PubMed Health⌕ Search

Biomedical subjects

K A Bollen

Publications and source records attributed to K A Bollen.

16 recordsLinked to original sources
PubMed2000

A tetrad test for causal indicators.

The authors propose a confirmatory tetrad analysis test to distinguish causal from effect indicators in structural equation models. The test uses "nested" vanishing tetrads that are often implied when comparing causal and effect indicator models. The authors present typical models that researchers can use to determine the vanishing tetrads for 4 or more variables. They also provide the vanishing tetrads for mixtures of causal and effect indicators, for models with fewer than 4 indicators per latent variable, or for cases with correlated errors. The authors illustrate the test results for several simulation and empirical examples and emphasize that their technique is a theory-testing rather than a model-generating approach. They also review limitations of the procedure including the indistinguishable tetrad equivalent models, the largely unknown finite sample behavior of the test statistic, and the inability of any procedure to fully validate a model specification.

Causality↗
PubMed2000

How well do perceptions of family planning service quality correspond to objective measures? Evidence from Tanzania.

This study examines the relationship between common objective measures of quality and perceptions of the quality of family planning facilities. Results of prior research indicate that such perceptions are an important determinant of contraceptive use in rural Tanzania. The data for this study are drawn from two surveys conducted in rural Tanzania. Three models are tested separately for women and for men. The important determinants of perceptions of quality among women and men are: perceived travel time to the facility, availability of immunizations, and availability of maternal and child health services. Additionally, the ratio of the number of staff to outpatients is important to men. The data explain a moderate amount of the variance in the quality measures, indicating that perceived quality is not fully predicted by common objective measures of quality. Future surveys of facility quality should develop objective measures to better predict the perceived quality, with the underlying goal of increasing contraceptive use.

Adult↗
PubMed1999

Quality, accessibility, and contraceptive use in rural Tanzania.

We examine how informants' reports on community perceptions of the quality and accessibility of family planning facilities relate to the use of modern contraceptives by individuals in rural Tanzania. Using information on individual-level contraceptive use in conjunction with community-level information on the accessibility and quality of family planning facilities, we employ two distinct statistical procedures to illustrate the impacts of accessibility and quality on contraceptive use. Both procedures treat the community-level variables as imperfect indicators of characteristics of the facilities, and they yield nearly identical implications. We find that a community-level, subjective perception of a family planning facility's quality has a significant impact on community members' contraceptive use whereas other community measures such as time, distance, and subjective perception of accessibility have trivial and insignificant direct impacts, net of the control variables. Future research that uncovers the determinants of perceptions of both community-level and individual-level quality could provide key insights for developing effective and efficient family planning programs.

Adolescent↗
PubMed1997

A resident-based reimbursement system for intermediate care facilities for the mentally retarded.

In this article, the authors present a resident-based reimbursement system for intermediate care facilities for the mentally retarded (ICFs-MR), which represent a large and growing proportion of the medicaid budget. The statistical relationship between resident disability level and the expected cost of caring for the individual is estimated, allowing for the prediction of expected resource use across the population of ICF-MR residents. The system incorporates an indirect cost rate, a base direct care rate (constant across all providers), and an individual-specific direct care rate, based on the expected cost of care.

Disability Evaluation↗
PubMed1995

Binary outcomes and endogenous explanatory variables: tests and solutions with an application to the demand for contraceptive use in Tunisia.

Many demographic studies examine discrete outcomes, and researchers often suspect that some of the explanatory variables may be influenced by the same unobserved factors that determine the discrete outcome under examination. In linear models, the standard solution to this potential endogeneity bias is an estimator such as two-stage least squares. These methods have been extended to models with limited dependent variables, but there is little information on the performance of the methods in the types of data sets typically used in demographic research. This paper helps to fill this gap. It describes a simple analytic framework for estimating the effects of explanatory variables on discrete outcomes, which controls for the potential endogeneity of explanatory variables. It also discusses tests for exogeneity and joint determination of the outcomes and the explanatory variables. It summarizes the results of a Monte Carlo study of the performance of these techniques and uses these results to suggest how researchers should approach these problems in practice. We apply these methods to the examination of the impact of fertility intentions on contraceptive use, based on data from the 1988 Tunisia Demographic and Health Survey.

Adolescent↗
PubMed1987

Cross-cultural comparability of the Philadelphia Geriatric Center Morale Scale: an American-Japanese comparison.

This study involves an American-Japanese comparison of the factorial structure of the Philadelphia Geriatric Center (PGC) Morale Scale. A model containing 11 of the original PGC Morale Scale items was found to fit both the American and Japanese data adequately. Factorial invariance was analyzed by testing a series of nested hypotheses involving various equivalence constraints. No major differences were found in the 11-item PGC Morale Scale between the American and Japanese data sets.

Aged↗
PubMed1987

Race differences in factorial structures of two measures of subjective well-being.

This research examines race differences in the factorial structure of two popular measures of psychological well-being: the Philadelphia Geriatric Center (PGC) Morale Scale and the Life Satisfaction Index A (LSIA). In particular, we view the covariance structure of the items of each scale as a function of several parameter matrices. We analyzed the factorial invariance by testing hypotheses involving the equivalence constraints of one or more parameter matrices in the White and the Black samples. Data for the research came from the 1968 National Senior Citizens Survey and the 1974 Harris National Council on Aging Survey, Myths and Reality of Aging in America. Analysis of covariance structures, or LISREL, was used to assess the factorial invariance of both the PGC Morale Scale and the LSIA. Race differences were found in the factorial structure of the PGC Morale Scale, but not in the structure of LSIA. In particular, consistent race differences in measurement error variances were found for two items of the PGC Morale Scale.

Black or African American↗
PubMed1986

Age differences in the structure of the Philadelphia Geriatric Center Morale Scale.

In this research we examined age differences in the factorial structure of the Philadelphia Geriatric Center (PGC) Morale Scale. In particular, we viewed the covariance structure of the PGC Morale Scale items as a function of several parameter matrices. We analyzed the factorial invariance by testing hypotheses involving the equivalence constraints of one or more parameter matrices in the young-old (65-74) and the old-old (75 and over) populations. Data for this research came from the 1968 National Senior Citizens Survey. Analysis of covariance structures, or LISREL, was used to assess the factorial invariance of the PGC Morale Scale. Although there are some statistically significant age differences in the factorial structure, substantively they are less important.

Aged↗
PubMed1985

Sex differences in the structure of the Philadelphia Geriatric Center Morale Scale.

This research examined sex differences in the factorial structure of the Philadelphia Geriatric Center Morale Scale. In particular, we view the covariance structure of the scale items as a function of several parameter matrices. We analyzed the factorial invariance by testing hypotheses involving the equivalence constraints of one or more parameter matrices in the male and female populations. Data for this research came from the 1968 National Senior Citizens Survey. Analysis of covariance structures (or LISREL) was used to assess the factorial invariance of the scale. We found some statistically significant sex differences in the first-order factor loadings and measurement error variances. Substantively, however, these differences are less important.

Aged↗
PubMed1983

The structure of the Philadelphia Geriatric Center Morale scale: a reinterpretation.

This research examines the structure of the Philadelphia Geriatric Center (PGC) Morale Scale. A multiple indicator structural equation model including three first-order factors and one second-order factor is proposed to account for the dimensionality of the PGC Morale Scale. Based on data from the 1968 National Senior Citizens Survey, the proposed model is consistently supported and replicated across four randomly divided subsamples. Given these results, the claim that the PGC Morale Scale is multidimensional is only appropriate for the first-order factors. On the level of the second-order, a unidimensional interpretation is more accurate.

Aged↗
PubMed1983

Temporal variations in mortality: a comparison of U.S. suicides and motor vehicle fatalities, 1972-1976.

Among the various types of mortality, motor vehicle fatalities (MVF) and suicides are two that are very responsive to social, economic, and psychological factors. The analysis of the temporal patterns of suicides and MVF points to important but neglected forces affecting them. This paper examines the daily patterns of MVF and suicides for the United States from 1972 to 1976. The effects of the day of the week, month, year, and holidays on these types of mortality are estimated. Total MVF and suicides, as well as daily differentials by sex, age, and race, are analyzed with regression techniques. The results show that temporal variations are stronger in MVF than in suicides but are present in both. MVF tend to peak on Saturday, in the summer months, in 1972 and 1973, and on holidays. Suicides are generally highest on Monday, in the spring months, in 1975 and 1976, and on nonholidays. Differentials in the temporal patterns are evident. For instance, whites generally have a sharper decline in suicides on holidays than do nonwhites. The holiday peak in MVF for the 60 and over population is Christmas, while it is New Year's for the younger age groups. Males exhibit a greater increase in suicides from March to May than do females. Finally, the results of a comparison of the daily patterns of MVF and suicides revealed a negative correlation. This contrasted with the strong positive relationship between MVF and suicides generally found in cross-sectional studies. The implications of these findings are briefly discussed.

Accidents, Traffic↗