Psych 2C03 - 2c03 notes

Psych 2C03 Week 2- Research Methods The Replication Crisis ● Failures to replicate science findings has led to what the APS calls a “credibility crisis” ● If failed to replicate findings, is it real? ○ Are not able to replicate findings ● Why these failures? ○ May not be well trained ○ Desperate to find something- fudging things ○ May make statistical errors ● Solutions? ○ Preregister- put in data plan, etc ■ Write up a proposal where you discuss hypothesis and methods ○ Open materials- whatever i used to set up this experiment, i will make it available to you ■ Can see everything ○ Open data- here’s my spreadsheets of all my data ■ Again can see everything Research Methods: The Scientific Method The theory-Data Cycle ● Start with an observation or existing theory ● From that, generate a hypothesis ○ What specific outcome you would expect ● From that, generate predictions ● Then, design your research and run it ● Then do some stats ○ If your data support your predictions, then you reinforce the validity of your observations or theory ○ If your data fail to support then, modify your theory, hypothesis, and/or predictions and try again ● Scientists are supposed to design research so that they prove themselves wrong instead of proving themselves right- because of confirmation bias Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

■ Need to have an experience that is like the experience they would have in the real world ■ The more real life like, the more generalizable, but the less control you have ○ Cannot maximize both at once- need to pick which is more important Design choices by validity type ● Observational studies - lowest internal validity, most external validity ○ Going out and watching people/animals, without knowing you are observing them ○ Usually correlational- this is associated with that but never this is what caused that ○ May lower external validity a bit since these people sometimes know you are studying them ● Quasi- and Field Experiments ○ Happens out in real world but can make a causal claim- do not have control over what happens in the real world ○ Cannot randomly assign- lose the ability to rule out alternative explanations for your affect ● Lab experiment- highest internal validity ○ Not desirable if you want to say your research generalizes to people other than those in your study Research Methods: Design Choices Maximizing Generalizability: Observational Studies ● Simply observe people you are researching ● No independent or control variables ● No random assignment ● No causal statements ● Three types: ○ Naturalistic ■ Going into the world and watching object of study ■ Most common ● Eg. watching animal behave in natural environment ■ May lower generalizability if they know they are being watched ■ Either hide or let participants know that they are being watched which will cause participants to change behaviour ○ Participant ■ Join the group of people you are interested in studying with a plan of what you will study Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

● Use when you’re interested in subject variables (where they come from, whether they smoke, etc.) ● Why do quasi experiments reduce our ability to make causal claims? Field experiments ● Done out in the real world ● In the street, at a park, in the mall, not in the lab ● Takes away control in experimental situation ○ Makes you less certain in drawing causal conclusions ■ Can't do them in lab space, or seeking generalizability ● Seeking generalizability ● Everything’s randomly assigned, but people in your experiment will have distractions and etc. which adds noise/confusion to data- reduces our ability to make causal claims ● ○ Research Methods 3: Confounds A few more things to worry about ● Self selection bias ○ Hear about research and are either very interested in research or not interested at all which causes a bias ○ Need to advertise as neutrally as possible- bland advertising ● Self presentation bias/ social desirability ○ Are my participants telling me the truth? ○ Are they behaving normally? ■ People like to look good and don’t want to seem weird- don’t show their true selves ○ Need to guarantee participants anonymity or confidentiality ● Bogus pipeline ○ Situation in which you make participants believe that you will know if they alter/fake anything ● Confirmation bias Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

2. The measured used in a study (experimental or otherwise) really measure the conceptual variables they were designed to measure ● Self reports ○ Interval-contingent self reports ■ Respondents report their experiences at regular intervals, usually once a day ○ Signal-contingent self reports ■ Respondents report their experiences as soon as possible after being signaled to do so, usually by means of a text message/special app ○ Event-contingent self reports ■ Report on a designated set of events as soon as possible after such events have occurred ● Interrater reliability ○ The level of agreement among multiple observers of the same behavior ● Correlational research ○ Research designed to measure the association between variables that are not manipulated by the researcher ○ Can be conducted using observational, archival, or survey methods ○ Measure the relationship between different variables ○ Do not manipulate variables, just measure them ○ Advantages ■ Can study the associations of naturally occurring variables that cannot be manipulated or induced- such as ethnicity, age, income ■ Can examine phenomena that would be difficult or unethical to create for research purposed, such as love, hate, abuse ○ Disadvantages ■ Correlation is not causation ■ Cannot demonstrate a cause-and-effect relationship ● Correlation coefficients ○ A statistical measure of the strength and direction of the association between two variables ■ The larger the absolute value of the number, the stronger the association between two variables ● Subject variables ○ Variable that characterizes preexisting differences among the participants in a study 1. Theories in social psychology attempt to explain and predict social psychological phenomena. The best theories are precise, explain all the relevant Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

● Thought about themselves then their mothers ● Thinking about mother and self in the same brain region vs ● Chinese- nothing going on because no overlap when they think about themselves vs mother ● USA- activation in brain since they use different regions when thinking about themselves vs mother Week 3 Self-esteem What is self-esteem? Rosenberg’s SE scale ● If SE is high: ○ On the whole i am satisfied with myself ○ I feel that i have a number of good qualities ○ I am able to do things as well as most people ○ I take a positive attitude towards myself ● Is SE is low: ○ At times i think i am no good at all ○ I feel i do not have much to be proud of ○ I certainly feel useless at times ○ All in all i am inclined to feel that i am a failure Self esteem is important! ● 3 sources of evidence: ○ Most people have high SE ○ Most people engage in self-serving cognitive biases to enhance or maintain their SE ○ Most people will do whatever it takes to maintain their positive SE 1. Most individualists have high SE ● scores are skewed high in individualist ● 93% canadians have higher self esteem ● Small correlations between SE and actual success ● Large correlations between SE and self-perceived success ● Even larger correlations between SE and self-perceived success on SE-relevant tasks 2. Self-serving cognitive biases ● automatic/unconscious: biases we use to enhance self esteem ○ Not aware that we are using them Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

Tesser’s self-esteem maintenance model ● 2x2 model ● Factor 1: relevance of task to self ○ Is it important to who you are? ○ How relevant is that task ● Factor 2: relationship to target ○ If you care about the person you should ○ If person is close to you but task is low relevance to you, will bask and feel happy for them ○ If person is close to you and task is high relevance, you will compare and feel bad about yourself ○ If distant relationship and don't care about task will not matter ○ If distant relationship and task is high relevance will bask in reflected glory Self-handicapping ● Sometimes, we’ll even hurt ourselves to preserve our self-esteem ○ This way they have an excuse to fail ● Males more likely to ● Cultural differences ● Collectivist cultures less likely to engage in self handicapping ● Had a task either really easy or really hard, told them they did really well ○ Were asked to do it again (those with hard task thought they got lucky and could not do it again) ○ After first time, they had to take a drug that either enhances or disrupts intellectual performance ○ Those with hard test chose disruptor drug more- self handicapping Summary ● We use all sorts of mechanisms to maintain or enhance our self esteem ● These mechanisms bias what’s available in long term memory ● They therefore bias how we perceive our successes and failures ● They can be harmful to ourselves, others, and our relationships Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

Culture and the FAE ● Indians and americans were asked to describe others ○ India: she brings pigs to my house (use concrete behaviour) ■ This is something someone does ○ America: she is friendly (use abstract traits) ● Morris & Peng experiment ○ Two different newspapers about a murder (american and asian) ○ American stories were more dispositional, asian were most situational ● Miller experiment ○ Asked different people to explain why people did what they did ○ As participants increased in age for americans, older the group, the more dispositional attributions they made ○ Not the same for indians- americans are trained to be more dispositional ○ Situational: no statistically significant differences, indians make more situational attributions as they get older Summary ● We have a tendency to believe that behaviours match dispositions- even in collectivist cultures ● Individualists have a tendency to ignore situational factors when making attributions for others behaviour ● Collectivists have a tendency to consider situational factors when making attributions for others behaviour Attribution Theory 4: Actor Observer Bias and Mechanisms Errors in Causal Attribution Actor-observer bias ● What is it? ○ Actors will tend to attribute the causes of their own behaviour to the situation, whereas observers will tend to attribute other’s behaviour to their stable dispositions ○ FAE plus ○ Eg. student has spent hours studying. She studied so hard because she has a hard exam and wants to do well- blaming intense studying on situation at task. Observers of her studying are going to say stuff like she’s very hard working Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

● After watching the videos, had to fill out a questionnaire, had questions like at what degree were the defendants coerced ● I n suspect-focused video - least coercion ● Detective-focused - most coercion ● Equal - fairest since you are seeing both which is least biased ● DV: perceive ed coercion (1-9 likert) ○ Likelihood of guilt Research 2: Diener and Wallbom ● Effect of manipulating perceptual salience on engagement in social norm violations ● Tell them they may not be back by when the buzzer is over- may use that to cheat- have some people in front of a mirror and some without ● Method ○ IV: Perceptual Salience (mirror vs no mirror) ○ DV: % cheating ■ 7% cheat when have to look at themselves ■ 70% cheat when they can avoid thinking about cheating Summary ● We make attributions about our own and others’ behaviours every moment of every day ● These attributions can be helpful or harmful, correct or incorrect ● Errors in causal attributions are due to: ○ How we are socialized ○ How well we know the targets of our attributions ○ What is most salient to us Attribution Theory 5: Two Factor Theory of Emotion Two-Factor Theory of Emotion ● We deduce our emotional state by assessing our: ○ Physiological arousal ○ Situation ■ Have to make a decision based on situation ● Our assessment usually leads to correct attributions ○ Eg. feeling fear, and hear a strange noise in house (physiological arousal shoots up, there is a situation) ● Our assessment can also lead to false attributions, thereby creating emotions we might not otherwise have had Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

○ 3. Random assignment to order of allocations of rewards- participants decide who gets rewards ○ 4. Allocation of rewards (DV) ■ Choice for one group predetermines choice for other group ■ No bias in allocation for rewards for ingroups (13;13) we see fairness ■ For outgroup-outgroup allocation, still gave 13:13 ■ Chose to give themselves 7 tokens so that the other group only gets 1 (harming their own group as well) ● Showing bias and prejudice towards outgroup Is this effect really about self esteem? (lemyre, smith) 1. Minimal groups assignment 2. Between groups IV: opportunity to make allocations (yes/no) ● Randomly assigned to make allocations 3. Within-grps IV: allocation type ● Those who get to make allocations make the same allocations 4. DV1: Allocations ● Found that people ended up using the same allocations 5. DV2: SE ● People who made allocations had significantly higher self esteem Prejudice 4: Relative Deprivation Theory Relative Deprivation Theory ● Lynchings between 1882 and 1930 ○ Looked at why they were so many lynches ○ When cotton prices failed, there were more. ● Explanation 1: frustration/aggression hypothesis ○ When they get frustrated, it leads to aggression- they were frustrated because of the not enough cotton, needed a scapegoat, blamed black people ○ If cotton prices become low, there is less lashing out ● Explanation 2: Relative deprivation theory ○ We have expectations for how our group is supposed to be doing, and when they are not doing as well, we get violent ■ Expectations based on past experience (temporal comparisons) ■ Expectations based on intergroup comparisons ● If someone in group is doing better, this makes us feel bad Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

● This practice “facilitates” activation of these concepts and “inhibits” activation of other concepts ○ Eg. saying doctor, first thing that comes to mind is a male Where do stereotypes come from? ● Usually born in culture- get it simply by contact from others stereotypes ● Where do they initially come from? ○ Got it from shared distinctiveness ■ Stereotypes are developed when two distinct events co-occur, and we mistakenly assume that they will always co-occur ● Eg. see a martian, highly distinct event, see them walking on their hands this is memorable, will assume this will always co-occur ■ Illusory correlation- not a real correlation, just made it up because we saw two distinct things happening at the same time Outgroup homogeneity ● Thinking that this one specific thing is the same for everyone in outgroups, but in ingroups everyone is different and has different characteristics ● Outgroup homogeneity and ingroup heterogeneity ● Role in stereotype development ● Role in stereotype maintenance Prejudice 6: Cognitive Approaches (pt 2) Encoding Bias ● Intentional Bias- notice some things and ignore other things ● Only encode what you notice ● Too much work to encode everything we see ● Have to encode stereotype irrelevant information ● Have to notice the opposite of your stereotype and encode it Research Example (Rothbart) ● IV: Target’s group membership (friendly vs intelligent) ● Dv: Recall ○ When target is intelligent people, focus more on intelligent behaviours rather than friendly behaviours ○ Friendly target; opposite results Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

○ ○ Time 2: no difference ○ People initially persuaded, but overtime go back to initial opinion ○ Low credibility source over time had no more of an impact than high credibility source after time Summary: Source Credibility ● Messages from high credibility sources become less persuasive over time ○ Normal decay ● Messages from low credibility sources become more persuasive over time ○ Sleeper effect ● Mechanism? ○ Dissociation hypothesis ■ Has to do with the source; discounting or acceptance use ■ Remember the message, but don’t remember connection of message to source over time Source characteristics: nonverbal cues ● Took away sound and cues; just looked at news reporters facial expression ● Tom was extremely neutral ● Dan was also extremely neutral ● Peter very happy, but more happy when talking about one person ○ Person asked which news source they watched and who they voted for ○ More people voted for reagan if they watched Peter (who had very positive facial expressions when talking about her) ○ But cannot prove this for sure, cannot infer causation due to correlation Persuasion 3: Target Characteristics Target Characteristics 1: Nonverbal Cues ● Hear pro-tuition-increase argument (from $587-$750) while: Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

● Enter the elaboration likelihood model Elaboration Likelihood Model ● Two ways of processing messages: ○ Central processing ■ Involves target working hard at processing a message ■ Work hard to come up with a conclusion ■ Carefully consider the message, compare pros and cons ■ Very effortful and time consuming ○ Peripheral processing ■ Using shortcuts and heuristics ■ Being persuaded in any way but deep thought ■ Superficial aspects- is it attractive etc (people who i like i tend to agree with, people who talk fast, people who are good looking) ■ People who I like, experts, people who talk fast, good looking people ● What predicts the use of central vs peripheral processing? ○ How much they care about the message Relevance and Mindlessness experiment ● 120 different people using copier approached by research assistant just before they deposited money into machine, asked if they could do theirs first ● IV1: size of request- 5 pages or 20 pages ● 2: reason provided- either really didn’t have a reason, or had a good/bad reason ● With the less papers, whether the reason was bad or good, lots of people said yes as long as they had some kind of reason ● With large papers, paying more attention to reason ● As long as they have a reason it works (93-94%) Summary ● Persuading someone to change their mind requires the persuader to consider interactions among the following ○ Who is the source and what are they doing? ○ Who is the target and what are they doing? ○ Is the message important to the recipient? ○ What medium should I use to convey this message October 26 Dissonance Theory Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

Defining psychological discomfort Aronson’s self-esteem explanation ● We feel psychologically uncomfortable when we see ourselves acting at odds with out self concept ● This harms our self esteem (SE) ● To restore our SE, we engage in dissonance reduction (DR) through attitude change ● But: if there is another way to restore our SE, then we need not to engage in attitude change Goffman on how to restore self-esteem 1. Making excuses- “i had no choice!” 2. Make justifications- “it was my job! I did it for the right reasons” 3. Focus on whether our behaviour caused harm- “no one (important) was hurt!” Methodological notes: use of counter-attitudinal essays to induce dissonance ● Eg. if you are pro life, you will write about pro choice ● Convince participants to write a strong essay ● Convince participants that they chose/did not choose to write the essay ○ Some participants felt that they chose to write the essay whereas some feel that they are forced ○ Those who were told “i would really appreciate it if you could do this” felt that they chose to write it ● Convince participants that writing the essay is their job Dissonance Theory 4: SE Explanation Research Impact of choice and reward on dissonance ● Counter-attitudinal essay in favour of law forbidding communists from speaking on college campuses ● IV1: Excuse (choice vs. no choice) ○ Have a choice and being paid high, have a choice and not being paid well, don’t have a choice and being paid well, don't have a choice and don’t being paid well ● IV2: justification (high vs low compensation) ● DV: attitude towards the law Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631

● 1. I don’t think that pot should be legalized ● 2. I wrote an essay for no real reward, saying that pot should be legalized ● 3. I wrote this essay for an audience that had already made up its mind ● 4. My essay won’t have any effect ● No dissonance, no distortion Harm done to whom? Cooper et al ● Replicate festinger and carlsmith ○ First: person perception task ○ Entirely scripted interview ○ Two different scripts, interviewee was very nice or very mean ○ Participant will have to lie to one of them ○ Second: boring peg turning task ○ Third: lie (unless control group) ○ Fourth (DV): attitude toward peg task (0 to 60) ● IVs ○ Liking for victim (high vs low) ○ Harm done (victim is convinced vs. not convinced) ○ ○ Only feel dissonance when have to lie to someone you like Possible cognitions ● 1. That experiment was very boring ● 2. I told the next subject that it was great fun ● 3. I liked him ● 4. He believed me ● Dissonance! Distortion of #1 ● 1. That experiment was very boring ● 2. I told the next subject it was great fun ● 3. I liked him ● 4. He didn’t believe me Downloaded by Kevin Lang (dilas34153@fesgrid.com) lOMoARcPSD|31076631