Observable Variables Research Articles

Actigraphic sleep patterns are associated with bone turnover and bone mineral density among university students.

Poorer sleep is associated with poorer bone health among older adults but the role of sleep in bone health during younger adulthood is understudied. In this observational study, the averages and variability in total sleep time (TST), sleep efficiency (SE), and sleep midpoint of university students were examined in relation to levels of bone turnover markers (BTMs) and bone mineral density (BMD) at the lumbar spine and femur. A sample of healthy, university students (N = 59, aged 18-25 years, 51.8% female, body mass index <30 kg/m2 ), wore a wrist actigraph for 7 days, completed a dual-energy X-ray absorptiometry scan, and underwent blood sampling to assess serum BTM concentrations of osteocalcin (OC) and N-terminal telopeptide of type 1 collagen. A sub-sample (n = 14) completed a one-year follow-up. Multiple regression models examined the associations between each sleep metric and bone health outcome at baseline and 1-year follow-up. At baseline, greater variability in sleep midpoint was cross-sectionally associated with greater OC (β = 0.21, p = 0.042). In the exploratory, follow-up sub-sample, lower average TST (β = -0.66, p = 0.013) and SE (β = -0.68, p = 0.01) at baseline were associated with greater increases in OC at follow-up. Greater delays in mean sleep midpoint over follow-up were significantly associated with decreases in lumbar spine BMD (β = -0.49, p = 0.03). In a sample of young adults, variable sleep schedules were associated with greater bone turnover suggesting the potential importance of regular sleep for optimising bone health into early adulthood.

Seeing and turbulence profile simulations over complex terrain at the Thai National Observatory using a chemistry-coupled regional forecasting model

Abstract This study utilized advanced numerical simulations with the Weather Research and Forecasting model coupled with Chemistry (WRF-Chem) to predict anticipated astronomical seeing conditions at the Thai National Observatory (TNO). The study evaluated the effects of both gas-phase and aerosol-phase chemical processes in the Earth’s atmosphere, along with the impact of spatial and temporal resolution on model performance. These simulations were validated against measurements from the Differential Image Motion Monitor (DIMM) and the Slope Detection and Ranging (SLODAR) technique. Due to the inherent temporal variability of the DIMM observations, a 24-h moving average window was applied to both DIMM data and WRF-Chem model outputs. This reduced the percentage root-mean-square error in the comparison between the two data sets from 23 per cent to 11 per cent and increased the correlation coefficient from 0.21 to 0.59. Chemistry played a minor role during the study period, contributing 3.49 per cent to astronomical seeing. However, it did affect the model’s accuracy. Additionally, the study revealed that higher spatial and temporal resolution simulations did not necessarily improve the model’s accuracy. When compared to SLODAR observations of the refractive index structure constant (Cn2dh), the simulations captured altitude variations within ±25 per cent above 5 km and 25–50 per cent below 5 km. Dome seeing also played a role, contributing to around 90 per cent or more in the lowest altitude layer. The results emphasized the significance of seeing predictions in providing valuable insights into complex atmospheric phenomena and how to mitigate the effects of atmospheric turbulence on telescopes.

Quality and Quantity of Hydroponic Rice Forage Ciherang and IR64 Varieties at Different Harvest Ages

The purpose of this study is to determine how different harvest ages and varieties affect the quantity and quality of fodder produced by hydroponically cultivated rice plants. The study design was a 2 x 3 factorial, completely randomized design (CRD) experiment with five replications. Tiller count, plant height, plant weight, and graph power were among the characteristics that were observed. The data were analyzed using analysis of variance (ANOVA), and any differences were found using Duncan’s multiple area test. The plant height, number of tillers, plant weight, and dry matter production were all significantly impacted by the harvest age factor, according to the data. The variety factor is significantly influenced by the observation variables. The interaction between the parameters at 60 days of yield after planting (DYAP) has a significant impact on the variables plant height, plant weight, number of tillers, and dry matter production. The IR64 variety outperformed the Ciherang variety, according to the results of the proximate analysis, and the best harvest age in terms of crude protein, crude fiber, and crude fat content was at 40 days after planting, while the best TDN was at 60 DYAP. The IR64 cultivar produced the most biomass at a harvest age of 60 DYAP. In summary, the study’s findings suggest that hydroponically grown paddy forage has great promise in terms of the nutrients it can produce, yet biomass output is still lower than that of cutter grass, while dry matter production is higher than that of fodder.

Evidence of validation in a strategy development model structure in public health schools

Background Scientific investigations utilizing the questionnaire methodology for data acquisition mandate a rigorous validation process for their research instruments. This imperative arises from the need to mitigate biases and uphold impartiality in the derived results. Regrettably, a lack of awareness persists among researchers regarding this pivotal requirement within the methodological framework. Although comprehensive validation typically unfolds across multiple stages, the present study is dedicated exclusively to the structural validation phase. Methods A questionnaire with 18 closed questions (observable variables) was administered across 12 units in different regions of Brazil. Data were classified according to a Likert scale, which ranges from 1 to 5, representing the participants’ degree of agreement in relation to 8 characteristics proposed for the model (latent variables). The research involved 75 participants who answered the research instrument between March and August 2023. To define the structure of the model, the data were subjected to an exploratory factor analysis, with the support of the Factor Analysis software. Results Findings suggested a structure was organized into two factors, called internal and external factors, according to the degree of governability that the SPH has in relation to the proposed items. Conclusion The validation evidence indicates a structure composed of two factors for the better organization of the items of a strategy development model in Schools of Public Health, in Brazil.

Vertical Velocity Diagnosed From Surface Data With Machine Learning

AbstractSubmesoscale vertical velocities, w, are important for the oceanic transport of heat and biogeochemical properties, but observing w is challenging. New remote sensing technologies of horizontal surface velocity at O(1) km resolution can resolve surface submesoscale dynamics and offer promise for diagnosing w subsurface. Using machine learning models, we examine relationships between the three‐dimensional w field and remotely observable surface variables such as horizontal velocity, density, and their horizontal gradients. We evaluate the machine learning models' sensitivities to different inputs, spatial resolution of surface fields, the addition of noise, and information about the subsurface density. We find that surface data is sufficient for reconstructing w, and having high resolution horizontal velocities with minimal errors is crucial for accurate w predictions. This highlights the importance of finer scale surface velocity measurements and suggests that data‐driven methods may be effective tools for linking surface observations with vertical velocity and transport subsurface.

Intelligent control decision integrating fuzziness and randomness for automatic management of cash flow.

Automatic management of cash flow from the perspective of cybernetics decisions can improve work efficiency and accuracy of cash flow management. Disadvantage of traditional fuzzy control method is that it only expresses fuzziness and ignores randomness. The automatic management of cash flow involves variables representing the fuzziness and randomness of human cognition which need new calculation methods to solve. Based on fuzzy control this paper proposes a cloud set control decision method for cash flow management. Cloud set and its I operation and P operation are described. Methods are studied including observation variables and control variables, fuzziness of observation variables and control variables, description of rules, and cloud reasoning based on cloud set. The method is applied successfully in automatic management of cash flow in which control amount of expenditure intensity is -2.285. It is shown that this method can effectively obtain reasonable control quantities considering fuzzy and random properties by the comparison with fuzzy control method. The method for automatic management of cash flow proposed has greater objectivity and effectiveness for the integration of fuzzy and randomness representing human cognition and decision.

Potential of The Biological Agent Bacillus spp in Inhibiting Fusarium Wild Disease and its Effects on The Growth and Production of Cayenne Papper (Capsicum frutescens L.)

Cayenne pepper (Capsicum frutescens L.) is a horticultural crop in Indonesia that often has many obstacles. The main obstacle that is often faced is wilt disease caused by the fungus Fusarium oxysporum. Bacillus spp. bacteria have the potential to control the fungus F. oxysporum. The purpose of this study was to determine the ability of Bacillus spp. bacteria in controlling fusarium wilt disease. This research was conducted from August to November 2023 at the Plant Health Laboratory and screen house of the Faculty of Agriculture, National Development University "Veteran" East Java. The study used a Factorial Complete Randomized Block Design with three replications and nine treatments so that there were 27 experimental units. The observation variables included incubation period, disease intensity, plant height, number of leaves, first flower appearance and fruit wet weight. The research data were analyzed using ANOVA analysis of variance procedure to determine whether there is an effect of each treatment. If it is known that there is a significantly different effect from the treatment, it will be continued with the DMRT test at the 5% level. The results showed that the combination of Bacillus spp bacteria and dosage had an effect on the incubation period and disease intensity. The results showed that the combination of Bacillus spp bacteria and doses had an effect on vegetative phase parameters, namely plant height and number of leaves. However, the combination of Bacillus spp and dosage showed that the generative phase parameters did not significantly affect the first flower appearance and fruit wet weight

Automated mitral inflow Doppler peak velocity measurement using deep learning

Doppler echocardiography is a widely utilised non-invasive imaging modality for assessing the functionality of heart valves, including the mitral valve. Manual assessments of Doppler traces by clinicians introduce variability, prompting the need for automated solutions. This study introduces an innovative deep learning model for automated detection of peak velocity measurements from mitral inflow Doppler images, independent from Electrocardiogram information. A dataset of Doppler images annotated by multiple expert cardiologists was established, serving as a robust benchmark. The model leverages heatmap regression networks, achieving 96% detection accuracy. The model discrepancy with the expert consensus falls comfortably within the range of inter- and intra-observer variability in measuring Doppler peak velocities. The dataset and models are open-source, fostering further research and clinical application.

Polish version of the questionnaire for measuring post-traumatic stress disorders PTSD-8 in studies of four groups of medical professions

There are several techniques to diagnose post-traumatic stress disorder (PTSD) symptoms that correspond to the Diagnostic and Statistical Manual of Mental Disorders (DSM) classification of mental disorders. This study aimed at testing the reliability and construct validity of the Polish version of the PTSD-8. The results obtained in four professional groups of healthcare workers (physicians, nurses, paramedics, other occupations) were compared. The PTSD-8 questionnaire includes questions on stress reactions manifested by intrusion (4 items), avoidance (2 items), or hypervigilance (2 items). Respondents were asked about their reactions to the traumatic work-related experiences occurring during the COVID-19 pandemic. The nationwide survey conducted in 2022 included 2303 employees, among whom 746 had such an experience and completed the PTSD-8 questionnaire. Based on confirmatory factor analysis (CFA) results, it was shown that a single-factor model is acceptable, but under the condition of modification by correlation of random errors at selected observable variables. The reliability of this scale is very good (McDonald's Ω = 0.890), and the goodness-of-fit of the overall CFA model is confirmed by: CMIN/DF = 3.969, NFI = 0.982, RFI = 0.964, IFI = 0.986, TLI = 0.983, RMSEA = 0.063. The model estimated for female nurses has the best psychometric properties. In the study group of 2303 respondents, 16.76% met the criteria for PTSD, accounting for 52.74% of those who were traumatized by a work-related event. The PTSD-8 index is M±SD 20.01±5.55 pts (range: 8-32 pts). The PTSD-8 index is significantly higher in women than in men, and higher in nurses compared to other professions. However, the differences between professional groups are significant only for women, and the largest when comparing female physicians and nurses. The analyses confirmed the rationale of further implementation of the PTSD-8 questionnaire in Poland, both as a tool for identifying PTSD and for studying the severity of the syndrome, its determinants and health effects. Med Pr Work Health Saf. 2024;75(1):45-56.

Noninvasive diagnosis of liver cirrhosis: qualitative and quantitative imaging biomarkers.

A diagnosis of cirrhosis initiates a shift in the management of chronic liver disease and affects the diagnostic workflow and treatment decision of primary liver cancer. Liver biopsy remains the gold standard for cirrhosis diagnosis, but it is invasive and susceptible to sampling bias and observer variability. Various qualitative and quantitative imaging biomarkers based on ultrasound, CT and MRI have been proposed for noninvasive diagnosis of cirrhosis. Qualitative imaging features are easy to apply but have moderate diagnostic sensitivity. Elastography techniques allow quantitative assessment of liver stiffness and are highly accurate for cirrhosis diagnosis. Ultrasound elastography are widely used in clinical practice, while MR elastography has narrower availability. Although not applicable in clinical practice yet, other quantitative imaging features, including liver surface nodularity, linear and volumetric measurement, extracellular volume fraction, liver enhancement on hepatobiliary phase, and parameters derived from diffusion-weighted imaging, can provide additional information of liver morphology, perfusion, and function, thus may increase diagnosis performance. The introduction of radiomics and deep learning has further improved diagnostic accuracy while reducing subjectivity. Several imaging features may also help to assess liver function and outcomes in patients with cirrhosis. In this review, we summarize the qualitative and quantitative imaging biomarkers for noninvasive cirrhosis diagnosis, and the assessment of liver function and outcomes, and discuss the challenges and future directions in this field.

Investigasi Aspek-aspek Promosi Penjualan

In today's highly developed e-commerce era, sales promotions have become a reliable marketing communication technique for many companies. There is also much research in this field. However, many researchers need to define and operationalize sales promotions as a research variable correctly. With critical review techniques, this study found that aspects of sales promotions that can be used as observation variables are perceived value, credibility, fairness, time frame, uncertainty, congruence, requirements, and consequences. Every sales promotion technique requires different aspects. The researchers should match the sales promotions technique properly with their relevant aspects.

Forest Canopy Height Estimation by Integrating Structural Equation Modeling and Multiple Weighted Regression

As an important component of forest parameters, forest canopy height is of great significance to the study of forest carbon stocks and carbon cycle status. There is an increasing interest in obtaining large-scale forest canopy height quickly and accurately. Therefore, many studies have aimed to address this issue by proposing machine learning models that accurately invert forest canopy height. However, most of the these approaches feature PolSAR observations from a data-driven viewpoint in the feature selection part of the machine learning model, without taking into account the intrinsic mechanisms of PolSAR polarization observation variables. In this work, we evaluated the correlations between eight polarization observation variables, namely, T11, T22, T33, total backscattered power (SPAN), radar vegetation index (RVI), the surface scattering component (Ps), dihedral angle scattering component (Pd), and body scattering component (Pv) of Freeman-Durden three-component decomposition, and the height of the forest canopy. On this basis, a weighted inversion method for determining forest canopy height under the view of structural equation modeling was proposed. In this study, the direct and indirect contributions of the above eight polarization observation variables to the forest canopy height inversion task were estimated based on structural equation modeling. Among them, the indirect contributions were generated by the interactions between the variables and ultimately had an impact on the forest canopy height inversion. In this study, the covariance matrix between polarization variables and forest canopy height was calculated based on structural equation modeling, the weights of the variables were calculated by combining with the Mahalanobis distance, and the weighted inversion of forest canopy height was carried out using PSO-SVR. In this study, some experiments were carried out using three Gaofen-3 satellite (GF-3) images and ICESat-2 forest canopy height data for some forest areas of Gaofeng Ridge, Baisha Lizu Autonomous County, Hainan Province, China. The results showed that T11, T33, and total backscattered power (SPAN) are highly correlated with forest canopy height. In addition, this study showed that determining the weights of different polarization observation variables contributes positively to the accurate estimation of forest canopy height. The forest canopy height-weighted inversion method proposed in this paper was shown to be superior to the multiple regression model, with a 26% improvement in r and a 0.88 m reduction in the root-mean-square error (RMSE).

Pengaruh Analisis Fundamental, Teknikal dan Bandarmologi terhadap Harga Saham Syariah yang Terdaftar di Jakarta Islamic Index Periode 2018-2021

Equity-based sharia stocks that comply with Islamic principles. One of the sharia stock indices is the Jakarta Islamic Index (JII). JII is an index consisting of the 30 most liquid Islamic stocks on the IDX which was launched in 2000. To select 30 Islamic stocks to be included in the index calculation, there are additional criteria carried out by IDX, namely liquidity criteria. The liquidity indicators used are based on market capitalization and daily transaction value. Usually calculated based on the last one year period and every day there must always be transactions. This study uses a quantitative descriptive research method, namely research that emphasizes the analysis of numerical data processed by statistical models. This method is called the quantitative method because the research data uses numbers that indicate the value of the observation variable that leads to a certain time period. This research results that from the results of the t test table it is known that the fundamental analysis variables partially influence the prices of Islamic stocks listed on the Jakarta Islamic Index (JII) in 2018-2021. From the results of the t test table it is known that the technical analysis variable has a partial effect on the prices of Islamic stocks listed on the Jakarta Islamic Index (JII) for 2018-2021. From the results of the t test table it is known that the bandarmological analysis variable has a partial effect on the prices of Islamic stocks listed on the Jakarta Islamic Index (JII) for 2018-2021. From the results of the f test table it is known that the variables of fundamental analysis, technical analysis, and bandarmological analysis simultaneously influence the prices of Islamic stocks listed on the Jakarta Islamic Index (JII) for 2018-2021.&#x0D; Keywords: Fundamental Analysis, Technical, Bandarmology, Sharia Stocks, Sharia Market

Machine learning prediction of factors affecting Major League Baseball (MLB) game attendance: algorithm comparisons and macroeconomic factor of unemployment

PurposeThis study aims to identify an automated machine learning algorithm with high accuracy that sport practitioners can use to identify the specific factors for predicting Major League Baseball (MLB) attendance. Furthermore, by predicting spectators for each league (American League and National League) and division in MLB, the authors will identify the specific factors that increase accuracy, discuss them and provide implications for marketing strategies for academics and practitioners in sport.Design/methodology/approachThis study used six years of daily MLB game data (2014–2019). All data were collected as predictors, such as game performance, weather and unemployment rate. Also, the attendance rate was obtained as an observation variable. The Random Forest, Lasso regression models and XGBoost were used to build the prediction model, and the analysis was conducted using Python 3.7.FindingsThe RMSE value was 0.14, and the R2 was 0.62 as a consequence of fine-tuning the tuning parameters of the XGBoost model, which had the best performance in forecasting the attendance rate. The most influential variables in the model are “Rank” of 0.247 and “Day of the week”, “Home team” and “Day/Night game” were shown as influential variables in order. The result was shown that the “Unemployment rate”, as a macroeconomic factor, has a value of 0.06 and weather factors were a total value of 0.147.Originality/valueThis research highlights unemployment rate as a determinant affecting MLB game attendance rates. Beyond contextual elements such as climate, the findings of this study underscore the significance of economic factors, particularly unemployment rates, necessitating further investigation into these factors to gain a more comprehensive understanding of game attendance.

Assessment of left ventricular ejection fraction in artificial intelligence based on left ventricular opacification.

Left ventricular opacification (LVO) improves the accuracy of left ventricular ejection fraction (LVEF) by enhancing the visualization of the endocardium. Manual delineation of the endocardium by sonographers has observer variability. Artificial intelligence (AI) has the potential to improve the reproducibility of LVO to assess LVEF. The aim was to develop an AI model and evaluate the feasibility and reproducibility of LVO in the assessment of LVEF. This retrospective study included 1305 echocardiography of 797 patients who had LVO at the Department of Ultrasound Medicine, Union Hospital, Huazhong University of Science and Technology from 2013 to 2021. The AI model was developed by 5-fold cross validation. The validation datasets included 50 patients prospectively collected in our center and 42 patients retrospectively collected in the external institution. To evaluate the differences between LV function determined by AI and sonographers, the median absolute error (MAE), spearman correlation coefficient, and intraclass correlation coefficient (ICC) were calculated. In LVO, the MAE of LVEF between AI and manual measurements was 2.6% in the development cohort, 2.5% in the internal validation cohort, and 2.7% in the external validation cohort. Compared with two-dimensional echocardiography (2DE), the left ventricular (LV) volumes and LVEF of LVO measured by AI correlated significantly with manual measurements. AI model provided excellent reliability for the LV parameters of LVO (ICC > 0.95). AI-assisted LVO enables more accurate identification of the LV endocardium and reduces observer variability, providing a more reliable way for assessing LV function.

Alternative Methods for Handling Iron (Fe) Deficiency of Oil Palms (Elaeis guineensis Jacq.) Grown in Sandy Soil

Iron (Fe) deficiency is one of the general problems that occur in Oil Palms (Elaeis guineensis Jacq.) which are planted on sandy soil. Fe is an important micronutrient in enzyme activities, including cytochromes in the process of electron transfer, chlorophyll synthesis, and in maintaining the chloroplast structure. The research was conducted in 2 series of field observations. The first observation was conducted on the application of clay mineral soils (Plinthic Kanhapludults and Lithic Dystrudepts) in a silt pit, while the second observation was on the application of liquid nano-sized Fe fertilizer through root infusion and trunk injection methods, in addition, Fe chelated fertilizer application were also observed. The observation variable was the level of Fe in leaves 17th. The results showed all treatments significantly increased levels of Fe in leaves with different plant response rates. Root infusion treatment shows a faster response than clay mineral soil treatment, namely at 2 - 4 months after application (50.3 ppm and 38.2 ppm). Meanwhile, clay mineral soil treatment has a slower plant response rate on increasing Fe level in the leaves with a significant difference against the control, precisely at 24 months after application. Normally, regular Fe fertilization is enough to fulfill plant nutrient needs. Application of Fe through trunk injection or root infusion can be an alternative treatment for short-term use, whereas the application of clay mineral (iron-rich) soil can be optional for long-term needs.

Digital Microscopy Augmented by Artificial Intelligence to Interpret Bone Marrow Samples for Hematological Diseases.

Analysis of bone marrow aspirates (BMAs) is an essential step in the diagnosis of hematological disorders. This analysis is usually performed based on a visual examination of samples under a conventional optical microscope, which involves a labor-intensive process, limited by clinical experience and subject to high observer variability. In this work, we present a comprehensive digital microscopy system that enables BMA analysis for cell type counting and differentiation in an efficient and objective manner. This system not only provides an accessible and simple method to digitize, store, and analyze BMA samples remotely but is also supported by an Artificial Intelligence (AI) pipeline that accelerates the differential cell counting process and reduces interobserver variability. It has been designed to integrate AI algorithms with the daily clinical routine and can be used in any regular hospital workflow.

Two is better than one: longitudinal detection and volumetric evaluation of brain metastases after Stereotactic Radiosurgery with a deep learning pipeline.

Close MRI surveillance of patients with brain metastases following Stereotactic Radiosurgery (SRS) treatment is essential for assessing treatment response and the current disease status in the brain. This follow-up necessitates the comparison of target lesion sizes in pre- (prior) and post-SRS treatment (current) T1W-Gad MRI scans. Our aim was to evaluate SimU-Net, a novel deep-learning model for the detection and volumetric analysis of brain metastases and their temporal changes in paired prior and current scans. SimU-Net is a simultaneous multi-channel 3D U-Net model trained on pairs of registered prior and current scans of a patient. We evaluated its performance on 271 pairs of T1W-Gad MRI scans from 226 patients who underwent SRS. An expert oncological neurosurgeon manually delineated 1,889 brain metastases in all the MRI scans (1,368 with diameters > 5mm, 834 > 10mm). The SimU-Net model was trained/validated on 205 pairs from 169 patients (1,360 metastases) and tested on 66 pairs from 57 patients (529 metastases). The results were then compared to the ground truth delineations. SimU-Net yielded a mean (std) detection precision and recall of 1.00±0.00 and 0.99±0.06 for metastases > 10mm, 0.90±0.22 and 0.97±0.12 for metastases > 5mm of, and 0.76±0.27 and 0.94±0.16 for metastases of all sizes. It improves lesion detection precision by 8% for all metastases sizes and by 12.5% for metastases < 10mm with respect to standalone 3D U-Net. The segmentation Dice scores were 0.90±0.10, 0.89±0.10 and 0.89±0.10 for the above metastases sizes, all above the observer variability of 0.80±0.13. Automated detection and volumetric quantification of brain metastases following SRS have the potential to enhance the assessment of treatment response and alleviate the clinician workload.

Implementasi Algoritma Centroid Linkage dan K-Medoids dalam Mengelompokkan Kabupaten/Kota di Sulawesi Selatan Berdasarkan Indikator Pendidikan

Cluster analysis is a multivariate analysis technique that aims to cluster the observational data or variables into clusters in such a way that each cluster is homogeneous according to the factors used for clustering. This study used the Centroid linkage algorithm that was useful for forming groups based on the distance between centroids and the K-Medoids algorithm that was based on the use of the most centered object (medoid) to group districts/cities and obtained comparison results based on the education indicator data in South Sulawesi. The implementation of the Centroid Linkage Algorithm and K-Medoids on the education indicator data in South Sulawesi in 2018, showed that the grouping of districts/cities in South Sulawesi produced 2 clusters with cluster 1 of 21 districts/cities, and cluster 2 of 3. To determine the best method, it was seen from the value of the Standard Deviation ratio in the cluster 〖(S〗_W) and Standard Deviation between Clusters 〖(S〗_B) showed the same standard deviation ratio (S) in the Centroid Linkage algorithm and K-Medoids that was equal to 104,967.

Teach to comply? Evidence from a taxpayer education program in Rwanda

AbstractThere is virtually no evidence on the role of taxpayer education to improve tax compliance. We address this gap by providing the first evaluation of a taxpayer education program on compliance behavior, as well as taxpayer knowledge and perceptions. Using a unique dataset of administrative and survey data, we show that training new taxpayers leads to a large and significant improvement on three compliance outcomes: the probability to file declarations, to report a zero-tax amount, and tax due. These results are robust to three estimation strategies: a simple regression exploiting baseline balance across a wide set of observable variables, propensity score matching, and an IV strategy using random assignment to take part in our survey. We identify reduced compliance costs as the key mechanism at play in explaining the relationship between tax education and compliance—particularly increased knowledge and better perceptions on complexity. The effects we document persist over time, beyond the year of implementation.