Search Results

Author: O'Keefe, Patrick
Resulting in 5 citations.
1. O'Keefe, Patrick
Rodgers, Joseph Lee
A Simulation Study of Bootstrap Approaches to Estimate Confidence Intervals in DeFries-Fulker Regression Models (with Application to the Heritability of BMI Changes in the NLSY)
Behavior Genetics 50 (2020): 127-138.
Also: https://link.springer.com/article/10.1007/s10519-020-09993-9
Cohort(s): NLSY79
Publisher: Behavior Genetics Association
Keyword(s): Body Mass Index (BMI); Modeling

Permission to reprint the abstract has not been received from the publisher.

The univariate bootstrap is a relatively recently developed version of the bootstrap (Lee and Rodgers in Psychol Methods 3(1): 91, 1998). DeFries-Fulker (DF) analysis is a regression model used to estimate parameters in behavioral genetic models (DeFries and Fulker in Behav Genet 15(5): 467-473, 1985). It is appealing for its simplicity; however, it violates certain regression assumptions such as homogeneity of variance and independence of errors that make calculation of standard errors and confidence intervals problematic. Methods have been developed to account for these issues (Kohler and Rodgers in Behav Genet 31(2): 179-191, 2001), however the univariate bootstrap represents a unique means of doing so that is presaged by suggestions from previous DF research (e.g., Cherny et al. in Behav Genet 22(2): 153-162, 1992). In the present study we use simulations to examine the performance of the univariate bootstrap in the context of DF analysis. We compare a number of possible bootstrap schemes as well as more traditional confidence interval methods. We follow up with an empirical demonstration, applying results of the simulation to models estimated to investigate changes in body mass index in adults from the National Longitudinal Survey of Youth 1979 data.
Bibliography Citation
O'Keefe, Patrick and Joseph Lee Rodgers. "A Simulation Study of Bootstrap Approaches to Estimate Confidence Intervals in DeFries-Fulker Regression Models (with Application to the Heritability of BMI Changes in the NLSY)." Behavior Genetics 50 (2020): 127-138.
2. O'Keefe, Patrick
Rodgers, Joseph Lee
Double Decomposition of Level-1 Variables in Multilevel Models: An Analysis of the Flynn Effect in the NLSY Data
Multivariate Behavioral Research 52,5 (2017): 630-647.
Also: http://www.tandfonline.com/doi/full/10.1080/00273171.2017.1354758
Cohort(s): Children of the NLSY79
Publisher: Taylor & Francis
Keyword(s): Flynn Effect; I.Q.; Modeling, Multilevel

This paper introduces an extension of cluster mean centering (also called group mean centering) for multilevel models, which we call "double decomposition (DD)." This centering method separates between-level variance, as in cluster mean centering, but also decomposes within-level variance of the same variable. This process retains the benefits of cluster mean centering but allows for context variables derived from lower level variables, other than the cluster mean, to be incorporated into the model. A brief simulation study is presented, demonstrating the potential advantage (or even necessity) for DD in certain circumstances. Several applications to multilevel analysis are discussed. Finally, an empirical demonstration examining the Flynn effect, our motivating example, is presented. The use of DD in the analysis provides a novel method to narrow the field of plausible causal hypotheses regarding the Flynn effect, in line with suggestions by a number of researchers.
Bibliography Citation
O'Keefe, Patrick and Joseph Lee Rodgers. "Double Decomposition of Level-1 Variables in Multilevel Models: An Analysis of the Flynn Effect in the NLSY Data." Multivariate Behavioral Research 52,5 (2017): 630-647.
3. O'Keefe, Patrick
Rodgers, Joseph Lee
The Corrosive Influence of the Flynn Effect on Age Normed Tests
Multivariate Behavioral Research 54,1 (2019): 155.
Also: https://www.tandfonline.com/doi/full/10.1080/00273171.2018.1562322
Cohort(s): Children of the NLSY79
Publisher: Taylor & Francis
Keyword(s): Children, Academic Development; Cognitive Ability; Flynn Effect; I.Q.; Test Scores/Test theory/IRT

This project provides empirical evidence for this built-in FE [Flynn Effect]. Using the National Longitudinal Survey of Youth-Children dataset (the NLSYC) and a variety of multilevel models, we: (1) Show a within person effect with individuals scoring higher over time and (2) Do not find evidence for practice. Previous work (O’Keefe & Rodgers, 2017 O’Keefe, P., & Rodgers, J. L. (2017) with this sample suggests the within person effect is not the FE itself. The NLSYC is well-suited to the task because it includes a known FE and longitudinal data. We conclude that there may be an artificial FE built into ability instruments because of this norming bias.
Bibliography Citation
O'Keefe, Patrick and Joseph Lee Rodgers. "The Corrosive Influence of the Flynn Effect on Age Normed Tests." Multivariate Behavioral Research 54,1 (2019): 155.
4. O'Keefe, Patrick
Rodgers, Joseph Lee
The Flynn Effect Can Become Embedded in Tests: How Cross-sectional Age Norms Can Corrupt Longitudinal Research
Intelligence 82 (September-October 2020): 101481.
Also: https://www.sciencedirect.com/science/article/pii/S0160289620300593
Cohort(s): Children of the NLSY79
Publisher: Elsevier
Keyword(s): Flynn Effect; I.Q.; Peabody Individual Achievement Test (PIAT- Math); Test Scores/Test theory/IRT

The Flynn Effect (FE; Flynn, 1984, Flynn, 1987) is the decades-long increase in measured mean IQ of approximately 1/3 point per year, observed in industrialized nations over the course of at least a century. An obvious and practical implication of the FE is that the FE can cause test norm obsolescence. If norms from 1970 were used today, the average score would be approximately a standard deviation above the original mean. A more subtle effect was suggested by Mingroni (2007): Age-normed tests could have a FE "built-in" through the norming process. His observation can be true in any case where there are cohort differences (between- or within-family); it is almost certain to occur in cases where cross-sectional samples are used to age norm in the presence of cohort effects. We illuminate this process in several ways, because it can significantly impact longitudinal research. If the "built in FE" hypothesis is supported, then the FE potentially affects resulting scores assigned to test-takers from all age-normed cognitive tests exhibiting a FE. A series of graphic simulations demonstrate the logic. Following, analysis of the National Longitudinal Survey of Youth Children data suggest that the Flynn Effect is indeed built into the PIAT-Math scores.
Bibliography Citation
O'Keefe, Patrick and Joseph Lee Rodgers. "The Flynn Effect Can Become Embedded in Tests: How Cross-sectional Age Norms Can Corrupt Longitudinal Research." Intelligence 82 (September-October 2020): 101481.
5. Rodgers, Joseph Lee
Garrison, S. Mason
O'Keefe, Patrick
Bard, David E.
Hunter, Michael D.
Beasley, William H.
van den Oord, Edwin J. C. G.
Responding to a 100-Year-Old Challenge from Fisher: A Biometrical Analysis of Adult Height in the NLSY Data Using Only Cousin Pairs
Behavior Genetics 49,5 (September 2019): 444-454. https://link.springer.com/article/10.1007/s10519-019-09967-6
Cohort(s): Children of the NLSY79, NLSY79 Young Adult
Publisher: Behavior Genetics Association
Keyword(s): Family, Extended; Height; Kinship

Permission to reprint the abstract has not been received from the publisher.

In 1918, Fisher suggested that his research team had consistently found inflated cousin correlations. He also commented that because a cousin sample with minimal selection bias was not available the cause of the inflation could not be addressed, leaving this inflation as a challenge still to be solved. In the National Longitudinal Survey of Youth (the NLSY79, the NLSY97, and the NLSY-Children/Young Adult datasets), there are thousands of available cousin pairs. Those in the NLSYC/YA are obtained approximately without selection. In this paper, we address Fisher's challenge using these data. Further, we also evaluate the possibility of fitting ACE models using only cousin pairs, including full cousins, half-cousins, and quarter-cousins. To have any chance at success in such a restricted kinship domain requires an available and highly-reliable phenotype; we use adult height in our analysis. Results provide a possible answer to Fisher's challenge, and demonstrate the potential for using cousin pairs in a stand-alone analysis (as well as in combination with other biometrical designs).
Bibliography Citation
Rodgers, Joseph Lee, S. Mason Garrison, Patrick O'Keefe, David E. Bard, Michael D. Hunter, William H. Beasley and Edwin J. C. G. van den Oord. "Responding to a 100-Year-Old Challenge from Fisher: A Biometrical Analysis of Adult Height in the NLSY Data Using Only Cousin Pairs." Behavior Genetics 49,5 (September 2019): 444-454. https://link.springer.com/article/10.1007/s10519-019-09967-6.