This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Chapter 2: Data Who, What, When, Where, Why, and How tell the context for data values If you cant answer Who and What, there is no data!! Must know Who, What, & Why to analyze data! Data systematically recorded info with its context; data is useless without context! Units tell how each variable has been measured and the scale Categorical variable names categories and answers questions about how cases fall into those categories Quantitative variable answers quantity of what is measured (has units) ordinal variable reports order without natural units (like rating 15 on an evaluation) Chatper 3: Displaying and Describing Categorical Data Frequency table lists the categories in a categorical variable and gives the count for each Relative frequency table lists the percentage rather than count for the values in each category Both tables show the distribution of a categorical variable Area principle the area occupied by a part of the graph corresponds to the magnitude of the value it reps Bar chart displays distribution of categorical variable, showing counts for each category right nxt to each other Relative freq bar chart shows the percentage rather than count Pie chart shows the whole group of cases as a whole and shows proportions Contingency table displays counts and/or percents of individuals falling into categories on two or more variab., categorizes the individuals on all variab. at once to reveal possible patterns in one variab. that may be contingent on the category of another Marginal distribution in a contingency table, the distribu of either variable alone, the counts are the toals found in the margins of the table Conditional distribution distrib of variable restricting the Who to consider only a smaller group of individ Segmtd bar chart treats the bar as a whole and breaks it up proportionally into segments Simpsons paradox when averages are taken across diff groups, they can appear to contradict the overall avgs Chapter 4: Displaying and Summarizing Quantitative Data Histograms must display quantitative data!! And quantitative data with units only on both axes!! Relative freq histogram replaces count with percents; shape of histogram is the same, only labels differ When describing distribution tell 1) Shape, 2) Center, 3) Spread Shape uniform, unimodal, bimodal, multimodal Center tell symmetry, mean, mode(splits data in half); thinner ends of distrib are tails, skewed to the longer side of the tail Spread tell about outliers and mention gaps in the distribution Median splits the data in half, resistant to values that are extra big or small and ignores their distance from center, use for asym grph Mean better used for unimodal, symmetric distrib, gives each value equal weight Interquartile range (IQR)...
View
Full
Document
This note was uploaded on 04/02/2008 for the course ILRST 2100 taught by Professor Vellemanp during the Spring '07 term at Cornell University (Engineering School).
 Spring '07
 VELLEMANP

Click to edit the document details