Ch. 2 - Frequency Distributions
Wednesday, September 12, 2012 3:04 PM
- Statistics for summarizing and organizing sample data, often for a general overview
- Frequency distributions
○ How many times (frequency) each value occurred in the data set for a given sample
○ Tables and graphs
- X & Y columns: categories on X measurement scale (X) & their frequencies (f).
○ X column: all values listed, highest to lowest
○ f column: Tallies, # times each value for X occurs in the data set
Tallies are frequencies
○ Sum of all f = N (total # cases in data set).
- Proportion (p) of each category: p = f/N
- % of distribution @ X value (=p*100). %.
- Humorous: "If at first you don't succeed, you're probably not related to the boss"
- Non-Humorous: "People related to the boss often succeed the very first time"
- 16 participants, 8 humorous + 8 non-humorous sentences, # recalled
Tables are nice for calculation,
but are hard to visualize. FrequencyDistributionTables
- Regular: all individual X values
- Grouped: groups of X scores (class intervals)
○ Used when X has a wide range of values
○ Intervals must all have the same width [same size]
○ Width chosen so table has ~10 intervals (min f = 5 per interval)
- X (score) categories listed on X (horizontal) axis, f (frequencies) on Y (vertical axis).
○ Scores nominal or ordinal scale: histogram [Histogram = bar graphs, but must count "how many"/ frequency]
○ Scores interval or ration scale: either one
- Histogram: bar centered over each score (or class interval). Bar height shows S/C