Search

Post-hoc power estimates (power calculated for hypothesis tests after performing them) are sometimes requested by reviewers in an attempt to promote more rigorous designs. However, they should never be requested or reported because they have been shown to be logically invalid and practically misleading. We review the problems associated with post-hoc power, particularly the fact that the resulting calculated power is a monotone function of the p value and therefore contains no additional helpful information. We then discuss some situations that seem at first to call for post-hoc power analysis, such as attempts to decide on the practical implications of a null finding, or attempts to determine whether the sample size of a secondary data analysis is adequate for a proposed analysis, and consider possible approaches to achieving these goals. We make recommendations for practice in situations in which clear recommendations can be made, and point out other situations where further methodological research and discussion are required. [ABSTRACT FROM AUTHOR]

Little research has evaluated whether conflicting evidence for gender and racial/ethnic differences in nicotine dependence (ND) may be attributed to differences in psychometric properties of ND symptoms, particularly for young Hispanic smokers. Inadequate racial/ethnic diversity and limited smoking exposure variability has hampered research in young smokers. We used integrative data analysis (IDA) to pool DSM-IV ND symptom data for current smokers aged 12-25 (N = 20,328) from three nationally representative surveys (1999, 2000 National Surveys on Drug Use and Health (NSDUH) and Wave 1 National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). Moderated nonlinear factor analysis (MNLFA) tested symptom measurement invariance in the pooled sample containing greater ethnic and smoking exposure variability. There was study noninvariance for most symptoms. NESARC participants were more likely to report tolerance, using larger amounts or for longer periods, inability to cut down/quit, and more time spent smoking at higher levels of ND severity, but reported emotional/physical health problems at lower ND severity. Four symptoms showed gender or race/ethnicity noninvariance, but observed differences were small. An ND severity factor score adjusting for symptom noninvariance related to study membership, gender, and race/ethnicity did not differ substantively from traditional DSM-IV diagnosis and number of endorsed symptoms in estimated gender and race/ethnicity differences in ND. Results were consistent with studies finding minimal gender and racial/ethnic differences in ND, and suggest that symptom noninvariance is not a major contributor to observed differences. Results support IDA as a potentially promising approach for testing novel ND hypotheses not possible in independent studies. [ABSTRACT FROM AUTHOR]

The present study sought to extend prior research by using data from the National Comorbidity Survey-Replication (NCS-R) to examine the relationship between number of lifetime traumas, posttraumatic stress disorder (PTSD) and 15 self-reported chronic medical conditions. The goal was to determine whether the commonly found relationship between PTSD symptomatology and physical health were better explained by the number of lifetime traumas experienced. The NCS-R is a representative US household survey that assessed lifetime experience of a variety of traumas, lifetime diagnosis of PTSD and 15 chronic medical conditions (e.g. pain conditions, cardiovascular disorders, etc.). Two major findings emerged: (1) there was a graded relationship between trauma exposure, PTSD, and the majority of chronic medical conditions where individuals with PTSD had the highest likelihood of chronic medical condition and non-traumatized individuals had the lowest risk and; (2) with the exception of headaches, the relationship between PTSD and chronic medical conditions was explained by the number of lifetime traumas experienced when analyses were subset to traumatized individuals. The present study supports prior research suggesting that multiple traumas have a cumulative effect on physical health. The impact of trauma on health may be independent of PTSD symptomatology. [ABSTRACT FROM AUTHOR]