions of ethnic identity predict drug outcomes in opposite ways, and these relations are different for minority students and non‐Hispanic White students. Generally, African American, Mexican American, and mixed‐ethnicity students with a strong sense of ethnic pride reported less drug use and exposure, while ethnically proud White students reported more. Ethnic minority students who viewed their behavior, speech, and looks as consistent with their ethnic group reported more drug use and exposure, while their White counterparts reported less. These findings are discussed, and recommendations for future research are provided.