PSYC 2320 Lecture Notes - Lecture 2: Frequency Distribution, Percentile Rank, Level Of Measurement
19 Sep 2018
School
Department
Course
Professor
Document Summary
How often scores/values occur in data (frequency) Distributions have both spread (x axis) and difference in frequency (y axis) Collection of observations produced by sorting observations into classes and showing frequency (f) of occurrence in each class. Ungrouped: one value per class, use less than 20 values. Grouped: classes of more than one value. Similar to traditional stem and leaf plots, but ultimately help to build frequency distribution curves. Don"t know how many of each value. Each observation should be included in only one class. List all classes - even those with zero frequencies. All classes should be equal in width, with upper and lower boundaries. The lower boundary of each class should be a multiple of the width. Aim for approximately 10 classes, too many classes defeats the purpose. Too many classes defeats the purpose, too few may mask important info. Find the width required to span data range by the number of classes.
