Comprehensive Guide to Psychological Research Methods and Ethics

Hey folks! Welcome to lecture two. Today, we're  talking about uh as it says on the tin here:   uh psychological research methods. This  is one of the most important lectures   that we're going to have over the course  of the semester because this represents  

how we lay the Bedrock for what we're going to  consider fact and Theory uh in in Psychology.   um Whenever we talk about uh sort of a fact of  psychology it was arrived at using one of these   methods or more of these methods one or more of  these methods um and and so it's important to  

understand that uh when we talk about psychology  being the systematic study of mind and behavior   these are the systems. This is  the systematic uh study part   um which is again the Bedrock of Psych right  um this is how we distinguish uh facts about  

people from myth from folklore from folk  wisdom uh things that might be useful   sometimes right but uh this is this this is  where we get closer to uh being able to uh to call what we are saying fact. So let's Dive In

So uh psychology follows the scientific method.  um It's not uh what determines whether something   is a science is not uh the subject matter  right it is the approach that you use right   um this is this comes up in my clinical work all  of the time because I I'll direct clients to do  

experiments on on their own emotions on their own  experience on their own experiences right and so   that brings that brings it closer to a science  as opposed to uh sort of going with their gut   right that kind of thing but to be even more  specific about what we're focused on here today  

um the scientific method is is the thing  that many of you learned really early in   your uh in your education um which  is uh this this process where we uh   have a this sort of circular relationship between  we have a hypothesis or an educated guess right  

it's important educated guess I have done research  based on This research it seems like the answer is   blank right and if the answer is blank right this  is the deductive part here if the answer is blank   then right uh rather if the answer is you know  x y z right we can use deductive reasoning  

right to uh to make predictions right so if  the Earth is uh orbiting around the sun right   um ooh better idea the if the if the Earth is  turning right that's our uh that's our hypothesis   the Earth rotates right then uh we can uh we  we can assume right that the sun will appear  

to move uh through the sky throughout the day and  will come up the next day at a pretty you know uh   predictable time period so what do we do we do we  we kind of do an experiment really what we do is   we stop and observe right based on our hypothesis  right so if this is true about how the Earth moves  

right then these you know then this will happen  right the sun will come back up the next day   right we make that prediction we see what happens  with our empirical observations and we go oh yeah   the sun went down and then the sun came up again  right um and that happens over and over and over  

again and that strengthens our hypothesis right  and this and this and this goes on right notice   there's not an end to this this keeps going  right as our observations continue to support   our hypothesis our general premise this  strengthens into a scientific theory and in  

some cases sign like a scientific law right um  Theory and law both very strong Concepts right   I think uh there's a tendency rhetorically to  say well that's just a theory you have a theory   um something doesn't become a scientific theory  until we've done many many experiments and the  

results all support it right and then it  becomes a theory that we might teach right   and so this and this goes on and on and on right  so if that's true then this then you know if   that's true then we use deduction right we use  a rule to draw a conclusion we make observations  

many observations many individual observations  and based on those many observations we decide   if our rule works or not right and this goes  on and on and on forever science is never going   to be finished right um there's always more  Nuance there's always new data there's always  

uh new ideas and uh we can always become more  precise we can always become more accurate right   um a thing that I want to to say here is for  instance somebody who goes and gets their PHD   um you write something called the dissertation  right uh and what a dissertation is is it's a  

really long paper that you've done a lot of  hard work on on a lot of research on and what   that dissertation really is right to put it  in the simplest terms is in order to get your   PhD you write this big paper and the result of  that paper is you you sort of create a new fact  

that's a very simple way of describing it an  overly simplified way of describing it but the   fact of the matter is is that's what every single  person who gets a PhD does is they make a new fact   foreign and they have a uh like a sort of a a  committee of other instructors who also have  

phds who look at this and go yeah all right  this is this is this is good right and and   to give you uh that should give you an idea  of how much there is for us to figure out   right if every person who gets a doctor in front  of their name has generated a new fact you can  

think about how small those facts must be right um  and think about how uh the the process of science   will go on forever we will always we will always  have a psychology right um and so there is no   finish line there's just always more to learn and  so we have our general theory from that theory we  

generate a hypothesis right if this is true then  I bet when I do a b c x y z will happen right   um if we take something like oh well uh you know  a person with with depression has uh is really   not so much sad they are that is a disease of  motivation somebody might say right so that's our  

Theory right actually depression is a disease of  motivation right not a disease of sadness okay so   then if that's the case right then uh individuals  with depression their reports of their mood will   be pretty diverse but uh they'll for the most  part report issues with school and work related  

to like you know completing work that's our  hypothesis right so what do we do right if   that's my hypothesis right so if it's true that  depression is really more about motivation than   it is about uh uh you know just feeling bummed  right um then uh then these things will be true  

so we could we go and we collect data we we grab  a uh and we'll talk more about this we grab 500   people with depression and we ask them a bunch of  questions or um We Gather a bunch of data about   their lives or we observe them or we put them in a  a special situation right we got 500 people and we  

do an experiment on them each individually right  there's a lot of different things we can do we'll   talk about many of those different things but we  collect data and we collect a lot of it and then   we analyze that data that is what statistics is  for that's why statistics is incredibly important  

and it's a class that you should take very  seriously and you should dive into with boldness   uh because statistics are everywhere they're  used constantly to convince you of things and   understanding what gen where those statistics  come from will make you more literate about what  

is what is accurate versus what is propaganda and  um statistics really comes alive when it helps you   answer your own questions about the world right  and so I I really uh those of you who maybe are   a little nervous about numbers I encourage you  to uh to dive in feet first to this process of  

analysis because it's very important so we analyze  that data and we make sure that what we found   right what we saw when we collected data we can  trust it to make conclusions and then whenever we   write up a report where we summarize our data  and Report our findings and either we go yeah  

it seems to support the theory that I based my  hypothesis on or you go well not quite right this   actually seems to you know refute my prediction  or refute the underlying Theory and then we we all   as a scientific Community got to talk about  what this means right so that's what things  

like conferences are for uh that's what things  like going out to lunch with your colleagues is   for it's like did you hear what so-and-so found  what is it like oh my gosh we gotta you know   start from scratch or we got to go back or we got  to rethink this right and those are very exciting  

and scary times right um both at the same time  often okay and that's the and that's the process   you should recognize this right because you uh  generally speaking if you if you had your if   you had your primary education here in the United  States particularly in California you very likely

um had to do like a science fair kind  of thing where you had a question and   then you had a you did some reading in the  library you checked out three books right   and then you generated a hypothesis right  and then you did an experiment or you made  

an observation right I watered you know like  I watered three different seeds I watered one   seed with water one with salt water and one  with Coca-Cola which one will grow more right   and then you saw you measured which plant grew  more right and then you wrote a little report  

about it right it was great that's it that's  it you did it you've done it already right   um are the procedures a little bigger and a little  more strict when you do it for real absolutely   but most of you again those of you educated in  California have done a version of this already

so uh the first kind of research we want  to talk about is descriptive research   where we describe a phenomenon the goals  of science there are four goals of science   not everything we do in science does all four but  ultimately we're trying to accomplish four things  

we want to describe explain predict and control  right we want to describe what's going on this   is what I'm seeing this is how bright it is this  is how loud it is this is how like long or short   or you know how long it takes that kind of thing  right we want to explain it the reason that this  

is happening is but likely because of blank right  likely because of ABC likely because of XYZ right   we want to predict and because you know  because of this explanation because of   this description I bet this is going to  happen in the future or when you do that  

it's going to explode right or it's going to grow  right or it's not going to grow right and then   control is okay so knowing all of these things  being able to make these predictions means that   if I build this particular apparatus I build this  structure I do this procedure this is going to  

happen right if for instance I know the positive  reinforcement is a really good way to motivate   a person to do what I want then if I want to get  a person to do what I want I need to praise them   when I see them doing that thing right I need to  reward them I need to catch them being good right  

it's like a basic child rearing thing catch your  kids being good and they'll be good for you right   um and so that's about prediction and control  right I bet if I spend a lot of energy   highlighting the things my kid does that they like  a lot they'll do more of those things right and  

then from there you get to this you know this idea  of control right well you know I can raise a kid   who is motivated to do well by doing XYZ right can  I totally control the kid's Behavior no absolutely   not but we can exert some influence right which  still counts as a kind of control at the end of  

the day so those are the four goals of science  descriptive research is about describing right   um and so uh there's a couple ways we do  this the first is observation this is where   we walk into a room and we watch we go to a  place and we watch we uh there are a lot of  

ways we can do this really systematically  um there's a procedure called coding   in which your observers are given like really  careful instructions about how to take notes   about the presence of certain things uh an example  of this is you're going to go to a playground and  

you're going to take data on aggression  That You observe right on the playground   and so then you know if you're thinking carefully  about this you have to ask yourself what counts as   aggression right and in a good observation  The Observers have been provided with very  

clear instructions about what observation about  what aggression is right you know aggression   is you know anytime a a kid comes  into contact with another kid um you know with with Force right  and that's that's already I've  

already done a bad job of uh of  what that coding might look like   um but the idea here is that ultimately  you know uh you get observers who are all   um all looking for the same thing right and doing  it in a way where two different observers right  

um would agree on what they were what they  were looking at so over time you get this this   information about like oh yeah so kids are way  more like we see more aggressive aggression before   launch right for during first recess than we do  during uh during lunch recess or the final like  

that 210 recess I don't know kids still get that  but the later in the day recess it's interesting   it seems like maybe kids these kids are hungry I  don't know right um that's uh that guessing why   that is is a different thing the observation  part is just when do we see more aggression  

oh morning recess uh and then after that it kind  of drops off all right that's that's interesting   another kind of descriptive research is  any kind of survey or interview right and   so what we're talking about here is you know  uh many of you have have taken surveys already  

um if you've taken a class at Golden West before I  I'm fairly certain you're given um uh like student   satisfaction service they're called they all have  every institution has different names for them   um but you're given a you know a questionnaire  about you know what was you know what ways did  

this teacher promote a good learning  environment right that kind of thing   um and that data gets collected and it  gives you a sort of description of how   the teacher is doing when you look at  all the results of their students right

um interviews same deal you sit across from  somebody and you you ask them a bunch of   questions sometimes those those questions are  set in advance and other times it can flow   a little more conversationally ultimately  after that happens you take a look at what  

was said you probably recorded that  interview and you are again looking   for particular themes right but always  again based on Theory right so you have   your theory about what you're looking for  and so you start looking for those things  

finally case studies so uh what a case study is  where we take one particular example one person   one phenomenon one thing that's happening right  and we ex exhaustively research it we look at   it very very very carefully right and get as  much information as we can about it right and

what that does right and I'm going to go back  I'm going to go back through each of these   um uh but what that does is that uh  gives us really really really detailed   information about something right a really good  description about something uh the weakness  

of a case study is that it's not really what  we we would call it it's not generalizable   we can't take what we learned from a  case study and apply it to other people   right uh a good example of this again kind of  leaning on my clinical background here a case  

study of an individual schizophrenia doesn't  actually tell us a ton about schizophrenia   because the presentation of that disorder  is radically different from person to person   right um that said we could have lots of you very  useful information about one person's experience  

that might apply to some people maybe uh but at  the very least it's going to give us lots to go on   when we start treating that person right surveys  and interviews are a great way to gather lots of   information from lots of people very quickly but  we get weird things some weaknesses are we get  

weird things like the person's interpretation  of the question can vary from participant to   participant uh and ultimately we can only really  ask very narrow questions a lot of the time right   um and so uh it's difficult to sometimes  capture information about more complex phenomena  

with observation uh what's cool about that is  we're not interfering right we the researcher   don't touch what's going on and so we get this  sort of unblemished naturalistic um representation   of what we're trying to study the bad thing  is because we're not doing any manipulation  

or exerting no control over it um it's very easy  to miss the factors uh that influence what we're   trying to understand right so uh I can observe  kids are more aggressive during the morning recess but I I get no closer to  understanding why that might be  

right because I haven't done anything I  haven't manipulated anything I haven't   introduced anything I haven't tried to change  anything or I'm not really like looking for   any other variables right and because of that any  other any of the other things that could be going  

on so because of that all I can do is say this  is how aggression changes throughout the day   and now there's much more work to do before  we can really use that information right   so ultimately descriptive research does not  answer the questions about why things are  

the way they are right we don't get to that  explanation piece that explanation goal of   science right we can do things like oh yeah well  then as you can see here you know we can go oh   well the Northern Lights appear at this time of  year in these places right but we get no closer  

to understanding well why does  this why does this happen right so one of the ways uh another uh one of the ways  we can start to get a little closer to this is   through something called correlational research  which is all about identifying relationships  

um and we measure a correlation from on a scale from negative one to one right  and what that does is that tells us how strong   the relationship is right so for example a good  uh um a really strong positive correlation right  

is the relationship between hours spent studying  right and your grade in the class right as you   spend more time studying we can predict right  that your grade in the class will go up now this tells us that there's a strong  relationship between those things right  

um a good example of a negative correlation and  this is one that I had to reckon with when I was   uh when I was in school is the more time you  spent a negative correlation is the more time   you spend watching TV or playing video games right  as that goes up your grade in the class is going  

to come down right so a positive relationship  is when one goes up the other goes up and a   negative relationship is when one goes up the  other goes down right now what I haven't said   is that they cause one another right so we'll uh  we'll get to that right so this negative one to  

one category right gives us the magnitude a one  is a perfect relationship or with with pretty   good certainty whenever right for every hour you  study your GP go GPA goes up by 0.1 right if I   can say that right we might have close to a an R  right a correlation coefficient an R value of one  

right um and that can get lower and lower  and lower it can get really really low   right a correlation of zero means  there's almost no relationship right   so the plus or minus gives us the direction of  the relationship here some uh uh kind of to give  

you an idea right so an R of 0.25 right we really  need a period here right an R of 0.25 right is the   very top of kind of weak between 0.25 and 0.5 is  moderate 0.5 to 0.75 is strong right and then 0.75   to 1 is very strong right in Psychology we love  strong we we will take strong we will break out  

the break out the champagne and the streamers for  for strong right um a lot of other than that of   a lot of other Natural Sciences don't really use  correlation because they can focus on experiments   um which we'll get to in a little bit but for  social sciences we like correlation because it  

um well we'll talk about why um but uh in Psych  we're very very happy about a strong and we're   over the moon super lucky if we see a very strong  and then one an R of one is a perfect relationship   we chart correlation on Scatter Plots to visualize  them right so this is an example of belong to the  

lecture the more yawns students have right  so here's a we got a long lecture here and   we're going to see many yawns right medium  electromedium ions if we take a bunch of   different lectures over time right so this was  a lecture that was in you know this was lecture  

in September a lecture in November a lecture  in December you know a lecture in the second   week of December third week of December  right and they're all of varying lengths   and they'll have varying numbers of yawns right if  we if we plot those on a graph we get the scatter  

plot and you can see the line what is called  the line of best fit is already drawn in here   but even if this line wasn't here you  can see these kind of go in a line   right and what that tells us that then the  factors vary in the same direction right  

when one goes up the other goes up so an example  of a negative correlation right the longer the   lecture the lower student attentiveness so  we have a long lecture with low attention   a short lecture with high attention a medium  lecture with medium attention you're getting  

this you're getting the idea here and then when we  chart many different lectures and we get this this   line and that's a negative correlation right  and here's our line of best fit right which   tells us oh yeah this is definitely a negative  right because this is this Clump sort of linear  

thank you and so these factors  vary in the opposite direction what you need to know is that  correlation does not equal causation   right let me I'm going to go back here  and that's annoying but here we go

length of time listening to a lecture does  not cause students to stop paying attention   right being in a long lecture does not cause yawns  right there's some there's some stuff that that   happens as a lecture gets longer right that's  really causing these changes correlation does  

not equal causation for instance hot weather is  strongly correlated with physical violence right   there are more violent crimes when the weather is  high does being hot cause people to be violent no   right no but there's a relationship between those  things right when it's hot there's more people out  

more people interacting with each other means  more opportunity for those interactions to go   badly right we can start to explain those  relationships right but we have to but the   correlation just tells us about a relationship  and so when you see that there's a a strong  

relationship between two things you can't assume  there's a causal relationship right well you have   to assume is that it's possible there's some other  variable causing those things to be connected   right so for instance why is Parental harshness  correlated with child Rebellion right parents  

who tend to be harsher tend to have more  rebellious children are harsh parents driving   their kids to Rebellion are rebellious  kids driving their parents to harshness   in stressed out families it's just if you're  stressed out it's just everybody that's just  

everybody doing badly right so parents have to  be harsh kids after our kids are rebelling right   um our families displayed genetically right our  ordinary families just so they just genetically   predispose to you know well if you've got parents  who happen to naturally be very harsh they're  

going to get together and have a kid who's  going to be who's also going to be harsh but   when you're a child you don't have any power so  instead you just are rebellious right um and more   right there's a million in perhaps infinite right  if you get creative enough explanations for why  

these might be related to each other right and so  all correlation can really do ultimately is point   you in the direction things to experiment on right  that's all it can really do in the end and so excuse me I had to adjust my C here um and so

you have to remember right I want to say  it again it's written not it's written up   here for you correlation does not equal  causation so when you see a correlational   result which are everywhere it's all  the time right such and such is linked  

to blank whenever you hear that right there's  a relationship between blank and blank right   it's correlational thing which doesn't mean you  can dismiss it entirely right because correlation   is great for identifying so-called risk factors  right which is like oh a person with x y and z  

we can predict pretty strongly that they're  going to have you know problem a right um so   an example an example of this uh where we use risk  factors a lot is thinking about things like eating   disorders again I'm leaning on clinical work here  um uh families uh kids who have parents who are  

preoccupied with weight and appearance who do  a lot of dieting themselves right so kids with   parents who diet right that increases the risk  of that kid getting an eating disorder right   observing a parent who diets period right and you  might go well there's good diets and there's bad  

diets it's like no it doesn't matter right just  a parent who is vocal and open about the fact   that they are restricting their food intake  increases the risk does it cause anorexia no   right but a family who has that a family that's  very critical right there's two risk factors  

right a family with a history of mood disorders  there's three risk factors right um and so uh   again those things do not cause anorexia but if  I see those things right it's worthwhile too in   the back of my head be like okay I need to be  on the lookout for maybe these concerns right

right and again we listed these these  possibilities here it was one you know   one through five it could be any combination right  does not settle why a behavior occurs right again   because we get this what I've been describing  is called the third variable problem we talk  

about things like confounding variables  it's the same thing it's a variable that   um or a factor right that we're not considering  right so here in this example we have parental   harshness and child Rebellion right those are the  variables right the level of how harsh a parent is  

and the level of childhood Rebellion those are our  variables right because they can vary if that's   all a variable is it's a thing that can vary so  don't get too uh caught up on that terminology and   so there might be a third variable here you know  um these things could relate to each other but the  

third variable could be stress the third variable  could be genetic neuroticism right just like how   kind of unpleasant of a person are you genetically  and have you passed that to your kids right another example happy mood predicts sociability  people who tend to be happy tend to be more social  

just being with others lift our spirits  to happy people seek out company right uh   is it that working on on your own  or being more isolated just based   on life circumstance make you sadder  right um do the same mechanisms that  

produce the feeling of good mood also produce  the desire of the desire to be social right again   could be more rain so again when you see a  relationship between two things just remember   one correlation design equal causation people  say it over and over again it's to the to the  

point where it begins to lose its meaning  here's what I'll say uh for you all when   you walk out of this class at the end of the  semester correlation just means you're not done   you have more work to do before you  can draw strong conclusions right

all right when we want to find  out the why we do an experiment   we have to do an experiment if  you didn't do an experiment you   don't know why something is happening  right this helps us determine causation  

so in an experiment right be randomly assign  people to groups we have an experimental   group and a control group so in an experimental  group we are going to do something to that group   right uh we look at the thing we think  might be causing a difference right that's  

our hypothesized cause and we manipulate it  that's our independent variable so for instance so harsh parents rebellious kids I  want to know what causes that to happen   why is that relationship there that's  interesting and my hypothesis is stress  

so my in my experiment group I take a bunch of  families and they are in my experimental group   and if I hypothesize that stress causes parents to  be harsh in kids to Rebel right stress becomes my   independent variable and stress is what I'm going  to manipulate what is that what on Earth does that  

mean it means I'm going to stress these families  out I'm going to subject them to an experimental   condition that is stressful I got to be gentle  about this we'll talk about ethics toward the end   here but I have to be a gentle right in the way  that I stress these families out for instance I  

might I might give them a really difficult puzzle  to solve with a timer or I might annoy them with   uh with loud music for a long period of time um uh  I could do something real extra and I'm thinking   of all the things that could be annoying here I  could um you know I could say hey family come to  

room 101 right and then I go oh Family actually  I'm so sorry we're going to be in room 102. oh oh   102 is full I think we got to go up to 201 right  up some stairs right oh really this is full let's   go to 205 right move them around a bunch and  and try to get them kind of keyed up of like  

the kid get you know kid gets settled down then  we gotta get the kid up and moving again right   I'm gonna annoy them I'm gonna stress them out   right um oh I could do this is the last one  of six and now my now I'm getting creative uh  

um which is gonna be bad for for time here another  thing you might you might do is you might tell the   family when they arrive um goodness you're a half  hour late because you told them you told them to   get here at two o'clock and then you're gonna  pretend that the time was actually 1 30 right  

so now the family feels bad or they're just like  ah or you're miscomicated with and now we've done   something wrong that's one group they're getting  they're getting the experimental manipulation   the control group are treated equally but  they don't get any manipulation so this  

family is going to come in right to the lab  still and we're going to put them in a room   with an activity a real like a neutral  activity here's a puzzle here's some toys   um you're just going to be here for a half  hour we don't do anything to them right

and then we look at the difference  between those rooms right so we're going to look at the difference between  those groups so we've got one group who doesn't   get stressed out another group who does get  stressed and then from there we're going to  

measure parental harshness and Rebellion Maybe  it's easier to pick one of those but that's not   terribly important right now um so we're gonna  so we're gonna measure parental harshness and   childhood child Rebellion how rebellious this kid  being right how much are they acting out and how  

harsh are the parents being in response to that  and we're gonna guess our hypothesis is that the   stressed out group will have more of those things  right and the group the control group who didn't   get stressed out we didn't make them stressed  out uh those levels are going to be lower right

and so what's important to remember right is  the reason we have a control group The reason   we have to have a control group is so that we can  get a baseline for how families tend to be right   for how these families tend to be um so that way  we can say look this is a group without stress  

when we add stress this shoots up otherwise  we just if we don't have a control group we   just have a number where it's like this group  experienced a level seven stress and without   the control group we have nothing to compare  that to what does that number mean right so  

out of this we get to make some statements about  causation right if child rebellion and parental   harshness go up when we stress the family out  we can say look parental harshness doesn't cause   kids to Rebel stress pushes both parts of this  family to do this negative behavior so stress is  

the cause of this right and the reason they vary  together is because stress affects them both right all right so when we do an experiment or  we make any uh draw any conclusion from   research we've done we need to be concerned  with two things uh validity and reliability  

so with validity we have a couple different  types there's external validity which is does   our experiment translate to the real world right  so for my example where we bring the family and   we stress them out um did I stress them out in  a way that they're likely to experience in real  

life personally I'm pretty proud of my two ideas  of bringing them from room to room over and over   again right letting the kid get settled and then  stressing everybody out by moving the whole family   again doing that a couple times or lying to them  basically and being like hey you're late you know  

um because that kind of stuff happens  all the time especially to families who   are disorganized right and so would we say that  experiment has external validity which is to say   um does this actually tell us anything about how  families are in real life outside the experiment  

right do the results generalized  to the real world right internal validity right um our department are  the dependent variable changes the result of   the independent variable or manipulation so we  have this uh you know we have this mischievous  

mischievous cook here who perhaps has snuck in  a secret ingredient or other variable into the   mix and the idea here is our dependent variable  right our dependent variable changes the result   of the independent variable manipulation so we did  the experiment we saw this change am I allowed to  

say that it was my stress condition right that  my experimental manipulation actually produced   the change in harshness and Rebellion or  did something sneak into my experiment   and I'm not talking about like a like an animal  snuck in and did something what I'm talking about  

is is there something about the way I'm doing  this experiment that allowed some other factor   to influence what's going on right it's  hard to ask these two things ultimately   does our experiment address the constructs  or ideas that I was trying to address  

right and we can think about that  question externally and internally   right uh also the thing to keep in mind is  are there bias or logical errors right in   the approach we take right um did I randomly  assign people to different groups or did I  

accidentally with my you know internal bias put  really disorganized families in one category   and the nice looking family so I just put  them in the control group so like oh you   all deserve a nice quiet time right um and  that would obviously mess up our experiment  

so there are there are a couple of biases  uh that can uh that can sneak into our   um that can sneak into our process here that  we need to make sure that we are aware of   um and these are uh these are  everywhere um everybody has biases um  

that uh that influence uh how we see the  world right nobody is nobody's unbiased right   um no individual can be objective right um there  are many people who really Pride themselves on   the kind of super reasonable and logical and I'm  I see things objectively um and that is a um uh uh

I don't want to give too much life advice here  but you're making people around you miserable by   pretending that that's true um is the gist  of it because um the only way to eliminate   that kind of bias is one it's impossible can't do  it but you can reduce it and the only way to do  

that is through consensus and cooperation with  others right and having many many observations   in many many instances and right so your  own personal perspective is never objective   um it is it is uh how you perceive the world  as a product of all of your experiences up  

to that point which are like as a matter  of course radically different from anyone   else so um the re the way that we can we try  to deal with that is by uh doing experiments   on uh many people right um and trying to  again like get consensus is the idea so  

um one you can talk about these biases everybody  can fall victim to these no individuals is you   know again without careful uh careful systemic  process no individual can really overcome these   um uh again without like doing a very careful  experimental design for instance so experimental  

bias is really straightforward um this is when you  the experimenter behave in a way that influences   what's going on in the experiment for instance  easy one of these my favorite example of these   is um you were doing research on a drug for  depression right one group is getting Placebo  

pills fake pills sugar pills they take the pill  nothing's gonna happen they might they get like   a small dose of sugar very very small to the point  where no one would notice it other group gets the   the actual you know antidepressant right if you  are aware of which pill is which you are going to  

be much just whether you like it or not you're  non-verbals right your demeanor right is going   to be different when you give a person the placebo  or the fake versus when you give them the one that   you expect as the experimenter you want this to  work because you're probably trying to prove my  

idea for a drug works so you're going to give  the you're going to as you're handing over the   uh the antidepressant that works you're going  to probably be a little more enthusiastic you're   going to communicate non-verbally very subtly  uh hope optimism right whereas the person who's  

getting the sugar pill you're going to hand  it to them and you're going to be like well   I'm wasting your time right which is in turn  going to influence the mood or the expectations   about that medication of the experimenter right  so the experimenter can again influence things in  

subtle ways um really straightforward ones  are things like you you the experimenter   can be tricked into like looking with  your eyes at like a correct answer or   the expected response right and so  that can influence what's going on  

um so demand characteristics this  one oh this one's rough this is   it's just so demand characteristics are basically  like do the conditions of the experiment give away   what's being measured right the conditions of  the experiment give away what's being measured  

right where then participants are being like Oh  you're trying to figure out this about me right   so um you know uh again um thinking about the  stressing families out thing um uh a family that's   stepping in knows they're in an experiment you  know the you know the the experiment comes out and  

says oh you're a half hour late and then the mom  goes like I think you're trying to stress us out   right because I know I've carefully read the  instructions I got here at two when I was   supposed to I think this is oh this is pretty  experiment right and then all of a sudden boom  

it's blown up right and so uh yeah the structure  of the experiment can give things away research   participant bias is a little bit different  it's very similar but a little bit different   by virtue of knowing you're in an experiment that  your behavior is being observed you change your  

behavior right there's a concept for instance  called social desirability and when especially   when we're doing stuff with surveys isn't just  experiments but when we're doing stuff with   surveys we have to watch out for this um if you're  researching stuff that is socially undesirable  

right or researching stuff about like activities  that generally are looked down on by Society or   people experience experience discrimination  because of these activities um you'll have   a harder time getting accurate data because many  people won't be open or honest right about those  

things right because there's shame or stigma  around them right so for instance an example   of this is trying to do research on the um the  prevalence of various like sexual fetishes right   um sexual fetishes that Society kind of goes out  of its way to kind of tease or make fun of right  

um fewer people are going to report having those  right um and they're gonna hide they're gonna   hide that right and so beyond this though um uh  when we bring somebody into a laboratory right or   you know a room that's relatively featureless  you know fluorescent lights plain Furniture  

um people get nervous people feel  uncomfortable and people are going to   act differently than they would in real life  which is going to hurt our external validity   and then the placebo effect this is a fun one  right when you ex when you expect medicine to work  

you start to feel it working before it's actually  doing anything chemically right so you take Advil   you feel a little bit better um but uh you know it  hasn't started working yet right um your your body   is is pretty powerful most of you are feeling all  of your feelings and thoughts and experiences are  

the product of right your body and so it stands  to reason that when you have an expectation   right A Feeling a different way there would  be some mechanisms that would start to kick in   throughout the rest of your body right  so you you take a sugar pill be after  

being told that's a it's a you know it's a pain  pill you're going to feel a little bit better   all right what's fun about the placebo effect  is um the the manner in which the placebo is   delivered it changes the effect so for instance if  I give you an injection of like a saline solution  

right so I give you I take a needle and I stick it  in you um you'll feel the placebo effect stronger   you'll feel better than if I just give you a pill  right because uh because your body anticipates   really strong medicine comes from Needles right  the color of the pill can affect this as well  

um and then my favorite example of this is and  I I see fewer and fewer students drinking these   all the time which is warms my heart because  they're gross and bad for you but um uh energy   drinks for instance uh do not have to taste the  way they do energy drinks could taste really good

um they could taste like  any other soda for instance   um however people feel like they work more  when they are made to taste more like medicine   right and so that weird kind of bitter  medicine taste that energy drinks have  

is there to engage your placebo effect and trick  you into being like Oh well this is gonna work   right and then when you have that  expectation it feels like it works right   um so that's a fun example of how placebo  effect is just used in product design  

so the way that we address this right remember  I talked about like well you can't you like an   individual can't really be objective um the way  that we address this is we and I talked about how   do we get past that right how do we do science if  an individual can't be objective well the way to  

get closer to this um is a double-blind experiment  in which the groups are randomly assigned right   and neither the participant nor the experimentor  knows which group is which so let's go back to the   um the antidepressant the depression pill example  um the way that works is um the participants don't  

know what group they're in right and the person  who is handing out the medication each week or   whatever isn't given uh is given an unlabeled  bottle right or rather on the bottle it just   says this is for participant ABC and then the  participant shows up and says hi I'm ADC and  

you just hand them that bottle of pills and you  have no idea if that's the experimental group or   the control group but the person who's going  to analyze the data later does know right oh   ABC got the placebo cool let me make sure they're  in that category right and that's how that works  

um it works really well and it IT  addresses a lot of these things right   um now uh bias uh sneaks into research in  all kinds of ways for instance um you know   who decides what research gets done for instance  what questions get asked um what sorts of uh um  

we'll get to this later I suppose but um uh  what sorts of sampling procedures are used   all of those things kind of get dictated  by people within the power structure of   um uh of a university and so um bias can  sneak in that way too it's not just how  

the experiment is conducted but which  questions are being asked right uh it   also matters right because it's a question of  like you know will this get us more funding   um are people generally interested in this are we  going to get featured you know on CNN or you know  

um or you know are we gonna get put on the news  right we're gonna get covered in the newspaper   if we do this experiment or it's just nobody  gonna care right other than us in this room   we're really excited about this topic um so buy  it that you know a double-blind experiment is not  

sufficient to remove all bias because the Again  by virtue of being a system made of people uh   the like the scientific Community  is also riddled with bias right   um and so it's much like I talked about  the scientific method is ongoing there  

is no end to science there's also no end to  addressing the impact of the human factor right   so that's not a Finish Line we can reach I think  a lot of people um feel exasperated when it's like   okay we fixed this issue but now there's this  other issue and the idea is it's not good it's  

not going to end because there's always going to  be um we're always going to have biases and stuff   that need to be addressed right and so tackling  bias needs to be part of this ongoing process all right

so um we can take different approaches to  uh um uh we can take different approaches   um to understanding a phenomena right  there's not one best approach right   um because you might say well the experiment  gives us well it gives us the answers as to  

why so we should just always do an  experiment but figuring out what the   heck we should do an experiment on takes a  lot of research first right so let's start   with this the apparent increase in belief  in conspiracy theories right like what

there seems to be a rise right with the rise  of like juvenon um I feel like uh I feel like   this peaked and it's finally going down but like I  think more like more people than ever more people   than like since the 1650s right believe the Earth  is flat right just a absurd stuff right um and so  

the rise of Rise of conspiracy theory um we might  have questions about where that comes from right   um and so uh a thing to to ask ourselves is  what are some different ways we could come to   understand this right so one is observation a  way we could do this is an experimenter could  

uh join a web form or web Community right like  a Facebook group or a subreddit or a Discord   Channel probably most likely these days right of  people who support this conspiracy let's just use   Q Anon right because there's you know there's a  fun documentary on HBO you can watch about that  

um uh and so I'm not going to get into  what that conspiracy theory covers but   um you join one of those Discord servers and just  observe right make notes about what conversations   you're seeing who's most active who's in charge  who posts the most right where is you know is  

there one individual who sort of like you know  feeds the ideas into this group or is it sort of   a much more Community more diffuse kind of thing  right also with observation this counts as well um   you know if you got the cooperation of the members  or leadership of this community you might be able  

to just be like hey can I just like get a a just  dump of your whole website for me to like that I   can search through and you know code like hey this  word appears a bunch or like oh look this this   you know this community also like is really like  super racist or this this community also like is  

really into you know it was weirdly in the video  games or right like there's you know what are the   other you know threads moving through this right  and we that would be up that would be observation   that would help us understand the shape of this  conspiracy theory right again doesn't tell us why  

we're getting more conspiracy theories but it  could give us a hint that we could then research   surveys interviews was as we reach out we contact  people who are part of these groups people who   espouse these theories we ask them questions  about things like well how did you get into  

this right like would be an interview question  like how did you learn about this um surveys of   things like how much do you believe this how much  do you believe this other conspiracy theory right   so we could start to get at things like hey  you know a person who believes one conspiracy  

theory very quickly will endorse a bunch of other  ones like we went from Q Anon to Bigfoot right   um and so we could be asking  questions right of these groups   studies we could take one person right and this  is this is like a 48 you know four hours uh like  

60 Minutes sort of um uh part of their approach  usually when they're talking about something is   they will hone in on one or two people to be  like this is the face of this right and so   um with the conspiracy theory thing I think the  thing we see a lot is it'll usually be a younger  

person somebody in their 20s or 30s right like  an adult like a young not young adult but like   a not yet middle-aged adult let's say that being  like yeah I lost my parents to this right they   fell down the hole and now they're you know deep  into this you know this stuff um and that would be  

closer to a case study where we're really going to  learn everything about this person's grandparents   like what's their background how are they raised  like okay Grandpa was in the military right so he   um and he fell in with like this particular  group and uh you know been over time his politics  

shifted right what caused that shift uh you know  oh you know the uh the impeachment of Nixon or   something right um you know did XYZ and so what  we're doing is we're we're working through the   steps sort of historically of this person's life  that led them to be somebody who was vulnerable  

to falling for conspiracy theories or um maybe we  don't talk about it as vulnerability but more like   uh cultivate an interest in these kinds of ideas  right and we try to use one person's journey and   really understand it super keenly super well  so that we can um start to wonder right about  

um how it you know we can then start to ask  questions is how similar is this journey to   everybody else's right and also too a useful  thing about case studies where I see this a lot is   um uh the idea of like I was able to save my  parents from this I got my parents to like not  

do it right and uh you know then we look at a  case study of that process of like oh so how do   we like heal a person who has then Fallen like  out of touch with reality right because again   the thing about conspiracy theories is um they  bring you out of touch with reality and then also  

because of that they make you unable to actually  solve the problems you're so concerned about right   if you're so concerned about a single like cabal  running the world you are then actually powerless   to solve the problems because um there is no such  cabal right um that's fiction and so if you're  

worried about this fake thing you can't actually  solve problems right um which is really useful   really useful to kind of maintaining the current  like power structure isn't it um and so uh uh   it's very helpful to understand these these  things correlational research right so we look  

at relationships between like one variable might  be depth of a person's Devotion to like I am 100   sure this is true right and I want to devote you  know 100 of my life to it right and then looking   at what variables predict that right the level  of endorsement right so we might do things like  

you know how isolated does this person feel or  did this person feel before they found this group   right um what else for conspiracy theories  um I could be mean and say we could look at   intelligence right um that's kind of a dubious  one you should be careful about but um we might  

be able we might find relations between yeah it  turns out you know the lower your right the lower   your IQ or there's a particular range of IQ that  really strongly predicts endorsing these right or   um it could just be like political affiliation  right like super strongly left leaning versus  

super strongly right leaning um and like which  end of the spectrum predicts you know uh um uh you know endorsement of these  things right so we could look for or   make guesses about what factors predict  an interest in conspiracy theory right

finally an experiment and so this is this is  where this is this is where this example I've   chosen gets a little rough because I believe  personally it would be unethical to do anything   that might encourage a person to support a  conspiracy theory because again it brings  

you out of touch with reality and then stops  you from actually being able to make your life   better right because again you're focused on this  fictional thing um right you're so like concerned   about like we have to show the world the World  is Flat right we have to show the world but the  

Illuminati wrote everything right um you know we  have to we have to show the world that um you know   this pizza shop is involved in child trafficking  right which brings us that's closer like kind of   a q Anon thing right and we get these people  get so preoccupied with these with with these  

with these things it's like it's bad for them it  isolates them from you know people who you know   from their families from their their friends who  you know uh want what's best for them and they get   stuck in these online communities that are again  wrapped up around this fiction so um I think it  

would be unethical to do an experiment that was  like Hey if we take this group of people and we   you know manipulate their Facebook or Instagram  feed to only show them conspiracy theory stuff   right I bet you know exposure to these ideas  will cause more people to subscribe to them right  

hopefully that makes sense right this idea of you  know okay well let's manipulate their feet which   by the way fun fact uh Facebook um and Instagram  I know it's the same company but um they uh were   caught recently doing experiments like some people  right where they looked and they were like hey I  

bet if we increase the amount of negative stuff  we show this person their posts will become more   negative and they did these kinds of experiments  and manipulations and they got caught and had to   stop it's a mean thing to do to somebody without  their consent right um we'll talk about that when  

we get to ethics so I think it would be unethical  to incur to do anything that would make a person   more likely to join a conspiracy theory so then  for me the experiment that I would then design   is I would take a group of people who are part of  a uh a conspiracy theorist community and examine  

which interventions can reduce their belief in it  and their support of it right and then we start to   get an idea of maybe what causes these ties it's  um and that is um uh you know I'm kind of like   one degree removed from what we're actually  interested in which is why are more people  

subscribing to these and so then do an experiment  on what gets people to stop believing in them is   not quite the same thing but it's important to  consider again the ethics of how you're doing   research right and so the experiment again here is  I want to uh try a particular intervention of like  

oh well we're gonna we're gonna um something  along the lines of like we're gonna sit these   people down with an actual person they believe  is part of the Illuminati or we're going to sit   these people down like we're gonna take these  flat earthers and we're gonna sit them down with  

um uh you know a pilot who has seen the curvature  of the Earth or an astronaut or that kind of   thing right maybe it direct human contact with the  so-called enemy right we'll fix this right so that   then we do an experiment where we take a control  group and we have the meat with just like a normal  

person and then we have an experimental group  where we have them meet with like an astronaut   and we see who still thinks about who thinks  the Earth is flat after those meetings right and so you can see that yeah I mean this  we've been on the slide for a while I realize  

um you know there's there's tons of different ways  and and I hope my hope is that you heard each of   these ideas and were kind of like oh well each  of those it'd be really interesting to see the   results of each of those right even though the  experiment might get us closer to why we get a  

lot of really valuable information from each of  these right um that could Inspire other research   so when we do research we need that we need a  sample we cannot do an experiment on the whole   world right um and so we have to think about  okay so we're doing an experiment and after the  

experiment we're going to draw some conclusions  who are we drawing conclusions about is a question   that's important to ask um and the answer to  that question is your population right so if   I want to know okay well if I'm interested  in college success right my population is  

probably college students right what factors  ensure that students graduate on time right   so my population is college students if I'm  going to do an experiment or I'm going to   learn I send out surveys I'm going to do  anything I have to choose who I'm going  

to look at so I'm going to take my population  of college students right and then I'm going   to draw a sample which is the portion of  the population we actually look at right ideally we want a representative sample so we take  a group of people out of this so we take let's say  

all of Golden West right and the population that I  take should be representative of the population of   Golden West so for instance this is an online  class that you're in right now so um if I am   going if I'm interested in college success and  like let's just say success transferring right  

um I need to be looking I need to make sure  that my sample isn't just taken from in-person   people right I can't just walk around the  quad or whatever and ask people questions   because then I'm leaving out everybody who's  taking classes online which is a large part  

of gwc right so if let's just say 50 of the  population of 50 of Golden West students are   in person and the other 50 are online my sample  should also be 50 50 in person and online so in   this example here right uh where we see these  people if I've got you know again 50 50 right  

and I'd get a sample of four which is really bad  ha it's a you need a big sample but half of them   should be in person half of them should be online  this does also true demographically this is also   true based on gender sexual orientation country of  origin all of these things we should try and think  

of as many demographic factors as possible and  try to build a sample that's representative right   um and so representative sample is the opposite of  a bias sample so for instance if I sit out on the   quad and give out surveys about transfer success  right I'm creating a bias sample because I'm only  

talking to people who won are on who are walking  on the quad which is like not in the I don't think   let's just say for sake of argument it's not in  the center of Campus so it's biased toward people   taking particular classes who parked at particular  parking lots right who happen to be there at a  

specific time of day I don't think Golden West  offers a lot of night night timing classes but   um uh but you know biases toward a particular  time of day right and all of these things might   be sneaky factors here right um for instance I  would bet students who are you know taking that  

8 30 or whatever like super first thing in the  morning class those students who are like get up   and go I bet they're trying I bet I bet they're  I bet they're more successful right I could be   totally wrong about that um but a student who's  willing to you know keep a uh keep a you know a  

decent sleep schedule who's willing to wake up  early for their education probably has a bunch   of other factors that are pushing them towards  success right do this is not to say that students   who aren't mourning people right um uh I think at  least because I'm including myself here at least  

half of your psych Department we're not morning  people right and so I'm I you know don't feel   if you're like an afternoon person don't worry  about it um the great philosopher Descartes not a   morning person and um uh the queen of Sweden made  him wake up super early and he died because of it  

so don't feel bad if you're not a morning person  um and don't look up that story because then but   it's more complicated than that um it was also  really cold in the morning in Sweden so I might   have not might have done it poor guy um and so uh  but I would I would guess generally speaking that  

um you know if you're interviewing students mostly  first thing in the morning you're going to get   a bunch of students who are you know a little  maybe a little bit more proactive I don't know   um and that could be my own personal bias  I'll really admit that right you know sorry  

for discriminating against uh afternoon people  and night owls um uh I respect all of you so   with that type with that terrible  digression out of the way I apologize   um uh hopefully you start to understand that  once we fail to get a representative sample  

all kinds of other things could start sneaking  into what we end up measuring right so usually we   try to get a representative sample by sampling  randomly and you might go well wait how does   that how does that work well if you have a big  enough sample drawn randomly you will eventually  

get a representative sample and the best way to  think about this is to really shrink the numbers   most of you can imagine flipping a coin  four times and getting heads three times   right so you flip a coin four times  and it comes up heads three times  

you and I would predict right if we flip a coin  four times we're gonna get heads twice and Tails   twice right we're gonna go 50 50 because the  coin has two sides but we all can imagine   flipping a coin four times and getting  those three heads in a row right

however and most most students  y'all can you'll get this kind   of intuitively if I flip that coin a million times I'm gonna be very very very close to 500  a million times yeah 500 000 heads and 500  

000 Tails because over time right that  50 50 probability is going to even out   right and the random sampling is the same thing if  we're truly drawing randomly from the population   like you know putting everybody's student ID  in a big hat and drawing it randomly right  

if we do that enough right if instead of taking  four people we instead take a thousand students   right we are much more likely  to get a sample that represents   the population right and then we don't have  to worry about you know somebody going through  

and being like okay well we need another  person with like a Vietnamese background   in here where we need another person with an  African-American background we need another   um we need another you know uh uh we need another  like we need another gay person we need three more  

straight people we need to you know we have to  like build we have to build the sample and when   you start really putting in that human factor you  let in lots of room to kind of make mistakes right   and so if you can get enough people randomly  you are more likely to have a sample that is uh  

um accurate right and so that's and so  Random large random samples are what you want   that's a big plus so uh there are a couple  different places where we can then do research   one is in the laboratory which is sort of an  artificial world uh the advantage of that is  

it's controlled right we can uh for instance  we can have do cool things like because we're   in a building with very few features in a room  with very few features we don't have to worry   if it's sunny or rainy right affecting somebody's  mood right you just come in and you're in a room  

no windows and it's just boring fluorescent  lighting and that's what everybody gets right   in a natural setting however we get to do more  naturalistic observation and so the advantage   of a natural setting is we're doing more  naturalistic observation which means we get  

to see people kind of out in the wild right we are  going to be able to generalize our findings more   right um possibly generalizer findings  more there's actually arguments either way   um but with this naturalistic setting uh  we're again we're seeing people in the wild  

we're seeing people as they are right and so  all of the little factors that might change a   person's Behavior are still present which could  either which make it so that it's hard for us to   make strong conclusions because it could be  some other sneaky factor in the environment  

but we can be a little bit more sure that  when this person goes somewhere else in the   real world right that will get similar results  whereas with a laboratory setting we have much   more control so we really isolate that variable  we can really say look everybody went through  

the exact same thing except for this one thing  I manipulated so I'm sure the changes we saw   are from this exact thing but then we have to  wonder can we take those results outside the lab   right an example of this is wow this person  really did this did this I did this wrote this  

report really well in school but then they had  to go and write a report for a job and all of   a sudden without a rubric all of a sudden without  um without knowing the sort of Industry standards   right that report actually doesn't hold up well  right so School trained you to write you know  

five paragraph essay right but when you got to  the Working World it turns out if you write more   than a paragraph your boss is annoyed with you for  writing so much and so really the skill you needed   outside of the quote the laboratory of the  classroom right was to be able to write  

really concisely and effectively right instead  of expanding your idea right into five paragraphs that's kind of an example of how that controlled  environment kind of backfire when you step outside   of it all right so that's how we gather our data  right and so then we have to analyze and interpret  

our data we do that with Statistics right which  is the math mathematical methods used to report   data so we have a couple different types we have  descriptive and inferential descriptive statistics   help us describe or summarize the mean median and  mode those are descriptive statistics right so the  

average number you've got is descriptive right so  let's see how let's start here so the measures of   central tendency are descriptive that's mean  median and mode right so let's go back here so   mean is like we have our average so we have three  instances three zeros 110 113 and two twenties  

when we total this and do the division we get an  average of of nine right we take the median we   line them all up and it's the one that sits in the  middle at the midpoint so that gives us 10 right   and then the mode is the thing that occurs  the most which is zero right and you can  

kind of see like oh these different Central  Tendencies actually tell us different things   about the data right the mean and the median  give us pretty similar things in nine and ten   um but the mode gives us this zero right so for  instance um a useful thing to think about is um  

housing prices it's you more useful  to use the median for housing prices   um because averages are pushed up by incredibly  expensive and Incredibly cheap houses there's   a lot more incredibly expensive houses right  and so that messes with the average a little  

bit but the median will put you right  in where the actual middle is generally   like so there's also measures of dispersion which  is how variable the data is how different is one   piece of data from another so we get the range  right which is 0 to 20 here for this group right  

and then the standard deviation which which  is I'll read this to here the square root   of the of the uh of the average squared  deviation from the mean namely 9.147 so   I'm not necessarily going to walk you through  that equation necessarily but what the standard  

deviation tells us right is you know so we  have our average right and so then how uh   in general how much is one how far is one  piece of data away from any other piece   right um and so that will uh that will  that will help us get a sense of uh how  

far from the average any particular data point is  the easiest way to think about this is thinking   about IQ the average IQ is 100 and the standard  deviation is 15. right so what that means is if   you have an 85 to a 115 IQ you are we can we can  we can make really accurate predictions um about  

63 of the population Falls between 85 and 115 IQ  and we know this because standard deviation um   uh that's the standard deviation works right and  as we get more standard deviations away from the   from the mean from the middle right from the the  average as we get more standard deviations away

we know we're talking about fewer and fewer people  so if you have a for instance if you have a 145   IQ you are a vanishingly rare intellect right uh  and then if somebody tries to tell me that they   have 160 IQ we're talking about fractions  of a percentage right of these existing  

right and I know all of this information because  I know that uh because I know the average   and I know the standard deviation of the data  I can go wow that is an extraordinarily unusual   number right that's what the measures of  dispersion are useful for we can go oh  

that's right at the end like with range we know  that's all the way at the end or all the way up   kind of the beginning it's the lowest possible  the most standard deviation we can go wow that's   a pretty like that number's above average  but not by much all right so I have 110 IQ  

cool sick that's you that's you got a good you  got a good old brain there well done somebody   tells me that 160 I'm gonna first be like well  you're probably lying or you took a test on the   internet and then then you know if if they like  well no here's my you know here's my score on the  

uh the weschler adult intelligence scale that was  done by a psychologist I'm gonna go oh my goodness   right that's incredibly rare so that's why it's  useful to have these measures of dispersion   because they're going to tell us uh we can then  derive a lot of information from a single point of  

data right we can go oh that's very very far above  or not that not that far above the average right inferential helps us draw conclusions they are the  bridge between the sample and the population right   so um uh they help us ask or help us answer the  question does the data confirm the hypothesis so  

let's go back to the coin example that I gave  right where you flip a coin four times and   you've got three heads and one tails if you flip a  coin four times and get three heads and one tails   the question I have for you and I want you  to take a second to answer this for yourself  

is it reasonable to assume if you flip a coin  four times and get three heads and one tails is   it reasonable to assume you have a trick coin  right there's something special about the coin   that the coin's been manipulated in some way  right the answer well you're still thinking  

is no it's not a reasonable assumption right  flip four coins I could grab a quarter right   now and I'd probably get three tails and one  heads right or three heads and one tail right   um it's not it's not reasonable it's you have so  few instances of flipping the coin right that it's  

it you might get it might come out a little weird  and you might get an extra right you might get an   extra you know an extra heads when you know the  probability suggests that over time you'll have   50 50 of each right now if you flip the coin  if you if you flipped the coin a million times  

or let's say a hundred times so the numbers are  always there if you flip a coin a hundred times   and you get 75 heads and 25 Tails right the same  ratio as before but now you flipped the coin a lot   it is reasonable to assume you've got a trick  coin right and inferential statistics help you  

look at more like so a coin is not complicated  it's a 50 52-sided thing inferential statistics   let you look at more complicated things and  then say was it your experiment that made the   difference appear right was it the when you when  you stressed that family out and they became more  

um rebellious and harsher right was that probably  because of your experiment or was it just a chance   right where sometimes the coins comes the coin  comes up with three heads sometimes the family is   just harsher right it has nothing to do with what  you did right and so inferential statistics help  

us answer this question and they help us identify  something called statistical significance right   um yeah so basically uh you are allowed  to say your experiment observed something   right you owe it what I did caused this difference  

right you're allowed to say that  if your statistics tell you that there's a five percent or less chance  that it was all random chance rather   that you can say with 95 certainty it was your  experiment that caused the difference to appear  

right and so we call this the alpha level or  the confidence level right um and so you and   your statistics and research methods classes you  will you will go over how to obtain how to get   these numbers right and uh we'll go into  a lot more detail about how to interpret  

your results right because it gets a little  more complicated than this but ultimately the   thing that you know the the thing that people  are obsessed with is this Alpha level right   um 0.05 right it's it's almost like a jokey value  among statisticians 0.05 that's it you know and so  

um what this does what most inferential statistics  will tell you right is at amongst many other   things right inferential statistics the different  kinds of analyzes like t-tests Anova and kova   mancova you know um uh you know uh regression  right that kind of thing all of those will  

tell you lots and lots of different things but the  thing that you generally tend to look for first is   you know did the differences I find are they  likely right is there a 95 chance that they're   up they're because of my manipulation right or is  there or is it you know right is there a greater  

than five percent chance that it was just my  sample was weird right or some other Factor foreign so your participants have rights [Music]  um hopefully that doesn't surprise you or   hopefully that doesn't make you angry rights what  are you talking about um uh your participants have  

uh have rights um that should be respected um the  medical and psychological Community the medical   community in particular just because there's  worse stuff that they can do but the medical   community has a long history of um not really  respecting the rights of various kinds of patients  

um I just I think time wise we're running along  here and so I I'm not going to necessarily get   into too many specifics because they're really  Dreadful stories but um uh the Tuskegee syphilis   experiment in particular is one example  um uh that is worth is worth investigating  

um and demonstrates the need for strong  ethical oversight for all experiments   um and so I hope um here in 2022 you can kind of  accept on face value that ethical oversight for   all experiments involving humans um and maybe even  animals we can go there too um is uh is worthwhile  

um to ensure that everybody is protected right so   uh the American Psychological Association has  some guidelines for how to do experiments that   generally need to be followed any University  that conducts experiments on people has  

um uh an internal review board an IRB that's  meant to make sure that all all experimenters   are educated and are following those guidelines so  that one component is informed consent you have to   to some extent tell people ahead of time here's  what's going to happen right now there's room to

in some cases you're not going to tell people  like here's what we're trying to learn right   but instead you're going to say like oh well  you're going to do um you know you're going   to be asked to do an activity together as a  family and here's what that's going to entail  

and here's what kind of feelings you might  have right that kind of thing confidentiality   you have to protect people's identities right  um this is I it was wild I saw somebody that   somebody was obviously not a scientist or not  part of Academia was very much like I can't  

trust this data because the participants haven't  been identified it's like no you can't do that   um because if people think that their names  are going to get out they're going to behave   differently and act differently and so we have  to protect people's identity one because it's  

worth researching sensitive topics or topics that  maybe that there's societal stigma around and two   um when people know that their names are  going to be published next to their data right   um they get nervous right imagine being like  oh we're going to give an IQ test to a thousand  

people and we're going to print everybody's IQ  score right and their name right people are going   to not want to do that right only people who  are confident in their intelligence are going   to sign up or be willing to do that and so we  protect people's identities to go hey come as  

you are right and no one will find out it was  you because all your data your your data is   just going to be one de-identified line amongst  thousands of participants right so that protects   people and it encourages people to you know behave  realistically there's debriefing afterward where  

after they're done with an experiment we tell  participants hey here's here's actually what we   were looking for right here's what we're looking  at you know any questions um here's an email you   can reach out to if you like feel sad about what  happened right and that doesn't that you know  

nope basically nobody calls the number but um you  know you give it to people anyway right so if you   feel upset because they're you know I have some  big feelings about what what happened you know   um call us here's what we were looking at  here's why this part of the experiment happened  

um here here let us let's reveal to you  here's an element of deception right   um we you know for instance with the the  stressed out families like we totally knew   we told you to come here at two we told you 1  30 to you know increase the stress level of the  

situation right we might go over those pieces  right debriefing I think uh um comes a lot from   the fact that there are some experiments you can  do in Psychology where people will learn things   about themselves that they maybe weren't  quite prepared to learn about themselves  

um well and we'll we will probably talk about some  of those experiments his some those historical   experiments over the course of the class deception  you have to be very you have to make a good case   for why you're deceiving people if you're going  to lie to people you have to there has to be a  

good reason for it at the end of the day good  reasons include things like look if we were   open and upfront about exactly what we wanted  this would totally disrupt the process and it's   pretty easy to make that case but um you need  to at least have good reason to deceive folks  

and so yeah institutional review boards um uh  are super important I'm speaking of somebody   my thesis uh got super tangled up in my in my  IRB at my institution at Cal State Fullerton   um because I was working with kids my my thesis  was on an efficacy study about a therapeutic  

summer camp and because I was working with kids  um there was a lot of really intense scrutiny   paid in my project and it really delayed  me and kind of messed me up a little bit   um and uh I was very sad at the time  but even then and especially now  

um that was hugely necessary super important that  we're making sure that these again these kids who   are suffering from very psychological disorders  and social emotional behavioral issues are being   protected and not being subjected to something  that's going to like hinder their um their healing  

right like he didn't hinder their progress um  and so again I'm speaking from somebody who was   massively inconvenienced by their IRB to the point  where it almost delayed my graduation um and I I'm   I'm here telling you super necessary super super  important cooperate with your IRB and recognize  

that they are they're doing something to ensure  that our discipline broadly remains respected   and remains a force for like good and progress  right and actually improving people's lives so   um yeah uh so I I kind of wanted to mention  that again you know coming from a perspective  

where you might expect me to be a little bitter  about it but now um super important um and I'm   glad that those kids um I I was certainly uh  advocating for and uh wanting very badly to   take care of the the kids I worked with on this  experiment but also I'm glad that they had a  

um uh even more even more Advocates um yeah so  that is those are sort of the basics of how we   approach research and psychology right this is  how we the what we've gone over is is how we   um is how we we build knowledge in Psychology  every fact you're going to read in your textbook  

for the rest of the semester is the product of  an experiment or the product of really intense   correlational investigation right using  using many like you know lots and lots of   um carefully designed correlational studies and  that have been repeated ideally not necessarily  

all of them have been successfully repeated  but the idea is to replicate as many of these   studies as possible so that we can confirm  that they're accurate and so everything we're   going to talk about is the product of this  process that we just went over right and so  

um whenever we talk about an experiment in the  findings of those experiments you can assume right   that the methods I've covered were used right and  so when you hear a fact or something that we talk   about that you feel skeptical about it's okay or  it's like that's really matched my experience what  

I want to encourage you to remember is that um uh  your individual experience is unique and distinct   right and the sort of uh the concepts  that have been identified right that   we're describing uh are usually the sum of  the the experiences of thousands of people  

right hundreds of people dozens of people  right and I hope that you'll then take   your own personal experience and understand  that maybe your experience is the exception   right or an exception or that perhaps the way that  you're thinking about your experience maybe is is  

subject to your your kind of your personal bias or  expectations right whereas these things that we'll   be talking about this semester are the product  of processes designed to iron those things out   right and so that basically what I'm saying is  it is safe in this class to approach some of  

the stuff we're talking about with an open mind um  because uh nothing I tell you is made up right so   um I gotta go to Great pains to make sure that  nothing I tell you is just made up and if I do   tell you something that's just made up or  is my best guess I will tell you ahead of  

time I will I will say something like I'm  taking my professor hat off for a second   to say that sort of make this yesterday answer  this question so you know carefully read this   chapter there's much more in the chapter uh  well not much more but there's there's there's  

more in the chapter so make sure you take a  look at that that's always going to be true   um and um yeah I think carefully about these these  processes and think carefully about the questions   you have and let's see if we find you some answers  over the course of the semester um yeah so I will  

uh you will you'll hear more from me uh next week  and that's going to be it for now take care folks