While entirely understandable given that their work was a "candidate pathway" analysis rather than a comparative analysis of pathway significance, it nonetheless points to the necessity of systematically controlling for the number of SNPs in a pathway (model complexity) when evaluating the significance of a candidate pathway association (model fit).
Careful research is needed to disentangle these associations by systematically controlling for the severity of depression and/or the environmental context of stress.
In these experiments, we systematically controlled for bird density, pen size, and previous social experience in groups of different sizes.
Surprisingly, while many fMRI researchers routinely take pains to eliminate response accuracy differences as a potential confound (e.g., by only analyzing trials with correct responses), relatively few studies have systematically controlled for trial-by-trial RT differences [e.g.], [5], [7,47]; a recent survey of 170 fMRI studies found that only 9% had explicitly modeled RT [3].
Last, statistical comparisons between the present data and statistical information from other samples were not systematically controlled for gender and age.
Unless otherwise specified, multiple regression was done as in Lartillot and Delsuc (2012), whereby correlations are systematically controlled for all other variables.
Theoretically, asymptomatic nocturnal hypoglycemic episodes occurring in the night before the experiment, which were not systematically controlled for in our study, could have influenced our results.
Pearson correlations between control variables and infant outcomes are presented in Table 2. Age at testing and social environment correlated with several of the outcomes, supporting the decision to systematically control for these variables in the regression models.
In all analyses, we systematically controlled for sickness absence, with nonpsychiatric and missing diagnoses as dichotomous variables to ensure that the associations observed were net of the effect of sickness absence due to other reasons.
Further, data on risk factors were ascertained only at baseline; therefore, we could not systematically control for differences in the severity of risk factors or perform time-dependent analyses.
