STAT 1350 Study Guide - Midterm Guide: Bar Chart, Pie Chart, Sample Size Determination

• School: Ohio State University
• Department: Statistics
• Course: STAT 1350
• Professor: Soboleva

Document Summary

Simple random sample: each subject has an equal chance of being selected. Stratified random sample: divide the population into subgroups using a categorical variable (gender; state; major), take mini srs within each subgroup. Convenience sampling: selection of whichever variables are easiest to reach. Voluntary response: people choose themselves by responding to a general appeal. **sample size is what is important; the size of the population does not matter. Larger samples have less variability the population size is not important to variability. Margin of error = 1 (cid:3041) (n = sample size) If the sample size increases the margin of error decreases. If the population size increases the moe stays the same: only covers random sampling error. Response error: subject gives incorrect response to question (lying) Nonresponse error: failure to o(cid:271)tai(cid:374) data fro(cid:373) a su(cid:271)je(cid:272)t (cid:894)(cid:272)a(cid:374)"t (cid:271)e (cid:272)o(cid:374)ta(cid:272)ted(cid:895) Response variable = dependent (measures result of study) Experimental design: control (subjects are equally affected, randomize (reduces bias, use enough subjects (reduces variability)