Week 2 Notes
Week 2 Notes Psych 100B
Popular in Research Methods in Psychology
Popular in Psychlogy
This 4 page Class Notes was uploaded by rallen17 Notetaker on Friday April 8, 2016. The Class Notes belongs to Psych 100B at University of California - Los Angeles taught by Darvick, E.R. Fourquet, N. Middlebrooks, C.D. in Spring 2016. Since its upload, it has received 73 views. For similar materials see Research Methods in Psychology in Psychlogy at University of California - Los Angeles.
Reviews for Week 2 Notes
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: 04/08/16
100B Monday Lecture Week 2 - You had to do 100B work over the weekend - Hypothesis: testable prediction - Correlation: two variable are correlated (part of hypothesis) - If we find out that these two variables are correlated we can make predictions o But we cannot apply causation because Third variable problem We don’t know directionality (which causes which) o We can’t apply causal claims - Example: coffee consumption and fertility are correlated o Of women trying to get pregnant, women who drank more than three cups of coffee per day take at least more than a year to conceive. o Predictor variable: amount of coffee that women drink per day o Criterion variable: the time it took these women to conceive o X axis: predictor variable: amount of coffee o Y axis: criterion variable (months) o this is a positive correlation. Use scatterplot whenever you plot correlations o third variable could be lack of sleep o third variable could cause increase in both variables o or taking longer time to conceive could cause coffee o criterion variable can cause predictor variable - it would be useful to make causual conclusions o (we would have to eliminate directionality and third variable issue) o We would need to perform new studies o In order to say A causes B, we need to manipulate A o By manipulating A, we need to control everything in the situation and control A o A is independent variable - Independent variable o Manipulate by controlling evyerhting else (dependent variable) - Dependent variable o Measure - Experimental design: manipulating independent and measuring dependent o True experiment: actually manipulating independent - Does watching violent TV cause aggressive behavior? o We need to manipulate independent variable o Manipulate violent TV and see if that affect dependent variable How Tell children how much violent tv to watch Random assignment: randomly assign each child to different level of IV (different amount of TV) Violent tv/aggressive behavior are constructs Operational definition: take a construct and define that construct o Define interference and discrete task o Totally up to us how we define construct o Operation definition up to us Interference defined: how long it takes Independent Variable o is number of hours children watch TV o At least 2 levels o Randomly assign children into those two level o Can be manipulate quantitative or qualitative o Quantitative: how much is the difference o Qualitatively: type of violent TV One group watches violent drama, one violent show etc IV for our experiment o We will use quantitative (effect of amount of TV) o X axis: independent variable: amnt violent tv o Might find that watching violent TV increases aggressive behavior might decrease aggressive behavior, or it might stay the same o By adding a third level: we can get a lot more information Adding third level o Turns into functional relationship, not linear o Range is really important If you found that there is no difference in aggressive behavior with change in minutes, that wouldn’t be sound conclusion if range wasn’t big enough (you did 0, 5, 10 minutes) IV changed to type of violent TV Aggressive behavior vs type violent TV Use bar graph when IV is qualitative (use line graph when it’s quantitative) Our experiment Operationally define IV: only thing that is different is amount of tv watched Everything else is controlled Put more specifics on it - Types of measurement that we have o We now need to operationally define our dependent variable o Different types of measurements self report: when you are asking someone to report on themselves ex: how aggressive are you not always accurate behavior measures: behavior watching them and measuring them ex: number of times a kid hit another physiological: anything from your body ex: blood pressure - Types of scales o Different scales Nominal: often “yes” or “no” Ex: aggressive or not Ordinal: categories with an order! Ex: first place, second place, third place Ex: not, somewhat, very Although we can order them, the magnitude between them is not constant or equal Interval: differences are constant or equal Ex: rate 5-15 o These are all arbitrary Ratio: 0 matters; means there I an absence of whatever you’re measuring Not ex: A 0 on an IQ test does not mean that there is an absence of what you’re measuring Ex: time, weight - We need to think about o Reliability and validity Reliability: consistency of our measures Example: whether or not a drug makes a rat go through a maze faster Types of reliability Test- retest: test yourself and then test yourself again, tests reliabilty Interrater: judges, must be consistent with eachother o If they are not consistent, then low interrater reliability Replicability: imagine that your results are pulished and someone wants to replicate your study o If your results aren’t replicable and people can’t get those results again then your results are going to be discounted Validity: are your measures measuring what you say they are Accuracy Something can be reliable without being valid or vice versa Types o Construct validity: also called “face”: how well your operational definition represent your construct Ex: poor construct validity: if you use yawning to understand how smart you are o Internal validity: checks the integrity between the causual relationship between independent and dependent Measures whether the manipulation of IV really casues DV We need high internal validity Ex: you want to be able to say IV causes DV. Extraneous variable: something that would be causing the DV that’s not the IV We must control everything (extraneous variables) so that they can’t penetrate IV-> DV relationship Confounding variable: extraneous variable that varies systematically with the levels of the IV How to increase internal validity? Controlling extraneous variables o External validty: how well does your study represent what happens in the real world Low external validty: does not represent what is happening in the real world o Biggest issues Dance between internal and external validity Depends on your research question which is more important You hope for perfect balance between the two o Review First thing is to figure out IV Operationally define it o First question: quant or qual Qualitative; 3 levels: news, cartoon, crime show o All need to watch tv Random assignment to condition Put specifics on it Control extraneous variables DV Aggressive behavior Type: behavioral o Number of punches: ratio (difference beween each punch is different) o =
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'