PSY201H1 Lecture Notes - Lecture 4: Frequency Distribution, Level Of Measurement, Categorical Variable
Document Summary
Goal: organize data to communicate how many observations exist at each category. Note: norms differ from source to source, so read label! Column of f should add up to sample size. Each score represented as a portion/percent of the total. % adds up to 100: cumulative fd. Show total f / p /% at each value and all lower ranked values. Begin at the bottom, add f upwards. Percentile rank (aka. cumulative percent): % of observations at / below same/smaller value. If the exact percentile does not appear, choose the smallest percent that first contain/passes the desired percentile (choose 97% in this case) This way, we are being conservative and reduces the chance of making an error. When the data cover a wide range of scores, representing all possible score is not feasible! Rule: if number of rows > 20, then you have too many! Arrange categories into meaningful orders and record frequencies.
