Study Guides (238,586)
Canada (115,230)
Psychology (1,813)
PSYB01H3 (82)

PSYB01H3 Detailed class notes for Final exam 2012 fall From Oct 18- NOv 29

24 Pages
Unlock Document

University of Toronto Scarborough
Connie Boudens

Readings for Final exam: Oct 18 Experimental design Chapter 8, 10, Nonexperimental vs experimental methods (pp. 77-83) Oct 25 Midterm exam Nov 1 Conducting experiments Chapter 9 Nov 8 Additional research designs Chapter 11, lecture notes, Choosing a method (pp. 86 – 90) Nov 15 Data analysis: description and correlation Chapter 12 Nov 22 Data analysis: inferential statistics Chapter 13, Appendix B Nov 29 Generalizing and presenting results, Ethics Chapter 14, Chapter 3, Appendix A 2012-10-18 Lecture 6 Experiments A demonstration: the taste test before Coca Cola Vs Pepsi - Not good experiment because many things she talked may affect the results of the experiment - Reminding of relationships b/t variables like height and weight; the pet ownership and wealth o Eating bbreakfast are positively related to academi performance in school age children o Eating disorders and wtching soap in teenaged famales - But, does the above mean that eating breakfast causes better academic performance, or that watching soap operas causes eating disorders? - The correlation only tells you the two are correlated but not the causal Problems with establishing causality in correlational research  Direction of influence problem: there are may be two different directions o For ex: class attendance and grades  It could be higher attendance make a higher grade, but also, the students with higher grade are more likely to attend classes  Third variable problem o For ex:positive mood and charitable donations  These two are positively correlated  But there may be a 3 variableslike the day after they got the payment are more likely to donate o Ice cream and drawning:  The 3 variable is high temperature  In some cases, maybe the two above problems come together: problems with both direction of influence and third variables. o For example: depressed mood and impaired sleep, and also family conflicts Three things needed to establish causality  Temporal order must be correct: temporal ordercasue has to come before outcome  Variables have to covary: they have to in some way varying together  No other variable is causing the outcome: like in the expereiment of Coca, the different ice cold temperature of the two drinks, and this is the “other” variables How experiments allow you to do this  Hold all variables constant o May still have influence, but the influence will be the same on all participants  Use random assignment for anything that you cannot hold constant o in the random assignment to assign different ppl in group 1 or 2; the ppl have the equal opportunity to put in one condition or another  Present i.v. first to see what happens after: b.c the IV causes the DV  Present i.v. carefully and consistently, measure the d.v. rigorously(strictly) Extraneous vs confounding variable - Confounding variables: type of extraneous variables - Confounding variables are related to the IV o Ex: temp of drink confounded with type of drink, length of relationship confounded with gaze behaviour o Ex: when man has to apologize for, when the man is doing this, the woman was gazing him. Are we able to predict whether she will forgive him by calculating the length of time that she was looking at her; could we estimate the possibility based on the time she spent on looking at him; the problem was the confound was, the new lovers spending more time looking at each other. So that gazing is not a good way to tell whether she will forgive him. Basic experimental design  One i.v. two levels o Experimental group/the treatment group: experimental stimulus administered –in other words, something happens to this group o Control group :no experimental stimulus applied but all other things should be same as experimental group  One d.v. Experimental Group Administer the IV Measure the DV Control group Measure the DV Ex: effect of noise on memory Experimental group Memorize words in a loud room Measure performance on memory task Control group Memorize words in quiet room Measure performance on memory task - There is a problem here: ppl may remember things in different ways: like someone prefer memorizing in loud or quiet circumstance; so you need the random sampling Pre-test- posttest design - Pre-test: doing a test before the experiment. - Reasons for pre-test  Make sure groups are actually equivalent: like to make sure the average of ppl to memorizing ability are same  Identify people with high/low pre-existing characteristics to figure out that whether there is interaction o Ex: looking at effects of noise on introverts Vs extraverts  In case of mortality: not ppl dying/dropping out the experiment especially in a long time experiment; it is not a problem if the types of ppl dropping in two groups are same o Ex: the ppl don’t like crowd may be leaving and the experimental group ppl may become as the ppl who do not mind the crowd and they are doing good in the experiment as the ones in the quiet room  Measure change in each individual: especially when teaching something—measure the ability before and measure the ability afterwards Independent groups vs. repeated measures  Independent: all p’s exposed to one level of i.v.  Repeated measures: all p’s exposed to all levels of i.v. o less participants o No individual differences o Order effects: the problems of order effect  Order Effects: People behave differently in second condition because of 1 condition  Practice effects  Boredom or fatigue  Counterbalance conditions to counteract order effects  Example: the experiment about how much medication is appropriate o 10mg --depression measure -- 30Mg--depression measure o 30mgdepression measure 10gmdepression measure  But it there are many levels of IV the method used above will be problematic  Latin square design (many conditions)  Controlled for order effects when there are many conditions o Every condition appears in each ordinal position o For example: MAGIC  And then AGICM  In total you will have 5 different orders M A G I C A G I C A G I C I C M C M A  Matched Pairs (small # of participants)  When oly small # of participants is available or running participants is expensive  Each participant matched with another on something related to the DV o Ex: using a new technology to find information  Then randomly assigned to conditions Factorial Designs  More than one i.v.  Need to consider independent effect of each i.v, and joint effects. (i.e. does effect of the first IV depend on level of 2nd? As well as how personality interact with noisy or quiet condition—introvert or extraverts; like extra ones may do bad in the quiet section. ) o Consider how they act alone and who they act together  Example: If we were looking at gender and time of exam, these would be 2 IVs o Gender 2 levels o Time of exam: multiple levels—ex: morning, noon, evening o The DV: the performance of the exam Experimental Design Names  2 lo 2x3 design: make sure the orders are same, like 2 is for the 1 IV,  Ex: coffee drinking X time of day  Coffee has 2 levels: coffee or water  Time of day has 3 levels: morning, noon, night  3 levels of the first iv, 2 levels of the second,4 levels of the third o 3x2x4 design: it is not common cuz you need many designs and the interpreting is very complicated  Ex: coffee drinking X time of day X exam during  Coffee has 3 levels: 1, 2, 3 cups  Time of day has 2 levels: morning, night  Exam durations has 4 levels: 30 min, 60min, 90 min, 120min - Independent groups design o All IVs given to independent groups - Repeated Measures design o All IVs repeated measures - Mixed or Split-Plot design o If at least one IV is between groups and at least one IV is repeated measures - the difference like: ppl first time drinking coffee, the coffin has different affect on ppl who coffee drinking histories; st nd rd o time of day (am, pm), coffee (1 cup, 3 cups), year in school (1 , 2 , 3 )  you may measure one ppl in 1 year, and then measure again in 2 year, and then finally in 3 year - Main effect = effect the IV on the DV, collapsed across levels of the other i.v. . o Means ignoring the other variables - Interaction = joint effect of the 2(+) IV’s on the DV o If there is an interaction, the effect of one IV depends on the level of the second IV - Examples of Main effect... o  The line is just connecting the two dots  Males are more likely to seek approaching behaviour  But at this point, you may want to see if there is relationship b/t the gender and birth orders - Example of Interaction of the 2 IV’s indicated above o  It tells that: it does not matter for man if he is first born and final born  But in females it is very different  So it is not alone but gender and birth order matters together; and also there may be other variables matter too. Example: Investigating regional differences in aggression  Greater likelihood of fighting in the US South, and the West versus the North and Midwest - greater frequency of feuds, duels, homicides, lynchings, sniper attacks, violent pastimes, and greater support for corporal punishment, gun ownership, and war.  People (esp. men) expected to maintain reputation for vigilance – toughness is a strong cultural value  They are seen as the person who should stand up  Perhaps related to history of herding, and need to protect land  Insults viewed as highly offensive public challenges that must be met with an aggressive response  Contrasted white male students who had either grown up in the North or in the South.  Half were assigned to an “insult” condition where a confederate insulted them  Assessed changes in cortisol and testosterone levels  Cortisol level is positively related to southern participants from control to insult group  Cortisol level is negatively ……………………  In the second chart, Conducting Experiments 2012-11-01 Decisions to be made—steps to conduct an experiment - Who to use as participants and how many - Setting - Manipulation of the iv - Measuring the dv - Exact procedure - Controlling for expectancy effects - Manipulation checks? - Debriefing Example: Effects of eye contact on perceptions of honesty in eye witnesses - Does not matter if the witness make eye contact with the juries; Participants  How many? o A bigger sample is better up to a point  What is the point of cutting off? o 30 per condition considered minimum  No knowing where the number coming from  This is the least number you have to have in each group 1 2 IV 1 MINN30 MINN30 IV 2 MINN30 MINN30 IV 3 MINN30 MINN30  Who? o Convenience: most of the case; but the internet make things easier  If you are doing a thesis with someone have access to mass resources o Snowball sample: you ask your friends to ask their friends and going on  Similarly, you sending email to friends for them to forward o Internet-based: but you will have little control over who will participate, but usually you would get broad range of ppl  Probability or non-probability sample? o Probability: everything will be equal and more generaliable, but this is not always what you need; it depends on many situations like if there are any resources available; Setting o Computer lab space o Video capable sometimes will help to see the movements o Regular lab space o Sometimes, an office could be a lab; like a enclosed space, you could control the situation, and ppl’s mobility o AV equipped?—Audio Visualized equipment o Ex: a gaze direction equipment; used in advertising research to figure out where ppl are looking at; o Another example: measuring the oxygen ppl taking while exercising; o Usually it takes more money and more time to set up the participants o More natural environment o Ex: irregular hallway –experiment Manipulating the IV o Straightforward o Low realism: like showing ppl a picture, or a video and asking them the decisions or other reactions to the stimulus o Usually involve stimuli in the form of photos, video, text, etc.  In the example of the witness,, like asking the participant to look at the jury; the participant may not feel involved in the study  Ex: you try to figure out whether ppl take responsibility for differently for they have done  Usually, there is only one differences b/t the experimental and controlled group  Sometimes, it could present in the computer or in the form of questionnaire the person fills out  It requires very few from the researchers and participants to do something; usually in these cases are easier and convenient o Staged manipulation AKA high impact experiment o Intended to involve the participant in what is going on in the study o The participant is actually experiencing something  Like in the Mil’s Shock experiment: the participants know they are delivery the pain/shock to the learner; they set it very realistic; and the participants believe the things are really happening; very involving;  Another example of Line study in Conformity:  They have one participants and 7confederates to see ppl will agree the group even if the answer is very obvious o Often make Use of confederate(s):pretending; but the participants don’t know what is going on; the experiment of Line study;  The confederates have to act exactly same way and acting consistently  Sometimes, they are using actors in here; or using friends or researching system;  Usually, it is in cheat; and it takes time to run this; and make sure same confederates attending all the experiments Measuring the DV o Self-report measures o Advantages:  Convenient, easy to construct and administer: unusually something that the participants could respond in their own; or paper and pencil; researcher not necessarily to be there;  Allow for Greater precision: you will get lots of detail and the detail usually précised;  Example of self-filling questionnaire for shyness test;  Can reveal private behaviours, opinions, thoughts, emotions,  Can expose thoughts, attitudes that ppl may not act on o Disadvantages: like ppl may not tell you the truth  Rely on honesty, memory  Potential for Social desirability responding  How many alcohol did you consume every week; usually doctor write down the double number of what patient responds Behavioural Measures  What someone actually does; sometimes, may be videotapes; o Only observable behaviours o Can include facial expressions, proximity, large movements, micro-movements, reaction time  Micro-movement: sometimes things are easier to be observed on videotapes than the actual happening  Reaction time o Must be (1) concrete (2) codeable  Is it codeable?  How many ppl going through with business suits and with partners or alone?  It is hard to tell, like ppl is waiting for someone counted as alone or with somebody? What about the girl is in the cellphone? It is hard to count here…so that it is not codeable. o Advantages: o Visible, external indicators of inner states  Facial expression usually representing the inner states; but for some ppl it is hard to tell their inner states; most of the case, it is hard to cover o More spontaneous, less filtered than verbal measures  for verbal measures, ppl may take time to answer b/c ppl are thinking;  for movement, it is more direct o No requirement of verbal skills: children or animals o Facial Expression as a Behavioural Measure: muscles of the face expand and contract in different configurations depending on the emotion  Facial expression sometimes is the inner expression  To measure the movement of muscles  Particular patterns of muscles create certain expressions  Usually the patterns of contraction or expressions to observe the feeling of that person  It is useful sometimes to figure out whether ppl is lying; o Implicit and physiological measures for the questions that ppl are not comfortable/not telling the truth when answering the question o Implicit measures access automatic reactions and evaluations.  IAT  Google: implicit association test o  Most of the time, it is usually to measure your attitude because you cannot control how much time you will take to answer the question  Bogus pipeline: they set up a machine like it may tells the lie and the truth; at this condition, ppl ―knows‖ they are testing the truth and they are more likely to tell the truth. o Physiological measures  Autonomic arousal: measuring the body, like ppl are under threat  Skin conductance, GSR( measuring the skin, about how nervious the ppl is feeling ), Heart rate, respiration  Stress  EMG, or Cortisol levelgive you data about the heart,  Testosterone level: is increased that when someone is under threat  Neural activity: many options, ranging from very simple to very seficicated  EEG, MRI, fMRI data showing the patterns in the brain  it is usually not the research on the brain, but when ppl are making certain decisions; it is intersection between philosophy and physiology o the way ppl weight certain information o whether particular part of brain is active when ppl are making decisions; to see how the brain solves out  this part will be more and more usedit is impossible for ppl to manipulate on this Exact procedure o Needs to be very specific o Should allow for replicationsomeone could continue your research without contact you o Lay out what participants will experience in step-by-step fashion Not test in the exam for the following details: - Past tense; - Research proposal: future tense - Write down what experimenter exactly said in the experiment - For other researcher: they could just use your report and do what exactly you did The experimenter began by telling participants that they were going to participate in two unrelated tasks. The first task would involve concept mapping as a tool for representing concepts graphically. The second task would involve a new version of the Graduate Record Examination (GRE) in mathematics. Next, the students were randomly assigned to participate in one of three experimental conditions: self-concept maps with few nodes, self-concept maps with many nodes, or no self-concept maps. In the self-concept maps with many nodes condition (35 women, 12 men), the experimenter also explained concept mapping, but noted that “the best way to create a concept map is to create a complete description of the topic” and “provide plenty of information on the topic using multiple connections.” To illustrate the procedure, the experimenter showed participants a self-concept map that had the same five nodes as in the few nodes condition, but also had 46 additional nodes branching out from the central node in a hierarchical network of connections. Then these participants drew self-concept maps with the same materials as in the few nodes condition. In the no maps condition (23 women, 8 men), participants drew no self-concept maps, but instead proceeded directly to the additional experiment. Participants in all three conditions were told that they were “about to complete a portion of the quantitative GRE, and it is diagnostic of your mathematical reasoning abilities. Please carefully complete each of the test items, making sure not to skip any items.” The 20-minute test consisted of 30 difficult items drawn from sample GRE tests. Difficult items were used because previous research had shown that stereotype threat affects women's performance on relatively difficult rather than easy mathematics problems (Spencer et al., 1999). In a thorough debriefing, no participant guessed the experimental hypothesis. Controlling for expectancy effects o Demand characteristics o Participants behave in ways consistent with what they think experimenter wants or does not want; they tend to think like this is an experiment and behave accordingly; they thinking about they are smarter than the experimenter and they could figure out what the experiment really about; o Can happen when participant gets clues about nature of experiment: that is the beginning they tend to act inconsistently  Examples:  Experiment title: the Effects of Horror Movie on Mood  Obvious manipulation: Ten psychology students looking straight up  Biased or leading questions: don’t you think it is bad to murder unborn children? They may think this study is having something to do with abortion; o Controlling for Demand characteristics  Make participants “blind” to the condition: usually in the drug test with placebo exactly same as the manipulation drug  Make cues as neutral as possible: try to keep things same for the both groups;  Participants may need to “habituate” to being observedAllow participants to “habituate” to. o When ppl know they are videotaping or observed they may feel conscious;  Automation may standardize in the interaction with the participants may use the videotaping or computer o Use the tape in Mil’s experiment; the sound are same to different participants  Unobtrusive measurement: it is for less affected outcomes  Placebo for control groups: usually in the drug experiment  Use deception / cover story if none of these others will work; o Ex: self-esteem and changes in the sign of handwriting to see if ppl with lower self-esteem with write smaller; make sure the ppl in the low part acting same as the way when they are interacting with high part; o Sometimes, making the participants what is exactly you are looking for  Check for suspicions—do this in debriefing o If anyone knows everything about the study, you may have to look his data more carefully o Controlling for expectancy effects  Experimenter expectations: o Experimenter knows what condition p’s are in, and subtly influences results o Solutions:  Consistent behaviour  Automation  Naive experimenters  Double-blind procedure both the participants and researchers don’t know which conditions they are in; usually in the drug trail Manipulation Checks:  to make sure the manipulation of the IV is actually working; like to see ppl is really scared or not;  Assess if IV had intended effect on the participants  Two ways to check manipulation  explicit measure of the independent variable o like 1-10 of scared degree; or like a self esteem measure o eg. A measure of mood in an mood inducing experiment  questionnaire for participants after the experiment whether they perceived and how they interpreted the manipulation Debriefing  Can be written or oral: giving out a sheet about the details  Must have: o Purpose of experiment, some background, logic of experiment  What is for?  Something like the research in the past and to explain the background and logic of the study o Contact info for key people: who is the researcher and who could contact to o Option to withdraw data: if a person saying they don’t want to be used as data, and they could withdraw any time  Can also include: o Options for counselling or other assistance  If the person is affected psychologically; like in Mil’s study; ppl’s self-esteem is manipulated; the option of professional counselling is needed; o Request to maintain study details confidential: like not telling anybody else about the study to affect the studies on other ppl “Experiments on TV” - Evaluate this TV experiment to figure out the mistakes in the TV o Is this a good experiment? o What could be done to improve it?  The manipulation of the IV is not very clear, some ppl may misunderstand what is going on there  The time/place/physical appearance of the actors matter  The measure of DV is not consistent enough. Additional research designs 2012-11-15 Quasi-Experiments: similar to a experiment, but does not mean all the requirement as the experiment; also called field- experiment - Key difference to the real experiment: Less control of extraneous variables. o Use random assign is missing: groups often naturally occurring - Some common features of quasi-experiments: - matching instead of randomization - time series analysis - unit of analysis not people—could be an organization, a group or other things One Group Pre-post—it is very basic; instances of bullying training for staff instances of bullying - you start this in a school; it is very hard to isolate the school or staff to only monitor certain types of bullying so that it would only possible to observe the whole school. - Problems would happen as: threats to internal validityyou have strong evidence to determine your dependent variables. But in here we cannot really make the conclusion. o Threats to Internal Validity  History: event affects study outcomeit not only just meaning the things happen before but could be present condition  It is whatever happening between the pretest and posttest  Maturation:  Change in subjects over a period of time  Ex: you look at the Gr 8 in the beginning, but ppl are growing olderstaturally  term, ppl will change a lot and understand more about university. 1 year university  Testing / Repeated testing: they have to do the same test in 3 or more different ways; repeated test make ppl getting used to the test and may act differently; for example, just filling a questionnaire may make ppl aware or think differently about their own behaviours according to the definitions given by the researchers; which may have effect on decreasing or increasing in bullying  Mortality:  Subjects dropping outleaving the study; not just because ppl drop out but more importantly, it is about which types/groups of ppl are dropping; it would not be a problem if the dropping are random as the sample, but it matters there is particular group of ppl like ppl are low in suggestibility are more likely drop out a smoking- stop experiment  Regression to the mean  High or low measurements followed by measurements closer to group mean  you set up to do experiment in a school having high than average bullying in your pre-test, and later the bullying dropped to the average… but what about the school was just emerged from two different school, in which case, there always more conflicts at the first period of emerging…. And this would affect your experiment results. o It is like you catch a happy person in a low day; Non -equivalent control group design Experimental group instances of bullying training for staff instances of bullying Control group instances of bullying instances of bullying Selection bias: preexisting difference between the groups Time Series Analyses - When you cannot randomize participants, but can get assessments of the DV pre and post-treatment o Most of the time is the weather of a day;??? Usually you have a control group. Single-Case Research Designs: - Use only one case or group to investigate a specific phenomenon. - Not the same as a case study. - Uses time-series design. - Take multiple pre and post-treatment measures. Advantage of Small-N Design - Participants from hard to find populations—like it is only one participant for particular activity: ex the person walking through the Niagara Fall - Results easy to interpret (often no stats)—looking at the relationship about the height of the hill and the oxygen level—it is usually very straightforward; - Can focus on helping one (few) participant(s) A-B-A Design - Problems: end up with the baseline condition - Effect may not be reversible. - Ex: You take the sleep pills prescription for sleeping but you did getting better after one month; but you cannot really 100% sure it is the effect of the pills; the only way to find out is stopping taking the pills o If you stop taking the pills, and you return to base line; it could be some the residual effect of the pills or it may be you have new things to disturbing your sleep o The reason is ppl keeping changing; you cannot give the reason because you don‘t have the control on that to find out it is because of the treatment or because of other issues o So that you have to go back to the treatment Multiple Baseline Design • Testing a treatment effect when effect is irreversible. • What you need to here is you collect more baseline data • Baseline data collected on: • 2 or more behaviours for same individual
More Less

Related notes for PSYB01H3

Log In


Don't have an account?

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.