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: Who: what is being studied, individual cases What : the variables Relative frequency table : displays percentages instead of counts Area principle: area occupied by a part of the graph should correspond to its value Bar chart : displays the distribution of a categorical variable Relative frequency: shows percentages Contingency table : shows how individuals are distributed along each variable Histogram: shows distribution of quantitative variables Stem and leaf : shows individual values (having a key is very important) Marginal distribution: the frequency distribution of variables Dot plot : dot along axis for each data point Segmented Bar Chart: treats each bar as a whole and divides it up proportionally into segments corresponding to the percentage of each group Simpsons paradox: When averages are taken between numbers that cannot be evenly averaged unfair averaging When describing, always use shape, center, and spread- humps in histogram are called modes (unimodal, bimodal, multimodal, uniform)- symmetric, or skewed (Graphs are skewed towards the tail)- mention outliers- median is resistant to outliers Interquartile Range (IQR)= Q3 Q1 Measures of Center: Use mean when graph is symmetric, median when skewed Measures of Spread: Use standard deviation when graph is symmetric, IQR when skewed, range can also be used to generally describe data Fences of Box plot = Q1 - 1.5IQR (lower fence) Q3 + 1.5IQR (upper fence) Z score = (y-mean of y)/STD- 68-95-99.7 Rule- 68% of values lay within -1STD - 1STD, 95 lay between -2 2 STDs, 99.7% lay within 3 STDs Standard deviation = ( - )- y y 2n 1 5 Number Summary:...
View Full Document
This note was uploaded on 04/04/2008 for the course ILRST 2100 taught by Professor Vellemanp during the Fall '07 term at Cornell University (Engineering School).
- Fall '07