Your cart is empty
Add prompt packs to continue
This definitive collection of statistical prompts represents the gold standard for data analysts, researchers and scientists seeking absolute precision in their workflows. Each prompt has been designed under principles of mathematical rigor and instructional logic, allowing raw data to be transformed into robust conclusions with unprecedented analytical depth in artificial intelligence environments. Optimize your validation, modeling and visualization processes using tools that cover everything from classical probability to the most complex non-parametric methods. By integrating this resource into your technical arsenal, you guarantee bias-free data interpretation, based on proven methodologies and aimed at excellence in strategic decision making.
100 resources included
Acts as a Senior Data Scientist and Statistician specializing in non-parametric methods. Your main task is to execute, analyze and interpret a **Mann-Whitney U Test** to determine if there are statistically significant differences between two independent groups: [Nombre_Grupo_A] and [Nombre_Grupo_B]. This analysis is critical given that the response variable [Nombre_Variable_Metrica] does not follow a normal distribution or is measured on an ordinal scale, invalidating the use of traditional parametric tests. To begin the procedure, carry out an exhaustive validation of the test assumptions: independence of the observations, scale of measurement of the variables and similarity in the shape of the distributions to establish whether the comparison is made on the medians or on the general range distribution. Please take into account the following data or description of the sample: [Insertar_Datos_o_Resumen_Estadistico]. It is vital that you properly handle ties in ranks, applying the necessary statistical corrections to avoid biasing the U statistic. Accurately calculate the U statistic for both groups (U1 and U2), identify the minimum value, and determine the resulting p-value. If the sample size is considerable (n > 20), use the approximation to the normal distribution by calculating the Z score and apply the Yates continuity correction. The significance level established for this analysis is [Alfa_ej_0.05]. You must be extremely rigorous in reporting whether the results allow us to reject the null hypothesis (H0) that the distributions of both groups are identical. Finally, generate a structured results report that includes: 1. Non-parametric descriptive statistics (Median, Interquartile Range, Average Ranges). 2. Test results (U-value, Z-score, p-value). 3. Calculation of the effect size (Rosenthal's r or Range-Biserial Correlation) to quantify the magnitude of the difference. 4. A technical conclusion written in professional language adapted to the context of [Contexto_del_Proyecto_o_Industria], providing clear recommendations based on the statistical evidence obtained.
He acts as an expert consultant in biostatistics and non-parametric methods. Your objective is to perform an exhaustive analysis using Fisher's Exact Test to evaluate the association between two dichotomous categorical variables, especially indicated for small samples where the expected frequencies in a 2x2 contingency table are less than 5, invalidating the Pearson Chi-square test. Context of the study: The analysis is framed in [Describe the area of study, for example: Clinical Trials, Quality Control or Social Sciences]. The aim is to determine whether there is significant statistical independence between the independent variable [Name of the predictor variable] and the dependent variable [Name of the outcome variable]. It is imperative that the analysis considers the underlying hypergeometric distribution to calculate the exact probability of observing the given data configuration or a more extreme one. Input data for the 2x2 Contingency Table: - Group A / Condition 1: [Value A] - Group A / Condition 2: [Value B] - Group B / Condition 1: [Value C] - Group B / Condition 2: [Value D] Please verify that the total sample sum is [Sample Total] and proceed to calculate the two-tailed probability p (p-value), unless it is specified that the study requires a one-tailed approach due to prior hypothetical directionality. Required technical interpretation: Once the p-value is calculated, compare it with an alpha significance level of [Significance level, e.g. 0.05]. If p < alpha, reject the null hypothesis (H0) of independence and conclude that there is a significant association. Additionally, calculate and interpret the Odds Ratio (OR) with its respective 95% confidence interval, explaining what this value means in terms of the magnitude of the effect and the relative risk for the context of [Target population]. Presentation of results: Generate a detailed report that includes: 1) Summary of the observed data. 2) The exact p value obtained. 3) The statistical decision based on evidence. 4) A narrative conclusion in technical but understandable language about the implications of the finding for [Final objective of the research]. Be sure to mention why Fisher's Exact Test is the correct methodological choice over other non-parametric tests in this specific scenario.
He acts as an expert in biostatistics and advanced data analysis with specialization in Parametric Inference and linear models. Your objective is to perform a comprehensive and professional analysis of variance (ANOVA) on the data set detailed below: [Dataset]. You must evaluate with mathematical precision whether there are statistically significant differences between the means of the groups defined by the categorical factor [Independent_Variable] in relation to the continuous numerical response variable called [Dependant_Variable]. First, before proceeding with the F-statistic calculation, it is imperative that you perform a rigorous validation of the fundamental assumptions of the ANOVA model. This necessarily includes the test of normality of the residuals using the [Test_Normality: Shapiro-Wilk or Kolmogorov-Smirnov] test and the evaluation of homoscedasticity (homogeneity of variances) using the [Test_Homoscedasticity: Levene or Bartlett]. If the data does not meet these prerequisites, you should suggest specific mathematical transformations or propose the use of an appropriate nonparametric alternative to ensure the validity of the inference. Once the assumptions have been validated, execute the [ANOVA_Type: One-factor, Two-factor or Repeated Measures] procedure using a confidence level of [Confidence_Level: 95% or 99%]. Generate the complete ANOVA table that accurately breaks down the Degrees of Freedom (GL), Sum of Squares (SS), Mean Square (MS), the F-statistic value, and the resulting p-value. Interpret the results in a technical manner, clearly establishing whether there is sufficient evidence to reject the null hypothesis of equality of means based on the alpha of [Significance_Level]. In the scenario of obtaining a statistically significant result, you should automatically proceed to perform a post-hoc multiple comparisons analysis using the [Post_Hoc_Method: Tukey, Bonferroni or Scheffé] method. This analysis should specifically identify which groups differ from each other, providing confidence intervals for the differences and corresponding adjusted p-values. Do not skip calculating effect sizes (such as Eta-squared or Omega-squared) to determine the practical relevance of the findings found. Finally, synthesize all the information into a structured executive report. This report should include a conclusions section where you explain the implications of the analysis and a recommendation for data visualization, suggesting the use of boxplots with error bars or interaction plots if necessary. Make sure the language is technical but understandable for data-driven decision making.