Estimation of treatment effects for heterogeneous matched‐pairs data with probit models

Jun Wang; Wei Gao; Man‐Lai Tang

doi:10.1111/sjos.12363

A conditional test with demonstrated insensitivity to unmeasured bias in matched observational studies

Biometrika ◽

10.1093/biomet/asaa032 ◽

2020 ◽

Vol 107 (4) ◽

pp. 827-840

Author(s):

P R Rosenbaum

Keyword(s):

Observational Study ◽

Null Hypothesis ◽

Observational Studies ◽

Treatment Effects ◽

R Package ◽

Causal Effects ◽

Rank Statistic ◽

Matched Pairs ◽

Response Patterns ◽

Signed Rank

Summary In an observational study matched for observed covariates, an association between treatment received and outcome exhibited may indicate not an effect caused by the treatment, but merely some bias in the allocation of treatments to individuals within matched pairs. The evidence that distinguishes moderate biases from causal effects is unevenly dispersed among possible comparisons in an observational study: some comparisons are insensitive to larger biases than others. Intuitively, larger treatment effects tend to be insensitive to larger unmeasured biases, and perhaps matched pairs can be grouped using covariates, doses or response patterns so that groups of pairs with larger treatment effects may be identified. Even if an investigator has a reasoned conjecture about where to look for insensitive comparisons, that conjecture might prove mistaken, or, when not mistaken, it might be received sceptically by other scientists who doubt the conjecture or judge it to be too convenient in light of its success with the data at hand. In this article a test is proposed that searches for insensitive findings over many comparisons, but controls the probability of falsely rejecting a true null hypothesis of no treatment effect in the presence of a bias of specified magnitude. An example is studied in which the test considers many comparisons and locates an interpretable comparison that is insensitive to larger biases than a conventional comparison based on Wilcoxon’s signed rank statistic applied to all pairs. A simulation examines the power of the proposed test. The method is implemented in the R package dstat, which contains the example and reproduces the analysis.

Download Full-text

Bootstrap Inference for Quantile Treatment Effects in Randomized Experiments with Matched Pairs

Review of Economics and Statistics ◽

10.1162/rest_a_01089 ◽

2021 ◽

pp. 1-47

Author(s):

Liang Jiang ◽

Xiaobin Liu ◽

Peter C.B. Phillips ◽

Yichong Zhang

Keyword(s):

Propensity Score ◽

Treatment Effects ◽

Tuning Parameter ◽

Matched Pairs ◽

Negative Dependence ◽

Randomized Experiments ◽

Parameter Choice ◽

Tuning Parameters ◽

Quantile Treatment Effects ◽

Bootstrap Inference

Abstract This paper examines methods of inference concerning quantile treatment effects (QTEs) in randomized experiments with matched-pairs designs (MPDs). Standard multiplier bootstrap inference fails to capture the negative dependence of observations within each pair and is therefore conservative. Analytical inference involves estimating multiple functional quantities that require several tuning parameters. Instead, this paper proposes two bootstrap methods that can consistently approximate the limit distribution of the original QTE estimator and lessen the burden of tuning parameter choice. Most especially, the inverse propensity score weighted multiplier bootstrap can be implemented without knowledge of pair identities.

Download Full-text

A Phonomotor Approach to Apraxia of Speech Treatment

American Journal of Speech-Language Pathology ◽

10.1044/2020_ajslp-19-00116 ◽

2020 ◽

Vol 29 (4) ◽

pp. 2109-2130

Author(s):

Lauren Bislick

Keyword(s):

Phase I ◽

Treatment Effects ◽

Single Case ◽

Motor Planning ◽

The Other ◽

Multimodal Approach ◽

Apraxia Of Speech ◽

Duration Of Treatment ◽

Treatment Gains ◽

Future Work

Purpose This study continued Phase I investigation of a modified Phonomotor Treatment (PMT) Program on motor planning in two individuals with apraxia of speech (AOS) and aphasia and, with support from prior work, refined Phase I methodology for treatment intensity and duration, a measure of communicative participation, and the use of effect size benchmarks specific to AOS. Method A single-case experimental design with multiple baselines across behaviors and participants was used to examine acquisition, generalization, and maintenance of treatment effects 8–10 weeks posttreatment. Treatment was distributed 3 days a week, and duration of treatment was specific to each participant (criterion based). Experimental stimuli consisted of target sounds or clusters embedded nonwords and real words, specific to each participants' deficit. Results Findings show improved repetition accuracy for targets in trained nonwords, generalization to targets in untrained nonwords and real words, and maintenance of treatment effects at 10 weeks posttreatment for one participant and more variable outcomes for the other participant. Conclusions Results indicate that a modified version of PMT can promote generalization and maintenance of treatment gains for trained speech targets via a multimodal approach emphasizing repeated exposure and practice. While these results are promising, the frequent co-occurrence of AOS and aphasia warrants a treatment that addresses both motor planning and linguistic deficits. Thus, the application of traditional PMT with participant-specific modifications for AOS embedded into the treatment program may be a more effective approach. Future work will continue to examine and maximize improvements in motor planning, while also treating anomia in aphasia.

Download Full-text

Brief Report. The PCL-R and Gender

European Journal of Psychological Assessment ◽

10.1027//1015-5759.16.3.147 ◽

2000 ◽

Vol 16 (3) ◽

pp. 147-149 ◽

Cited By ~ 44

Author(s):

Martin Grann

Keyword(s):

Gender Differences ◽

Sexual Behavior ◽

Discriminant Analysis ◽

Juvenile Delinquency ◽

Violent Offenders ◽

Matched Pairs ◽

Psychopathy Checklist ◽

Psychopathy Checklist Revised ◽

And Gender

Summary: Hare's Psychopathy Checklist - Revised (PCL-R; Hare, 1991 ) was originally constructed for use among males in correctional and forensic settings. In this study, the PCL-R protocols of 36 matched pairs of female and male violent offenders were examined with respect to gender differences. The results indicated a few significant differences. By means of discriminant analysis, male Ss were distinguished from their female counterparts through their relatively higher scores on “callous/lack of empathy” (item 8) and “juvenile delinquency” (item 18), whereas the female Ss scored relatively higher on “promiscuous sexual behavior” (item 11). Some sources of bias and possible implications are discussed.

Download Full-text