Psych250 Week 6 Notes
Psych250 Week 6 Notes Psy 250
Popular in Research Methods in Psychology
Popular in Psychlogy
verified elite notetaker
This 4 page Class Notes was uploaded by Kyra Ferguson on Monday February 29, 2016. The Class Notes belongs to Psy 250 at Colorado State University taught by Tori Crain in Spring 2016. Since its upload, it has received 23 views. For similar materials see Research Methods in Psychology in Psychlogy at Colorado State University.
Reviews for Psych250 Week 6 Notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 02/29/16
Chapter 6: Sampling Population- thinking of it as the people you want to be able to say that your findings apply to. Large group- includes all potential participants May be defined in many ways Sample- you can't include all members of your population in your study usually, so we get a small subgroup of participants from a population Representative or unrepresentative Goal of sampling: be able to apply results obtained from a sample to the population Generalizability- the ability to apply results obtained from a sample of the population Should apply more to external validity Two main types of samples Probability Sample- probability of choosing an individual in a population is known Convenience Sample- probability of choosing an individual in the population is not known Not all members of the population are considered for participation Sampling Error Is there a difference between the observations you get from a sample and the observations you would get from the larger population? Different types of sampling result in more or less sampling error Trade off is feasability Probability Sampling Simple Random Sampling- every member of the population has an equal probability of being selected Advantages: low sampling error Disadvantages: difficult to ensure that each member could be chosen Stratified Random Sampling- Population divided into subgroups (strata) Proportion of group in sample is equal to proportion of group in population Cluster Sampling- Identify "clusters" of individuals and then randomly select a number of complete clusters to sample from Advantages: easier to choose members randomly from smaller clusters to represent population Disadvantages: can miss part of the population Convenience Sampling When choosing from a sample randomly is not necessary or not possible or don't know the probability of choosing someone room larger population Volunteer (Haphazard) sampling = participant selection based on availability (convenience); "take them where you can find them", Ie, research pool Very common Advantages: Very feasible, and usually does not greatly reduce generalizability Disadvantages: Extraneous/confounding factors, might be missing part of the population Quota Sampling- sample reflects numerical composition of various subgroups Similar to stratified random sampling but participants are chosen out of convenience until quote is met Advantages: Very feasible, still able to sample according to proportions of characteristic Disadvantages: Extraneous/confounding factors, might be missing part of the population Probability vs Convenience Samples Probability samples can be difficult to obtain (depending on specified population)! But probability samples are likely important when demographic factors are likely to influence results Participants' behavior and attitudes are likely to differ Trade-offs Probability Samples More difficult Lower sampling error, more likely it looks like the population Higher internal validity Higher external validity Convenience samples More feasible Higher sampling error, not usually enough to really matter Lower internal validity Lower external validity How important is external validity? Ie, undergrad research pools. Will depend on research question Basic vs Applied Research Both are important! Basic tends to have more internal validity Applied tends to have more external validity You'll potentially see both basic and applied within certain studies Considerations for choosing sampling technique Goal of research Research design Population (target vs accessible) Recruitment how will you contact participants Can influence sample size Examples: Phone book (becoming less popular) internet research pool in-person survey at location flyers others? Incentives Will you provide something to participants for their time and effort? Can influence sample size Can't "coerce" people Examples: money gift certificate course credit/extra credit others? Chapter 7: Summarizing Data Descriptive vs inferential statistics Descriptive Allow us to summarize data sets Tell us in general what data sets look like Inferential Procedures that use descriptive statistics from sample to test hypothesis about populations Summarizing data Central tendency- how participants scored overall, a typical score in distribution Mean- add all scores and divide by number of scores, not used on nominal or ordinal Median- the middle most score (50% should be below and 50% should be above), shows central tendency on ordinal, interval, and ratio scales Mode- most frequently occurring score, indicates central tendency with all scales including nominal Outliers- extreme high or extreme low, skew central tendency towards top or bottom (skews the mean), usually report the median when there are outliers Variability- How widely the distribution of scores is spread Range- difference between highest and lowest score, but ignores middle scores Variance- standard deviation squared, represents total amount of similarity or dissimilarity of scores Standard Deviation- average difference between scores, square root of variance, becomes larger when more scores lie farther from the mean value, expressed in original unit of measurement Degrees of Freedom- number of scores that can vary in the calculation of a statistic n-1, where n = number of scores Tables and Graphs Quickly summarize with a visual Typically slow central tendency or variability Frequency Distribution Graph Graph indicating the number of individuals how recieved each possible score on a variable Bar Graph Can also communicate measure of central tendency, but demonstrates the relationship between different variables Line graph Scatterplot