A new method to explore inferential risks associated with each study in a meta-analysis: An approach based on Design Analysis

In the last ten years, scientific research has experienced an unprecedented “credibility’s crisis” of results. This means that researchers couldn't find the same results as in the original ones when conducting replication studies. In fact, the results showed that effects size were often not as strong as in the original studies and sometimes no effect was found. However, an important side-effect of the replicability crisis is that it increased the awareness of the problematic issues in the published literature and it promoted the development of new practices which would guarantee rigour, transparency and reproducibility. In the current work, the aim is to propose a new method to explore the inferential risks associated with each study in a meta-analysis. Specifically, this method is based on Design Analysis, a power analysis approach developed by @gelmanPowerCalculationsAssessing2014, which allows to analyse two other type of errors that are not commonly considered: the Type M (Magnitude) error and the Type S (Sign) error, concerning the magnitude and direction of the effects. We chose the Design Analysis approach because it allows to put more emphasis on the estimate of the effect size and it can be a valid tool available to researchers to make more conscious and informed decisions.

Soft on errors? The correcting mechanism of a Skolt Sami speller

The article discusses correcting of typos due to erroneous use of the so-called soft sign in Skolt Sami, one of the most common orthographic symbols, and the most common source of typographic errors. The discussion is based upon the suggestion mechanism of an existing open source Skolt Sami speller. The discussion shows that with an improved suggestion mechanism, the speller is able to restore a single soft sign error in over 97 % of the cases, and remove a hypercorrect soft sign as first correction in 90 % of the cases. Allowing the target form to be within top-5, the correction performance is well above 99 %. Improving the suggestion mechanism also had a positive impact of its overall performance, rising the percentage of target forms within top-5 from 74.1 % to 84.7 %.

A COMPARISON ON VARIANTS OF LMS USED IN FIR ADAPTIVE NOISE CANCELLERS FOR FETAL ECG EXTRACTION

Fetal electrocardiogram (FECG) non-invasively obtained through abdominal recordings serves as a promising diagnostic tool for fetal health monitoring during pregnancy. However, in the abdominal ECG (AECG) signal, FECG overlaps with maternal ECG (MECG) in both temporal and spectral domains in addition to interference from various sources like electromyogram, electrogastrogram, motion artifacts and other noises. The objective of this paper is to eliminate MECG components from AECG signal to extract FECG signal through FIR adaptive noise canceller (ANC) with filter coefficients updated using adaptive algorithms. Adaptive filters are suitable for current problem of interest and Least Mean Square (LMS) and its variants are analyzed for the problem of FECG extraction. We have compared the four variants of LMS such as normalized LMS (NLMS), sign-error algorithm, least mean fourth (LMF) algorithms for FECG extraction. The algorithms are evaluated using real-time abdominal ECG recordings acquired from daisy database. The performance of each algorithm is evaluated using various parameters like sensitivity, accuracy, positive predictive values and [Formula: see text] score. Further, the convergence rate for different algorithms are plotted and analyzed. From the simulation results, it is observed that the LMF algorithm outperforms its counterparts by providing an accuracy and positive predictive value of 73.3%, sensitivity of 100% and [Formula: see text] measure of 84.5%. The convergence plots obtained justify that LMF algorithm has a faster convergence rate compared to the other variants of LMS.

What’s in a looking time preference?

Looking time preference methods are an ubiquitous tool for tapping into infants' early skills and knowledge. However, predicting what preference infants will show in these paradigms can be difficult, and studies investigating the same ability oftentimes report opposing patterns of preference. For example, most studies investigating infant pattern learning report preferences for novel stimuli, but some report preference for familiar stimuli. How should such differences in preference direction be interpreted? One possibility is that any statistically significant preference is evidence for discrimination, such that all preferences provide similar evidential value. But another possibility is that the less-frequent preferences are so-called “sign errors”, in which a result is statistically significant, but the estimated effect size has the incorrect sign, e.g., showing a familiarity rather than novelty preference. In this paper, we use meta-analytic methods and statistical modeling to examine whether, when literatures show a heterogeneous pattern of looking time preferences, those preferences provide consistent evidential value, or whether one direction of preference may be a sign error. [Summary of included meta-analyses, results, implications]

Corrigendum: Spatial and temporal characterisation of sea-ice deformation

Correcting a sign error results in no changes to the key conclusions of Hutchings and others (2011). However, there is an improved agreement with previous work. Mean total sea-ice deformation scales log linearly with distance and the scaling exponent was found to be dependent on time. We find a linear relationship between the temporal scale and spatial scaling exponent, for timescales of an hour to a day. Extrapolating to the timescales of deformation resolved by RADARSAT, we find total deformation and distance scale with an exponent of between −0.16 and −0.19.