How accurate are gender detection tools in predicting the gender for Chinese names? A study with 20,000 given names in Pinyin format

Objective:We recently showed that the gender detection tools NamSor, Gender API, and Wiki-Gendersort accurately predicted the gender of individuals with Western given names. Here, we aimed to evaluate the performance of these tools with Chinese given names in Pinyin format.Methods:We constructed two datasets for the purpose of the study. File #1 was created by randomly drawing 20,000 names from a gender-labeled database of 52,414 Chinese given names in Pinyin format. File #2, which contained 9,077 names, was created by removing from File #1 all unisex names that we were able to identify (i.e., those that were listed in the database as both male and female names). We recorded for both files the number of correct classifications (correct gender assigned to a name), misclassifications (wrong gender assigned to a name), and nonclassifications (no gender assigned). We then calculated the proportion of misclassifications and nonclassifications (errorCoded).Results:For File #1, errorCoded was 53% for NamSor, 65% for Gender API, and 90% for Wiki-Gendersort. For File #2, errorCoded was 43% for NamSor, 66% for Gender API, and 94% for Wiki-Gendersort.Conclusion:We found that all three gender detection tools inaccurately predicted the gender of individuals with Chinese given names in Pinyin format and therefore should not be used in this population.     

The effects of work:rest duration on physiological and perceptual responses during intermittent exercise and performance

In this study, we examined the effects of different work:rest durations during 20 min intermittent treadmill running and subsequent performance. Nine males (mean age 25.8 years, s = 6.8; body mass 73.9 kg, s = 8.8; stature 1.75 m, s = 0.05; VO(2max) 55.5 ml x kg(-1) x min(-1), s = 5.8) undertook repeated sprints at 120% of the speed at which VO(2max) was attained interspersed with passive recovery. The work:rest ratio was constant (1:1.5) with trials involving either short (6:9 s) or long (24:36 s) work:rest exercise protocols (total exercise time 8 min). Each trial was followed by a performance run to volitional exhaustion at the same running speed. Testing order was randomized and counterbalanced. Heart rate, oxygen consumption, respiratory exchange ratio, and blood glucose were similar between trials (P > 0.05). Blood lactate concentration was greater during the long than the short exercise protocol (P < 0.05), whereas blood pH was lower during the long than the short exercise protocol (7.28, s = 0.11 and 7.30, s = 0.03 at 20 min, respectively; P < 0.05). Perceptions of effort were greater throughout exercise for the long than the short exercise protocol (16.6, s = 1.4 and 15.1, s = 1.6 at 20 min, respectively; P < 0.05) and correlated with blood lactate (r = 0.43) and bicarbonate concentrations (r = 0.59; P < 0.05). Although blood lactate concentration at 20 min was related to performance time (r = - 0.56; P < 0.05), no differences were observed between trials for time to exhaustion (short exercise protocol: 95.8 s, s = 30.0; long exercise protocol: 92.0 s, s = 37.1) or physiological responses at exhaustion (P > 0.05). Our results demonstrate that 20 min of intermittent exercise involving a long work:rest duration elicits greater metabolic and perceptual strain than intermittent exercise undertaken with a short work:rest duration but does not affect subsequent run time to exhaustion.     

The effects of arm crank strategy on physiological responses and mechanical efficiency during submaximal exercise

The purpose of this study was to compare submaximal physiological responses and indices of mechanical efficiency between asynchronous and synchronous arm ergometry. Thirteen wheelchair-dependent trained athletes performed eight steady-state incremental bouts of exercise (0 to 140 W), each lasting 4 min, using synchronous and asynchronous arm-cranking strategies. Physiological measures included oxygen uptake (VO2), heart rate, and blood lactate concentration. The power outputs corresponding to fixed whole blood lactate concentrations of 2.0 to 4.0 mmol x l(-1) were calculated using linear interpolation. Mechanical efficiency indices - gross efficiency, net efficiency, and work efficiency - were also calculated. An analysis of variance with repeated measures was applied to determine the effect of crank mode on the physiological parameters. Oxygen uptake was on average 10% lower (P < 0.01), and both net efficiency (P < 0.01) and gross efficiency (P < 0.01) were higher, during the asynchronous strategy at both 60 and 80 W (gross efficiency: 16.9 +/- 2.0% vs. 14.7 +/- 2.4% and 17.5 +/- 1.8% vs. 15.9 +/- 2.6% at 60 and 80 W respectively). There were no differences in heart rate, blood lactate concentration or power output at either of the blood lactate reference points between the asynchronous and synchronous strategies (P > 0.05). In conclusion, test specificity is an important consideration. If a synchronous strategy is to be adopted, it is likely to result in lower efficiency than an asynchronous strategy. The exercise testing scenario may help dictate which method is ultimately chosen.     

Seasonal progression and variability of repeat-effort line-drill performance in elite junior basketball players

To determine gender differences, positional differences, and patterns of change in the performance of the basketball line-drill test, 93 male (mean age 16.8 years, s = 1.1) and 95 female (mean age 16.5 years, s = 1.0) basketball players undertook 516 line-drill tests over a 5-year period. Log-transformed performance times were analysed using a mixed model that included quadratic within-participant fixed effects for time in the season and time in the programme. Changes and differences were standardized for interpretation of magnitudes. Mean performance times were 28.0 s (s = 1.3) for males and 30.4 s (s = 1.3) for females. The mean pattern of change in performance within a season differed substantially between the sexes and playing positions: male guards and female centres showed moderate to very large improvements mid-season of 1.1% and 3.5% respectively (90% confidence limits +/-2.1% and +/-3.0%), while female guards and male forwards showed large to very large decrements of -1.6% (+/-2.6%) and -2.4% (+/-2.0%). Over 3 years, males improved performance across all three playing positions by 1.4% (+/-1.3%) and females by 2.9% (+/-1.4%). Males improved performance by 0.2% (+/-0.5%) per year, whereas the performance of females deteriorated by 0.6% (+/-0.4%) per year. The differing patterns of performance change presumably reflect variations in training and competition loads, with short-term fluctuations in performance being managed to promote longer-term improvements.     

A knock to the system: A new sociotechnical systems approach to sport-related concussion

Broader contextual factors that influence concussion management have tended to be overlooked. To address this, the present study used a sociotechnical systems approach to identify perceived responsibilities and applied strategies for three domains of concussion management (i.e., prevention, identification and treatment). Participants were 118 members of the community rugby union system in Australia (69.2% male). Participants from throughout the rugby system (e.g., players, parents, coaches, club management) were asked open-ended questions about their perceived responsibilities and the strategies they use for concussion management. It was found that (a) proper training, technique correction and education were recurrent prevention themes; (b) the majority of key stakeholders felt that they could consistently identify concussion; however, medical aids (medics) were the only system actors who stated a responsibility to use standardised concussion assessment measures and (c) less than one third of the respondents indicated their involvement in treating concussion. This study identifies specific junctures in the system that prevents effective concussion management strategies. A sociotechnical systems approach improves the understanding of concussion prevention, and management beliefs and behaviours.     

The validity and reliability of a sample of 10 Wattbike cycle ergometers

ABSTRACT

The purpose of the study was to assess the validity and inter-bike reliability of 10 Wattbike cycle ergometers, and to assess the test–retest reliability of one Wattbike. Power outputs from 100 to 1000 W were applied using a motorised calibration rig (LODE) at cadences of 70, 90, 110 and 130 rev · min^?1, which created nineteen different intensities for comparison. Significant relationships (P < 0.01, r² = 0.99) were found between each of the Wattbikes and the LODE. Each Wattbike was found to be valid and reliable and had good inter-bike agreement. Within-bike mean differences ranged from 0.0 W to 8.1 W at 300 W and 3.3 W to 19.3 W at 600 W. When taking into account the manufacturers stated measurement error for the LODE (2%), the mean differences were less than 2%. Comparisons between Wattbikes at each of the nineteen intensities gave differences from 0.6 to 25.5 W at intensities of 152 W and 983 W, respectively. There was no significant difference (P > 0.05) between the measures of power recorded in the test–retest condition. The data suggest that the Wattbike is an accurate and reliable tool for training and performance assessments, with data between Wattbikes being able to be used interchangeably.