Class Notes (836,580)
Canada (509,856)
Psychology (3,518)
PSY100H1 (1,627)

PSYB01 - Lec 7 - Conducting Experiments (near-verbatim)

9 Pages
Unlock Document

Caroline Barakat

Lec 7: Conducting Experiments: Nov 1, 2012 Decisions to be made Outline: Who to use as participants and how many  Setting  Manipulation of the i.v.  Measuring the d.v.  Exact procedure o Level of description when writing up study  Controlling for expectancy effects  Manipulation checks o Checks to ensure that the manipulation of the iv actually worked the way you wanted it to  Debriefing Participants Ex) eye-witness testimony – how accurate – can we see how much eye-contact they make and determine this? Court-room behavior – what makes them more credible or less credible?  How many? o In general, bigger sample = better up to a certain point o Conducting study, want to get as many as possible o Getting participants can be hard o Try to get as many ppl as possible, but there is an upper limit, if you get more than this, even the smallest differences between ppl are going 2 seem like significant differences o To figure out what the maximum = use statistical consultant = what point you need to cut it off, where #s won’t give you results of true situation  Usually, it’s harder to get participants and this isn’t a problem o In general, minimum # that you’re supposed to have in each group = 30 – nobody knows where this # came from  # of participants in each cell/group  Ex) 3X2 design = 6 cells = 30X6 = 180 participants needed  So consider how many variables and how many levels you want – be practical because then you’ll need a lot of participants  Who? – best case = randomly select sample from population of interest but most of the time, not done – most of the time = convenience samples with undergrad @ uni o Convenience o Snowball  You ask your friend to ask some of their friends and so on – forward by email o Internet-based  Good source but have less control of participants; but will get broad range of ppl  Probability or non-probability sample? o All other things being equal, probability sample will be better b/c more generalizable but that’s not always your goal – sometimes, need to get things done quickly and won’t have time to get probability sample o And sometimes prob sample isn’t necessary – depends on research question o And what types of resources are available to you – money and time Setting o Computer lab space o Ppl interacting with computer o Regular lab space o Any kind of space; doesn’t necessarily mean lab with test tubes; it could be an office etc; o An enclosed space where you can control type of stimuli exposed to, ppl coming an going, keep materials locked up – an area where u can keep control of things o AV equipped? o Any other kind of special equipment; determine which area to conduct experiment o Some equipment you can take and move but others you can’t o Ex) gaze direction assessment equipment – measure where person looked on cpu screen  Advertising research – which place they looked @. Want person to look @ the product but if they don’t, then it’s not a good ad o If equipment off-site = more no-shows; more expensive and take more time for each participant (for set-up) o MRI = takes long time, set-up, questions about metal objects o More natural environment o Ex) quasi-experiment in outdoors o Natural enviro in lab – regular hallway in a building – Northern and Southern participants insulted o Depends on research question – what sort of space you use Manipulating the IV *doesn’t mean that you’re changing the IV during the experiment – just means that you’re introducing diff levels of the IV o Straightforward manipulation o Low realism  This is a downside; although participant is involved, they don’t feel that they are o Usually involve stimuli in the form of photos, video, text, etc.  Then ask them to make decision or assessing their rxn to stimulus o Another form of straightforward manipulation = presenting ppl with scenarios  Can be on computer or questionnaire paper that person fills out o More commonly done b/c it involves very little participation from research assistants/experimenter ; can be done more quickly; can be done in a # of diff settings  Data collected pretty quickly o Staged manipulation aka high-impact experiment o Intended to involve the participant; participant actually experiences something  Ex) Milgram study – obedience study – deliver electric shocks to student  High level of realism  Ex) line study of conformity; 3 lines of diff lengths – ask if same length – 8 ppl in the room and asked which lines were the same, 1 participant + other 7 were confederates – wanted to see if participant would cave to the pressure of conformity even though the answer was obviously wrong (stated by confederates) o Use of confederate(s) o High realism; see beh that ppl engage in when they don’t really know what’s going on o Need confederates that act the same way each time for each participant – can’t substitute confederates if they’re feeling sick one day – must keep constant  Extraneous variable that must be held constant o Not cheap, because one participant at a time; also time-consuming *not just doing one type of manipulation or another; sometimes do diff types of studies and they’ll be published in journal with the high-impact experiment bringing home the point Measuring the DV – many diff ways you can operationalize DV, esp in psychology 1) Self-report measures – can be written or verbal; report on own thought, feeling, beh o Advantages:  Convenient, easy to construct and administer  Esp if person can respond to on their own  Administer to more than one person at a time  Most measures already constructed and well-tested  Greater precision compared to others  Know that instrument is highly valid, get a lot of detail and it’s precise  Ex) questionnaire based on shyness – gives you a lot of info o Direct ppl to this website or copy onto your own  Private behaviours, thoughts, feelings etc  Not necessarily see these things if observing that they won’t do in public  Thoughts, attitudes not acted on  Weak link between attitudes and beh; there’s lot of attitudes that ppl don’t act on but you may just want 2 know ppl’s attitudes o Disadvantages:  Rely on honesty, memory  If what you’re asking has a social desirability component, they may not tell you the truth  Social desirability 2) Behavioural Measures – look @ what someone actually does; sometimes in lab setting (record) o Observable – not necessarily observable with naked eye o Can include facial expressions, proximity (how close someone is standing to something), large movements , micro-movements (see on videotape), reaction time (usu computer) o Must be (1) concrete (2) codeable o Advantages: o Can be used as proxy for inner states; facial expressions representing person’s inner state for ex  It’ll be hard for ppl to cover up what’s going on inside them o Visible, external indicators o More spontaneous, less filtered than verbal measures  How funny is this? How happy do you feel? You have to think about it and then you reply verbally  More direct way such as observing behavior, it’s not filtered, more direct way of knowing if someone is happy o No verbal skills  Ex) kids, ppl who speak another language, animals o Video: o Subway station in Tokyo – using behavioral measure after intervention o See how many ppl in business clothes and how many in casual attire; how many alone and how many with another person o How to code? Behaviors that you’re looking for have to be codeable if you’re using observational methods Facial Expression as a Behavioural Measure  Some expressions can be seen very clearly  Sometimes, can’t be observed; sometimes, person will have particular feeling and muscles in their face will move but not in a way that’s observable to humans – but you can still measure using electrodes – measure contraction and expansion of certain muscles  Particular patterns of expansion and contraction of certain muscles creates particular expressions; if someone is trying to conceal certain emotion, will see some expression in those muscles  Can use actual observable facial expression or use pattern of expansion and contraction of certain muscles to infer what the person is feeling  Can use this type of research to figure out whether ppl are lying – will be discordant with what they’re saying o Use in everyday life situations – like border control, police officers etc; 3) Implicit and physiological measures – even more direct than behavioral methods; participant can’t manipulate (if social desirability effect may be present, use this) o Implicit measures access automatic reactions and evaluations.  IAT (implicit associations test)  Ppl’s differences in reaction times to determine ppl’s attitudes towards diff things  Asks attitude towards particular thing without asking you directly; uses differences in reaction times – using keyboard  Switches the pairings and looking for differences in how quickly you can press correct button o If it takes longer to press button when female paired with career as opposed to male paired with career – slightly stronger association for femal
More Less

Related notes for PSY100H1

Log In


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.