Barely Significant

Excerpts

33,758 sentences · 1 filter applied

Other areas where we feel we have demonstrated appropriate judgement of significance is the inclusion of maternal age in our multivariate model based on our a priori hypothesis, and presentation of other ‘borderline significant’ outcomes, for example the inclusion of hypertension as a risk factor for admission (aOR 3.63, 95% CI 0.99 – 13.30).
The sentence has been amended by pointing out that numbers are small and the aOR from the sensitivity analysis have been added so that the readers can see the association for hypertension is of borderline significance: Line 222-223: “In the sensitivity analysis, there was some evidence that asthma and hypertension specifically increased the risk of admission, although the numbers in some groups were small (aOR 2.12, 95% CI 1.25-3.58 and aOR 3.63, 95% CI 0.99 – 13.30 respectively, Table S3).” line 255 when noting symptomatic SARS and pregnancy loss less than 24 weeks , if the authors do not wish to add the historical comparison can they explain why that is not used.
With regard to: ‘Specifically, intervention participants were less likely to report their lifestyle as ‘very healthy’ or ‘healthy’ (53.4%) compared with controls (59.5%) at the end of the study (PRR=0.94; 95%CI= 0.87, 1.01)’, that CI’s cross over 1.0 would suggest that though approaching significance, the finding is not statistically significant.
Therefore the findings of the present study and those from Treiman, Kessler & Evans are not in direct contradiction, given the nature of participants (bilingual pupils being proficient in two languages, compared to monolingual English-speaking students, who just started to learn French as a second language) and nature of direction (interference between equally mastered languages, compared to interferences from L2 to L1) even though our result was only marginally significant.