STAT 110, Notes for Week of 9/6/16
STAT 110, Notes for Week of 9/6/16 STAT 110
Popular in Introduction to Statistical Reasoning
verified elite notetaker
Popular in Statistics
This 4 page Class Notes was uploaded by runnergal on Friday September 9, 2016. The Class Notes belongs to STAT 110 at University of South Carolina taught by Dr. Wilma J. Sims in Fall 2016. Since its upload, it has received 9 views. For similar materials see Introduction to Statistical Reasoning in Statistics at University of South Carolina.
Reviews for STAT 110, Notes for Week of 9/6/16
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/09/16
STAT 110 – Notes for Week of 9/6/16 Chapter 6 Continued o You can only generalize results (apply statistics from a sample to an entire population) if the results are: Statistically significant Useful The sample that produced the results is representative of the population The experiment that produced the results was similar enough to real life o Completely Randomized Experimental Design: a type of experimental design where all of the subjects are randomly allocated among the treatments. This type of design allows more than one treatment (treatment combinations) to be isolated and studied at the same time. o Matched Pairs Design: where either one subject compares two treatments by trying them both (pair of treatments) or two very similar subjects compare two different treatments (pair of people comparing treatments). These treatments are randomized within the pairs, not within the full group; that way, no pairs get the same two treatments. o Within Subject Variability: uncontrollable variability between subjects in a sample. Researchers cannot avoid this small amount of variability. o Block: a group of experimental subjects that the researchers have already predetermined are very similar in some way(s), such as age, gender, race, etc. These groups are not random; they are specifically chosen in order to reduce the chance of lurking variables, like age, gender, race, etc., from influencing the experiment. o Block Design: a type of experimental design where the subjects within each block are randomly assigned to a treatment. This helps researchers determine if any lurking variables affected how the subjects reacted to the treatment. o Blocking Variable: accounts for a potential lurking variable. Chapter 7 o Ethics: a set of principles concerning moral values that lead people and organizations to determine which actions are right and which actions are wrong. o Institutional Review Board: all researchers must submit their experiment proposals to one of these boards. These boards determine if an experiment is ethically responsible and protect subjects from any harm or injury. o Informed Consent: all subjects in the experiment must be informed of any potential risks and then agree to participate. The subjects must also understand that the treatment will be randomly applied to the subjects in the sample. The American Psychological Association (APA) does not require consent from subjects whose behavior is being observed in a public place. Additionally, subjects are not always given all of the information before the experiment, assuming the lack of information does not harm the subjects, because it may alter the true effects of the treatment. o Confidentiality: all information that could be used to identify specific subjects must be kept secret. All research reports may only report on the aggregate data, not the specific data of a certain subject. o Clinical Trials: medical experiments that are carried out on human subjects. While these definitely are risky for the subjects, it is also the only way to find out the true effects of a medical treatment on humans. Chapter 8 o Measure: assign a number to represent a characteristic of a subject. Measurements turn concepts into defined variables. o Instrument: something that measures a characteristics of a subject. o Units: things used to record measurements, like meters, pounds, etc. o Variable: the result of a measurement that is unique to each subject. The variable should be a valid and accurate way to measure the characteristic. o Count: the number of times a characteristic occurs in the sample. o Rate: the number of times a characteristic occurs in the sample divided by the number of subjects in the sample. For example, if there are 40 freshman in this STAT 110 class of 200, there is a rate of 40/200, or 20%. o When the population of the group that researchers study varies over time, they should use rates because it takes into account the fact that the population has changed. Rates are usually more valid than counts. o Predictive Validity: a measurement has this if it can predict the success on tasks related to the measurement. For example, if I can run a 5 minute mile, that measurement has good predictive validity that I can run a fast 5K. o Reliability: this is when numerous measurements produce about the same results. Use the average of many measurements to improve reliability. o Bias: this occurs when the true results are consistently overstated or understated. Use a different instrument to reduce or prevent bias.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'