A Bayesian Power Analysis Procedure Considering Uncertainty in Effect Size Estimates from a Meta-analysis

In conventional frequentist power analysis, one often uses an effect size estimate, treats it as if it were the true value, and ignores uncertainty in the effect size estimate for the analysis. The resulting sample sizes can vary dramatically depending on the chosen effect size value. To resolve the problem, we propose a hybrid Bayesian power analysis procedure that models uncertainty in the effect size estimates from a meta-analysis. We use observed effect sizes and prior distributions to obtain the posterior distribution of the effect size and model parameters. Then, we simulate effect sizes from the obtained posterior distribution. For each simulated effect size, we obtain a power value. With an estimated power distribution for a given sample size, we can estimate the probability of reaching a power level or higher and the expected power. With a range of planned sample sizes, we can generate a power assurance curve. Both the conventional frequentist and our Bayesian procedures were applied to conduct prospective power analyses for two meta-analysis examples (testing standardized mean differences in example 1 and Pearson's correlations in example 2). The advantages of our proposed procedure are demonstrated and discussed.     

How Much Older Do You Get When a Wrinkle Appears on Your Face? Modifying Age Estimates by Number of Wrinkles

ABSTRACT

The present study investigated the influence of wrinkles on facial age judgments. In Experiment 1, preadolescents, young adults, and middle-aged adults made categorical age judgments for male and female faces. The qualitative (type of wrinkle) and quantitative (density of wrinkles and depth of furrows) contributions of wrinkles were analyzed. Results indicated that the greater the number of wrinkles and the depth of furrows, the older a face was rated. The roles of the gender of the face and the age of the participants were discussed. In Experiment 2, participants performed relative age judgments by comparing pairs of faces. Results revealed that the number of wrinkles had more influence on the perceived facial age than the type of wrinkle. A MDS analysis showed the main dimensions on which participants based their judgments, namely, the number of wrinkles and the depth of furrows. We conclude that the quantitative component is more likely to increase perceived facial age. Nevertheless, other variables, such as the gender of the face and the age of the participants, also seem to be involved in the age estimation process.     

Estimating the population size of transgender women in San Francisco using multiple methods, 2013

Purpose: To estimate the number of adult transgender women (TGW) in San Francisco and to provide an example of how estimation can be conducted.

Methods: Using nine different services accessed by the TGW population, the service multiplier, integrated into a respondent-driven sampling (RDS) survey of TGW, was used to calculate nine estimates of the TGW population in San Francisco. Both the successive sampling method, also integrated into the RDS survey of TGW, and a population proportion from the literature each calculated an additional estimate of the TGW population. The median of all 11 estimates was calculated to present a single estimate of the TGW population in San Francisco.

Results: The median of estimates was 978 TGW in San Francisco (plausibility bounds 635 – 1,400). This estimate is consistent with known cases of HIV among TGW and surveys on the prevalence of infection in the TGW population.

Conclusion: Using multiple methods, we estimate the TGW population size to be 0.13% of the San Francisco population. Despite small numbers, TGW bear a disproportionate burden of HIV infection in San Francisco. TGW are a priority for HIV care and prevention services.     

Judging the accuracy of a likelihood judgment: the case of smoking risk

A standard method for assessing whether people have appropriate internal representations of an event's likelihood is to check whether their subjective probability or frequency estimates for the event correspond with the assumed objective value for that event. When a person's estimate for the event exceeds its assumed objective probability or frequency, the person's expectancy for the event is concluded to be greater than warranted. This paper describes three lines of reasoning as to why conclusions of this sort can be problematic. Recently published findings as well as data from two new experiments are described to support this main thesis. The case of smoking risk is used to illustrate the more general problem, and issues that must be considered to avoid or contend with the problem are discussed. Copyright © 2002 John Wiley & Sons, Ltd.     

书写习惯对数字空间表征SNARC效应的影响

本文设计两个实验,采用大小判断任务,选取读写习惯相反的维、汉大学生为研究对象,探讨两类被试对阿拉伯数字和母语数字空间表征的一致性与差异性。结果发现：（1）维、汉被试均表现出对阿拉伯小数反应左手快于右手,对大数反应右手快于左手,即出现正向SNARC效应;（2）维吾尔族被试表现出对母语小数反应右手快于左手,对大数反应左手快于右手,即对母语数字出现反向SNARC效应;汉族被试表现出对母语小数反应左手快于右手,对大数反应右手快于左手,即对母语数字出现正向SNARC效应。结论：（1）维、汉被试对于阿拉伯数字的SNARC效应存在一致性;（2）维、汉被试母语数字的SNARC效应存在差异性;（3）数字空间对应性与书写习惯是导致结果的重要因素。     

On the Origins of Children's Compliance and Opposition: Family Context, Reinforcement, and Rules

Interactional synchrony and context dependent reinforcement are presented as similar phenomena in this discussion of children's dispositions to comply with or to oppose parental instructions. I argue that consistent and appropriate parental reactions to the full range of a child's response repertoire will establish a family context conducive to positive reinforcement of child compliance. In contrast, inconsistent parental reactions appear to create a chaotic family context conducive to the negative reinforcement of child opposition. Depending on these two forms of context, the parents and children seem to generate distinctive personal rules which outline the functional arrangements of context, behavior and reinforcement. The importance of these personal rules in helping parents and children to improve the lawfulness of family context is highlighted.     

Bayesian estimation in the two-parameter logistic model

A Bayesian procedure is developed for the estimation of parameters in the two-parameter logistic item response model. Joint modal estimates of the parameters are obtained and procedures for the specification of prior information are described. Through simulation studies it is shown that Bayesian estimates of the parameters are superior to maximum likelihood estimates in the sense that they are (a) more meaningful since they do not drift out of range, and (b) more accurate in that they result in smaller mean squared differences between estimates and true values.The research reported here was performed pursuant to Grant No. N0014-79-C-0039 with the Office of Naval Research.     

A modified beta binomial model with applications to multiple choice and taste tests

A modified beta binomial model is presented for use in analyzing ramdom guessing multiple choice tests and certain forms of taste tests. Detection probabilities for each item are distributed beta across the population subjects. Properties for the observable distribution of correct responses are derived. Two concepts of true score estimates are presented. One, analogous to Duncan's empirical Bayes posterior mean score, is appropriate for assessing the subject's performance on that particular test. The second is more suitable for predicting outcomes on similar tests.This research was made possible by a grant from the Center for Food Policy Research, Graduate School of Business, Columbia University.     

Self- and other-estimates of multiple abilities in Britain and Turkey: a cross-cultural comparison of subjective ratings of intelligence

This study is part of a programmatic research effort into the determinants of self-assessed abilities. It examined cross-cultural differences in beliefs about intelligence and self- and other-estimated intelligence in two countries at extreme ends of the European continent. In all, 172 British and 272 Turkish students completed a three-part questionnaire where they estimated their parents', partners' and own multiple intelligences (Gardner (10) and Sternberg (3)). They also completed a measure of the 'big five' personality scales and rated six questions about intelligence. The British sample had more experience with IQ tests than the Turks. The majority of participants in both groups did not believe in sex differences in intelligence but did think there were race differences. They also believed that intelligence was primarily inherited. Participants rated their social and emotional intelligence highly (around one standard deviation above the norm). Results suggested that there were more cultural than sex differences in all the ratings, with various interactions mainly due to the British sample differentiating more between the sexes than the Turks. Males rated their overall, verbal, logical, spatial, creative and practical intelligence higher than females. Turks rated their musical, body-kinesthetic, interpersonal and intrapersonal intelligence as well as existential, naturalistic, emotional, creative, and practical intelligence higher than the British. There was evidence of participants rating their fathers' intelligence on most factors higher than their mothers'. Factor analysis of the ten Gardner intelligences yield two clear factors: cognitive and social intelligence. The first factor was impacted by sex but not culture; it was the other way round for the second factor. Regressions showed that five factors predicted overall estimates: sex (male), age (older), test experience (has done tests), extraversion (strong) and openness (strong). Results are discussed in terms of the growing literature in the field.