DOC PREVIEW
UH KIN 4310 - Sampling and Confounding Variables
Type Lecture Note
Pages 5

This preview shows page 1-2 out of 5 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 5 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

KIN 4310 1st Edition Lecture 3Outline of Last Lecture I. Excel FunctionsII. VariabilityIII. Why Variability is Important?IV. Measures of VariabilityV. DefinitionVI. DefinitionVII. Sample Standard Deviation FormulaVIII. Why n-1?IX. Standard Deviation – Important PropertiesX. Things to RememberXI. Computing VarianceXII. Standard Deviation or Variance?XIII. Excel FunctionsXIV. Methodological ApproachesXV. Descriptive StudiesXVI. Correlation StudiesXVII. Experimental StudiesOutline of Current Lecture I. ExampleII. Sample SelectionIII. Sample SelectionIV. Methods of SamplingV. Random SamplingVI. Systematic SamplingVII. Convenience SamplingVIII. Stratified SamplingIX. Cluster SamplingX. DefinitionsXI. DefinitionsXII. Definitions – Methodological DesignXIII. DefinitionsXIV. Strategies to Avoid ConfoundingCurrent LectureThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.I. Example a. The Effect of Increased Speed Limitsi. On Nov 28, 1995, the National Highway System Designation Act was signed into lawii. Abolished the federal mandate of 55 mph maximum speed limits on roads in the U.S. iii. Of the 50 states (plus DC), 32 increased their speed limits during 1996b. This was an observational study because nothing was manipulated here, the states chose on their own if they were going to increase their speed limitsc. From the data, we cannot conclude that high speed limits increased fatalities because results may be from random noised. Experimental studies are the only ways to prove causationII. Sample Selectiona. Samples should be selected randomlyi. Best method is random selectionb. Random selection corrects for systematic bias that may confound resultsi. Eliminates bias – this is why random selection is so goodc. Non-random sample selection has to be accounted fro statistically i. Can still use non-random sampled. Very common to have a convenience samplee. All human samples are volunteer samplesi. Has to be a volunteerii. Ethical reasonsIII. Sample Selectiona. Pick a cardi. Red King of Hearts 1. 16ii. Black 7 of Clubs1. 15iii. Red Ace of Diamonds1. 48iv. Red 4 of Hearts1. 27v. Red 9 of Diamonds1. 17b. There was a lot of bias to pick the Ace because normally, the ace is the best card in card gamesIV. Methods of Samplinga. Randomb. Systematicc. Convenienced. Stratifiede. Clusterf. Know these different types of sampling and what makes them good or faulty in certain situation (TQ)V. Random Samplinga. Random Samplei. Members of the population are selected in such a way that each individual member has an equal chance of being selectedii. Computer generated programiii. Draw strawsiv. Numbers out of a hatb. Need a master list from the population to achieve thisc. A random sample is almost impossible to administrate but still good idea becauseit eliminated biasVI. Systematic Samplinga. Select some starting point and then select every Nth element in the populationb. Hard to put in order though c. Almost truly randomi. Almost as good because people aren’t coming in in any particular orderd. Involves ordere. Often used in clinical studiesVII. Convenience Samplinga. Data or results that are easy to getb. Take individuals that are closest to youc. Total opposite of random selectionVIII. Stratified Samplinga. Subdivide the population into at least two different subgroups, then draw a sample from each subgroup (or stratum)b. When you take entire population and divide them demographically (gender, age, disease condition) into stratac. Then do random sampling within each strataIX. Cluster Samplinga. Divide the population into sections (or clusters); randomly select some of those clusters; choose all members from selected clustersb. Usually geographicallyc. No one is left outd. Randomly select clusterse. In the state and county example, they didn’t want to travel all over the state so just chose 3X. Definitions (TQ)a. Parameteri. A numerical measurement describing some characteristic of a populationb. Statistici. A numerical measurement describing some characteristic of a sample1. Ex. mean, standard deviationXI. Definitionsa. Sampling Errori. The difference between a statistic and the associated parameter; such an error results from chance when it’s a random sampleii. Occurs naturallyiii. Any number off due to random chanceiv. The more people you select, the more x bar and mu will be the sameb. Nonsampling Errori. Sample data that are incorrectly collected, recorded, or analyzed (such as by selecting a biased sample, using defective instrument, or recording thedata incorrectlyii. Not due to the sample, but due to measurementsXII. Definitions – Methodological Design (these have to do with time frame of study)a. Cross Sectional Studyi. Data are observed, measured, and collected at one point in timeii. Snap shot in timeb. Retrospective Studyi. Data are collected from the past by going back in timeii. Looking backiii. Data that’s already been collected and study those numbers iv. Ex. chart reviewsc. Prospective (of Longitudinal) Studyi. Data are collected in the future from groups (called cohorts) sharing common factorsii. Looking into the futureiii. Passage of time is very important elementXIII. Definitionsa. Confoundingi. Occurs in an experiment or observational study when the experimenter isnot able to distinguish between the effects of different factorsii. Try to plan an experiment to avoid confoundingiii. Ex. people who carry lighters and canceriv. Smoking is a confoundv. The lighter doesn’t cause cancervi. It is important to see possible confounders in an experimentXIV. Strategies to Avoid Confoundinga. Blindingi. Participant does not know whether he or she is receiving a treatment or placebob. Matchingi. Select participants with similar characteristicsii. If you’re selecting certain groups, you match and make sure ages are the same and the number of men and women are similarc. Randomized Controlled Triali. Randomly assign participants to each experimental groupii. Gold standard of scientific


View Full Document

UH KIN 4310 - Sampling and Confounding Variables

Type: Lecture Note
Pages: 5
Download Sampling and Confounding Variables
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Sampling and Confounding Variables and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Sampling and Confounding Variables 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?