100 likes | 344 Views
Chapter 2 – Experimental Design and Data Collection. Math 22 Introductory Statistics. Error Due to Bias and Chance. Bias - A systematic tendency to misrepresent the population. The object of any experimental design is to eliminate bias and reduce chance error as much as possible.
E N D
Chapter 2 – Experimental Design and Data Collection Math 22 Introductory Statistics
Error Due to Bias and Chance • Bias - A systematic tendency to misrepresent the population. • The object ofany experimental design is to eliminate bias and reduce chance error as much as possible.
Types of Design • Confounding factor (lurking variable) – a hidden factor that has an effect on the response we are attempting to measure. • Experimental (treatment) Group – those persons or objects that receive the treatment of interest in an experimental design. • Control Group – those persons or objects that do not receive the treatment of interest. Rather, they may receive the “old” treatment or a placebo.
Statistical Designs • Experiments - Attempts to determine a cause and effect relationship between two or more variables. • Blind Experiment – the test subjects do not know if they are getting the experimental treatment or the placebo. • Double Blind Experiment – neither the test subject nor the experimenter measuring the response knows to which group the test subjects have been assigned (treatment or placebo).
Statistical Designs • Prospective Study – Study of future events. • Randomization – An excellent way to reduce bias. • Retrospective Study - Study of past events. • Cross Sectional Study – Study of events at the current time (one point in time). Data represents what is going on at a certain cross section of time.
Collecting Data To obtain reliable information that will help answer your research questions, follow these steps: • Determine the objectives of the study you are undertaking. • Define the population of interest. • Choose the variables that you will measure in the study.
Collecting Data • Decide on an appropriate design for producing data. • Collect the data. • Determine the appropriate descriptive and/or inferential data analysis techniques.
Types of Random Sampling • Simple Random Sample - To select the sample in such a way that every sample of that size has the same chance of being chosen. • Systematic Random Sample • Stratified Random Sample
Surveys A properly designed survey reports the following information: • A description of the sampled population • A description of the method of contact for interviews • The response rate • The exact wording of the questions
Surveys • The timing of the interview. • The size of the sample (or the margin of error) • The sampling technique