QTM 100 Lecture Notes - Lecture 15: Random Variable, Statistical Inference, Box Plot
13 Dec 2017
School
Department
Course
Professor
Document Summary
Three main components of statistic: design, description, inference. Summarizing and visualizing data (lectures 2, 3, and 4) Summarize in a frequency table with frequency, percent, proportion. Describing the shape of a distribution: unimodal, bimodal, left-skewed, right-skewed, symmetric, bell-shaped, uniform. Measures of spread: standard deviation, variance, range, iqr. Measure of position - identifying quartiles, understanding percentiles. Graphs: dot plot, histogram, time series, box plot (and how to construct them) Identifying potential outliers: in graphs (box plots, histograms), numerically (empirical rule, q1-1. 5iqr, q3+1. 5iqr) Using z-scores: z = observation mean standard deviation. Two quantitative variables: scatterplot (positive vs negative association) Two categorical variables: contingency tables for two categorical variables; calculating overall and conditional proportions/percents. One quantitative and one categorical variable: side by side boxplot, calculate measures of center and spread in each group. Random sampling methods: simple random sample, cluster random sample, strati ed random sample. Non-random sampling methods: convenience sample, volunteer sample. Potential sources of bias: sampling bias, nonresponse bias, response bias.
