Psychology 3980 lecture notes, week 3
Psychology 3980 lecture notes, week 3 Psych 3980
Popular in Research Methods in Psychology
verified elite notetaker
Popular in Department
verified elite notetaker
This 4 page Class Notes was uploaded by Britney Beckett on Saturday September 3, 2016. The Class Notes belongs to Psych 3980 at University of Georgia taught by Trina Cyterski in Fall 2016. Since its upload, it has received 7 views.
Reviews for Psychology 3980 lecture notes, week 3
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: 09/03/16
Chapter 3: Three claims, four validities: What is a variable? A variable must have at least two levels (values) o “ 60% of teens text while driving” o Variable: testing while driving Level 1: texts while driving Level 2: does not text while driving Measured: observed and recorded as it occurs naturally Manipulated: controlled by researcher o Some variables can not be manipulated because it is impossible (gender, IQ), others can not be manipulated because it is unethical o Many variables can be recorded as measured or manipulated Conceptual variables: o A conceptual variable is an abstract concept such as shyness or intelligence. o Operationalized- turning a concept into a measured or manipulated variable. o Operational definition: the process by which the variable of interest is being measured or manipulated in a study/ experiment. Example: Intelligence can be measured by IQ test, GPA, accuracy/problem solving test, etc. 3 claims: Frequency o Claim made through observation and collection of data o Claims that mention percent of variable, number of people who engage in some activity, or a certain groups level on a variable o Variable is always measured o Focuses on one variable Association o Claim made through collecting data o Variables are measured o Measurements for two variables o Types of association- represented by scatter plots Positive association- as A increases, B increases (or vice versa) Negative (inverse) association -As A increases, B decrease (or vice versa) Zero association – no correlation Causal o Claim made through experimental research o Starts with a type of association o Claim goes beyond associating the two variables to say that one causes the other *A claim can still be a causal claim even if the words may, seems, sometimes, etc. are used **Advice is a causal claim. If you do X, then Y will happen. Type of claim dictates research design A valid claim is reasonable, accurate, and justifiable 4 validities: Construct-accuracy of measurement of the variable o Needs a good operational definition o Stresses reliability/ consistency of measurements. (Getting about the same results every time.) *Measurements can be reliable but not valid. **If measurement is valid then reliability is implied External-extent to which the results of the study can be generalized to (represent) people besides those in the study o Needs a representative sample Statistical- Probability that the outcome of the study is due to chance (low probability= more statistical validity) o Margin of error- range in which true value should lie. o Type I error- concluding from a sample that there is an association between two variables when there is no association in the real population o Type II errors- concluding from a sample that there is no association between two variables when there is an association in the real population. o Statistical significant- likelihood results are not due to chance Internal- ability to rule out any alternative explanations. What validities are needed to make each claim? Frequency Claim o Construct validity o External validity Association Claim o Construct validity o External validity o Statistical validity Casual Claim o Construct validity o External validity o Statistical validity o Internal validity *Internal validity is only important in casual claims but is the most important validity when making a causal claim. **For casual claim you can sacrifice another validity for internal validity (often external validity will be the validity sacrificed) o 3 things to demonstrate for casual claim Covariance: as A changes, B also changes Temporal Precedence: A comes before B; A is the manipulated variable Internal validity: there are no other possible explanations for changes in B then change in A. Internal Validity in experiments: Achieved by controlling all extraneous variables and by randomly assigning people to different experimental conditions. o Random assignment- process ensures all participants have an equal chance of taking part in any condition of an experiment. *Most important part of experimental method. Internal vs. external validity: Random vs. representative samples “Realism” is needed to be able to generalize results to other situations. Quasi- experiment No random assignment of subjects to conditions. Does not fit well with any of the claims *Can not be used for causal claims Practice: (answer on next page) What type of claim is this? Heavy kids face heart risks as they age. What validities are needed? Questions you should have for each validity Construct Validity o How was the variable measured? i.e. How was heart risk measured? Was it measured using a method that has been approved by medical professionals? o External Validity Is the study representative of the population it would be applied to? i.e. Where were the kids from? How were participants chosen? o Statistical Validity What is the correlation coefficient? Is the correlation statistically significant? Practice answer: Association claim External, statistical, and construct validity Reading Notes: Constant: could potentially vary but is held constant in study/ experiment. To manipulate a variable means to assign participants to be at one condition or another
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'