R Concepts and Data Handling

R Concepts

• R is a powerful programming language used for statistical computing and graphics.
• R code can be commented for explanation, improving readability and maintainability.
• Packages provide extra functionalities; packages must be installed and loaded to be used.
• R's workspace manages data and variables within the R session.
• Navigating the file system lets you access and manage files in your computer.
• R allows loading and saving of various data types efficiently.
• Variables in R have specific properties. Factors are categorical variables.
• Data frames are used to organize data in rows and columns with different variable types.
• Lists in R are useful for storing different data types (not just numbers).
• Formulas in R define the statistical relationships between variables.
• Generic functions perform similar tasks on various object types.
• Help in R provides details on functions and other elements. Summary of R is available.

Working with Data

• Descriptive statistics summarize data characteristics.
• Measures of central tendency (mean, median, mode) describe the center of the data.
• Measures of variability (standard deviation, variance) measure the spread of data points.
• Skew and kurtosis measure the asymmetry and shape of the data distribution.
• Summary of a variable provides overall statistics (maximum, minimum).
• Descriptive statistics for groups help to examine separate data subsets.
• Standard scores (z-scores) quantify the relative position of data points within a distribution.
• Correlations measure the relationship between two variables.
• Missing values can be handled using imputation techniques. A summary of handling missing values is available.

Drawing Graphs

• R offers comprehensive options for creating various types of graphs and plots. An overview of R graphics exists.

Berkeley Admissions Data (Example)

• Figure 1.1 shows Berkeley college admissions data.
• The admission rate varies depending on the percentage of female applicants.
• Departments with high admission rates tended to get more female applicants.
• This example demonstrates Simpson's paradox: seemingly contradictory relationships in subgroups disappear when the datasets are combined.
• Data analysis isn't just about calculating statistics. Choosing appropriate methods and properly interpreting results (even with appropriate metrics) is crucial. Statistics are tools, interpretation is key.
• Research can be complex and counterintuitive.
• Statistical knowledge is vital for evaluating and verifying information. This is true for daily life.

Statistics in Everyday Life

• Many sources of information include poor or misleading statistics or fail to include baseline data.
• Statistical awareness helps to identify errors or misrepresentations.

Research Methods Beyond Statistics

• Research methodology encompasses a broader range of topics beyond statistical analysis. Methodology can vary.
• Methods for data measurement can be self-report (surveys), or via authority (parents for kids).
• Using scales and grouping responses can allow for effective interpretation of results (e.g. statements endorsing a belief can be ordered or grouped to understand the subject matter).
• Example of variables: response time, ordering of beliefs, self-reported age data.
• Distinguishing between continuous and discrete variables: Continuous variables have infinitely many possible values between measures, unlike discrete variables which have limited options.
• Response time (RT) is a continuous variable because values can be found between the measured values.
• Scales of measurement (nominal, ordinal, interval, ratio) and the differences between continuous and discrete variables are crucial.