“No Evidence for a Relationship between Intelligence and Ejaculate Quality”, 2020-09-18 (; backlinks; similar):
Genetic quality may be expressed through many traits simultaneously, and this would suggest a phenotype-wide fitness factor. In humans, intelligence has been positively associated with several potential indicators of genetic quality, including ejaculate quality. We conducted a conceptual replication of one such study by investigating the relationship between intelligence (assessed by the Raven Advanced Progressive Matrices Test-Short Form) and ejaculate quality (indexed by sperm count, sperm concentration, and sperm motility) in a sample of 41 men (ages ranging 18 to 33 years; M = 23.33; SD = 3.60). By self-report, participants had not had a vasectomy, and had never sought infertility treatment. We controlled for several covariates known to affect ejaculate quality (eg. abstinence duration before providing an ejaculate) and found no statistically-significant relationship between intelligence and ejaculate quality; our findings, therefore, do not match those of Arden, Gottfredson, Miller et al or those of previous studies. We discuss limitations of this study and the general research area and highlight the need for future research in this area, especially the need for larger data sets to address questions around phenotypic quality and ejaculate quality.
[Keywords: phenotype-wide fitness factor, ejaculate quality, intelligence, fertility, Raven Advanced Progressive Matrices test]
…An important limitation of the current research is the small sample of 41 men, as small sample sizes increase the risk of both Type I and Type II errors. Our analyses, therefore, may have lacked sufficient power to detect the small effect sizes, r = 0.14 to 0.19, reported by Arden, Gottfredson, Miller, and 2009. Small sample sizes are a recurrent limitation of psychological research investigating ejaculate quality (eg. 1989; et al 2005), perhaps due to difficulties recruiting participants outside a clinical setting. Arden, Gottfredson, Miller, and Pierce analyzed data from a sample of 425 men, which afforded the analyses over 80% power to detect small effects. However, it is important to note that the correlation coefficients we obtained were similar in magnitude to those reported by Arden, Gottfredson, Miller, and Pierce, ranging from −0.18 to 0.30, and the repeated-measures nature of our study gave it greater power despite the small sample size.