Do significant labor market events change who does the chores? Paid work,housework, and power in mixed-gender Australian households

We examine how men and women in mixed-gender unions change the time they allocate to housework in response to labor market promotions and terminations. Operating much like raises, such events have the potential to alter intra-household power dynamics. Using Australian panel data, we estimate couple-specific fixed effects models and find that female promotion has the strongest association with housework time allocation adjustments. These adjustments are in part attributable to concurrent changes in paid work time, but gender power relations also appear to play a role. Further results indicate that households holding more liberal gender role attitudes are more likely to adjust their housework time allocations after female promotion events. Power dynamics cannot, however, explain all the results. Supporting the sociological theory that partners may “do gender,” we find that in households with more traditional gender role attitudes, his housework time falls while hers rises when he is terminated.     

Muslim and Non‐Muslim Differences in Female Autonomy and Fertility: Evidence from Four Asian Countries

On the basis of research on paired Muslim and non‐Muslim communities selected in India, Malaysia, Thailand, and the Philippines, the authors test the hypothesis that greater observed Muslim pronatalism can be explained by less power or lower autonomy among Muslim women. Indeed, wives in the Muslim communities, compared to the non‐Muslim ones: 1) had more children, 2) were more likely to desire additional children, and 3) if they desired no more children, were less likely to be using contraception. However, the authors do not find that Muslim communities consistently score lower on dimensions of women's power/autonomy. Thus, aggregate‐level comparisons provide little evidence of a relationship between lower autonomy and higher fertility. Individual‐level multivariate analysis of married women in these paired settings similarly suggests that women's autonomy differentials do not account for the higher fertility, demand for more children, and less use of contraception among Muslim wives. These results suggest that explanations for Muslim/non‐Muslim fertility differences lie elsewhere.     

SOME TESTS OF BINOMIAL PROBABILITY ASSESSMENTS

A replication of the Alpert-Raiffa probability assessment experiment is used to motivate the development of several tests for evaluating assessors. The tests permit an evaluation of an assessor's accuracy in estimating target quantities and in estimating his own uncertainty as well. Assumption of normal subjective distributions permits the determination of subjective variances. Data from trial assessments are employed to test models which assert that assessed quantities are equal to subjectively assessed error variances. Several tests of the latter type are compared using data from individual, as well as pooled, assessors.     

Two-Group Classification Using Neural Networks*

Artificial neural networks are new methods for classification. We investigate two important issues in building neural network models; network architecture and size of training samples. Experiments were designed and carried out on two-group classification problems to find answers to these model building questions. The first experiment deals with selection of architecture and sample size for different classification problems. Results show that choice of architecture and choice of sample size depend on the objective: to maximize the classification rate of training samples, or to maximize the generalizability of neural networks. The second experiment compares neural network models with classical models such as linear discriminant analysis and quadratic discriminant analysis, and nonparametric methods such as k-nearest-neighbor and linear programming. Results show that neural networks are comparable to, if not better than, these other methods in terms of classification rates in the training samples but not in the test samples.     

The probabilistic analysis of testimony

The probabilistic analysis of testimony is surveyed. The coverage is not comprehensive; attention is focused on several problems of particular interest or complexity. The theory often contains implicit assumptions, and some attempt is made to clarify the role these play. The theory originally arose in an attempt to understand the logic of belief. It was not empirically grounded, however, and later died out in the 19th century when its conclusions became largely self-evident, and its non-empirical background suspect.     

Bayesian sequential methods for choosing the best multinomial cell: some simulation results

Suboptimal Bayesian sequential methods for choosing the best (i.e. largest probability) multinomial cell are considered and their performance is studied using Monte Carlo simulation. Performance characteristics, such as the probability of correct selection and some other associated with the sample size distribution, are evaluated assuming a maximum sample size. Single observation sequential rules as well as rules, where groups of observations are taken, and fixed sample size rules are discussed.     

Some tests for detection of faulty inspection

Some heuristic tests for detection of existence of errors in inspection are proposed. Some of these tests are only effective if the sampling fraction ((sample size)/(lot size)) is rather large, and in all cases their application predicates special experiments to provide the requisite data. Feasibility of these experiments will vary according to specific circumstances.     

A GOODNESS OF FIT TEST BASED ON THE LINKED LINE CHART1

In this paper, we propose a graphical representation of data and a test statistic based on it for testing the goodness of fit of a completely specified null distribution. The graph is constructed as a linked line chart given by vectors which reflect the pattern of order statistics. The test statistic is defined as an area defined by our chart and its asymptotic distribution is derived under the null hypothesis. Computer simulations performed to study the power properties of our chart indicate that the test is powerful for scale alternatives. Furthermore, it is shown that our test is closely related to the Watson test.     

CHARACTERIZATION OF EFFICIENT SEQUENTIAL SCHEMES FOR POPULATION GROWTH MODELS1

For population growth models, which include standard processes like Poisson, birth, birth-death, immigration-death, simple epidemic and M/M/1 queue as particular cases, we obtain efficient sequential schemes and characterize all efficiently estimable functions of the parameters by first deriving a sequential version of the information inequality.