Análisis de relación entre dos variables cuantitativas Coeficiente de correlación de Pearson Módulo3
New Section
In this module, the focus is on analyzing the relationship between two quantitative variables, emphasizing the Pearson correlation coefficient. The discussion starts with defining quantitative variables and setting up hypotheses about their independence or relationship.
Analyzing Hypotheses
- Two types of hypotheses are introduced:
- Null Hypothesis (Chisucero): Assumes independence between the two variables under study.
- Alternative Hypothesis (Chesua): Suggests a relationship between the variables.
- Researchers typically aim to support the alternative hypothesis, assuming a relationship until data proves otherwise, akin to innocent until proven guilty in legal contexts.
Data Inspection and Visualization
- Data inspection involves graphing or analytical methods using a dataset with x and y variables represented by values like xuí and yuí.
- Plotting these values on a Cartesian plane forms a scatter plot showing a trend where increasing x values correspond to decreasing y values.
New Section
This section delves into interpreting graphical patterns in scatter plots to infer relationships between variables and introduces the concept of stochastic dependencies.
Understanding Scatter Plots
- Observing trends in scatter plots helps draw initial conclusions about variable relationships.
- Differentiating mathematical dependencies from stochastic dependencies is crucial; stochastic relationships show trends but not precise functional forms like straight lines or curves.
Quantifying Relationships
- The covariance statistic is introduced as a measure of association between two variables, calculated using formulas involving deviations from means for each variable pair.
New Section
In this section, the concept of covariance is discussed in relation to points on a graph and their contributions to covariance.
Understanding Covariance Calculation
- Points on a graph are analyzed by drawing lines parallel to the axes.
- Focus is placed on a specific point with coordinates x and y.
- Points painted in red contribute positively to covariance, while those in green contribute negatively.
- Positive covariance indicates a direct relationship between variables.
New Section
The discussion shifts towards the practical application of correcting covariance through Pearson's correlation coefficient.
Practical Application of Pearson's Correlation Coefficient
- Pearson's correlation coefficient corrects for units by dividing covariance by the product of standard deviations.
- The coefficient provides an adimensional value, indicating the strength and direction of the relationship between variables.
New Section
Exploring scenarios where correlation coefficients vary based on data dispersion.
Impact of Data Dispersion on Correlation Coefficients
- Correlation coefficients change as data dispersion varies, with higher concentration leading to stronger correlations.