SAGE Journal Articles
Click on the following links. Please note these will open in a new window.
Standardized effect sizes and confidence intervals thereof are extremely useful devices for comparing results across different studies using scales with incommensurable units. However, exact confidence intervals for standardized effect sizes can usually be obtained only via iterative estimation procedures. The present article summarizes several closed-form approximations to the exact confidence interval bounds in the two-independent and two-dependent samples design. Monte Carlo simulations were conducted to determine the accuracy of the various approximations under a wide variety of conditions. All methods except one provided accurate results for moderately large sample sizes and converged to the exact confidence interval bounds as the sample size increased.
Questions to Consider
1. Explain how standardized effect sizes and confidence intervals are useful devices for comparing results across different studies using scales with incommensurable units.
Cognitive Domain: Comprehension
Difficulty Level: Medium
2. Summarize a couple of examples from the article regarding the closed-form approximations to the exact confidence interval bounds in the two-independent and two-dependent samples design.
Cognitive Domain: Analysis
Difficulty Level: Hard
3. How accurate were the results for moderately large sample sizes?
Cognitive Domain: Knowledge
Difficulty Level: Medium
Mixed-effects models are increasingly used in many areas of applied science. Despite their popularity, there is virtually no systematic approach for examining the homogeneity of the random-effects covariance structure commonly assumed for such models. We propose two tests for evaluating the homogeneity of the covariance structure assumption across subjects: one is based on the covariance matrices computed from the fitted model and the other is based on the empirical variation computed from the estimated random effects. We used simulation studies to compare performances of the two tests for detecting violations of the homogeneity assumption in the mixed-effects models and showed that they were able to identify abnormal clusters of subjects with dissimilar random-effects covariance structures; in particular, their removal from the fitted model might change the signs and the magnitudes of important predictors in the analysis. In a case study, we applied our proposed tests to a longitudinal cohort study of rheumatoid arthritis patients and compared their abilities to ascertain whether the assumption of covariance homogeneity for subject-specific random effects holds.
Questions to Consider
1. Describe the mixed-methods model and explain why it has begun to be used more.
Cognitive Domain: Comprehension, Knowledge
Difficulty Level: Medium
2. Describe and explain the two tests presented in this article for evaluating the homogeneity of the covariance structure assumption across subjects.
Cognitive Domain: Comprehension
Difficulty Level: Medium–Hard
3. The authors discuss the need for adjusting the inflated type I error; why is this recommended?
Cognitive Domain: Comprehension, Analysis
Difficulty Level: Hard
It is shown that the procedure of applying the Wilcoxon test after Levene’s transformation can have an inflated type I error rate when distributions are skewed. Thus, when the data may come from an asymmetric distribution, the Wilcoxon test should not be applied as a test for homogeneity of variances after Levene’s transformation.
Questions to Consider
1. Explain some of the consequences of inflated type I error. Why is this important?
Cognitive Domain: Comprehension
Difficulty Level: Medium
2. Why do the authors posit that the Wilcoxon test should not be applied as a test for homogeneity of variances after Levene’s transformation?
Cognitive Domain: Comprehension
Difficulty Level: Medium
3. When distributions are skewed, what happens in regards to running Wilcoxon test after Levene’s transformation? What did previous studies fail to recognize?
Cognitive Domain: Comprehension, Analysis
Difficulty Level: Hard
Men succumb to sexual temptations (e.g., infidelity, mate poaching) more than women. Explanations for this effect vary; some researchers propose that men and women differ in sexual impulse strength, whereas others posit a difference in sexual self-control. These studies are the first to test such underlying mechanisms. In Study 1, participants reported on their impulses and intentional control exertion when they encountered a real-life tempting but forbidden potential partner. Study 2 required participants to perform a reaction-time task in which they accepted/rejected potential partners, and we used process dissociation to separate the effects of impulse and control. In both studies, men succumbed to the sexual temptations more than women, and this sex difference emerged because men experienced stronger impulses, not because they exerted less intentional control. Implications for the integration of evolutionary and self-regulatory perspectives on sex differences are discussed.
Questions to Consider
1. Why did the authors conduct a one-tailed test when comparing behavioral enactment between men and women? Do you have any concerns about the statistics that they report to support their conclusion?
Learning Objective: Directional hypotheses
Cognitive Domain: Analysis
Difficulty: Hard
2. Is sex a(n): (a) IV, (b) DV, (c) quasi-IV, (d) confounding variable.
Learning Objective: Independent variables
Cognitive Domain: Comprehension
Difficulty Level: Easy
3. What additional information would the authors need to report for you to compute the effect size, Cohen’s d, for the difference in behavioral enactment between men and women? (a) The standardized relationship between the two variables. (b) The confidence interval for the difference between the means. (c) The standard error of the difference. (d) The standard deviation of behavioral enactment for men and women separately.
Learning Objective: Effect size
Cognitive Domain: Application
Difficulty Level: Medium
When we buy our daily cup of coffee, sometimes we engage in a social interaction with the barista, and sometimes we are in a rush. Every day we have opportunities to transform potentially impersonal, instrumental exchanges into genuine social interactions, and the happiness literature suggests that we may reap benefits by doing so; in other words, treating a service provider like we would an acquaintance (i.e., weak tie) might make us happier. In the current study, people who had a social interaction with a barista (i.e., smiled, made eye contact, and had a brief conversation) experienced more positive affect than people who were as efficient as possible. Further, we found initial evidence that these effects were mediated by feelings of belonging. These results suggest that, although people are often reluctant to have a genuine social interaction with a stranger, they are happier when they treat a stranger like a weak tie.
Questions to Consider
1. The authors computed ANOVAs to compare two-groups instead of t-tests. When there are only two groups, the relationship between F and t is: t2 = F. Using this formula, convert the F values for positive affect and negative affect for experiences to t-tests. Does this change the author’s conclusions?
Learning Objective: One IV with two levels
Cognitive Domain: Comprehension
Difficulty Level: Medium
2. For positive affect in experiences the pooled standard deviation is 0.63. What is Cohen’s d? (a) −0.11. (b) 12.41. (c) 0.98. (d) 23.41.
Learning Objective: Effect size
Cognitive Domain: Analysis
Difficulty Level: Medium
3. In the limitations and alternative explanations section, the authors compare the two experimental groups to the control group. Evaluating their t-tests, which conclusion could we draw? (a) Participants in the social condition reported feeling greater belonging than the control group. (b) Participants in the efficient condition reported feeling less belonging than the control group. (c) Participants in both conditions were significantly different than the control group. (d) Participants in neither condition were significantly different than the control group.
Learning Objective: Significance testing
Cognitive Domain: Evaluation
Difficulty Level: Easy