You are here: CTSPedia>CTSpedia Web>ContentInterest>StatisticalConceptsToMaster (11 Aug 2011, MaryBanach)EditAttach

- Difference between a functional language with live access to data and a procedural or macro language
- Meaning of a variable and what is the most commonly used object in S to store the values of a variable
- What constitutes legal names for S objects
- The most common data types in S
- How to compose S commands for doing arithmetic on constants and variables
- Names of the most commonly used algebraic and statistical functions
- Difference between character constants and names of objects
- Purpose and names of commonly used attributes of S objects
- When to use
`{ [ ( ) ] }`

- How to create, use, and subset vectors
- Making systematic sequences of integers and floating point numerics
- How to create logical expressions for checking equality and inequality, unions, and intersections of conditions
- Notation for missing values and how to check for them
- How to invoke functions having more than one argument
- Purpose and components of data frames
- How to subset rows and/or columns of data frames
- How to extract or access a single variable from a data frame
- Purpose of a
`factor`

variable and how to create one - How to make functions in add-on S libraries available for execution
- Purpose of the
`data.dump`

and`data.restore`

functions. - Methods for carrying out an overall inspection of the quality and completeness of data in an imported dataset.
- Understand the
`search`

list. - Understand the special purpose of search position one.
- Know how to make the variables contained in a data frame available for computation without using the name of the data frame as a prefix to a variable name.
- How the general setup for using the Hmisc
`upData`

function to change, add, or delete variables in a data frame. - Understand some of the ways for repeating analyses over various data subgroups (stratified analyses).
- Know how to create basic derived variables and how to recode categories of categorical variables.
- Know an easy way to categorize a continuous variable into intervals.
- Know the order that data import / annotation and other changes to variables / analyses / attaching data frames should be done.
- Know how to use S functions to compute probabilities from a few of the most commonly used statistical distributions. For discrete distributions such as the binomial, know how to compute the probability of a specific value and how to compute cumulative probabilities.
- Know how to use S statistical functions for the normal,
*t*,*F*, and chi-square distribution to compute*P*values for statistical tests. - How to get a Spearman test of association, Wilcoxon two-sample test, and Kruskal-Wallis test, as special cases of a more general rank test, using a single S function.
- Know the syntax for an S statistical formula.
- Know in general terms the capabilities of the
`summary.formula`

function. - Understand that in many cases stratifying on a continuous variable by categorizing into intervals does not require creating a new variable.
- Know the major elements of graphical perception.
- Know how best to make a graph so that people can perceive differences.
- Understand Weber's law and its ramifications for graphical design.
- Know causes of common optical illusions in statistical graphics.
- Know the ordering of perception tasks by how well humans interpret information from them.
- Understand problems caused by pop charts.
- Know how to determine a good aspect ratio for graphing curves.
- Know how dot charts overcome problems with other types of charts.
- Know how to best summarize distributional characteristics of data for graphics.
- Know the best types of graphs for representing various types of data, and understand why these types are preferred.
- Understand various methods for conditioning on other variables.
- Know how to interpret and when to use the following types of plots:
- rug plots (one-dimensional scatterplots)
- histograms
- density plots
- empirical cumulative distribution plots
- box plots
- scatter plots
- thermometer plots
- bubble plots
- scatterplot matrices

-- FrankHarrell - 29 Feb 2004

Edit | Attach | Print version | History: r2 < r1 | Backlinks | Raw View | Raw edit | More topic actions

Topic revision: r2 - 11 Aug 2011 - 12:47:21 - MaryBanach

Copyright &© by the contributing authors. All material on this collaboration platform is the property of the contributing authors.

Ideas, requests, problems regarding CTSPedia? Send feedback

Ideas, requests, problems regarding CTSPedia? Send feedback