Novel prediction methods should always be compared to a baseline to determine their performance. Without this frame of reference, the performance score of a model is basically meaningless. What does it mean when a model achieves an $F_1$ of 0.8 on a test set? A proper baseline is, therefore, required to evaluate the ‘goodness’ of a performance score. Comparing results with the latest state-of-the-art model is usually insightful. However, being state-of-the-art is dynamic, as newer models are continuously developed. Contrary to an advanced model, it is also possible to use a simple dummy classifier. However, the latter model could be beaten too easily, making the comparison less valuable. Furthermore, most existing baselines are stochastic and need to be computed repeatedly to get a reliable expected performance, which could be computationally expensive. We present a universal baseline method for all binary classification models, named the Dutch Draw (DD). This approach weighs simple classifiers and determines the best classifier to use as a baseline. Theoretically, we derive the DD baseline for many commonly used evaluation measures and show that in most situations it reduces to (almost) always predicting either zero or one. Summarizing, the DD baseline is general, as it is applicable to any binary classification problem; simple, as it can be quickly determined without training or parameter tuning; and informative, as insightful conclusions can be drawn from the results. The DD baseline serves two purposes. First, it is a robust and universal baseline that enables comparisons across research papers. Second, it provides a sanity check during the prediction model’s development process. When a model does not outperform the DD baseline, it is a major warning sign.

There is serious concern for the future of a wide range of birds in Java and elsewhere in Indonesia due to both loss of habitat and trapping for the cagebird trade (the so-called “Asian Songbird Crisis”). Despite this concern, few data on presence and abundance of key species exist. We provide such data on 184 bird species from over two years of biodiversity surveys from 37 sites on 12 mountains in West and Central Java. Many of these species are heavily traded, endemic, and globally threatened. Several of the threatened endemics, notably Javan Trogon and Javan Cochoa, were often recorded, in terms of both geographical spread and numerical abundance. Rufous-fronted Laughingthrush, Spotted Crocias, and Orange-spotted Bulbul, believed to be threatened by trapping for the songbird trade, appear to remain fairly widespread. By contrast, Brown-cheeked Bulbul, Chestnut-backed (Javan) Scimitar-babbler, Javan Oriole, and especially Javan Blue-flycatcher, recorded on just a single occasion, and Javan Green Magpie, which we failed to record with certainty, now appear to be extremely rare. Our encounter rates, while not pinned to specific mountains for security reasons, represent an important baseline against which future changes in abundance can be gauged.

The New York Bight is undergoing rapid anthropogenic change amidst an apparent increase in baleen whale sightings. Though survey efforts have increased in recent years, the lack of published knowledge on baleen whale occurrence prior to these efforts impedes effective assessments of distributional or behavioural shifts due to increasing human activities. Here we synthesize opportunistic sightings of baleen whales from 1998–2017, which represent the majority of sightings data prior to recent survey efforts, and which are largely unpublished. Humpback and fin whales were the most commonly sighted species, followed by North Atlantic right whales and North Atlantic minke whales. Important behaviours such as feeding and nursing were observed, and most species (including North Atlantic right whales) were seen during all seasons. Baleen whales overlapped with multiple anthropogenic use areas, and all species, but of particular importance North Atlantic right whales, were sighted outside the spatial and temporal bounds of the Seasonal Management Areas for North Atlantic right whales. These opportunistic data are vital for providing a baseline and context of baleen whales in the New York Bight prior to broad-scale efforts and facilitate interpretation of current and future observations and trends, which can more accurately inform effective management and mitigation efforts.

The sophisticated spatial reference tools that exist today greatly facilitate studies of spatial ecology. Historically, however, the lack of such tools meant that spatial data were often imprecise, ambiguous or sometimes inaccurate. This can hinder or confound assessments of whether species distributions have changed in the past over decadal timescales. This is the case for Adélie penguins breeding at the southern limit of their breeding range in East Antarctica. In this short note, we resolve uncertainties in the locations of Adélie penguin breeding sites observed in the first population survey in Prydz Bay in 1981 by examining the original working notes of that work in combination with data from a recent survey in 2009 and a recently published spatial reference and identification system for coastal East Antarctica. By clarifying the historical locations, we conclude that the Adélie penguin breeding distribution has remained unchanged in this region over the past three decades, and we provide a robust baseline for assessing change in the future.

This study introduces a new real-time kinematic (RTK) positioning method which is suitable for baselines of different lengths. The method merges carrier-phase wide-lane, and ionosphere-free observation combinations (LWLC) instead of using pseudo-range, and carrier-phase ionosphere-free combination (PCLC), or single-frequency pseudo-range and phase combination (P1L1). In a first step, the double-differenced wide-lane ambiguities were calculated and fixed using the pseudo-range and carrier-phase wide-lane combination observations. Once the double-differenced wide-lane integer ambiguities were known, the wide-lane combined observations were regarded as accurate pseudo-range observations. Subsequently, the carrier-phase wide-lane, and ionosphere-free combined observations were used to fix the double-differenced carrier-phase integer ambiguities, achieving the final RTK positioning. The RTK positioning analysis was performed for short, medium, and long baselines, using the P1L1, PCLC, and LWLC methods, respectively. For a short baseline, the LWLC method demonstrated positioning accuracy similar to the P1L1 method, and performed better than the PCLC method. For medium and long baselines, the positioning accuracy of the LWLC method was slightly higher than those of the PCLC and P1L1 methods. In conclusion, the LWLC method provided high-precision RTK positioning results for baselines with different lengths, as it used high-precision carrier-phase observations with fixed ambiguities instead of low-precision pseudo-range observations.

The longitudinal relationship between muscle strength, dietary intake and physical activity among adolescents is not well understood. We investigated the trend and longitudinal effects of dietary intakes and physical activity scores on muscle strength in adolescents. This prospective cohort study consisted of 436 adolescents (134 males; 302 females) aged 13 years at baseline (2012) who were followed up at the ages of 15 (2014) and 17 (2016) years, respectively. We measured muscle strength using a calibrated hand dynamometer, estimated dietary intake with a 7-d dietary history and physical activity scores with a validated physical activity questionnaire for older children. A generalised estimating equation was used to examine the effect of dietary intakes and physical activity on muscle strength changes. The analysis was performed separately by sex. The muscle strength for males and females had increased within the 5-year period. The dietary intakes (energy and macronutrients) also increased initially but plateaued after the age of 15 years for both sexes. Females recorded a significant declining trend in physical activity scores compared with males as they grew older. A significant positive longitudinal relationship was found between protein (β = 0·035; P = 0·016), carbohydrate intake (β = 0·002; P = 0·013) and muscle strength among males. However, no longitudinal relationship was found between dietary intake, physical activity and muscle strength among females. Higher protein and carbohydrate intake among males was associated with higher muscle strength but was not observed in females. Nutrition and physical activity focusing on strength building are required in early adolescence and need to be tailored to males and females accordingly.

Coral reefs have experienced extensive degradation across the world over the last 50 years as a result of a variety of stressors operating at a range of spatial and temporal scales. In order to assess whether declines are continuing, or if reefs are recovering, detailed baseline information is required from across wide spatial scales. Unfortunately, for some regions this information is not readily available, making future reef trajectories difficult to determine. Here we characterized the current benthic community state for coral reefs in the Wakatobi region of Indonesia, one of the most biodiverse marine regions in the world. We surveyed 10 reef sites (5, 10 and 15 m depth) to explore spatial variation in coral reef benthic communities and provide a detailed baseline. Previous data (2002–2011) were available for coral, sponges, algae and soft coral at six of our study sites. Using this information, we determined if any changes had occurred in dominance of these benthic groups. We found that benthic assemblage composition differed significantly over relatively small spatial scales (2–10 km) and hard coral cover was highly variable, ranging from 7–48% (average 19.5% ± 1.5 SE). While coral cover appears to have declined at all sites where data were available since 2002, we found little evidence for widespread increases in other benthic groups or regime shifts. Our study provides a comprehensive baseline dataset for the region that can be used in the future to determine rates of change in benthic communities.

Inadequate protein quality may be a risk factor for poor growth. To examine the effect of a macronutrient–micronutrient supplement KOKO Plus (KP), provided to infants from 6 to 18 months of age, on linear growth, a single-blind cluster-randomised study was implemented in Ghana. A total of thirty-eight communities were randomly allocated to receive KP (fourteen communities, n 322), a micronutrient powder (MN, thirteen communities, n 329) and nutrition education (NE, eleven communities, n 319). A comparison group was followed cross-sectionally (n 303). Supplement delivery and morbidity were measured weekly and anthropometry monthly. NE education was provided monthly. Baseline, midline and endline measurements at 6, 12 and 18 months included venous blood draws, diet, anthropometry, morbidity, food security and socio-economics. Length-for-age Z-score (LAZ) was the primary outcome. Analyses were intent-to-treat using mixed-effects regressions adjusted for clustering, sex, age and baseline. No differences existed in mean LAZ scores at endline (−1·219 (sd 0·06) KP, −1·211 (sd 0·03) MN, −1·266 (sd 0·03) NE). Acute infection prevalence was lower in the KP than NE group (P = 0·043). Mean serum Hb was higher in KP infants free from acute infection (114·02 (sd 1·87) g/l) than MN (107·8 (sd 2·5) g/l; P = 0·047) and NE (108·8 (sd 0·99) g/l; P = 0·051). Compliance was 84·9 % (KP) and 87·2 % (MN) but delivery 60 %. Adjusting for delivery and compliance, LAZ score at endline was significantly higher in the KP v. MN group (+0·2 LAZ; P = 0·026). A macro- and micronutrient-fortified supplement KP reduced acute infection, improved Hb and demonstrated a dose–response effect on LAZ adjusting consumption for delivery.

Beetroot juice (BJ) consumption has been associated with improved cardiovascular health owing to an increase in NO bioconversion. This study evaluates the effect of BJ consumption on macrovascular endothelial function (flow-mediated dilation (FMD)) and muscle oxygen saturation (StO2) parameters in pregnant women within a randomised, crossover, double-blind design in which twelve pregnant women consumed a single dose (140 ml) of BJ or placebo (PLA). Urinary nitrate was assessed before (T0) and 150 min after BJ/PLA consumption. FMD was used to evaluate macrovascular endothelial function, and near-IR spectroscopy was used to evaluate muscle StO2 parameters during the occlusion and reperfusion phases, which were taken at baseline (PRE) and 120 and 140 min after BJ/PLA consumption, respectively. A significant increase in urinary nitrate was observed at 150 min after BJ consumption when compared with T0 (BJ: 0·20 (sd 0·13) v. T0: 0·02 (sd 0·00), P=0·000) and PLA intervention (PLA: 0·02 (sd 0·00), P=0·001). FMD improved after BJ consumption when compared with PRE (BJ: 11·00 (sd 1·67) v. PRE: 5·53 (sd 1·17), P=0·000) and PLA (5·34 (sd 1·31), P=0·000). No significant difference between PLA and PRE in FMD (P=1·000) was observed. In StO2 parameters, a difference was not observed after BJ consumption compared with PRE and PLA intervention. The data demonstrate that a single dose of 140 ml of BJ consumption improves macrovascular endothelial function, but not StO2 parameters.

Pertussis morbidity is highest in infants too young to be fully protected by routine vaccination schedules. Alternate vaccine strategies are required to maximise protection in this age-group. To understand baseline pertussis epidemiology prior to the introduction of the maternal pertussis vaccination program in 2014, we conducted a retrospective case series analyses of 53 901 notifications and temporal trends from 1997 to 2014. Notifications were highest in infants younger than 4 months of age and highest annual notification rates in infants younger than 1 month of age (308/100 000 per year). Amongst Aboriginal and Torres Strait Islander infants aged younger than 1 month, this rate was 576/100 000 per year. Notification rates were 40% higher amongst women 15–44 years, 62·4/100 000 population compared with men (44·5/100 000) and 90% higher in Aboriginal and Torres Strait Islander women of the same age (38·2/100 000) compared with men (19·7/100 000). Six infant deaths were identified, all younger than 2 months of age. Monitoring epidemiology in at-risk groups – infants too young to be vaccinated, women of childbearing age and Aboriginal and Torres Strait Islander peoples – following implementation of the maternal pertussis vaccination program will be important to assess its impact and safety.

Objectives: The purpose of this study was to assess the contribution of socioeconomic status (SES) and other multivariate predictors to baseline neurocognitive functioning in collegiate athletes. Methods: Data were obtained from the Concussion Assessment, Research and Education (CARE) Consortium. Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) baseline assessments for 403 University of Florida student-athletes (202 males; age range: 18–23) from the 2014–2015 and 2015–2016 seasons were analyzed. ImPACT composite scores were consolidated into one memory and one speed composite score. Hierarchical linear regressions were used for analyses. Results: In the overall sample, history of learning disability (β=−0.164; p=.001) and attention deficit–hyperactivity disorder (β=−0.102; p=.038) significantly predicted worse memory and speed performance, respectively. Older age predicted better speed performance (β=.176; p<.001). Black/African American race predicted worse memory (β=−0.113; p=.026) and speed performance (β=−.242; p<.001). In football players, higher maternal SES predicted better memory performance (β=0.308; p=.007); older age predicted better speed performance (β=0.346; p=.001); while Black/African American race predicted worse speed performance (β=−0.397; p<.001). Conclusions: Baseline memory and speed scores are significantly influenced by history of neurodevelopmental disorder, age, and race. In football players, specifically, maternal SES independently predicted baseline memory scores, but concussion history and years exposed to sport were not predictive. SES, race, and medical history beyond exposure to brain injury or subclinical brain trauma are important factors when interpreting variability in cognitive scores among collegiate athletes. Additionally, sport-specific differences in the proportional representation of various demographic variables (e.g., SES and race) may also be an important consideration within the broader biopsychosocial attributional model. (JINS, 2018, 24, 1–10)

Policymakers are concerned about nitrogen and phosphorus export to water bodies. Exports may be reduced by paying farmers to adopt practices to reduce runoff or by paying performance incentives tied to estimated run-off reductions. We evaluate the cost-effectiveness of practice and performance incentives for reducing nitrogen exports. Performance incentives potentially improve farm-level and allocative efficiencies relative to practice incentives. However, the efficiency improvements can be undermined by baseline shifts when growers adopt crops that enhance the performance payments but cause more pollution. Policymakers must carefully specify rules for performance-incentive programs and payments to avoid such baseline shifting.

In our opinion, the challenges of ongoing measurement, the ever-moving behavioral baseline, and strategic self-ignorance return us full-circle to a sensible point made by Peter Bohm–Benefit-Cost Analysis for environmental goods should use an interval method.

Human activities have increasingly affected biodiversity in the Mediterranean Sea. Data on the distribution and abundance of species allows researchers to assess the possible degradation of wild populations. These data could act as a baseline to assess the magnitude of the effects of human activities on a bioindicator species. The distribution and relative abundance of the south-western populations of the endemic Astroides calycularis in the South Iberian Peninsula were studied to establish a baseline for future studies. The rocky shoreline was studied at a depth range of 0–12 m, including more than 650 km of Spain's Andalusian coastline. The species was present in 135 of the 585 dive points sampled. ANOVA analysis showed differences in depth in the four provinces studied, and there was no interaction between the two factors. As human activities on the Mediterranean coast are reducing the A. calycularis populations, a baseline on marine populations is greatly recommended for monitoring, assessment, and management studies, especially for endangered or bioindicator species. This baseline could be useful as a reference tool to assess the effects of human activities on marine biodiversity, including global change.

The present study investigates the species distribution patterns and macrofaunal assemblages along the English Channel and the environmental factors contributing to observed patterns. Seven distinct macrofaunal assemblages were identified based on Hamon grab samples. In the western Channel, an Echinocyamus/Nemertea assemblage dominated, giving way to an Abra/Scalibregma assemblage in inshore waters. A Verruca/Sabellaria assemblage was identified to occupy large regions of the mid-Channel while those of the east Channel were more spatially variable. While variations in depth, sediment particle size and wave and tidal stress were significantly related to variations in assemblage structure, temperature did not appear to be greatly influential. In addition, acoustic data obtained for each station sampled allowed us to determine how representative the observed biological communities are over spatial scales somewhat larger than the sampling points. In contrast to earlier comparable studies using anchor dredges, the biological communities sampled in this study did not exhibit a large east–west distinction; the most abundant species and key assemblages were found to be present along the length of the Channel. The possible reasons for these differences with earlier studies are discussed.

Hydrocarbons in the environment occur as complex mixtures. The detection of low-level anthropogenic contamination can be difficult as many of the compounds can also be found in biogenic materials. Results obtained in the Antarctic show that accepted methods of distinguishing biogenic from contaminating hydrocarbons are inconsistent. The samples taken from South Georgia in the vicinity of derelict whaling stations had odd-to-even carbon number ratios in the range 0.8–1.0. An offshore site had the highest levels for phytane and the lowest for pristane. The pristane-to-phytane ratio varied between 0.4 and 1.4 and contradicted the interpretation of the odd-to-even ratio at several sites. A value of less than 1.0 for these ratios has been taken to indicate anthropogenic origin. A number of polycyclic aromatic hydrocarbons were detected at all the sites, but the concentration gradient did not correspond with that of the n-alkanes. It is therefore recommended that recognition of anthropogenic input is based on quantification of all compounds against a well-defined baseline. The source of the pollution can then be estimated from the distribution patters of the hydrocarbons.