PSY 2116 Lecture Notes - Lecture 3: Statistical Power, Central Limit Theorem, Alternative Hypothesis
5 Mar 2016
School
Department
Course
Professor
Document Summary
Scientists build rw processes to predict how these operate under certain conditions. Can"t assess rw directly, make inferences based on these models. Want an accurate model so predictions are valid, therefore must build a model based on data we collect. Degree to which a stat model represents data collection known as fit of the model. Good fit - excellent rep of rw a. Since rep"s r (reality) so closely, can be used confidently to make predictions. Moderate fit - predictions from model may be inaccurate a. Some similarities to reality but some important differences. Poor fit - predictions likely be completely inaccurate. Interpretation: data we observe is predicted from the model we choose to fit the data, plus error. Models vary depending on study design, type of data, purpose of the model. Measured constructs that vary across the sample (data we have collected)
