Transcript Document
Unit 6 Sampling Why Sample • To save time and money. • To make generalization or predictions from the sample to the population • Example: Exit Polls Let’s use the following research questions as an example. • What percentage of college students use MDMA. • What are the most important predictors of MDMA use among college students? • What are the effects of MDMA use? • What are the consequences of chronic MDMA use? MDMA • • • • • Ecstasy Pills Rolling Common Effects Hug Drug or Love Drug Populations and Samples Population: A complete set of individuals, objects, or measurements having some common observable characteristic. Sample: A subset of a population that is used to represent the population. Runyon, R., Haber, A, and Coleman, K. (1994). Behavior Statistics: The Core. McGraw Hill, Inc.: New York, NY. Probability Sampling o 1. There is a list of persons or things to sample from. o 2. Everyone has a equal chance of being selected. o 3. Probability sampling helps researchers and practitioners generalize from observed cases to unobserved ones. o 4. Often used in quantitative research Probability Sampling Techniques Probability Sampling Simple Random Sample Systematic Multistage Cluster Simple Random Sample Assigning a single number to each element in the population, and a table of random numbers is used to select elements for the sample. Most statistical programs have this function. Babbie, E. (1989). The Practice of Social Research 5th Ed. Wadsworth Publishing Company: Belmont, CA. Systematic Sampling Every Kth element in the total population is chosen for inclusion in the sample. If a population contains 300 cases and you want a sample of 50, you select every sixth element (300 / 50). The first case would be selected at random to start the subsequent selection. Babbie, E. (1989). The Practice of Social Research 5th Ed. Wadsworth Publishing Company: Belmont, CA. Multistage Cluster Sampling Example – Monitoring the Future Study – Drug Use Among High School Youth Example: Research Question: Do Southern Baptist visit Disney World? Babbie, E. (1989). The Practice of Social Research 5th Ed. Wadsworth Publishing Company: Belmont, CA. Response Bias • There a few returns = low response rate • Increase Response Rate – Token of Appreciation • If you have a low response you cannot make generalization or predictions from the sample to the population. Nonprobability Sampling 1. There is often not a list to build a sample. 2. No everyone has a equal chance of being selected 3. One cannot make generalization or predictions 4. Representativeness can not be assured. Imagine now we are going to conduct field research in which we will examine MDMA use at Raves Rave • • • • • • All night dance party Peace, Love, Respect, and Unity Celebration of Techno Music High use of MDMA Sexual Overtone Go back to youth – simpler life Rave Videos • Please watch the first three minutes of the following videos: • http://www.youtube.com/watch?v=8ljg-6MXEJ4&feature=fvw • http://www.youtube.com/watch?v=gUOeFN4vXHU Nonprobablity Sampling Techniques • Convenience Sampling Reliance on available subjects. Examples: Individuals on a street corner, at a mall, university students, or Ravers Nonprobablity Sampling Techniques Quota Sampling Example: Joe is asked to interview students at a Rave Joe’s sample consists of 70 percent females. Joe is asked to interview 50 percent males and 50 percent females. Nonprobablity Sampling Techniques • Snowball Sampling Build sample from respondents or subjects Example: Heroin Addicts Nonprobablity Sampling Techniques Purposive Sampling Choose subjects with unique characteristics. Examples: Interview DJs at Raves Mapping • • • • • Touchy Feely Room Cool Down Room Outside a Rave The Dance Floor The VIP Room Your Research Paper • 1. Cover Page • 2. Abstract • 3. Introduction – A problem statement – Write your Hypothesis – Identify your dependent and independent variables. • 4. Your Literature Review – – – – Summarize 5 relevant research studies Rely on academic journal articles Use Kaplan’s Online Library Critique studies • 5. Research Design – Give a description of your research design and a justification of why this particular design was chosen. – Experiment or Survey Experimental Design – – – – – Parts of a Classical Experiment 1. Variables (independent and dependent variables) 2. Control and Experimental Groups 3. Pretest and Post-test 4. Randomization or Matching (Sampling) Survey Design • You will make up an actual survey. • Question or questions measuring your dependent variable • Question or questions measuring your independent variable Example: Predictors of binge drinking • Dependent variable = How many times in the last two weeks have you drank 4 or more beers or alcoholic drinks in less than an hour? • Independent variables? Example: Rape on Campus • Dependent variable = In the last two years, have you had sexual intercourse because someone made you by using force or by threatening you? • There is a difference between rape and sexual assault. • Independent variables (characteristic of both the victim and offender) • 6. Sampling – Create heading and leave blank, but may write this part after listening to this seminar. • 7. Data collection – same as research design • (I do not know why Kaplan has this as a separate section)